EUROPEAN COMMISSION

Brussels, 2.6.2021

SWD(2021) 409 final

COMMISSION STAFF WORKING DOCUMENT

In-Depth Review for The Netherlands



















in accordance with Article 5 of Regulation (EU) No 1176/2011 on the prevention and correction of macroeconomic imbalances

Accompanying the

COMMUNICATION FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN CENTRAL BANK, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE, THE COMMITTEE OF THE REGIONS AND THE EUROPEAN INVESTMENT BANK

Economic policy coordination in 2021: overcoming COVID-19, supporting the recovery and modernising our economy

{COM(2021) 500 final} - {SWD(2021) 401 final} - {SWD(2021) 402 final} - {SWD(2021) 403 final} - {SWD(2021) 404 final} - {SWD(2021) 405 final} - {SWD(2021) 406 final} - {SWD(2021) 407 final} - {SWD(2021) 408 final} - {SWD(2021) 410 final} - {SWD(2021) 411 final} - {SWD(2021) 412 final}

Contents

Executive summary

1.Assessment of Macroeconomic Imbalances

2.Thematic Issue: External Sector

3.Thematic Issue: Private Indebtedness and Housing

LIST OF Tables

Table 1.1:Outward spillover heat map for The Netherlands

Table 1.2:Assessment of Macroeconomic Imbalances Matrix - The Netherlands

Table 1.3:Selected economic and financial indicators, The Netherlands

Table 2.1:Selected external sector indicators, The Netherlands

Table 3.1:Household debt indicators, The Netherlands

Table 3.2:Selected housing indicators, The Netherlands

LIST OF Graphs

Graph 2.1:    Thematic Graphs: External sector    

Graph 3.1:    Thematic Graphs: Private indebtedness and housing    

Graph 3.2:    Thematic Graphs: Private indebtedness and housing (cont.)    

LIST OF Boxes

Box 1.1:    Spillovers The Netherlands    

Executive summary

The 2021 Alert Mechanism Report concluded that an in-depth review should be undertaken for the Netherlands to examine further the persistence of imbalances or their unwinding. In February 2020, under the previous annual cycle of surveillance under the Macroeconomic Imbalances Procedure, the Commission identified “macroeconomic imbalances” in the Netherlands. These imbalances related to the large current account surplus and high private-sector debt. The analysis shows that these vulnerabilities remain. It should be noted that the context of the assessment of vulnerabilities in this year’s in-depth review (IDR) for the Netherlands is markedly different from last year. Also, the evolution of the COVID-19 pandemic, the strength of the recovery, and possible structural implications of the crisis are all still surrounded by high uncertainty, requiring caution in the assessment. In general, policy action over the past year focused on cushioning the impact of the COVID-19 shock and facilitating the recovery. This should support adjustment in the medium-term. Looking forward, the Recovery and Resilience Plan provides an opportunity to address imbalances, investment and reforms needs.

Main observations and findings of this IDR analysis are:

·This IDR is informed by the 2021 spring forecast, which expects a recovery in economic activity in the Netherlands with the easing of the COVID-19 crisis. After the steep drop of 3.7% in 2020, real GDP is projected to increase by 2.3% this year and 3.6% next year, allowing the economy to recover its pre-pandemic level by the end of this year.

·The current account surplus declined to 7.8% of GDP in 2020 but is expected to remain elevated. The COVID-19 shock caused considerable shifts in sectoral positions. The savings surplus in the household and corporate sectors widened, but that was more than offset by the government sector which moved sharply into net borrowing territory due to the implementation of crisis-related fiscal support measures. Despite its recent decline, the overall savings surplus remains high both relative to fundamentals and by international comparison. Despite an ongoing pension reform and recent tax changes addressing incentives to retain earnings within SMEs, the structural drivers underpinning high household and corporate savings remain in place. The balance-of-payments picture also points to the current account surplus remaining at an elevated level going forward, as the strong recovery in global goods trade supports net exports.

·The private debt-to-GDP ratio increased somewhat in 2020, to 235%, and is expected to remain high. While both headline corporate and household debt levels are high, a large share of corporate debt represents intra-group debt of multinationals. Household debt is primarily driven by mortgage debt, and is expected to continue growing at a somewhat accelerated pace in nominal terms as a result of recent sharp house price rises. The maximum applicable rate of mortgage interest deductibility is being reduced by 3 percentage points per year until 2023, but a substantial subsidy for mortgage borrowing remains. Measures to discourage buy-to-let investment may further reinforce existing policy distortions favouring owner-occupancy over rental housing and thus risk undermining efforts to boost the private rental sector.

1.    Assessment of Macroeconomic Imbalances

Introduction

In February 2020, over the previous annual cycle of surveillance under the Macroeconomic Imbalances Procedure, the Commission identified “macroeconomic imbalances” in the Netherlands. These imbalances related to the large current account surplus and high private-sector debt linked to distortions in the housing market. The 2021 Alert Mechanism Report published in November 2020 concluded that a new in-depth review (IDR) should be undertaken for the Netherlands with a view to assessing the persistence or unwinding of imbalances.

The context of the assessment of vulnerabilities this year is markedly different from last year's IDRs, which took place before the COVID-19 pandemic. The evolution of the pandemic, the strength of the recovery, and possible structural implications of the crisis are still surrounded by high uncertainty requiring caution in the assessment. Policy action over the past year focused on cushioning the impact of the COVID-19 shock and on facilitating the recovery. Follow-up to country-specific recommendations from 2019 and 2020, including those that are MIP-relevant, is taking place in the context of the assessment of the Recovery and Resilience Plans (RRPs). The analysis of policies in the present report was finalised before the formal submission of RRPs and does not draw on information included in RRPs. It is therefore without prejudice to the Commission’s assessment of RRPs, which is ongoing at the time of publication of this report.

The assessment follows a similar structure as the IDRs that were included in Country Reports in recent annual cycles. This chapter presents the main findings for the assessment of imbalances, also summarised in the MIP assessment matrix. The assessment is backed by selected thematic chapters that look more at length at the external position and private debt developments. Spillovers and systemic cross‑border implications of imbalances are also taken into account. In addition, assessments of structural issues made in previous IDRs and in the context of fiscal assessments are also considered.

Macroeconomic context

The Dutch economy is forecast to rebound swiftly to its pre-pandemic level on the back of a robust recovery in consumer demand and more favourable external factors. The economy is expected to grow by 2.3% and 3.6%, in 2021 and 2022, respectively, the latter also driven by strong carry-over effects. By the end of the forecast horizon, Dutch real GDP is forecast to be 2.2% above its level at the end of 2019. The output gap is set to improve significantly but still remain slightly negative, at -0.2% in 2022. Following a decline to 7.8% of GDP in 2020, the current account surplus is forecast to gradually increase to 8.7% of GDP in 2022. The gross saving rate of households is forecast to remain above its long term average of around 16%, albeit declining gradually to around 18% in 2022. The unemployment rate is forecast to remain relatively low, at 4.3% and 4.4% in 2021 and 2022, respectively, compared to 3.4% in 2019. HICP inflation is set to increase from 1.1% in 2020 to around 1½% in 2021 and 2022, reflecting a recovery in energy prices and the services sector.

The economic recovery is expected to be broad-based, gaining steam in the second half of this year. The implementation of the vaccination strategy and the following easing of confinement measures are set to lead to a rebound in private consumption, also to the benefit of the recovery in the contact-intensive services sector. The rebound in 2021 is also projected to be supported by robust investment dynamics on the back of a strong manufacturing sector, where previously postponed investment plans are being revived. Net exports are also expected to contribute substantially to the near-term recovery. In 2022, the pace of growth is expected to slow down in view of the expected withdrawal of the emergency support measures, a limited increase in unemployment and subdued wage growth. Risks to the outlook are broadly balanced. On the downside, possible slower pace of vaccination and new variants of the virus could delay the easing of containment measures. On the upside, stronger than projected global growth, particularly in the US, could have a more positive impact on the recovery.

Imbalances and their gravity

Despite its recent decline, the current account surplus remains high by international standards and relative to fundamentals. Following a peak of 10.8% in 2018, the current account surplus narrowed slightly to 9.9% in 2019. However, it remained among the highest in the euro area as a share of GDP and is well above fundamentally justified levels. A persistently high trade surplus in goods is the main driver from a trade perspective.

Prior to the COVID-19 shock, all domestic sectors were in surplus, with the corporate sector making the largest contribution to net lending. Both financial corporations and non-financial corporations recorded a structural surplus, with the latter being the main driver. Net lending by non-financial corporations amounted to 4.7% of GDP in 2019. Compared to the rest of the euro area, profitability and net property income were relatively high for Dutch firms, whereas domestic investments were lower. This was linked in part to the large presence of multinationals in the Netherlands, but small and medium-sized enterprises were also significant contributors to net lending. Households have been recording surpluses since the global financial crisis, amounting to 2.3% of GDP in 2019. The dip in the housing market following the crisis initially led to a decrease in residential investment, while at the same time boosting personal savings via deleveraging pressures linked to high household debt. Pension funds were also an important driver of household net lending due to relatively high second-pillar pension contributions which are largely invested abroad. Before the COVID-19 pandemic, the government sector also recorded a headline surplus (of 1.7% of GDP in 2019), driven by past consolidation measures and increasing tax revenues.

Private debt continued to decline in 2019, but remained high. It reached 234% of GDP in 2019, down from 244% the year before. Non-financial corporate debt accounted for 134% of GDP. However, around 60% of this debt is owed by multinationals and largely consists of intra-group loans, implying limited macro-economic risks. The household debt ratio stood at 100% of GDP in 2019 – among the highest in euro area and well above relevant benchmarks. Although household debt continued to increase in nominal terms, growth remained relatively muted at around 1-1.5% annually since the 2008-2009 financial crisis. Household debt largely consists of mortgage debt, fuelled by mortgage interest tax deductibility coupled with distortions in the housing market, including an underdeveloped rental market and an overall housing shortage due to construction persistently falling short of demographic requirements. Real house prices continued to rise in 2019 (by 4.8% in real terms) to levels beyond what fundamentals such as income would suggest (overall valuation gap is +10% but fundamental-based gap is +20%).

Cross-border spillovers to other EU countries are relatively moderate given the size of the Dutch economy. Table 1.1 shows that exports to the Netherlands constitute a relatively large share of GDP for Belgium (14%). From the Dutch perspective, Germany is the largest export destination. On the financial side, Belgium, Cyprus, Ireland, Luxembourg and Malta have a relatively high exposure to the Netherlands. Box 1.1 shows an analysis of Dutch cross-border demand on other Member States' value added in 2020, confirming significant adverse impact on Belgium.

Evolution, prospects and policy response

The current account surplus declined further to 7.8% of GDP in 2020 but is expected to increase somewhat again in the coming years. The COVID-19 shock caused considerable shifts in sectoral positions. The savings surplus in the private sector widened as household savings were boosted by curtailed consumption combined with income support measures, and the corporate sector’s savings surplus also slightly increased as firms cut back both earnings distributions and investment. However, this was more than offset by the government sector which sharply moved into net borrowing territory due to fiscal support measures. Despite its recent decline, the overall savings surplus remains high both relative to fundamentals and by international comparison. The structural drivers underpinning high household and corporate savings remain in place, notwithstanding the ongoing reform of the second-pillar pension system and recent tax reforms addressing incentives to retain earnings within SMEs, The balance-of-payments picture also points to a somewhat wider current account surplus going forward, as the strong recovery in global goods trade supports net exports.

The private debt-to-GDP ratio increased somewhat in 2020, to 235%, and is expected to remain high. In nominal terms, household debt growth appears likely to pick up somewhat as a result of sharp house price rises in recent years, and in particular in 2020, when price growth accelerated again (to 7.3% in real terms as of Q3). However, the household debt-to-GDP ratio is likely to remain roughly stable over the next few years due to denominator effects, as economic activity gradually recovers from the COVID-19 crisis. In terms of policy measures, the maximum applicable rate of mortgage interest deductibility is being reduced by 3 percentage points per year. However, the eventual rate of 37%, to be reached in 2023 still implies that a substantial subsidy for mortgage borrowing remains. Measures to discourage buy-to-let investment may further reinforce existing policy distortions favouring owner-occupancy over rental housing and thus risk undermining efforts to boost the private rental sector. This could further increase upwards price pressure for owner-occupied homes.

Overall assessment

The Netherlands has recorded persistently large current account surpluses and high private debt levels. The headline current account surplus is largely driven by non-financial corporations, with relatively high savings and low domestic investment. Both large corporations, including multinationals, and small and medium-sized enterprises have substantial surpluses. Households also make a sizable contribution (particularly when correcting for statistical distortions linked to the treatment of retained earnings), among other things due to high mandatory pension contributions. Household debt as a share of GDP is around 50 percentage points higher than the euro area average and well above the relevant benchmarks (see subsection 4.2.3), as tax incentives encourage households to take on mortgage debt. While household debt is coupled with substantial housing and pension assets, these assets are often illiquid, leaving households vulnerable to shocks.

The external surplus and private debt level are expected to remain elevated. The current account surplus is expected to remain well above the threshold, as the strong recovery in global goods trade supports net exports and the structural factors underpinning high private-sector savings surpluses remain in place. Household debt is expected to continue growing in nominal terms, linked to sharp house price rises. Despite recent policy steps, strong incentives to take on mortgage debt remain, also against the background of an underdeveloped private rental market.

2.    Thematic Issue: External Sector

The Netherlands has been recording a sizable current account surplus for more than two decades. It reached a peak of 10.8% in 2018, before narrowing to 9.9% in 2019 and further to 7.8% in 2020 under the effect of the crisis (Table 1). The structural surplus in the balance of trade in goods accounts for nearly all of the surplus, while net exports of services and the primary income balance are much lower, though more volatile (Graph 2.1a). ( 1 ) The decline in 2020 was due to declining income account balances, both primary and secondary, that more than offset marginal improvements in the trade balance.

The longstanding current account surplus can partly be explained by the economy’s fundamental characteristics. The current account surplus explained by the fundamentals (‘current account norm’) ( 2 ) equalled 4% of GDP in 2019 (Table 2.1), with important contributions coming from high per capita output (relative to other countries), as well as from the financial centre status of the Netherlands. The fact that the Dutch surplus has consistently exceeded the norm points to a number of structural factors closely linked to savings and investment decisions, including compulsory savings via the second-pillar pension system combined with mortgage repayments, relatively low construction investment and private credit growth and the comparatively high structural fiscal balance (Graph 2.1b). ( 3 )

Prior to the COVID-19 crisis, all institutional sectors added to the substantial net lending position of the economy. Following a switch from net borrowing to lending by households in 2009, all private institutional sectors have consistently been recording a saving surplus, leading to an overall savings surplus between 5% and 11% of GDP over the last decade (Graph 2.1c). The government’s position moved from considerable borrowing of around 5% of GDP after the financial crisis, into a surplus over the last three years before the COVID-19 pandemic. While high net lending could point to underinvestment domestically, the comparison of aggregate investment with the euro area average does not point to a significant investment gap (Graph 2.1d). ( 4 ) Thus, high net lending is primarily explained by excess savings, with some important specific drivers across the corporate and household sectors.

The largest contribution to the headline savings surplus stems from the corporate sector. On average, the net lending of the (non-financial) corporate (NFC) sector (Graph 2.1c), accounted for 80% of the economy’s saving surplus since the early 2000s. Within the sector, the net lending of multinational enterprises (MNEs) amounted to 4% of GDP on average, exhibiting considerable volatility ( 5 ), while the surplus of small and medium companies has been more stable, ranging between 2% and 3% of GDP. High net lending by MNEs arises due to their global profits, which are attributed to their Dutch headquarters, even though these are largely being retained and invested abroad. Thus, their (global) earnings exceed (domestic) investment leading to a structural saving surplus. As for the small and medium companies, especially those with just a few shareholders, they typically distribute only a small portion of their profits, likely due to fiscal incentives to retain earnings and defer or avoid tax payments. ( 6 )

Households’ net lending position turned positive after the global financial crisis. On average, it equalled slightly less than 3% of GDP since 2009. Its evolution was driven by housing market developments (low housing investment following the crisis and substantial deleveraging of households), relatively weak wage and disposable income growth following the 2008-2009 financial crisis and large savings in the form of relatively high (second-pillar) pension contributions, which are widely invested abroad (European Commission, 2020). The pension savings aspect is of a more structural nature and also reveals that the contribution of households to the overall savings surplus is higher than recorded in the headline statistics on sectoral net lending. This is because the Dutch pension funds, whose ultimate beneficiaries are households, are large portfolio investors in foreign companies with substantial retained earnings that remain attributed to their home country in national accounts until they are distributed. ( 7 ) Even without adjusting for these statistical distortions, the household (gross) saving rate was consistently higher than the saving rate of the euro area following the global financial crisis, with the difference exceeding 3 percentage points of gross disposable income over most of the period (Graph 2.1e). ( 8 )

The Covid-19 shock caused considerable shifts in sectoral positions, but overall net lending remained large in 2020. The general government moved to a substantial deficit related to fiscal support measures, mainly to paid subsidies on production. As a result, the government sector net borrowing amounted to 4.3% of GDP in 2020, versus net lending of 1.7% of GDP in 2019. This decline has partly been offset by a higher household savings surplus, which rose from 2.3% to 6.0% of GDP in 2020, mostly due to higher savings amid a strong decline in consumption (especially in 2020Q2), but partly also due to increased disposable income (mainly linked to support measures). Over the same period, the NFCs’ saving surplus expanded from 4.7% to 5.7% of GDP, despite a large negative contribution from the oil sector amid low oil prices (see also footnote 5). The overall corporate sector recorded a milder increase from 5.9% to 6.1%, as financial corporations’ surplus declined to 0.4% of GDP in 2020, versus 1.1% in 2019. The NFCs’ operating surplus slightly increased (in nominal terms) during 2020, supported by higher subsidies on production, which more than offset the decline in their value-added. Their savings recorded a stronger increase as companies cut back shareholder pay-outs to preserve liquidity and investments have been reduced due to increased uncertainty. ( 9 ) Based on the Commission Spring Forecast, the corporate net lending position is projected to remain large at roughly the same levels as before the COVID-19 pandemic. Household and general government positions are expected to gradually move toward their pre-crisis levels, so that the overall net lending position should increase somewhat over the forecast horizon.

The impact of recent policy steps on the structural drivers of the savings surplus is uncertain. An agreement between social partners and government on a major reform of the second-pillar pension reform is currently going through the legislative process and will be phased in from January 2023, with full implementation envisaged by 2027. While this addresses some key challenges with the existing system and lays the groundwork for more individual flexibility, it will not directly reduce the high level of compulsory household savings via the pension system. ( 10 ) Moreover, in the meantime, pension contributions have actually been raised in 2020; this implies a further increase of the household savings surplus in the near-term. With regard to the corporate sector, some recent tax reforms ( 11 ) could help address incentives for SMEs favouring the accumulation of retained earnings within the company. However, in practice this may shift part of small companies’ savings surplus to the household sector. The overall impact on the savings surplus therefore remains uncertain.

3.    Thematic Issue: Private Indebtedness and Housing

Overall situation

The private debt-to-GDP ratio continued to decline until 2019 but increased somewhat in 2020. High private debt has been a long-standing challenge for the Netherlands. It reached 234% of GDP in 2019, down from 244% the year before. Preliminary data for 2020 suggest that it saw a slight increase (to 235% of GDP in Q4 2020). Both household debt and corporate debt are among the highest in the EU relative to GDP and well above relevant benchmarks (see Graphs 3.1(a) and 3.1(b)). Household debt (consolidated) stood at 100% of GDP in 2019 (103% in Q4 2020), above both the prudential threshold (61% of GDP) and fundamental benchmark (77% of GDP). ( 12 ) Non-financial corporate debt amounted to 134% of GDP in 2019 (131% of GDP in Q4 2020), exceeding the prudential threshold (94% of GDP) and fundamental benchmark (112% of GDP). However, around 60% of this debt is owed by multinationals. As multinationals’ debt largely consists of intra-group debt, the macro-economic risks appear to be limited. Therefore, household debt represent the main reason for concern.

While the COVID-19 crisis has reinforced some existing vulnerabilities, its overall impact on the private debt stock is still uncertain. The COVID-19 shock reversed the decline of private debt observed in previous years. However, the increase of debt-to-GDP ratios in 2020 was driven by the contraction of GDP by 4%, with the nominal debt stock actually declining somewhat. While there was a slight pick-up in non-performing loans (NPLs), they remained at a very low level and overall financial stability has not been affected so far (see below). Although this is partly linked to temporary support measures and debt moratoria and there remains some broader macroeconomic uncertainty following the COVID-19 crisis, at the current juncture the vulnerabilities associated with the high private debt stock do not pose an immediate significant risk to financial stability.

Household debt

Household debt as a share of GDP declined in 2019 but increased significantly in 2020. The net credit flow to households stood at 1.2% of GDP in 2019 and is estimated at 1.3% for 2020 (Table 3.1). While robust real GDP growth and inflation contributed to passive deleveraging in 2019 (Graph 3.1(c)), the GDP decline in 2020 drove an increase of the household debt-to-GDP ratio by around 3 p.p. The stock of household debt consists mainly of mortgage debt and there the COVID-19 pandemic has only had a limited impact as house prices continued to rise (see details below). By contrast, consumption loans experienced a major decline in 2020, but they represent only a small fraction of total household borrowing (Graph 3.1(d)). Therefore, the overall credit flow to households remained positive and even slightly accelerated. At the same time, there has been a significant increase in household savings as the COVID-19 pandemic containment measures constrained consumption (Graphs 3.1(e)). The ratio of household debt to gross financial assets has been on a long-term declining path (29% in 2019), which is likely to continue.

Policy distortions underpin the high household debt stock. High mortgage debt has been driven by long-term policy distortions in the housing market (see below) and tax incentives for mortgage borrowing. In particular, mortgage interest on owner-occupied homes is treated as a fully deductible expense in income taxation. Dutch authorities have taken some steps to limit this, both by disqualifying new interest-only mortgages from tax deductibility (in force since 2013) and by reducing the maximum applicable rate at which mortgage interest is deductible. However, this reduction only affects households in the top tax bracket (some 10% of the labour force) and the eventual rate of 37% (to be reached in 2023) still signifies a strong implicit subsidy on mortgage borrowing.

Financial stability risks from elevated household debt appear contained. While the share of household debt on gross disposable income stands at 200%, almost double than on GDP, NPLs on Dutch mortgages have historically remained very low, even during major macroeconomic and housing market downturns, e.g. following the 2008-2009 financial crisis. The impact of the COVID-19 crisis on mortgage loan performance has so far been very limited as well, in part thanks to income support measures and mortgage moratoria. Once these are wound down, some households may face repayment difficulties. However, the crisis has disproportionally affected lower-income workers (who tend not to be home‑owners) and the impact on employment has been very limited so far. Furthermore, Dutch banks entered the crisis with robust capital levels (the Tier 1 ratio stood at 18.9% in 2019) and house prices are expected to continue growing, contrary to the financial crisis. Therefore, near-term risks to financial stability from elevated household debt appear limited.

Corporate debt

Despite the COVID-19 shock, corporate debt appears to have remained broadly stable relative to GDP in 2020. Following active NFC deleveraging in 2019, which was also supported by inflation (see Graph 3.1(f)), there was a sharp pick-up in short-term borrowing and bond issuance in early spring 2020 due to acute liquidity needs and precautionary build-up of financial reserves. Preliminary data indicate this has more than reverted back in the second half of 2020, ultimately leaving NFC borrowing somewhat below the 2019 level in nominal terms (see Graph 3.1(c)). Overall, this reflects the offsetting effects of increased demand for credit from firms with impaired revenues to cover ongoing costs, and lower borrowing needs as firms cut back production and investment. In addition, government support measures and moratoria granted by banks have met a significant part of firms’ liquidity needs. In terms of credit supply, banks have also tightened their lending standards for corporates (see Graph 3.2(a)).

While the corporate debt-to-GDP ratio is rather high, a significant share of corporate debt is intra-group debt of multinational companies. The headline debt ratios overstate actual exposure, as about 60% of total NFC debt is linked to multinationals and consists largely of intragroup debt. When this is excluded, corporate debt levels are not particularly high by international comparison and relative to relevant benchmarks. ( 13 )

Despite the impact of the COVID-19 crisis, vulnerabilities associated with corporate debt appear unlikely to raise immediate financial stability concerns. So far, the crisis has had limited impact thanks to support measures and moratoria, with the corporate bankruptcy rate in 2020 actually significantly lower than in 2019. Once these are phased out, there may be a rise in bankruptcies and credit losses for banks. While the overall NPL ratio (to total loans and advances) remains low in a cross-country comparison, the NPL ratio for NFCs has already started to increase (from 1.5% in Q1 2020 to 2% in Q3). This could contribute to further tightening of credit standards for NFCs beyond that observed in 2020 (Graph 3.2(b)) and thus create renewed deleveraging pressure. However, in light of the comparatively small bank exposure to riskier corporate lending (e.g., total SME lending represents about 7% of overall banking assets) and their robust capital levels, risks to financial stability from NFC debt seem contained as well.

Housing market

House prices have been growing sharply over the past year and there are signs of overvaluation. Real house prices started to recover in 2013 and were gradually accelerating until 2018 (7.1% YoY growth), and following a slowdown in 2019 (4.8%), house price growth (7.3% in 3Q) and market activity picked up again in 2020 (Graph 3.2(c)). In addition to the impact from the COVID-19 shock (see below), house price growth was likely also boosted by transfer tax changes: a rise in transfer taxes for buy-to-let investors from January 2021 led to a spike in investor demand in the second half of 2020, and the abolishment of transfer taxes for younger households from the same date subsequently increased the bidding power of this cohort. There has been some regional differentiation and, in a break with long-standing prior trends, growth has been stronger outside the major cities in 2020. Valuation metrics suggest that prices in the Netherlands are getting into overvaluation territory (Graph 3.2(d)), and beyond what fundamentals such as income would suggest (overall valuation gap is +10% but fundamental-based gap is +20%).

Following the COVID-19 shock, house price growth has accelerated. On the demand side, mortgage rates further declined and household disposable income increased due to supportive policy measures. As in other countries, households may gradually shift their preferences in favour of more time spent at home, inducing willingness to spend a higher share of their incomes on housing and/or to purchase larger homes away from major urban centres, thus pushing up house price growth in regional locations in particular. Moreover, increased demand can come also from investors facing uncertainty and lack of alternative investment opportunities. There was indeed a strong increase in investor interest following the COVID-19 outbreak, with the share of owner-occupied homes bought by private landlords nearly doubling from pre-COVID-19 levels towards the end of 2020, although this was primarily driven by the transfer tax changes mentioned above. On the supply side, the COVID-19 shock appears to have had limited impact, as housing completions and starts were broadly in line with pre-crisis expectations and residential building permits picked up in 2020 (Graph 3.2(e)).

Several structural factors contribute to upward pressure on house prices. The short-term developments discussed above were coupled with some long-term factors underpinning high housing prices. On the demand-side, there is favourable tax treatment of owner-occupied housing via mortgage interest deductibility in personal income taxation as well as owner-occupied homes not normally being subject to asset taxation under ‘box 3’. ( 14 ) At the same time, there is a lack of suitable alternatives to owner-occupancy due to an underdeveloped rental market. On the supply side, while the share of residential investment on GDP has been increasing for several years (Graph 3.2(e)), housing construction has fallen short of demographic requirements since the housing market downturn following the 2008 crisis. This is linked to unaddressed capacity shortages in the construction sector and planning constraints. ( 15 )

While house price growth is expected to slow down, upward pressure on house prices remains. The negative income impact of the crisis was to a large extent cushioned by support measures (GDI of households increased by about 5% in 2020). However, the crisis disproportionally affected lower-income households, who tend to be tenants rather than home-owners. Moreover, some households will exit the crisis with additional savings resulting from curtailed consumption. The aforementioned structural factors supporting house prices are also set to remain in place. Overall, house prices are therefore likely to continue growing, albeit at a slower pace. While this does not represent a major immediate risk for financial stability, the combination of high household debt and house prices diverging from fundamentals points to long-term vulnerabilities. High house prices in the owner-occupier market combined with and underdeveloped private rental sector also represent a significant burden for younger and middle-income households.

The Netherlands has taken policy steps to discourage buy-to-let investment. As mentioned above, the transfer tax for private landlords has been raised from 2% to 8% of the property value from January 2021 (while for owner-occupiers up to 35 years old it has been abolished ( 16 ). In addition, rent increases (Graph 3.2 (f)) in the private rental sector have been capped, and legislation is under preparation allowing municipalities to ban buy-to-let purchases for a five-year period. These measures may further reinforce the existing policy distortions favouring owner-occupancy over rental housing and risk undermining efforts to boost the availability of private rental housing for middle-income households.

(1) ()    While at least 3% of GDP of the goods surplus can be attributed to Rotterdam’s role as a global trade and logistics hub (and in particular to re-exports), trade in services and primary income transactions are partly related to activities of multinational enterprises (MNEs), i.e. their intra-group payments. For more details, please see Suyker, W., Wagteveld, S. (2019), A Fresh Look at the Dutch Current Account Surplus and its Driving Forces, CPB (Netherlands Bureau for Economic Policy Analysis), The Hague.

(2) ()    ‘Current account norm’ is the current account balance that can be explained by fundamentals. It is based on an empirical setup similar to IMF’s EBA (Phillips, S., Catão, L., Ricci, L., Bems, R., Das, M., Di Giovanni, J., Unsal, D.F., Castillo, M., Lee, J., Rodriguez, J., and Vargas, M. (2013), "The external balance assessment (EBA) methodology", International Monetary Fund Working Paper 13/272). Fundamentals are slow-moving variables including, e.g. natural resources, demographics, or relative income. For details see Coutinho, L., Turrini, A., Zeugner, S. (2018), "Methodologies for the assessment of current account benchmarks", European Economy, Discussion Paper 86/2018..

(3) ()    The structurally high current account surplus has also led to an elevated net international investment position (NIIP). The Dutch NIIP turned positive in 2009 and has increased to 90% of GDP by 2019, driven by the accumulated current account surpluses. The NIIP excluding the non-defaultable instruments (NENDI) is balanced, so that the excess of external assets over liabilities is due to the equity component. However, gross external assets and liabilities exceed GDP tenfold (amounting to 1171% and 1056% of GDP, respectively) due to strong MNEs presence.

(4) ()    The same holds true also for the general government investment specifically. Importantly, although at a macro level there is no underinvestment relative to peer countries, there are some ad hoc and industry-specific investment challenges in some areas, for instance linked to the ‘nitrogen problems’ affecting various sectors, including the construction industry. For more details, see European Commission (2020), Country Report The Netherlands 2020, Commission Staff Working Document SWD(2020) 518 final, Publications Office of the European Union, Luxembourg, subsection 4.4 and box 4.4.1 in particular.

(5) ()    This volatility is largely due to a single MNE in the oil sector which pays a fixed dividend to shareholders, implying that oil price swings translate into net savings and retained earnings figures almost mechanically. Please see European Commission (2020), "Interpreting external sector statistics: taking into account the role of financial globalization and MNEs", Note for the attention of the EPC LIME Working Group, January.

(6) ()    For more details, please see European Commission (2020), Country Report The Netherlands 2020, Commission Staff Working Document SWD(2020) 518 final, Publications Office of the European Union, Luxembourg, and De Nederlandsche Bank (DNB) (2019), Het Spaaroverschot van Nederlandse Bedrijven Ontrafeld, DNB Occasional Study Volume 17-4 9 december 2019, Amsterdam.

(7) ()    There is a roughly offsetting effect for Dutch MNEs, which often have an internationally diversified shareholder base. For more details on this and other statistical discrepancies in the allocation of net lending across sectors, please see European Commission (2020), Country Report The Netherlands 2020, Commission Staff Working Document SWD(2020) 518 final, Publications Office of the European Union, Luxembourg. In addition, the aforementioned fiscal incentives to retain profits in small and medium companies also artificially shifts a part of the saving surplus to the corporate instead to household sector.

(8) ()    While the investment rate of households also exceeded the corresponding rate for the euro area for most of the last decade, the difference was smaller on average, but increasing to more than 3 percentage points in 2019.

(9) ()    As for the financial corporations, their position was mainly affected by the decline in net (received minus paid) distributed income of corporations.

(10) ()    From a policy perspective, high compulsory savings via the pension system (in combination with mortgage repayments) is a key structural driver of the high household savings surplus. It can also lead to suboptimal consumption smoothing across different lifetime phases. Pension premiums are generally set as a fixed contribution percentage (typically around 20%) of gross earnings. For more details, please see European Commission (2020), Country Report The Netherlands 2020, Commission Staff Working Document SWD(2020) 518 final, Publications Office of the European Union, Luxembourg, subsections 4.2.5 and 4.2.6.

(11) ()    These include the phasing out of tax-advantaged internally managed pension plans (‘Pensioen in eigen beheer’) since 2017, and making large debts owed by controlling shareholders to their companies partially subject to tax from 2022.

(12) ()    Country-specific prudential thresholds are the values beyond which countries are deemed more vulnerable to crises given characteristics such as relative income, public debt, and the ratio of regulatory bank capital to risk-weighted assets. Fundamentals-based benchmarks are derived from regressions capturing the main determinants of credit growth (e.g. expected growth, house prices, unemployment) and taking into account a given initial stock of debt.

(13) ()    For more details, please see European Commission (2020), Country Report The Netherlands 2020, Commission Staff Working Document SWD(2020) 518 final, Publications Office of the European Union, Luxembourg, subsection 4.2.3.

(14) ()    For more details, please see European Commission (2020), Country Report The Netherlands 2020, Commission Staff Working Document SWD(2020) 518 final, Publications Office of the European Union, Luxembourg, p. 34-35.

(15) ()    For more details, please see European Commission (2020), Country Report The Netherlands 2020, Commission Staff Working Document SWD(2020) 518 final, Publications Office of the European Union, Luxembourg, subsection 4.2.4.

(16) ()    From April 2021, this will only be the case for purchases below EUR 400,000; for properties above this price level, the standard 2% transfer tax will apply.

(I) ()The estimates derive from a two-step analysis: 1) Compiling 2020 output declines at the sector-level across the globe. For the EU Member States, detailed information for 2020 is available in Eurostat. For non-EU countries, sectoral output changes in 2020 are approximated by IMF WEO GDP changes, thus implicitly abstracting from sectoral heterogeneity; 2) Using output changes in all country-sector pairs to trace back changes in final demand, based on global supply chain interlinkages as captured by the OECD ICIO tables. The resulting set of final demand changes can then be used to simulate the impact of the COVID-19 crisis on each country-sector's value added. Two important assumptions are made to allow for such a translation of output changes into demand changes. First, the technological coefficient matrix, which captures the required amount of supplies from any sector to produce a given sector's output, is assumed to have remained fixed since the latest ICIO data (2016). Second, the country allocation final demand for a specific sector is assumed to have remained proportional to 2016.

(II) ()Distinguishing by the source of the demand allows to quantify the impact of demand changes in a certain country on VA production in all its trading partners, by type of good or service. Given the particular nature of the COVID-19 crisis, with a strong impact on hospitality sectors, the analysis distinguishes between tourism (NACE sector I), all other services (G-N excl. I) and industry (A-F).

(III) ()For example, a drop in Dutch demand for German cars reduces Slovak production of engines, with a knock-on effect on Czech suppliers of engine parts.