Contents

.............

Executive summary and conclusions................................................................... 3

1........... Introduction........................................................................................................ 5

2........... Macroeconomic situation and potential imbalances.................. 5

2.1........ Macroeconomic scene setter........................................................................................... 5

2.2........ Sustainability of external positions.................................................................................... 6

2.3........ Competitiveness and trade performance.......................................................................... 9

2.4........ Private sector indebtedness........................................................................................... 13

2.5........ Housing market development........................................................................................ 17

3........... In-depth analysis of selected topics..................................................... 19

3.1........ Housing market and household sector indebtedness....................................................... 19

3.1.1.     Swedish housing market overview.............................................................. 22

3.1.2.     Challenges linked to the construction sector............................................... 26

3.1.3.     A squeezed residential rental market........................................................... 30

3.1.4.     Household indebtedness............................................................................... 36

3.2........ Indebtedness of the non-financial corporate sector......................................................... 43

3.2.1.     Recent developments in non-financial corporate debt................................ 43

3.2.2.     Factors behind Swedish corporate indebtedness......................................... 48

3.3........ Linkages from private debt to the real economy, the financial sector and public finances.. 49

3.3.1.     Risks for the real economy........................................................................... 50

3.3.2.     Risks for the financial sector....................................................................... 51

3.3.3.     Risks for public finances.............................................................................. 52

4........... Policy challenges............................................................................................ 53

References.......................................................................................................................... 57

              Executive summary and conclusions

In May 2012, the Commission concluded that Sweden was experiencing macroeconomic imbalances, in particular as regards developments related to private sector debt and the housing market. In the Alert Mechanism Report (AMR) published on 28 November 2012, the Commission found it useful, also taking into account the identification of an imbalance in May, to examine further the persistence of imbalances or their unwinding. To this end, this In-Depth Review (IDR) takes a broad view of the Swedish economy in line with the scope of the surveillance under the Macroeconomic Imbalance Procedure (MIP). The main observations and findings from this analysis are:

· The Swedish current account surplus is large, but it does not seem to point to an underlying imbalance in the economy. The current account surpluses have not resulted in an accumulation of net external wealth, but have rather corrected large current account deficits from the past. They reflect primarily prudent fiscal policy and high pension savings by households, and to a minor extent also lower construction investment. Apart from bottlenecks in the housing supply, no policies seem to be in place that would artificially suppress demand. The current account surplus has already corrected markedly since 2007 and is expected to decline further in the medium term. Besides contributing to the high surplus, constraints in the construction sector have been conducive to strong dynamics in house prices, although at the same time preventing an excessive construction boom.

· Despite a continued reduction in the export market shares, the Swedish economy remains one of the most competitive in the EU. Disregarding the crisis years 2008-2009, the losses in market shares have been marginal over the last 10 years. Market share losses in goods have been largely compensated by gains in services. The losses reflect structural changes rather than unfavourable domestic cost developments or eroding quality competitiveness. Unit labour costs and prices have been growing more slowly in Sweden than in its main trading partners. However, the strengthening of the Swedish krona and higher growth in unit labour costs in 2012 indicate competitiveness challenges in the near term. In a broader context, moderate shrinking of market shares can be seen as a positive development in view of the strong current account surplus and euro-area rebalancing.

· Macroeconomic risks related to high indebtedness of the private sector continue to be important, despite a number of mitigating factors and recently introduced policy measures on housing and corporate taxation. Corporate debt has declined substantially since 2009 but remains high even if the large bulk of inter-company loans is disregarded. Household debt has stabilised only recently and is not likely to decline in the near future given the continued credit growth and a low mortgage amortisation pace. On the other hand, various indicators of financial health of corporations and stress tests for households, as well as the low rate of defaults during the 2008-2009 crisis, alleviate to some extent concerns about the private actors' capacity to service their debts. Credit supply and demand conditions do not indicate any imminent deleveraging pressures.

· The Swedish housing market has been stable over the last year, but remains a potential source of instability for the future. Despite some recent measures, inefficiencies still weigh on the housing supply, especially linked to cumbersome planning processes connected to the local planning monopoly of the Swedish municipalities, limited competition within the construction sector and the remaining regulation of the rental market. Together with debt-inducing housing taxation, these inefficiencies tend to create an upward-bias in house prices and thus contribute to increasing household indebtedness.

The IDR also discusses the policy challenges stemming from these developments and possible policy responses. A number of elements can be considered.

· Household indebtedness could be tackled within the framework of a global reform addressing the various incentives biasing demand towards property ownership. A gradual phasing-out of the tax deductibility of interest payments could thus be combined with strengthening the recurrent property taxation. This would restore neutrality among investment alternatives, but also free up fiscal space to reduce taxes that are more harmful to growth. Mortgage amortisation requirements could be envisaged to curtail excessive borrowing and lending. 

· With a view to lowering the level of corporate debt, it could be worth exploring how the tax-induced debt bias could be further reduced beyond measures already taken to tackle international fiscal optimisation by multinational companies.

· In order to improve housing market efficiency, further reforms to the rent-setting system might be envisaged to allow market forces to establish an optimal supply of rental housing at an adequate price. Such reforms could be staged in two steps whereby the first might be to allow rents of newly produced rental units to reflect tenants' willingness to pay. As a second step, rent levels of the remaining stock of rentals could be gradually adapted. It might also be worthwhile to explore how the role of individual tenants in rent negotiations might be strengthened, and to allow for an increased freedom of contract. Streamlining the planning and zoning processes would increase the flexibility of housing supply, foster competition in the construction sector and decrease construction costs. This could be achieved for example by limiting the appeal procedures, clarifying cost responsibilities, reconciling building and environmental regulations and by giving municipalities, which have a de facto local planning monopoly, clearer incentives to decrease the administrative burden on construction companies and provide more efficient services.

These measures are generally mutually reinforcing, as an improvement in the functioning of the housing market is likely to help reduce household debt levels.  A well-considered timing and gradual implementation of the measures would be crucial to achieve the foreseen long-term benefits without destabilising the housing market in the short-term. Although the cyclical situation in the housing and mortgage market does not encourage important reform steps in the near future, it is important to start preparatory works for measures requiring longer administrative or legislative procedures, or wider political and public support.

Interlinkages between external and internal imbalances also should be kept in mind. On the one hand, measures to tackle constraints in the construction sector are likely to spur construction investment, and by making it accessible to foreign competition, work in favour of a lower current account surplus. On the other hand, measures resulting in fast deleveraging would lead to higher savings and, thereby, stronger current account surplus.

1.           Introduction

On 28 November 2012, the European Commission presented its second Alert Mechanism Report (AMR), prepared in accordance with Article 3 of Regulation (EU) No. 1176/2011 on the prevention and correction of macroeconomic imbalances. The AMR serves as an initial screening device helping to identify Member States that warrant further in depth analysis to determine whether imbalances exist or risk emerging. According to Article 5 of Regulation No. 1176/2011, these country-specific “in-depth reviews” (IDR) should examine the nature, origin and severity of macroeconomic developments in the Member State concerned, which constitute, or could lead to, imbalances. On the basis of this analysis, the Commission will establish whether it considers that an imbalance exists and what type of follow-up it will recommend to the Council.

This is the second IDR for Sweden. The previous IDR was published on 30 May 2012, on the basis of which the Commission concluded that Sweden was experiencing macroeconomic imbalances, in particular as regards developments related to private sector debt and the housing market.  Overall, in the AMR the Commission found it useful, also taking into account the identification of an imbalance in May, to examine further the persistence of imbalances or their unwinding. To this end this IDR takes a broad view of the Swedish economy in line with the scope of the surveillance under the Macroeconomic Imbalance Procedure (MIP).

2.           Macroeconomic situation and potential imbalances

1.

2.

2.1. Macroeconomic scene setter

Prior to the 2008/2009 crisis, Sweden enjoyed a long period of strong economic growth, averaging 3.5% between 1998 and 2007. This period was characterised by far reaching economic reforms which set the ground for sound public finances, increasing employment, low-inflationary environment and excellent competitiveness position. As a relatively small and open economy, Sweden was hit forcefully by the 2008/2009 crisis, but recovered quickly once  demand from Sweden's main trading partners resumed. The rapid recovery was helped by a sharp depreciation of the Swedish krona as well as expansionary fiscal and monetary policies. The Riksbank cut the repo rate by 4.5 pps within a few months and the fiscal injection amounted to 1¾ % of GDP in 2009. Despite rapid credit growth and a house price boom before the crisis, the housing market proved rather resilient and provided another buttress to the crisis-hit economy. In 2010 and 2011, the Swedish economy experienced buoyant growth of 6.1% and 3.9% per year, respectively.

Since late 2011, the economy has decelerated again in the wake of the euro-area sovereign-debt crisis and another global slowdown. A recession has been avoided but GDP growth is expected to be meagre in 2012-2013, around 1% per year. The unemployment rate which has never fully recovered after the 2008/2009 crisis is likely to turn around 8% again in 2013. The gloomy unemployment outlook is driven by diminished demand from both consumers and firms and as the latter are pressurised by dropping profitability and lower sales, lay-offs are a likely reaction to counter the negative development. The strong krona is also taking a toll on the prospects for an export-driven recovery. Sluggish economic growth is expected to bring general government finances from slightly positive to slightly negative numbers in 2012 and 2013. Given the different starting position, the fiscal and monetary policy – although accommodative - are not likely to provide the same cushioning as during the 2008/2009 slowdown. Growth is expected to resume in the second half of 2013 as confidence returns and households and companies start realising their pent-up consumption and investment plans. The recovery will gain full speed in 2014 when the economy is expected to grow by 2.5-3%.

Graph 1. GDP growth and unemployment

Source: Commission services (Eurostat)

Note: Forecast for 2012 GDP growth

2.2. Sustainability of external positions

Sweden's current account surplus has been above the indicative scoreboard threshold since 2004. It corrected from 9.1 to 7.2% of GDP between 2008 and 2012.

Developments over the last year confirmed previous trends in the composition of the current account surplus. In particular, the trade surplus in services (4% of GDP in 2012) which became the main contributing element in 2009, continued to be significantly larger than the trade surplus in goods (2.1% of GDP). The contribution of investment income has been stable (2.3% of GDP) but lower than in 2007-2008 due to lower repatriated profits from Swedish investment abroad. Current transfers show a stable negative balance (-1.3% of GDP) reflecting Sweden's position as a net contributor to the EU budget and an aid donor on a global level. The capital account does not have any significant impact on the external balance of the country.

The financial account provides a more stable picture than before the crisis.  Although cross-border financial flows are typically very volatile, their net value has had a stable sign since 2009. Since the crisis, Swedish banks have intensified their lending abroad, causing net outflows in the other investment item. At the same time, they have increasingly financed themselves through bonds issuance to foreign investors, which resulted in net portfolio inflows. The latter has coincided with increasing foreign demand for Swedish debt securities, including government bonds. Net outflows of foreign direct investment have also remained important, reflecting lower FDI coming to Sweden after the crisis but also repayment of debts to foreign owners. Official reserves have been more or less stable over the last year.

Graph 2. Current account decomposition || Graph 3. Financial account decomposition

Source: Commission services (Eurostat) || Source: Commission services (Eurostat)

The net international investment position (NIIP) has deteriorated since 2011 and remains negative (-14.4% of GDP in 2012),[1] despite continued strong current account surpluses. Sweden is the only surplus country in the EU which has seen its NIIP worsening between 2007 and 2012. As shown in the 2012 IDR, in that period Sweden recorded significant negative valuation effects which outweighed the effects of the current account surpluses.  This means that total returns on Swedish assets abroad (taking into account current yields and the change in the asset price) have been lower than returns on assets held by foreigners in Sweden.[2] The negative valuation effects have come mainly through the FDI category and to a lesser extent through portfolio investment. They have been mainly related to (i) the steady appreciation of the Swedish krona since 2009,[3] (ii) a decreasing value of Swedish FDI abroad relative to foreign holdings in Sweden and, (iii) an increasing value of Swedish bonds held by foreigners relative to the value of foreign bonds held by Swedes (see Graph 7). Swedish net external debt also increased over the last year (preliminary figures suggest 63% GDP in 2012) along with increased sales of Swedish corporate bonds to foreign investors. The correction which started in 2009 and was driven by repayment of inter-company loans to foreign affiliates has thus ceased. Given the high level of Swedish net external debt (fourth highest in the EU as % of GDP), its further expansion could be a reason for concern (a large part of external debt is attributable to cross-border inter-company lending which are discussed in section 3.2).

 

Graph 4. NIIP decomposition || Graph 5. Net lending and borrowing by sector

Source: Commission services (Eurostat) || Source: Commission services (Eurostat)

Graph 6.  Valuation effects || Graph 7. Decomposition of valuation effects

Source: Commission services (Eurostat) || Source: Commission services (Eurostat) * For 2012, preliminary figures on the basis of quarterly data

Looking at sectoral saving-investment balances, all sectors show net creditors position. The expanding current account surplus between 1995 and 2007 can be attributed mainly to increased savings by both households and the government which resulted from the tax and pension reforms and the adoption of an ambitious fiscal target in the 1990s. Apart from higher savings, net lending has also been to some extent related to lower investment. The investment ratio was far below the EU level for a decade following the crisis at the beginning of the 1990s, mainly on account of low construction investment, but also due to a structural shift from manufacturing to services. A number of constraints in the construction sector prevented it from responding excessively to strong housing demand after 2000 (see also section 3.1.2). Public investment in construction has also been rather limited, possibly as a result of a strong political commitment to fiscal surplus targets. Beside construction, investment might have been also hindered by several shortcomings in the market for services (e.g. fixed tariffs for provision of certain listed services or limited access of foreign service providers to Sweden until 2011). At present, the overall investment ratio have already caught up with the EU levels, but investment in construction and housing as % of GDP still remains below the EU average .

In view of the above arguments, it would be difficult to establish that the current account of Sweden should point to an underlying imbalance in the economy. The current account surpluses have not resulted in an accumulation of net external wealth, but have rather corrected large current account deficits from the past.  Domestic demand does not seem to be artificially suppressed domestic demand, with the exception of the construction sector and the service sector where domestic regulation might have led to lower levels of investment and imports, respectively. Sweden allows its currency to float freely, wages are set among social partners with no political intervention and, if anything, fiscal policy has provided support to domestic demand in recent years. The surplus has been rather underpinned by a high level of saving in the government and household sector which increased in response to the severe economic crisis in the early 1990s and the subsequent reforms of the tax, pension and fiscal framework.

The current account surplus has already corrected substantially - from 9.1% to 7.1% of GDP between 2008 and 2011 – and is likely to decrease further in the medium term. Several factors are in place which are likely to continue to push the Swedish current account further to balance also in the future: continued deleveraging in other parts of Europe which leads to lower demand for Swedish exports as well as their lower cost-competitiveness, appreciation of the Swedish krona, decreasing household savings as the post-war boomers retire and ageing of the population continues, and growing competition from developing countries. Lower current account surpluses would most likely be compatible with a sustainable external position of Sweden: ECFIN calculations suggest that a balanced current account would be sufficient to bring NIIP close to zero by 2025, disregarding valuation effects.

2.3. Competitiveness and trade performance

Taking a global view on Sweden's competitiveness, it is clear that the Swedish economy remains one of the most competitive of the EU. It is characterised by a high technological advancement, solid R&D and innovation performance, reliable and transparent institutions as well as a higher education and training that deliver satisfactory results.

Nevertheless, over 2007-2011, Sweden's export market shares (EMS) for goods and services have continued to drop (Graph 8), recording a negative growth rate of -11.6% over the five year period, similar to the 2006-2010 time period, highlighted in the 2012 in-depth review for Sweden (-11.9%); in both cases, the MIP scoreboard indicator flashed (-6% being the threshold). Nevertheless, when decomposing the latest five-year-period, the lion's share of this loss occurred in the crisis years of 2008 and 2009 (-4.1% and -6.9% respectively), whereas during remaining years the drop in EMS has been more subdued (in 2011 it stood at -0.5%).

The loss in EMS is linked mostly to the market for goods and the sector-level analysis contained in the 2012 IDR is still relevant. Hence, the decline is mostly due to structural factors linked to the traditional product mix of Swedish exports: paper products are increasingly losing ground as carriers of information, and saw mill products are facing constraints due to the weak economic outlook of the construction sector. Likewise, the numerous suppliers to the car industry are facing increased competition from fast-growing, export-oriented economies that are in the process of being integrated into the world economy.

Sweden needs to be observant in terms of the competitive advantages the country has over low-cost emerging economies, to make sure that its products remain demanded by export markets. The intensity of competition in internal markets can be approximated by an indicator of overlapping or similarity of export structures, the so called Finger-Kreinin index. The index ranges from 0 to 100; where 100 indicates perfect similarity and, on the contrary, 0 indicates that the export structures are completely different. For Sweden, the overlapping index with China is 30.5%, being among the higher in the EU (Graph 9).  As for many other EU member states, the overlap is particularly high for textiles and textile articles, mineral products and miscellaneous manufacturing but Sweden also competes with China in sectors such as basic metals and articles of base metal, machinery and mechanical appliances.

Graph 8. Export market share growth (% y-o-y) || Graph 9. Export similarity index: China vs. EU countries – Partner: World

Source: Commission services (Eurostat) || Source: Comm.services calculations using UN's COMTRADE

More than half of the Swedish exports of goods are destined to other EU countries, and with the crisis of the euro-zone, combined with the strong appreciation of the Swedish krona during the third quarter of 2012 (5% compared to the second quarter of the year), the majority of all EU member states (including Germany, one of Sweden's largest trade partner) recently showed less interest in Swedish goods. Official data on Swedish exports to Germany in 2012 point to a reduction of -4%, which is identical to the overall drop. However, Sweden's share of high-technology products in exports has remained fairly constant over the last decade, and stood at 13.9% in 2011.

In parallel, Sweden's net position is somewhat compensated by a corresponding gain in the export market share for services (Graph 10), driven to some extent by the on-going servicification of the manufacturing industry which increases the interdependency of trade in services and merchandise. The services sector accounted for 1.81% of the world market shares for services in 2010-2011, and among the different types of services, other business services and computer and information services contribute the largest surplus.[4] Under the heading other business services, the largest share is comprised of exports of merchanting, which typically involves the mediation of goods from a foreign supplier to a foreign customer. Merchanting can also cover the purchase by a Swedish resident of goods from a non-resident, and subsequent resale of the goods to another non-resident. During the process, the goods neither enter nor leave the compiling economy. The income received from the service is entered under exports of services irrespective of the direction of the stream of goods. Should these types of services be removed from the exports data, the development would be much more subdued. The inherent volatility of these types of services poses some questions linked to the sustainability of what now accounts for roughly one third of the trade in services.

In total, although Sweden's trade balance can be explained partly by the goods surplus (2.4% of GDP in 2011), the services surplus is however even larger (3.6% in 2011), bringing the total Swedish trade balance to a fairly large surplus of around 6% of GDP. This decomposition is likely to continue strengthening in the future, i.e. the share of services in total exports is expected to increase further going forward, also reflecting the fact that roughly 70% of the Swedish GDP comes from the production and consumption of services, and that this share is continuously rising.

Graph 10. Evolution of exports market shares in goods and services (in % of world share) || Graph 11. Contribution to overall trade balance  for goods by Broad Economic Categories (%)

Source:  UN COMTRADE data || Source: Eurostat COMEXT, Commission services calculations

More recently, the loss in export market shares is linked to changes in the real exchange rate of the Swedish krona vis-à-vis the currencies of its trade partners. During the summer of 2012, the krona appreciated substantially against in particular the euro (by some 8% in Q3 2012, compared to the same period in 2011) and thereby squeezing margins of Swedish export-reliant industry further. Especially saw mills and pulp and paper industries voiced concerns over the development, and demanded further repo rate cuts. In September, the Riksbank cut the rate to 1.25% and in December it was further lowered to 1% in order to support the Swedish economy. Meanwhile, the exchange rate also stabilised to some extent.

When comparing Swedish cost developments against those of trading partners, no major challenges in terms of competitiveness emerge. While unit labour costs have been growing more slowly than in the euro-area over the last few years (Graph 12), there are signs that this situation may be reversed in the years to come (appreciation of the Swedish krona, employees' compensation growing faster than GDP etc), which in that case could be conducive to euro-area rebalancing. Nevertheless, no particular trends are discernible at this stage, and it will be important to observe the outcome of the on-going wage bargaining round to assess possible wage increases going beyond productivity growth indications, and the real compensation per employee. Looking at the recent past, unit labour costs grew in both services and industry over 2007-2010. At the sectoral level, manufacture of (i) basic metals, (ii) motor vehicles, trailers and semi-trailers, (iii) electricity, gas, steam and air conditioning, and (iv) construction have the highest sectoral contribution to total ULC, suggesting that Sweden might be dropping behind in terms of its competitiveness in these areas.

It still holds true that the real effective exchange rate (REER), based both on HICP and unit labour costs, is characterised by some degree of volatility, as described in last year's IDR. In particular, the REER being strongly linked to the nominal exchange rate and co-varying with the developments of the Swedish krona vis-à-vis the currencies of trading partner economies, currency fluctuations have an immediate impact on developments (Graph 13).

A new foreign trade and investment promotion structure under the name of Business Sweden has been created recently through the merge of what used to be the Swedish Trade Council and the Invest Sweden agency. It is owned jointly by the Swedish government (represented by the Ministry for Foreign Affairs) and the private business via the Swedish Foreign Trade Association. It is hoped that there will be synergies from the two types of activities (promoting Swedish exports and attracting FDI to Sweden) and that the joint structure will also improve the efficiency of government funding.

To sum up, despite some challenges Sweden counts among the top performers in terms of its relative competitiveness, and this is something which has also been high-lighted by various competitiveness rankings carried out by international observers.. For example, the World Economic Forum's 2012-13 Global Competitiveness report ranks Sweden as number four out of 144 countries. The World Bank's 2013 "Ease of Doing Business" report lists Sweden as the 13th most favourable country (out of 185 countries, but slipping from the 8th place of the 2012 edition) in which to run a business. The OECD describes the Swedish innovation performance as one of the best in the world (OECD, 2013). Finally, the European Commission's 2012 Innovation Union Scoreboard ranked Sweden as the best EU country in terms of innovation performance, dominating in three out of eight dimensions measured by the scoreboard. However, Sweden faces a challenge in terms of the unsatisfactory degree of commercialisation of innovative products and counts among the countries that need to raise their R&D intensity more quickly in order to reach the tentative national R&D target of 4% of GDP in 2020. R&D intensity has decreased from a peak of 4.13% of GDP in 2001 to 3.42% in 2010 (1.07% public + 2.35% private), according to Eurostat.[5]

Graph 12. Decomposition of ULC developments || Graph 13. Decomposition REER  developments ||

Source: Commission services (AMECO) Note: Forecast for 2012-2014 || Source: Commission services (AMECO) Note: Forecast for 2012-2014

2.4. Private sector indebtedness

Private debt remains at a high level (231% of GDP in 2011), far above the threshold of 160% of GDP. Following three years of correction, Swedish private debt as a percentage of GDP is expected to increase in 2012 since its absolute growth has not decreased as quickly as GDP decelerated (preliminary data based on first-three quarters suggest 235% of GDP for 2012).  The correction in 2009-2011 was on account of decreasing indebtedness of non-financial corporations which generate about two thirds of total private debt. On the other hand, household debt, which account for the remaining third, has not seen any correction and has remained stable at around 81% of GDP since 2009.  To complete the picture, financial corporations also contribute significantly to the country's debt (with 129% of GDP in 2011). On the other hand, public debt stands at a relatively low level of around 40% of GDP.

Graph 14. Decomposition of debt (% of GDP), consolidated data || Graph 15. Credit developments (% of GDP)

Source: Commission services (Eurostat) * For 2012: preliminary figures on the basis of quarterly data || Source: Commission services (Eurostat) * For 2012: preliminary figures on the basis of quarterly data

At 149% of GDP in 2011, corporate debt in Sweden was the fifth highest in the EU, but a number of circumstances mitigate concerns about its sustainability. The previous review showed that the debt-to-GDP ratio is overstated by a large share of cross-border lending within multinational companies (almost 50% of GDP) which tend to minimise their tax payments by exploiting differences between corporate taxation in countries they operate in. Lending within multi-national groups involves lower risks than other types of lending, since liabilities of Swedish affiliates are typically matched by assets of foreign affiliates. The Swedish government has recently made changes to corporate taxation in order to limit the tax minimisation by companies. Various indicators of Swedish corporations' financial health, such as debt-to-assets, debt-to-equity, default rate or interest burden, do not point at any significant sustainability risks. The very low number of company defaults during the 2008/9 crisis also argues for a rather strong financial standing of Swedish corporations. Annual credit growth declined from 6.3% to 1.4% between January 2012 and January 2013 and is expected to remain subdued also in the coming months.

Despite these mitigating factors, the high level of corporate indebtedness calls for further investigation of the driving forces and potential sources of vulnerability. In section 3.2 of this review, the role of corporate taxation in the debt-inclined behaviour of corporations will be discussed, as will. credit conditions, in particular the role of low risk weights attributed by banks to corporate lending. The size of deleveraging pressures will be assessed against various indicators of debt overhang and credit supply and demand pressures to determine the likelihood of further deleveraging in the Swedish corporate sector.

 

Graph 16. Leverage ratios, Non-financial corporations || Graph 17. MFI lending to households (% of GDP)

Source: Commission services (Eurostat) * For 2012: preliminary figures on the basis of quarterly data || Source: Commission services (Eurostat), ECB

Household debt stands at more than 80% of GDP, or 170% of disposable income in Sweden. The lion's share of household debt is made up of mortgages (Graph 17), the build-up of which has gone hand in hand with the rising housing prices over the last 15 years. At present, credit growth for mortgage loans is less expansive but still amounts to approximately 4.5% per year and still outpaces the growth of disposable income. Based on this development, the Riksbank Governor has issued a recommendation of not letting the household indebtedness exceed 200% of disposable income, due to the apparent risks for the financial stability of the country. The vulnerability caused by the substantial household indebtedness is also the reason why this topic, and its relation to the developments of the Swedish housing market, is analysed in-depth in the subsequent chapter of this review. 

Last year's review identified housing taxation and weak amortisation as the main factors contributing to the build-up of household debt. A very low property tax together with generous tax relief on interest payments tilt the incentive structure towards debt-financed investment in property. As for amortisation,   100% loan-to-value mortgages were not uncommon in Sweden before the crises. Increasingly, borrowers opted for amortisation-free loans and banks were not particularly interested in obliging clients to reduce their debt.

These factors contributed to the debt build-up, which has to some extent already been addressed by the national authorities. As a first step, an 85% loan-to-value cap was introduced in October 2010. It has however been possible to circumvent the cap by taking up other types of loans to complement mortgages, a practice which some Swedish banks have on offer (Swedish Financial Supervisory Authority, 2012(b)).To hold back household debt accumulation, the FSA has recommended the introduction of a 15% risk-weight floor for mortgages,[6] compared to the current 6% in order to reflect a changing risk profile. Through the stricter rules, the amount of capital that banks have to set aside to protect against potential mortgage loan losses would increase sharply. Although default rates have historically been very low in Sweden, the FSA argues that caution is necessary. The tougher rules imply that an additional 20 billion Swedish kronor would be locked in as core capital under the so called Pillar 2 (Financial Supervisory Authority, 2012). However, to a large extent Swedish banks have already taken account of the capital levels brought about by the measure, which is currently being consulted with stakeholders. Another important step is the foreseen strengthening of capital adequacy and liquidity requirements for banks, which in Sweden goes beyond the minimum Basel III requirements and is introduced earlier than in other EU countries (10% Core Tier 1 capital as soon as possible, and 15% as from 2015). To a large extent, Swedish banks have already adjusted to the tougher requirements. Possible mortgage amortisation requirements are also being evaluated. The Governor of the Riksbank has spoken out in favour of regulating amortisation practices. In March 2013, the government also tasked the FSA with preparing an action plan for the establishment of a sound culture of amortisation, whereby banks are to become more active in advising clients to amortise at a reasonable pace. The action plan is meant to be finalised by mid-October 2013.

The assets owned by Swedish households do, however, substantially exceed their debt, but these assets are typically bound in either housing or pension savings which means that it may prove difficult to rely on them should Swedish households need to deleverage quickly. A problematic linked issue is that households might reduce spending to rebuild their balance sheets even after a modest decline in real estate prices. This could impact negatively on GDP growth and employment and induce a negative feed-back loop affecting house prices. In addition, the distribution of assets tends to be uneven by generational, regional and income-related axes. This means an increased vulnerability of new borrowers, which tend to be young adults without accumulated wealth, having entered the housing market at an elevated price level.

High private debt entails considerable risks for the financial sector as well as for the government finances. Although Swedish public finances and the banking sector look generally sound, a more substantial deleveraging could likely reduce the perceived robustness of these sectors. For the banking sector, the risks include potential loan losses on mortgages or corporate loans and a subsequent potential loss in investor's confidence leading to a more difficult access to funding in international markets. For public finances, the risks include a drop in tax revenues (property taxes, stamp duties, corporate income tax) in case of a house price correction or a significant contraction of the corporate sector. In addition to this direct effect, deleveraging pressures affect the fiscal position indirectly by negative feedback effects on growth. Even without any deleveraging, the preference for debt-financing by both households and corporations and the tax minimisation practices by multinationals translate into substantial losses in government revenues from direct taxes.  In this review, these feedback effects between individual sectors will be investigated more closely.

2.5. Housing market development

The development of the housing market in Sweden has stabilised over the last couple of years, after experiencing massive price hikes over more than a decade. The MIP scoreboard indicator reports a 1% price increase of deflated house prices in 2011. In 2012, the latest figures also suggest modest price increases, for the country as a whole with a real price increase of roughly 1%.

Graph 18: Deflated house prices in the Nordic region, Index  1995Q1 = 100 || Graph 19. House price cycle  

Source: Sveriges Riksbank || Source: Eurostat, ECB, OECD, BIS

Comparing price developments in Sweden with neighbouring countries, it is clear that Sweden has not experienced the large correction that has taken place over the last few years, for example in Denmark (see Graph 18). Nevertheless, the Swedish housing market has some characteristics which justify a continued in-depth monitoring of this sector. Views diverge as to whether the prices are estimated to be in line with fundamentals or whether signs of a misalignment can be established. This issue was analysed exhaustively in the 2012 IDR, which concluded that although some indicators pointed at an overvaluation (for instance the price-to-rent ratio, affordability ratios etc), others suggested a development in line with the underlying conditions which have been characterised by a strong rise in disposable incomes, a relatively resilient labour market, specific debt-biased tax measures which contributed to price increases etc. Yet, the position of Sweden in comparison with other countries' housing cycles calls for some caution (see Graph 19). Looking ahead, the nearest year is however not expected to include dramatic price movements in either direction.

With global economic uncertainty, the bleak outlook for 2013, rising unemployment and dropping household consumption there are evident downside risks to the asset market. In case of a future house price correction, Swedish households would risk having difficulties to deleverage at a reasonable pace, as assets are for the most part tied up in long-term pension savings and in real estate. The effects could be exacerbated as described in section 2.4 by a negative feed-back loop resulting from a reduction in consumption. Moreover, assets are not evenly distributed and highly indebted households risk ending up insolvent even after a relatively modest price fall. With most mortgages having interest rate terms of at most five years, any future interest rate increases could put additional strains on the already highly indebted households. 

The Swedish housing market is also characterised by some inefficiencies on the supply side. For example, Swedish construction investments are only half of those of other Nordic countries, both in relation to GDP as well as population (Swedish Construction Federation, 2012). In 2009-2011 the number of households also increased by 43,000 more than the number of new housing units (Confederation of Swedish Enterprise, 2012). Recent data on housing starts point towards a continued decline for both residential one-dwelling-buildings and multiple-dwelling-buildings and in total, housing investments dropped by almost 13% (from already low levels) during the third quarter 2012 compared to the same period in 2011 (Statistics Sweden, 2012 (c)), and the construction industry does not expect investments to pick up this year. Moreover, the rental market is afflicted by structural problems linked to several decades of stringent rent regulation, which is only recently showing signs of some softening.

As a result of the above-mentioned difficulties linked to the development of the Swedish asset market, a detailed analysis of some of the key challenges is warranted.

 

3.           In-depth analysis of selected topics

3.

3.1. Housing market and household sector indebtedness

In Sweden, and in particular in the three main urban areas of Stockholm, Gothenburg and Malmö, the price of real estate has increased substantially over the last 15 years, both as concerns single family dwellings and apartments (Graph 20). This development has spurred the increase in household indebtedness, which as mentioned above now stands at ca. 170% of disposable income or ca. 80% of GDP. The worryingly high private debt levels, as well as housing market price developments, were key themes of the European Commission's in-depth review for Sweden of 30 May 2012.

House prices have been characterised by modest nominal increases since the 2012 IDR, with prices having increased by 2 per cent on an annual basis in the last quarter of 2012, compared to the same period last year for one- and two-dwelling buildings (Statistics Sweden, 2013). Nevertheless, household indebtedness remains at historically high levels. Credit growth is still positive, albeit at levels under 5% per year (4.5% in January 2013).

Graph 20. Evolution of House Price Index and MFI Loans for House Purchase

Source: Commission services (Eurostat), ECB

Last year's review did not conclude on a possible misalignment of house prices. Whereas most traditional indicators, such as price-to-rent or affordability indices,[7] suggested a significant overvaluation, the adjusted price-to-rent ratio or econometric estimates found house prices to be close to their fundamental value.[8] Due to stable nominal prices, affordability ratios have improved over the last year (Graph 21), indicating a smaller possible overvaluation than in last year's report, ranging from a suggested 13% overvaluation for the price-to-disposable income ratio, to 22% for the price-to-earnings ratio in 2012Q3.

Graph 21: Housing affordability indices

Source: Ecowin, Sweden Statistics

The price-to-rent ratio[9] has not changed compared to last year and remains some 34% above its historical average at the end of 2012. However, the price-to-rent ratio adjusted for the evolution of the user cost of home ownership (fundamental ratio),[10] suggests a much smaller misalignment (4% if mortgages of five-year maturity are considered, and 9% if an average mortgage rate is used as demonstrated in Graph 22). It should be noted that according to these indicators, house prices have exceeded their fundamental value since last year. This could be found counterintuitive given the decreasing mortgage rates. The reason is that expected inflation declined more strongly than mortgage rates, resulting in higher user costs.

Graph 22: Price-to-rent ratios (actual vs. fundamental)

… for mortgages with a five-year maturity || … for mortgages with an average maturity

  ||

Source: Sweden Statistics, own calculations

The most important factors explaining the price increases over the last fifteen years are fundamental factors such as an increased income level, low interest rates, and increased population density in urban areas etc. (accounting for 2/3 of the price increase), according to a recent analysis of the house price developments carried out by the Swedish National Board of Housing, Building and Planning (February 2013). The Board, using a dynamic panel data model, also finds that the remaining 1/3 of increases are explained by factors related to backward-looking expectations of continued price increases. The latter is found to be a risk factor, as in the event of even moderate price falls, these expectations may create strong pro-cyclical patterns.

In addition to analysing the supply and demand side reasons behind, as well as the sustainability of, the price increases, the 2012 review also pointed towards other factors causing inefficiencies in the Swedish housing sector, namely the depressed construction activity as well as the still regulated rental market. These particularities have also been raised by the OECD in its 2012 Economic Review of Sweden, which recommends a reform of the housing market. 

The sustained household indebtedness linked to the substantial mortgage loan portfolio merits a renewed in-depth evaluation, in particular as the price-to-disposable income ratio is above the historical average and as the interest burden in Sweden, measured as the share of the disposable income, is also the fifth highest of the EU-27, standing at 4.5%, which adds further weight to the affordability analysis (European Commission, 2013). Nevertheless, when looking at the housing market it also seems pertinent to address some of the supply side restrictions identified in last year’s In-Depth Review in more detail in this assessment. Therefore, the focus lies on assessing the bottlenecks present in the Swedish housing market structure, with a special focus on the construction industry in terms of planning, zoning and appeal processes, as well as the specificities of the Swedish rental market.

3.1.1  Swedish housing market overview

Sweden counts roughly 4.5 million dwellings, split between approximately 2.5 million apartments in multi-dwelling-buildings and 2 million single-family-buildings.[11] The apartment segment is made up of roughly 1.6 million rental apartments (63%), 930,000 tenant-owned[12] ones (37%) and a mere 182 owner-occupied condominiums.[13](Statistics Sweden, 2012 (d)). Corresponding figures for Stockholm County, with a population of roughly 2.1 million, are roughly 970,000 dwellings distributed between 700,000 apartments in multi-dwelling-buildings and some 270,000 single-family-buildings. However, in Stockholm a majority of apartments are tenant-owned (51%), which is a share significantly above the national average.

Over time, tenant-owned apartments have been gaining in popularity. Compared to 1990, there are some 33,000 more rental apartments whereas the number of tenant-owned ones have become ten times more numerous and rose by 317,000 (Graph 23). In addition to the fact that numerous tenant-owned apartments were constructed over this period, a significant share of these has previously been rental apartments which were subsequently converted into tenant-owned apartments. In fact, since 2000 more than 150,000 such conversions have taken place, out of which 72% are to be found in Stockholm County. The conversions concerned both privately and publicly owned properties. However, conversions managed by public housing companies have been a hot political issue over the last decade, especially in Stockholm which has seen its stock of rental apartments diminish substantially.[14] Nevertheless, the rather far-reaching sell out of publicly owned rental apartments in Stockholm show some signs of a slowdown, as in 2011 only 1,740 rental apartments were sold (Swedish National Board for Housing, Building and Planning, 2012(a)).

The exponential growth in the number of tenant-owned apartments over the last decade has been coupled with an unprecedented price increase for dwellings of this type of tenure, as well as for one-dwelling-buildings albeit to a smaller extent. Housing price developments, and the possible existence of a housing bubble in Sweden, are topics that were investigated extensively in the 2012 in-depth review for Sweden. In the present report it shall therefore be sufficient to summarise the developments having taken place since then. Graph 23: Changes in stock of tenant-owned and rented dwellings 1990-2011. Sweden. Index: 1990=100

Source: Statistics Sweden

At the national level, housing prices for one-dwelling-buildings have remained reasonably stable over the last year, with a 2% nominal increase between the fourth quarter of 2012 and the same period 2011 according to official data, which would translate into a ca. 1% increase in real terms. The area of Greater Stockholm witnessed a nominal price increase of 3% for this type of dwellings. In the Greater Gothenburg area prices climbed by 4 %, whereas the Greater Malmö area saw them decrease by 1%, which is partly linked to deleveraging strives of Danish residents (Statistics Sweden, 2013). Compared to the previous three-month-period, nominal prices remained unchanged in December 2012-February 2013. As concerns the price developments of tenant-owned apartments, there was a more expansive nominal increase of 8% over the period from December 2011 to December 2012 (Svensk mäklarstatistik, 2013).

The prices of tenant-owned apartment show more dynamics also in a somewhat longer perspective. Analysing price developments over the last five years since the pre-crisis peak in mid-2007 until end-2012,[15] the nominal prices for tenant-owned apartments increased by 11.1% at the national level (by 7% in Stockholm), whereas the inflation measured as CPI over the same period was 7.6% (9% if measured by HICP), suggesting only a mild increase in real prices. For one-family dwellings, nominal prices increased by 9% over that period, suggesting stable prices in real terms. Whereas prices of one-dwelling houses have declined slightly since their all-time peak, tenant-owned apartments are still climbing (Graph 24). This tendency is also being reflected in the latest real estate agents' sales statistics, which point at a 3% nominal price increase for dwellings of this type of tenure in January 2013, compared to a corresponding figure of ±0% for one-dwelling-buildings (Svensk mäklarstatistik, 2013).

Graph 24: Price development of Swedish tenant-owned apartments and one-dwelling-buildings 2000-2012. Index: 2000=100.

Source: Svensk mäklarstatistik, Swedish Board of Housing, Planning and Building

The Swedish housing market has benefitted from certain reform measures during the course of the last couple of years, which might on the one hand have played a role in stabilising prices. For instance, in October 2010, a loan-to-value cap of 85% was introduced for new mortgage loans and there are currently expectations of the introduction of an upward-adjusted risk weight floor of 15%. Moreover, there seems to be an increased pressure on households to amortise on their mortgages. On the other hand, the Swedish government has not signalled any intention to review the generous mortgage interest rate deductibility scheme, nor to strengthen and widen the recurrent property tax. It may also be argued that the gradual cuts of the Riksbank’s policy rate, down to 1% in December 2012, and the linked drops in mortgage interest rates (standing at roughly 3.25% at the beginning of 2013) have spurred price developments.

Expectations about the future have been mixed. At the end of 2012, surveys suggested that households’ outlook on the housing market was increasingly pessimistic. At the time, fewer Swedes believed in rising prices ahead, whereas more expected price drops (Demoskop, December 2012 and SKOP, January 2013). Nevertheless, it should be noted that these survey results tend to be quite volatile, and following the repo rate cut of the Swedish Riksbank in December 2012, the sentiments about future housing market development shifted towards the positive among households (Demoskop, January and February 2013). The larger Swedish banks have put forward rather cautious forecasts for the near future, ranging from expectations about stable prices (with the possible exception of the capital region, where price increases cannot be excluded) to predictions of price drops of up to -5% in 2013.[16] In terms of their purchasing power related to housing investments, recent indicators suggest that households' possibilities of buying real estate and taking on mortgages improved during the fourth quarter of 2012, compared to the previous quarter on the back of real wage increases combined with low mortgage interest rates and in spite of slightly rising prices (Swedbank, 2013 (b)).

Focussing on the economic fundamentals, the Swedish economy is still experiencing a downturn with weak exports and labour market. The linked uncertainty may on the one hand cause risk averse households to hesitate to take on substantial mortgages. On the asset side, most Swedish savings are tied up either in long-term pension funds or in real estate, which means that even moderate price drops in the short run could risk leading to a situation where Swedish households face trouble deleveraging at an appropriate pace and thereby cause more substantial drops in the long run. It is also the case that with a repo rate of 1% following the December 2012 cut by the Riksbank[17], mortgage interest rates are unprecedentedly low and possible future increases would put strain on households.

On the other hand, wage expectations are rather stable. Real interest rates are also expected to stay low for a prolonged period of time making the expected cost of assuming debt lower[18]. In addition, at least in the Stockholm area, prices are also spurred through the insufficient construction activity coupled with the constant population increase.

However, price variations are rather large across the country, both as regards one-dwelling-buildings and tenant-owned apartments. The former sell at an average of ca. SEK 2 million, with the average price at the county level ranging from roughly SEK 3.7 million in Stockholm county to roughly SEK 1 million in northern parts of the country. The latter, on the other hand, sell at a national average of about SEK 25,000 per square meter. In Stockholm city, prices are more than twice as high and apartments sell for roughly SEK 58,000 per square meter whereas for the county as a whole the average sq. m. price level is ca. SEK 36,000. Whereas income levels are in general higher in the capital than in other areas, the differences are not of a corresponding magnitude. The clearly specific Stockholm characteristics are rather linked to longstanding supply restrictions and to demographic factors, which have been driving prices for many years.

Looking more closely at demographic developments, data suggests that there may be a gap between the population growth and the supply of housing, a gap which is exacerbated in urban areas and cities with university colleges[19]. For instance, during 2006-2011 the number of households increased by 80,000 more than the number of newly constructed dwellings in Sweden (Confederation of Swedish Enterprise, 2012), and the Swedish National Board of Housing, Building and Planning estimates that more than 30,000 new dwellings should be constructed yearly to accommodate the population growth while the projected housing starts for 2012 is just slightly above 20,000 projects. This figure should be compared to the situation in the early 1990s, before the housing bubble burst, when approximately 50,000 dwellings were constructed yearly (Swedish National Board of Housing, Building and Planning, 2012(b)). Stockholm county alone, which grows by ca 40,000 inhabitants per year, would need approximately 15-20,000 new dwellings per year, whereas in reality only roughly half of that number is achieved (County Administrative Board in Stockholm, 2012).

To conclude, rigidities on the housing market may be linked to foregone employment opportunities and hampered growth. This is linked to the fact that supply-side restrictions cause a number of difficulties at different levels, as young adults cannot find proper housing, population density increases in spite of income growth, students may take decisions on where to study based on the housing situation, rather than the academic merits, of a given university etc. Swedish urban areas also need to be able to attract innovative companies and talented employees, often in competition with cities located within the same region, or even part of Europe. In particular Stockholm has an important role as growth engine and in this context, the described challenges stand out as particularly problematic. For this reason, it seems warranted to further examine the above-mentioned supply side challenges of the Swedish housing market.

3.1.2 Challenges linked to the construction sector

Following the crisis of the early 1990s, there was a severe contraction of the Swedish construction sector, a contraction which may still be making itself felt through a suboptimal activity level, although the sector was in a clearly expansive phase until the 2008-2009 crisis (Graph 25). The Swedish construction sector (in particular as far as the construction of dwellings is concerned) is also characterised by a high degree of price sensitivity (OECD, 2012), which in the current situation with uncertainty about price developments in the housing market means that firms cut down on investments. Construction activities are also expected to remain subdued until the economic outlook becomes more stable and the pick-up that follows could be weaker than that after the 2008-2009 financial crisis. The OECD recently reported that construction costs in Sweden have also risen rapidly from already high levels, and now count among the highest in Europe, a fact which is not conducive to an expansion of housing starts. In fact, for the country as a whole, these costs have made construction largely unprofitable over the period 1980-2005, as measured by the average Q ratio of house prices to construction costs (OECD, 2007).

Graph 25. Residential investment and building permits

Source: Commission services (Eurostat)

In 2012, construction of new dwellings dropped substantially and is expected to remain depressed in 2013. Total investments in dwellings’ construction is estimated to have amounted to SEK 130 billion, roughly equally distributed between rebuilding activities and the building of new dwellings. This represents a drop (as compared to 2011 levels) of -1% and -17% respectively. Estimations for 2013 predict unchanged levels (Swedish Construction Federation, 2012). The exceptionally low number of building permits that were granted in 2012 is expected to hold back investments this year. The number of granted building permits, fell with an average of more than 40% during the first three quarters of 2012, which should be seen in the perspective of a high cumulative demand for housing investments, especially in urban areas (Swedbank, 2013). In terms of the output, preliminary figures suggest that 20,050 housing starts (4,750 units in one- or two-dwelling-buildings and 15,300 units in multi-dwelling-buildings) took place in 2012, which corresponds to a drop of 23% compared to 2011 when the construction of slightly more than 26,000 dwellings was started (Statistics Sweden, 2013 (b)).

These output levels are clearly below par, when seen from a historical perspective, as they were several times higher, a few decades ago. Whereas the construction of roughly 100,000 dwellings per year during the 1960s and 1970s, which was instrumental in addressing the housing shortages of those decades, was not upheld during the 1980s, a new peak was reached in the early 1990s before all construction activities collapsed as a consequence of the bubble bursting. Since then, Sweden has built up an investment gap in the construction of dwellings of SEK 1,500 billion, compared to the OECD-average. Also when looking at housing investments as a share of GDP, Sweden is placed persistently below other European countries. Over the last decade (2002-2011), housing investment amounted to 3.2% of GDP on average, while the corresponding figure for the EU stood at 5.5% of GDP.

The restricted construction can hence be argued to constitute a supply side explanation to the big house price increases discussed in chapter 3.1.1 and risks affecting growth negatively by putting limits to the mobility of workers, students, immigrants etc. It is therefore important to assess the efficiency of the sector and whether there are institutional factors holding back activity levels.

It often takes several years to launch a new project, as each one needs to go through lengthy zoning and planning processes. Before building permits are granted, a detailed development plan designating residential, commercial and industrial areas first has to be put in place and appeals may be launched at both stages. As a result, it is not uncommon that before a final green light is given by the municipality in which the land is located, several years may have elapsed. The rules governing the zoning and planning processes are also perceived as opaque and there are unwanted variations in their applicability across municipalities (Kalbro, T., Lindgren, E. and Paulsson, J., 2012). Uncertainty lies both with the developer as well as with the civil servants having to carry out an assessment before granting or rejecting building permits. Different pieces of legislation may point in opposite directions (for instance as regards environmental protection and the reconversion of industrial sites into housing) and local politicians may be cautious to support projects that risk disturbing incumbent residents. Municipalities, representing a local planning monopoly, face no direct consequences when projects are delayed or become more costly due to inefficiencies in the processing, and may therefore be inclined to rather preserve the status quo.

These challenges have to some extent been addressed by the Swedish government. However, substantial results in terms of increased construction have so far failed to emerge, in spite of several amendments to the Planning and Building Law (Plan- och bygglagen) both in 2007 and 2010 in order to simplify the planning process. In a continued effort, the government commissioned a public inquiry in November 2011 to assess how the planning process could be rendered more efficient. The inquiry, published in January 2013, proposes limiting appeal procedures, clarifying cost responsibilities and increasing transparency. The increased predictability is expected to lower the risks for construction companies and investors, which in turn is expected to lead to larger housing investments, increased competition and resulting lower prices (Statens Offentliga Utredningar, 2012:91). Nevertheless, the proposals were instantly met by opposition by the Swedish Association of Local Authorities and Regions (SALAR), which claims that by regulating the process more thoroughly, increased transparency may well be achieved but only at the price of an even increased administrative burden.

Other reform strands include a public inquiry looking into how the municipalities apply the rules in place on technical characteristics' requirements of the Planning and Building Law. The inquiry (Statens Offentliga Utredningar, 2012:86) found that the unharmonised application leads to additional costs, which are estimated to hamper construction. Proposals brought forward include limiting the possibilities for municipalities to introduce specific technical requirements going beyond the common set of rules, and streamlining appeal procedures. Moreover, the Swedish National Board of Housing, Building and Planning is to assess how more student housing and dwellings specifically dedicated to young adults could be constructed through specific measures such as the introduction of more generous rules relating to temporary dwellings, lowering requirements in terms of the standard of the dwellings etc. The review is to be concluded by mid-2013. Another review has been commissioned in order to avoid diverging interpretations of existing legislation linked to noise levels in construction and housing (to be finalised by 31 August 2013). The current rules are often interpreted in different ways by various actors on municipal and county level, which slows down the conversion of industrial sites into dwellings and contributes to limiting construction. With a clearer set of limits regarding noise levels, the planning process should be rendered more efficient.

The 2013 Budget also foresees additional resources for processing appeals at the county administrative boards, thereby hopefully speeding up the process. Finally, a legislative proposal relating to the responsibility of municipalities for the provision of housing will be put before the Parliament before the end of 2013, clarifying that in addition to simply evaluating the housing needs within its own boundaries, each municipality also has to take the regional perspective and, together with neighbouring municipalities, ensure that housing needs are being met. This new piece of legislation is of particular relevance for metropolitan areas, such as the Greater Stockholm area, and is meant to stimulate a coordinated approach towards reducing the housing shortage (Swedish National Board of Housing, Building and Planning, 2012 (d)). However, the proposal also contains the possible reintroduction of state aid for construction, a measure which the Swedish Competition Authority is strictly against as it risks distorting the competition and further drives up prices.

The burdensome processes preceding the issuance of a construction permit weigh on efficient competition in the construction sector. The sector has traditionally been dominated by large, national companies which have sufficient financial and human resources to face the challenges.  Concerns about the insufficient competition in the construction sector have been previously voiced also by the Swedish Competition Authority who pointed at deficiencies in public procurement practices and the existence of corruption (Swedish Competition Authority, 2011). This assessment is to some extent confirmed in a recent survey, where it was found that four out of ten Swedish construction companies believe that the public procurement system works badly and that it is characterised by insufficient competition and quality concerns. The risks for corruption are still judged to be high, but few actors take action to minimise them and as many as 90% of the surveyed companies have no, or very little, knowledge of the new Swedish legislation linked to bribery that entered into force in mid-2012 (PWC, 2013). In this context, the government has commissioned a public inquiry to assess how a sounder competition could be achieved and, specifically, how to reduce tax evasion and black labour in the construction sector. The inquiry is to be finalised by January 2014.

Finally, a parliamentary committee related to housing supply in urban growth areas is to be set up by the government. The committee is to evaluate the current system and the role of the municipalities with a view to addressing the housing shortage in the major cities through improved regional integration. The committee will be established during the spring of 2013 and shall finalise its task at the beginning of the next mandate period, starting in autumn of 2014.[20]

In light of the above, it seems somewhat premature to evaluate the usefulness of these inquiries and proposals, especially as the government has announced that in the months ahead, they will be scrutinised with the objective of reforming the planning and building process, and that there is hence no clear legislative proposal at this stage. However, should the reforms indeed lead to an increased transparency and predictability, the market might open up for new players, in particular smaller and foreign-based companies which have so far had a very small market share due to the high entry barriers of the sector. The proposed changes should also involve public procurement[21], where divisions into smaller lots, better specified call for tenders and a shorter time period from call to submission and outcome would favour a diversification of the market actors.

3.1.3 A squeezed residential rental market

There are approximately 1.6 million rental apartments in Sweden, accounting for 63% of the total stock of apartments. In terms of the owner structure, the distribution is fairly even between private landlords and public housing companies. A very small share is also owned by the Swedish state, counties and directly by municipalities (Graph 26).

Graph 26: Rental apartments in Sweden by type of owner, 2011

Source: Statistics Sweden

Rent levels have traditionally been negotiated exclusively between the Swedish Union of Tenants (Hyresgästföreningen) and the publicly owned housing companies. Private landlords have been obliged not to surpass the price ceiling established during these negotiations with more than maximum 5%. The rent regulation system has its roots in the post-war period, and it has limited rents' responses to market signals over several decades. According to the OECD, Sweden is characterised by an exceptionally high degree of rent control in the private rental market, for example as concerns control of rent levels, how rent increases are negotiated etc (OECD, 2011). In Sweden, the latter takes place through a rent valuation model based on a set of characteristics defined as the so called utility value of any given dwelling. These characteristics include factors such as the level of standard, services offered and the condition of the dwelling, with the age of the building often becoming a proxy for these factors (Graph 27). However, factors pertaining to the actual market value of rental leases (for example the relative proximity to the city center) have not, until recently, been included.

Graph 27: Development of rents in Sweden 2000-2011

Source: Statistics Sweden[22]

This has led to landlords not being able to charge tenants market-clearing rents in central city locations as compared to suburban areas, although the size of the investment differs substantially, thereby discouraging investments in rental housing. The insufficient supply on the rental market creates an emphasised pressure on dwellings of other types of tenure, most notably on tenant-owned apartments (which are the closest substitute in urban areas) the prices of which escalate as has been showcased in Sweden over the last fifteen years. It also creates an undesired indebtedness and risk exposure for individuals who are in a sense forced to own although they would actually rather rent if the opportunity were available. Graph 27 demonstrates, however, that also rent developments have outpaced inflation over the last decade.

Until 2011, there has also been a situation of unfair competition on the rental market, as municipally owned housing companies did not operate on market-based conditions, having had access to implicit credit guarantees from their respective municipalities. In 2005, the Swedish Property Federation therefore filed a complaint with the European Commission objecting precisely to this form of state aid, and to the bench-marking role of the two partite negotiations. The complaint was found justified and in 2011 a new system was put in place, under which the approximately 300 municipally owned housing companies must no longer benefit from preferential conditions, and their exclusive role in rent-setting disappeared through the introduction of separate negotiations between the Swedish Union of Tenants on the one hand, and municipally owned and private landlords on the other hand As a result of the new law that took effect on 1 January 2011, municipally-owned housing companies (MHCs) are henceforth meant to act in a business-like way, striving to achieve business aims, and should not have a social function but rather be open to all sorts of tenants. Given that MHCs are to compete on equal terms with private companies, behaving as any other long-term private investor, the role of these companies may be put into question.

Rents are still being set according to the principles of the utility value system, irrespective of the change in legislation, hence not allowing any real rent level dispersion between, for instance, metropolitan city centers and suburbs. Tenants that have been able to sign contracts for centrally located apartments, with rent levels far below the market clearing rate, therefore tend to be unwilling to let go of their contracts even if they no longer need them, which substantially reduces mobility on the rental market.[23] As a consequence, most municipalities located in urban areas face a shortage of available rentals and administer queuing systems whereby future tenants may wait to be allocated an apartment, often after having queued for several years (please see Box 1 for a description of the Stockholm Housing Service).

A public inquiry looking into the unsatisfactory situation on the rental market was presented to the government in December 2012. The inquiry criticises the current system and proposes “a gradual transition to entirely new market conditions” as being “necessary in order to create a functioning rental market in the long term”. Somewhat pragmatically, it however also lists a number of measures that could be immediately put in place, under the system launched in 2011. Among the several useful avenues proposed to rebalance the situation, i.e. to give investors incentives to meet the demand for rental housing, it is suggested to allow for an increased spread whenever rent tribunals examine rental levels. Moreover, the geographical location of rental apartments should be given an increased weight in the utility value basket of characteristics, which forms the basis for rent negotiations. In parallel with the supply side restrictions existing in metropolitan areas, other parts of the country are characterised by an excess supply in the rental market with rents higher than justified by market conditions. This is also a consequence of the utility value system, whereby rental apartments located in relatively unattractive municipalities with limited employment opportunities, may nevertheless be coupled with high rents given that they are of a high standard, well-serviced etc. More market-oriented rent levels would benefit tenants in such areas (Statens offentliga utredningar, 2012:88).

Considering the more long-term objective of moving towards market-clearing rents, a similar transition, which may be evaluated as a reference, was carried out in Finland in the 1990s. The Finnish deregulation concerned only the private rental market while the publicly owned rentals still apply a cost-coverage method for rent setting and the dwellings are allotted according to a social needs-based approach. The transition was carried out in two steps with the first allowing for rents reflecting tenants' willingness to pay only for newly produced rental units, and the second involving a gradual transition also for the existing stock of rentals. The effect implied a 57% rent increase on average over the 1990-2004 time period, however variations where noticeable across regions with Helsinki seeing rents increase by 71%. Nevertheless, it is also estimated that the reform was coupled with new rentals entering the market, as potential landlords realised the potential benefit of letting their property (Kulander et al, 2008).[24]

Looking at rental markets of other Nordic countries, it may be observed that similarly to Sweden, in Denmark the rental market structure is also that of a broad tenants' base with a unified rent setting system for the majority of rental dwellings. However, the public housing is not owned by the Danish municipalities but rather by non-profit organisations. In Norway, there are no publicly owned housing companies and in general policies are biased towards housing ownership. The supply of rentals is generally limited to private individuals letting (parts of) their own dwellings (Stockholm County Administrative Board, 2013).

In addition to the traditional rental market, with contracts being allotted through municipal services, a smaller secondary market for privately let dwellings exists. Different sets of rules apply depending on whether the dwelling is a sub-let rental unit or a let tenant-owned apartment, but until early 2013 joint characteristics have been that the permission of the land lord or the housing cooperative association respectively has been necessary, and that it has not been possible for lettors to go beyond the rental level of comparable rental objects located in the same area.[25] This rule is believed to have limited the supply of privately let tenant-owned apartments especially in urban areas, as with such a set-up not even the housing expenses of the lettor, which in addition to the fees to the housing cooperative association typically includes substantial mortgage interest rates, are covered. In an attempt to widen the base of available rentals, the government proposed in its Budget Bill for 2013 to relax these rules and allow tenant-owners to demand rents that would make private letting profitable (limited to one unit per person). Moreover, it was proposed to abolish the permission requirement. The proposal was met by protests by the opposition, but was partially endorsed by the Parliament in December 2012. The compromise means that private lettors and tenants may agree on rent levels based on the actual costs borne by the lettor, and as a general guidance for rent-setting in the future a 4% return on the market value plus the monthly fees will be applicable. However, lettors are still obliged to ask the housing cooperative association, which will grant permissions in justified cases such as work or studies elsewhere, trial cohabitation etc. These changes entered into force on 1 February 2013 and there are some expectations of rising rent levels for privately let apartments in urban areas, in particular in Stockholm city. Such adaptations would replicate the potential market-clearing rents of rental apartments, and in this light a continued evaluation of the price developments and market effects would be useful.

Box 1: An outlier at 59°19′46″N 18°4′7″E : Stockholm's rental market - benefits insiders at the expense of outsiders A sub-optimally functioning housing market can have detrimental effects on growth and employment. Mobility is hampered as workers and students may hesitate to take up offers of employment and/or studies knowing that they may be faced with a grim housing situation. This is of course a more pronounced problem in urban areas, where the costs of buying an apartment are also far above the national average. Debt levels of new mortgage borrowers in the municipality of Stockholm stand at more than six times their disposable income (Sveriges Riksbank, 2012). A price per square meter of roughly SEK 58,000 limits investment possibilities. In the present economic situation, with apparent downside risks on the housing market, and a likely stable price development in the short run, investment decisions tend to be long-term. Persons brought to Stockholm to work or study for shorter periods, generally prefer other options. A well-functioning rental market therefore seems crucial for the Stockholm region to continue to attract young adults, immigrants and people in search of better job prospects or education. Changing structure of the Stockholm housing market The development of Stockholm county’s housing market over the last couple of decades has mirrored that of the national changes in tenant-owned and rented dwellings, but in an even more exaggerated way. At present, the distribution between rentals and tenant-owned apartments is even (Graph 28) at somewhat less than 40% each of the total stock, which is to be compared with a predominance of roughly 60% of rentals in 1990. The difference corresponds to more than 100,000 rental apartments having been converted into tenant-owned ones. Private property owners have found the regulated structure dissatisfying and have therefore chosen to exit the market by converting rental units into property that can be sold at market value, a strategy which has proved to be profitable also for the former tenants. As for publicly owned housing companies, the income from the sales (amounting to SEK 35 billion, or roughly EUR 4 billion, for the Stockholm-based MHCs) has been used to renovate and upgrade the remaining stock, as well as to a limited extent to launch production of new rental dwellings; investments that cannot fully be financed via the regulated rents.  Graph 28: Distribution of various types of tenure in Stockholm county, 1990 and 2010

Source: Country Administrative Board of Stockholm

Social considerations

All municipalities in Stockholm county report that there are too few rental dwellings available, which makes it especially difficult for young adults and immigrants to find appropriate housing (County Administrative Board of Stockholm, 2012). Immigrants, and persons with foreign-born parents, are also more likely than persons born in Sweden of Swedish origin to have low incomes, creating an additional barrier to finding suitable accommodation in the Stockholm area. In Sweden, social housing, in the meaning of a separate stock of housing being reserved for low-income households at low rents, has been rejected by the legislator. Instead, targeted groups (notably pensioners and single-parent households with children) may benefit from housing allowances with certain eligibility criteria such as income, number of children etc. However, as a consequence of political decisions the rules have become tougher and the number of households receiving such support has dwindled over the last couple of decades. Another strand of possible support for weak households goes via the municipality, and the share of households living on such economic support is higher among households with a foreign background. For the Stockholm municipality taken as a whole, the figures are 9.2% for foreign-born persons and 2.2% for persons born in Sweden, but in certain immigrant-dense suburbs the corresponding share climbs up to 20% (Lind, forthcoming 2013). The difficult situation linked to the regulation of the Stockholm rental housing market, which has been further aggravated by population growth and suppressed construction, puts additional strain on integrating low-income, immigrant households. The conversion from rentals to tenant-owned apartments has also, to some extent, created a shift in the composition of the stock of remaining rentals, as the conversions predominantly took place in high- and/or middle-income areas where tenants could afford to take on mortgages, whereas low-income households remained as tenants. Hence, although Sweden does not de jure have a social housing sector, de facto the country, and its capital in particular, may be moving in this direction.

Households not having the option to buy tenant-owned apartments, or to come by rental contracts via contacts etc., can queue to be allotted a rental apartment. The Stockholm Housing Service manages a queue of almost 400,000 prospective future tenants, and the queue is growing by the year (for reference, the queue grew with 9% in 2012 and as recently as a decade ago it stood at 82,000 members). In order to be allocated an apartment in the central areas of the city, future tenants have to wait for 14-17 years. In the suburbs, the waiting time is not quite as long but as a general rule suburbs within closer commuting distance require roughly a decade’s queuing. Nevertheless, the picture is not quite as bleak for newly produced dwellings, where due to the higher rent levels the average queuing time required was roughly five years in 2012. On the other hand, due to the very limited construction activities in 2012 less than two thousand such contracts were allocated (Stockholm Housing Service).

Pressure is mounting on rent levels

In theory, landlords of newly constructed rentals may charge market rents as new apartments are exempt from the normal rent negotiations under the utility-value-system. However, they need to obtain prior agreement on the rent level from a local tenants’ association, otherwise the rent may be questioned by the Rent Tribunal upon request of tenants. Moreover, even with such an agreement, the exemption only covers the first fifteen years (only the first ten years until end-2012). Thereafter rents may be disputed by referring to comparable rentals. Hence, charging full market rents is a risky strategy for landlords, who often opt for a middle way. In fact, some estimations suggest that market rents in the city center would be up to 50% higher than the actual rents in new developments. At the same time, the latter were 45% higher than average conventional rents in 2010 (Lindblad, 2010).

There have also been some attempts by landlords to increase rents on the stock of rental dwellings through the so called Stockholm model (Stockholmsmodellen). The model was negotiated between the Swedish Property Federation, the Swedish Association of Public Housing Companies and the Swedish Tenants' Union in 2008-2011 with the aim of widening the application of the utility value system to better take tenants' real preferences into account when setting rents.  For Stockholm, this entails a bigger focus on the location of the building. However, the practical implementation of the model has caused friction between the parties as landlords demand rents closer to market-clearing levels. The Rent Tribunal is currently assessing demands of roughly 10-15% rent increases for a number of pilot properties. Should the verdict side with the landlords, widespread rent increases are likely to follow in the whole Greater Stockholm area.

A parallel development towards market-clearing rents can be observed on the market for privately let apartments. As from February 2013, tenant-owned apartments and one-dwelling-buildings may be privately let at a rent linked to the market value of the property (and not having to be equivalent to the rent of a comparable rental dwelling), a fact that should also put an upward pressure on rents especially in the urban areas.

Black market activities:

Over the last decades the current set up of the rental market has however provoked a substantial black market with different characteristics, the most common ones being i) sub-letting without the permission of the landlord and at rent levels higher than those of equivalent rentals, and ii) trade with rental contracts.

There are indications that subletting without permission takes place widely in Stockholm. In fact, Stockholm county's largest housing companies estimate that between 10 and 20% of their total stockis being sub-let without permission. In 2012, these companies agreed to 782 sub-lettings but at the same time they estimate that approximately 9,000 rental apartments among their stock were being sub-let without the necessary permissions. Engaging is such behaviour is not particularly risky for the individual tenant: while in theory one risks losing one's rental contract, in 2012 less than 500 persons actually did so. Nine out of ten sub-let apartments have rent levels that widely surpass those of comparable rentals (County Administrative Board of Stockholm. 2013).

Trade with rental contracts must also be assumed to be a widespread phenomenon, although it is by definition difficult to pinpoint the extent of these illegal activities. The Swedish Property Federation estimates that every year, five to seven per cent of all rental units get new tenants in Stockholm, but only roughly 0.5% are allotted by the landlord or the Stockholm Housing Service while swapped tenancies account for the remaining share. Approximately 50% of all such swapped tenancies (or roughly 2,200 cases per year) are accompanied by illegal transactions whereby the tenant of the less attractive rental pays money to the tenant of the more attractive one in order to sign a rental contract. This branch of the black market is estimated to account for approximately SEK 660 million (or almost EUR 75 million) per year, which represents on average SEK 300,000 per apartment, only in Stockholm city. Including the Greater Stockholm area, black market transactions are estimated to amount to roughly SEK 1.2 bln per year (The Swedish Property Federation, 2006).       

With a sequenced transition to market-clearing rents, such market anomalies would gradually be phased out, which should lead to better access to the rental market for, in particular, vulnerable groups such as young adults and immigrants who are most at risk of having to pay for tenancies while at the same time being characterised by low and insecure incomes. 

Future outlook:

It is clear that the Stockholm rental market is challenged by severe distortions resulting from the several decades of rent regulation. What was once introduced as a crisis measure during the 1940s has remained in place and hampered efficient solutions to the detriment of the inhabitants of the capital. The Greater Stockholm area is important as a catalyst for increased prosperity and employment, and a key challenge for the future is making sure that insufficient housing supply does not become a drag on growth. The reform steps outlined have the potential of addressing some of the problem areas, but can, most likely, not substitute for a deregulation of the rental market in the longer run.

3.1.4 Household indebtedness

As noted in chapter 2, household indebtedness increased strongly over the last decade and reached slightly over 170% of disposable income in 2010 (Graph 29). The lion’s share of household borrowing is made up of mortgage debt, the growth of which has been intimately linked with the increases in housing prices discussed above.[26] Nevertheless, debt levels seem to have stabilised and in its forecasts until and including 2015, the Riksbank expects unchanged, or slightly decreased, debt levels.

Graph 29. Households' liabilities || Graph 30. Households' mortgage credit growth, annual % change

Source: Eurostat || Source: Sveriges Riksbank

Mortgage credit growth continued to slow down during the course of 2012, and the pace of mortgage expansion has fallen to a year-on-year rate of around 4.5% in January 2013 (Graph 30). Factors behind this continued slowdown in new mortgage issuance include moderate house price growth, heightened uncertainty about the economic outlook reflected in economic sentiment indicators and forecasts of rising unemployment and, possibly, more restrictive lending practices by banks, partly reflecting the introduction in October 2010 of a 85%-cap on the loan-to-value for new loans and, to a lesser extent, possibly also the introduction of amortisation requirements by SEB and the linked expectation that other banks may follow. Contrary to the credit growth contraction in 2010 and 2011, lately it has been coupled with falling mortgage interest rates. The Riksbank cut the repo rate on three occasions in 2012, bringing it down to 1%[27], which was also to some extent reflected in the variable mortgage rates, despite the fact that the margins on variable-rate mortgages have increased as an effect of banks not having reduced their lending rates at the same pace as their funding costs have fallen due to the decrease in risk premiums for the Swedish banks during 2012 (Sveriges Riksbank, 2012).

In a situation of high and increasing household debt, the developments in the mortgage and housing market have become a factor in formulating monetary policy.  Over the last few years, the Executive Board of the Riksbank has been divided as regards whether housing market considerations should be taken into account when taking decisions on monetary policy. Increasingly over the course of 2012, the bleak economic outlook spoke in favour of further cuts of the repo rate. Nevertheless, the fear that such cuts would translate into lower mortgage rates, and, as a consequence, increased borrowing by households, held back some Members from arguing for too substantial cuts.[28]

In the current context, with low mortgage interest rates, households' interest ratios (i.e. interest expenditure-to-disposable income) are also low, and standing at 4.4% in December 2012. Nevertheless, going forward interest rates are expected to pick up. Higher rates would of course have a negative consequence for the interest ratio, which is expected to climb to 6% towards the end of 2015. Measures of affordability suggest that the mortgage servicing costs are manageable for Swedish households. Stress tests (Swedish Financial Supervisory Authority, 2012 (b) and 2013) also indicate that a vast majority of households would be able to honour their obligations also under various adverse scenarios. Nevertheless, the stress tests do not evaluate what would happen to households' consumption when income and expenditure levels shift, how banks' costs and access to funding would be affected or whether households could quickly lower other types of expenditure and/or use their savings to continue to service their debt in adverse conditions. The net worth of households indicates that on average assets (corresponding to roughly 630% of disposable incomes in mid-2012 and including the market value of real estate) are more than three times larger than the liabilities, but the corresponding figure for liquid assets is approximately ca. 150% which is lower than the accumulated debt (Graph 31).

Almost half of all Swedish real estate assets are located in the three major urban areas, where big house price increases have also been observed over the last fifteen years. In the event of future price falls, this situation may become problematic as households in these areas are therefore particularly sensitive to price changes and adapt by changing their consumption patterns. In this context, Stockholm is most at risk, with 35% of all mortgages but only 25% of the employment income (Swedish National Board of Housing, Planning and Building, 2012).

The distribution of debt between different income groups is another aspect of the assets/liabilities matching issue. In this context, it is noteworthy that 57% of the total stock of mortgages is held by the households with the top 20% income level (having an average indebtedness level of 277% of disposable income). Similarly, the 20% of households with the lowest incomes (having an average indebtedness level of 38% of disposable income) accounted for only 2% of total mortgage debt. This result suggests that households' ability to service their debts should be rather satisfactory (SEB, 2012).

Graph 31: Households' wealth and debts as % of disposable income

Source: Sveriges Riksbank

Moreover, comparing debts to the value of assets is not a reliable indicator of the leverage. The value of assets is volatile – it is prone to speculation and tends to increase in boom periods and contract in bust periods. Therefore, it is appropriate to discount for these evaluation effects, e.g. by comparing debt to deflated assets (see Graph 32). So while the standard debt-to-assets ratio has not increased compared to its 1995 level (30%), the adjusted ratio of "notional leverage" shows an increasing trend.[29] This indicates that indebtedness of Swedish households has to a large extent been driven by the increasing value of assets. In Denmark, the rapid increase in notional leverage between 2003 and 2007 signalled an unsustainable debt built-up which subsequently led to a pronounced house price correction.

 

Graph 32: Debt-to-asset ratio – standard vs. notional leverage (in %), selected Nordic countries

||

Source: Commission services (Eurostat, own calculations)

Another less reassuring characteristic of the composition of households' debt is the fact that there are large variations in the degree of indebtedness within the segment. The already high average indebtedness of slightly above 170% of disposable income includes households without any debt which means that for households with debt, the debt levels are often much higher. This is particularly the case in urban areas (Graph 33), where housing expenses are substantial, but even when the three metropolitan areas are excluded, the indebtedness level of new borrowers is on average four times as high as the disposable income (Sveriges Riksbank, 2012). 

Graph 33: Debt levels of new mortgage borrowers in major urban areas vs. remaining parts of the country

Source: Sveriges Riksbank

The increase in indebtedness has been coupled with a trend towards a higher share of mortgages taken up at variable interest rates, which in last 15 years have been on average lower than fixed rates and therefore perceived as attractive for the consumers who perhaps do not fully take the risk into account that interest rates may increase. In Sweden, it remains costly to change from one type of contract to the other, as many banks fully compensate for foregone interest rates whenever clients wish to switch from fixed to variable mortgage rates or change the total duration of the mortgage, hence discouraging them from altering the conditions of mortgage contracts.

In the context of especially the urban areas, but also for new borrowers in general, it is also interesting to look at the amortisation behaviour of households. This group of borrowers tend to prefer amortisation-free mortgages, or in any case mortgages with an amortisation period of more than 50 years (57% and 21% respectively), with the underlying risky assumption that real estate investments will almost by default increase in value over time. Less than 15% foresee to pay back their mortgages within 30 years. While a slow pace of amortisation may not seem as problematic for young people, it is certainly a matter of concern for households approaching or beyond retirement age. Loan-to-value ratios for new loans for these age groups amounted to 50-60% in 2011 (Financial Supervisory Authority, 2012). It is with this development in mind that the Swedish National Board for Housing, Building and Planning issued a recommendation in the autumn of 2012 that households should increase their amortisation pace. It is based on factors such as that high indebtedness is not a phenomenon concerning only young households but that the average indebtedness level of pensioners is also substantial and stands at 53% of disposable income, and that households have become used to seeing real estate as a virtually risk free investment which will always increase in value. Nevertheless, Swedish households have experienced price drops rather recently (2007/2008) and although the housing market seems to have stabilised, mortgage loans are still associated with a certain level of financial risk. Based on the above, the Board recommends an amortisation pace which at the very least would imply that indebtedness levels do not increase further in the future. In practice, this corresponds to amortisations of roughly 2-4% per year in real terms and while this essentially corresponds to the actual figure it is counterbalanced by the fact that consumers constantly take up new mortgages, meaning that for the economy as a whole there was a negative amortisation of roughly -3.6% in 2012 (Swedish National Board of Housing, Building and Planning, 2012 (e)).

The development has also been scrutinised by the Riksbank, which together with the Swedish Financial Supervisory Authority look into the issue of whether measures to stimulate amortisation are needed. The Swedish Bankers' Association's recommendation that loans with a loan-to-value of more than 75% should be amortised down to at least that level[30] has not been fully implemented by banks, at least not in the past, as they may hesitate to rein in the highly profitable mortgage loan segment.[31] Out of the four major Swedish banks,[32] only Swedbank is in favour of introducing legal requirements relating to amortisation while the others refer to the existing recommendation. SEB has already introduced an amortisation requirement on its own account, meaning that new borrowers are obliged to amortise mortgages with loan-to-value rates of more than 70%, i.e. a somewhat more stringent requirement as compared to the recommendation.[33] In a recent survey, the FSA concludes that the 85% LTV mortgage cap continues to have a positive effect in stabilising mortgage developments, as the households' loan-to-value ratio for new loans is currently around 70 per cent. The survey also shows that nine out of ten households with a loan-to-value ratio above 75 per cent amortise. The number of households taking loans that have a loan-to-value ratio above 85 per cent is limited (roughly 11%). However, only four out of ten households with a loan-to-value ratio of less than 75 per cent amortise, and the average repayment period for mortgages being amortised is more than 140 years. The government has tasked the FSA with preparing an action plan on how to promote a better amortisation culture among banks and borrowers, where, while not introducing any new legal requirements, banks will still be obliged to issue specific guidance to clients on a suitable amortisation pace etc. by mid-October 2013.

Certain policies are likely to have contributed to the debt accumulation and the slowing amortisation pace, as pointed out in last year's IDR. A reform which is very likely to have affected the housing market and tilted the incentives of households towards debt-financing is the drastic reduction of the property tax that took place in 2008[34], leaving the deductibility scheme intact. The generous interest rate deductibility stands out as particularly conducive to debt accumulation (European Commission, 2012). In Sweden, mortgage rates are 100% deductible against capital income, which is quite unique as most other countries, for example Denmark and Belgium, have set a ceiling for the maximum allowed deductibility. In case of a deficit, interest payments are deductible against other types of income (for example labour) at 30% up to SEK 100,000 (roughly EUR 11,000). Thereafter, it decreases to 21%, but there is no upper limit meaning that all borrowers get a tax subsidy irrespective of the size of their mortgages (Swedish National Board of Housing, Building and Planning, 2012 (e)). In its December 2012 economic survey of Sweden, the OECD also brings up these issues as areas for potential reform, as does the Swedish Centre for Business and Policy Studies in its 2013 Economic outlook report. However, the Swedish government has so far not taken any taxation-related steps to propose measures to reduce the debt bias.[35] In fact, the Minister of Finance rather confirmed that no proposals targeting neither the interest rate deductibility scheme, nor the property tax, are foreseen, claiming that it would risk provoking heavy price falls on the housing market.[36] Whereas the timing and sequencing of a possible phase out should of course be subject to careful analysis, the measures as such should assist in balancing the situation. It is clear that a global reform, touching all the above-mentioned domains, would be preferential as compared to a piece-meal approach targeting only specific aspects thereof.

The Swedish system of giving tax payers the right to defer capital gains tax resulting from a sale of private real estate (given that they buy a replacement residence and that certain other conditions are fulfilled), is another issue linked to housing taxation which may be reviewed. This system was first introduced in 1968 to counter possible lock-in effects resulting from tax payers not being able to change residence to an equally priced dwelling due to the obligation to pay the capital gains tax up front, and in order to stimulate mobility in the labour and housing market. Since the introduction, it was reformed in several steps and even abolished in 1990-1994. At present, capital gains amounting to SEK 256.6 billion have been deferred, which translates into SEK 56.5 billion in deferred taxes distributed over roughly 600,000 approved deferrals. Taking into account the magnitude of the measure, as well as the complexity of the rules linked to the system, the Swedish Tax Agency recently recommended scrapping the system (Swedish Tax Agency, 2012). According to their proposal, the new rules would enter into force as from 1 January 2015 and leave current deferrals unchanged. At the same time, the tax rate is proposed to decrease from 22% of the gain to roughly 19% to compensate the tax payers. The proposal is currently assessed by the Ministry of Finance.

The consequences of the current system, and its possible abolition, need to be well understood. The system likely has adverse price effects on housing, as the deferrals enable tax payers to finance more expensive real estate. Also, it creates other types of lock-in effects, such as the reluctance of home owners to sell property and move to a rental unit as that would lead to an immediate and full taxation of the capital gain. The link to the level of indebtedness of the Swedish households is also worthwhile to consider in this context, as the deferred tax payments constitute an additional debt weighing on households' balance sheets. The downside risks for consumption are larger in a stagnating market, as tax payers having previous deferrals and subsequently selling their dwellings at a loss, are still obliged to pay a certain share of the capital gains tax related to the initial investment.

In the Swedish context, consumption loans are also increasingly warranting some concern. In particular short-term credits (so called text loans, or SMS loans, due to the instant application procedures via text messages) may be troublesome as they risk attracting predominantly vulnerable groups such as young adults. The Swedish Enforcement Authority reports that in 2012, the number of enforcement cases grew by 62% and that approximately one fifth of all cases concern young adults between 18-25 years old. The Authority finds that the procedures applied to assess the credit worthiness of borrowers are insufficient and that interest rate levels are problematic, in particular as the text loans target consumers with an already weak standing (Swedish Enforcement Authority, 2013). In Finland, the government has recently proposed a new piece of legislation introducing a cap of the reference rate plus 50% effective interest rate for consumer credits under EUR 2,000 combined with stricter application procedures[37]. The Swedish Enforcement Authority has proposed that similar limitations should be introduced also in Sweden. The Swedish government has declared that as a first step, the Swedish FSA will get increased possibilities to impose sanctions on market actors, and as a second step a legislative proposal to address the situation will be presented most likely by mid-2013 giving the Consumer Protection Agency the possibility to fine credit providers that fail to check the credit worthiness of applicants appropriately.[38]

3.2. Indebtedness of the non-financial corporate sector 3.2.1.  Recent developments in non-financial corporate debt

Debt of non-financial corporations accounted for 149% of GDP in 2011 and was the fifth highest in the EU, far above the euro area average (99% of GDP). Corporate debt has gradually declined by 20 percentage points since its peak in 2009. In 2012, the correction ceased because of a sharp drop in GDP growth which was not followed by the same deceleration in credit to corporations. While the GDP is forecast to have grown by a mere 1% in 2012, credit to corporations grew by 3.8%. Therefore, it is expected that corporate debt might increase marginally as a percentage of GDP in 2012. 

In terms of consolidated data, Swedish corporate debt is lower but still high in EU comparison. In 2011, consolidated debt of Swedish corporations stood at 133% of GDP, which was the fourth highest level in the EU (to compare with the euro area average of 84% of GDP). In Graph 34, consolidated debt is illustrated by the total of the red and blue areas. Due to reasons explained below as well as specific statistical issues, consolidated data are more appropriate in assessing the debt burden of Swedish corporations. [39]   

Graph 34. Corporate debt as % of GDP: consolidated and super-consolidated view

Source: Statistics Sweden

A large share of corporate debt takes the form of cross-border intra-group loans which entail lower risks than other types of borrowing.  Corporate debt consists mainly of ordinary loans (46% in 2011) and inter-company loans and trade credits (42%), while borrowing through securities forms a smaller part (10%). Inter-company loans from abroad constitute about one third of consolidated corporate debt (corresponding to 47% of GDP in 2011), which is among the highest shares in the EU. It was also this type of borrowing which increased most markedly over the past decade. As explained in the last year's review and further in section 3.2.2, the wide use of intra-group borrowing from abroad is motivated by efficient tax minimisation and not by a need to cover up for insufficient profits. From a sustainability point of view, intra-group loans entail much lower risks than ordinary loans from banks or from other corporations. Multinational companies typically cover net debts in one country by net assets in affiliates in another country. A majority of intra-group lending to the country (55% for Sweden) is directly channelled to other affiliates abroad. Therefore, risks attached to intra-group lending would have to be assessed on the basis of net lending/borrowing positions of multinational groups, the data on which are, however, not available.

In this light, the debt-to-GDP ratio draws an exaggerated picture of indebtedness of Swedish corporations. If cross-border inter-company lending were deducted from the consolidated data, Swedish corporate debt would amount to 87% of GDP in 2011, a level comparable with the European averages (see the blue area in Graph 34). Also, the development of this measure over time would seem less worrying since it has been very stable between 1995 and 2006 and accumulation of debt was limited only to period 2006-2008 (and a part of this accumulation has already been corrected). 

Given the drawbacks of the debt-to-GDP indicator, it is useful to complement the analysis with other indicators of corporate sector leverage. Balance-sheet indicators suggest a rather healthy situation of non-financial corporations. The ratio of debt to financial assets (51% in 2011) in Sweden is below the euro area average (56%). The debt-to-equity ratio (58%) is even far below the euro area average (78%). Over the past decade, both indicators have been relatively stable, or even declined slightly, confirming that liabilities grew hand in hand with financial assets/equity. In contrast, the ratios have increased moderately in the euro area during the same period.  

Graph 35. Leverage ratios, Non-financial corporations, Sweden || Graph 36. Leverage ratios, non-financial corporations, euro area 

Source: Commission service (Eurostat) || Source: Commission servces (Eurostat), ECB

However, some caution is needed when regarding growing assets as a mitigating circumstance for mounting debt. First, it is useful to determine to what extent accumulation of debt was facilitated by an increasing value of assets. These valuation effects are unstable and tend to evaporate in bust periods, therefore they may overestimate the firms' ability to incur liabilities. If these effects are discounted for, the leverage ratio (called also "notional leverage"[40]) for Sweden appears to follow an upward trend (Graph 37). This contrasts with the situation in the euro area where the average notional leverage has not diverged significantly from the standard ratio compared to 1995. Second, it is pertinent to look at the maturity composition of assets and liabilities. Sweden shows a rather high and rising share of volatile assets[41] (76 vs. 66% in the euro area), whereas almost the entire debt is long-term (97 vs. 74% in the euro area, 2011), suggesting a growing maturity mismatch between assets and liabilities.

Graph 37. Debt-to-assets ratio – standard vs. notional leverage

Source: Commission services (Eurostat)

Indicators of the capacity to repay do not suggest a clear need for deleveraging. If gauged against gross operating surplus, Swedish corporate debt remains the fifth highest in the EU and on an increasing trend. However, this indicator suffers from the same bias as the debt-to-GDP ratio, given the large share of intra-group lending.[42] The interest burden (9.6% of value added in 2011) is above the euro area average (5.7%) but has come down significantly since the 1990s along with declining interest rates.

Graph 38. Debt-to-gross operating surplus,  Sweden  vs. euro area || Graph 39. Interest burden, Sweden vs. euro area 

Source: Commission services (Eurostat) || Source: Commission services (Eurostat), ECB

Debt sustainability analyses do not point to imminent deleveraging pressures. Cluster and principal component analysis which relates the level of debt with the size of accumulation (both for the debt-to-assets and debt-to-operating surplus ratios) identifies Sweden as prone to suffer from deleveraging pressures, but risks are considered lower than in other similarly-indebted countries (Cuerpo, Drumond, Lendvai, Pontuch and Raciborski, 2013).[43] Typically, if the size of debt accumulation over time is considered, Sweden appears in a relatively low-risk position. Indeed, compared to the 2000 level, non-consolidated corporate debt as a percentage of GDP has increased only by some 14 percentage points, which is less than in most other countries (in the euro area, debt-to-GDP increased by 28 pps).[44] On the other hand, analyses based on the level of the debt indicate a clear need for deleveraging (see the above discussion on debt-to-GDP or debt-to-operating surplus). Following the approach presented in Cecchetti et al.[45] or Arcand et al.,[46] Swedish corporate debt would also be far above the levels where debt becomes a drag on growth.  However, the utility of all indicators relating corporate debt to GDP or operating surplus is undermined in the case of Sweden by the overstated level of the debt-to-GDP ratio due to a high share of intra-group borrowing from abroad. 

The relatively healthy financial situation of Swedish corporations is also illustrated by a low rate of company defaults, even in weak economic environment. The default rate for Swedish companies declined from 1.-1.5% before the 1990s crisis to around 0.8% in 2012. It did not increase substantially even during the crisis of 2008/09, suggesting that Swedish companies can cope with their debt levels even under a sharp economic downturn (Graph 40).

Graph 40. Company default rate (in %)

  

Source: Sveriges Riksbank

Notwithstanding the indications of contained deleveraging pressures, the high level of Swedish corporate debt still calls for vigilance. The level of corporate debt remains high even if the inflating effect of intercompany loans is disregarded (above 100% of GDP in 2011, non-consolidated). The strong deleveraging in 2009-10 indicates that the rapid debt built-up in the preceding years might have been excessive pushing for a return to a more sustainable path. Although intra-group loans involve clearly lower risks than other forms of debt, they should not be fully ignored since some minor risks may remain with respect to the overall debt-asset position of the multinational groups, as well as with respect to maturity and currency composition of this cross-border lending.

3.2.2. Factors behind Swedish corporate indebtedness 

The higher level of corporate debt in Sweden can be explained by several Sweden-specific factors. These would include a strong presence of multinational companies, company taxation and a long-period of an easy access to credit.

The fact that Sweden has traditionally been the cradle of a number of large, multi-national companies tends to amplify corporate debt vis-à-vis the Swedish GDP. While recording their debts in Sweden, the multinational companies incur income from global sales. Therefore, their indebtedness should be viewed against their global activities.

As mentioned earlier, another feature distinguishing Sweden is the wide use of borrowing from foreign affiliates of the same corporate group, a phenomenon which is motivated by the set-up of company taxation in Sweden. The generous tax deductibility of interest payments has given an incentive to Swedish companies to finance their investments with debt rather than equity. If combined with a relatively high corporate income tax and double taxation agreements with several countries, it invites to an aggressive tax planning by multinationals operating in Sweden. Multi-national companies typically channel loans from affiliates in countries with lower corporate taxes to their Swedish affiliates which then deduct the interest payments from taxable income. The multinationals can determine the interest costs paid on loans to their Swedish affiliates (be it an owner or a subsidiary) to influence the size of profits to be taxed in Sweden. By charging a higher than the market interest rate on these loans, multinationals succeed in avoiding profit taxation in Sweden.[47]

The tax minimisation also explains why FDI into Sweden takes predominantly the form of share acquisitions mirrored by a loan from the foreign parent company. In 2010, 35% of FDI in Sweden was financed through loans, while the comparable figure for Swedish FDI abroad stood at mere 4%.[48] Foreign investors typically establish a holding company in Sweden with a small equity capital but a large loan which allow the holding company to buy a decisive share of the Swedish company (Sveriges Riksbank, 2012(b)) . This behaviour can be illustrated by the fact that loans from parental companies account for more than 60% of all intra-group lending from abroad and have seen a steeper increase than loans from foreign subsidiaries of Swedish-owned companies. Given the tax deductibility of interest payments, it is more profitable for foreign investors to hold assets in Sweden in the form of interest-bearing loans to holding-companies in Sweden and not as shares in the acquired Swedish company. Foreign owners typically charge interest payments on loans to their Swedish subsidiaries (which have been deductible from taxable profits) that are higher than possible dividends from share ownership (which would have been derived from after-tax profits).

The government has recently taken some measures to limit the most far-reaching tax minimisation practices by the multinationals. In 2009, tax regulations were introduced which reduced the tax deductibility of interest payments for intra-group loans related to the acquisition of shares from an affiliate. Following a series of public inquiries on various related issues, new legislation came into force in January 2013 which extends the restrictions on interest deductibility to all types of loans, regardless of their purpose (PWC, 2012)[49]. These measures are likely to eliminate a large share of corporate debt which is solely driven by tax avoidance, in particular the intra-group loans. Intra-group lending from abroad has become further less attractive following the reduction in the corporate income tax from 26.3 to 22% as of beginning of 2013. Nevertheless, it remains to be seen how these steps will eventually affect the level of corporate debt. Some aspects of the new legislation bring about a high degree of uncertainty about how different cases will be treated by the Swedish Tax Agency (namely uncertainties about the interpretation of "sound commercial reasons" which may justify the use of intra-group lending). Also, since the stripping of interest deductibility is limited to internal loans, it may not fully avoid external loans in acquisition structures and it does not solve the systemic debt-bias in company taxation.

Finally, escalation of corporate debt in 2003-2008 has been also facilitated by the low-interest-rate environment and low perceived risk of lending to corporations. Following the economic slowdown in 2001-03 related to the burst of the IT bubble, interest rates remained at very low levels for several years. This spurred high demand for credit by Swedish companies. Low interest rates coincided with a rather lenient lending attitude by Swedish banks towards both corporations and households. In view of a long period of a low rate of company defaults, the banks may have started to underestimate the risks associated with lending to corporations. This led to an excessive expansion of credit to corporations in 2006-2008 (annual growth of 16%). Still today, there are concerns that banks may in their internal risk assessment models assign too low risk weights to loans to the corporate sector. 

  

3.3. Linkages from private debt to the real economy, the financial sector and public finances

At moderate levels, debt is beneficial for the economy and leads to higher growth. If economic agents are able to borrow and save, they can consume and produce even without current income. However, if debt exceeds a reasonable level, it can act as a trigger of long and painful financial and economic crises, marked with a number of bankruptcies and contraction in consumption and investment. Given the interlinkages between the sectors of the economy, high indebtedness of one sector tends to have negative consequences for other sectors. Even if the banking and public sectors are regarded as sound, a major need for deleveraging of households or non-financial corporations can change the status of the sector substantially.

3.3.1.     Risks for the real economy

Sweden has already experienced a financial crisis driven by overindebtedness already in the early 1990s. The crisis was preceded by a housing and credit boom triggered by the abolition of credit regulations in 1985. The corporate debt level had risen to about 120% of GDP in 1992 and household debt exceeded 60% of GDP.  The necessary deleveraging process was painful and sent GDP growth to negative figures for three consecutive years. Consumption contracted as house price falls forced households to repair their balance sheets. The investment ratio plummeted from 23% in 1990 to 15.5% in 1994 and it took a further 15 years for the investment ratio to catch up with the EU levels. The debt-to-GDP levels decreased by some 25 percentage points for the corporate sector and household leverage declined by 15-20 percentage points between the peak and the trough.

Deleveraging typically affects output via falling consumption and investment but also through a debt-deflation spiral. On the former, in reaction to a fall in asset prices, households repair their balance sheets by cutting on consumption and lowering their demand for housing investment. On the latter, despite deleveraging in nominal terms, falling prices slow down deleveraging in terms of real debt, which forces households to deleverage even more aggressively. Low inflation or deflation results in high real interest rates which further discourage investment.  In the Swedish context, it could be mentioned that an independent monetary policy is able to cushion these effects of deleveraging. While in deleveraging parts of a monetary union, falling prices typically lead to high real interest rates, an independent monetary authority can lower policy rates drastically to fight deflation and to keep real interest rates low. Also, positive inflation helps to erode the ratio of debt to nominal GDP.

Deleveraging of the private sector may also leave permanent trace in terms of lower potential output. The capital stock can be negatively affected due to a decrease in the investment level by deleveraging companies. The stock of investment often does not reach the pre-crisis level. There may be also permanent negative effects on the labour force due to exclusion from the labour market following long-periods of unemployment. 

Notwithstanding the relatively low imminent risks of deleveraging suggested by various analyses discussed earlier in the note, the effects of potential deleveraging on the real economy should not be neglected. An internal QUEST model[50] simulates the impact of household deleveraging combined with a fall in house prices on the main macro-economic aggregates (Cuerpo, Drumond, Lendvai, Pontuch and Raciborski, 2013). It shows that a shock leading to a 20 pps reduction in the households' debt-to-GDP ratio and a 15-percent fall in house prices in the long run would decrease output by 3% in the first three years and have a negative permanent effect on GDP of about 1.5%. The unemployment rate would rise by 1.7 pps. Consumption would decline by 3% and investment drop by 6% within the first three years after the shock. The model also predicts a prolonged period of low housing investment. These results are broadly consistent with the deleveraging experience in Sweden in the 1990s. However, while the model estimates that household consolidation lasts more than  ten years, with half of the consolidation happening in the first six years, the Swedish experience suggests a much faster pace of deleveraging (the main consolidation phase was completed in four years, reducing the debt to GDP ratio by 10 pps).

3.3.2.     Risks for the financial sector

Compared to the situation before the 1992 crisis, but also in comparison with the other EU countries, the Swedish financial sector appears to be in a relatively good shape. Swedish banks are well capitalised,  their profitability measured as returns on equity has improved over the last year and the share of underperforming loans is very low, banks' exposure to countries in receipt of EU or IMF financial assistance is marginal. However, the financial sector continues to be vulnerable to risks related to its size, strong international exposure and heavy reliance on short-term funding, largely in foreign currencies. Market funding still being a major source of funding, the ratio of loans to currency and deposits is high, amounting to 117% in 2011, compared to 79% for the euro area. Moreover, the majority of wholesale funding in the form of securities issuance takes place in foreign currencies (Sveriges Riksbank, 2012).

These risks have been addressed thoroughly by the Swedish authorities since the 2008/2009 crisis through a number of measures to strengthen the resilience of the financial sector. The major step was a decision to apply stricter requirements for capital adequacy and liquidity by banks ahead of the Basel III schedule.[51] The risks have been also picked up by the Financial Crisis Committee, which was set up by the Swedish government in 2011 with the task of preventing and managing financial crises. The Committee proposes the establishment of a macroprudential council in Sweden which, among other tasks, should "analyse and present proposals on how to improve crisis prevention" for instance by developing tools that manage "risky interdependence and links between players and markets, for example, rules on debt levels" (Statens Offentliga Utredningar, 2013:6).

Although private sector debt is of course a cornerstone of a modern economy, it also constitutes a direct risk to the financial sector through two channels. First, there is a risk of loan losses related to defaults on mortgages or non-performing loans to corporations. At the moment, this risk seems to be rather limited. However, a long period of low default rates may have led banks to underestimate risks related to household and corporate lending. The second channel - risks related to a loss in investor's confidence - appears more likely. A downturn in sentiment driven by international developments or an asset price fall in Sweden could result in a more expensive or difficult access to funding in international markets.

An analysis of credit supply and demand conditions suggests that the probability of deleveraging is rather low in Sweden. A composite indicator of credit supply and demand pressures places Sweden among countries with the lowest short-term deleveraging pressures in the EU (Cuerpo, Drumond, Lendvai, Pontuch and Raciborski, 2013). The supply of credit is sufficient[52], facilitated by an easy access by the Swedish banks to cheap foreign funding. This is related to the relatively high perception of the soundness of Swedish banks, if measured by capital adequacy, return on equity or the share of non-performing loans. The risk of a spill-over from the sovereign debt is low, as public debt-to-GDP ratio stands below 40% and sovereign credit default swaps are among the lowest in the EU. On the demand side, the current risks are slightly higher given the recent deterioration in the economic outlook and economic sentiment. Higher expected unemployment and continued uncertainties about the external development are likely to have a depressing effect on credit demand. On the other hand, the housing market has been resilient over the past year and given the high sensitivity of Swedish households to interest rates, which have fallen considerably over the past year, sustained demand for mortgage lending can be expected also in the future.

3.3.3.     Risks for public finances

Financial crises and deleveraging affect negatively also the government budget balance and public sector's indebtedness. In the wake of the Swedish financial crisis of 1992-1993, the government balance has plummeted from a surplus of 2% of GDP in 1990 to a deficit of 11% of GDP in 1993. The hand-over of debt from the private to the public sector was also particularly evident in the countries which experienced a property bubble, such as Ireland, the UK, the USA or Spain.

On the revenue side, tax receipts tend to drop significantly with falling asset prices. Experience from countries with housing busts show that construction booms can lead to an overestimation of the soundness of the public finances. The decline is particularly noticeable in the performance of capital taxes, such as capital gain tax, property taxes and stamp duties. In Ireland, collected capital taxes dropped by 80% between 2006 and 2009 and their share in total tax revenue declined from 16 to 5%. In addition, tax receipts further fall due to the second-round effects on consumption and production. Households cut back on consumption in order to restore their balance sheets. Companies decrease investment and restructure their balance sheets by selling assets. Lower production leads to higher unemployment and thereby significant losses in revenues from taxes on income.

In the Swedish context, the unlimited tax deductibility of interest payments for both households and corporations translates into significant foregone revenues for the state budget. To illustrate the size of these revenues, the partial stripping of the tax relief on interest payments on intra-group loans in 2013 is expected to generate an additional SEK 8.8 billion, corresponding to 0.2% of GDP.

On the expenditure side, private sector deleveraging affects public finances via the automatic stabilisers. Higher unemployment requires more spending on unemployment benefits, retraining, subsidised jobs and other social allowances.  Also, a drop in confidence of investors can make servicing of public debt more expensive due to higher interest rates charged on government bonds.

The effects on the public debt are substantial and long-lasting. The above mentioned QUEST simulation suggests that deleveraging amounting to 20-pps reduction in the households' debt-to-GDP ratio, combined with a 15-percent fall in house prices in the long run leads to a deterioration in the public debt-to-GDP ratio by about 11 percentage points in the first ten years. The Swedish experience from the early 1990s indicates that effects on public debt can be even stronger. Public debt jumped from 43% of GDP in 1990 to 78% in 1994 and it took fifteen more years of responsible consolidation to bring the debt ratio back below 40%.

 

4.           Policy challenges

The analysis in sections 2 and 3 indicates that developments in the areas of private debt and the housing market are the main potential sources of macroeconomic imbalances in Sweden.

These challenges were already identified under the MIP in the first IDR and relevant policy responses were reflected and integrated in the country-specific recommendations issued for Sweden in July 2012 (CSR 2). Progress in the implementation of those recommendations will be analysed when assessing the Swedish National Reform Programme and Convergence Programme under the European Semester. Against this background, this section discusses different policy avenues that could be envisaged to address the challenges identified in this IDR.

Concerning the challenge of high private debt, a number of different avenues can be considered:

Notwithstanding recently introduced policy measures, risks related to the high indebtedness of the private sector remain. Further reforms along the lines detailed below would reduce the exposure of the Swedish economy to potential imbalances as high debt levels pose a downside risk to consumption and investment, but also to the health of the financial sector, and in extremis to the currently sound public finances.

Housing taxation: The analysis identified a clear debt-bias in housing taxation that creates incentives for households to finance their investment in housing through debt rather than own savings. The Swedish tax system offers a generous deductibility of interest payments on mortgages. At the same time, following the reform of property taxation, the recurrent property taxes have practically disappeared and ceased to counterweight the effect of the tax relief on mortgages. Due to the reduced relative price of debt financing, households have often preferred to save in other forms (such as investing in the stock market) rather than repaying their mortgages. This has contributed to an increased leverage of the household sector.

To reduce the risk of an unsustainable accumulation of household debt, one avenue could be to correct the debt bias in housing taxation by a global reform gradually phasing-out the tax deductibility of interest payments while at the same time strengthening again the recurrent property taxation. Such a reform would not only restore neutrality among investment alternatives, but also free up fiscal space to reduce taxes that are more harmful to growth, such as labour income taxes. In this context, possible risks of adverse social effects of a property tax reform need, however, to be kept in mind.[53] Introducing amortisation requirements and favouring lending based on real repayment ability, rather than the value of the asset, could also be envisaged. Making it less costly for borrowers to switch between fixed and variable mortgage rates could also give incentives to households to opt for fixed rates.

It is clear that the timing of reforms would need to be well chosen, in order not to feed the remaining uneasiness of the housing market. A sequenced approach also seems warranted in this context. However, although the timing of the actual implementation of the reforms could be discussed, their content and structure may already be established so that the roll out is facilitated once the timing is judged to be appropriate. 

Financial supervision: Household debt accumulation has been facilitated by the rather lenient lending standards of Swedish banks. Since 2000, there has been a trend towards ever longer amortisation periods, with new mortgages often including no or very limited amortisation. Debt-servicing consisting mainly of interest payments, has enabled households to take on larger loans, especially as mortgage rates have trended down at the same time. In addition, risk weights on mortgages applied by Swedish banks have so far been among the lowest in the EU and may not reflect the real risks related to high household indebtedness.

To prevent a continued increase in household indebtedness, amortisation requirements could be envisaged for new mortgage loans. Moreover, prudent lending could be fostered by increasing the risk weights for mortgages. On both strands, the authorities are already taking some steps. The Financial Supervisory Authority has launched a public consultation on a possible introduction of a risk weight floor of 15% for mortgages. The Association of Swedish Bankers also put in place a recommendation on amortisations in December 2010, which many banks claim to follow. Discussions are also continuing between the FSA and the Riksbank on this topic.

Corporate taxation: As described in the analysis above, corporate taxation has been a decisive contributing factor behind increasing corporate indebtedness in Sweden. The combination of a generous tax deductibility of interest payments and a relatively high corporate income tax in Sweden has inspired advanced tax planning by multinational companies and foreign owners of Swedish companies. This has led to an escalation of corporate debt, in particular in the form of inter-company loans from abroad. Even for local companies, the tax deductibility has made it more beneficial to finance their investments with debt rather than equity. Although intra-group loans clearly involve lower risks than other types of debt, this form of debt should not be fully ignored since some risks may remain with respect to the overall debt-asset position of the multinational groups, as well as with respect to maturity and currency composition of this cross-border lending. Also, the tax planning entails non-negligible amounts of foregone tax revenues for the state budget.

Within the 2013 budget, the government has introduced several reforms to company taxation which address these issues. Firstly, the corporate income tax has been lowered from 26.3 to 22%. Secondly, new legislation on the tax deductibility of interest payments entered into force to curtail the aggressive tax minimisation practices by multinational companies. These steps go in the right direction, but it remains to be seen what effect they will eventually have on the level of corporate debt. Also, further ways to reduce the debt-bias in company taxation could be considered, such as a generally applicable cap on the tax relief for interest payments. It would not only bring benefits in lowering corporate indebtedness, but also higher tax receipts.

Concerning the challenges in the housing market, a number of different avenues can be considered:

The Commission's analysis shows that inefficiencies still weigh on the Swedish housing market. These are likely to have significant effects on the economy as a whole since they contribute to a) long-term underproduction of dwellings; b) higher prices of tenant-owned apartments and one-dwelling-building due to lack of available rental units. The latter subsequently spills over into a risk for higher household indebtedness. Therefore, even though some measures have been undertaken recently, further policy actions could be envisaged to improve housing supply as well as market efficiency.

Rental market: The analysis shows that rent regulation tends to impede a rational development of the housing stock and to keep house prices at an elevated level. The functioning of the rental market might be enhanced in the long term by moving further towards market-clearing rents. The characteristics of the rent-setting system have limited rents' response to market signals over several decades. Rents are set according to the utility value system, which does not fully allow for dispersion in rent levels between metropolitan city centres and suburbs. For urban areas, this has led to landlords not being able to charge tenants market-clearing rents in central city locations as compared to suburban areas, although the size of the investment differs substantially, thereby discouraging investments in rental housing. The insufficient supply on the rental market creates an emphasised pressure on dwellings of other types of tenure, most notably on tenant-owned apartments (which are the closest substitute in urban areas) the prices of which escalate as has been showcased in Sweden over the last fifteen years. It also creates an undesired indebtedness and risk exposure for individuals who are in a sense forced to own although they would actually rather rent if the opportunity were available. Restrictions on private letting of tenant-owned apartments have until now further contributed to a suboptimal supply of housing services. Most municipalities located in urban areas face a shortage of available rentals and administer queuing systems for allotting tenancies.

The government has already taken some measures to reduce the negative effects of rent regulation, e.g. by a partial reform to the rent setting system in 2011 and by a simplification of private lettings. With a view to increasing the flexibility of the housing market and limiting upward price pressures, further easing steps in the direction of market-clearing rents might be taken to allow market forces to establish an optimal supply of rental housing at an adequate price. Possible avenues to explore in this context could be to open up for allowing rents of newly produced rental units to reflect tenants' willingness to pay as a first step. As a second step, rent levels of the remaining stock of rentals could be gradually adapted (cf. the Finnish experiences of deregulating the private rental market in the 1990s). It might also be worthwhile to explore how the role of individual tenants in rent negotiations might be strengthened, and to allow for an increased freedom of contract.

Housing supply: The analysis also highlighted some restrictions and rigidities in the construction sector that tend to hold back construction activity, create an upward-bias in house prices and at the same time affect growth negatively by limiting mobility of workers and households in general. These institutional features include lengthy and intransparent zoning and planning processes and an uneven application of rules for the issuance of building permits by municipalities that in practice have a local planning monopoly. Low predictability of potential returns from construction investments has a clear discouraging effect for potential investors. Cumbersome and intransparent rules create high entry barriers for new entrants, especially for small or foreign companies. The Swedish construction sector has traditionally been dominated by large national companies with sufficient financial and human resources to face the challenges, a situation which has limited the competition in the sector.

In this context, a streamlining of the planning and zoning processes could be considered. In particular, these processes could be rendered more efficient by limiting the appeal procedures and clarifying the cost responsibilities. Also, municipalities could avoid insisting on specific technical requirements going beyond the common set of rules. A more regional view when evaluating municipal housing needs, and increased cooperation between neighbouring municipalities in planning the necessary infrastructure could also be a way forward. Within public procurement, splitting tenders into smaller lots with shorter call periods and more specific requirements in the call for tenders would also favour more competition and the involvement of SMEs and foreign companies in the market. Some initiatives along these lines have been already introduced by the Swedish authorities and others are under consideration. It could also be evaluated whether giving municipalities clearer incentives (for example through the introduction of fees in case of delays, allowing counterparts to appeal in case of turned down building permit demands etc.) to process planning and zoning requests more efficiently might alleviate the administrative burden for the construction companies and at the same time shorten the time required to construct.

  

              References

Arcand, J.-L., Berkes, E. and Panizza, U., Too much finance?, IMF Working Paper No. 161, 2012.

Cecchetti, S., Mohanti, M.S. and Zampolli, F., The real effects of debt, BIS Working Paper No.352, 2011. 

Confederation of Swedish Enterprise, Släpp strypgreppet kring nya bostäder – förutsättning för jobb och tillväxt, December 2012.

County Administrative Board of Stockholm (Länsstyrelsen i Stockholms län), Läget i länet: Bostadsmarknaden i Stockholms län 2012, Rapport 2012:16, 2012.

County Administrative Board of Stockholm (Länsstyrelsen i Stockholms län), Bostäder i andra hand – en översikt, Rapport 2013:4

Cuerpo, C., Drumond, I., Lendvai, J., Pontuch.P. and Raciborski, R.: `Indebtedness, Deleveraging and Macroeconomic Adjustment', European Economy - Economic Papers, forthcoming, 2013.

Demoskop (for SEB), Boprisindikatorn, December 2012, January 2013, February 2013.

European Commission, Possible reforms of real estate taxation: Criteria for successful policies, European Economy, Occasional papers 119, October 2012.

European Commission, Innovation Union Scoreboard, 2012.

European Commission, Employment and Social Developments in Europe, 2012 (b).

European Commission, Macroeconomic Imbalances – Sweden, Occasional Paper No. 108, 2012(c). 

Girouard, N., Kennedy, M., van den Nood, P. and Andre, Ch., Recent house price developments: the role of fundamentals, Economic Department Working Papers No. 475, OECD, 2006.

Kalbro, T., Lindgren, E. and Paulsson, J., Detaljplaner i praktiken – är PBL i takt med tiden?, Royal Institute of Technology (KTH Arkitektur och Samhällsbyggnad), 2012.

Kulander, M., Lind, H., Lundström, S., Hur skulle hyresmarknaden i Stockholm påverkas av friare hyressättning?, uppsats nr 45, Royal Institute of Technology (Bygg- och fastighetsekonomi, Institutionen för fastigheter och byggande, KTH), 2008.

Lind, H., Social housing in Sweden, Article in Social Housing in Europe, London School of Economics, forthcoming new edition 2013.

Lindblad, M., Marknadshyror och nyproduktion – en hyresmodell applicerad på Stockholms bostadsmarknad, Examensarbete nr 505, KTH Arkitektur och samhällsbyggnad (Royal Institute of Technology), 2010.

National Institute of Economic Research, Business Tendency Survey, 2012.

OECD, The Swedish Housing Market: Better Allocation via Less Regulation, OECD Economics Department Working Papers, No. 559, 2007.

OECD, OECD Economic Surveys: Sweden, Volume 2012/Supplement 3, December 2012

OECD, Reviews of Innovation Policy: Sweden, preliminary version, 2013.

PWC, Real Estate Tax Services News Alert, Sweden – October 2012.

PWC, Entreprenadbarometern 2013, February 2013.

SEB, Hög skuldsättning – bara ett problem för de rikaste, November 2012.

SKOP, Hushållens förväntningar om bostadsmarknaden, 3 January 2013.

Statens Offentliga Utredningar, Ett effektivare plangenomförande – Betänkande av Plangenomförandeutredningen, 2012:91

Statens Offentliga Utredningar, Ökat bostadsbyggande och samordnade miljökrav – genom enhetliga och förutsägbara byggregler, 2012:86

Statens Offentliga Utredningar, Att förebygga och hantera finansiella kriser, 2013:6

Statistics Sweden, (Statistiska centralbyrån), Bostadsbyggandet lågt under lång tid, SCB-indikatorer - Ekonomisk månadsöversikt, Nummer 6, 28 June 2012 (b).

Statistics Sweden (Statistiska centralbyrån), Sveriges ekonomi – statistiskt perspektiv, Nummer 4, 2012 (c).

Statistics Sweden, (Statistiska centralbyrån), Bostadsrätterna fortsätter öka på hyresrätternas bekostnad, Nr 2012:760, 30 May 2012, 2012 (d).

Statistics Sweden (Statistiska centralbyrån), Medelpriset för ett småhus över en miljon i samtliga län, Nr 2013:27, 7 February 2013.

Statistics Sweden (Statistiska centralbyrån), Bostadsbyggandet minskade med 23 procent, Nr 2013:47, 21 February 2013 (b).

Svensk mäklarstatistik, Prisuppgång för bostadsrätter under 2012, (http://www.maklarstatistik.se/media/13408/f%c3%b6rdjupad_statistik_201302.pdf), February 2013.

Svensk mäklarstatistik, Priser på bostadsrätter fortsätter stiga, 14 February 2013 (b).

Sveriges Riksbank, Finansiell stabilitet 2012:2, 2012.

Sveriges Riksbank, Tax planning may have contributed to high indebtedness among Swedish companies, Economic Commentaries No.3/2012 (b)

Swedbank, Economic Outlook, 16 January 2013.

Swedbank, Boindex, 7 March.2013 (b).

Swedish Centre for Business and Policy Studies (SNS), Konjunkturrådets rapport 2013, January 2013

Swedish Competition Authority, Konkurrens- och upphandlingsproblem i byggbranschen, speech given by Dan Sjöblom, Director General, at the Conference "Entreprenadrisker inom bygg och anläggning in Stockholm" on 28 September 2011.

 

Swedish Competition Authority, Statement of opinion Dnr 562/2012, 18 January 2013.

Swedish Construction Federation (Sveriges Byggindustrier), Byggkonjunkturen Nr 4, 12 December 2012.

Swedish Enforcement Authority, Obetalda snabblån. Konsekvenser hos Kronofogden, report of 31.01.2013.

Swedish Financial Supervisory Authority (Finansinspektionen), Riskviktsgolv för svenska bolån, Promemoria FI Dnr 12-11920, 26 November 2012.

Swedish Financial Supervisory Authority (Finansinspektionen), Den svenska bolånemarknaden, March 2012 (b).

Swedish Financial Supervisory Authority (Finansinspektionen), Den svenska bolånemarknaden 2013, March 2013.

Swedish National Board of Housing, Building and Planning (Boverket), Bostadsmarknaden 2012-2013, 2012(a).

Swedish National Board of Housing, Building and Planning (Boverket), Bostadsbristen ur ett marknadsperspektiv, Rapport 2012:18, 2012(b).

Swedish National Board of Housing, Building and Planning (Boverket), Boverkets indikatorer: Analys av utvecklingen av bygg- och bostadsmarknaden med byggprognos, November 2012 (c).

Swedish National Board of Housing, Building and Planning (Boverket), Boverkets översyn av bostadsförsörjningslagen, Rapport 2012:12, 2012 (d)).

Swedish National Board of Housing, Building and Planning (Boverket), Varför amortera, Marknadsrapport, October 2012 (e)).

Swedish National Board of Housing, Building and Planning (Boverket), Choices on the Housing Market, Market report, May 2012 (f)).

Swedish National Board of Housing, Building and Planning (Boverket), Drivs huspriserna av bostadsbrist?, Market report, February 2013.

Swedish Property Federation (Fastighetsägarna), Missbruket av bytesrätten – en rapport om svarthandeln med hyreslägenheter i Stockholm, 2006.

Swedish Tax Agency (Skatteverket), En avskaffad uppskovsrätt – bilaga 1, Promemoria, Annex to Dnr 131-203632-12/1152, 17 December 2012.

World Bank, Doing Business report, 2013.

World Economic Forum, Global Competitiveness Report, 2012-13.

[1] If measured by market value, Swedish NIIP would be positive (5.7% of GDP in 2011).

[2] European Commission, Current Account Surpluses in the EU (2012).

[3] Foreign assets are mainly denominated in foreign currencies (88% in mid-2012), whereas the majority (58% in mid-2012) of foreign liabilities is denominated in the national currency.

[4] Other business services comprise mainly merchanting and other trade-related services, operational leasing services and miscellaneous business, professional and technical services

[5] http://epp.eurostat.ec.europa.eu/statistics_explained/index.php?title=File:Gross_domestic_expenditure_on_R%26D,_2000-2010_(%25_share_of_GDP).png&filetimestamp=20121016060906 and http://epp.eurostat.ec.europa.eu/statistics_explained/index.php?title=File:Gross_domestic_expenditure_on_R%26D_by_sector,_2005_and_2010_(%25_share_of_GDP).png&filetimestamp=20121016060835

[6] It is necessary for banks to hold a certain amount of Tier 1 capital on their balance sheets for all loans they issue. The risk profile of the loans determines how much capital has to be set aside. So far, Swedish banks' risk weights for mortgages have been much lower than those of other European banks (5-10% compared to 20-25%). (WSJ, 26.11.2012).

[7] In order to identify unsustainable developments in housing markets, affordability (price-to-income) and dividend (price-to-rent) ratios can be compared to their long-term averages; the gaps between the actual value of the former and the corresponding long-term averages provide indicators of the degree of over- or under-valuation of house prices. Conclusions based on such indicators must be treated with caution due to the underlying assumptions. Meaningful comparisons with long-term averages generally require stationary series. However, unit root tests point to non-stationary affordability and dividend ratios in many countries.

[8] European Commission, 2012(c).

[9] In view of the characteristics of the Swedish rental market, which includes rent setting based on the so called utility-value-system (described in section 3.1.3), the value added of analysing price-to-rent ratios might be discussed.

[10] The fundamental ratio takes into account structural changes, such as the development in user cost of home ownership. User cost reflects changes in mortgage rates, housing taxation, depreciation and expected capital gains/losses. For further details on the approach, see Girouard, Kennedy, van den Noord and André (2006).  

[11] This should be seen against a population of roughly 9.5 million.

[12] Tenant-owned apartments ("bostadsrätter") are apartments owned by an association, in which the respective residents own a share giving them the right to reside in a particular apartment. This means that you do not in theory own a particular apartment and decisions regarding more substantial renovations and sub-letting of particular apartments (as well as decisions about major renovations affecting the collective parts of the property such as main ducts, stairways, entrance hall, garden etc) have to be taken at collective level.

[13] Owner-occupied apartments constitute a new form of tenure in Sweden, having been introduced in May 2009 and so far being applicable only to newly constructed buildings. While tenant-owned apartments are actually not the property of the resident-buyer (please see above), owner-occupied dwellings represent a three-dimensional property. The new form of tenure has not yet succeeded in attracting the full attention of investors, but to widen the accessibility the government has commissioned a public inquiry which is to look into the possibility of opening up the market for conversion from rental units to owner-occupied ones. The inquiry is to be finalised in spring 2014.

[14] The conversions have often taken place on very beneficial terms for the tenants that accept them, as the market value of the newly created tenant-owner's apartments usually by far exceeds the initial investment required.

[15] http://www.affarsvarlden.se/hem/bostad/article3618032.ece

[16] http://www.di.se/#!/artiklar/2012/12/21/makropanelen-kallare-bostadsmarknad-2013/

[17] The repo rate was left unchanged by the Executive Board in February 2013.

[18] The growth of debt is linked to the estimated rate of return on investment. As housing is immobile, housing investments are normally discounted using the relevant rate of inflation. Borrowers deflate nominal interest rates with national price indices. When interest rates drop and housing prices rise as has been the case in Sweden, the real cost of borrowing decreases, hence making it reasonable for households to take on more debt and hence spur further price increases.

[19] Since many years, almost the entire population growth in Sweden, amounting to roughly 70,000 persons per year, takes place in metropolitan areas and university cities. Metropolitan Stockholm accounts for roughly half of the population growth of the entire country (Swedish National Board of Housing, Building and Planning, 2012).

[20] General elections are planned for 14 September 2014.

[21] The Swedish Association of Public Housing Companies has recently taken an interesting initiative to spur the construction of new rental units at competitive prices. By launching a call for tenders for a framework contract giving their member companies the chance to construct new rental dwellings with an upper price ceiling, the traditional market players have been challenged to come up with innovative solutions. With the prospect of being able to replicate identical buildings in numerous municipalities across the country, construction companies may rely on economies of scale to ensure a satisfactory level of profitability, although each new building offers rental housing at prices approximately 25% below the average of other newly constructed rentals. The first such building was inaugurated in January 2013 and several other projects are under way.

[22] No data for 2010 on rent of newly built apartments.

[23] According to Swedish law, tenants having a direct lease are entitled to continue to rent the apartment for as long as the rent is paid and neighbours are not disturbed, etc. This is known as ‘security of tenure’.

[24] Deregulation of the rental market in the Czech Republic which took place in 1992-2013 led in its second stage to an equalisation of market and regulated rents, with the former decreasing by about 15% in large cities. 

[25] A very limited surcharge has been allowed for instance in cases of furnished apartments etc.

[26] This development was also described in the 2012 In-depth review for Sweden (European Commission, 2012(c)).

[27] As mentioned earlier on, the Exectuive Board decided to hold the repo rate unchanged in February 2013, although two Deputy  Governors entered reservations against the decision, arguing for further cuts.

[28] In particular, the Governor Stefan Ingves has warned against further built-up of mortgage debt in a situation of extremely low interest rates. On the other hand, Deputy Governor Lars E O Svensson argued that monetary policy has little effect on household debt in the short run and (with low and stable inflation) no effect in the long run. Hence, an expansionary monetary policy would not tangibly increase the risks from household debt (see minutes of the Executive Board's meetings at http://www.riksbank.se/en/Press-and-published/Minutes-of-the-Executive-Boards-monetary-policy-meetings/).

[29] For the concept of notional leverage, see Cuerpo Drumond, Lendvai, Pontuch and Raciborski (2013).

[30] http://www.swedishbankers.se/web/bfmm.nsf/lupGraphics/PM%20Principer%20för%20bolån%20slutlig.pdf/$file/PM%20Principer%20för%20bolån%20slutlig.pdf

[31] In January 2013, the Governor of the Riksbank Stefan Ingves stated that it would be preferable if banks could take care of this issue via self-regulation but that unless banks are willing to take their responsibilities, legal measures should be foreseen. The Ministry of Finance reacted by refuting this proposal while at the same time confirming the government's intention to come up with proposals strengthening the willingness of households to amortise in spring 2013 (http://www.dn.se/ekonomi/regeringen-inget-lagkrav-pa-amortering)

[32] SEB, Svenska Handelsbanken, Nordea and Swedbank.

[33] http://www.dn.se/ekonomi/fyra-av-fem-banker-sagar-ingves-forslag

[34] The property tax was lowered from 1.2% of the cadastral value (the cadastral value amounts to 75% of the market value) to the lower of either 0.75% of the cadastral value or 6512 SEK (or roughly 700 EUR, a very low ceiling that would apply to the vast majority of houses), which drastically reduced the taxation of housing.

[35] A minor reduction of the tax on apartment buildings tends to indirectly increase the debt bias for tenant-owned apartments, as the tax balancing the interest deduction is reduced.

[36] http://sverigesradio.se/sida/artikel.aspx?programid=83&artikel=5384141 and http://www.regeringen.se/content/1/c6/20/72/49/895b99ad.pdf.

[37] Approved by the Finnish Parliament in the first reading on 7 February 2013: http://web.eduskunta.fi/Resource.phx/pubman/templates/22.htx?id=5519.

[38] http://www.regeringen.se/sb/d/17003/a/208537

[39] Consolidated data are more appropriate in the Swedish case due to a wide use of intra-group lending and also due to statistical issues (a statistical revision in 2004 and implausible data for inter-company lending within Sweden since 2006).

[40] For the concept of notional leverage, see Cuerpo, Drumond, Lendvai, Pontuch and Raciborski (2013).

[41] Volatile assets include financial derivatives, shares and other accounts receivable/payable (notably trade credits).

[42] Since debts are incurred by Swedish affiliates of multinational groups to be reallocated within the group, the operating surplus of the entire group would be more appropriate as a benchmark. The indicator thus provides a biased picture of Swedish corporate sector's indebtedness.

[43] The composite indicator points at certain risks of deleveraging pressures stemming from the part of capacity to repay, but the pertinence of this indicator is questionable due to the impact of the inter-company loans.

[44] However, given a statistical revision in 2004, consolidated data should rather be used to assess the debt dynamics. These point at a larger accumulation of debt by 41 pps, which is still lower than in other countries.

[45] Cecchetti et al.(2011) estimates that corporate debt above 90% of GDP is detrimental to growth. 

[46] Arcand et al. (2012) conclude that the marginal effect of financial depth on output becomes negative if credit to the private sector reaches 100% of GDP.

[47] Indeed, data show that the average interest rate paid by foreign-owned subsidiaries in Sweden on their loans from their foreign owners is one percentage point above the average interest rate paid on loans from Swedish subsidiaries to  foreign parents.  Interest payments from Sweden are also systematically higher than interest payments flowing to Sweden from abroad. Economic commentary, Riksbank, 2012/3.

[48] According to Swedish Tax Agency, Most FDI has originated from Luxembourg (low corporate taxes) and Belgium (where taxation can be minimised through using the "notional interest deduction").

[49] However, as was the case before, interest payments are tax deductible if a) the creditor is taxed on the interest at a rate of at least 10% should the interest income be the only income for the creditor (10% tax test), or b) the debt has been undertaken for "sound commercial reasons". 

[50] QUEST is a new-Keynesian dynamic  stochastic general equilibrium  model developed by the European Commission (see http://ec.europa.eu/economy_finance/research/macroeconomic_models_en.htm)

[51] The original entry into force of the new requirements was planned for January 2013 but has been delayed due to negotiations at EU level about the final shape of the Capital Requirements Directive IV. 

53 See the Business Tendency Survey by the National Institute of Economic Research.

[53] For more details on the social effects of property taxation see Chapter "Taxation in the context of the Europe 2020 strategy on employment and poverty" in European Commission 2012 (b).