EUR-Lex Access to European Union law

Back to EUR-Lex homepage

This document is an excerpt from the EUR-Lex website

Document 52019SC0001

COMMISSION STAFF WORKING DOCUMENT Accompanying the document REPORT FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS Energy prices and costs in Europe

SWD/2019/1 final

Brussels, 9.1.2019

SWD(2019) 1 final

COMMISSION STAFF WORKING DOCUMENT

Accompanying the document

REPORT FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS

Energy prices and costs in Europe

{COM(2019) 1 final}


Contents

Introduction    10

1    Electricity prices    13

1.1    Wholesale electricity prices    13

1.1.1    Evolution of wholesale electricity prices    14

1.1.2    Factors impacting the evolution of wholesale prices    18

1.1.3    International comparisons    23

1.2    Retail Electricity Prices    26

1.2.1    Household Electricity Prices    30

1.2.2    Industrial Electricity Prices    38

1.2.3    Small vs. Large Industrial Electricity Prices    44

1.2.4    International comparisons    51

2    Gas prices    54

2.1    Wholesale gas prices    54

2.1.1    Evolution of wholesale gas prices    55

2.1.2    Factors impacting the evolution of wholesale gas prices    60

2.1.3    International comparison    64

2.2    Retail gas prices    68

2.2.1    Household Natural Gas Prices    70

2.2.2    Industrial Natural Gas Prices    73

2.2.3    International comparisons    76

3    Oil and oil product prices    79

3.1    Crude oil prices    79

3.2    Wholesale prices of oil products    81

3.3    Retail prices of oil products    82

3.3.1    Methodology    84

3.3.2    General findings    84

3.3.3    Gasoline    87

3.3.4    Diesel    90

3.3.5    Heating oil    94

3.3.6    Gasoline vs diesel    97

3.3.7    International comparison    103

4    The EU energy bill    107

4.1    Introduction    107

4.2    Methodology    108

4.3    Drivers    110

4.4    Import bill calculation    114

5    Household energy expenditure and energy poverty    117

5.1    Energy products in household budgets    118

5.1.1    Energy expenditure (excluding transport) in households with low income    118

5.1.2    Energy expenditure (excluding transport) in households with middle income    120

5.1.3    Energy expenditures in the transport sector    123

6    Industry energy costs    125

6.1    Energy costs and competitiveness at macroeconomic level    128

6.1.1    Competitiveness drivers: EU vs G20    129

6.1.2    Impact of energy on the economy's competitiveness    139

6.2    Energy costs for industry    146

6.3    Exploring energy intensities    159

6.4    Energy costs drivers    163

6.4.1    Drivers of energy costs (Purchases of energy)    163

6.4.2    Impact of energy costs on Total Production Costs    168

6.4.3    Drivers of the energy costs as a share of production costs    169

6.5    International comparisons    171

6.5.1    Energy costs vs other G20 countries    172

6.5.2    Energy intensity of EU sectors vs other G20    174

6.5.3    Industrial electricity prices: EU vs G20 countries    177

6.5.4    Industrial gas prices: EU vs G20 countries    184

6.6    Case studies of selected Energy Intensive Industries    193

6.6.1    Scope and samples    193

6.6.2    Cross-Sectorial comparisons    195

6.6.3    Prices across Member States    199

6.6.4    Overview of results of specific case studies    208

7    Energy subsidies    211

7.1    Subsidies in the energy sector    216

7.1.1    Fossil fuel subsidies in the energy sector    217

7.1.2    Renewables    218

7.1.3    Nuclear    219

7.1.4    Other specific subsidies    221

7.2    Subsides beyond the energy sector    224

7.2.1    Fossil fuels subsidies beyond the energy sector    224

7.2.2    Sectors receiving subsidies beyond the energy industry    225

7.3    Impact of subsidies on energy prices for consumers    227

7.4    International comparisons of fossil fuel subsidies    231

8    The role of energy for government revenues and inflation    233

8.1    Government revenues from the energy sector    233

8.1.1    Energy taxes    233

8.1.2    Excise duties    237

8.1.3    Value added tax (VAT)    241

8.1.4    Tax revenues from oil products    242

9    Prices and costs and future investments    245

9.1    Introduction and definitions    246

9.2    Drivers of electricity costs and prices    247

9.2.1    Renewable costs are declining    247

9.2.2    Fossil fuel prices and carbon price    249

9.2.2.1    Gas price    249

9.2.2.2    Coal price    250

9.2.2.3    Oil price    250

9.2.3    Carbon price    251

9.3    Need for new capacities    252

9.3.1    Evolution of electricity demand    252

9.3.1.1    Age structure of fossil fuel fired power plants    254

9.4    Electricity prices vs costs up to 2030 in the EU    257

9.5    Investment in power production assets    258

10    Price regulation    263

10.1    Impact of price regulation on competition, prices, quality of services and investments    264

10.2    Price regulation in household gas and electricity markets    266

10.2.1    Share of households under regulated prices    267

10.2.2    Share of consumers under social tariffs    270

10.3    Impact of regulated prices on household retail markets    272

10.3.1    Impact on competition and consumer engagement    272

10.3.2    Market concentration    279

10.3.3    Impact on prices and costs    282

10.3.4    Impact on consumer perception    290

10.4    Impact of regulated prices on non-household retail markets    292

10.5    Impact of regulated prices on tariff deficits and investments    297

10.6    Dynamic pricing    298

10.6.1    Objective, methodological approach and assumptions    298

10.6.2    Preliminary results    301

10.6.3    Robustness of results and further research    307



List of Figures

Figure 1 - Evolution of monthly average wholesale day-ahead baseload electricity prices in Europe, showing the European Power Benchmark and the range of minimum and maximum prices across the markets    14

Figure 2 - Evolution of wholesale electricity prices in Europe since 2016    15

Figure 3 - Regional market prices in the North-Western Europe coupled area    16

Figure 4 - The Central Eastern Europe average wholesale price and the EPB benchmark    17

Figure 5 - Regional market prices in Italy and South Eastern Europe    18

Figure 6 - Electricity consumption and economic growth    19

Figure 7 - Electricity generation mix in the EU-28 (actual power generation)    19

Figure 8 - Monthly electricity generation in the EU and the share of some generation sources in the EU electricity mix    20

Figure 9 - Monthly coal, natural gas and carbon price indexes, compared to the 2008 average price and the share of renewable energy (right hand scale)    21

Figure 10 - Net electricity flow position compared to domestic electricity generation in the European power regions    22

Figure 11 - Comparison of wholesale electricity prices in the EU with global trade partners    24

Figure 12 - Comparison of wholesale electricity prices in the EU with global trade partners    24

Figure 13 - Evolution and composition of the EU household price (DC). MR = Most representative    29

Figure 14 - Household rices in 2017 (most rep.)    31

Figure 15 - Composition of hosehold prices in 2017 (most rep.)    32

Figure 16 - Breakdown of household prices (DC)    33

Figure 17 - Composition of taxes in 2017 (Most rep.)    33

Figure 18 - Taxes, ees, levies and charges for EU households (DC)    37

Figure 19 - Evolution and composition of the EU industrial price (ID)    38

Figure 20 - Industrial (ID) electricity prices in 2017    39

Figure 21 - Composition of industrial (ID) electricity prices in 2017    39

Figure 22 - Breakdown of EU prices in 2017 (ID)    40

Figure 23 - Composition of taxes in 2017 (ID)    41

Figure 24 - Taxes, levies, fees and charges of industrial electricity prices    43

Figure 25 - Evolution of small (IB) and large (IF) industrial electricity prices    44

Figure 26 - Composition of small (IB) and large (IF) industrial electricity prices    45

Figure 27- Composition of small (IB) and large (IF) industrial prices in 2017. First 16 countries alphabetically.    46

Figure 28 - Composition of small (IB) and large (IF) industrial prices in 2017. Last 15 countries alphabetically.    47

Figure 29 - Composition of taxes on prices for small (IB) and large (IF) industrial electricity prices (2008-2017)    47

Figure 30 - Composition of taxes on prices for small (IB) and large (IF) industrial electricity consumers in 2017    47

Figure 31 - RES support costs for electricity consumers (DC=Household, IB=small, ID=median, IF=large)    48

Figure 32 - RES support in 2017 by consumer type and country. First 13 countries alphabetically    49

Figure 33 - RES support in 2017 by consumer type and country. Last 13 countries alphabetically    49

Figure 34 - Difference between household retail electricity prices and electricity wholesale prices 2008-2018, EUR2017/MWh    51

Figure 35 - Difference between industrial retail electricity prices and electricity wholesale prices, EU28 and other G20 countries, 2008-2018, EUR2017/MWh    52

Figure 36 - Selected wholesale gas prices in Europe    55

Figure 37 - The difference between the Platts North West Europe Gas Contract Indicator (GCI) and the Dutch hub price (TTF)    55

Figure 38 - Price formation in Europe    56

Figure 39 - Daily day-ahead prices at selected gas hubs from 2008 to mid-2018    57

Figure 40 - The monthly average price of oil (Brent) and oil-indexed gas contracts (Platts GCI)    59

Figure 41 - Daily spot prices of oil (Brent) and gas (at the Dutch TTF hub)    60

Figure 42 - Daily change of spot prices of oil (Brent) and gas (at the Dutch TTF hub)    60

Figure 43 - The monthly average price of oil (Brent) and gas (at the Dutch TTF hub), measured in EUR/MWh    61

Figure 44 - Monthly average gas price at the Dutch TTF hub and heating degree days in the Netherlands    62

Figure 45 - Daily gas price at the Dutch TTF hub and average daily temperature in the Netherlands from the beginning of 2013 to mid-2018    62

Figure 46 - Comparison of European, US and Japanese wholesale gas prices    63

Figure 47 - The ratio of European, US and Japanese wholesale gas prices    64

Figure 48 - Gas wholesale prices in the EU (weighted average), China, Japan and the US    65

Figure 49 - Gas wholesale prices in the EU (weighted average) and selected markets    66

Figure 50 - Composition of the EU household gas price (DC)    70

Figure 51 - Household gas prices in 2017    70

Figure 52 - Composition of household gas prices in 2017    71

Figure 53 - Composition of EU taxes on household gas prices    72

Figure 54 - Composition of EU prices for small (I3) and large (I5) industrial gas consumers    73

Figure 55 - Median (I3) and large (I5) industrial gas prices in 2017    74

Figure 56 - Composition of median (I3) and large (I5) industrial gas prices in 2017    74

Figure 57 - Composition of EU gas prices for median (I3) and large (I5) consumers    75

Figure 58 - Composition of taxes for median (I3) and large (I5) industrial gas consumers    75

Figure 59 - Difference between household retail natural gas prices and wholesale prices, EU28 and trading partners    76

Figure 60 - Difference between industrial retail natural gas prices and wholesale prices, EU28 and trading partners    77

Figure 61 - The Brent crude oil price from 2008 to mid-2018    79

Figure 62 - Crude oil (Brent) and European wholesale gasoline, diesel and heating oil prices from 2008 to mid-2018    80

Figure 63 - Crack spreads of gasoline, diesel and heating oil from 2008 to mid-2018    81

Figure 64 - Average retail price of oil products in the EU    84

Figure 65 - Average excise duty rates for oil products in the EU    85

Figure 66 - Average retail price of oil products in the EU, without taxes    85

Figure 67 - The retail price of gasoline in the EU    86

Figure 68 - The retail price of gasoline in the EU, without taxes    87

Figure 69 - The exercise duty rate of gasoline in the EU    88

Figure 70 - Average retail price of gasoline in the EU by price component    88

Figure 71 - Average retail price of gasoline in the first half of 2018 by Member State and price component    89

Figure 72 - The retail price of diesel in the EU    90

Figure 73 - The retail price of diesel in the EU, without taxes    90

Figure 74 - The exercise duty rate of diesel in the EU    91

Figure 75 - Average retail price of diesel in the EU by price component    92

Figure 76 - Average retail price of diesel in the first half of 2018 by Member State and price component    92

Figure 77 - The retail price of heating oil in the EU    93

Figure 78 - The retail price of heating oil in the EU, without taxes    94

Figure 79 - The exercise duty rate of heating oil in the EU    95

Figure 80 - Average retail price of heating oil in the EU by price component    95

Figure 81 - Average retail price of heating oil in the first half of 2018 by Member State and price component    96

Figure 82 - Average retail price of gasoline and diesel in the EU    97

Figure 83 - Average retail price of gasoline and diesel in the EU, without taxes    98

Figure 84 - Average excise duty rates for gasoline and diesel in the EU    98

Figure 85 - The difference between the average excise duty rate on gasoline and diesel    99

Figure 86 - Excise duty rates in individual Member States in 2008 and the first half of 2018    100

Figure 87 - the change of the difference between the gasoline and diesel excise duty rates between 2008 and the first half of 2018    101

Figure 88 - Excise duty rates for motor fuels in Belgium    101

Figure 89 - International comparison of retail gasoline prices    102

Figure 90 - International comparison of retail gasoline prices    103

Figure 91 - International comparison of retail diesel prices    103

Figure 92 - International comparison of retail diesel prices    104

Figure 93 - EU import dependency by fuel    107

Figure 94 - EU net imports of energy in 2016 (mtoe)    108

Figure 95 - EU net imports    110

Figure 96 - Comparison of European oil, gas and coal prices    111

Figure 97 - The USD/EUR exchange rate since 2013    112

Figure 98 - The estimated EU import bill    115

Figure 99 - Expenditures on household energy products for the poorest households in the EU Member States), and the share of energy in the total household consumption expenditure    118

Figure 100 - Expenditures on household energy products for the lower-middle income households in the EU Member States, and the share of energy in the total household consumption expenditure    119

Figure 101 - Expenditures on household energy products for households with middle income in the EU Member States, and the share of energy in the total household consumption expenditure    120

Figure 102 - The ratio of homes being not adequately warm for households being below the 60% of the median income and the share of energy products within the total expenditure for households in the third decile (lower-middle income)    121

Figure 103 - Expenditures on transport energy products for the poorest households in the EU Member States, and the share of transport energy in the total household consumption expenditure    122

Figure 104 - Expenditures on transport energy products for households with middle income in the EU Member States, and the share of transport energy in the total household consumption expenditure    123

Figure 105 - Overall competitiveness in EU and non-EU G20    129

Figure 106 - Results for selected favourable factors of competitiveness    133

Figure 107 - Results for selected unfavourable factors of competitiveness    134

Figure 108 - Overall productivity in the EU and G20    135

Figure 109 - Productivity and Efficiency of EU and non-EU G20    136

Figure 110 - Labour productivity in industry in EU and G20    137

Figure 111 - Labor productivity in Services in EU and G20    137

Figure 112 - User satisfaction on Energy infrastructure in EU and G20    138

Figure 113 - Elecriticity prices for industry in the EU and G20 in 2017    139

Figure 114 - Real Unit Energy Cost - manufacturing excluding refining    140

Figure 115 - Contribution to growth of RUEC by Real Prices and Energy Intensity    140

Figure 116 - Evolution of energy costs shares in production value    141

Figure 117 - Evolution of energy costs shars in production value for Manufacturing    142

Figure 118 - Breakdown of the energy consumption per energy carrier, EU, 2008-2015 averages    150

Figure 119 - Average energy cost shares per sector – based on available data points, split by energy carrier, 2008-2015 averages    150

Figure 120 - Energy costs shares in total production costs in manufacturing sectors, 2008-2015    153

Figure 121 - Energy costs shares in total production costs in non-manufacturing sectors    154

Figure 122 - Gross Operating Surplus in manufacturing sectors (average 2008-2015)    156

Figure 123 - Gross Operating Surplus in manufacturing in the EU and Member States, 2008-2015    157

Figure 124 - Energy intensity (consumption/value added in nominal terms) for the most energy intensive manufacturing sectors (average of available countries)    159

Figure 125 - Energy intensity (consumption/value added in nominal terms) for other manufacturing sectors (average of available countries)    160

Figure 126 - Energy intensity (consumption/value added in nominal terms) for non- manufacturing sectors (average of available countries)    161

Figure 127 - Drivers of energy costs (absolute changes)    164

Figure 128 - Drivers of change in energy costs in EU manufacturing sectors over 2010-2015 (%)    165

Figure 129 - Classification of sectors by comparing the dynamics of energy costs dynamics vs productions costs    170

Figure 130 - International comparision of energy costs shares for selected energy intensive sectors    171

Figure 131 - Energy costs shares in production value for the most energy intensive sectors in manufacturing, 2008-2015    172

Figure 132 - Energy costs shares in production value for other manufacturing sectors, 2008-2015    172

Figure 133 - Energy intensity international comparisons for the most energy intensive manufacturing sectors    174

Figure 134 - Energy intensity international comparisons for other manufacturing sectors    175

Figure 135 – Retail electricity prices for industry: EU vs China, Japan & US, 2008-2018    176

Figure 136 - Retail electricity prices for industry: EU vs other G20, 2007-2018    177

Figure 137 – Retail electricity indexes prices for industry: EU vs Argentina, Australia & India, 2008-2018    178

Figure 138 - Range of retail electricity prices for industry in the EU    181

Figure 139 - Box plot of EU28 industrial retail electricity prices 2008-2017    181

Figure 140 - EU28 industrial retail electricity prices 2008-2017, individual Member States lines visible, outliers named    182

Figure 141 - Retail gas prices for industry: EU vs China, Japan and the US, 2008-2018    183

Figure 142 - Retail gas prices for industry: EU vs other non-EU G20 countries, 2008-2018    184

Figure 143 - Retail gas indexes prices for industry: EU vs AU and MX, 2008-2018    185

Figure 144 - Max-min range of retail gas prices for industry in the EU, 2008-2018    188

Figure 145 - Box plot of industrial gas prices, 2008-2017    188

Figure 146 - EU28 industrial retail natural gas prices 2008-2017, individual Member States lines visible, outliers named    189

Figure 147 - Average industry electricity (left) and gas (right) prices in the EU and among the top 15 EU trade partners (current prices)    190

Figure 148 - Impact on EU industry unit costs in a counterfactual scenario where EU energy prices over 2007-2016 are comparable to energy prices faced by the EU’s main trading partners    191

Figure 149 - Impact on EU industry prices in a counterfactual scenario where EU energy prices over 2007-2016 are comparable to energy prices faced by the EU’s main trading partners    191

Figure 150 - Plant electricity consumption and average electricity prices by sector, 2016 (185 observations)    194

Figure 151 - Average electricity prices by consumption level, 2008 – 2017 (2008: 120 observations; 2009: 73; 2010: 141; 2011: 86; 2012: 154; 2013: 161; 2014: 161; 2015: 158; 2016: 185; 2017: 182)    195

Figure 152 - Plant electricity consumption and average electricity costs by sector, 2016 (185 observations)    196

Figure 153 - Plant natural gas consumption and average natural gas prices by sector, 2016 (165 observations)    197

Figure 154 - Average natural gas prices by consumption level, 2008 – 2017 (2008: 110 observations; 2009: 74; 2010: 129; 2011: 81; 2012: 137; 2013: 145; 2014: 149; 2015: 147; 2016: 165; 2017: 163)    198

Figure 155 - Structure of average electricity prices in the surveyed plants in the NWE region (France, Germany, the Netherlands and the UK) in absolute terms (€/MWh), 2008 – 2017    201

Figure 156 - Structure of average electricity prices in the surveyed plants in the SE region (Portugal, Italy, Spain and Greece) in absolute terms (€/MWh), 2008 – 2017    202

Figure 157 - Structure of average electricity prices in the surveyed plants in the CEE region (Poland and Bulgaria) in absolute terms (€/MWh), 2008 – 2017    203

Figure 158 - Structure of average natural gas prices in the surveyed plants in the NWE region (France, Germany, the Netherlands and the UK) in absolute terms (€/MWh), 2008 – 2017    204

Figure 159 - Structure of average natural gas prices in the surveyed plants in the SE region (Portugal, Italy and Spain) in absolute terms (€/MWh), 2008 – 2017    205

Figure 160 - Structure of average natural gas prices in the surveyed plants in the CEE region (Poland and Bulgaria) in absolute terms (€/MWh), 2008 – 2017    206

Figure 161 – Financial support by sector in the EU-28 (expressed in €2017bn)    213

Figure 162 – Financial support by energy group (expressed in €2017bn)    214

Figure 163 – Financial support in 2016 by energy source and EU Member States (expressed in €2017bn and as the percentage of Gross Domestic Product - GDP)    215

Figure 164 – Subsidies in the energy sector, by generation technology    216

Figure 165 – Subsidies to fossil fuels in the energy sector, by generation technology    217

Figure 166 – Support to renewable energy sources in the EU Member States    218

Figure 167 – Nuclear R&D support in the EU Member States    219

Figure 168 – Support to nuclear decommissioning in some EU Member States    220

Figure 169 – Free emission allowances allotted to stationary installations, to the aviation sector, and the evolution of the carbon price in the ETS system    221

Figure 170 – Capacity payment support in some EU Member States    222

Figure 171 – Financial support to interruptible load schemes in some EU Member States    223

Figure 172 – Fossil fuel subsidies in different sectors in the EU    224

Figure 173 – Subsidies to oil and petroleum products in different sectors in the EU    224

Figure 174 – Distribution of support among energy sources in 2016 in the EU    226

Figure 175 – Retail electricity prices, recoverable taxes and tax relieves paid by large industrial (energy intensive) and median level electricity customers in some EU Member States in 2016    227

Figure 176 – Retail gas prices, recoverable taxes and tax relieves paid by large industrial (energy intensive) and median level gas customers in some EU Member States in 2016    228

Figure 177 - Retail electricity prices, recoverable taxes and tax relieves paid by households in some EU Member States in 2016    229

Figure 178 - Retail gas prices, recoverable taxes and tax relieves paid by households in some EU Member States in 2016    229

Figure 179 - The impact of loans and grants for energy efficiency measures and/or other investments on EU28 household and industry energy consumption between 2008 and 2015    230

Figure 180 – Fossil fuel subsidies in the world and in the EU    231

Figure 181 - Energy taxes in the EU-28    233

Figure 182 - Energy taxes as a percentage of tax revenue and of GDP in 2016    234

Figure 183 - Energy taxes by economic activity    234

Figure 184 – Average energy tax for 1 toe of gross inland energy consumption in the EU-28    235

Figure 185 – Average energy tax for 1 toe of gross inland energy consumption in 2016    236

Figure 186 - Excise duty revenues from energy consumption    237

Figure 187 - Exercise duty revenues from energy consumption, adjusted for inflation (in 2015 euros)    238

Figure 188 - The share of excise duty revenues by energy product    239

Figure 189 - The share of excise duty revenues by energy product, 2017    239

Figure 190 - The average standard VAT rate in the EU    240

Figure 191 - Estimated tax revenue from gasoline, diesel and heating oil    242

Figure 192- Ratio between volume traded annually on day-ahead market and electricity final    246

Figure 193 - Evolution of investment costs of renewables    246

Figure 194 - Expected development of investment costs of renewable energy technologies in the long term under different scenarions (lines) and uncertain factors (range): example of solar and wind (left: utility scale photovoltaics, right: offshore wind turbines)    247

Figure 195 - Monthly gas import price    248

Figure 196 - Monthly coal import price    249

Figure 197 - Daily oil price (Brent)    250

Figure 198 - Daily ETS price    250

Figure 199 - Daily ETS price in 2018    251

Figure 200 - Final electricity consumption in the EU28, historical data and projections.    252

Figure 201 - EU 2015 Installed gas capacity by age [MW]    253

Figure 202 - EU 2015 installed coal capacity    254

Figure 203 - EU 2015 installed oil capacity    254

Figure 204 - additions and retirement of gas power plants.    255

Figure 205 - additions and retirement of coal power plants.    255

Figure 206 - additions and retirement of oil power plants.    256

Figure 207 - EU28: electricity prices and cost    257

Figure 208 - New power generation capacity - PRIMES projections.    258

Figure 209 - New power generation investments - PRIMES projections.    258

Figure 210 - Household price regulation in EU Member States in 2016    266

Figure 211 - Share of consumers under regulated prices in EU Member States in 2016    267

Figure 212 - Evolution in the share of households under regulated prices    268

Figure 213 - Share of households under social tariffs in EU Member States    269

Figure 214 - Evolution in the share of households under social tariffs in EU Member States    270

Figure 215 - Number of active suppliers per 100,000 citizens in 2016    272

Figure 216 - Evolution in the number of active suppliers per 100,000 citizens in 2016    274

Figure 217 - Annual switching rates in 2016    275

Figure 218 - Evolution in annual switching rates    276

Figure 219 - Type of electricity and gas contracts available and offers per supplier    278

Figure 220 - Market share of the 3 largest suppliers    279

Figure 221 - Evolution in market concentration    280

Figure 222 - Prices for electricity (2016) and gas (2015) on household markets    281

Figure 223 - Evolution in prices for electricity (2016) and gas (2015) on household markets    282

Figure 224 - Energy component mark-ups    284

Figure 225 - Evolution in energy component mark-ups    285

Figure 226 - Energy expenditure as share of disposable income    286

Figure 227 - Energy poverty    287

Figure 228 - Evolution of energy poverty    288

Figure 229 - Consumer perception of the market in 2015    290

Figure 230 - Non-household price regulation from a geographical perspective    291

Figure 231 - Share of non-household consumption under regulated prices    292

Figure 232 - Energy and supply component for electricity (2016) and gas (2015) on the non-household consumer market    293

Figure 233 - Energy and supply component for electricity (2016) and gas (2015) on the non-household consumer market    294

Figure 234 - Energy and supply component for electricity (2016) and gas (2015) on the non-household consumer market    295

Figure 235 - Potential savings for households from switching to dynamic pricing in selected Member States ; by consumption band    302

Figure 236 - Elements to understand consumer behaviour and agregation for analysis    305

Figure 237 - Potential savings from switching to DPC with alternative mark-ups in Italy    307



List of Tables

Table 1 - Key figures on the evolution and drivers of retail electricity prices    26

Table 2 - Composition of small (IB) and large (IF) industrial prices    44

Table 3 - The ratio of European, US and Japanese wholesale gas prices    64

Table 4 - Key figures on the evolution and drivers of retail gas prices    68

Table 5 - Estimated average gas import prices by supplier (€/MWh)    109

Table 6 - EU crude oil import bill in 2013-2017    113

Table 7 - EU gas import bill in 2013-2017    113

Table 8 - EU hard coal import bill in 2013-2017    114

Table 9 - Coverage of manufacturing sectors    146

Table 10 - Coverage of other sectors, excluding manufacturing    147

Table 11 - Energy costs shares in total production costs for manufacturing and non-manufacturing, 2008-2015    151

Table 12- Drivers of total production costs in manufacturing sectors    167

Table 13 - Changes in energy costs and total production costs by sector    168

Table 14 - Changes in retail industrial electricity prices compared to EU prices, constant 2017 EUR/MWh    179

Table 15 - Factors in observed industrial retail electricity price changes per country, nominal prices, per MWh    180

Table 16 - Changes in the industry retail natural gas price differential compared to EU prices, constant 2017 euros per MWh    186

Table 17 - Factors in observed industrial retail natural gas price changes per country, nominal prices, per MWh    187

Table 18 - Size and EU representativeness of the samples in the bottom-up analysis    193

Table 19 - Key plant level results from the case studies (2016) – EU values    207

Table 20 - Overview of tariff deficits in Member States between 2008-2016    296

Table 21 - Composition of user group for estimating the DPC    303

Table 22 - Estimated average annual savings per Band, EUR, 2016    303

Table 23 - Estimated average annual savings per Band, share of energy supply component, 2016    304

Table 24 - Estimated number of consumers losing out from the switch to DPC    304



Introduction

The European's Commission Clean energy for all Europeans package adopted in November 2016 1 included capital energy policy initiatives to roll out the Energy Union and set strengthened foundations to meet the EU's climate and energy targets and its international commitment under the Paris Agreement. The (previous) 2016 energy prices and costs report was part of that package and provided evidence assessing and underpinning the need for various European energy policy actions.

Two years after, most of the initiatives of that package have already been agreed between Parliament and Council and will start to be implemented soon. In November 2018, the European Commission has just presented the Long-term EU strategy for the reduction of greenhouse gas emissions in accordance with the Paris Agreement 2 . In a similar vein as in 2016, the 2018 edition of the energy prices and costs report comes at a timely moment and contributes to assess energy policies on the basis of the new evidence coming from the analysis of energy costs, prices and support interventions. The evidence of the report highlights the important role of international fossil fuel prices in driving energy prices in the EU making the case for pursuing our efforts to decarbonise our energy system. Data also shows the impact of dollar-denominated international energy prices on our energy bill, underpinning our efforts to reduce dependence on fossil fuels and highlighting the benefits of pricing the transactions of energy products in euros to reduce the uncertainty brought by exchange rate volatility 3 .

The first part of the report (Part I – Energy Prices, comprising Chapters 1, 2 and 3) looks at the developments on wholesale and retail energy prices for electricity, gas and oil products between 2008 and 2017-18. On retail prices, the European Commission has again conducted an ad hoc data collection of the cost elements making up retail prices. The report presents the most detailed available breakdown of these elements affecting prices, in particular the various taxes and levies, and provides an insight on the evolution, composition and drivers of retail prices. International comparisons of the prices for petroleum, gas and electricity products are also presented in the chapters of Part I.

The impact of the energy costs on the economy, the industry and households is addressed in the second part of the report (Part II – Energy costs). Chapter 4 on the import bill assesses the developments in the EU energy bill and the reasons behind them. Chapter 5 looks at impact of energy prices and costs on energy poverty including an assessment of the energy expenditure shares by income level. Finally, Chapter 6 analyses the impact of energy costs on the (cost) -competitiveness of the European industry. The analysis includes an assessment of the costs for whole industry and services sectors and for 45 specific manufacturing and non-manufacturing sectors, including energy intensive industries. Their energy costs shares, energy intensities and energy prices are examined and, to the extent possible by the available data, compared with those in the third countries. The chapter presents a combination of results coming from studies using highly aggregated statistical data and sectorial data collected at plan level. Part II is complemented by Annex 1, which presents case studies for a number of sectors/subsectors of energy intensive industries, and by Annex 2, which includes sectorial decomposition analyses of energy costs.

Government interventions and revenues from energy products are addressed in Part III of the report. Chapter 7 presents the result of the analysis of the data collected on support interventions across member states (more than 1400 types of interventions) from 2008-2017. This constitutes the most updated and detailed inventory of energy subsidies of the EU Member States. The impact of these subsidies on the prices and costs for consumers and industry, particularly energy intensive sectors, is also estimated using a combination of analytical and econometric techniques (that distinguish the impact on prices at wholesale and retail level). Chapter 8 also assesses the nature and importance of energy tax revenues in government's budgets.

Part IV of the report (Chapter 9), looks forward, and assesses the relation between future expected prices and costs in the electricity market and how this can affect the incentives for investment in the different energy technologies in the 2030 horizon. It analyses the various underlying factors driving the price-cost relation together with expected developments on future demand which is significantly influenced by our policy decisions.

Finally, Part V looks at the impact of regulation on prices. Chapter 10 assesses in detail the impact of price regulation on the existence of competitive prices, quality of service and the propensity to invest between 2008 and 2016. It looks at how price regulation/de-regulation affects Member States by analysing indicators on various groups of Member States, those which with have been fully liberalised prices (prior or during the period of study) versus those which are in transition to price deregulation (less than 50% of the market is with price regulation) and those which still have significant part of their market (more than 50%) with price regulation. Finally, on the basis of data collected from Member States, section 10.2 of the chapter includes estimates of the benefits from switching from regulated prices to fully liberalised dynamic price contracts.

Country-factsheets with energy prices and costs key indicators are included in Annex 3.


PART I

ENERGY PRICES



1Electricity prices

1.1Wholesale electricity prices

¾Over the last ten years wholesale electricity prices showed a significant volatility in Europe, reaching their peak in 2008 and their lows in early 2016. Since spring 2016, when the European Power Benchmark (EPB) index, in parallel with many wholesale electricity market prices, fell to ten year lows and reached 30 €/MWh, there was a general price recovery and by August 2018 the EPB went up to 60€/MWh.

¾Price convergence across the European regional wholesale markets varies over time. Wholesale electricity prices tended to diverge more in general price spike periods (e.g.: in 2008 or in early 2017), or when low prices could be observed in some local markets (e.g.: periods of abundant hydro power generation in the Nordics).

¾Different regional markets in Europe face different prices, owing to differences in local generation mixes and interconnection capacities.

¾EU and national level policies resulted in increasing wind and solar generation in many European countries. Integration of renewable generations sources to the grid and investments in infrastructure capacities are indispensable to accomplish a properly working internal electricity market in the EU. Market couplings now cover the majority of EU wholesale markets, enabling more efficient cross-border trade and better convergence of prices between neighbouring markets.

¾Several factors impacting wholesale electricity prices can be identified both on the demand side and on the supply side of the electricity market. Since 2008 GDP in the EU-28 was up by 10%, while at the same time electricity consumption decreased by 6%, pointing to less electricity intensity of the EU economy. On the supply side there were fundamental changes in the EU electricity generation mix since 2008: the share of fossil fuels decreased while the share of renewable energies practically doubled, reaching an estimated 38% in April 2018.

¾The price of coal and natural gas has an important role in shaping electricity generation costs, as coal-fired and gas-fired generation ensures the marginal generation costs on the electricity market. Coal and gas prices peaked in 2008; then fell sharply in 2009. After a recovery, between 2012 and 2016 they decreased again, whereas in the last two year they rebounded. After being at low levels for several years, emission allowance prices tripled between mid-2017 and mid-2018, starting to impact again wholesale electricity prices. Wholesale electricity prices showed a strong co-movement with fossil fuel prices.

¾In international comparison, which is useful for analysing the competitiveness of electricity intensive industries, wholesale electricity prices in the EU were higher than in the US, Canada, Australia and Russia during most of the time in the last ten years, however, they were lower than prices in China, Japan, Mexico, Turkey, Brazil and Indonesia. Electricity price differentials between the EU and its trading partners do not give in all cases a comprehensive explanation for competitiveness of products manufactured by energy intensive industries, other factors of competitiveness need to be taken into consideration.

1.1.1 Evolution of wholesale electricity prices

Over the last decade, wholesale electricity prices in the European markets showed significant volatility. In 2008, as all energy commodity prices reached their top, wholesale electricity prices also peaked. Subsequently during 2009, in parallel with the economic crisis the wholesale electricity prices fell back. Between 2010 and 2012 a solid price recovery took place, however, in the following four years a downward trend could be witnessed, and in early 2016 wholesale electricity prices fell to a decade low in many European markets. Compared to these lows, wholesale electricity prices increased until mid-2018 again.

The next chart ( Figure 1 ) shows the evolution of the European Power Benchmark (EPB) and the range of minimum and maximum wholesale electricity prices in each month between 2008 and 2018. EPB is an index computed as the weighted average of the day-ahead prices of the most liquid wholesale electricity markets, serving as a general European benchmark,. In different periods convergence across European markets reached a different degree; in the periods of general price spikes (e.g.: in 2008 or in early 2017) prices became more divergent, however, very low prices in some national markets (e.g.: Nordic markets during summertime, a high level of hydro power generation, or markets with high share of variable renewable penetration) can also lead to significant price differentials across Europe.

Figure 1 - Evolution of monthly average wholesale day-ahead baseload electricity prices in Europe, showing the European Power Benchmark and the range of minimum and maximum prices across the markets

Source: Platts, European power markets

Zooming on the market developments of the last two years (since the publication of the 2016 edition of the Energy Prices and Costs in the EU report), , we can see that in the second half of 2016 wholesale electricity prices started to increase, owing to higher electricity generation costs in parallel with increasing coal and gas prices (see Figure 2 ). This increase was reinforced at the beginning of 2017 as an ongoing cold weather, coupled with lower than usual nuclear electricity generation in some Western European countries (mainly France and Belgium), resulted in high wholesale electricity prices. At the end of 2017 safety inspections also reduced the availability of nuclear fleet in the same markets. All-in-all, since spring 2016, when the EPB index was around 30 €/MWh, a measurable wholesale price increase could be observed by August 2018, as the benchmark was above 60 €/MWh in that month.

Figure 2 - Evolution of wholesale electricity prices in Europe since 2016

Source: Platts, European power markets

As it was already mentioned, there might be significant differentials in wholesale electricity prices across Europe in some periods, thus it is worth taking a look at the price evolution of the regional markets.

Figure 3 shows the regional wholesale electricity prices in the North Western Europe (NWE) market coupling area, including Central Western Europe (Germany, France, Austria and the Benelux), the UK, the Nordic markets (Norway, Sweden, Denmark, Finland and the Baltic States) and Iberian market (Spain and Portugal). Nordic markets had normally the lowest wholesale price across Europe over the last two years, owing to the important role of hydro power generation and increasing of wind and solar. In Central Western Europe (CWE), where prices were also lower than the EPB during most of this time period, increasing renewable penetration and the important role of nuclear power resulted in competitive electricity generation costs.

In the Iberian region prices significantly impacted by hydro reserves and generation; from mid-2016 to the beginning of 2018 a dry period resulted in lower than usual hydro power generation that increased the wholesale price level above the EPB. During most of the time in 2016-2018 the highest wholesale prices could be observed in the UK; owing to significant electricity import needs of the country and the climate change levy that directly impacts the UK wholesale electricity price.

Although the NWE region is the largest flow based market coupled area in Europe, significant differences in wholesale prices exists across its different regions. This suggests further integration of all electricity generation sources (especially variable renewables like wind and solar) and further investment in infrastructure (e.g.: interconnection capacities) could help to diminishes cross-market price differentials.

Figure 3 - Regional market prices in the North-Western Europe coupled area

Source: Platts, European power markets

In the Central and Eastern Europe region (CEE – Poland, Czech Republic, Slovakia, Hungary, Romania, Croatia and Slovenia) prices behaved similarly to the European benchmark between 2016 to (August) 2018 (see Figure 4 ). Besides abundant fossil fuel (mainly coal and lignite) and nuclear power generation in the region as a whole, market prices in the Central Eastern Europe are impacted by electricity imports from Central and Western Europe and the Balkans, where hydro generation is important. At the end of 2016 and 2017, when nuclear availability in the CWE region was low due to the aforementioned reasons, prices in the CEE region were lower than in Western Europe.

In the CEE region four national markets (Czech Republic, Slovakia, Hungary and Romania) are coupled and wholesale electricity prices are well aligned in the majority of trading hours. Poland is price coupled with Sweden (and thus with the NWE region). Croatia and Slovenia are not coupled with the rest of the CEE region.

Figure 4 - The Central Eastern Europe average wholesale price and the EPB benchmark

Source: Platts, European power markets

In Italy and Greece wholesale electricity prices between 2016 and 2018 so far were usually higher than the EPB benchmark ( Figure 5 ). Italy has traditionally been a net electricity importer, as the cost of import (mainly from the CWE region) is competitive to the domestic, primarily gas-fired power generation. During the summer period renewable generation picks up in Italy and imports are lower. Greece is also traditionally net electricity importer, the country's domestic production is largely based on domestic lignite and gas fired generation. Imports from the Balkans are competitive against domestic generation.

Bulgarian wholesale prices were however lower than European prices between 2016 and 2018 so far, owing to favourable costs of domestic electricity generation (largely based on solid fuels and nuclear). Bulgaria is normally a net electricity exporter, but in some periods (e.g.: the cold snap in January 2017) exports were banned 4

Figure 5 - Regional market prices in Italy and South Eastern Europe

Source: Platts, European power markets

1.1.2 Factors impacting the evolution of wholesale prices 

Wholesale electricity prices are determined by market forces. In this section we look at demand side and supply side factors that explain their evolution.

On the demand side of the electricity market, residential electricity consumption is determined by needs for various purposes, for example lighting, heating and using household appliances. For businesses, the consumption of electricity is mainly determined by the level of economic activity, which can be measured by the evolution of the Gross Domestic Product (GDP).

The next chart ( Figure 6 ) shows that electricity consumption in the EU has strong seasonality (see blue dashed line in Figure 6). During wintertime the industrial activity and the lighting and heating needs of households are higher than in summertime, all this resulting in higher electricity consumption during the winter period.

In order to assess the relation between electricity consumption and economic activity we need to compare their trends. The seasonality of electricity consumption can be mostly eliminated by applying a four-quarter moving average (red line), which can then be compared to the evolution of the overall economic activity. In the second quarter of 2018 GDP in the EU was up by more than 10% compared to 2008; (after the crisis in 2008-2009 the recovery took several years) at the same time electricity consumption was down by 6%. This decoupling of the GDP trend from the electricity consumption shows a decreasing electricity intensity of economic activity over the last ten years.

Figure 6 - Electricity consumption and economic growth

Source: Eurostat

On the supply side of the wholesale electricity market, it is the electricity generation mix, the marginal costs of the generation technologies and the electricity imports, as competing alternative that determine the overall generation costs and electricity prices. The next chart ( Figure 7 ) shows the changes in the shares of the generation technologies in the EU electricity mix between 2008 and the first quarter of 2018. The share of fossil fuel generation (lignite, coal, gas and oil) decreased significantly (from 54% in 2008 to 37% in 2017 and to 34% in January-August 2018). At the same time the share of renewables (including wind, solar, hydro and biomass) increased from 17% to 33%. The share of nuclear remained practically constant, showing minor changes from one year to another.

Figure 7 - Electricity generation mix in the EU-28 (actual power generation)

Source: Eurostat and ENTSO-E. *2017 and 2018 Jan-Aug data are not fully comparable with earlier periods, as a part of biomass generation seems to be reported under 'Other'

Within renewables, the share of hydro power remained constant over time (although the hydro share can vary on the short term depending on weather conditions, namely the amount of precipitation).

The increase in the share of renewables within the EU electricity generation mix was largely owing to wind power, whose share went up from 3.5% to 12% between 2008 and 2017 (in some periods, for example in December 2017 the share of wind was almost 17%).

The amount of generated electricity from the main sources and their share in the total generation can also be seen in Figure 8 , Nuclear energy had an important but slightly decreasing share over the last two-three years in the EU. Its variability is small and mainly due to maintenance cycles. Both wind and solar power had an intra-annual seasonality in the EU over the last few years: while the share of solar increased during summertime, the share of wind generation was the highest in the stormy wintry period. In April 2018 the combined share of wind and solar was 19% in the EU electricity generation mix. Biomass had a relatively stable share over time. Hydro availability strongly depended on weather and the amount of precipitation. Both coal and gas had higher share in winter periods, however, coal, as baseload technology had a smaller seasonality than natural gas.

Figure 8 - Monthly electricity generation in the EU and the share of some generation sources in the EU electricity mix

Source: ENTSO-E

Besides the generation mix the marginal cost of each generation technology impacts the supply of electricity on the wholesale market. Generation technologies like wind, solar, hydro have very low or negligible marginal generation costs. Nuclear and biomass also have low marginal costs, whereas coal, gas and oil fired generation have higher marginal generation costs, so these latter normally set the marginal cost on the market; they are on the higher end of the electricity supply curve (the so-called merit order curve). Variable renewables (wind, solar) and other low marginal cost technologies can impact the total electricity supply by shifting the merit order curve towards the right (more generation at the same cost level) and thus result in lower equilibrium price assuming the same electricity demand curve.

Figure 9 shows the monthly coal, gas and emission allowance (carbon) prices, compared to the average of 2008, and the share of renewable sources in the EU electricity generation mix. Both coal and gas prices, after reaching a peak in 2008, fell back amid of the economic crisis in 2009, and recovered in 2010-2011. After 2012 both coal and gas prices started to decrease and in the first half of 2016 they reached the lowest since the crisis year of 2009. In 2016-2018 they rebounded from these low levels. If looking at the evolution of wholesale electricity prices (see Figure 1 ) there is a high degree of correlation between fossil fuel prices and the wholesale electricity market. However, increasing share of renewables (between 2008 and 2018 the share of renewables practically doubled in the EU generation mix) had a downward impact on the wholesale market. A recent study estimates conducted by the EC (Trinomics et altri, 2018) that one percentage point increase in the share of renewables in Germany results in a decrease of the wholesale electricity price by 0.5 €/MWh.

Carbon prices showed a sharp decrease between 2008 and spring 2013, as they went down by 80% (in June 2008 the average carbon price peaked at 28 €/MtCO2e, while in April 2013 it was below 4 €/MtCO2e). Since then until mid-2017 they showed only minor variations, not being able to significantly impact competition between coal, gas and renewables, and not being effective in driving investments towards the decarbonisation of the European electricity sector. However, as market players anticipated that regulatory changes in the so-called Market Stability Reserve of the EU Emission Trading System might reduce the oversupply of allowances in the carbon market – a measure to enter into force as of January 2019 – the price of emission allowances doubled between mid-2017 and mid-2018, reaching 21 €/MtCO2e 5 by the end of August 2018, which was the highest since October 2008.

Figure 9 - Monthly coal, natural gas and carbon price indexes, compared to the 2008 average price and the share of renewable energy (right hand scale)

Source: ENTSO-E

National and EU level policies played a role on incentivising the increase of the share of variable renewable technologies. EU Member States have to fulfil their 2020 renewable objectives, for which different instruments, such as feed-in-tariffs, feed-in-premiums, renewable quota obligations, etc. have been implemented. Increasing renewable capacities and less reliance on fossil fuels, in parallel with moderate growth in demand for power, have led to overall overcapacities in many European electricity markets, resulting in a downward pressure on wholesale electricity prices.

Furthermore, as the integration of the European wholesale electricity markets moves forward, the importance of cross-border electricity trade increases and market-based instruments, such as the aforementioned coupling of neighbouring markets have been implemented, contributing to more efficient cross-border electricity trade, and generally better convergence of wholesale electricity prices. EU policies aim at achieving a better market integration through improving market liquidity, cross border trade and better integration of variable renewable generation sources to the power grid.

At EU level electricity imports do not have significant influence on wholesale market prices as extra-EU electricity imports are negligible compared to the bloc's total consumption. However, looking at individual power regions (see Figure 10 ) the situation is different, as some regions (e.g.: the Baltic states and Italy) might have significant import needs on the top of their domestic production to satisfy all consumption needs. Other regions, such as Central and Western Europe, the Nordic region or South Eastern Europe recently, produce more electricity than their domestic needs, therefore they are net exporter. As electricity normally flows from low priced areas to higher priced ones, net exporter regions have lower wholesale prices as net importers.

Figure 10 - Net electricity flow position compared to domestic electricity generation in the European power regions

Source: ENTSO-E

1.1.3International comparisons

Comparing the European electricity benchmark wholesale price index (EBP) with the wholesale prices in the most important trading partners of the EU can provide a useful analysis on how energy costs differentials can impact the competitiveness of European energy intensive industries with a high trade exposure. Electricity costs are only one factor of international competitiveness and other aspects (such as business environment, labour costs, etc.) are also important. A more detailed analysis of the impact of prices on competitiveness can be found in chapter 6.

Figure 11  shows that, since 2008, wholesale electricity prices in the US have been for most of the time lower than in the EU, with the EU-US price ratio reaching the magnitude of 2 sometimes. In contrast, prices in Japan showed a huge increase after the Fukushima nuclear incident in March 2011, and as nuclear capacities were put offline, and the increasing reliance on gas fired generation resulted in Japanese prices being 3-4 times as the EU average between 2012 and 2014. Since 2016, as nuclear capacities were gradually put back in operation, the wholesale price gap between Japan and the EU decreased.

In China the wholesale electricity prices have been constantly higher by several magnitudes (2-3) than in the EU, implying that competitiveness problems of some energy intensive industries (e.g.: steel sector) vis-à-vis China do not actually stem from electricity prices.

Figure 12 shows some further examples on wholesale prices of important EU trade partners. In Canada wholesale prices were one of the lowest over the last ten years among countries presented below. In Australia, based on competitive domestic coal fired generation, wholesale prices were also lower than in the EU (however in the summer period in 2017 there were some price spikes). In Russia wholesale electricity prices were also lower than in the EU.

On the other hand, prices in Mexico and Turkey, albeit following a decreasing trend, were higher than the European benchmark. Prices in Indonesia were comparable with the EPB between 2009 and 2013, however, the price gap with Europe has widened since and at the end of 2017 local wholesale electricity prices were twice as high as in Europe. Among all analysed countries wholesale electricity prices were the highest in Brazil.

Figure 11 - Comparison of wholesale electricity prices in the EU with global trade partners

Source: Trinomics et atri study (2018)

Figure 12 - Comparison of wholesale electricity prices in the EU with global trade partners

Source: Trinomics et altri study (2018)



1.2Retail Electricity Prices

Main findings

The last two years brought major departures from decade long trends in the evolution, composition and drivers of electricity prices:

¾The EU household price decreased for the first time. This means that from 2016 to 2017 prices fell for all consumer analysed types.

¾The decade long trend of taxes and network charges driving household prices up also came to an end. Taxes kept increasing for all industrial consumer types, albeit these increases were smaller than the decreases in energy components, leading to falling prices.

¾Progress towards the completion of the single energy market continued. This is reflected by the fact that national energy components are gradually converging: they became 15% less spread out since 2008. For industrial consumers even total prices converged by 8%, as such prices are less impacted by varying national taxation. 6

¾The energy component, which consists mostly of wholesale prices, remained on a steadily decreasing trajectory due to EU policies, such as market coupling and increased interconnection capacities. The energy component diminished both in absolute and relative terms. In other words the only part of the price which is set by market forces is contracting while the share of the regulated part is growing, reaching 40% EU- wide.

¾Wholesale prices have been increasing since the spring of 2016. This development has not yet factored in to the energy component of retail prices. Similarly, decreasing wholesale prices in the period 2012 to 2016 were not fully passed on to retail prices.

¾Electricity prices remained heavily impacted by policy support costs and fiscal instruments, albeit to a varying degree across Member States.

¾The cost of supporting renewable energy also started to fall for households after a decade of steep increases. This is remarkable as the share of renewables in the EU's generation mix kept growing. RES support costs decreased in the last reporting year by 1% for households, but increased by 7% for medium industrial and by 17% for large industrial consumers. RES levies ranged from 1 to 73 EUR/MWh across reporting countries.

¾Excise duty rates range from 5 to 122 EUR/MWh, while VAT rates spread from 6% to 27% displaying a highly differentiated picture of energy taxation across the EU.

¾The EU household electricity price grew annually by 2.1% since 2008 and reached 195 7 EUR/MWh in 2017. The EU medium industrial electricity price grew at the annual rate of 1% and averaged at 103 EUR/MWh in 2017. The EU price for large industrial consumers even experienced a decrease of 0.3% annually and reached 80 EUR/MWh by 2017, down from 83 EUR/MWh in 2008.

Methodological framework

Table 1 - Key figures on the evolution and drivers of retail electricity prices

Consumer type

Household (DC)

Industrial (ID)

Large Industrial (IF)

Component

Annual growth

Share 2017

∆ Share

Annual growth

Share 2017

∆ Share

Annual growth

Share 2017

∆ Share

Energy

-1.5%

33%

- 12 p.p.

- 4.7 %

40%

- 28 p.p.

- 4.8%

49%

-25 p.p.

Network

+2.5%

27%

+ 1 p.p.

+ 3.5%

22%

+ 5 p.p.

+ 2.7%

17%

+ 4 p.p.

Taxes

+ 6.1%

40%

+ 11 p.p.

+12.3%

38%

+ 23 p.p.

+ 11%

34%

+21 p.p.

Total

2.10%

 

 

+ 1%

 

 

- 0.3%

 

 

Source: DG ENER in-house data collection

Aim and scope of the chapter

The following chapter analyses retail electricity prices. It takes an in- depth look at the evolution, composition and drivers of prices paid by final consumers on EU as well as on national level in 30 European countries from 2008 to 2017.

The analysis serves as an objective, evidence based tool to determine how the composition of retail prices changed over time, how did various policies and fiscal instruments impact prices and which elements contribute the most to increasing or decreasing prices. The data collection designed and conducted specifically for the purpose of this report, introduces a high level of harmonization and transparency. This allows for the comparison of price developments over time and across countries.

A Decade of Data

The chapter is based on an in-house data collection by the Directorate General for Energy of the European Commission (DG Energy). Data for this in-house survey was provided by the competent authority of each reporting country, in most cases the statistical office, ministry or regulator representing the country in the European Statistical System. Data was provided by 26 EU Member States and 3 non-member countries, Montenegro, Norway and Turkey.

Greece and the United Kingdom provided no data. Figures for these countries are substituted by Commission estimates.

Structure along Eurostat legislation

The chapter is structured along different consumer types of the two energy products. Consumer types are defined by Eurostat methodology under Regulation (EU) 2016/1952 of the European Parliament and of the Council of 26 October 2016 on European statistics on natural gas and electricity prices. It differentiates household and industrial consumers 8 , whereas both consumer types are further broken down into consumption bands. Consumption bands cover different volume ranges of annual consumption. Different bands are applied to electricity and natural gas.

The chapter commences by examining the most representative household electricity price on EU level and in each reporting country. Next, the chapter looks at electricity prices paid by industrial consumers. It differentiates between 3 levels of industrial consumption in order to provide the best possible picture of a diverse group of consumers, ranging from small businesses to manufacturing industries consuming large amounts of energy. The chapter first examines prices paid by industrial consumers with a median volume of consumption. This consumer band is often used to describe economy- wide industrial price developments. The analysis is completed by a comparison of prices paid by consumers of small versus large volumes of electricity.

General observations that hold for all bands are presented only under the section of household prices and are not repeated in each section.

 

Harmonized methodology for comparable results

Total prices provide no information on the drivers of price developments. To facilitate a more focussed identification of price increase drivers, total prices are further decomposed into three main components. The components Energy, Network and Taxes disaggregate the total price along the value chain. DG Energy further disaggregates taxes into 10 sub-components. These were designed to showcase national characteristics based on harmonized sub-components to the fullest extent possible, while minimizing the number of elements designated as "other". The same components and sub-components are applied to both electricity and natural gas.

DG Energy provided extensive guidance to the reporting national authorities to ensure that elements are assigned to components and sub-components in a harmonized manner across all countries.

All EU figures are weighted averages of EU Member States only. It is to be noted that the number of countries included in each EU average changes according to energy product and consumption band. For example, there are no consumers in the largest electricity bands.



Data cooperation

DG Energy would like to thank for the cooperation of national authorities from 29 countries who provided essential data for the retail prices sections of the report:

Country

Organization

Expert

Austria

E- Control

Esther Steiner

Belgium

Belgian Ministry of Economic Affairs

Marc Vos

Bulgaria

National Statistical Institute of Bulgaria

Iveta Minkova

Ivanka Tzvetkova

Cyprus

Electricity Authority of Cyprus

Marios Skordellis

Czech Rep.

Czech Statistical Office

 

Germany

BDEW Bundesverband der Energie- und Wasserwirtschaft e.V.

Christian Bantle

Denmark

Danish Energy Agency

Ali A. Zarnaghi

Estonia

Statistics Estonia

 

Spain

Ministerio para la Transición Ecológica

 

Finland

Statistics Finland

Marianne Rautelin

France

Ministère de la Transition écologique et solidaire

Pascal Levy

Croatia

Croatian Bureau of Statistics

Mirjana Petanjek

Željka O. Kelebuh

Hungary

Hungarian Energy and Public Utility Regulatory Authority

 

Ireland

Sustainable Energy Authority of Ireland

Mary Holland

Martin Howley

Italy

Italian Regulatory Authority for Energy, Networks and Environment

Gabriella Antonel

Lithuania

Statistics Lithuania

Virginija Jasionienė

Luxembourg

Institut national de la statistique et des études économiques du Grand-Duché de Luxembourg

Olivier Thunus

Latvia

Central Statistical Bureau of Latvia

Anna Paturska

Malta

National Statistics Office of Malta

Ronald Tanti

Netherlands

Statistics Netherlands

Eva Witteman

Poland

Polish Energy Market Agency

Krzysztof Dziedzina

Portugal

Direccao Geral de Energia e Geologia

Elisa Oliveira

Romania

Romanian National Institute of Statistics

Michaela Chirculescu

Sweden

Statistics Sweden

Viktor Ahlberg

Slovenia

Statistical Office of the Republic of Slovenia

Marko Pavlič

Slovakia

Statistical Office of the Slovak Republic

 

Norway

Statistics Norway

Thomas Aanensen

Montenegro

Statistical Office of Montenegro

Suzana Gojcaj

Turkey

Turkish Statistical Institute

Mehmet Gedik

1.2.1Household Electricity Prices

The following section analyses prices paid by household electricity consumers. It examines weighted EU averages and the most representative band in each country. "Most representative" is defined as the consumption band accounting for the largest share in total household consumption, in other words the price for which the most electricity was sold. It is irrespective of the number of consumers in the band. Due to data availability restrictions, weighted EU averages are built uniformly from the consumption band defined by Eurostat terminology as DC, covering annual consumption of 2500 to 5000 kWh. For 2017 a "Most representative" weighted average is also presented.

Evolution of household electricity prices

Total prices grew at 2% annual rate from 2008 to 2017. In absolute terms the EU price grew from 166 to 200 EUR/MWh in the same period. When considering the most representative band in each country, instead of uniformly considering the consumption band DC, the EU average falls slightly lower, at 196 EUR/MWh. Prices grew faster than inflation, which averaged at 1.2% annually during the same period.

In 2017 the EU price fell for the first time by 3%. This is a significant departure from almost a decade of continuous increases. The decreasing EU average is however to be interpreted with caution. From 2016 to 2017 prices of the most representative bands actually increased in 13 reporting countries. Prior to 2016 the direction of developments on EU level, were mostly the same as the direction of developments in the majority of reporting countries. Since 2016, we can observe more divergent developments.

Figure 13 - Evolution and composition of the EU household price (DC). MR = Most representative

Source: DG ENER in-house data collection

Composition of household electricity prices

Over time the composition of prices changed significantly. The share of the energy component in the total price decreased by 13 percentage points from 46% to 33% in 2017. In the beginning of the observation period, the energy component was the largest of the three components in all reporting countries.

In absolute terms, the energy component decreased on average at an annual rate of 1.5% and reached 67 EUR/MWh in 2017. The contraction of the energy component can be linked to EU energy policies: increased competition resulting from market coupling and the growth of power generation capacity with low operating costs, such as wind and solar power, in addition to existing nuclear and hydro power. On national level 11 Member States reported actually higher energy components in 2017 than in 2016. In these countries wholesale prices either increased or their fall has not translated into a reduction of the energy component. Such results may imply that price competition in a number of retail markets is weak, allowing suppliers to avoid passing on wholesale price reductions to final consumers.

The share of the network component remained almost constant at about quarter of the price from 2008 to 2017. In absolute terms the network component grew at the annual rate of 2.5% and reached 54 EUR/MWh in 2017.

The share of the taxes component grew by 12 percentage points. It accounted for 28% of the weighted average EU price in 2008 and for 40% in 2017, meaning that it was the largest of the three components. In absolute terms, taxes grew at the annual rate of 6% and reached almost 80 EUR/MWh in 2017. The section "Composition of taxes, levies, fees and charges" analyses in detail which specific policies and fiscal instruments were driving this increase.

Box – Definition of most representative band

Household electricity consumption is broken down into 5 bands. The most representative band is defined as the one of these five bands with the highest share in total consumption. In other words, the price for which the most electricity was sold.

The 2016 edition of the Energy Prices and Costs series as well as all regular Eurostat press releases uniformly analyse consumption band DC for each country. In many of our reporting countries however only a smaller portion of consumption falls into DC. Household consumption varies highly across countries. It is determined by several factors including household size, climatic conditions (availability of sunlight and consequent lighting needs, heating and cooling needs), GDP (number and size of electric appliances on one hand and the efficiency of these appliance son the other hand), and prevalence of electrification in the transport and heating sectors. In northern countries consumption is typically above DC while in southern countries typically below.

To analyse prices that are truly representative, DG Energy introduced the possibility of reporting the price of the most representative band in each country. This feature of the in-house data collection serves the purpose of better catering to needs of Member States. One third of the countries made use of this possibility. It is important to note that if a country did not provide data for the most representative band, it was automatically assumed that DC is such. Typical consumption falls in the following ranges in the reporting countries.

Source: DG ENER in-house data collection

Drivers of Household Electricity Prices

2017 brought a significant departure from the decade long trend of increasing prices: the EU household electricity price fell for the first time. Between 2008 and 2016 the price was steadily increasing, driven by the combined impact of network charges and taxes – both components steadily growing until 2016. At the same time, smaller decreases in the initially large energy component slightly moderated the growth of the total price. Since 2017 both network charges and taxes have been decreasing on average, in addition to the energy component that has been contracting throughout the whole period. Decreases in all three components led to an overall price reduction.

Figure 14 - Household rices in 2017 (most rep.)

Source: DG ENER in-house data collection

In 2017 Germany reported the highest price of 305 EUR/MWh, overtaking Denmark with a price of 289 EUR/MWh. Denmark reported the highest price from 2008 to 2017. Bulgaria reported the lowest price of 97 EUR/MWh among all EU and non- EU countries. The ratio of the largest to smallest price across the EU decreased by 4% over the last decade, indicating progress towards the completion of the internal energy market. In 2017 the largest price was 3.1 times of the smallest.

Denmark and Germany reported the highest tax components of almost 200 and 166 EUR/MWh respectively. In Germany this tax component consists of a number of elements. Support to renewable energy is the largest of those elements (45%), followed by excise duties and concession fees (each 13%). In Denmark the tax component consists mostly of excise duty (61%) and to a much lesser extent of support to renewable energy (10%). The three countries (BE, DE, DK) that reported the largest total prices also reported the highest tax components, indicating a strong correlation between overall price levels and taxation.

Belgium reported the largest network component of 105 EUR/MWh which is double that of the EU average (52 EUR/MWh), followed by Sweden with a network component of 85 EUR/MWh.

The largest energy components were reported by the islands of Cyprus and Malta. Relatively high energy components result from the characteristics of non- interconnected island systems: limited economies of scale, higher proportion of costs to ensure security of supply and the lack of gas and electricity interconnections. Limited land availability and resulting expensive land occupation fees might also contribute to higher energy components.

Italy and Ireland reported the highest energy components among interconnected countries. In Italy, it results mostly from the prominent and increasing role of natural gas in the country's generation mix. There is no nuclear generation in Italy and coal plays a very limited role. Production from renewable sources decreased by 3.3% in 2016, mostly due to a significant drop in hydroelectric production, resulting from reduced availability of water resources. As hydro generation decreased by 3.3 TWh and gas fired generation grew by 7 TWh, the latter's share increased to almost half (47%) of the country's gross electricity production. At the same time, wholesale prices of gas also returned to growth. These developments are reflected in the energy component.

In Ireland, much like in the case of isolated island systems, the lack of interconnections is contributing to a higher energy component. The level of interconnection in Ireland is relatively low, at below 5% of all installed capacity or at 6.6% of dispatchable generation. The share of natural gas is also a contributing factor in Ireland: at 50% the highest in Europe. Natural gas is mostly the price setting, marginal plant of the merit order and as such, relatively expensive.

Figure 15 - Composition of hosehold prices in 2017 (most rep.)

Source: DG ENER in-house data collection

Box – European Commission efforts to increase interconnection capacities

As we saw above, interconnections – or more precisely the lack thereof – is a key contributing factor to high energy components. The socioeconomic value of electricity interconnectors comes from their ability of reducing costs by increasing the efficiency of the electricity systems and in parallel improving security of supply and facilitating the cost effective integration of the growing share of renewables.

The framework for the trans-European energy networks (TEN-E) and the Projects of Common Interest (PCIs) are the main tools of the EU energy policy to increase the physical electricity exchange capacity between Member States. The PCIs aim particularly to better connect the peripheral regions such as for example the Iberian Peninsula with the rest of Europe or to integrate rapidly growing share of renewables from remote generation areas such as the Northern Seas. On the current third Union list, there are 110 PCIs in electricity, which benefit from a streamlined permit granting procedures, improved regulatory conditions and under certain conditions are eligible for funding through the Connecting Europe Facility.

In addition, the 10% electricity interconnection target by 2020 has provided political momentum to advance key cross-border projects. As a result, seventeen Member States have already reached the target and seven more are on the path to reaching the target by 2020 through the completion of PCIs currently under construction.

In November 2017, the Commission proposed to operationalise the 15% interconnection target by 2030 through a set of additional and more specific thresholds which serve as indicators of the urgency of the action needed. The new thresholds reflect the three headline goals of European energy policy: increasing competitiveness through market integration and better prices, guaranteeing security of supply and achieving the climate targets through increased use of renewable sources. In total, approximately 22 electricity infrastructure PCIs have been completed or will be in operation by the end of 2018. Another 31 important projects are scheduled to be completed around 2020.

Composition of taxes, levies, fees and charges

In order to better understand how specific policies and fiscal instruments impact taxation levels- which in turn impact total prices- taxes and levies are broken down into 10 sub-components. These sub-components were designed to showcase national characteristics to the fullest extent possible, while minimizing the number of elements designated as "other". DG Energy provided extensive guidance to reporting authorities to ensure that the wide range of tax elements that exist across the countries are assigned to sub-components in a harmonized manner.

It is important to note that we consider only policy support costs that directly impact retail prices. Also, not every tax sub-component exists in each Member State. The following graph displays EU averages.

Figure 16 - Breakdown of household prices (DC)

Source: DG ENER in-house data collection

Figure 17 - Composition of taxes in 2017 (Most rep.)

Source: DG ENER in-house data collection



Value Added Tax

VAT is imposed on household electricity prices in all reporting countries. The EU VAT Directive explicitly allows Member States to apply reduced rates to electricity. As a result, VAT rates range from 6% in the United Kingdom to 27% in Hungary. As the largest sub-component, VAT accounted for 37% of the tax component and 17% of the total price. VAT is an ad valorem tax, its absolute value is based on the value of all other elements in the price. Even if VAT rates remain unchanged but other elements increase, the absolute amount of VAT increases. The average amount of VAT paid by households across the EU was 29 EUR/MWh in 2017, an increase of 31% since 2008.

Environmental taxes incl. excise duty (Non- Earmarked Taxes)

The sub-component includes any manifestation of excise duty, environmental, greenhouse gas emission, transmission and distribution taxes. Their common characteristic is that normally revenues from these taxes are not earmarked to energy, climate or environment related policies. In other words, revenues flow into the central state budget regardless of the name of the tax. Minimum excise duty levels on energy products are harmonised on the EU level and are defined by the Council Directive 2003/96/EC22. The sub-component excludes VAT.

Non- earmarked taxes grew from 11 to 16 EUR/MWh by 2017. They made up 21% of the taxes component and 8% of the total price, representing the third largest sub-component after VAT and RES&CHP. This reflects an annual growth rate of 1%, significantly lower than for industrial consumers.

Renewable Energy and Combined Heat and Power

This sub-component includes any support to renewable energy (RES) and combined heat and power generation (CHP). Explicit RES & CHP support costs are presented for 25 EU Member States and Montenegro. In Finland and Malta the renewable energy support scheme is not financed through an explicit levy on consumer bills but from the central state budget. France has been following the same example since 2016.In Hungary household electricity consumers, unlike their industrial counterparts, are exempted from the RES levy (a value of 0 is reported). Therefore in these 4 countries the direct cost to households is zero. It is important to note that consumers still pay for the support of renewable energy, albeit in an implicit way. Data for the United Kingdom is estimated and includes support to energy efficiency. In several countries renewable energy is supported also from sources additional to taxes on consumer bills.

The average EU household paid 24 EUR/MWh for RES&CHP support in 2017. This figure equals 30% of the taxes component and 12% of the total EU price. The sub-component experienced a fast increase from 2008 to 2015 as it grew annually by 15%. An important departure from this trend is to be observed in 2016 when the cost of RES&CHP support on EU level started to fall. A 2% decrease is remarkable especially as the share of renewable energy in the EU's energy mix grew by 16% from 2015 to 2016. These developments mean not only that the average EU household pays less but also that the lower cost enables more renewable energy.

Box - The French energy taxation and support to RES

France recently overhauled its energy taxation. Until 31 December 2015 a Public Service Obligation was levied on electricity (CSPE) and natural gas (CSPG) bills. The PSO consisted of three main elements: support to renewable energy, support to vulnerable consumers and national tariff equalization 9 . All three PSO elements are accounted for separately in our report and the latter two are classified as social levies. The RES support levy ranged from 2 EUR/MWh for large industrial consumers to 11 EUR/MWh for households. 10 The RES element of the PSO on natural gas bills specifically supported bio-methane. At the same time, a separate consumption tax was applied to both electricity (TICFE) and gas (TICGN). Rates of this tax on electricity rates ranged from 1.5 EUR/MWh for large industrial consumers to 12 EUR/MWh for households. 11

From 1 January 2016 the explicit RES levy, as well as both other PSO elements were discontinued. Vulnerable consumers and national tariff equalization are financed through the general budget. The dedicated budget line "Public service of energy" also includes smaller elements of the discontinued PSO such as, support to combined heat and power and CDC (Caisse des dépôts et consignation, the French National Promotional Bank) management fees. Until 2017 RES was supported through a transitory system.

In 2017 the electricity consumption tax (TICFE) increased to 9 EUR/MWh for large industrial consumers and to 31 12 EUR/MWh for households. The natural gas consumption tax increased to about 5 EUR/MWh for households and median industrial consumers, while large industrial consumers remained exempted. Revenues from these taxes feed into the general budget and do not support any specific policies.

From 2017 onwards, renewable energy is supported through the revenues of two fossil fuel taxes. TICPE, which is applied to the end use of petroleum products and TICC, which is applied to coal consumption, including coal used to generate electricity. Consequently, the financing of renewable energy is no longer directly connected to electricity or natural gas bills. This is reflected in our data: From 2017 onwards the RES sub-component is zero.

CSPE: Contribution au service public de l'électricité. Discontinued PSO.

CSPG: Contribution au service public du gaz. Discontinued PSO.

TICFE: Taxe intérieure sur la consommation finale d’électricité. Current consumption tax on electricity.

TICGN: Taxe intérieure sur la consommation de gaz naturel. Current consumption tax on gas.

TICPE: Taxe Intérieure sur les Produits Énergétiques. Fossil fuel tax on petroleum products.

TICC: Taxe Intérieure sur la Consommation de Charbon. Fossil fuel tax on coal.



Social Charges

Social charges include support to vulnerable consumers, social tariffs, last resort supplier, national tariff equalization, pension funds and support to sectorial employment policies. They were the second most important energy policy relevant charges across the EU in terms of the cost to consumers as well as in terms of the number of countries applying them 13 . Still, the impact of social charges remained limited. They made up 3% of the taxes component and 0.8% of the total price. They grew at the annual rate of 3% and reached 2.15 EUR/MWh in 2017. More telling is the number of countries that applied explicit social charges: it grew from 6 in 2008 to 9 in 2017. In Bulgaria and Portugal in certain years between 2015 and 2017 the sub-component was negative as rebates were effective, which decreased total prices in these countries. The United Kingdom's Warm Home Discount is a redistributive levy. All households pay towards this levy via their gas and electricity bills. All the revenues collected are recycled back as £140 annual rebates on the electricity bills of eligible low-income and vulnerable households. As such, the net effect of the policy on average dual fuel (gas and electricity) bills is £0. The underlying report however reflects the gross cost, as the rebate applies only to smaller a subset of households.

Security of Supply

Security of supply related levies were imposed by 9 Member States in 2017, up from 6 in 2008. The impact of security of supply related charges remained limited, at below 1% of the EU price.

Concession Fees

Concession fees, in most cases for the occupation of public land, accounted for 2% of the EU price and 4% of the taxes component. Such fees, averaging at 3.7 EUR/MWh, were imposed by 5 Member States (AT, BE, DE, PL, PT) on household electricity prices. The presence of concession fees in the EU average is due to a higher fee of 20.5 EUR/MWh in Germany.

Other energy policy relevant charges

The explicit impact of other energy policy relevant charges remained limited. Cost elements imposed to support energy efficiency, the nuclear sector or the national regulatory agency, each made up less than 1% of the average EU price. Charges for the financing of the National Regulatory Agency or a market operator are levied in 6 Member States (BE, CZ, PT, SK, SI, ES). Nuclear sector levies are imposed by 4 Member States (BE, IT, SK, ES). The nuclear sub-component includes only levies that directly impact retail prices by supporting the sector. Taxes paid by generators of nuclear energy - imposed on the wholesale level - are therefore not considered. Support to energy efficiency measures is explicitly levied in 4 Member States (BE, IT, SI, UK 14 ). The residual sub-component "other" includes a limited number of elements such as, deficit annuities in Spain, public television fee in Turkey or R&D in Denmark.

Figure 18 - Taxes, fees, levies and charges for EU households (DC)

Source: DG ENER in-house data collection

1.2.2 Industrial Electricity Prices

The following section analyses prices paid by non- household electricity consumers on EU and national levels. It examines prices of the Eurostat band ID, covering annual consumption of 2000 to 20 000 MWh. This band is often considered as the best indicator of economy wide non- household prices. It can be considered industrial as far as it might include prices paid by other non-household sectors, such as services, agriculture, fisheries and transport. 15

Box - The role of households and industry in our energy consumption

Households account for about 22% of our energy consumption 16 . This has not changed significantly over the last decade as the share of household consumption fluctuates close to this figure. Improvements in energy efficient appliances however slowly dampen these fluctuations that no longer reach as high as around 2010.

Industries accounted for 25% of the EU’s energy consumption in 2016, reflecting a 3 percentage point decrease since 2008, the beginning of the economic and financial crisis that dampened industrial production and thus energy consumption.



Evolution of Industrial Electricity Prices

Total prices grew at 1% annual rate from 2008 to 2017. In absolute terms the EU price grew from 94 to 103 EUR/MWh. This growth was slightly higher than inflation for industrial prices, which averaged at 0.5% annually during the same period. 17

The evolution of the EU industrial (ID) price conceals two distinct periods. From 2008 to 2014 it steadily grew by 2% annually. Since 2015 it has been decreasing by 1% annually These EU figures are made up of relatively homogenous national developments, as ID prices grew only in one third of the reporting countries from 2016 to 2017.

Figure 19 - Evolution and composition of the EU industrial price (ID)

Source: DG ENER in-house data collection

Composition of Industrial Electricity Prices

Over time, the composition of the EU industrial price changed even more significantly than the composition of the household price. The share of the energy component in the total price decreased by 28 percentage points from 68% to 40% in 2017. Still, unlike in the case of households, the energy component of industrial electricity prices remained the largest of the three components by a small margin. Its share in the total price was 2 p.p. higher than the share of taxes. In absolute terms, the energy component decreased at the annual rate of 5% and reached 41 EUR/MWh in 2017. The contraction of the energy component reflects falling wholesale prices which make up most of the component and can be linked to EU energy policies: increased competition resulting from market coupling and the growth of power generation capacity with low operating costs (such as wind and solar power, in addition to existing nuclear and hydro power). On country level, 9 countries reported actually higher energy components in 2017 than in 2016. In these countries wholesale prices either increased or their fall has not translated into a reduction of the energy component. Such results may imply that price competition in a number of retail markets is weak, allowing suppliers to avoid passing on wholesale price reductions to final consumers.

The share of the network component increased by 5p.p. from 17% to 22% in 2017. In absolute terms the network component grew at the annual rate of 4% and reached 22 EUR/MWh.

The share of the taxes component grew by 23 percentage points, taking over most part of the falling share of the energy component. It accounted for 15% of the EU price in 2008 and for 38% in 2017. In absolute terms, taxes grew at the annual rate of 12% and reached 39 EUR/MWh in 2017.

In 2017 Germany reported the highest medium industrial price (142 EUR/MWh), followed by Italy (133 EUR/MWh) and Cyprus. Germany and Italy reported also the highest taxes across the EU, while in Cyprus a high energy component results in overall higher prices, 18 In Germany RES support costs made up 80% of all taxes while excise duties accounted for an additional 18%. The composition of Italian taxes is similar to those of Germany, with RES accounting for four-fifths of all taxes. Sweden reported the lowest price of 55 EUR/MWh among all EU and non- EU countries.

For medium industrial consumers, Latvia, Lithuania, Germany and Slovakia reported the highest network components (in descending order) ranging from 36 to 31 EUR/MWh. Belgium, where network charges for household consumers are double that of the EU average, reported the 6th smallest network component for industrial consumers (18 EUR/MWh).

The largest energy components were reported by the islands of Cyprus and Malta, followed by Ireland and the United Kingdom.

Figure 20 - Industrial (ID) electricity prices in 2017

Source: DG ENER in-house data collection

Figure 21 - Composition of industrial (ID) electricity prices in 2017

Source: DG ENER in-house data collection

The previous graphs present medium consumption figures in each country. Several countries grant tax reductions to energy intensive industries. As energy intensity is not based on consumption volume, but on the share of energy related costs in total production, this median band can include enterprises that benefit from such reduced tax rates.

Drivers of Industrial Electricity Prices

The EU industrial price grew continuously from 2008 to 2015. This increase was driven by the combined impact of network charges and taxes, as both components were steadily increasing. At the same time, smaller decreases in the initially large energy component moderated the growth of total prices. A departure from this trend is to be observed in 2015 when the total EU price fell for the first time. This can be attributed entirely to the accelerated reduction of the energy component as the other two components kept increasing. From 2016 to 2017 the average industrial price further decreased from 106 to 103 EUR/MWh.

It is to be underlined that EU averages conceal divergent national developments. Total prices increased in one third of the reporting countries from 2016 to 2017. The energy component actually increased in 9 countries. On the national level network charges increased in 14 countries and taxes grew in 16 of the 28 EU Member States 19 .

Composition of taxes, levies, fees and charges

The following section considers only policy support costs that directly impact retail prices.

Figure 22 - Breakdown of EU prices in 2017 (ID)

Source: DG ENER in-house data collection

Figure 23 - Composition of taxes in 2017 (ID)

Source: DG ENER in-house data collection

Value Added Tax

VAT is recoverable for most industrial consumers in all reporting countries. Therefore this report analyses industrial prices excluding VAT. Other recoverable taxes are also excluded prom the price we report.

Environmental Taxes incl. excise duty (non- earmarked taxes)

Non- earmarked taxes grew from 6 EUR/MWh to 11 EUR/MWh in 2017. This reflects an annual growth rate of 3%, the third highest among the 10 different categories of taxes. Non- earmarked taxes made up 25% of the taxes component and 9% of the total price, representing the second largest sub-component after RES&CHP.

Renewable energy and Combined Heat and Power

This sub-component includes any support to renewable energy and combined heat and power generation. Explicit RES & CHP support costs are presented for 25 EU countries and Montenegro. In Finland and Malta the renewable energy support scheme is not financed from a levy on electricity consumption but from the central state budget. France has been following the same example since 2016. Therefore in these 3 countries the explicit cost of supporting renewable energy is zero for industrial consumers. It is important to note that consumers still pay for the support of renewable energy, albeit indirectly. In Hungary industrial consumers are subject to a RES levy while households are exempted. Data for the United Kingdom is estimated and includes support to energy efficiency. In several countries renewable energy is supported also from sources in addition to levies on electricity consumption.

RES & CHP accounted for 25 EUR/MWh in 2017 (up from 23.2 EUR/MWh in 201). This figure equals 64% of the taxes component and 24% of the total EU price. These relative shares are almost twice of the household figures. This is due to the fact that VAT, which accounts for a high share of household taxes, is recoverable for industrial consumers.



Social charges

 

Social charges include support to vulnerable consumers, social tariffs, last resort supplier, national tariff equalization, pension funds and support to sectorial employment policies. The impact of social charges on industrial prices diminished over time, as they decreased from 0.9 EUR/MWh to 0.6 EUR/MWH in 2017. They made up 2% of the taxes component and 0.6% of the total price (slightly less than for household consumers) in 2017.

Security of supply

Security of supply related levies were imposed by 10 countries in 2017, up from 6 in 2008. The impact of security of supply related charges remained limited, at below 1% of the EU price.

Concession fees

Concession fees for industrial consumers were much smaller than for their household counterparts. As a result, the impact of such fees remained insignificant (below half a percent of the total price).

Other energy policy relevant charges

The explicit impact of other energy policy relevant charges also remained limited. Cost elements imposed to support energy efficiency, the nuclear sector and national regulatory agency made up less than 1% of the average EU price. Charges for the financing of the National Regulatory Agency or a market operator are levied in 6 Member States (BE, ME, PT, SK, SI, ES). Nuclear sector levies are imposed by 4 Member States (BE, IT, SK, ES). The nuclear sub-component includes only levies that directly impact retail prices by supporting the sector. Taxes paid by generators of nuclear energy - imposed on the wholesale level - are therefore not included. Support to energy efficiency measures is explicitly levied in 4 Member States (BE, IT, SI, UK 20 ). The residual sub-component "other" includes a limited number of elements such as, deficit annuities in Spain, public television fee in Turkey or R&D in Denmark.



Figure 24 - Taxes, levies, fees and charges of industrial electricity prices

Source: DG ENER in-house data collection

1.2.3Small vs. Large Industrial Electricity Prices

The following section compares prices paid by two types of non- household consumers: consumers with small volume of annual consumption (20 to 500 MWh defined as band IB) and consumers with large volume of annual consumption (70 000 to 150 000 MWh defined as IF). Band IB typically includes small enterprises and services. Band IF includes large industrial consumers such as the chemicals, metals and construction materials industries. Band IF covers many energy intensive industries, however not all of them. Energy intensity is not defined based on consumption volume but on the relative importance of energy costs in total production costs. As a result, an enterprise that consumes small amount of electricity can still be energy intensive. The opposite holds as well. There are enterprises consuming large amounts of electricity that are not considered energy intensive. An example is the automobile industry. Band IF is therefore a mixed band that covers prices paid by energy intensive and non-energy intensive industries. This is reflected in our data: large consumers pay lower taxes due to partial exemptions from levies according to the Energy and Environment State Aid Guidelines (EEAG). These partial exemptions apply to some, not all IF consumers.

Data for IB was reported by 26 EU member States (except the United Kingdom and Greece), Montenegro, Norway and Turkey. Data for the UK and Greece was estimated by DG Energy. IF data was reported by 25 EU Member States and Turkey. Data for the UK and Greece was also estimated by DG Energy for this band. In Luxembourg IF data is confidential (there are less than three consumers in the band).

Evolution of Small vs. Large Industrial Electricity Prices

Total prices grew at annual rate of 1.8% for IB consumers. IF prices experienced a different evolution as they fell slowly, by 0.3% a year. In absolute terms the IB price grew from 121 to 143 EUR/MWh and the IF price fell from 83 to 80 EUR/MWh. Inflation averaged at annual 0.5% during the same period. 21 The above figures cover the period 2008 to 2017, by 2015 year-on-year prices were falling for both consumer types. The EU averages conceal relatively homogenous evolutions across the reporting countries as IB and IF prices increased in 9 and 11 countries respectively from 2016 to 2017.

In 2017 Germany reported the highest (192 EUR/MWh) IB price while Sweden reported the smallest (78 EUR/MWh) across the EU. Both Norway and Turkey reported prices lower than any EU Member State. Cyprus reported the highest price for large industrial consumers (117 EUR/MWh) followed by Germany (114 EUR/MWh) and the United Kingdom (113 EUR/MWh). Also Sweden reported the smallest price for IF (40 EUR/MWh). Over the last decade the ratio of the largest to smallest price evolved differently for the two consumer types. For IB consumers the ratio decreased by 13% and the largest price was 2.7 times of the smallest. For IF consumers the ratio increased by 18% to 4.7.

Figure 25 - Evolution of small (IB) and large (IF) industrial electricity prices

Source: DG ENER in-house data collection

Composition of Small vs. Large Industrial Electricity Prices

Albeit the two bands reflect prices for quite different consumer types, the composition of prices evolved in a notably similar fashion. For both consumer types the energy component lost about 25 p.p. of its share in the total price. Also for both consumer types taxes took over most of the decreasing share of the energy component, as they experienced around 20 p.p. increase. A smaller increase in the share of network charges accounted for the remaining changes.

Table 2 - Composition of small (IB) and large (IF) industrial prices

 

IB

IF

IB

IF

 

2008

2017

2008

2017

Change (p.p., 2008-2017)

Energy

62%

38%

74%

49%

-24%

-25%

Network

25%

28%

13%

17%

3%

4%

Taxes

13%

34%

13%

34%

20%

21%

Source: DG ENER in-house data collection

In absolute terms, the energy component decreased at an annual rate of 3.5-5% for the two consumer types. By 2017 it reached 54 and 39 EUR/MWh for IB and IF consumers respectively. The contraction of the energy component reflects falling wholesale prices, which make up most of the component

In absolute terms the network component grew at the annual rate of 2.7% (IB) and 3.3% (IF) for the two consumer types. For IB consumers the network component grew from 30 to 40 EUR/MWh from 208 to 2017. The IF network component grew from 10 to 13 EUR/MWh during the same period.

In 2008 taxes made up 34% of the total price for both consumer types. In absolute terms they grew from 16 to 48 EUR/MWh for IB consumers and from 10 to 27 EUR/MWh for IF consumers. Taxes grew 2 p.p. faster for large consumers (by 13% annually).

In 2017 the total IB price was 143 EUR/MWh, 62 EUR higher than the IF price. The reason for this difference lies in all three components. Some IF consumers are auto producers, who generate their own electricity or have over the counter, long term power purchase agreements which are reflected by lower energy components in our data (54 vs. 39 EUR/MWh). IF consumers are often connected directly to the high voltage transmission grid and therefore do not need to pay for using the distribution grid. This reduces the network component in our data set (40 vs. 13 EUR/MWh). Lastly, for IB consumers taxes at 48 EUR/MWh were 17 EUR higher than for IF consumers. The difference stems mostly from partial reductions in RES levies, according to the State Aid Guidelines and from reduced excise duty rates according to the Energy Tax Directive.

As all three components are proportionally smaller, taxes made up the exact same share of the total price for both consumers types: 34%. While relative shares for the two consumer types were equal, as we saw above, large consumers pay lower taxes. Reduced taxes reflect efforts to safeguard the competitiveness of EU industries. These reductions, which cannot exceed 85% of the applicable RES levy, qualify as aid compatible with the internal market. Reductions are partial, therefore no consumer is completely exempted. Qualifying industries are energy intensive, meaning that they face higher share of energy costs in their production (around 10% compared to 3% cross- industry average). Furthermore, due to the trade intensity of the goods they produce, they are exposed to competition with producers outside of the EU, who are not subject to comparable climate legislation. As a result of reductions in the applicable RES levies, not only the share of taxes in the total price is equal for the two consumer types, but also the subset of taxes supporting renewable energy. From 2008 to 2016 the share of RES support in the total price was equal in each year for IB and IF consumers. A difference appears only in 2017 when the RES levy for IB consumers was 20% of the total price, compared to 23% for IF consumers.

Figure 26 - Composition of small (IB) and large (IF) industrial electricity prices

Source: DG ENER in-house data collection

Drivers of Small vs. Large Industrial Electricity Prices

Both for small and large industrial consumers we observe the continuation of the trends of the last decade, with one exemption. The energy component continued to fall, taxes continued to increase. Both unbroken trends since 2008. The network component continued to slightly increase for IB consumers but fell for the first time for IF consumers from 2016 to 2017. This development represents the only new trend, however an important change lies in the speed of increases and decreases of the three components. We observe falling total prices for both consumer types since 2015, while the evolution of the three components mostly didn’t change direction. Falling total prices result from the fact that the energy component kept decreasing at the same speed (4-7% year-on-year) while the increase of network charges and taxes slowed down to around 2% compared to double digit growth rates in the early years of our observation period. Since 2015 the decrease of the energy component was no longer only mitigating increases of the two other components but overtook them, resulting in overall falling prices both for small and large industrial consumers.

These EU level figures conceal relatively homogenous developments on national level. For IF consumers total prices increased in 9 countries and the energy component increased in 10 countries from 2016 to 2017. For IF consumers both total prices and the energy component increased in 11 countries. The list of countries where total prices increased and where the energy component increased is not identical but notably similar. This indicates, as also noted above, that in most countries the evolution of the network and taxes components slowed down so much that the overall price evolution is now set by the energy component.

Figure 27- Composition of small (IB) and large (IF) industrial prices in 2017. First 16 countries alphabetically.

Source: DG ENER in-house data collection

In 2017 Germany reported the highest small industrial price of 192 EUR/MWh followed by Italy (172).

In Germany and Italy taxes are made up mostly of RES support (above 70% in both countries) and excise duty (17%), complemented with smaller elements, such as concession fees in Germany.

In 2017 Latvia, Ireland and Belgium reported the highest network components of 62, 59 and 56 EUR/MWh respectively for small industrial consumers (IB). While the lists of countries with highest level for taxation for small and large industrial consumers are very similar, we observe a difference in terms of network charges. For large industrial consumers the United Kingdom, the Czech Republic, Malta and Germany reported the highest network components in descending order. Belgium, where network charges for household consumers are double that of the EU average, reported the 3rd smallest network component in the EU for large industrial consumers.

The energy component was the highest in the same countries as in the case of household and medium industrial consumers. For further explanation, please consult the corresponding section of the chapter on household prices.

Figure 28 - Composition of small (IB) and large (IF) industrial prices in 2017. Last 15 countries alphabetically.

Source: DG ENER in-house data collection

Composition of taxes, levies, fees and charges

Figure 29 - Composition of taxes on prices for small (IB) and large (IF) industrial electricity prices 
(2008-2017)

Source: DG ENER in-house data collection

Figure 30 - Composition of taxes on prices for small (IB) and large (IF) industrial electricity consumers 
in 2017

Source: DG ENER in-house data collection

Value Added Tax

VAT is recoverable for most industrial consumers in all reporting countries. Therefore the analyses in this section focus on industrial prices excluding VAT. Other recoverable taxes are also excluded prom the price we report.

Other non- earmarked taxes

Non- earmarked taxes for small industrial consumers grew from 7 to 12 EUR/MWh by 2017. This reflects an annual growth rate of 2.4%, the third highest among the 10 different categories of taxes. Non- earmarked taxes made up 26% of the taxes component and 8% of the total IB price. For large industrial consumers, non- earmarked taxes grew even faster, at the annual rate of 3%, albeit from an initially lower level. They averaged at 7 EUR/MWh across the EU in 2017.

Renewable energy and Combined Heat and Power

This sub-component includes any support to renewable energy and combined heat and power generation. Explicit RES & CHP support costs were reported by 25 EU countries. In Finland and Malta, the renewable energy support scheme is not financed from a levy on electricity consumption but from the central state budget. France has been following the same example since 2016. Therefore in these 3 countries the cost of supporting renewable energy through electricity bills is zero. It is important to note that consumers still pay for the support of renewable energy, albeit in an implicit way. Data for the United Kingdom is estimated and includes support to energy efficiency.

RES& CHP support cost were 29 and 18 EUR/MWh for small and large consumers respectively in 2017. RES support made up 59% of all taxes and 20% of the total price for small industrial consumers. The same shares for large industrial consumers were slightly higher at 65% (of taxes) and 22% (of the total price). The RES sub-component grew at the annual rate of 16% and 18% for IB and IF respectively. These growth rates covering the period 2008 to 2017, conceal volatile sub- periods: after strong initial growth from 2008 to 2012 we observe almost stagnation from 2014 to 2016 for both consumer types. From 2016 to 2017 the cost of RES support grew again slightly faster.

The following graphs display the cost of supporting renewable energy for different consumer types. They reflect the extent to which Member States make use of the above described exemptions and reductions regulated by the State Aid Guidelines. Of the 26 countries that reported RES support costs for both consumer types, burden sharing is equal in 6. In 3 (HR, HU, IT) households pay less than small and medium industrial consumers (albeit more than large industrial consumers). In 17 countries the levy is degressive, it decreases as the consumption increases.

Figure 31 - RES support costs for electricity consumers (DC=Household, IB=small, ID=median, IF=large)

Source: DG ENER in-house data collection

Figure 32 - RES support in 2017 by consumer type and country. First 13 countries alphabetically

Source: DG ENER in-house data collection

Figure 33 - RES support in 2017 by consumer type and country. Last 13 countries alphabetically

Source: DG ENER in-house data collection

The share of all other policy support costs, including energy efficiency, security of supply and nuclear sector policies, remained below 1% of the total price.

1.2.4 International comparisons

Component level data (energy, network, taxes) is not available for countries outside of the European Statistical System, which encompasses the 28 EU Member States, 4 EFTA members, 9 Energy Community contracting parties of the western- Balkans and Turkey. Consequently Turkey is the only G20 trading partner of the EU for which component level data is available.

Component level data enables the identification of price drivers. As this data is not available for G20 trading partners, the difference between wholesale and retail prices can serve as a proxy. The difference consists of network charges, taxes, levies as well as of the costs and profit margins of supply companies. Consequently, the difference includes elements from all three components. The non- regulated, supply related costs account for only a small share of the total difference in most countries.

We observe that the difference between wholesale and retail prices, therefore the impact of the regulated part, is larger in the EU than in its G20 trading partners. This holds for both electricity and natural gas and both households and industry. On average across the EU the non- regulated part of the [price has been contracting since 2008 while the regulated part kept growing until recently. At the same time we observe that retail prices are below wholesale prices in some trading partner countries, indicating that prices are subsidized and regulated at low levels. Consumers pay less than the actual cost of their energy use.

Electricity wholesale prices in the EU are often comparable to those in G20 countries. This however does not translate to retail prices as such are on average higher in the EU than in all G20 trading partners. This is a result of various taxes and levies that provide revenue to treasuries through excise taxes and finance policies, such as renewable energy and energy efficiency. These contributions result in higher prices but also allowed the EU to be a global driving force in combatting climate change and a leader in the development and deployment of sustainable energy technologies.

Household Electricity Prices

The EU28 average difference between household retail prices and wholesale prices has increased from around 100 EUR/MWh in 2008 to more than 160 EUR/MWh in 2017.

The difference in the US is lower than in the EU28 at around 80-90 EUR/MW, albeit it has been increasing since 2008. The same analysis using the wholesale proxy for China shows a negative outcome of 30-40 EUR/MWh, this highlights that household consumers in China are not paying the full cost of their electricity use. The difference in Japan varied considerably over the observation period, with the Fukushima effect on wholesale prices likely to have played an important role in the 2011 peak. For the other G20 countries the differences are significantly lower than for the EU28 average. In Mexico (MX), Indonesia (ID) and Russia (RF) the difference is small, indicating that retail prices are regulated and kept at low levels in these countries. In Canada (CA), Turkey (TR) and Brazil (BR) the difference is greater, but still significantly smaller than for the EU28 average.

Figure 34 - Difference between household retail electricity prices and electricity wholesale prices 2008-2018, EUR2017/MWh

Source: Trinomics et altri study

Industrial Electricity Prices

The EU28 average difference between household retail prices and wholesale prices has increased from around 30 EUR/MWh in 2008 to around 70 EUR/MWh in 2017. The difference in the US is lower than in the EU28 at around 15-40 EUR/MWh, albeit it has been slowly increasing since 2008. The difference in Japan is in the same order of magnitude as the EU28 average and US levels, but has varied considerably over the period, with the annual frequency of the data playing a role, and the Fukushima effect on wholesale prices likely to have played an important role in the 2011 peak. The same analysis using the wholesale proxy for China shows virtually no difference, likely due to the proxy being similar to the industrial price, it is an interesting contrast to household prices, pointing towards energy policy priorities and price interventions in favour of households rather than industry.

For the rest of the G20 countries the difference compared to the EU28 average is typically lower, although the difference in Canada (CA) has generally been similar to the EU. We can observe a small divergence in Mexican (ME) and Russian (RF) prices which remained mostly constant over time, while the difference in the EU increased. In Turkey (TR) the difference in prices was often negative highlighting that retail prices are regulated at levels below the cost of supplying electricity.

Figure 35 - Difference between industrial retail electricity prices and electricity wholesale prices, EU28 and other G20 countries, 2008-2018, EUR2017/MWh

Source: Trinomics et altri study


2Gas prices

2.1Wholesale gas prices 

Main findings

¾European wholesale gas prices plummeted in the wake of the 2008-2009 financial crises but recovered by 2011-2013, helped by the economic recovery and the Fukushima accident which increased global LNG demand. This was followed by a period of declining prices as low oil prices and increasing global LNG supplies, coupled with weak demand put pressure on European gas prices. After bottoming out in 2016, wholesale gas prices have been on the rise, driven by the economic recovery, rising gas demand (both in Europe and globally) and increasing oil and coal prices.

¾European wholesale prices move in a rather wide range, with regional price differences driven mainly by the level of competition: in general, markets with higher levels of competition show a lower price level than markets with only one supply source. Since 2015, falling oil prices, the decreasing role of oil-indexation and, in some cases the diversification of supply sources contributed to converging wholesale prices in Europe.

¾Among the different pricing mechanism, oil-indexation is losing ground in the European market but continue to play an important role in certain regions, in particular in the Mediterranean, in Southeast Europe and the Baltics. On the other hand, hub prices gained significant ground in Central Europe: wholesale prices in this region are more and more aligned with Northwest European hub prices, rather than with oil-indexed prices.

¾From time to time, daily prices can show extreme volatility, typically when cold snaps sharply increase demand while supply is limited by infrastructure constraints or other factors. Two such extreme periods occurred during the 2017-2018 winter, leading to unprecedented prices at certain gas hubs. On both occasions, rising prices provided the right signal to market participants and gas supplies were not interrupted but the extent of the price rise seems to point toward the inflexibility of demand.

¾While oil-indexed prices have a diminishing role in the European market, European wholesale gas prices continue to be closely aligned with the oil price, reflecting the close relationship between the gas market and the wider energy complex. There is a strong correlation between oil and gas prices in the long term but during shorter periods the price trend of the two commodities can diverge. Compared to oil, the price of gas is more exposed to seasonality.

¾EU gas demand shows a strong seasonality which is reflected in the development of wholesale gas prices: prices tend to be higher when temperatures are low. Periods of extremely cold temperatures often trigger price peaks.

¾In international comparison, European wholesale gas prices are well above those in major gas producing countries (Canada, Russia, US) but in general lower than in other G20 economies, especially those which solely or largely rely on LNG imports (e.g. China, Japan, Korea). International prices have converged since 2015 which means that the absolute value of the regional differences decreased but, nevertheless, these differences proved persistent.

2.1.1Evolution of wholesale gas prices

Gas prices, similarly to the price of most other commodities, peaked in 2008, driven by robust global economic growth and rising demand from emerging markets, particularly China. In the wake of the economic crisis, prices sharply decreased: in less than one year, European gas hub prices fell by around 70%. Prices gradually recovered between 2009 and 2013, helped by the economic recovery and the Fukushima accident which increased global LNG demand. In March 2013, hub prices exceeded the record levels reached in 2008.

In 2013-2016, wholesale gas prices were on a declining trajectory and, by 2016, European wholesale gas prices fell to the lowest levels since 2009. In this period, low oil prices and increasing global LNG supplies, coupled with weak demand put pressure on European gas prices.

Wholesale gas prices have bottomed out in 2016 and have been on the rise since then, driven by the economic recovery, rising gas demand (both in Europe and globally) and increasing oil and coal prices. Hub prices have nearly doubled since 2016 and, at the same time, they have been exhibiting strong seasonal volatility.

The Commission follows the development of a number of wholesale gas prices across the EU, including prices at trading hubs, estimated border prices calculated based on customs data and other prices reported by commercial data providers or other sources. Wholesale prices move in a rather broad band: in 2008-2014, the average difference between the highest and lowest price was close to 20 EUR/MWh. From 2015, prices have perceivably converged, with the widths of the band narrowing to around 10 EUR/MWh in 2015-2017. In case of extreme events (cold spells and/or supply disruptions) affecting specific regions, e.g. in the first quarter of 2018, the price band can become much wider.

Hub prices, especially those in the liquid Northwest European markets have been near to the lower border of the price range for most of the last decade, as demonstrated in Figure 36  by the price at the Dutch (TTF) and the UK (NBP) hubs. Oil-indexed prices, on the other hand, have been typically closer to the upper border of the price band for most of the period, as indicated by the development of the Platts North West Europe Gas Contract Indicator (GCI), a theoretical index showing what a gas price linked 100% to oil would be.

Regional price differences are largely explained by the different pricing mechanisms and the different levels of competition. In general, markets with higher levels of competition show a lower price level than markets with only one supply source. Lower oil prices, the decreasing role of oil-indexation and, in some cases, new supply sources (e.g. LNG in Lithuania and Poland) contributed to converging wholesale prices in Europe in 2015-2017.

Figure 36 - Selected wholesale gas prices in Europe

Source: Platts, BAFA, Eurostat Comext

The difference between GCI and the price at the Dutch hub (TTF) averaged around 10 EUR/MWh in 2011-2014. In the wake of the oil price fall in 2014-2015, oil-indexed gas prices have significantly decreased, facilitating the convergence of European wholesale gas prices. In certain periods, oil-indexed prices were actually lower than the price at the most liquid gas hubs in Northwest Europe. This was the case during most of the 2016-2017 winter and in March 2018 when hub prices grew driven by high seasonal demand and supply disruptions.

Figure 37 - The difference between the Platts North West Europe Gas Contract Indicator (GCI) and the Dutch hub price (TTF)

Source: Platts

Over the period, there has been a trend of moving from oil-indexed prices towards hub pricing. After years of a declining trend, the share of oil-indexation in Europe seems to have stabilised around 30% in 2014-2016 but there are significant regional differences.

The development of the German border price (a weighted average gas import price for the country) on Figure 36 clearly demonstrates the trend of moving towards hub pricing: while in 2009 it was very close to the Platts GCI index, it gradually approximated the Dutch and UK hub prices, showing that the pricing of German imports, including those coming from Russia, is increasingly based on hub prices, rather than oil-indexation. As Figure 38 shows, such a trend could be observed also in the countries of Central Europe (e.g. Czech Republic, Hungary, Slovakia) where wholesale prices are more and more aligned with Northwest European hub prices, rather than with oil-indexed prices. On the other hand, oil-indexation continued to have an important role in the Mediterranean, in Southeast Europe and the Baltics.

Figure 38 - Price formation in Europe

Source: IGU Wholesale Gas Price Survey, 2018 Edition

Northwest Europe: Belgium, Denmark, France, Germany, Ireland, Netherlands, UK

Central Europe: Austria, Czech Republic, Hungary, Poland, Slovakia, Switzerland

Mediterranean: Greece, Italy, Portugal, Spain, Turkey

Southeast Europe: Bosnia, Bulgaria, Croatia, FYROM, Romania, Serbia, Slovenia

Scandinavia & Baltics: Estonia, Finland, Latvia, Lithuania, Norway, Sweden

OPE: oil price escalation, GOG: gas-on-gas competition; BIM: bilateral monopoly, NET: netback from final product, RCS: regulated cost of service, RSP: regulated social and political, RBC: regulated below cost, NP: no price

The monthly average prices depicted in Figure 36 often hide a high degree of daily volatility. For short periods, daily prices can reach exceptionally high levels, typically when cold naps sharply increase demand while supply is limited by infrastructure constraints or other factors. Figure 39 shows that two such extreme periods occurred during the 2017-2018 winter, leading to unprecedented prices at certain gas hubs.

Figure 39 - Daily day-ahead prices at selected gas hubs from 2008 to mid-2018

Source: Platts

On 12 December 2017, an explosion at the Baumgarten facility in Austria cut Russian supplies to the country. In Austria, domestic supplies were secured from increased storage withdrawals. The accident also cut all Russian imports to Italy, the main source of supply to the country. In Italy, the supply-demand balance was rather tight already before the incident, because of strong winter demand and reduced import capacity through Switzerland. In the wake of the disruption, the Italian day-ahead price settled at a record 80 EUR/MWh on 12 December, up from 23.7 EUR/MWh the day before. The price spike allowed demand to fall and alternative supplies to grow. Italy has a number of import sources and, within hours, pipeline imports were ratcheted up on all supply routes, with the biggest additional volumes coming from Switzerland and Algeria. To a lesser extent, Libyan supplies also increased. In addition, LNG send-out from the Adriatic LNG facility was also stepped up. Growing imports were not sufficient to replace the missing volumes; increased withdrawals for domestic storages had to fill the gap. A longer-lasting disruption would have also attracted additional LNG cargoes. However, the outage was resolved in less than a day, with gas flows from Russia resuming before midnight and prices quickly receding to previous levels.

More or less at the same time, amid strong seasonal demand, supplies in Northwest Europe were constrained by the unplanned outage of the Forties pipeline system in the UK North Sea and a 2-day loss of output from Norway’s Troll field. The outage of the Forties pipeline lasted from 11 to 30 December 2017 and triggered a reduction of UK gas production by 30-40 million cubic meters a day. As a result, UK prices also soared, reaching the highest level in nearly three years and incentivising additional imports. On 12 December, the NBP settled at 26.1 EUR/MWh, 3.9 EUR/MWh above the TTF and the premium remained around 2 EUR/MWh between 13 and 20 December. Increasing imports through the Interconnector (from Belgium) and the BBL pipeline (from the Netherlands) were instrumental in replacing missing volumes and, after 20 December, rising temperatures also contributed to the easing of the market tightness. On the other hand, the price signal was not strong enough to boost LNG imports.

During a late-winter cold snap in late February/early March 2018, temperatures dropped around 10oC below the seasonal average in most of Europe, causing a significant price spike. The cold spell triggered record gas consumption in a number of Member States and tightened the supply-demand balance in Europe. The cold spell arrived at a time when gas storage levels were already rather low (especially in Northwest Europe), increasing the impact on day-ahead prices. Hub prices in Western Europe reached unprecedented levels, with TTF and NBP closing at 79.0 Euro/MWh and 88.4 Euro/MWh, respectively on 1 March 2018. Storage withdrawals typically tail off in February but this was not the case in 2018. Skyrocketing spot prices provided a boost to withdrawals which was a key source of gas supply during the cold spell. On 28 February, withdrawals in the EU reached 11.4 TWh (nearly 1.1 bcm), the highest daily rate on record. On the other hand, the price spike did not last long enough to attract additional LNG supplies: LNG imports remained below 2017 levels in both February and March 2018.

On both occasions, rising prices provided the right signal to market participants and gas supplies were not interrupted but the extent of the price rise seems to point toward the inflexibility of demand.

Figure 39 also depicts the price in Southern France which, on certain occasions, well exceeds the price in the Northern part of the country. Northern France has access to the diverse supply sources available in Northwest Europe and, as a result, the price is typically very close to the price at the Dutch TTF hub. Southern France, however, is largely relying on the LNG terminals on its Mediterranean coast; constraints on the north-south link within France mean that prices can be quite divergent. For example, the premium of PEG Sud/TRS over PEG Nord reached a record 23.0 EUR/MWh on 20 January 2017 when high seasonal demand coupled with low LNG imports and the persistent capacity restrictions on the north-south link caused supply tightness in the Southern part of the country. In this period, LNG supplies from Algeria, France's main LNG supplier, were hindered by technical issues. Under such circumstances, prices had to increase substantially and for a sustained period in order to attract LNG cargoes from the high-priced Asian market. By early February, milder weather and additional LNG cargoes allowed the situation to ease and the premium of TRS over PEG Nord has practically disappeared. Ongoing infrastructure upgrade projects will help to debottleneck the north-south link; once these are completed, such big price differences are not likely to occur any more.



2.1.2Factors impacting the evolution of wholesale gas prices

The development of wholesale gas prices is influenced by a number of factors. In this section we look into the impact of two important variables, the oil price and the weather.

There is a persistent strong correlation between oil and gas prices. By definition, this is the case for oil-indexed prices, as shown by Figure 40 which depicts the movement of the Brent oil price and the Platts North West Europe Gas Contract Indicator (GCI), a theoretical index showing what a gas price linked 100% to oil would be. Typically there is a 6-9 month time lag in the pricing formulas used which means that oil-indexed gas prices react to changes in the oil price with a delay. For example, Brent started its steep fall in mid-2014 but this was reflected in the development of oil-indexed prices only from the beginning of 2015.

Figure 40 - The monthly average price of oil (Brent) and oil-indexed gas contracts (Platts GCI)

Source: Platts

The observation of a strong correlation between oil and gas prices also holds for European gas hub prices, as shown in Figure 41 through the example of the Dutch TTF, Europe's most liquid hub. While oil-indexed prices have a diminishing role in the European market (see section 2.1.1), hub prices continue to be closely aligned with the oil price, reflecting the close relationship between the gas market and the wider energy complex.

Looking at the last decade, this correlation is apparent in the long term but in shorter periods the price trend of the two communities can diverge. For example, in the second half of 2014, when oil prices started to fall steeply, gas hub prices moved in the opposite direction. Also, the price of gas is more exposed to seasonality, as demonstrated lately by the distinct peaks during the 2016-2017 and the 2017-2018 winters.

In certain cases there seems to be a time gap between changes in oil and gas hub prices: in 2008-2009, TTF plunged a couple of months later than Brent while in 2014 the fall of TTF preceded the collapse of the oil price.

Figure 41 - Daily spot prices of oil (Brent) and gas (at the Dutch TTF hub)

Source: Platts

Figure 42 depicts daily changes of Brent and TTF. Dots represent individual days, showing the change of oil price (on the horizontal axis) and the gas price (on the vertical axis) compared to the previous day, expressed in percentage. While oil and gas prices do not necessarily change in the same direction every day, there is a weak positive correlation, particularly the increasing oil prices often coincide with increasing gas prices.

Figure 42 - Daily change of spot prices of oil (Brent) and gas (at the Dutch TTF hub)

Source: Platts

Measured in energy content, oil has traditionally been more expensive than natural gas. This was the case in the last decade, except for a short period at the end of 2008 and the beginning of 2009 when European gas hub prices followed the plunge of oil price with some delay. Between the beginning of 2008 and mid-2018, the price of Brent (measured in EUR/MWh) was on average 86% higher than the price of gas at the TTF hub. This ratio has gradually decreased over the period, meaning that the relative price of gas compared to oil increased. In the last three and a half years, the average "premium" of oil over gas was 58%.

Figure 43 - The monthly average price of oil (Brent) and gas (at the Dutch TTF hub), measured in EUR/MWh

Source: Platts

Note: a conversion rate of 1.7 MWh/barrel was used for Brent

EU gas demand shows a strong seasonality, reflecting the fact that a large proportion of gas is used for space heating. According to 2016 data, the residential sector covered 27% of the gross inland consumption of gas in the EU; this is very close to the share of transformation input (gas used mainly in power stations and district heating plants) which was 33%. 22 Depending on temperatures, the level of gas consumption can be rather volatile during the winter months which can obviously have an impact on the price of gas.

Accordingly, one would expect a seasonality in prices reflecting the seasonality of demand, with lower temperatures associated with higher prices. Looking at the Netherlands, this is indeed the case in several years (in particular 2009, 2014, 2016 and 2017) with lower prices in summer and higher prices in winter. On the other hand, there are years when such a trend cannot be observed, indicating that gas prices are also impacted by other factors. But in general, periods of extremely cold temperatures (e.g. January 2009, January 2010, December 2010, January 2012, January 2017) trigger smaller or larger price peaks. Cold spells at the end of winter, when gas stocks are largely depleted (e.g. March 2013 and March 2018), can induce exceptionally high prices.

Figure 44 - Monthly average gas price at the Dutch TTF hub and heating degree days in the Netherlands

Source: Platts, Eurostat

On Figure 45 , dots represent individual days, showing the daily average temperature in the Netherlands (on the horizontal axis) and the gas price at the TTF hub (on the vertical axis). The linear trendline confirms the negative correlation between temperatures and prices. It is also clearly visible that exceptionally high prices occur on days with low temperatures. Since 2013, the TTF price never exceeded 30 EUR/MWh on days with an average temperature of more than 4oC. The highest price was observed on 1 March 2018 when temperatures were 9oC below the seasonal average.

Figure 45 - Daily gas price at the Dutch TTF hub and average daily temperature in the Netherlands from the beginning of 2013 to mid-2018

Source: Platts, Thomson Reuters/Point Carbon

Note: only weekdays are depicted

2.1.3International comparison

Comparing European gas wholesale prices with those in the EU's major trading partners provides an insight into how energy costs can impact the international competitiveness of energy intensive industries which are exposed to global trade.

After the 2011 Fukushima accident and the start of the US shale gas revolution, the three main regional benchmarks depicted in Figure 46 had significantly diverged and price differences remained large through 2011-2014.    Since the second half of 2014, there has been a convergence of international prices. Japanese LNG prices significantly decreased, facilitated by weak demand in Asia and the increasing global LNG supplies, and compounded by the fall of oil prices. These factors also contributed to the decrease of prices in Europe, thereby lowering the premium above the persistently low gas wholesale price in the US.

In the second quarter of 2017, the convergence among the key international gas prices reached the greatest level since the Fukushima accident. However, the trend of converging regional prices was interrupted during the last two winters (2016-2017 and 2017-2018) when Asian prices showed a steep rise due to strong seasonal demand. European and US prices also increased but to a lesser extent, resulting in a widening gap between regional benchmarks.

Figure 46 - Comparison of European, US and Japanese wholesale gas prices

Source: Platts, Thomson Reuters

There has been a slow convergence between US and European prices in 2012-2016, with the ratio of European and US prices dropping below 1.5 in the third quarter of 2016. Since then, however, European prices increased while US prices remained rather stable, leading to a diverging trend. European prices remain stubbornly high compared to those in the US: in 2017, the TTF price was on average almost two times higher than the US Henry Hub benchmark.

For most of 2011-2014, Japanese spot LNG prices were 40 to 100% higher than the Dutch TTF benchmark. In the last few years, during the summer months, the gap between European and Japanese prices has almost disappeared. However, during the winter months, driven by the strong seasonal demand in Asia, Japanese prices rise substantially and as a result show a distinctive premium over Europe. In 2017, the Japanese spot LNG price was on average 20% higher than the price at the Dutch TTF hub but in January 2018 the mark-up was well above this level, reaching 64%. In March, when TTF surged due to a late-winter cold spell, the Japanese spot LNG price was exceptionally lower than the Dutch benchmark.

Figure 47 - The ratio of European, US and Japanese wholesale gas prices

Source: Platts, Thomson Reuters

Table 3 - The ratio of European, US and Japanese wholesale gas prices

Source: Platts, Thomson Reuters – The 2018 values refer to the period of January-August 2018

The study prepared by Trinomics 23 provides a more comprehensive international comparison of gas wholesale prices, covering most G20 economies, with the findings shown in Figure 48 and Figure 49 . Prices are expressed in constant (2017) euros. In case of the EU, a weighted average of national wholesale prices was calculated and depicted.

The analysis reveals a very large dispersion of prices in 2011-2011, followed by a noticeable convergence from 2015. Part of the gas wholesale prices is indexed to oil prices and hence the price convergence was largely driven by the lowering of the crude oil price.

Major gas producing countries, including Canada, Russia and the US have the lowest gas wholesale prices in the G20. This was also the case in Australia until 2016 but then domestic supply shortages triggered a significant price rise.

Apart from the producing countries, wholesale prices in the G20 countries tend to be higher than the EU average, often showing a high degree of volatility.

Chinese wholesale prices follow a similar trend to the Japanese price but in 2011-2014, after the Fukushima accident, the absolute level of the price remained somewhat lower, probably because – unlike Japan– China is not fully reliant on LNG (the country also has indigenous production and pipeline imports from a couple of sources). In addition, Chinese prices exhibit less seasonality.

Figure 48 - Gas wholesale prices in the EU (weighted average), China, Japan and the US

Source: Platts, Thomson Reuters, Knoema (World Gas Intelligence; World Bank), World Bank Commodities Price Data (The Pink Sheet).

Figure 49 - Gas wholesale prices in the EU (weighted average) and selected markets

Source: Platts, Thomson Reuters, Knoema (World Gas Intelligence; World Bank), World Bank Commodities Price Data (The Pink Sheet).



2.2Retail gas prices 

Main findings

¾Progress towards the completion of the single gas market continued. This is reflected by the fact that national energy components are gradually converging: they became 20% 24 less spread out over the last decade.

¾Natural gas prices remained largely determined by the international wholesale price of the commodity and followed its evolution with a slight time lag. Consequently, the energy component - containing wholesale prices- retained its positions as the largest of the three components for all consumer types, with shares ranging from 50 to 80%.

¾In absolute terms the energy component increased by half a percent annually for households but decreased at a faster pace for industrial consumers (by 1.75% annually 25 ).

¾Network charges continued to be on divergent trajectories for different gas consumer types. They increased annually by 4% for households and 5% medium industrial consumers but decreased by 1.6% annually for large industrial consumers.

¾The impact of taxation on natural gas prices remained limited. Taxes made up 26% of household bills (compared to 40% for electricity) and only 10% of large industrial gas bills.

¾The composition of taxes on natural gas bills also differs significantly from their electricity counterparts. Almost 90% of imposed additions are fiscal instruments generating revenue for the state budget and only a small portion are levies supporting specific policies. Non- harmonized taxes (other than VAT and excise duty) are more common compared to electricity bills.

¾Support to renewable energy has played an insignificant role in gas price developments. By 2017 only 3 countries imposed a RES levy on gas.

¾The EU household gas prices grew by 2.5% annually since 2008 and reached 60 EUR/MWh in 2017. Industrial gas prices evolved in the opposite direction as prices for both small and large consumers contracted over the same period and reached 29 EUR/MWH and 22 EUR/MWh respectively.

Table 4 - Key figures on the evolution and drivers of retail gas prices

Consumer type

Household (D2)

Industrial (I3)

Large Industrial (I5)

Component

Annual growth

Share 2017

∆ Share

Annual growth

Share 2017

∆ Share

Annual growth

Share 2017

∆ Share

Energy

0.5%

49%

- 10 p.p.

-2%

69%

- 14 p.p.

- 1.5%

81%

- 4 p.p.

Network

+ 4.1 %

24%

+ 3 p.p.

+ 4.8 %

18%

+ 7 p.p.

- 1.6%

8%

0

Taxes

+ 5.8 %

26%

+ 7 p.p.

+ 10%

13%

+7 p.p.

+ 5%

11%

+4 p.p.

Total

+ 2.5 %

 

 

- 0.4%

 

 

- 1%

 

 

Source: DG ENER in-house data collection

Scope of the chapter

According to Regulation (EU) 2016/1952 the report analyses prices of natural gas sold to consumers who purchase gas for their own use. Therefore prices paid by consumers who purchase gas for electricity generation in power plants or for non-energy purposes (e.g. for use in the chemicals industry) are excluded.

Box - The role of electricity and natural gas in our energy consumption

This chapter analyses consumer prices of electricity and natural gas. What is the role of these two energy products in our economy and everyday lives? Electricity accounts for 22% of our energy consumption 26 across the whole economy. This share remained remarkably constant over the last decade as a combined result of progressive electrification and increasing energy efficiency at the same time. The share of electricity varies considerably across reporting countries. It ranges from 13% in Lithuania to 34% in Sweden. The highest shares of electricity in final energy consumption are recorded by 2 non- EU member countries: Montenegro and Norway with 34% and 51% respectively.

The high share of electricity in Norway's energy mix results from the abundance of available hydro power resources. About 96% of electricity in Norway is produced from hydro power and most of it is consumed domestically. Relatively low cost and low carbon electricity helped Norway to develop a large power intensive industry. The availability of hydro electricity also impacts household energy consumption: electric heating sector is more wide spread than in other EEA countries. In Montenegro the dominant role of electricity results partially from the fact that there is no natural gas consumption in the country. On the other hand, different end uses of electricity, such as space cooling or electro intensive industries are prevalent.

Natural gas accounted for the exact same share on average across the EU as electricity, namely 22%. The share of natural gas in our final energy consumption has also not changed over the last 10 years. The use of natural gas differs even more across countries than the use of electricity. No gas is used on the islands of Cyprus and Malta. This is reflected in our data. The share of natural gas in final energy consumption of households is the lowest in Sweden at 1 %, closely followed by Finland and Norway with 2% share in both countries. The relative importance of natural gas is the highest in Hungary at 31%, followed by Italy and the United Kingdom with 29% share in both countries. Our analysis of national energy components reflects on the high shares of natural gas in the energy mix of these countries.

Electricity and natural gas together account for less than half of our energy consumption. The rest is covered mainly by liquid fuels for transport and heating as well as biomass.

2.2.1Household Natural Gas Prices

Household gas prices were reported by 23 EU Member States and Turkey. Natural gas is not used on the islands of Malta and Cyprus and in Montenegro. Regulation (EU) 2016/1952 lays down that reporting countries, where natural gas accounts only for an insignificant share of final energy consumption, are exempt from the obligation of providing price data. According to this Finland, where the share of household consumption of gas in final energy consumption is below 1.5%, is not reporting such data. Figures for Greece and the United Kingdom are estimates by DG energy as these two countries have not reported the breakdown of gas prices. No household gas data or estimates are available for Greece after 2015.

The following chapter analyses gas prices paid by household consumers whose annual consumption falls in the range of 20 to 200 GJ. This consumption band is defined by Eurostat as D2. It is the most representative consumption band in all but one reporting country.

Evolution Household Gas Prices

Total prices grew at 2.5% annual rate from 2008 to 2017. In absolute terms the EU price grew from 48 EUR/MWh to 60 EUR/MWh. This growth is faster than inflation, which averaged at 1.2% annually during the same period. The overall increase recorded throughout the period 2008 to 2017 however conceals two distinct periods. Prices steadily grew from 2008 to 2014 but have been increasing ever since. The EU average price peaked at 69 EUR/MWh in 2014 and decreased to 60 EUR/MWh by 2017. This EU average conceals relatively homogenous developments on national level as prices increased only in 10 countries in the last reporting year. In 2017 Romania reported the smallest and Sweden the largest price. The ratio of the largest to smallest price increased by 6% to around 4:1 over the last decade.

Composition of Household Gas Prices

Over time, the composition of prices changed albeit less significantly than in the case of electricity. In 2017 the energy component, which mainly consists of wholesale prices, still made up almost half of the price even after its share decreased by 10 percentage points from 59% to 49% by 2017.

In absolute terms, the energy component decreased at an annual rate of 0.5% and reached 29 EUR/MWh in 2017. This collides with developments on the national level, as the energy component increased in only 6 reporting countries from 2016 to 2017.

The share of the network component increased slightly from 21% to 24% of the total price. In absolute terms the network component grew at the annual rate of 4% and reached 14 EUR/MWh by 2017.

The share of the taxes component grew by almost 7 percentage points and reached 26% in 2017. In absolute terms, taxes grew at the annual rate of 6% and reached almost 15 EUR/MWh by 2017.

The impact of taxes was smaller on household gas prices than on their electricity counterparts as the energy component, driven mainly by international commodity prices, remained the dominant component.

Figure 50 - Composition of the EU household gas price (DC)

Source: DG ENER in-house data collection

Drivers of Household Gas Prices

The EU natural gas price for household consumers peaked in 2014 and has been decreasing ever since. The trend results from a steady decline of the energy component, continued smaller increases in the network component and a volatile evolution of taxes. 2017 was the first year when all three components contracted, leading to the largest year-on-year fall of the total price in our observation period.

Figure 51 - Household gas prices in 2017

Source: DG ENER in-house data collection

Sweden's high gas prices stem from a carbon tax, which aims to curb greenhouse gas emissions. This holds both for households and industrial consumers. The tax was introduced as early as 1991 on all fossil fuels in proportion to their carbon content. Combustion of sustainable biofuels doesn't result in a net increase of carbon in the atmosphere and hence are not subject to the carbon tax.

Relatively high gas prices in Portugal result from a combination of factors, namely the very low unit consumption of households (mild climate), a modern network resulting in higher access tariffs (Portugal’s natural gas industry is only 20 years old), and a higher tax burden than in most other Member States. Additionally, natural gas has a CCGT backup role in Portugal, complementing the relevant, non-dispatchable and renewable-based electricity production. This balancing role led to access tariff volatility, affecting price convergence with other EU countries. This is reflected in our data: in other countries high prices typically stem from high taxes. In Portugal the network component of 30 EUR/MWh is more than double that of the EU average (13 EUR/MWh).

In the Netherlands relatively high prices stem also from taxation. 41 EUR/MWh taxes account for more than half of the total price and are significantly above the 13 EUR/MWh EU average. This tax policy aims to prevent gas field earthquakes at extraction sites through reducing demand. The Netherlands has significant gas resources in the Groningen area, located in north of the country. The extraction of these resources causes seismic activity which in turn causes significant damage to local businesses and homes.

90% of taxes imposed on natural gas bills across all reporting countries, including those in Sweden and the Netherlands, are non- earmarked taxes. Revenues from these taxes feed into the general budget and do not directly support energy or climate policies.

Figure 52 - Composition of household gas prices in 2017

Source: DG ENER in-house data collection

Composition of taxes, levies, fees and charges

Natural gas prices remained less impacted by taxation than their electricity counterparts. Taxes made up 26% of the total price compared to 41% of household electricity prices. The number and composition of taxes imposed on household gas prices also significantly differs from electricity taxation. While Member States add dozens of different policy support costs to electricity prices, the variety of taxes on gas prices is much more limited.

Taxation of household gas prices is dominated by excise duties. Such are imposed according to the Energy Tax Directive from 2003. Excise duties accounted for 40% of all taxes, twice as much as in the case of electricity (20%). Renewable energy support costs, which make up 30% of all taxes on electricity prices, have almost no impact on household gas prices as they account for less than a percent of the total price. Their 0.5% share almost equals the share of social charges and energy efficiency (both below half a percent share). In the case of electricity household prices RES support cost exceed energy efficiency related charges by a factor of 100. While until 2016 all but 3 Member States levied explicit RES support costs on household electricity prices, only 5 did so on household gas prices.

Figure 53 - Composition of EU taxes on household gas prices

Source: DG ENER in-house data collection

2.2.2Industrial Natural Gas Prices

The following chapter compares gas prices paid by industrial consumers with medium versus large annual consumption. Medium industrial consumption is defined as band I3 covering annual consumption volumes between 10 000 and 100 000 GJ. Large consumption is defined as band I5 covering annual consumption between 1 million and 4 million GJ. Median industrial (I3) prices were reported by 24 EU Member States and Turkey. The breakdown of gas prices was not reported by Greece and the United Kingdom. Large industrial prices (I5) were reported by 19 EU Member States and Turkey (in other countries there are either no consumers in this consumption band or data is confidential). Figures for the United Kingdom are Commission estimates both for I3 and I5 prices.

Evolution Industrial Gas Prices

The observation period 2008 to 2017 conceals two distinct periods for both consumer types: Prices grew from 2008 to 2012 and have been on a decreasing trajectory ever since. Looking at the whole of the last decade, the median industrial price decreased at the annual rate of 0.4%. The large industrial price contracted even faster, by 1% annually. Inflation during the same period averaged at 0.5% 27 . In absolute terms the I3 price fell from 30 to 29 EUR/MWh by 2017. The I5 price decreased from 25 to 23 EUR/MWh. In 2017 Belgium reported the smallest I3 price and Sweden the largest. The ratio of the largest to smallest price across the EU increased by 22% since 2008 and reached almost 3:1 by 2017. The largest price for I5 consumers was also reported by Sweden, the smallest by Romania with a ratio of 3:1 in 2017. This ration decreased by 17% over the last decade. The convergence indicates progress towards the completion of the internal energy market.

Composition of Industrial Gas Prices

Over time the composition of industrial gas prices also changed, albeit to a different extent for the two consumer types. In 2008, the first year of our observation period, the energy component accounted for 82% of small and 85% of large industrial prices. The international price of the commodity, complemented by the commercial costs of suppliers, made up most of the final consumer price. The impact of network costs and taxes was limited. The share of the energy component decreased to 68% for small and to 81% for large industrial consumers by 2017. The 13 percentage point decrease in the composition of the I3 price is more significant than the smaller change in the I5 price. In I3 prices, the diminishing share of the energy component was taken over equally by network charges, which grew by almost 5% annually and taxes, which grew by 10% annually. Taxes increased at a higher speed, albeit from an initially lower level, reaching 4 EUR/MWh by 2017. As a result, both components grew in absolute terms by about 2 EUR/MWh in the period 2008-2017. The share of the network and taxes components grew by 6.6% and 7.7% respectively.

The composition of I5 prices changed less significantly as the energy component maintained its dominance accounting for over four-fifths of the price, even as it decreased in absolute terms from 21 to 18 EUR/MWh by 2017. The network component also decreased in absolute terms: large industrial consumers payed 20 Eurocents/MWh less in 2017 than they did in 2008. These network developments hold uniquely for I5 consumers among all analysed electricity and gas bands. As the share of both the energy and network components decreased, the relative weight of taxes grew from 6% to 11%, and reached 2.5 EUR/MWH in absolute terms.

Figure 54 - Composition of EU prices for small (I3) and large (I5) industrial gas consumers

Source: DG ENER in-house data collection

Drivers of Industrial Gas Prices

Industrial gas prices remained dominated by the energy component, which mainly consists of the commodity price. Consequently, consumer prices followed the developments on international gas markets, albeit with a slight time lag. Between 2008 and 2010 the large energy component slightly decreased, while the initially small tax component grew by 25% for I3 consumers. From 2010 to 2012 international commodity prices gradually recovered. As a result, I3 and I5 consumer prices grew by 18% and 25% respectively. Since 2012, both prices have been on a downward trajectory. Decreases are driven by falling energy components complemented by moderate decreases in network components since 2015.

The evolution of the three components for the two consumer types (small and large industrial gas) is fairly similar. One difference is to be observed in the evolution of taxes: for small industrial gas consumers taxes kept increasing throughout the whole observation period while they started to decrease for large consumers in 2015. Price decreases are driven by the continued contraction of the energy component, in the last year accompanied by contractions in all three components for small industrial consumers.

 

Figure 55 - Median (I3) and large (I5) industrial gas prices in 2017 28

Figure 56 - Composition of median (I3) and large (I5) industrial gas prices in 2017 29

Composition of taxes, levies, fees and charges

Gas prices are generally less impacted by policy support costs and fiscal instruments than electricity prices. Also, industrial consumers benefit from exemptions and reduced tax rates in most countries. As a result, taxes accounted for only 13% and 10% of the total price for medium and large consumers respectively. In comparison, taxes made up 26% of the household gas price and 40% of the household electricity price.

Taxes imposed on industrial gas prices consist mostly of excise duty and other non- earmarked taxes that do not support any specific policies (shown as environmental taxes incl. excise duty on our graphs). Non- earmarked taxes made up 89% and 86% of total prices for median and large consumers respectively. The impact of renewable energy support costs remained limited at 4% and 5% of all taxes. 3 countries imposed a levy on median industrial gas prices to support renewable energy and only 1 of all reporting countries (Italy) did so for large industrial consumers.

Several countries however impose taxes that are both non-earmarked and non-harmonized. These taxes are non- earmarked as their revenues do not support any specific policy and non- harmonized, as their minimum levels are not regulated on EU level (minimum levels of excise duty and VAT are set by the Energy Tax Directive of 2003). Imposing non-harmonized taxes on gas prices is more common across the EU than imposing such taxes on electricity prices

Figure 57 - Composition of EU gas prices for median (I3) and large (I5) consumers

Figure 58 - Composition of taxes for median (I3) and large (I5) industrial gas consumers

2.2.3 International comparisons

For a detailed explanation on the purpose and methodology of this analysis, please consult the international comparison section of the electricity chapter.

Household Natural Gas Prices

The EU28 average difference between retail and wholesale prices increased from 30 EUR/MWh in 2008 to 40 EUR/MWh in 2017. The difference in the US is lower, at around 20 EUR/MWh and experienced only a small increase since 2008. The small difference however allows for important conclusions in terms of the role of the components. As US wholesale prices have declined significantly, the share of the other components in the total price must have increased.

The difference in Japan has declined over the period from more than 120 EUR/MWh to around 80 EUR/MWh but remains around twice as high as EU average levels. The price difference in China fluctuates between low and negative levels (+/- 10 EUR/MWh) indicating that Chinese households don’t pay the full cost of their natural gas use. For all other G20 countries) the difference is lower than the EU28 average. In Mexico (MX) and Turkey (TR) there is only a very small difference, indicating that prices are regulated at low levels. The difference in Russia, Canada and Brazil is also lower, in the range of 20 EUR/MWh . Only in South Korea is the difference closer to the EU level. In Argentina the prices difference is negative, highlighting indicating that households pay subsidized prices that do not cover the full costs of their natural gas use.

Figure 59 - Difference between household retail natural gas prices and wholesale prices, EU28 and trading partners

Source: Trinomics et altri study

Industrial Natural Gas Prices

The EU28 average difference between industrial retail and wholesale remained at 5 EUR/MWh between 2008 and 2017. The difference in the US is slightly lower than in the EU28 difference in Japan has been greater than EU, but has substantially converged since 2011. The price difference in China has been greater than in the EU and finished the period at around 15 EUR/MWh. In both Japan and China this reflects both wholesale and retail prices that are similar or higher than in the EU.

Figure 60 - Difference between industrial retail natural gas prices and wholesale prices, EU28 and trading partners

(1)

COM(2016) 860

(2)

 COM(2018) 773

(3)

 COM(2018) 796

(4)

  https://ec.europa.eu/energy/sites/ener/files/documents/platts_report_final_version_rrr.pdf

(5)

At the end of November 2018 the was around 19 €/MtCO2e

(6)

Simple averages of the analysed bands.

(7)

Weighted average “Most representative” price.

(8)

'Industrial' consumers are currently referred to in Eurostat statistics as 'Non-households' consumers

(9)

National tariff equalization enables consumers on isolated island energy systems to pay prices similar to those on mainland France.

(10)

Graphs in the body of the report as well as in the French graphical country profile display a RES levy of around 18 EUR/MWh for households and 3.4 EUR/MWh for large industrial consumers in 2015. This is due to a late data revision request by the French Ministry of Ecology, Sustainable Development and Energy that could no longer be reflected on our graphs. This box contains the updated data.

(11)

Combined excise duty and local tax

(12)

Combined excise duty and local tax

(13)

VAT and non- earmarked taxes are not considered energy policy relevant as their revenues not necessary support energy or climate policies.

(14)

Data is available for the United Kingdom until 2015.

(15)

According to Directive 2008/92/EC of the European Parliament and of the Council: Industrial end-user may include other non-residential users.

(16)

Final Energy Consumption in 2016, ESTAT: nrg_110a

(17)

Eurostat Producer Price Index,   (sts_inppd_a)

(18)

For more information on factors determining the energy component and on non- interconnected island systems, please consult the corresponding section of the household electricity chapter.

(19)

No 2016 data available for Greece and the United Kingdom.

(20)

Data is available for the United Kingdom until 2015. From 2016 onwards Commission estimates are used as substitutes.

(21)

Eurostat Producer Price Index (sts_inpp_ma)

(22)

Source: Eurostat (http://ec.europa.eu/eurostat/web/energy/data/energy-balances)

(23)

Energy prices, costs and subsidies and their impact on Industry and Households (2018) by Trinomics et altri (2018)

(24)

Average of the 3 analysed natural gas consumption bands (D2,I3,I5).

(25)

Average of the two analysed industrial bands (I3,I5).

(26)

Final Energy Consumption in 2016, ESTAT: nrg_110a

(27)

Eurostat Producer Price Index (sts_inpp_a)

(28)

Energy Component for Greece includes also network charges

(29)

Energy Component for Greece includes also network charges

Top

Brussels, 9.1.2019

SWD(2019) 1 final

COMMISSION STAFF WORKING DOCUMENT

Accompanying the document

REPORT FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS

Energy prices and costs in Europe

{COM(2019) 1 final}


3Oil and oil product prices 

Main findings

¾After the dramatic fall seen in 2014-2016, crude oil prices have broadly been rising since mid-2017, driven by robust global demand growth, Middle East tensions, concerns over the impact of a return to US sanctions on Iranian oil, sliding output in Venezuela and the continued OPEC-led output cuts.

¾The crude oil price is the main driver for the development of the wholesale prices of oil products although other factors, like the supply-demand situation in the specific oil product market, refinery maintenance or seasonality can also influence the prices.

¾In addition to the crude oil price, the retail price of oil products is also influenced by the costs of refining and distribution, variations in exchange rates (crude oil is traded in US dollar but the finished products are sold at the pump in euros or other national currencies) and tax rates. In fact, the share of crude oil in the final price can be as low as 25% and, therefore, variations in the price of crude oil have a limited impact on the price at the pump. In contrast, the tax component (excise duty plus VAT) can reach up to 70% of the retail price

¾The high share of taxes and exchange rate developments moderate the pass-through of falling/rising oil prices to the retail prices of oil products in Europe.

¾In mid-2018, retail prices reached the highest levels since 2014-2015

¾There has been some convergence of gasoline and diesel prices, helped by some convergence of the excise duty rates but in several Member States the tax advantage of diesel actually increased.

3.1Crude oil prices

Crude oil prices reached unprecedented levels in 2008, with Brent exceeding 140 USD/bbl at the height of the "commodity super cycle" which was driven by the rising demand from emerging markets, particularly China. The price increase was interrupted by the financial crisis, with a sharp downturn in the second half of 2008. However, as demand recovered, prices began to rise and crossed the 100 USD/bbl level again in early 2011. This was followed by three and a half years of remarkable price stability, with Brent rarely leaving the 100-120 USD/bbl range.

Crude oil prices started to decline in mid-2014, driven by weak demand and robust supply growth, resulting in an oversupplied market. Global oil demand growth has significantly weakened in 2014, mainly because of lower than expected global economic growth and mild winter temperatures.

On the supply side, non-OPEC output showed a robust growth, driven by increasing unconventional oil production in North America. US light tight oil production proved to be rather resilient to low prices: improving efficiency and cost reductions allowed output to continue increasing in spite of the plummeting crude oil prices.

In spite of the falling prices, OPEC countries chose not to cut production in an attempt to maintain market share and to squeeze out high-cost producers. Furthermore, the lifting of the Iranian sanctions in January 2016 allowed Iran to increase its oil exports, adding to an already high OPEC output and further delaying the market rebalancing. OPEC and a few key non-OPEC producers finally agreed in November 2016 to limit their production, in order to accelerate the drawdown of the stock overhang and bring the rebalancing forward.

From a 115 USD/bbl peak in June 2014, Brent dropped to 26 USD/bbl on 20 January 2016, its lowest level since 2003. This means the price decreased by 77% in 19 months.

Despite the November 2016 agreement of OPEC and non-OPEC producers to reduce output, oil prices decreased in the first half of 2017 reflecting increasing production in the US, as well as growing output in Libya and Nigeria which were exempted from the OPEC cut. The rollover of the cut in May 2017 failed to reverse the trend: in the second half of June 2017, the price of Brent dropped below 45 USD/bbl, the lowest level since November 2016.

From mid-2017, however, oil prices have broadly been on the rise, driven by a combination of factors, including the robust growth of global demand, growing tensions in the Middle East, a number of actual supply disruptions (Northern Iraq, hurricanes in North America, closure of the Forties pipeline system in the UK North Sea, a sustained plunge of Venezuelan supply), the weakening of the dollar and a further extension of the OPEC cut in November 2017. In late December and early January, the protests in Iran provided support to prices.

Prices receded in early February 2018 as the market remained well supplied, but the price rise resumed afterwards as growing tensions in Syria and the expectation of the US withdrawal from the Joint Comprehensive Plan of Action (the Iran nuclear deal) raised concerns about future oil supplies. In mid-May 2018, after President Trump announced the re-imposition of US sanctions on Iran, Brent reached 80 USD/bbl, the highest level in three and a half years. Compared to the 44 USD/bbl low on 20 June 2017, Brent increased by more than 75%. Prices continued to rise despite the strengthening of the US dollar which in general is conducive to lower oil prices.

Brent receded to around 75 USD/bbl in late May/early June after Russia and Saudi Arabia indicated they would increase production in the second half of the year. On 22 June, OPEC and non-OPEC producers agreed to do away with the over-compliance with the cuts agreed back in 2016, implying a theoretical output increase of around 1 million barrels per day (mb/d) in the second half of the year. Despite the agreement, prices rose again in late June and early July, supported by production outages in Libya and Canada.

Figure 61 - The Brent crude oil price from 2008 to mid-2018

Source: Platts

The unilateral withdrawal of the US from the Iran nuclear deal casts doubt about the future of Iranian crude exports, at a time when we already see sliding output in Libya and Venezuela, as well as geopolitical risks in other parts of the world. This is expected to further tighten the global market, potentially leading to an additional price rise 1 .

It is far from certain whether OPEC and Russia have the capacity to fill this gap. Even if they do, this will significantly reduce global spare capacity, making the market exposed to any supply disruptions.

3.2Wholesale prices of oil products

Crude oil is the main feedstock to produce oil products and oil product prices closely follow the development of the crude oil price. This is clearly visible if we compare the Brent oil price with the representative wholesale prices of the main oil products in Western Europe.

Figure 62 - Crude oil (Brent) and European wholesale gasoline, diesel and heating oil prices from 2008 to mid-2018

Source: Platts, ECB

The following oil product prices were used: Gasoline Prem Unleaded 10ppmS FOB AR Barge (gasoline), ULSD 10ppmS FOB ARA Barge (diesel) and Gasoil 0.1%S FOB ARA Barge (heating oil)

The following conversion rates were used: crude oil 159 litre/barrel, gasoline 1350 litre/ton, diesel and heating oil 1186 litre/ton.

Looking at the crack spreads (i.e. the differential between the wholesale price of oil products and crude oil), one can see that these spreads are however rather volatile and often follow different paths for different products.

Figure 63 - Crack spreads of gasoline, diesel and heating oil from 2008 to mid-2018

Source: Platts, ECB

Crack spreads are calculated as the difference between the Brent crude oil price and the price of the following products: Gasoline Prem Unleaded 10ppmS FOB AR Barge (gasoline), ULSD 10ppmS FOB ARA Barge (diesel) and Gasoil 0.1%S FOB ARA Barge (heating oil)

The following conversion rates were used: crude oil 159 litre/barrel, gasoline 1350 litre/ton, diesel and heating oil 1186 litre/ton.

The supply-demand conditions of the different products are divergent (both from crude oil and from each other) which will affect their crack spreads. For example, the 2008 oil price rise was very much driven by industrial growth in China, leading to a big increase in the demand of middle distillates which is reflected in the high crack spreads of these products. There are also seasonal differences in demand, for example, gasoline demand is higher in the summer, typically resulting in a relatively high crack spread during that period while in times of low demand crack spreads can even turn negative (implying the gasoline is cheaper than crude oil). In the summer of 2015, gasoline crack spreads reached unusually high levels as low prices boosted gasoline demand.

Oil product supply can also fluctuate, for example as a result of refinery maintenance or natural disasters affecting refinery operations; this will also affect crack spreads. For example, Hurricane Harvey in the US triggered the spike of European gasoline crack spreads in late August 2017.

On Figure 63 one can see that European crack spreads have been relatively high in 2015, averaging 0.08 EUR/litre (around 13 EUR/barrel) for both gasoline and diesel. Afterwards, crack spreads diminished: in the period from the beginning of 2016 to mid-2018, both gasoline and diesel crack spreads averaged 0.06 EUR/litre (less than 10 EUR/barrel).

3.3Retail prices of oil products

In addition to electricity and gas, oil products constitute an important part of the energy costs of both households and industry. Oil products have a dominant role in transport where they have limited alternatives, particularly in road freight, maritime and air transport. In case of space heating, the share of oil products is on a declining trend but in certain Member States they still have an important role in this sector.

The retail price of oil products depends on several factors. Variations in the price of crude oil will obviously have an impact on retail prices but crude oil costs constitute just a part, often a relatively small part, of the final price paid by the consumer. Crude oil is traded in US dollar but the finished products are sold at the pump in euros or other national currencies. Therefore, variations in exchange rates will also influence the crude oil component.

Crude oil has to be refined to produce fuels which can be used in transportation, heating or other uses. After refining, the finished products have to be distributed and sold, typically at petrol stations. Refining and distribution costs are relatively stable and are not proportional to the crude oil price.

A significant part of the price goes to taxes: excise duties, other indirect taxes and VAT. These taxes make an important contribution to the tax revenue of Member States (see Chapter 8.1). In case of motor fuels (gasoline and diesel), taxes typically cover more than half of the final price.

Excise duties are generally a fixed amount per quantity (usually litre or kg), i.e. not influenced by the price of crude oil. VAT, on the other hand, is set as a percentage of the price of the product (including the excise duty) and, therefore, changes in the crude oil price will have an impact on the absolute value of the VAT component.

Rates of both the excise duty and VAT vary by product and by Member State, resulting in significant price differences across Europe. Nevertheless, Member States have no complete freedom when setting the tax rates. The Energy Tax Directive (2003/96/EC) sets minimum excise duty rates for gasoline, gasoil, kerosene, LPG and heavy fuel oil. New Member States were often granted a transition period to reach the minimum level; today, all Member States comply with minimum level.

In case of VAT, the VAT Directive (2006/112/EC) requires that the standard VAT rate must be at least 15%; currently the standard VAT rates applied by Member States range from 17% (in Luxembourg) to 27% (in Hungary). In case of oil products, Member States typically apply the standard VAT rate. Under certain conditions, however, Member States can set a lower VAT rate for specific products and services; for example, a few Member States apply a reduced rate for heating oil.

As the share of crude oil in the final price can be as low as 25%, variations in the price of crude oil will have a limited impact on the price at the pump. In fact, the high share of fixed taxes in the price acts as a buffer: fluctuations in the retail price of oil products (particularly motor fuels) are significantly lower than the fluctuation of the crude oil price. Variations in the exchange rate have a similar effect: the oil price and the value of the US dollar usually move in the opposite direction: a strengthening dollar typically coincides with decreasing oil prices and vice versa. This means that changes in the oil price, whether upwards or downwards, are mitigated by the exchange rate and the volatility of the oil price expressed in euros is smaller than the volatility of the price expressed in dollar.

During the decline of crude oil prices in 2014-2016, the above factors moderated the pass-through to oil product prices in the EU: while crude oil prices (expressed in USD) fell by 77% between mid-2014 and early 2016, in the same period 2 the average EU consumer price of gasoline and diesel decreased by 24% and 28%, respectively. In case of heating oil, the decrease was 45%.

Similarly, the comparably high taxes in the EU mitigated the feed-through of the recent oil price rise: between 3 July 2017 and 11 June 2018, retail prices of gasoline and diesel (including taxes and duties) increased by 12% and 18%, respectively, as compared to a more than 50% increase in international crude oil prices in the same period (measured in USD). In case of heating oil, where the tax component is smaller, the price increase was 32%.

Finally, although their current use is limited, alternative fuels provide an increasing share of the energy mix in transport and their importance is expected to grow in the future. At the same time, as shown by Trinomics et al. (2018), data on retail prices for compressed natural gas (CNG), liquefied natural gas (LNG), liquefied petroleum gas (LPG) and biofuels is not widely available. The growing importance of the market for alternative fuels shows the need of further efforts in collecting such retail prices in the future.

3.3.1Methodology

The analysis in this section is based on the data of the weekly Oil Bulletin. Pursuant to the Council Decision on Crude Oil Supply Costs and the Consumer Prices of Petroleum Products (1999/280/EC), Member States have to report to the Commission the retail prices of the main petroleum products on a weekly basis. Member States also have to report any changes in the tax rates (VAT, excise duty, other indirect taxes) applicable to these products, allowing us to break down the final price to three main components: the net price, excise duty 3 and VAT. The reported data are published on the website of DG Energy. 4

The analysis covers the three main petroleum products sold in the retail sector: gasoline (Euro-super 95), diesel (automotive gas oil) and heating oil (heating gas oil). The time horizon is from 2008 to the first half of 2018. All Member States are covered but data for Croatia is available only from 2013. In case of heating oil, Slovakia does not report prices since October 2011 while, from 2015, Greece does not report prices for the period from May to mid-October.

Prices reported in currencies other than the euro were converted into euro, using the ECB exchange rate of the day for which the price applies.

For each year and each Member State an average price was calculated as an arithmetic average of the weekly prices and an EU average price was calculated as the weighted average of these. In the absence of 2017 and 2018 consumption figures, for these years we used the 2016 consumption data as the weight.

3.3.2General findings

While the absolute level of the prices of the three oil products are different, their development over the last 10 and half years is very similar and basically reflects the evolution of the crude oil price in the same period. The price of all three products decreased significantly in 2009 when oil prices plummeted in the wake of the financial crisis. This was followed by years of gradual increase, with prices peaking in 2012. Prices decreased afterwards, with the decrease accelerating in 2015-2016. As crude oil prices recovered from 2016, oil product prices have been also rising in the last two years.

For comparison, Figure 64 also depicts the evolution of the Brent crude oil price (recalculated into EUR/litre).

Figure 64 - Average retail price of oil products in the EU

Source: Oil Bulletin, DG Energy, Platts

The difference in the absolute price of the three products can be mostly attributed to the diverging tax rates. In practically all Member States, the excise duty rate of gasoline is higher than that of diesel. The Energy Tax Directive also sets a higher minimum rate for gasoline (0.359 EUR/litre) compared to diesel (0.33 EUR/litre). The UK is the only Member State where the two motor fuels are taxed at the same level.

In case of heating oil, a few Member States (Bulgaria, the Czech Republic, Hungary and the Netherlands) apply practically the same excise duty rates than for diesel. In most Member States, however, heating oil is taxed at a lower level. The minimum rate established by the Energy Tax Directive (0.021 EUR/litre) is much lower than those for motor fuels. Ireland, Luxembourg and the UK also apply a reduced VAT rate for heating oil.

Although excise duty rates are set in absolute values, i.e. as a fixed amount per quantity of the product, several Member States increased the tax rates over the period, resulting in a gradually increasing (weighted) average tax rate. According to the Energy Tax Directive, the minimum excise duty rate for diesel increased from 0.302 EUR/litre to 0.33 EUR/litre on 1 January 2010, requiring some Member States to adjust their rates.

Contrary to the general trend, the weighted average excise duty rate for gasoline slightly decreased in 2016 and 2017. While a few Member States indeed reduced the excise duty rate for gasoline in this period, the decrease was driven mainly by exchange rate developments, in particular the depreciation of the pound sterling which made the UK excise duty (unchanged in the local currency) significantly lower when expressed in euros.

Figure 65 - Average excise duty rates for oil products in the EU

Source: Oil Bulletin, DG Energy

If the net price of the three products is compared, the difference is significantly lower. In fact, during the whole period the net price of diesel is slightly higher than that of gasoline.

Figure 66 also depicts the evolution of the Brent crude oil price (recalculated into EUR/litre), showing that crude oil is clearly the main component of the net price. Over the period, crude oil price represented on average 65-70% of the net price of gasoline and diesel but in 2015-2016, as crude oil prices dropped significantly, this share dropped below 60%.

Figure 66 - Average retail price of oil products in the EU, without taxes

Source: Oil Bulletin, DG Energy

3.3.3Gasoline

In most Member States, the evolution of gasoline prices clearly followed the trend of the crude oil price but there have been considerable differences in the absolute level, mainly explained by the diverging excise duty and VAT rates. Average prices moved in a relatively wide range, with the difference between the highest and lowest price being about 0.5 EUR/litre. This range has slightly narrowed between 2008 and 2014, from 0.52 EUR/litre to 0.45 EUR/litre, indicating some degree of price convergence. However, the range widened afterwards, reaching 0.54 EUR/litre in the first half of 2018.

Greece showcased the biggest relative increase in gasoline prices: while in 2008-2009 Greek prices were well below the EU average, since 2011 they are among the highest, mainly as a result of the sharp increase of the excise duty rate. In the first half of 2018, the EU average gasoline price was 6% higher than in 2008; in case of Greece, the increase was 39%. At the other end of the spectrum, prices in Poland decreased by 8%, mainly because of the depreciation of the national currency in the second half of 2008 (measured in Polish zloty, the average price increased).

Figure 67 - The retail price of gasoline in the EU

Source: Oil Bulletin, DG Energy

Looking at net prices, the dispersion is smaller, the difference between the highest and the lowest price is typically between 0.10 and 0.15 EUR/litre. The net price depends on a number of factors, including the source of supply (local refinery or import), industry structure and competition. In the first half of 2018, the highest net price was reported by Denmark while the lowest by the UK. Comparing the average net price with a representative wholesale price (Platts Gasoline Prem Unleaded 10ppmS FOB AR Barge), the difference is relatively stable, averaging 0.13 EUR/litre over the period.

Figure 68 - The retail price of gasoline in the EU, without taxes

Source: Oil Bulletin, DG Energy, Platts

The wholesale price is Gasoline Prem Unleaded 10ppmS FOB AR Barge reported by Platts

Excise duty is an important component of the retail gasoline price; in the first half of 2018, in half of the Member State it actually exceeded the net price. Over the years, we see a gradual increase of the average excise duty rate, with a slight decrease in 2016 and 2017. While in 2008 the average excise duty rate was 0.56 EUR/litre, by 2015 it increased to 0.64 EUR/litre. In the first half of 2018 the average rate was 0.62 EUR/litre.

The average VAT rate also increased during this period, from 19.3% in 2005 to 21.0% in 2014. Since then, the average VAT rate has not changed.

In most Member States, excise duty rates increased between 2008 and the first half of 2018, with the biggest increases in Greece (98%), Latvia (63%) and Cyprus (58%). Germany and Luxembourg are notable exceptions: in these countries, the excise duty rate for gasoline has not changed since 2003 and 2007, respectively. In Hungary and Poland, the excise duty rate measured in euro was lower in the first half of 2018 than in 2008, mainly because of exchange rate developments (in national currencies, the excise duty rates increased over this period). In 2015, the UK had the highest excise duty in the EU but since then, due to the depreciation of the pound sterling, the excise duty measured in euro has significantly decreased.

For most of the study period, the Netherlands applied the highest excise duty rate for gasoline while Bulgaria had the lowest rate, just above the minimum level prescribed by the Energy Tax Directive.

Figure 69 - The exercise duty rate of gasoline in the EU

Source: Oil Bulletin, DG EnergyIn 2014-2016, in line with the decreasing oil prices, the average retail price of gasoline decreased. However, because of the fixed (or, in case of several member States, increasing) excise duty rates, the share of the tax component gradually increased, from 55% in 2012 to 66% in 2016. In absolute terms, the tax component decreased from 0.90 EUR/litre in 2012 to 0.85 EUR/litre in 2016.

The average gasoline price increased in both 2017 and the first half of 2018 but remained well below the record level reached in 2012. In the first half of 2018, the average price was 1.40 EUR/litre, composed of a 0.54 EUR/litre net price (38%), 0.62 EUR/litre excise duty (44%) and 0.24 EUR/litre (17%) VAT.

Figure 70 - Average retail price of gasoline in the EU by price component

Source: Oil Bulletin, DG Energy

The next graph shows the composition of the average gasoline price by Member State in the first half of 2018.

Figure 71 - Average retail price of gasoline in the first half of 2018 by Member State and price component

Source: Oil Bulletin, DG Energy

3.3.4Diesel

Similarly to gasoline, the evolution of diesel prices clearly followed the trend of the crude oil price, with considerable differences in the absolute level, mainly explained by the diverging excise duty and VAT rates. Average prices moved in a relatively wide range and, contrary to gasoline, this range has widened between 2008 and 2015: it was 0.40 EUR/litre in 2008 but grew to 0.56 EUR/litre in 2015. In 2016-2018, the range has significantly narrowed, mainly because of the decrease of UK prices measured in euros.

If the three most expensive countries (Italy, Sweden and the UK) were disregarded, the range would be considerably narrower. In 2015, the UK was by far the most expensive, 0.18 EUR/litre above the second most expensive country, Italy. However, the depreciation of the pound sterling in 2016-2017 had a negative impact on UK prices measured in euros and, as a result, it was "only" the third most expensive country in the first half of 2018.

Cyprus experienced the biggest relative increase in diesel prices: in 2008 it had the lowest price in the EU but after significant increases in the excise duty rate the price reached the EU average by 2013. In the first half of 2018, EU average diesel prices were 1% higher than in 2008; in case of Cyprus, the price increased by 16%. In Slovakia, in turn, the price fell by 9% in this period, helped by a reduction of the excise duty rate introduced in 2010.

Figure 72 - The retail price of diesel in the EU

Source: Oil Bulletin, DG Energy

In case of net prices, the difference between the highest and the lowest price has been 0.10-0.12 EUR/litre in 2008-2014 but significantly increased afterwards, reaching 0.20 EUR/litre in the first half of 2018. The widening range was largely driven by a robust price increase in Sweden which is now by fare the most expensive country in terms of net prices. In the first half of 2018, somewhat surprisingly, the lowest net price was reported in Malta which in 2015 was the country with the highest net price. Comparing the EU average net price with a representative wholesale price (Platts ULSD 10ppmS FOB ARA Barge), the difference has increased from 0.11 EUR/litre in 2010 to 0.14 EUR/litre in the first half of 2018.

Figure 73 - The retail price of diesel in the EU, without taxes

Source: Oil Bulletin, DG Energy, Platts. The wholesale price is Gasoline Prem Unleaded 10ppmS FOB AR Barge reported by Platts

The average excise duty rate of diesel increased from 0.41 EUR/litre in 2008 to 0.50 EUR/litre in the first half of 2018 (an increase of 22% in 10 years). This increase is faster than in case of gasoline and, as a result, the difference between average excise duty rate of gasoline and diesel slightly narrowed: from 0.16 EUR/lire in 2010-2011, it decreased to 0.12 EUR/litre in the first half of 2018.

The average VAT rate of diesel also increased during the study period, from 19.1% in 2008 to 20.9% in 2014. Between 2014 and the first half of 2018, the average VAT rate for diesel has not changed.

With two exceptions, excise duty rates increased in all Member States between 2008 and the first half of 2018, with the biggest increases in Cyprus (80%), Belgium (76%) and Slovenia (64%). In Germany, the excise duty rate for diesel has not changed since 2003 (similarly to the excise duty of gasoline). Slovakia is the only country where the excise duty was lower in the first half of 2018 than in 2008, as a result of a cut in the rate in 2010.

The excise duty rate applied by the UK and Italy is significantly higher than in the rest of the countries. Even after the depreciation of the pound sterling in 2016-2017, the UK excise duty rate remained the highest in the EU. In contrast, Bulgaria imposes a rate at the minimum level prescribed by the Energy Tax Directive.

Figure 74 - The exercise duty rate of diesel in the EU

Source: Oil Bulletin, DG Energy

In 2012-2016, the average retail price of diesel decreased, with the share of the tax component increasing from 48% in 2012 to 61% in 2016. In absolute terms, the tax component decreased from 0.72 EUR/litre in 2012 to 0.68 EUR/litre in 2016.

Since 2016, the average retail price of diesel has been on the rise but remained well below the record level reached in 2012. In the first half of 2018, the average price was 1.30 EUR/litre, composed of a 0.57 EUR/litre net price (44%), 0.50 EUR/litre excise duty (39%) and 0.22 EUR/litre (17%) VAT. Compared to 2016, the absolute value of the tax component increased to 0.73 EUR/litre but its share in the total price fell to 56%.

Figure 75 - Average retail price of diesel in the EU by price component

Source: Oil Bulletin, DG Energy

The next graph shows the composition of the average diesel price by Member State in the first half of 2018.

Figure 76 - Average retail price of diesel in the first half of 2018 by Member State and price component

Source: Oil Bulletin, DG Energy

3.3.5Heating oil

The large differences in the excise duty rates result in a wide dispersion of heating oil prices across the EU. The difference between the highest and lowest price increased from 0.60 EUR/litre in 2008 to 0.79 EUR/litre in 2014 but decreased to 0.67 EUR/litre in the first half of 2018. In the most expensive Member State, Denmark, the price in the first half of 2018 was 109% higher than in the cheapest Member State, Luxembourg. Many of the most expensive countries have a rather low level of heating oil consumption. Germany is by far the biggest consumer of heating oil in the EU and its price has been consistently below the EU average.

Bulgaria experienced the biggest relative increase in heating oil prices: in 2008 its price was well below the EU average but today it is considerably higher. In the first half of 2018, the EU average heating oil price was 9% lower than in 2008; in case of Bulgaria, the price increased by 33%. Ireland experienced the biggest price drop between 2008 and the first half of 2018, 23%.

During most of the study period, Denmark had the highest heating oil prices in the EU, driven by a high excise duty.

Figure 77 - The retail price of heating oil in the EU

Source: Oil Bulletin, DG Energy

The difference between the highest and the lowest price is rather high also in case of net prices (0.21-0.43 EUR/litre), significantly higher than for motor fuels. The gap significantly increased until 2014 but narrowed afterwards.

Denmark had the highest net price of heating oil in the first half of 2018; the lowest net price was reported in the Netherlands. Comparing the EU average net price with a representative wholesale price (Platts Gasoil 0.1%S FOB ARA Barge), the difference has been stable in the 0.10-0.12 EUR/litre range. Curiously, the Dutch price is lower than the wholesale price.

Figure 78 - The retail price of heating oil in the EU, without taxes

Source: Oil Bulletin, DG Energy, Platts

The wholesale price is Gasoline Prem Unleaded 10ppmS FOB AR Barge reported by Platts

The average excise duty rate of heating oil increased from 0.08 EUR/litre in 2008 to 0.12 EUR/litre in the first half of 2018. Although most Member States apply a higher rate, the main consumer of heating oil, Germany, has an excise duty of only 0.06 EUR/litre.

The average VAT rate of heating oil also increased during this period, from 19.3% to 20.1%.

Several Member States increased the excise duty rate between 2008 and the first half of 2018, but in a couple of countries (Austria, Germany, Italy, Lithuania and Luxembourg) it remained unchanged. Bulgaria significantly increased the excise duty rate in 2011 but returned to the previous, lower rate the following year; the rate was increased again in 2016. The Netherlands has the highest excise duty rate (0.50 EUR/litre in the first half of 2018) – it is one of the few countries that apply the same rate for diesel and heating oil. Luxembourg reports the lowest excise duty rate, 0.01 EUR/litre. The rates applied by Belgium and Luxembourg are lower than the minimum level set by the Energy Tax Directive (0.021 EUR/litre); Lithuania uses the minimum level.

Figure 79 - The exercise duty rate of heating oil in the EU

Source: Oil Bulletin, DG Energy

The average retail price of heating oil significantly decreased between 2012-2016, with the tax component increasing from 26% in 2012 to 34% in 2016. In absolute terms, the tax component decreased from 0.26 EUR/litre in 2012 to 0.21 EUR/litre in 2016.

Prices have been rising since 2016 but remain well below the levels seen in 2012. In the first half of 2018, the average price was 0.78 EUR/litre, composed of a 0.53 EUR/litre net price (68%), 0.12 EUR/litre excise duty (15%) and 0.13 EUR/litre (17%) VAT.

Because of the low average level of the excise duty, the tax component of the average heating oil price is much lower than for gasoline and diesel.

Figure 80 - Average retail price of heating oil in the EU by price component

Source: Oil Bulletin, DG Energy

The next graph shows the composition of the average heating oil price by Member State in the first half of 2018.

Figure 81 - Average retail price of heating oil in the first half of 2018 by Member State and price component

Source: Oil Bulletin, DG Energy

3.3.6Gasoline vs diesel

The unequal tax treatment of the main motor fuels, gasoline and diesel, has been a contentious policy issue and was often blamed for the "dieselisation" of the European vehicle fleet. Most Member States impose a lower level of excise duty for diesel than for gasoil, resulting in a lower retail price, in spite of the fact that the wholesale price of diesel is typically slightly higher than that of gasoline. The price advantage of diesel, coupled with the improving fuel economy of diesel engines, made diesel cars increasingly popular in the passenger car and light duty vehicle segments, with their share from new registration reaching up to 70-80% in certain Member States. In contrast, in other regions of the world gasoline-engine cars continued to have a dominant role in the passenger car fleet. The dieselisation significantly contributed to the gasoline/diesel imbalance: European refineries produce too much gasoline which has to be exported while diesel output is insufficient to meet demand is Europe has to rely on imports.

More recently, the Volkswagen emission scandal which broke out in September 2015 put diesel-engine cars in the spotlight and raised renewed questions about the tax advantage of diesel.

Back in 2011, the Commission made an attempt to remove the distortive tax treatment of the two fuels in the proposed revision of the Energy Taxation Directive. 5 According to the proposal, the minimum tax rates of energy products would have been based on the energy content and the CO2 content of the fuel, resulting in a lower minimum rate for gasoline (diesel has a higher energy and CO2 content per litre). However, following the unsuccessful negotiations between Member States in the Council, the proposal was withdrawn.

In this section we compare the development gasoline and diesel prices in the EU and try to investigate whether there has been an approximation of excise duty rates imposed on the two fuels.

Over the last ten years, the average retail price of gasoline has been consistently above the price of diesel, with the difference averaging 0.13 EUR/litre in this period. The difference peaked in 2016 at 0.17 EUR/litre but, since then, has noticeably decreased: in the first half of 2018 it averaged 0.10 EUR/litre, the lowest level since 2018.

Figure 82 - Average retail price of gasoline and diesel in the EU

Source: Oil Bulletin, DG Energy

When comparing the prices without taxes, it is striking that diesel prices are actually higher than gasoline prices. The only exception is 2016 when the average gasoline and diesel price was practically identical. In this year, global gasoline demand was supported by record-low oil prices, resulting in a relatively high gasoline price. Over the ten and half year period, the net price of diesel was on average 0.04 EUR/litre higher.

Figure 83 - Average retail price of gasoline and diesel in the EU, without taxes

Source: Oil Bulletin, DG Energy, Platts

The average excise duty rate for gasoline has been 0.15 EUR/litre over the period, more than offsetting the lower net price of gasoline. The difference was largest in 2010 (0.16 EUR/litre) but since then there has been a clear declining trend, with the average difference dropping to 0.12 EUR/litre in the first half of 2018.

Figure 84 - Average excise duty rates for gasoline and diesel in the EU

Source: Oil Bulletin, DG Energy



In addition to the absolute difference, the relative (percentage) difference between gasoline and diesel excise duty rates also shows a decreasing trend: while in 2010 the excise duty on gasoline was on average 37% higher, by the first half of 2018 this difference decreased to 24%.

Figure 85 - The difference between the average excise duty rate on gasoline and diesel

Source: Oil Bulletin, DG Energy

In most Member States, excise duty rates increased for both gasoline and diesel in the last ten years. In case of gasoline, the average EU rate grew from 0.56 EUR/litre to 0.62 EUR/litre (+10%) while for diesel the average rate increased from 0.41 EUR/litre to 0.50 EUR/litre between 2008 and the first half of 2018. The faster growth of the diesel rate means that the difference has gradually diminished. Nevertheless, there is still only one Member State, the UK, which applies the same rate for the two fuels.

Figure 86 - Excise duty rates in individual Member States in 2008 and the first half of 2018

Source: Oil Bulletin, DG Energy

At EU level, the difference between the average gasoline and diesel excise duty rates decreased from 0.15 EUR/litre in 2008 to 0.12 EUR/litre in the first half of 2018. Looking at Member States, we can see that the difference decreased in only half of the Member States. In 12 Member States the absolute difference has actually increased, implying a growing tax advantage for diesel. For example, in Greece the gasoline excise duty rate has almost doubled (+98%) in this period while that of diesel grew by "only" 39%.

Figure 87 - the change of the difference between the gasoline and diesel excise duty rates between 2008 and the first half of 2018

Source: Oil Bulletin, DG Energy

In recent years, Belgium made the biggest step to remove the tax advantage of diesel: since 2016, the excise duty rate for diel has been gradually raised and by mid-2018 the difference between the gasoline and diesel rate dropped to 0.03 EUR/litre. As a result the difference between the retail price of the two fuels has practically disappeared.

Figure 88 - Excise duty rates for motor fuels in Belgium

Source: Oil Bulletin, DG Energy

3.3.7International comparison

This section is relying on the price data collected by Trinomics and covering G20 economies. 6

Comparing the average retail price of motor fuels in the EU with prices in other G20 countries reveals that the trajectory of prices is in general very similar, basically following the development of crude oil prices. However, there can be significant differences in the absolute level of prices which are largely affected by taxes.

In case of gasoline, retail prices in most G20 countries are lower than the EU average. The retail price in the US is typically less than half of the EU average level. While in the EU the tax component is on average about 60% of the final price, this share in the US is only around 25%. Excluding taxes, EU and US prices are comparable. A few G20 countries had higher prices than the EU average for most of the period, in particular Korea and Turkey, but even these have converged to the EU average level over the last decade.

To sum up, differences in tax treatment are instrumental in explaining the price differences across G20 countries. EU taxes on fuels are among the highest globally, resulting in a high retail price compared to most G20 countries.

Figure 89 - International comparison of retail gasoline prices

Source: Oil Bulletin, DG Energy; IEA, GIZ

Note: prices are expressed in real (2017) euros; dotted line highlights that it is unclear if the excluding taxes price actually excludes relevant taxes

Figure 90 - International comparison of retail gasoline prices

Source: Oil Bulletin, DG Energy; IEA, GIZ

Note: prices are expressed in real (2017) euros; dotted line highlights that it is unclear if the excluding taxes price actually excludes relevant taxes

For diesel, the price is similar: the EU average price is one of the highest among the G20 countries. This is explained by a high tax component which on average constitutes about 50% of the final price. The retail price in the US, where the share of the tax component is only about 25%, is less than half of the EU average. Excluding taxes, EU prices are very similar to those in the US and lower than those in the majority of G20 countries. Turkey is the country which had a consistently higher price than the EU average for most of the period but the difference has largely disappeared by 2015-2016.

Similarly to gasoline, differences in tax treatment are instrumental in explaining the price differences across G20 countries. EU taxes on fuels are among the highest globally, resulting in a high retail price compared to most G20 countries, in spite of the relatively low net price.

The EU is not the only region with gasoline retail prices exceeding diesel prices. This is the case in practically all G20 economies.

Figure 91 - International comparison of retail diesel prices

Source: Oil Bulletin, DG Energy; IEA, GIZ

Note: prices are expressed in real (2017) euros

Figure 92 - International comparison of retail diesel prices

Source: Oil Bulletin, DG Energy; IEA, GIZ

Note: prices are expressed in real (2017) euros; dotted line highlights that it is unclear if the excluding taxes price actually excludes relevant taxes



PART II

ENERGY COSTS for the economy, households and industry



4The EU energy bill 

In this chapter we outline the main drivers of the import bill and estimate its size in the last couple of years.

Main findings

¾High import dependency means that the EU faces an important energy import bill.

¾The price of oil, gas and coal decreased significantly in 2014-2016, resulting in a decreasing import bill. After bottoming out in 2016, energy commodity prices and the import bill have been on the rise

¾In 2013, the EU's estimated import bill reached EUR 400 billion. In 2013-2016, falling energy prices allowed the import bill to decrease significantly, although the weakening of the euro has partly offset this effect. In 3 years, the import bill has almost halved, thereby giving a boost to the economy.

¾The prices of all three fuels increased in 2017, resulting in a growing import bill, but still well below the 2013 level: in 2017, the estimated import bill amounted to EUR 266 billion, 26% more than in 2016 but 34% less than in 2013.

¾The increase in crude oil prices in 2018 could result in lower growth and higher inflation. Crude oil prices assumed at 75$/bbl on average in 2018 would result in an economic growth being 0.4% lower in the EU and an inflation rate higher by 0.6% than the baseline assumption with oil prices remaining at the level of 2017.

¾Crude oil is by far the main component of the import bill, making up 68% of the total in 2017. The share of gas and hard coal was 28% and 4%, respectively.

4.1Introduction 

The EU is a net importer of energy: in 2016, the import dependency 7 stood at 53.6%, practically the same as two years earlier. This means that the EU needs to import just over half of the energy it consumes. Import dependency is particularly high in case of fossil fuels: in 2016, it was 87.4% for crude oil and NGL, 70.4% for natural gas and 40.2% for solid fuels (from which 61.2% for hard coal).

Between 2014 and 2016, import dependency increased for gas (because of rising consumption and falling indigenous production) but decreased for solid fuels (the consumption of which decreased to a larger extent than production). The import dependency for oil has not changed significantly.

EU energy import dependency seems to have stabilised in recent years: since 2005, it has been fluctuating between 52% and 55%. While the import dependency of fossil fuels shows a long-term increasing trend, their share within the energy mix is gradually decreasing. The share of renewables, on the other hand, is steadily growing and these are typically produced within the EU.

Figure 93 - EU import dependency by fuel

Source: Eurostat

The high import dependency poses significant challenges in terms of energy security and the diversification of suppliers and supply routes but, in addition, it also means that the EU is facing an important energy import bill.

4.2Methodology

Scope

In this analysis, we focus on the import bill of the EU as a whole, therefore only extra-EU imports are considered. (When the import bill of an individual Member State is looked at, it is of course reasonable to take all imports into account, including those coming from other Member States.)

The analysis covers the main fossil fuels: crude oil, natural gas and solid fuels. These fuels still cover nearly three-quarters of the EU's gross inland energy consumption and the overwhelming majority (98% in 2016) of net energy imports. Crude oil alone makes up well over half of the EU's net energy imports while gas accounts for 30%.

Figure 94 - EU net imports of energy in 2016 (mtoe)

Source: Eurostat

In addition to crude oil, the EU is also an importer of petroleum products. However, considering the practical difficulties of finding reliable volume and price data for a multitude of products with different specifications and the fact that the EU is also exporting petroleum products and exports and imports are of a similar magnitude (the EU typically exports motor gasoline and imports middle distillates), petroleum products were not included in the calculation of the import bill.

Lignite/brown coal is typically not traded internationally and the imports arriving to the EU are negligible. Therefore, the analysis of solid fuels was restricted to hard coal.

In terms of time horizon, we provide import bill estimates for the period 2013-2017.

Data sources

In case of oil, we are in comfortable position as Member States report on a monthly basis the volume and the average CIF price 8 of imported oil under Regulation (EC) No 2964/95 of 20 December 1995 introducing registration for crude oil imports and deliveries in the Community. 9 Every year, the collected and aggregated information is published on the website of DG Energy. 10

For gas, the import volumes used are from the Transparency Platform of the European Network of Transmission System Operators for Gas (ENTSO-G) 11 which is based on the gas flows reported by gas transmission system operators. Gas imports arrive to the EU from Russia, Norway, Algeria and Libya through several pipelines while, in 2017, LNG was arriving from 12 supplying countries to around 25 terminals in 13 Member States. 12 Volumes were calculated by adding the gas flows at the relevant entry points to the EU gas network.

Gas import prices can vary across Member States depending on the supplier, the supply route, the type of contracts (spot or long-term), the way of pricing (hub-based or oil-indexed) and the level of competition. Based on available sources, including customs data, national agencies (e.g. BAFA in Germany) and commercial data providers, for each supplier (Russia, Norway, Algeria, Libya and LNG) and for each year an estimated average price was established.

Table 5 - Estimated average gas import prices by supplier (€/MWh)

Year

Russia

Norway

Algeria

Libya

LNG

2013

30.0

25.0

30.0

31.0

28.5

2014

25.5

20.0

27.5

29.5

25.5

2015

22.0

19.5

23.5

23.5

20.5

2016

16.0

14.0

16.0

14.5

15.5

2017

17.5

17.5

18.0

15.5

18.5

Source: DG Energy estimation

In case of coal, volumes are the imports of hard coal 13 , reported in Eurostat annual (2013-2016) and monthly (2017) statistics. For price, the CIF ARA spot price reported by Platts was used; this is deemed to be representative for most of the hard coal imports arriving to the EU.

For the conversion from US dollars to euros, we used the annual average of the daily official exchange rates published by the European Central Bank 14 : 1.3281 in 2013, 1.3285 in 2014, 1.1095 in 2015, 1.1069 in 2016 and 1.1297 in 2017.

4.3Drivers

The import bill basically depends on the volume and the average price of imports. Like most commodities, energy sources are typically traded in US dollars and therefore the development of the USD/EUR exchange rate will also influence the import bill (if expressed in euros).

Volumes

Import volumes will depend mainly on the level of consumption. In addition, the development of indigenous production (falling production results in increasing import dependency even if consumption is unchanged) and, to a smaller extent, stock changes can also affect import volumes. In principle, exports can also influence import volumes (higher exports has to be offset by higher imports) but extra-EU exports of crude oil, natural gas and coal are negligible.

Figure 95 - EU net imports

Source: Eurostat

EU imports of fossil fuels showed a marked increasing trend during the 1990s and for most of the 2000s. Since then, the tendencies of the different fuels are diverging.

In case of oil, imports have been decreasing since 2008 but bounced back in 2015, as the significant fall of oil prices triggered an increase in fuel demand.

Gas imports decreased in 2010-2014 when this fuel lost ground in the electricity sector where it had to face increasing competition from renewables and coal. The trend turned after 2014 as increasing gas consumption and the ongoing fall of indigenous production increased import needs.

In case of hard coal, imports increased from 2009-2010, helped by low prices (cheap shale gas squeezed out the fuel from the US power sector), coupled with the low carbon prices. In 2013-2014, the trend reversed and coal imports started to fall again. The competitiveness of gas has improved compared to coal and, in addition, many Member States announced plans to phase out coals.

Prices

International commodity prices generally decreased in 2014-2016 and have been rising since 2016. There is a strong correlation between international commodity prices; in particular, one can observe a strong correlation between Brent and TTF (the Dutch gas benchmark) since 2015.

In the short run, changes in the import volumes are usually moderate but prices can be rather volatile. For example, the price of oil fell by more than 70% between mid-2014 and early 2016, whereas coal prices have more than doubled between the beginning and the end of 2016.

 

Figure 96 - Comparison of European oil, gas and coal prices

Source: Platts; GCI is the North West Europe Gas Contract Indicator, a theoretical index showing what a gas price linked 100% to oil would be

As the EU is net crude oil importer, price volatility impacts the energy expenditure of EU consumers and at macroeconomic level the impact can be tracked in economic growth and in the inflation. According to an analysis carried out by the European Commission, in 2015 and 2016 decreasing oil prices resulted in an additional GDP growth of 0.8% and 0.5%, respectively. As since crude oil prices started to rise again, an opposite impact is anticipated.

If prices are estimated to be 75 USD/barrel on average in 2018, being measurably higher than the average of 2017 (54 USD/barrel), real GDP in 2018 is predicted to be 0.4% below a baseline reflecting constant 2017 oil prices. Compared to the baseline, rising production costs together with the direct effect of higher oil prices on consumer prices are expected to translate into an overall increase in the consumer price index (CPI) inflation by 0.6 percentage points in 2018.

Exchange rates

Most energy is traded in US dollars. Accordingly, the fluctuations of the USD/EUR exchange rate can directly affect the prices and the import bill when these are measured in euros.

Historically, there has been a consistently negative correlation between oil prices and the US dollar, although recently, with the decline of US oil imports, the relationship has weakened. In other words, it can be observed that the price of oil and the value of the US dollar generally move in an opposite direction: a strengthening dollar typically coincides with decreasing oil prices and vice versa. This means that changes in the oil price, whether upwards or downwards, are mitigated by the exchange rate and the volatility of the oil price expressed in euros is smaller than the volatility of the price expressed in dollars. In view of the correlation between oil, gas and coal prices, to a certain extent this is true for coal and gas prices, too.

The euro has considerably weakened compared to the US dollar in the second half of 2014: the exchange rate went down from nearly 1.40 USD/EUR in early May 2014 to 1.06 in March 2015, a depreciation of 24% in 10 months. In spite of the weakening of the euro in the second half of 2014, the 2014 average exchange rate was practically the same as in 2013, 1.33, but in 2015 it decreased to 1.11.

In 2015-2016, the exchange rate had been rather stable, moving in the 1.05-1.15 range during most of this period. The annual average was 1.11 in both 2015 and 2016.

Throughout 2017, the Euro strengthened compared to the US dollar but the annual average exchange rate was only slightly higher (1.13) than in the previous two years.

 

Figure 97 - The USD/EUR exchange rate since 2013

Source: ECB

The red dotted lines represent the annual average in 2013, 2014, 2015, 2016 and 2017

The European Union has a strong intention to "do more to allow the euro to play its full role on the international scene" 15 . As the EU is a net importer of petroleum products, gas and coal, broader deployment of euro in the international trade of these energy products could eliminate the risk of price volatility stemming from the fluctuation of euro against other major currencies, such as the US Dollar.


4.4Import bill calculation

Oil

Table 6 - EU crude oil import bill in 2013-2017

2013

2014

2015*

2016*

2017*

Volume (million bbl/day)

9.83

10.01

10.48

10.29

10.53

Average Brent price (USD/bbl)

108.66

98.95

52.39

43.73

54.19

Average CIF import price (USD/bbl)

108.83

98.65

51.72

42.11

53.16

EUR/USD exchange rate

1.3281

1.3285

1.1095

1.1069

1.1297

Import bill (bn USD)

390.6

360.4

197.8

158.6

204.3

Import bill (bn EUR)

294.1

271.3

178.3

143.2

180.8

Source: DG Energy, based on Member State reports under Regulation (EC) No 2964/95, Platts, ECB

*for confidentiality reason, from 2015 figures do not include the Czech Republic (in 2014, imports by the Czech Republic made up around 1.5% of total EU imports, implying an estimated annual import bill of 2-3 billion euros in 2015-2017)

In spite of the growing import volumes, the EU oil import bill significantly decreased in 2014-2016 as a result of the oil price fall. While in 2013 the oil import bill was close to USD 400 billion, in 2016 it dropped below USD 160 billion, a decrease of almost 60% within three years. The depreciation of the euro in the same period mitigated this trend: measured in euros, the import bill decreased from EUR 294 billion in 2013 to EUR 143 billion euros in 2016, a decrease of 51%.

2017 was the first year since 2012 when the average Brent price increased: it was 54 USD/bbl, 24% more than in 2016. The volume of daily imports also rose (by 2.3%), helped by falling indigenous production, rising fuel consumption and a relatively good refining environment. Driven mainly by the increasing oil prices, the EU's oil import bill increased from EUR 143 billion in 2016 to EUR 181 billion in 2017 (an increase of around 26%) but remained well below the level observed in 2013, the last year before the oil price fall. The euro slightly strengthened in 2017, which moderated the increase of the oil price bill.

Gas

Table 7 - EU gas import bill in 2013-2017

2013

2014

2015

2016

2017

Volume (TWh)

3 390

3 113

3 445

3 853

4 238

Estimated average import price (€/MWh )

28.1

23.6

21.0

15.2

17.7

Import bill (bn EUR)

95.4

73.5

72.4

58.7

74.8

Source: ENTSO-G, DG Energy estimations

Gas imports showed a robust increase in 2015, 2016 and 2017 but prices had been on the decline, bottoming out in 2016 and increasing in 2017. In spite of the rising volumes, the estimated import bill decreased in 2014, 2015 and 2016 (as a result of the falling prices) but bounced back in 2017 when both import volumes and prices increased.

Between 2013 and 2016, the estimated gas import bill decreased by 38%, from EUR 95.4 billion to EUR 58.7 billion. In 2017, the gas import bill increased by 27%, reaching EUR 74.8 billion.

Coal

Table 8 - EU hard coal import bill in 2013-2017

2013

2014

2015

2016

2017

Volume (million tons)

230.0

226.6

210.8

183.4

140.0

CIF ARA spot price (USD/ton)

81.57

75.20

56.84

60.18

84.73

EUR/USD exchange rate

1.3281

1.3285

1.1095

1.1069

1.1297

CIF ARA spot price (EUR/ton)

61.41

56.63

51.25

54.37

75.00

Import bill (bn USD)

18.8

17.1

12.0

11.0

11.9

Import bill (bn EUR)

14.1

12.8

10.8

10.0

10.5

Source: Eurostat, Platts, ECB

Similarly to oil and gas, the coal import bill also decreased between 2013 and 2016 although the absolute values are significantly lower. The estimated coal import bill decreased by 29%, from EUR 14.1 billion in 2013 to EUR 10.0 billion in 2016. International coal prices significantly increased in 2017 which offset the decrease of the imported volumes, resulting in a 5% increase of the import bill to EUR 10.5 billion.

Total

In 2013, the total import bill was about EUR 400 billion, more than EUR 1 billion per day. Falling prices allowed the EU to decrease its estimated import bill to EUR 358 billion in 2014 (-11%), EUR 262 billion in 2015 (-27%) and EUR 211 billion in 2016 (-19%). The cumulative decrease between 2013 and 2016 was 47%.

In 2017, however, the import bill increased by 26%, reaching EUR 266 billion.

When expressed as a percentage of EU GDP (at current prices), the share of the estimated import bill decreased from 3.0% in 2013 to 1.4% in 2016. This saving gave a significant boost to GDP growth in 2015-2016: lower energy prices meant more disposable income for households, lower energy costs for businesses and increasing activity of oil intensive sectors (e.g. transport, refining and chemicals). In 2017, the estimated import bill was equivalent to 1.7% of the GDP.

The per capita import bill decreased from around 800 euros in 2013 to 415 euros in 2016 and bounced back to around 520 euros in 2017.

Figure 98 - The estimated EU import bill

Source: DG Energy calculation



5Household energy expenditure and energy poverty 

Introduction

Energy poverty became an ever more important issue in the last decade, as retail energy prices underwent a significant increase and in the consequence of economic crisis that began in 2008 the income of many households, especially in the case of the poor, fell measurably, leading to an increasing burden of paying their energy bills, as the share of energy rose in their consumption expenditure. As energy as basic need cannot be replaced, at least not on the short run, increasing energy related expenditures imply less spending on other consumer purposes.

In its Clean Energy for All Europeans legislative package, the European Commission has proposed to include the concept of energy poverty, explicitly mentioning it in the proposal of the new Electricity Directive. This would foresee that all Member States measure energy poverty. Recently, the project on European Energy Poverty Observatory (EPOV) has been launched, aiming at collecting data, developing indicators and presenting best practices to tackle energy poverty in the EU Member States. In this chapter an analysis is provided on the importance of energy products and transport fuels for households with different income across the EU Member States.

Main findings

¾In 2015 the poorest households spent € 870 on energy products (electricity, gas, liquid and solid fuels, central heating) in the EU on average, representing around 10.4% of their total consumption expenditure. There were huge differences across the EU Member states, energy expenditures ranging from € 500 to € 2,300 per household.

¾When compared to the total expenditure, the poorest households in Sweden spent only 3% on energy, while in Slovakia this share was more than 23%.

¾Households with middle income, though spending higher amounts on energy products, spent proportionally less on energy within their total expenditure, only 7.1% on EU average, as opposed to the aforementioned 10.4% in the case of the poorest.

¾However, even middle income households in Central and Eastern Europe spent around 10-15% of their expenditure on energy, owing to lower income compared to North and Western Europe, where this share was typically around 4-8% in 2015.

¾The share of households being unable to keep their home adequately warm serves as a good complementary indicator on energy poverty, showing a positive correlation with the share of energy products within the total household expenditure. In 2017 around 19% of lower middle income households in the EU could not keep their home adequately warm, ranging from 2% in Finland to 60% in Bulgaria.

¾Expenditures on transport fuels (petrol and diesel) represented € 370 (3.1% of the total expenditure) on EU average in the case of the poorest, while for middle income households it reached € 980 (4.3% of the total expenditure).

¾Households with higher income spent proportionally more on transport fuels within their total expenditure than the poorer, and diesel had an increasing importance in their fuel spending compared to lower income households.

5.1Energy products in household budgets

In this chapter we primarily rely on data provided by national statistical authorities (NSIs) on expenditure on energy products of households in the twenty-eight EU Member States. Energy expenditure of the residential sector usually covers heating, lighting, cooking needs, and the operation of electrical appliances. Household Budget Survey (HBS) and Standard Income and Living Conditions (SILC) data, available in both Eurostat and NSI databases provide information on expenditures on products and services and the quality of living conditions. In order to analyse the burden energy expenditures mean to households, it is reasonable to take into account the major energy products households spend resources on (electricity, gas, solid fuels, liquid fuels and heating – mainly district heating), and to look at how much households with different income conditions spend on these products, in absolute figures and in the share of their total expenditure on products and services.

In the 2016 edition of the Energy prices and costs report, detailed data were provided on energy expenditures in each income quintiles (one fifth of the population regarding their income), in this report for most of the EU countries we have more refined data, detailed expenditures in each decile (one tenth of the population, arranged into income strata). Furthermore, detailed data are available on expenditures on transport fuels (fuels total, expenditures on petrol and diesel). The analysis in this report primarily focuses on the latest available data, as opposed to the 2016 report, which looked at the timely evolution of energy expenditures over a ten-year long period.

5.1.1Energy expenditure (excluding transport) in households with low income 

The next chart ( Figure 99 ) shows energy expenditure of households in the lowest decile (the poorest ten per cent of the population) in the EU countries 16 . In the EU € 870 was spent on energy on average by the poorest household according to the latest data 17 , which represented 10.4% of their total consumption expenditure. There were huge differences across the EU on both absolute expenditures and the share of energy in the total household expenditure. In Bulgaria and Romania the annual energy expenditure remained below € 500 18 , in contrast, in Denmark it was above € 2300 in 2014-2015. This nearly five-fold difference in energy expenditures reflect mainly differences in average household incomes in different EU Member States, however, differences between household energy prices also play a role. In the case of heating related expenditure the quality of residential building stock also has of particular importance, as energy expenditure can be reduced if buildings are more energy efficient.

Looking at the shares of energy products in the households' budget 19 , in Sweden the poorest households spent only 3% of their total expenditure on energy, whereas in Slovakia this share was higher than 23%. Countries in Central and Eastern Europe, primarily owing to lower incomes compared Northern and Western Europe, spent significantly higher share on energy within their household expenditure.

The role of different household energy products may also differ across the EU. A good example for this is the high share of district heating in Denmark, representing more than half of the total energy-related household expenditures 20 . In Estonia, Lithuania, the Czech Republic and Slovakia district heating also had an important share in household energy.

Electricity had high share in Sweden, Finland and France; in these countries this energy source is dominant not only in residential lighting but in heating as well. Natural gas had high share in the Netherlands, Italy, UK and Malta, and liquid fuels, mainly in the form of heating oil, are of importance within household energy in Slovenia, Belgium, Ireland and Luxembourg. Solid fuels only represented a fraction in the total energy expenditure in the EU, however, in Ireland, and in some Central and East European countries they still had a measurable share.

Figure 99 - Expenditures on household energy products for the poorest households in the EU Member States), and the share of energy in the total household consumption expenditure

Source: DG ENER ad-hoc data collection on household consumption expenditures

5.1.2Energy expenditure (excluding transport) in households with middle income

Beside the poorest households, it is reasonable to analyse the situation of the lower-middle income and middle-income households, which are represented by the third and the fifth income decile (or in the case of some countries, the second and the third quintile) in the expenditure data. As both Figure 100 and Figure 101 show, the order of the countries regarding the absolute spending on energy products and the distribution of the individual energy sources within the total spending on energy is similar in all income deciles, however, the higher income a given household has, the higher is the usual amount it spends on energy products.

On the contrary, households with higher income spend proportionally less on energy products, compared to their total consumption expenditure, than poorer households. In the third decile (lower-middle income households) the average share of energy in total spending was only 7.2% (as opposed to 10.4% in the case of the poorest), and in the fifth decile (middle income households it was 7.1%. However, even for middle income households differences across Europe are perceivable regarding the share of energy in total spending, as households in Northern and Western Europe spent typically between 4% and 8% of their expenditure on energy, in Central and East Europe this share was 10-15% in recent times, implying that amid current income levels energy represents a significant burden for households in these latter countries.

Figure 100 - Expenditures on household energy products for the lower-middle income households in the EU Member States, and the share of energy in the total household consumption expenditure

Source: DG ENER ad-hoc data collection on household consumption expenditures

Figure 101 - Expenditures on household energy products for households with middle income in the EU Member States, and the share of energy in the total household consumption expenditure

Source: DG ENER ad-hoc data collection on household consumption expenditures

There exist a few other indicators providing information on the burden of household relating to paying their energy bills and/or keeping their home sufficiently warm. Figure 102 shows the relation between spending on energy (in the share of the total) for lower-middle income households and the share of those being unable to warm up their home sufficiently.

Whereas in Finland only 2% of the households being under 60% of the median income were not able to keep their home adequately warm, in Bulgaria this share was more than 60% in 2017. The share of homes non-adequately warm shows a positive correlation (though not very strong, having a coefficient around 0.25) with the share of energy in total expenditures. The correlation is weakened by the data in some Mediterranean EU Member States, owing to lower energy expenditure amid warmer climate; however, this is not reflected in the perception of households on having a sufficiently warm home.

Figure 102 - The ratio of homes being not adequately warm for households being below the 60% of the median income and the share of energy products within the total expenditure for households in the third decile (lower-middle income)

Source: DG ENER ad-hoc data collection on household consumption expenditures and Eurostat

Box –Energy efficiency of the household sector

Energy consumption in the residential sector is impacted by several factors. Higher number of households, higher floor area of buildings, higher disposable household income all result in higher energy consumption (though as regards income proportionally less expenditure on energy products, as it was presented before, than in the case of poorer households). Increasing energy prices also result in decreasing consumption, however, energy is a price-inelastic product and necessary to living conditions. Energy consumption also shows strong correlation with climate conditions and has a strong intra-annual seasonality therefore, as two thirds of total energy consumed by households is related to heating needs.

Energy efficiency measures can mitigate the impact of these factors on the total residential energy consumption. Consumption of energy in the EU residential sector, accounting for around a quarter of total final energy use, rose by 7.4% between 2014 and 2016, largely owing to colder weather conditions in 2016 compared to the mild winter in 2014. Weather-corrected heating energy consumption has been relatively flat since 2010, following a decade of reductions. Non-heating related energy use, accounting for one third of the household energy consumption, went up by 3% a year between 2014 and 2016. This was related to large household appliances, such as refrigerators and televisions, in spite of their improving energy efficiency. For better understanding the reasons behind household energy consumption, the attention should be more focused on electricity, which seems to be driving up the total residential energy consumption in recent years.

In earlier periods significant decreases could be observed in EU household energy consumption (by 5.7 between 2008 and 2016), largely owing to decreasing heating related consumption, due to building refurbishments and more efficient heating systems (e.g.: replacement of boilers with low energy efficiency). However, over the last few years these factors could not any longer contribute to the decrease in energy consumption up to such extent as before, pointing to the need of seeking new sources of efficiency improvements in other areas, such as the aforementioned electricity use of appliances.

5.1.3 Energy expenditures in the transport sector

Figure 103 and Figure 104 show the expenditures on transport fuels (petrol and diesel, or in the case of some Member States where detailed data were not available, fuels and lubricants total). Similarly to household energy products, there were significant differences across the Member States, both in absolute spending on fuels and in their share in the total household expenditure.

There were six Member States (Romania, Bulgaria, Slovakia, Lithuania, Czech Republic and Latvia) where spending on transport fuels remained below € 100 per household in 2015, whereas in France and Italy it was above € 600. In the EU the poorest households spent € 370 on average on transport fuels, representing 3.1% of the total consumption expenditure. The lowest share of transport fuels within the total expenditure could be observed in Romania (2%), whereas in Malta the poorest households spent almost 8% on transport fuels of their total expenditure.

The share of petrol and diesel within transport fuels was different across the EU. In countries like the Netherlands, Sweden, Belgium, UK and most of the countries in Central and Eastern Europe expenditures on petrol dominated the transport fuel bill, whereas in France, Italy, Luxembourg or Austria diesel had a significant share (though with the exception of France it was lower than the share of petrol).

Figure 103 - Expenditures on transport energy products for the poorest households in the EU Member States, and the share of transport energy in the total household consumption expenditure

Source: DG ENER ad-hoc data collection on household consumption expenditures

Note: "Fuels and lubricants total" cover diesel, petrol and other fuels and lubricants. A split is not available by fuel in these EU Member States

In contrast to household energy products, the share of transport fuels within the total expenditure proportionally increases with the income of households, otherwise saying richer households tend to spend more on transport fuels within their total expenditure. As it was mentioned before, the poorest households spent 3.1% on transport energy on EU average, while those in the third income decile (lower-middle income) and in the fifth decile (middle income) respectively spent 3.9% and 4.3%. Expenditures on transport fuels reached €980 in the case of middle income households in 2015 in the EU.

Comparing the details of transport fuel expenditures of the poorest and middle income households, it seems that the share of diesel fuel is higher in the case of middle income households than for the poorest. Diesel engine cars are more popular among those who use their car more frequently, or having a higher annual mileage, as in many countries taxation of diesel fuels is more favourable (or at least it used to be in the past) compared to petrol.

As households with higher income rely more on private transport, they spend proportionally more on diesel than the poorer. However, in the future this might change as difference in taxation of petrol and diesel (mainly excise duties) will diminish and due to the changing environmental rules and public acceptance; thus diesel may not be as attractive alternative to petrol cars as in the past.

Figure 104 - Expenditures on transport energy products for households with middle income in the EU Member States, and the share of transport energy in the total household consumption expenditure

Source: DG ENER ad-hoc data collection on household consumption expenditures

(1)

However, oil market is very volatile; since the beginning of October 2018 crude oil prices fell and at the end of November 2018 the Brent crude oil price was slightly below 60 USD/bbl.

(2)

Between 30 June 2014 and 15 February 2016

(3)

In this section, other indirect taxes are reported in the excise duty component

(4)

https://ec.europa.eu/energy/en/statistics/weekly-oil-bulletin

(5)

http://europa.eu/rapid/press-release_IP-11-468_en.htm?locale=en

(6)

Trinomics et altri (2018)

(7)

Import dependency is calculated as net imports divided by gross inland consumption

(8)

The CIF price includes the FOB price (the price actually invoiced at the port of loading), the cost of transport, insurance and certain charges linked to crude oil transfer operations.

(9)

  http://eur-lex.europa.eu/legal-content/EN/ALL/?uri=CELEX:31995R2964

(10)

  https://ec.europa.eu/energy/en/statistics/eu-crude-oil-imports

(11)

  https://transparency.entsog.eu/

(12)

Including small-scale terminals in Finland and Sweden.

(13)

This includes anthracite, coking coal, other bituminous coal and sub-bituminous coal

(14)

  http://www.ecb.europa.eu/stats/exchange/eurofxref/html/index.en.html

(15)

See Commission President Juncker's speech on the State of the Union, 2018. https://ec.europa.eu/commission/sites/beta-political/files/soteu2018-speech_en_0.pdf

(16)

For some countries (Germany, Denmark and Poland) data of the lowest quintile (the poorest 20% of the population) was used for the computations as we did not receive decile data from the national authorities or there were issues with the data quality.

(17)

EU average is calculated as weighted average from Member States' expenditure data, using the number of households as weight. Latest available data in most cases mean 2015 or 2014 data, however, due to different data collection in different countries, in some cases data might be of earlier time period.

(18)

In this chapter expenditures are expressed per household

(19)

As HBS data are not fully harmonised in the EU, the actual shares might differ from the result of this analysis. In some countries the share of energy is low in the total expenditure, as energy bills are "hidden" in the rental payments in the housing sector.

(20)

According to the Danish District Heating Association, around 64% of all Danish households were connected in 2017 to the district heating grid, and district heating companies were legally bound to run on a non-profit basis. Otherwise saying, the high expenditures on district heating in Denmark was rather due to the broad deployment of district heating, not to the costs of this technology.

Top

Brussels, 9.1.2019

SWD(2019) 1 final

COMMISSION STAFF WORKING DOCUMENT

Accompanying the document

REPORT FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS

Energy prices and costs in Europe

{COM(2019) 1 final}




6Industry energy costs 

Introduction

The chapter mainly looks at the impact of energy prices and energy costs on the cost-competitiveness of selected European industrial sectors. In the following pages we first look at the context of the general competitiveness of the EU and the importance of energy costs for the overall industry and services. Energy costs are then mapped across several manufacturing, services and agricultural sectors. Emphasis is put in analysing energy intensive sectors which by nature are sensitive to energy costs fluctuations. The evolution of energy costs in the various sectors are assessed as well as the factors driving these costs like changes on energy prices, output and energy intensity. Finally the available international data is used to establish international comparisons of energy costs with third countries.

Main findings

On the overall impact of energy on the EU competitiveness

¾Energy plays a relatively modest role in the formation of the gross value added in the economy. On EU level, its share is estimated at around 2% of the total production value of manufacturing and around 1% of the combined group of industry and services in 2016.

¾The economic performance and the overall competitiveness of the EU Member States has remained stable compared to that of our main trading partners.

¾The macroeconomic effects of the wide variations of the global energy commodity prices (oil, gas and coal) in recent years are yet to unravel. It seems that up to 2017 the impact on the overall competitiveness of the EU economy was limited and the real unit energy costs of large industrial subsectors remained stable.

¾Yet, energy is at the very fabric of almost all products and services used in everyday life. In addition, several important manufacturing sectors (see section 6.2) also rely on energy as the biggest or the most critical factor of production.

Energy costs shares

¾Energy costs shares in production costs fell for the vast majority of the sectors studied between 2008 and 2015, with the most significant declines appearing in some of the most energy intensive sectors.

¾Energy costs for the selected manufacturing sectors accounted for around 1-10% of total (operational) production costs. For some sectors of the most energy intensive sectors energy costs accounted for more than 10% of production costs in at least one year like for paper, clay building materials, iron and steel and cement (on the latter sector the energy costs share was consistently above 10%).

¾Amongst the less energy intensive manufacturing sectors studied, energy costs are typically 1-3% of operational (production) costs. For computers and electronics, motor vehicles and other transport equipment costs do not reach 1% of total production costs

¾Amongst the non-manufacturing sectors studied, energy cost shares are comparable to or higher to those in the highly energy intensive manufacturing sectors for land transport, air transport, mining of metal ores, electricity-gas and other mining. It is notable that energy cost shares in waste management and accommodation and restaurants are 3-5%, while negligible in construction and trade. 

Drivers of energy costs for industry

¾The aggregated energy costs of the sectors studied at EU level fell by 8% over 2010-2015. The decomposition analysis shows that this was the result of increasing prices of energy (that induced +7% increase in energy costs), reduced energy intensity (-4% in energy costs) and almost close to zero impact from changes in output. Due to data limitations, a large part of the decrease in energy costs over the period could not be specifically identified and is behind the important reduction in energy costs over the period (the residual explains -10% of reduction in energy costs). The low quality of the energy consumption data may have prompted an underestimation on the reduction of energy intensity, which could come from the intense industry restructuring that follows economic crises and international competitive pressures in energy intensive sectors.

¾Energy costs have a zero or negative impact on the Total Production Costs in the vast majority of manufacturing sectors analysed over the period of study.

¾Energy costs shares in total production costs have fallen among nearly all the manufacturing sectors over the period of study (27 out of 30 sectors). Although half of the sectors experienced increases in energy costs, these costs have not increased by as much as other non-energy costs of production resulting in lower energy costs shares. For most of the less energy-intensity industries, the energy costs share fell by -0.1pp to -0.6pp while for the more energy intensive sectors, reductions were larger, between -5.8pp and -1.5pp.

Energy intensity

¾Energy intensity (energy consumption/GVA) varies considerably across the sectors studied in accordance to the various production processes (the highest values are displayed in steel, cement, refineries, paper and basic chemicals, in manufacturing, and land transport, electricity-gas, in non-manufacturing,)

¾Energy intensity fell in most of the highly energy intensive sectors in manufacturing, including steel, refineries and paper. There were, however, manufacturing sectors in which energy intensity increased (i.e. cement, grain products, sawmills and chemicals). In non-manufacturing, energy intensity decreased in sectors like land transport and other mining although increased in electricity-gas and agriculture. Energy intensity remained overall decreased or remained relatively stable in the less energy intensive sectors (manufacturing and non-manufacturing)

International comparisons

¾Energy costs shares in production costs in the EU are usually higher than in Asian partners (Japan, South Korea). EU costs shares are in most of the cases comparable to those in the US sectors, with the exception of sectors like non-ferrous metals (aluminium) or steel, which display lower energy costs shares in the US.

 

¾Energy intensity (proxy of energy efficiency) 1 in the sectors studied is systematically lower in the EU than those in China and Turkey. Energy intensity in the EU sectors is comparable to those in the US, although with a considerable variation per sector (EU sectors being more energy efficient (less energy intensive) in beverages, glass, fabricated metal products; and the less energy efficient (more energy intensive) in chemicals, man-made fibres and computers and electronics).

¾EU industrial prices for electricity are lower or comparable to Asian countries (lower than Japan, comparable to China) and higher than US prices (US prices are half the EU levels). Most other G20 countries (Canada, India, Russia, Mexico, South Korea, Saudi Arabia, and Turkey) also have lower prices than in the EU. Only Brazil has higher prices.

¾Industrial prices for electricity increased over the period (from 100 EUR/MWh in 2008 to 110 EUR/MWh in 2017). The price gap with the US and China (which was favourable for these two countries) has widened, slightly with the US and more significantly with China (where prices have declined since 2011). The gap with Japan (which was favourable for the EU) has shrunk as prices have converged in 2015-2016 to EU levels. Prices South Korea are increasing and converging to EU levels, whilst prices in Mexico (already below cost) are decreasing since 2014.

¾Industrial gas prices in the EU are overall lower than those in Asia (Japan, South Korea, China) but higher than the rest of G20, particularly gas producers (e.g. US, Canada, Russia, Brazil display price levels around half of those in the EU).

¾EU industrial gas prices have declined in nominal and real terms between 2008-2017 (to reach below 25 EUR/MWh in 2016) but prices declined even further in most of the other G20 countries. The price gap deteriorated with regard to the US and Canada (where prices reached 10 EUR/MWh in 2016) and Japan and South Korea (where after peaking around 2014 prices decreased to get closer to EU prices). Conversely, price gaps with China, Russia, Mexico and Australia improved for the EU as prices in these countries remained roughly constant at the end of period or increased.

¾Inflation and exchange rate changes played a significant role in the evolution of nominal prices. High inflation pushed up prices in countries like Brazil and Indonesia while in Russia and Turkey the inflationary effects were mitigated or offset by exchange rates depreciations. Exchange rate appreciations pushed up Chinese and US prices.

6.1Energy costs and competitiveness at macroeconomic level

To properly asses the cost-competitiveness of the various industrial sectors it is important to know more about the international competitiveness context of the EU. Competitiveness is a complex matter which depends not only on prices and costs (cost competitiveness) but also on aspects like the quality of the products produced and the institutional background (economy stability, legal certainty, etc.). All these different factors are weighed by investors and producers when taking decisions on where (in which country) to produce or invest.

This introductory section provides an overview of the competitiveness context of the EU and precedes the more detailed analyses on the sectorial competitiveness (section 6.2). In this section we first assess how the EU competitiveness is placed internationally by looking at international competitiveness indexes. Second, we assess the overall impact of energy costs in the cost competitiveness of the EU and its Member States. We do so by calculating the shares of energy costs in the total production value of the whole industry and services sectors in each Member State.

Defining competitiveness and the factors affecting it

The competitive positions of companies, industries and economies are impacted by a set of factors that go beyond prices, costs and factor productivity. The country risk, the political stability, the regulatory environment, the presence or absence of barriers to trade and investment and taxation policy (among others) are weighing heavily on economic decisions to invest or do business in a given area of the world.

The purpose of the current and the next section of the chapter is to position the energy-related aspects as a subset of the complex group of interdependent factors that impact competitiveness, productivity and economic decisions to invest. Seen from the macro-economic perspective, the importance of the energy compound might appear modest when compared to total production value. Yet, as energy is at the very fabric of almost all products and services used in everyday life its role is crucial for the entire economy.

Several international institutes and organisations have developed methodologies combining hard statistical data and surveys of experts' opinions to measure the factors that influence the competitiveness of a given economy.

The Global Competitiveness Indicator of the World Economic Forum (WEF)  2 , the World Competitiveness Scoreboard of the International Institute for Management Development (IMD) 3 and the Economic Freedom of the World Index of Fraser Institute (FI) 4 are using similar bottom-up approaches to measure economic performance and assess the competitive position of a country. A wide range of parameters are first collected and quantified, to be later regrouped in main themes which will then feed into global, composite indices. A methodological note at the end of section 6.1.2 presents the structure, the parameters and the factors of competitiveness used in the global indices of WEF, IMD and FI.

WEF defines 5 the competitiveness on a national level as "the set of institutions, policies, and factors that determine the level of productivity of an economy, which in turn sets the level of prosperity that the country can earn"

6.1.1Competitiveness drivers: EU vs G20

From 2008 to 2018, the EU economy has followed a pathway that is broadly comparable to that of its main trading partners, proxied here as the non-EU group of G20 countries 6 . Starting from 2014, the EU economy as a whole was actually improving its overall competitiveness and performance, as shown by the aggregate indices in Figure 105 .

The focus of this section is on general trends that are driving competitiveness; as such the use and comparison of groups of countries to track evolution is justified. It should nevertheless be pointed out that the EU and the non-EU G20 are far from being homogenous groups. They are rather composed of countries that are at undergoing different stages of their economic development. Caution should be used when general conclusions are applied to specific countries of the two groups. The charts that follow try to capture group diversity by reporting data for averages and dispersion (outliers and middle quartiles).



Figure 105 - Overall competitiveness in EU and non-EU G20

This finding on gradual improvement of competitiveness of the EU holds when the economic performance of the two groups is measured by actual score or by the ranking position of the countries that are being analysed.

Figure 106 and Figure 107 report on a set of components and factors of competitiveness where EU Member States and the non-EU G20 countries show a relative divergence in terms of performance. Those factors could also explain possible differences of the level of macroeconomic competitiveness of the two groups of countries.

The indices of WEF, IMD and FI produce similar results indicating that the EU group seems to enjoy a clear competitive advantage over the non-EU G20 group in the following areas: international trade 7 ; freedom to trade internationally 8 ; prices 9 ; public finance 10 ; institutional framework 11 ; sound money 12 ; business legislation 13 ; infrastructure – including electricity supply; security; health and primary education and ICT use. On the other hand, the relatively small size of EU economies (taken separately), the employment and flexibility of the labour market and fiscal policies seem to be pushing down the economic performance of the EU group compared to the group of non-EU G20 countries.

Figure 106 - Results for selected favourable factors of competitiveness



vz

Figure 107 - Results for selected unfavourable factors of competitiveness



The recent improvement of EU economic performance is also confirmed by the European Commission reports on Single Market Integration and Competitiveness 14 .

In terms of overall productivity, the majority of EU MS are competing with the high-income members of the non-EU G20 group, as shown in Figure 108 .

 

Figure 108 - Overall productivity in the EU and G20

Notes:

1. Overall productivity is defined as the ratio of GDP over employment; employment data are estimates and are provisional for the most recent period;

2. Data for Malta is not available

Over the recent years, the dispersion of productivity and efficiency rates across G20 economies has increased, with high income countries enjoying stronger growth rates than the low-income and less diversified economies of the non-EU G20. Whereas the group of EU countries perform better over the group on non-EU G20 (see Figure 109 ), the 2016 Single Market Integration and Competitiveness report mentioned above, as well as its 2015 version, point out that many EU Member States face a generalised fall in productivity growth rates and that productivity gaps are accumulating with respect to high income G20 countries. Reforms enhancing the labour and total factor productivity of European companies at both national and EU level are further discussed in separate reports of the European Commission 15 .

Figure 109 - Productivity and Efficiency of EU and non-EU G20

Whereas the underlying trends driving productivity remain the same, the productivity spread across high- and low- income countries is less pronounced when the monetary value used is switched from US Dollars to Purchasing Power Parities 16 . This is confirmed both for the industrial and services sectors, as shown in Figure 110 and Figure 111 .



Figure 110 - Labour productivity in industry in EU and G20

Notes:

1. Industrial productivity is defined as the ratio of related GDP (PPP) per person employed in industry; employment data are estimates and are provisional for the most recent period

2. Data for Canada and Brazil is not available

Figure 111 - Labor productivity in Services in EU and G20

Notes:

1. Productivity in services is defined as the ratio of related GDP (PPP) per person employed in services; employment data are estimates and are provisional for the most recent period

2. Data for Canada and Brazil is not available



6.1.2Impact of energy on the economy's competitiveness

Figure 112 reports on the results from the user satisfaction survey on the adequacy and efficiency of energy infrastructure. As a group, EU clearly outperforms the non-EU G20 countries, as EU users of energy grids tend to be more satisfied on average with the overall operation, reliability and quality of service of dispatching of energy than their counterparts from G20 countries.

Figure 112 - User satisfaction on Energy infrastructure in EU and G20

Looking at recent prices for electricity for industrial users, both the EU and the non-EU G20 country groups appear as quite dispersed. Within non-EU G20, and compared to prices from 10 years ago, Mexico and Turkey register decreases in the range of 10% - 20% whereas the highest increases were recorded in Indonesia and Korea, both above 30%, as shown on Figure 113 . Section 6.5.3 provides a comprehensive international comparison on the evolution industrial prices, using a richer data sources.



Figure 113 - Elecriticity prices for industry in the EU and G20 in 2017

Notes:

1. Prices refer to the simple average of the domestic monthly reference with tax for electricity for industry; data may be different

2. US prices are net of taxes;

3. Data for India is for 2013, data for Saudi Arabia is for 2014 data for South Africa is not available

Assessing the importance of energy as a production factor on the global macroeconomic level, and comparing the performance of EU countries against its main trading partners is not straightforward, with serious data availability and methodological issues.

In terms of the absolute value of energy costs, a report 17   from the European Commission estimated that the EU manufacturing sector had some of the lowest Real Unit Energy Costs (RUEC)  18 together with Japan and the US. While the USA is constantly performing better than the EU, looking at updated data 19 , Japan has recently decoupled from the EU and increased its RUEC, mainly on the back of rising energy prices. China and Russia, on the other hand, score worse than the EU on a permanent basis (see Figure 114 ).

Figure 114 - Real Unit Energy Cost - manufacturing excluding refining

Source: DG JRC (own calculations based on WIOD) and DG ECFIN

The Real Unit Energy Costs of the EU has remained broadly stable over the last 5 years. However, a slight increase was recorded recently, especially due to higher energy prices. To a large extent these were offset by decreasing energy intensity of the European manufacturing sectors ( Figure 115 ).

Figure 115 - Contribution to growth of RUEC by Real Prices and Energy Intensity

Source: DG JRC (own calculations based on WIOD) and DG ECFIN

The improvements of the EU industry in terms of energy intensity have helped to offset the increase in real energy prices. Compared to its world peers, EU manufacturers have developed best-of-class industrial processes in terms of energy efficiency and have continued to steadily improve their energy intensity levels.

The relative share of energy in total factor production costs can be proxied by the share of energy products in total production value, as reported by the Structural Business Statistics (SBS) tables in Eurostat. This approach has several important limitations, listed in the Box at the end of this section, but remains the only viable one in terms of harmonised and publically available data.

Figure 116  shows the evolution of the share of energy-related costs in total production value in the broad classes of industry and services 20 . On EU level, and for the last decade of observed data, this share has decreased from 1.5-1.7% to around 1.0%-1.3%.

Figure 116 - Evolution of energy costs shares in production value

Notes:

1. Data for Malta (prior to 2016), Poland (prior to 2015), Slovenia (prior to 2012) and Greece (prior to 2008) is not available. Data for Denmark, Sweden and the United Kingdom for 2016 was missing by the time of extraction

Figure 117  represents the share of energy-related costs for the manufacturing sector and across the EU Member States. Throughout the 2008-2016 period, and where data is available, the energy share has gradually decreased for the majority of Member States. On EU level, it went from 2.2%-2.5% at the beginning of the period to 1.5%-2.0% at the end. Member States with relatively smaller size would typically present a higher and more oscillating share than average; probably pointing to the fact that these economies have a relatively less diversified portfolio of manufacturing industries centred mainly on more energy-intensive sectors.

Figure 117 - Evolution of energy costs shars in production value for Manufacturing

Notes:

1. Data for Malta (prior to 2016), Poland (prior to 2015), Slovenia (prior to 2012) and Greece (prior to 2008) is not available. Data for Denmark, Sweden and the United Kingdom for 2016 was missing by the time of extraction



Box- Data limitations

·There is no one-on-one mapping between the economic indicators of SBS and the profit and loss account of real companies;

·Capital expenditure (CAPEX) is difficult to collect in SBS, forcing the estimation of the energy component to rely solely on operating expenditure (OPEX); as a result the provided estimation is not assessing the long term investment and cannot determine the relative share of investment in improved energy performance tools over the total stock of investment;

·The purchases of energy product data is available only for NACE Rev. 2 sections B (Mining and quarrying), C (Manufacturing), D (Electricity, gas, steam and air conditioning supply) and E (Water supply, sewerage, waste management and remediation activities). It is not available for important industrial such as Section F (Construction) and energy intensive sections such as H (Transportation and storage). More importantly, it is not available for all services sectors. According to the 2015 Commission report on single market integration and competitiveness, the relative share of the services sector in the 2014 Total Value Added in the EU 28 stood at almost 75%, as opposed to 15% for Manufacturing.

·Based on the definition of the Commission Regulation (EC) No 250/2009, the structural business statistics (SBS) code "20 11 0 Purchases of energy products" includes only energy products which are purchased to be used as a fuel. Energy products purchased as a raw material or for resale without transformation (such as crude oil) are excluded.



Methodological note

Structure, parameters and factors of competitiveness in WEF, IMD and FI indices



6.2Energy costs for industry

Sources, scope and methodology

The following sections of this chapter mainly rely on findings from studies commissioned by the European Commission to external consultants. The main source for this chapter is the study on 'Energy prices, costs and subsidies and their impact on industry and households' by Trinomics et altri 21 (2018), onwards study by Trinomics, which provides data and analyses on a wide range of manufacturing sectors and other relevant economic sectors (30 manufacturing sectors, including 15 energy intensive industries plus 14 sectors from agriculture, extractive and services). The study on 'Composition and drivers of Energy: case studies in selected Energy Intensive Industries' by CEPS and Ecofys (2018) 22 , onwards the study by CEPS, is the other main source for this chapter and focuses on case studies of 8 energy intensive subsectors.

The methodology of both studies is different but complementary. The study by Trinomics et altri follows a top-down approach using highly aggregated statistical data in particular for non-manufacturing sectors (30 manufacturing sectors at NACE 3 level, 7 non-manufacturing level at NACE 2 level and 7 other non-manufacturing at NACE level 1). Statistical data and estimates based on that data are used to understand the role of energy prices and costs in the competitiveness of these sectors. The CEPS study follows a bottom-up approach. CEPS calculations are based on direct collection of price and costs data at plant level via a questionnaire which allows analysing samples of varying representativeness.

These two approaches are complementary and provide a broader vision of the energy prices and energy costs paid by European industries. Highly aggregated data (used in a top down approach) usually is easily available in official statistics with well-established methodologies that cover long-time spans that makes it good for identifying long term trends. But aggregated information within one sector contains individual companies which may have rather heterogeneous industrial processes and products. Plant data or highly disaggregated data (used in bottom up approach) can be better for identifying targeted sub-groups of individuals and therefore represent better the analysed characteristics of these groups. But plant data is however scarce, normally based on ad-hoc methodologies which make comparisons of each studies or sources difficult and most importantly the actual effectiveness of the data sample for describing the targeted sub-groups of individuals depends critically on how large and how well the sample represents that group of individuals (something which is difficult to achieve).

Approach for the selection of sectors

The study commissioned by the European Commission to Trinomics et altri analysed energy costs and other indicators across 44 sectors of different levels of aggregation. Sectors were selected by looking at three aspects:

¾Importance of energy costs for the sector usually proxied by the energy cost per production value, calculated by dividing purchases of energy by the total production value of each sector 23 ;

¾Economic importance of the sector proxied by the share of sectoral value added in GDP of the country and by the assessment of the general economic or strategic relevance of the sector;

¾The trade exposure of the sector which was proxied by the trade intensity of the sector which was calculated by dividing the sum of imports and exports of a product to and from the EU in total, by the size of the market which is represented by the sum of production value and imports.

The selection resulted in 30 manufacturing sectors 24 . The first 15 sectors are the most intensive energy sectors and were already the object of a study in the 2016 energy prices and costs report (shaded sectors in Table 9 ). In addition to those sectors, 15 more manufacturing sectors were selected in order to be able to map energy costs in manufacturing (non-shaded sectors in Table 9 ). The mapping of energy costs was completed by selecting the other non-manufacturing sectors.

Table 9 - Coverage of manufacturing sectors

Coverage of Manufacturing

Study by Trinomics et altri

Study by CEPS et altri

Sector

Level of aggregation (NACE code)

Sector

Level of aggregation (NACE code)

Grain mill and starch products

C106

Weaving of textiles

C132

Sawmilling and planing of wood

C161

Pulp, paper and paperboard

C171

Refined petroleum products

C192

Refineries

C1920

Basic chemicals and fertilisers

C201

Nitrogen fertilisers

C2015*

Man-made fibres

C206

Glass and glass products

C231

Packaging glass

C2313*

Glass tableware

C2313*

Refractory products

C232

Clay building materials

C233

Wall and floor tiles

C2331

Bricks and roof tiles

C2332

Porcelain and ceramic products

C234

Cement, lime and plaster

C235

Cutting stone

C237

Basic iron and steel and of ferro-alloys

C241

Iron and steel

C2410

Non-ferrous metals

C244

Aluminium

C2442

Fruit and vegetables

C103

Articles of paper and paperboard

C172

Plastics products

C222

Abrasive products and non-metallic mineral products n.e.c.

C239

Casting of metals

C245

Beverages

C11

Basic pharmaceutical products

C21

Fabricated metal products (except machinery)

C25

Computer, electronic and optical products

C26

Electrical equipment

C27

Machinery and equipment n.e.c.

C28-

Motor vehicles, trailers and semi-trailers

C29

Other transport equipment

C30-

Other manufacturing

C32

Repair, installation of machinery

C33

* The sector analysed is a subsector of the NACE code mentioned.

Source: European Commission Services

Note: Shaded Sectors are those most intensive

Table 10 - Coverage of other sectors, excluding manufacturing

Coverage of other agriculture, mining, construction and services

Study by Trinomics et altri

Sector

Level of aggregation (NACE code)

Agriculture, forestry and fishing

A

Extraction of crude petroleum and natural gas

B06

Mining of metal ores

B07

Other mining and quarrying

B08

Electricity, gas, steam and air-conditioning supply

D35

Water supply, sewerage, water management and remediation activities

E38

Construction

F

Wholesale and retail trade

G

Land Transport

H49

Air Transport

H51

Accommodation and food service activities

I

Information and communication

J

Professional, scientific and technical activities

M

Administrative and support service activities

N

Source: European Commission Services



Energy costs shares

This section focuses in assessing the cost-competitiveness of the various industrial sectors. The share of energy costs in the total production cost is a good indicator of the impact that energy costs can have on the price competitiveness of the various industrial sectors. Using Eurostat SBS as the main data source, energy cost shares are calculated by dividing the purchases of energy by total production costs, where total production costs are equal to total purchases of goods and services (including energy) 25 plus personnel costs.

Before looking at the results we should bear in mind that the heterogeneity of the energy intensity of the industries aggregated in each sector makes that the results for the total sector usually underestimates the impact of energy costs on the industrial segments with the highest energy intensity. This is particularly true for sectors like chemicals, cement, non-ferrous metals, steel and paper which include companies producing high energy intensive primary products alongside companies producing low energy intensive secondary products. It is also important to be aware that the consumption of self-generated energy is not captured by the indicator analysed (energy costs shares using SBS data) and that self-consumption is not uncommon in the industrial segments with a high energy intensity. This is why the analysis in this section is complemented by a more exhaustive analysis of all the factors affecting the energy costs (self-consumption of self-generated energy, exemptions to energy taxes, etc.) at a more disaggregated level (NACE 4) on the basis of the results of the CEPS study 26  (see 6.6.4 and Annex 1)

Results on energy costs shares

and Figure 120 (see below) look at the main developments on energy costs shares the selected manufacturing sectors in the period 2008-2015 showing that:

¾Energy costs for the selected manufacturing sectors typically accounted for around 1-10% of total (operational) production costs, although for some sectors the costs significantly exceed 10% (e.g. Cement, lime and plaster and Clay building materials)

¾Amongst the 15 most energy intensive manufacturing sectors energy costs accounted for more than 10% of production costs in at least one year in the pulp and paper, clay building materials, iron and steel and in particular, the cement, lime and plaster sectors.

¾Amongst the 15 less energy intensive manufacturing sectors energy costs are typically 1-3% of operational (production) costs. For computers and electronics, motor vehicles and other transport equipment costs do not reach 1% of total production costs

¾Over the period 2008-2015, energy cost shares have fallen in almost every sector (except for the refineries sector) The largest declines in cost share were observed in the most energy intensive sectors like cement, lime and plaster (-7%). clay building materials (-4%), pulp and paper (-4%), glass (-1.7%) and iron and steel (-1.7%). Other sectors with smaller declines nevertheless see proportionally significant decreases like in non-ferrous metals, textiles and pharmaceutical products.

¾While the overall trend is for decline in energy cost shares across all sectors over the full period there are few exceptions (although more frequent in the second part of the period, 2011-2015) such as the refractory products, clay building materials, abrasive products, fabricated metal products and computer and electronics for which the cost shares increased by approximately 1-3%.

An account of the available data from MS to estimate the sector's energy cost share in production costs can be found in Annex D of the study by Trinomics et altri study (2018).

Box - Energy costs and the sector's fuel mix

Energy costs are determined by the price of energy products and the quantities of consumed for each product. Prices usually show high volatility while energy consumption volumes tend to more stable (as it depends on factors like the consumption patterns, the economic situation and energy efficiency). This makes that prices changes explain most of the changes in energy costs in the short term. The consumption fuel mix is thus very relevant for energy costs as it determines how much will be affected the energy costs by price changes in each energy product. It is also important to note that the sector's fuel mix tends to be rather stable (as it depend inter alia on the fuel requirements of the specific production process and the availability of the energy product (e.g. gas is not always available) which usually limit fuel switching)

Figure 118 displays the average importance of fuels in terms of energy consumption by sector and Figure 119 shows the importance of fuels in terms of their energy costs shares in total energy costs related to each sector.

These figures show that electricity and gas (this varies across sectors) are generally the most important energy products in terms of consumption. Electricity is the energy product having the biggest impact on energy costs shares (e.g. > 80% for Pharmaceuticals, non-ferrous metals and computers and electronics). This can be explained by its relatively high price compared to the other fuels. Natural gas has a major impact on the energy costs in glass, beverages and steel. Oil and coal have a relatively small impact on energy costs even when consumption is high. Oil costs are relevant for refineries, cement and chemicals while coal costs are for steel, abrasive products, cement and casting of metals. “Other energies”, in particular biomass, represent an important consumption share in some sectors like sawmills (>80% of consumption), man-made fibres (57%), stone (38%) and paper (29%) and can thus significantly impact their energy costs.

Figure 118 - Breakdown of the energy consumption per energy carrier, EU, 2008-2015 averages

Source: Trinomics et altri study

Note: “other” combines biomass and heat energy consumption

Figure 119 - Average energy cost shares per sector – based on available data points, split by energy carrier, 2008-2015 averages

Source: Trinomics et altri study



Table 11 - Energy costs shares in total production costs for manufacturing and non-manufacturing, 2008-2015

 

2008

2009

2010

2011

2012

2013

2014

2015

Changes 2008-2015

Changes 2008-2011

Changes 2011-2015

Level 2015

Average

Max. level

Low. level

Diff max-low level

Section C

C103 - Fruit and vegetables

3,6%

3,5%

2,8%

2,8%

3,0%

2,8%

2,9%

2,5%

-1,1%

-0,8%

-0,3%

2,5%

3,0%

3,6%

2,5%

1,1%

C106 - Grain products

3,8%

3,8%

3,3%

3,1%

3,3%

3,1%

3,3%

3,0%

-0,8%

-0,6%

-0,1%

3,0%

3,3%

3,8%

3,0%

0,8%

C132 - Textiles

4,3%

6,4%

3,6%

2,5%

2,7%

2,4%

2,3%

2,1%

-2,2%

-1,8%

-0,4%

2,1%

3,3%

6,4%

2,1%

4,3%

C161 - Sawmills

3,7%

4,1%

3,6%

4,1%

3,7%

3,6%

3,4%

3,1%

-0,6%

0,4%

-1,0%

3,1%

3,7%

4,1%

3,1%

1,0%

C171 - Pulp and paper

12,2%

13,0%

11,1%

11,2%

10,7%

9,9%

9,1%

8,4%

-3,9%

-1,1%

-2,8%

8,4%

10,7%

13,0%

8,4%

4,6%

C172 - Articles of paper

3,6%

3,7%

3,1%

2,8%

3,0%

3,0%

2,7%

2,5%

-1,0%

-0,8%

-0,3%

2,5%

3,0%

3,7%

2,5%

1,2%

C192 - Refineries

3,2%

2,4%

2,5%

2,0%

2,8%

3,1%

3,1%

3,7%

0,6%

-1,2%

1,7%

3,7%

2,8%

3,7%

2,0%

1,7%

C201 - Basic chemicals

7,1%

7,7%

6,8%

7,0%

6,7%

6,7%

6,1%

5,7%

-1,4%

-0,1%

-1,3%

5,7%

6,7%

7,7%

5,7%

2,0%

C206 - Man-made fibres

8,6%

12,4%

7,8%

7,1%

6,7%

8,5%

6,5%

6,2%

-2,4%

-1,6%

-0,9%

6,2%

8,0%

12,4%

6,2%

6,2%

C222 - Plastics products

3,5%

3,5%

2,9%

2,9%

2,8%

2,9%

2,7%

2,6%

-0,9%

-0,6%

-0,3%

2,6%

3,0%

3,5%

2,6%

0,9%

C231 - Glass

9,8%

10,1%

8,9%

9,1%

10,3%

10,1%

9,3%

8,2%

-1,7%

-0,7%

-0,9%

8,2%

9,5%

10,3%

8,2%

2,1%

C232 - Refractory products

6,9%

6,5%

6,2%

5,9%

6,5%

6,6%

5,8%

6,1%

-0,8%

-1,0%

0,1%

6,1%

6,3%

6,9%

5,8%

1,1%

C233 - Clay building materials

15,4%

14,1%

11,8%

11,0%

12,4%

12,4%

11,3%

11,1%

-4,3%

-4,4%

0,1%

11,1%

12,4%

15,4%

11,0%

4,4%

C234 - Porcelain and ceramics

6,0%

5,7%

4,8%

5,0%

5,3%

5,4%

5,0%

4,3%

-1,7%

-1,0%

-0,8%

4,3%

5,2%

6,0%

4,3%

1,7%

C235 - Cement, lime and plaster

22,1%

22,9%

22,1%

23,5%

21,4%

21,8%

20,9%

16,3%

-5,8%

1,5%

-7,3%

16,3%

21,4%

23,5%

16,3%

7,3%

C237 - Stone

4,8%

4,4%

3,3%

3,4%

2,6%

4,3%

3,1%

3,2%

-1,5%

-1,4%

-0,1%

3,2%

3,6%

4,8%

2,6%

2,1%

C239 - Abrasive products

5,8%

5,3%

4,9%

4,9%

5,0%

5,2%

4,8%

5,1%

-0,7%

-0,9%

0,1%

5,1%

5,1%

5,8%

4,8%

1,0%

C241 - Iron and steel

9,2%

11,9%

9,5%

7,7%

8,5%

8,5%

7,3%

7,5%

-1,7%

-1,4%

-0,3%

7,5%

8,8%

11,9%

7,3%

4,6%

C244 - Non-ferrous metals

4,6%

6,0%

4,2%

4,0%

3,9%

4,0%

3,6%

3,5%

-1,1%

-0,5%

-0,6%

3,5%

4,2%

6,0%

3,5%

2,5%

C245 - Casting of metal

6,4%

7,1%

6,0%

5,2%

5,4%

5,5%

5,3%

4,9%

-1,4%

-1,1%

-0,3%

4,9%

5,7%

7,1%

4,9%

2,2%

 

2008

2009

2010

2011

2012

2013

2014

2015

Changes 2008-2015

Changes 2008-2011

Changes 2011-2015

Level 2015

Average

Max. level

Low. level

Diff max-low level

C11 - Beverages

2,6%

2,6%

2,6%

2,7%

2,6%

2,6%

2,5%

2,4%

-0,2%

0,1%

-0,2%

2,4%

2,6%

2,7%

2,4%

0,2%

C21 - Pharmaceutical products

2,8%

1,7%

1,2%

1,2%

1,3%

1,3%

1,2%

1,1%

-1,7%

-1,6%

-0,1%

1,1%

1,5%

2,8%

1,1%

1,7%

C25 - Fabricated metal products

2,2%

2,4%

2,3%

1,9%

2,0%

2,1%

2,1%

1,9%

-0,2%

-0,3%

0,0%

1,9%

2,1%

2,4%

1,9%

0,5%

C26 - Computer and electronics

0,9%

0,9%

0,7%

0,8%

0,8%

0,8%

0,8%

0,8%

-0,2%

-0,2%

0,0%

0,8%

0,8%

0,9%

0,7%

0,2%

C27 - Electrical equipment

1,1%

1,3%

1,0%

1,0%

1,0%

1,0%

1,1%

0,9%

-0,3%

-0,2%

-0,1%

0,9%

1,0%

1,3%

0,9%

0,5%

C28 - Machinery and equipment

1,1%

1,2%

1,0%

0,9%

0,9%

1,0%

0,9%

0,8%

-0,3%

-0,2%

-0,1%

0,8%

1,0%

1,2%

0,8%

0,4%

C29 - Motor vehicles

1,0%

1,0%

0,8%

0,8%

0,8%

0,8%

0,7%

0,7%

-0,3%

-0,2%

-0,1%

0,7%

0,8%

1,0%

0,7%

0,3%

C30 - Other transport equipment

1,1%

1,0%

0,9%

0,8%

0,8%

0,9%

0,7%

0,8%

-0,3%

-0,3%

-0,1%

0,8%

0,9%

1,1%

0,7%

0,4%

C32 - Other manufacturing

1,3%

1,4%

1,3%

1,1%

1,1%

1,1%

1,1%

1,0%

-0,3%

-0,2%

-0,1%

1,0%

1,2%

1,4%

1,0%

0,4%

C33 - Repair of machinery

1,3%

1,2%

1,1%

1,1%

1,1%

1,2%

1,1%

0,9%

-0,4%

-0,2%

-0,2%

0,9%

1,1%

1,3%

0,9%

0,4%

Other sections

B - Mining and quarrying

3,4%

2,9%

2,9%

2,7%

2,8%

2,8%

2,7%

3,1%

-0,3%

-0,8%

0,5%

3,1%

2,9%

3,4%

2,7%

0,8%

B06 - Oil and gas

1,6%

0,6%

0,6%

0,5%

0,6%

0,7%

0,7%

0,7%

-0,9%

-1,1%

0,2%

0,7%

0,7%

1,6%

0,5%

1,1%

B07 - Mining of metal ores

15,8%

16,6%

19,7%

20,8%

19,6%

19,4%

17,7%

18,4%

2,6%

5,0%

-2,4%

18,4%

18,5%

20,8%

15,8%

5,0%

B08 - Other mining

10,3%

9,8%

10,4%

10,4%

10,9%

10,2%

9,6%

9,4%

-0,9%

0,1%

-1,0%

9,4%

10,1%

10,9%

9,4%

1,5%

D35 - Electricity, gas and steam

17,0%

16,8%

16,9%

16,4%

14,3%

12,3%

11,4%

11,5%

-5,5%

-0,6%

-4,9%

11,5%

14,6%

17,0%

11,4%

5,6%

E38 - Waste management

4,0%

3,0%

3,1%

3,5%

4,2%

4,3%

4,8%

4,3%

0,3%

-0,5%

0,8%

4,3%

3,9%

4,8%

3,0%

1,8%

F - Construction

1,5%

1,5%

1,5%

1,7%

1,7%

1,7%

1,6%

1,4%

0,0%

0,2%

-0,3%

1,4%

1,6%

1,7%

1,4%

0,3%

G - Wholesale and retail trade

0,7%

0,8%

0,7%

0,6%

0,7%

0,6%

0,6%

0,6%

-0,1%

0,0%

0,0%

0,6%

0,7%

0,8%

0,6%

0,2%

H49 - Land transport

36,3%

31,0%

33,2%

40,6%

37,0%

34,4%

32,1%

27,0%

-9,3%

4,3%

-13,6%

27,0%

33,9%

40,6%

27,0%

13,6%

H51 - Air transport

19,5%

16,7%

21,6%

20,1%

23,3%

20,0%

24,4%

20,2%

0,7%

0,6%

0,1%

20,2%

20,7%

24,4%

16,7%

7,8%

I - Accomodation and restaurants

3,9%

4,2%

4,7%

4,2%

4,5%

4,3%

3,7%

3,9%

0,0%

0,3%

-0,3%

3,9%

4,2%

4,7%

3,7%

1,1%

Figure 120 - Energy costs shares in total production costs in manufacturing sectors, 2008-2015

Source: Trinomics et altri study

Figure 121 - Energy costs shares in total production costs in non-manufacturing sectors

Source: Trinomics et altri study

Amongst the non-manufacturing sectors for which data was available energy cost shares are particularly high in 5 sectors, being comparable to or higher than cost shares in the most energy intensive manufacturing sectors. These 5 sectors are H49 Land transport (H49), Air transport (H51), Mining of metal ores (B07), Electricity, gas and steam (D36) and other mining (B08). Clearly fuel costs are important drivers of costs in the transport and electricity and gas sectors, whilst mining is also an energy intensive activity. It is notable that energy cost shares in Waste management (E38) and Accommodation and restaurants (I) also have cost shares of 3-5%, which is comparable to many of the energy intensive manufacturing sectors. Energy cost shares are negligible in the construction (F) and Wholesale and retail (G) sectors.

Box - Energy costs for refineries

In sectors like refineries (also chemicals) the impact of energy products on production costs goes beyond the purchases of electricity and gas from external energy suppliers. In refineries energy products are also used as feedstocks (e.g. crude oil) in the industrial process. Moreover, some energy products are also self-produced and self-consumed in the industrial process. Energy costs have thus a key relevance for refineries.

Estimating the importance of all energy costs for the refinery sector is complex due to the limited or confidential data. The 'purchases of energy' from Eurostat statistics (SBS) represented on average bit less than 3% of the production costs over the last years (Trinomics et altri, 2018). However, the SBS 'purchases of energy' does not include crude oil expenses. These crude oil expenses were estimated to account for more than 80% of the production costs of refineries (CEPS and Ecofys, 2018).

Finally, refineries also consume petroleum products, refinery gas and petroleum coke for its own use. Such products are self-consumed. The estimation of the monetary amounts from self-generated and self-consumed products (Trinomics et altri) signals that these amounts would represent a small share of total energy costs (few percentage points) which would tend to be smaller where products of the prices are lower (particularly as regards gas).

Results on Gross Operating Surpluses shares

Profit margins together with production costs make up the final sales price. Profits therefore play an important role in the cost-competitiveness of firms in the short term (when setting prices). But profits are also relevant for the competitiveness in the long term as they are necessary to attract and enable investment.

It is thus interesting to analyse the trends on Gross operating surplus 27 (GOS, a proxy for profits) for the sectors studied (see the two graphs below in Figure 122 ) shows the average GOS share in production costs for the manufacturing sectors between 2008 and 2015. For most of those sectors, the share was between 5-15%, higher for sectors like pharmaceuticals; cement, beverages and particularly lower for steel (only 3.2% and even negative in one year, in 2009).

The GOS shares in production costs increased and decreased across for some sectors alike. The proportionally higher increases were in textiles (+94%), casting of metal, paper, porcelain and ceramics (all the latter around 50%). The most significant declines were in steel (-57%) and refineries, cement, and motor vehicles (all the latter around minus 30%).

Figure 122 - Gross Operating Surplus in manufacturing sectors (average 2008-2015)

Source: Trinomics et altri study

Note: Average of for the sector based on the MS for which total production cost and GOS data available for all years

For the EU as a whole (see Figure 123 ), GOS shares in production costs oscillated in the range of 11-13%/year between 2008 and 2015, with significant differences between Member States. Poland, the UK and Ireland have the highest surpluses (over 16%) and are closely followed by Greece, Cyprus, Bulgaria and Romania. The lowest surpluses are found in France, Italy, Belgium and Sweden.

Figure 123 - Gross Operating Surplus in manufacturing in the EU and Member States, 2008-2015

Source: Trinomics et altri study



6.3 Exploring energy intensities 

Energy intensity is the result of dividing the energy consumption by the Gross Value Added (GVA). Although is not is not a direct measure of energy efficiency of production (which could be measured by dividing the energy consumption by the volume of production) it is used as proxy of it. This is because comparable production volume data is not easily available. In any case, when using energy intensity as proxy for energy efficiency one should bear in mind that the energy intensity is subject to the factors that change the value added of the production, i.e. there are price effects (which are common and volatile and that can be due to various reasons like demand changes, monetary issues, etc.) that affect the value added and are not related to changes in the volume of production.

Figure 124 , Figure 125 and Figure 126 (see next page/s) display the energy intensity of selected sectors in the period 2008-2015 showing that:

¾Energy intensity varies considerably across sectors in accordance to the various production processes (steel and cement (>2 toe/1000 Euros) and refineries and paper (> 1 toe/1000 thousand Euros) display the highest values).

¾Energy intensity decreased in most of the energy intensive sectors like steel (-1,9%/year since 2009), refineries and paper although it increased in cement (by around 3.1%/year since 2009), grain products, sawmills and chemicals. For the rest it remained relatively similar levels.

¾Energy intensity was particularly volatile for refineries, steel and man-made fibres probably reflecting price affecting the value added of production.

¾There were also proportionally significant decreases in the less energy intensive manufacturing sectors like in textiles, stone, articles of paper, motor vehicles (although for these sectors the indicator lies at a rather low level, i.e. between 0.3 and 0.1 toe/1000euros)

¾Energy intensity also decreased in non-manufacturing sectors with the highest energy intensity (e.g. land transport, electricity-gas, other mining and agriculture) and remained relatively stable for the less energy intensive sectors.

Figure 124 - Energy intensity (consumption/value added in nominal terms) for the most energy intensive manufacturing sectors (average of available countries) 28

Source: Trinomics et altri study

Figure 125 - Energy intensity (consumption/value added in nominal terms) for other manufacturing sectors (average of available countries) 29

Source: Trinomics et altri study

Figure 126 - Energy intensity (consumption/value added in nominal terms) for non- manufacturing sectors (average of available countries) 30

Source: Trinomics et altri study

6.4Energy costs drivers 

Changes in energy costs over time can be the result of several factors. In this section we estimate how changes on energy prices, output and energy intensity impact energy costs. The section relies on the main findings of the decomposition analyses undertaken in the Trinomics et altri study which assesses the extent to which these three factors affected the energy costs of selected energy-intensive sectors over 2010-2015.

The decomposition was carried out using the Log Mean Divisia Index (LMDI) which shows for a given percentage change in energy costs over the period, the extent to which this change is attributable to changes in each driver over the same period. To make that analysis it was necessary to estimate the prices and the consumption by sector. The purchases of energy resulting from multiplying the estimated prices and consumption were not always similar to the results from historical data coming from 'purchases of energy' collected in Eurostat. A residual (the difference between the two) was therefore introduced in the analysis to take into account for these data discrepancies and ensure a coherent approach in the analysis of the energy costs in this document 31 .

The ratio between energy costs and total production costs is also a good indicator of the impact on total costs and competitiveness of a sector. In this section after the analysis of the factors driving the real energy costs (numerator of the ratio) we also look at the factors that drive the production costs (denominator) so we can also assess the actual evolution of the ratio.

6.4.1 Drivers of energy costs (Purchases of energy) 

Using the LMDI decomposition, the key drivers of energy costs can be identified.

 

The analysis in this section aims to use LMDI decomposition to explain the behaviour of the energy costs observed as energy purchases of energy from SBS data. Thus, a residual is introduced in the analysis to account for the difference between estimated energy costs and the SBS data for energy purchases of energy. For the purposes of this analysis, the change in energy costs over time is defined as follows:



Where

¾Output effect: the effect of changes in real production (GVA),

¾(Real) Energy intensity effect: the effect of changes in energy per unit of real output (GVA) over time due to energy efficiency measures, behavioural changes and industry structural change;

¾Price effects: the effect of changes in coal, gas and electricity prices.

¾The residual, which includes the effect of unexplained data discrepancy with Eurostat SBS data on 'purchases of energy'.

Box – Interpretation of results

The interpretation of some of these effects is complex in some cases and may deserve some additional explanations.

The unexplained residual likely arises from missing data, in particular, on energy consumption. In these cases, data gaps were filled using sectoral energy-intensity figures for those countries where data is available. In some cases, that meant relying on trends of very few countries (Germany and few others) to predict the wider sectoral trends at the EU28 level. Therefore it is possible that the residual is partly reflecting some energy intensity effects that were impossible to identify from the limited energy consumption data available. Industry restructuring is indeed typical in periods following economic crises and intense international competition as the one suffered for some of the analysed energy intensive industries in the period under study. On the other hand, the residual was calculated as the difference from the Eurostat SBS data in order to ensure a coherent analysis in this section in line with the analysis on the previous sections of the chapter. However Eurostat SBS data could also present some inconsistencies as it is based on surveys which might also be partially incomplete.

The interpretation of the price effect is more complex that it seems and also deserves further explanations. The price effect captures the effect of changes in weighted-average energy prices on energy costs faced by firms. The prices used are nominal and exclude all recoverable tax and levies (such as VAT). The price effects are estimated by combining (fixed) estimates of the energy mix at a sectoral level and estimates of energy prices (by fuel) over the period 2010-2015. Energy price for each sector and fuel is estimated by using the Eurostat production price band in which most industrial production would fall into 32 . Therefore the price effect does not capture the behaviour of other fuel prices (price of biomass or heat) which are assumed to behave in line with the weighted average from coal, gas and electricity prices. Finally, price for each industry sector at the EU28 level, the Member State level prices are weighted by the total value of production (by Member State). Thus, the EU28 level results for each industry sector reflect a double-weighting of price: (i) (fixed) fuel shares used to derive a representative weighted-average fuel price for each industry and each Member State (ii) (dynamic) Member State production shares used to weight the Member State -level price effects, to derive an EU average price effect for each industry sector. The latter means that prices changes can be due to changes on the production structure of the sector at EU level (shifts of production across MS) which results on changes on the weights used for calculating the prices.



Results of the analysis of energy cost drivers at EU level

At aggregate level across all the manufacturing sectors analysed, the change of energy costs over the period 2010-2015 fell by 8%. This was the result of the following combined effects:

·energy price increases contributed to an increase of 7% in energy costs;

·real output changes had an impact close to zero on energy costs;

·lower energy-intensity contributed to energy savings that reduced energy costs by 4%.

·The residual (unidentifiable factors) drove energy costs down by 10%

Figure 127 - Drivers of energy costs (absolute changes)

Source: Trinomics et altri

Results of the analysis energy costs drivers by sectors

Figure 128 summarises the analysis of the impact of energy cost drivers by sectors. It shows that the magnitude of the various effects across sectors is very diverse. The value of the residual also varies significantly across sectors signalling which sectors have the most robust estimates of the effects. A more detailed analysis by sector can be found in Annex 2.

 

Figure 128 - Drivers of change in energy costs in EU manufacturing sectors over 2010-2015 (%)

NB: Industry sectors are ordered according to energy intensity

The price effect was positive across almost all industry sectors analysed contributing to a 5%-10% increase in current energy costs over the period 2010-2015. The price effect mainly reflects increases in prices of electricity, which accounts for the largest share of energy consumption for many of the sectors studied. Gas is also important for many energy-intensive industries but gas prices remained roughly stable over the period.

The price effect were modest (contributed a 0-5% increase in energy costs) in sectors like Beverages (C11); Weaving of textiles (C132); Cement (C235), Fabricated metal products, except machinery and equipment (C25); Basic chemicals (C201). In these sectors, the low price effect is largely because production took place in Member States where energy price rises were more modest and/or production shifted to Member States where energy prices are lower 33 . The decline on oil prices over the period also contributed to bring down the price effect in oil intensive sectors like Basic chemicals (C201) 34 and particularly Cement (C235). The largest price effect was estimated in Computer, electronic and optical products (C26) which relies on electricity very importantly (80% of the sector's energy costs) and is prominently located in Member States which experienced significant electricity price rises over 2010-2015 35 .

The real output effect was negative for most of the more energy-intensive industry sectors and positive for most of the less energy-intensive industry sectors, reflecting that for high energy-intensity industry sectors reductions in energy costs are closely linked to lower economic activity. The sectors (and Member States) that saw the largest falls in real output over 2010-2015 include Basic chemicals (C201) in France; Man-made fibres (C206) in the UK, the Netherlands, Belgium and Spain; Cement (C235) in Spain, Italy and Greece; Cutting stone (C237) in Italy and Spain.

The largest positive output effects appeared in sectors which are less affected by energy price increases like the Motor vehicles (C29) and Other transport equipment (C30), where rises in real output 36 contributed to a 23% and 10% increase in energy costs, respectively.

Energy-intensity effects suggest that, in most cases, the energy intensity of manufacturing industries fell over the period. These results have however to be taken with precaution as, in many cases, they are reliant on data (or estimations) from very few countries 37 . The negative energy intensity effect can represent improvements in energy efficiency (due to behavioural change or investments in more energy efficient processes in response to higher prices or policies 38 ) but also structural changes within sectors (as there can be considerable heterogeneity at the level of aggregation studied, i.e. NACE 3-digit level) 39 . It is interesting to see that reduction in (real) energy intensity appeared systematically among the ‘less energy-intensive’ sectors while it was not always the case for the 'most energy-intensive sectors'. This however might be due to the fact that data for the most energy-intensive industry sectors was sparser.



6.4.2 Impact of energy costs on Total Production Costs 

   

Based on Eurostat SBS data for energy purchases and total production costs, this section assesses the results of the decomposition of total production costs in order to assess the extent to which total production costs were driven by changes in energy costs.

 

The result of the analysis by Trinomics et altri estimates that, at aggregated level, energy costs had a close to zero impact on increasing total production costs over the period of study (2010-2015).

At sector level, the impact of changes in energy costs on total production costs was more diverse (See Table 12 ), ranging from -8% to +1%. In almost all cases the impact of energy costs was smaller than other cost drivers. Energy costs drove significant reductions in total costs of production in Cement (-8%), Basic iron and steel (-3%), Paper (-2%), Man-made fibres (-2%), Clay building materials (-1%), Basic chemicals (-1%), Casting of metals (-1%) and Cutting of stone (-1%). Cement was the only sector in which energy costs explained more than half of the reduction in total production costs, with energy costs falling primarily due to reductions in output. Among the less energy intensive sectors, energy costs (mainly driven by efficiency improvements, structural change and lower real output) had a negative impact on production costs only for Weaving of textiles (-2%).

Table 12- Drivers of total production costs in manufacturing sectors

Code

Description

Main energy carrier used by sector

Energy cost effect

Other cost effect

Total effect

High energy-intensity sectors

C235

Manufacture of cement, lime and plaster

Oil

-8%

-7%

-15%

C233

Manufacture of clay building materials

Natural Gas

-1%

2%

1%

C171

Manufacture of pulp, paper and paperboard

Natural Gas

-2%

9%

7%

C231

Manufacture of glass and glass products

Natural Gas

0%

9%

9%

C241

Manufacture of basic iron and steel and of ferro-alloys

Natural Gas

-3%

-8%

-10%

C206

Manufacture of man-made fibres

Natural Gas

-2%

-7%

-9%

C232

Manufacture of refractory products

Natural Gas

0%

-3%

-3%

C201

Manufacture of basic chemicals, fertilisers and nitrogen compounds, plastics and synthetic rubber in primary forms

Natural Gas

-1%

7%

7%

C239

Manufacture of abrasive products and non-metallic mineral products n.e.c.

Natural Gas

1%

10%

11%

C245

Casting of metals

Electricity

-1%

11%

11%

C234

Manufacture of other porcelain and ceramic products

Natural Gas

0%

14%

14%

C192

Manufacture of refined petroleum products

Oil (chemical feedstock); Natural Gas (energy input)

1%

-15%

-14%

C244

Manufacture of basic precious and other non-ferrous metals

Electricity

0%

17%

17%

C237

Cutting, shaping and finishing of stone

Electricity

-1%

-20%

-21%

C161

Sawmilling and planing of wood

Electricity

0%

22%

22%

Lower energy-intensity sectors

C106

Manufacture of grain mill products, starches and starch products

Natural Gas

0%

23%

23%

C222

Manufacture of plastics products

Electricity

0%

18%

18%

C172

Manufacture of articles of paper and paperboard

Natural Gas

0%

11%

11%

C103

Processing and preserving of fruit and vegetables

Natural Gas

0%

25%

26%

C11

Manufacture of beverages

Natural Gas

0%

6%

6%

C132

Weaving of textiles

Electricity

-2%

-2%

-3%

C25

Manufacture of fabricated metal products, except machinery and equipment

Electricity

0%

9%

9%

C21

Manufacture of basic pharmaceutical products and pharmaceutical preparations

Natural Gas

0%

17%

17%

C32

Other manufacturing

Electricity

0%

15%

15%

C33

Repair and installation of machinery and equipment

Electricity

0%

14%

14%

C27

Manufacture of electrical equipment

Electricity

0%

10%

9%

C28

Manufacture of machinery and equipment n.e.c.

Electricity

0%

22%

22%

C26

Manufacture of computer, electronic and optical products

Electricity

0%

-6%

-5%

C30

Manufacture of other transport equipment

Natural Gas

0%

28%

28%

C29

Manufacture of motor vehicles, trailers and semi-trailers

Electricity

0%

42%

42%

Source: Trinomics et altri study

6.4.3Drivers of the energy costs as a share of production costs 

The energy costs shares in total production costs () is a useful measure for assessing energy cost impacts. It is therefore interesting to see how this ratio has been affected by the dynamics in energy costs (the numerator of the ratio) and total production costs (the denominator).

Table 13 - Changes in energy costs and total production costs by sector

Code

Description

Main energy carrier used by sector

Change in energy costs (%)

Change in total production costs (%)

Percentage point change in ratio of energy costs in total costs

High energy-intensity sectors

C235

Manufacture of cement, lime and plaster

Oil

-37%

-15%

-5.8 pp

C233

Manufacture of clay building materials

Natural Gas

-5%

1%

-0.7 pp

C171

Manufacture of pulp, paper and paperboard

Natural Gas

-19%

7%

-2.7 pp

C231

Manufacture of glass and glass products

Natural Gas

0%

9%

-0.7 pp

C241

Manufacture of basic iron and steel and of ferro-alloys

Natural Gas

-29%

-10%

-2.0 pp

C206

Manufacture of man-made fibres

Natural Gas

-27%

-9%

-1.6 pp

C232

Manufacture of refractory products

Natural Gas

-5%

-3%

-0.2 pp

C201

Manufacture of basic chemicals, fertilisers and nitrogen compounds, plastics and synthetic rubber in primary forms

Natural Gas

-11%

7%

-1.1 pp

C239

Manufacture of abrasive products and non-metallic mineral products n.e.c.

Natural Gas

15%

11%

+0.2 pp

C245

Casting of metals

Electricity

-9%

11%

-1.1 pp

C234

Manufacture of other porcelain and ceramic products

Natural Gas

1%

14%

-0.6 pp

C192

Manufacture of refined petroleum products

Oil (chemical feedstock); Natural Gas (energy input)

31%

-14%

1.3 pp

C244

Manufacture of basic precious and other non-ferrous metals

Electricity

-3%

17%

-0.7 pp

C237

Cutting, shaping and finishing of stone

Electricity

-23%

-21%

-0.1 pp

C161

Sawmilling and planing of wood

Electricity

5%

22%

-0.5 pp

Lower energy-intensity sectors

C106

Manufacture of grain mill products, starches and starch products

Natural Gas

11%

23%

-0.3 pp

C222

Manufacture of plastics products

Electricity

6%

18%

-0.3 pp

C172

Manufacture of articles of paper and paperboard

Natural Gas

-10%

11%

-0.6 pp

C103

Processing and preserving of fruit and vegetables

Natural Gas

10%

26%

-0.4 pp

C11

Manufacture of beverages

Natural Gas

-1%

6%

-0.2 pp

C132

Weaving of textiles

Electricity

-44%

-3%

-1.5 pp

C25

Manufacture of fabricated metal products, except machinery and equipment

Electricity

-8%

9%

-0.4 pp

C21

Manufacture of basic pharmaceutical products and pharmaceutical preparations

Natural Gas

6%

17%

-0.1 pp

C32

Other manufacturing

Electricity

-10%

15%

-0.3 pp

C33

Repair and installation of machinery and equipment

Electricity

-3%

14%

-0.2 pp

C27

Manufacture of electrical equipment

Electricity

-9%

9%

-0.2 pp

C28

Manufacture of machinery and equipment n.e.c.

Electricity

4%

22%

-0.1 pp

C26

Manufacture of computer, electronic and optical products

Electricity

2%

-5%

+0.1 pp

C30

Manufacture of other transport equipment

Natural Gas

13%

28%

-0.1 pp

C29

Manufacture of motor vehicles, trailers and semi-trailers

Electricity

21%

42%

-0.1 pp

Source: Trinomics et altri study

Note: Energy costs are taken from the ‘Purchases of Energy Products’ data (from Eurostat SBS). Other costs comprise ‘personnel costs’ and ‘costs of goods and services, net of energy costs', calculated from the Eurostat SBS data. Results are rounded to the nearest percentage point and cases where the energy cost effect and the other cost effect do not sum to the total effect are due to rounding.

Table 13 shows that the energy costs shares in total production costs have fallen among nearly all sectors over the period of study. There was a wide variation in changes in energy costs (the numerator) over the period, with around half of the sectors experiencing increases in energy costs and the other half experiencing decreases. This means that, even though energy costs have increased among some sectors, they have not increased by as much as other non-energy costs of production resulting in lower energy costs shares. For most of the less energy-intensity industries, the ratio of energy costs in total production costs fell by -0.1pp to -0.6pp. For the more energy intensive sectors, there were typically larger reductions in the ratio of energy costs to total production costs, particularly in Cement (-5.8pp), Paper (-2.7pp), Basic iron and steel (-2 pp), Man-made fibres (-1.6pp) and Weaving of textiles (-1.5pp).

Figure 129  shows that the only sector in which energy costs increased at a faster rate than other non-energy costs of production over the period were Computer products, abrasive non-metallic products and refinery products.

Reduction in energy costs over 2010-2015

Increase in energy costs over 2010-2015

Energy costs grew at a slower rate than non-energy costs of production

·Fabricated metal products (C25)

·Repair and installation of machinery (C33)

·Cutting and shaping stone (C237)

·Weaving of textiles (C132)

·Beverages (C11)

·Refractory products (C232)

·Cement, lime and plaster (C235

·Clay building materials (C233)

·Electrical equipment (C27)

·Other manufacturing (C32)

·Paper and paperboard (C172)

·Pulp, paper and paperboard (C171)

·Glass (C231)

·Iron and steel (C241)

·Basic and non-ferrous metals (C244)

·Chemicals (C201)

·Man-made fibres (C206)

·Casting of metals (C245)

·Other transport equipment (C30)

·Fruit and vegetables (C103)

·Plastics products (C222)

·Grain mill products, starches (C106)

·Motor vehicles (C29)

·Sawmilling and planing of wood (C161)

·Machinery and equipment n.e.c. (C28)

·Pharmaceutical products (C21)

·Other porcelain and ceramic (C234)

Energy costs grew at a faster rate than non-energy costs of production

-

·Computer, electronic and optical C26)

·Abrasive non-mettalic minerals(C239)

·Refined petroleum products (C192)

Figure 129 - Classification of sectors by comparing the dynamics of energy costs dynamics vs productions costs

Source: Trinomics et altri study

6.5 International comparisons

International competitiveness is a relative matter. We need to compare the energy costs to the EU trading partner to assess what is their actual impact on the cost-competiveness of our industry. The same applies to the energy efficiency indicators which can influence the relative energy consumption and therefore the energy costs. Finally, prices of energy products should also be compared as they are usually the main drivers of the energy costs. While data on international prices is relatively robust, the data available for energy costs and energy efficiency is rather limited and the results of the latter should be taken with certain caution.

The chapter compares retail industrial prices for the EU industry with those in G20 Members. It relies on the results of the Trinomics et altri study. It focuses on the international comparisons of prices for electricity and gas, which are the most relevant for energy costs in most of manufacturing. International comparisons on oil products prices can be also found in section 3.3.7 as they are also relevant for the energy costs in some specific manufacturing sectors and non-manufacturing sectors.

6.5.1 Energy costs vs other G20 countries 

In this section the energy costs of EU sectors are compared to those in main EU trading partners in order to assess the competitiveness of EU sectors. Unfortunately, as noted in the previous sections, specific data on energy cost shares is relatively limited across the main non-EU G20 partners.

Figure 130  shows a comparison of energy costs shares in production costs for the sectors and countries for which equivalent energy cost and production cost data were found.

 

Figure 130 - International comparision of energy costs shares for selected energy intensive sectors

Source: Trinomics et altri study

The data for the available sectors show that EU shares in production costs are comparable to those of the US except for steel and non-ferrous metals. As to Japan, costs shares are higher in the EU energy costs shares in the 3 sectors for which data was available. These results are in line with those found in the 2016 energy prices and costs report where comparisons could be made for a larger number of sectors and were pointing to higher energy costs shares in energy intensive industries in the EU compared to Japan and similar energy costs shares in the EU and the US, with the exception of sectors subjected to higher international competitive pressure, for which EU cost shares were lower.

By using production value rather than production costs as the basis of the comparison the international analysis can be expanded to include South Korea and other sectors (See Figure 131 and Figure 132 ). It is important to bear in mind that comparing energy costs to production value is not the same as comparing to production costs. Production value includes profits and this variable is much more volatile and also depends of other factors than production costs.

Figure 131 - Energy costs shares in production value for the most energy intensive sectors in manufacturing, 2008-2015

Source: Trinomics et altri study

Figure 132 - Energy costs shares in production value for other manufacturing sectors, 2008-2015

Source: Trinomics et altri study

When we look at the energy costs shares in production value we observe that in most cases the EU shares continue to be higher than in those Japan although in some cases EU shares become comparable or even lower (i.e. for computer and electronics; cement and refineries). The EU energy cost shares were also higher than those in South Korea for almost all the sectors (particularly much higher than in sawmills and transport equipment; and with the only exception of clay building materials).

As to Norway, the result of the comparison is mixed. For some sectors the energy cost shares are much lower than in the EU (Grain, Glass, Refractory Products) while for some other they are much higher (paper, chemicals, steel and non-ferrous metals).

EU energy cost shares were in general lower in the EU than those of Turkey with the exceptions of few sectors (fruit and vegetables, chemicals, abrasive products, non-ferrous metals and repair of machinery.

Some observations can be drawn on some of the sectors. For Grain products, the EU average energy cost share is higher than in the US (where production is highly mechanised) and Norway; it is however similar to that in Turkey. For sawmills costs in the EU are a little lower than those in the US on average, but higher than in the other countries. For Glass, the EU cost shares are lower than in the US and Turkey, but higher than in Norway. For steel and non-ferrous metals the EU energy cost shares are lower than in Norway and comparable to those in Turkey, but higher than those in the US and Japan (and, for non-ferrous metals, particularly much higher than in South Korea).

6.5.2Energy intensity of EU sectors vs other G20

Energy efficiency can also be factor for international competitiveness (the more energy efficient a firm is, the lower its relative consumption and energy costs). By comparing energy intensities across one can have an indication of the different energy efficiency in these compared sectors and countries. This complements the understanding of the role of energy cost shares. One should also be aware that the international data on energy intensity is rather limited (with often only one or two other international comparators available) and that these results should not be generalised.

Figure 133  and Figure 134 display the trends in energy intensity on the available sectors and countries. Although it is difficult to draw any general conclusions it can be observed that that:

¾Energy intensities in the EU compared to those in the US show considerable variation per sector for which data is available, with the EU being less energy intensive in beverages, glass, fabricated metal products, and the US being less energy intensive in chemicals, Man-made fibres and computers and electronics.

¾The EU is less energy intensive than China in every sector for which data is available (except for refineries for which the EU have the highest energy intensity of the countries for which data was available).

¾The EU sectors are also systematically less energy intensive that those in Turkey (note that the data set was the most complete)



Figure 133 - Energy intensity international comparisons for the most energy intensive manufacturing sectors

Source: Trinomics et altri study

Note: data limited for available sectors and countries

Figure 134 - Energy intensity international comparisons for other manufacturing sectors

Source: Trinomics et altri study

Note: data limited for available sectors and countries

6.5.3 Industrial electricity prices: EU vs G20 countries

This section compares retail industrial prices for electricity in the EU industry with those in G20 Members. It relies on the results of the Trinomics et altri study. International comparisons of electricity prices are very relevant for an assessment of cost competitiveness of sectors. Indeed electricity is the energy carrier which most often has the highest potential to impact the energy costs differential between energy intensive sectors in manufacturing.

Retail electricity prices for industry have relatively complete datasets. The price data covers EU28 and G20 countries from 2008-2018. EU28 prices are based on consumption band assumptions (mainly Eurostat consumption band ID) while data for non-EU G20 countries is normally based on the average of the countries (not based on consumption bands). The price data is however widely comparable (i.e. comparability checks were undertaken can be found in the study by Trinomics et altri). Finally, prices are exclusive of VAT and recoverable taxes and levies but include (non-recoverable) excise taxes and levies.

The main conclusions that can be drawn from this data are:

¾EU28 average prices increased from around 100 EUR/MWh in 2008 to 120 EUR/MWh by 2013-2014, but since then prices have slowly declined to around 110 EUR/MWh.

¾US prices are around half the EU average levels and have not changed significantly between 2008 and 2018.(See Figure 135 )

¾Prices in Japan were higher than the EU28 average but have declined since 2012 and converged in 2015-2016 to a broadly similar level .(See Figure 135 )

¾Prices in China began at a comparable level to EU prices but have declined since 2011 and thus diverged from EU levels. (See Figure 135 )

Figure 135 – Retail electricity prices for industry: EU vs China, Japan & US, 2008-2018

Sources: Eurostat, CEIC and IEA

¾Most other G20 countries (Canada, India, Russia, Mexico, South Korea, Saudi Arabia, and Turkey) also have lower prices than the EU average. Only Brazil has higher prices. Prices in South Korea are lower but increasing and converging to EU levels, whilst prices in Mexico (already below cost) are diverging as they have significantly decreased since 2014. (See Figure 136 )

Figure 136 - Retail electricity prices for industry: EU vs other G20, 2007-2018

Sources: Eurostat, CEIC, IEA, ERRA

For Argentina, Australia, India there is only information from price indices (and not absolute price data). The evolution of the indices indicate that EU average prices have increased by around 10% since 2008 (+1.1%/year) while real price indices declined in India and particularly in Argentina. The Australian price index has increased in real terms by more than 60% over the period. – See Figure 137

Figure 137 – Retail electricity indexes prices for industry: EU vs Argentina, Australia & India, 2008-2018

Sources: Eurostat, CEIC and IEA

¾Price differentials (in real prices 2017) did not evolve favourably for the EU in most of the cases as EU prices increased (by 10%) while prices decreased in most of the G-20 countries (with the exception of Indonesia and South Korea). As a result of this divergent evolution, the price gap with the US and China (which was already favourable for these two countries at the start of the period analysed) has widened slightly. The gap with Japan (which was favourable for the EU) has shrunk. The price gap with the EU increased for most of the other G-20 countries, or in the case of Brazil, prices that were initially higher converged towards EU levels. Only the price gap with South Korea evolved favourably for the EU (while the gap with Indonesia remained roughly stable) – See Table 14  


Table 14 - Changes in retail industrial electricity prices compared to EU prices, constant 2017 EUR/MWh

Country

Start price [EUR2017]

End price [EUR2017]

Change EUR

Change %

Start Gap [EUR]

End Gap [EUR]

Difference [EUR]

Impact for EU

EU28

101.33

112.05

10.72

10.6%

 

 

 

 

Argentina

 

 

 

 

 

 

 

 

Australia

 

 

 

 

 

 

 

 

Brazil

151.51

144.48

-7.03

-4.6%

50.18

32.43

-17.75

Negative

Canada

71.53

70.66

-0.87

-1.2%

-29.80

-41.39

-11.59

Negative

China

109.83

99.65

-10.18

-9.3%

8.50

-12.40

-20.90

Negative

India

 

 

 

 

 

 

 

 

Indonesia

65.01

71.30

6.29

9.7%

-36.32

-40.75

-4.43

Negative

Japan

140.15

134.58

-5.57

-4.0%

38.82

22.53

-16.29

Negative

Mexico

127.10

63.20

-63.90

-50.3%

25.77

-48.85

-74.62

Negative

Russia

64.00

56.33

-7.67

-12.0%

-37.33

-55.72

-18.39

Negative

Saudi Arabia

47.25

43.60

-3.66

-7.7%

-54.08

-68.46

-14.38

Negative

South Africa

 

 

 

 

 

 

 

 

South Korea

62.94

85.82

22.88

36.3%

-38.39

-26.23

12.16

Positive

Turkey

70.21

56.80

-13.41

-19.1%

-31.12

-55.25

-24.13

Negative

USA

63.85

58.71

-5.14

-8.1%

-37.48

-53.34

-15.86

Negative

Source: Trinomics et altri study.

Note: a positive impact for the EU is recorded if the price gap has improved over time, e.g. that if a country had lower prices initially the gap is now smaller or prices are higher than the EU average, or if a country had higher prices and that the gap has increased. A negative impact is recorded if a country had lower prices than the EU, and that the gap has now increased, or if the country had higher prices than the EU but this gap has narrowed or the country now has lower prices.

¾The analysis of the drivers of international prices (see Table 15 ) shows that monetary effects (inflation and exchange rate changes) played a very significant role in the evolution of nominal prices. High inflation played a key role in pushing up prices in countries like Brazil and Indonesia. In Russia and Turkey the effects of high inflation were mitigated or offset by exchange rates depreciations. Exchange rate appreciations also played a role in pushing up Chinese and US prices. EU prices increased by around 12% in real terms between 2008 and 2017 (1.3%/year). In contrast, real prices declined in Brazil, Mexico, China, the US and Saudi Arabia. However, national prices increased in real terms in the remainder of the G20 in a comparable or higher magnitude.

Table 15 - Factors in observed industrial retail electricity price changes per country, nominal prices, per MWh

Country

Start date

End date

Nominal Start price EUR

Change due to inflation [EUR]

Change due to price change in national currency [EUR]

Exchange rate effect [EUR]

Total change [EUR]

Nominal End price EUR

Change due to inflation [%]

Change due to real price change in national currency [%]

Exchange rate effect [%]

Total change [%]

EU28

2008-1

2017-11

92.80

8.20

11.05

0.00

19.25

112.05

8.8%

11.9%

0.0%

20.7%

Argentina

No data

 

 

 

 

 

 

 

 

 

 

 

Australia

No data

 

 

 

 

 

 

 

 

 

 

 

Brazil

2008-12

2016-12

111.68

71.01

-15.02

-14.87

41.12

152.80

63.6%

-13.4%

-13.3%

36.8%

Canada

2008-1

2016-1

48.18

4.02

23.20

-2.86

24.36

72.55

8.3%

48.2%

-5.9%

50.6%

China

2008-1

2017-12

62.87

14.43