Op-ed in Dagens Nyheter, November 9, 2019, English translation: The amortization requirements and other credit tightening by Finansinspektionen (FI, the Swedish Financial Supervisory Authority) have large individual and social costs, but according to research they have no demonstrable social benefits. It is irresponsible to push through a credit tightening that fails a cost-benefit analysis, write Robert Boije, Chief Economist, SBAB; Harry Flam, Professor Emeritus, Institute for International Economic Studies; John Hassler, Professor, Institute for International Economic Studies; and Lars E.O. Svensson, Affiliated Professor, Stockholm School of Economics.
Every economic-policy measure should pass a cost-benefit analysis. It should show that the benefits exceed the costs. The FI’s amortization requirements and the other credit tightening that it has achieved fails such an analysis. The credit tightening has obvious, large costs but no demonstrable benefits.
According to the FI, the benefit is that the amortization requirements force new homebuyers to take out smaller loans, which would reduce the risks of household indebtedness. This is based on FI’s view that highly indebted households may reduce their consumption sharply in an economic downturn and that a high indebtedness thus constitutes an “elevated macroeconomic risk.” FI believes that international experiences from the financial crisis 2008-2009 supports this view.
However, research contradicts this view (see report to the Stockholm Chamber of Commerce and the latest report of the Fiscal Policy Council). The fall in consumption in other countries was not due to indebtedness itself, but to the fact that increased mortgages before the crisis had been used to finance “overconsumption” in relation to household incomes. This was shown, among other things, in an unsustainably low household saving rate (saving/income). When the financial crisis came, this overconsumption could not continue. The decisive research result is that the highly indebted households that had not engaged in mortgage-financed overconsumption did not reduce their consumption more than less-indebted households. Thus, the fall in consumption was due to mortgage-financed overconsumption, not to indebtedness in itself.
However, there are no signs of a large mortgage-financed overconsumption in Sweden. The household saving rate has risen to a historic high, which is incompatible with unsustainable overconsumption. The proportion of durable consumer goods (such as cars and boats) in household consumption has not increased. In a survey, an overwhelming majority of mortgagors respond that they have not used the mortgage for consumption. There is simply no evidence of the macroeconomic risk to which the FI refers and thus no evidence that the amortization requirements would reduce it.
On the contrary, the amortization requirements reduce the resilience of households and increase the risk of deeper recessions. The households’ ability to maintain their consumption in the event of income falls or interest-rate increases does not depend on the debt itself, but on households’ payments in relation to their income and what financial buffers they have. Amortization requirements increase households’ payments and make it difficult for households to build up such buffers. It also takes many years for households to amortize and reduce their loans so that their payments of interest and amortizations will be less than if they had had an interest-only loan. Meanwhile, they have lower resilience. If the FI is concerned about a negative effect on consumption during a recession, the amortization requirements should therefore be removed.
Also, the amortization requirements do not bring any benefit by reducing the risk to the stability of the financial system. This risk is already small according to FI’s own assessments. This is because mortgagors generally have good opportunities to continue to pay their interest and amortization even if interest rates rise or incomes fall. Households also have on average good margins to deal with a fall in housing prices. In addition, the Swedish mortgage firms are deemed to have satisfactory capital buffers should credit losses nevertheless occur.
However, if the credit tightening does not bring any demonstrable benefit, it is clear that it entails large costs for both individuals and society. The amortization requirements together with the mortgage firms’ requirement that households be able to pass an affordability stress test with a high interest rate have significantly has reduced the amount that households can borrow. This has reduced housing demand, led to falling housing prices and new housing construction, especially in Stockholm, and added to the already large housing shortage.
Amortization requirements and other credit tightening unequivocally affect first-time buyers who do not have high income, wealth or help from parents. Calculations show that with the amortization requirements and other credit tightening, only 20 percent of the 25-29-year-olds in Stockholm have sufficient income to get a loan and buy an average studio (one-room apartment). The amortization amounts to a full SEK 6,000 (EUR 570) per month.
Without amortization and other credit tightening, 50 percent would have had sufficient income. Although some banks when they previously granted interest-only loans nevertheless included some amortization in the affordability tests, the amortization requirements and other credit tightening contribute unnecessarily to exclude many young households from the market for owner-occupied housing.
According to information from the lenders, more and more young individuals see their mortgage application rejected. Because of the dysfunctional rental market (due to rent control) many of these individuals are forced to go to the secondary rental market, where they receive much higher housing costs compared to the alternative of buying a home. According to a report from Skandia, the proportion of young people with at least 18 years older co-borrowers, probably a parent, has increased from 15 percent in 2016 to 24 percent today. According to a Novus survey, 2 out of 3 young professional employees in Stockholm today are forced to seek help from their parents for housing purchases. However, not everyone has parents who can help.
Not only the FI but also the Riksbank regularly refer to warnings from international organizations, such as the European Systemic Risk Board (ESRB), the European Commission, and the OECD. However, with blunt indicators, which do not take into account important factors such as a falling trend for long real interest rates, these organizations conclude that Swedish housing would be overvalued by between 20 and 60 percent and warn that a price correction would have major negative consequences for the Swedish economy. On the other hand, two thorough Swedish studies by officials at the Riksbank and the Swedish National Debt Office do not show any overvaluation at all of Swedish housing prices.
In addition, a calculation shows that the so-called user cost of housing (the imputed rent, that this, the sum of real after-tax interest, cost of equity, and operating costs, disregarding any real after-tax capital gains) per month for the average owner-occupied studio in Stockholm is only SEK 2,800 (EUR 270). According to the Swedish Housing Agency, the rent for the corresponding rent-controlled studio one with an 11-year waiting list is approximately SEK 5,300 (EUR 500) per month, while the secondary rent is between SEK 10,000 and 13,000 (EUR 950 to 1250). How then can owner-occupied housing be considered overvalued? Indeed, neither the FI nor the Riksbank seem today to argue that Swedish housing is overvalued. To refer to the international organizations’ warnings because of faulty analyzes of an alleged overvaluation is therefore strange.
In summary, the FI’s credit tightening leads to great damage but not to any demonstrable benefit. It is irresponsible to enforce a credit tightening that clearly fails a cost-benefit analysis.