This is an English translation of an Ekonomistas post published on June 16, 2014.
During the week of June 2, 2014, Finansinspektionen (FI, the Swedish Financial Supervisory Authority) presented its report on financial stability, Stability in the Financial System, which is objective and balanced. During the same week, an IMF mission, as part of the regular Article IV consultation with member states, presented the mission’s Concluding Statement on the Swedish economy and economic policy. That statement is unfortunately partial and biased. The contrast between FI’s report and the mission’s statement could hardly be greater.
FI’s stability report, published June 2, presents the FI’s view of the stability of the financial system. Regarding household debt, FI writes (pp. 6-7):
In general, Sweden is a country with a high savings rate, as evidenced in, for instance, the fact that Sweden has had a current account surplus for many years. At the same time, household indebtedness is high. High levels of mortgage lending increase the risk to household finances because of increased sensitivity to a drop in house prices, interest rate hikes and unemployment. At the same time, households have substantial assets and sound ability to pay, as shown in e.g. FI’s mortgage survey. FI’s stress tests show that most households can cope with substantial interest rate hikes and high unemployment, even in combination with a drop in house prices. FI therefore believes that household indebtedness primarily poses risks to the real economy, rather than major credit losses on the banks’ mortgages.
The risk to the real economy is mainly about indebted households, as a consequence of e.g. a drop in house prices or higher interest rates, potentially cutting back on consumption. This would have negative consequences for growth. In order to reduce the risks that household indebtedness poses to society, FI can improve the resilience of the financial system on the one hand, and influence the supply of and demand for loans among households on the other. The overarching measures taken by FI to strengthen the resilience of the banking system in terms of capital and liquidity are of great importance in this context. This spring, FI also proposed increased risk weights for mortgages to 25 per cent, in order to address systemic risk in mortgages in particular. In order to curb the risks associated with the growth in household indebtedness, FI introduced a mortgage cap in 2010. In 2013 FI presented a proposal to strengthen amortisation culture by presenting amortisation plans for all mortgage customers, which has now been implemented by the banks.
FI finds that the various measures taken to reduce the risks associated with household indebtedness are currently sufficient. However, it is important to follow how the level of lending to households develops. FI’s annual mortgage survey will be an important part of the future follow-up. If the situation clearly deteriorates, further measures may be required. Should such a situation arise, FI finds it most effective to address the ability and willingness of households to assume debt, such as by means of amortisation requirements, tightening the mortgage cap or by introducing Loan-to-Income ratio or Debt-service-to-Income ratio restrictions. At the same time, measures aimed at increasing the supply of homes, and changes to taxation would be more effective in curbing the risks associated with household indebtedness in the long run. However, any new measures must be introduced carefully, and one step at a time, in order for their effect to be measured and so as not to adversely affect the recovery of the economy.
Regarding the risks of low inflation, FI writes (p. 12, italics added):
Low capacity utilisation and the weak recovery have led to gradually lower inflation in e.g. the euro area and Sweden, which has also started to affect inflation expectations. A lower-than-expected inflation rate contributes to pushing up the real debt burden, i.e. debt in relation to the general price level. This can add to the build-up of financial risks and make it more difficult for households, corporations, governments and countries to adapt their balance sheets. If inflation is negative for a long period of time, deflation occurs, in which case a further increasing debt burden and expectations about falling prices can lead to sagging demand and hence lower prices. As shown by what has happened in Japan since the 1990s, it can be hard to break such a circle.
The IMF mission
The IMF mission that visited Sweden in June is part of the so-called Article IV IMF’s regular consultations with member countries. A full report will be published after the summer. On June 13, 2014, a first report, the Concluding Statement of the mission, was presented.
The mission argues that Swedish inflation is low because global inflation is low. But it misses that, as Deputy Governor Martin Flodén has emphasized in a speech, with a floating exchange rate and appropriate monetary policy, Swedish inflation becomes independent of global inflation. The low inflation is thus mainly due to tight monetary policy and the associated strength of the krona.
The mission states that housing prices are increasing. But it misses that any risks will depend on the reasons why the prices are rising, especially whether prices rise due to other than fundamental factors. The mission also misses that housing prices on average are still lower relative to disposable income than they were in 2007 (see this updated post and this updated figure).
The mission states that household debt is growing rapidly, but it misses that the nominal growth rate is actually quite normal (about 5 percent per year), whereas the real growth rate is high because of low inflation. The real growth rate is about 5 percent per year, as inflation is approximately zero. With inflation at 2 percent, the real growth rate would only have been about 3 percent per year, with inflation at 2 percent; see this post and this figure.
The mission states that the Riksbank’s new data would show that household debt ratios are high in all income groups and that households in low income groups would be particularly vulnerable to disturbances in the form of income losses, interest-rate increases, housing price falls. But it misses that the Riksbank’s survey hardly shows anything new, and that the Riksbank survey ignores the fact that the share of households with debt is significantly lower in low income groups. It also misses that that the FI’s detailed stress test with individual data shows that the new borrowers, which are more vulnerable than the old borrowers, have good resilience precisely against these disturbances. The mission also misses that the income concept in the data from the source UC exaggerates the debt-to-income ratio, especially for low-income households, since the tax-free allowances (child benefit, student grants, etc.) are not included in the denominator.
The mission argues that additional measures to prevent the increase in household debt are required, in order to “free the Riksbank to pursue its inflation target with less concern for financial stability.” In line with this, it also claims that “absent decisive action to address financial stability concerns, the Riksbank is faced with a dilemma,” a dilemma between too low inflation and too high indebtedness. But the mission misses that FI has already taken a number of extensive and apparently effective measures (as mentioned above in the quotation from FI) and that, as mentioned, the nominal growth of debt is normal but the real growth is high because of low inflation.
The mission even claims that “the current monetary stance appropriately balances price and financial stability risks.” Underlying this is the Riksbank’s view that a lower rate would increase household indebtedness and increase the risk of a future crisis and of even higher unemployment. But the mission completely misses that, according to the Riksbank’s own calculations, the policy rate has only a small and uncertain effect on household indebtedness and therefore an insignificant effect on any risks associated with household debt. The mission also misses that, as stated in the above quotation from the FI, the unexpectedly low inflation has actually increased households’ real debt burden and therefore, if anything, increased any risks.
The mission thus fails to note that the high costs of current monetary policy, in the form of too low inflation and too high unemployment, is not matched by any noticeable benefit in the form of lower risks of crises and for financial stability. Contrary to what the mission seems to take for granted, any concern for financial stability is not a real constraint on monetary policy in Sweden. There is simply no dilemma for the Riksbank.
The mission argues that “it is time for a comprehensive set of macroprudential actions that, implemented gradually, would help steer mortgage credit demand towards a sustainable path.” But it misses that there is nothing to suggest that the demand for credit would now follow some unsustainable path.
The mission proposes a number of measures to be taken, but without being able to demonstrate that they are needed. It suggests that the mortgage loan-to-value cap of 85 percent should be reduced to 75 percent. But it misses that the average LTV ratio for new mortgages is only about 70 percent. This means that there is a significant average downpayment and a substantial average buffer against housing price falls. The mission warns that borrowers can avoid the mortgage cap with unsecured loans, but it fails to notice that unsecured loans only make up just over one percent of the total volume of new loans. The mission warns that borrowers would amortize too little. But it misses that household saving is historically high and that amortization is saving that is illiquid and tied up (Swedish), which is not necessarily the best form of saving for each individual household. A liquidity buffer and/or investments in different financial assets may be better options. The mission proposes limited deductibility and increased housing taxation as a means of limiting the increase in household debts. But it misses that increased deductibility and lower property taxation only results in a one-time increase in the level of housing prices and new debt and do not cause a permanent increase in the growth rate of housing prices and debt. The mission also proposes a cap on the debt-service-to-income ratio, but it misses that there is no current need for that, since the FI’s mortgage market report shows that new borrowers’ repayment capacity and resilience to disturbances are good.
Incidentally, the statements of IMF missions to various countries are far from consistent. One can compare this statement with the IMF mission’s very different Concluding Statement about the UK. It mentions the same risks with high debt in the household sector and rising house prices, but concludes that this is primarily a problem for macro-prudential policy and that Bank of England therefore is absolutely right in continuing its very expansionary monetary policy (although the UK is one of the few countries that do not have a problem with low inflation). See here for that statement.
Thus, the contrast between the FI’s report and the IMF mission’s statement could hardly be greater.