Over the past few years, we have on many occasions warned about the risks inherent in the Dutch housing market. After all, the close interaction between the housing market and the economy make our country highly vulnerable to declining house prices. The sharp price increases seen in recent years and exceedingly risky behaviour on the part of buyers have exacerbated systemic risk in the housing market. House prices have surged by nearly 8% on average in the past three years, significantly outstripping growth in household income. As a result, price/income ratios in the major cities are now higher than at the previous peak. Also, home buyers borrow ever larger amounts in relation to their income, and household indebtedness remains at a very high level.
At the same time, Dutch banks assess the risks associated with their mortgage loans as very low, because they have historically suffered low default losses on these types of loans. Compared with their peers in other European countries, too, Dutch banks hold relatively little capital against their mortgage loan portfolios. Some of these countries recently took measures aimed at shoring up capital requirements for mortgage loans, making the Netherlands stand out in even sharper contrast. While relatively low risk weights reflect the small likelihood of individual defaults, it fails to reflect the heightened systemic risk in the Dutch housing market.
We have been concerned about the combination of increased systemic risk in the housing market and the low risk weights assigned to mortgage loans for some time, and we discussed them in the Financial Stability Committee in early 2019. I also expressed these concerns in the House of Representatives in the spring. Basically, two options are available to address them, which are curbing the risks in the housing market, or increasing the banks’ shock resilience.We are not in a position to take any measures to tackle risks on the housing market, such as imposing stricter borrowing criteria, further reducing mortgage interest tax relief or addressing bottlenecks in the housing supply – these are actions for the government to take. It is our responsibility, however, to ensure that banks are sufficiently capable of withstanding shocks. The banks’ resilience is of fundamental importance to financial stability, as banks are the most systemically important financial institutions. It is the banks that are most exposed to risks in the housing market, more than pension funds or insurance firms.
Our analyses show that risk weightings could increase substantially in an adverse scenario. This would depress the banks’ capital ratios, putting pressure on market participants’ confidence in banks. For this reason, we intend to impose a floor for mortgage portfolio risk weights to improve the banks’ resilience. We feel supported by the European Systemic Risk Board, which in its recent Recommendation pointed out risks in the Dutch housing market and recommended raising buffers. The floor will be dynamic and depend on the loan to value ratios of underlying mortgage loans. If the ratio between a loan amount and the appraisal value of a home is higher, the floor will be higher.
Our estimates indicate that, taken together, the six Dutch banks that use internal models to calculate risk weights must hold an additional EUR 3 billion against their mortgage portfolios. This represents an increase of roughly 30%. The measure will provide banks with a more stable capital position, thereby reducing their likelihood of getting into funding difficulties in times of crisis. So we are effectively handing them a thicker coat to protect them if a chill wind should start blowing through the housing market.
The measure does not target the Dutch housing market, so we expect its impact on this market to be limited. The measure is aimed purely at bolstering the banks’ positions. We do expect it to push up banks’ funding costs slightly, part of which they can pass on through their lending rates. Our estimates indicate that its impact on mortgage interest rates will be marginal – no more than one or two basis points, which is 0.01 or 0.02 percentage points. So home buyers are likely to be barely affected at all.
We will be imposing the measure as part of a procedure set out in Article 458 of the Capital Requirements Regulation (CRR). Under this procedure, we must first consult the European Central Bank, the European Banking Authority, the European Systemic Risk Board and the European Commission before we can implement it. We are also launching the public consultation today. In principle, the measure could become effective in the autumn of 2020 for a two-year period, following successful consultations. So it is not an addition on top of the new capital requirements under the Basel 3.5 accord, which will probably be phased in only from 2022 onwards. We expect risk weight increases for mortgage loans under the accord to be of a similar order of magnitude.
We believe that making banks more robust will not be enough. Banks need thicker coats, but so do households. Due mainly to high mortgage indebtedness, Dutch households are vulnerable to declining house prices. In recent years, the Dutch government has already taken various measures that reduce incentives for debt financing. Examples include mandatory loan repayments as a precondition for eligibility for mortgage interest tax relief, the reduction of the maximum tax rate at which mortgage interest payments can be deducted, and the cap on a house buyer's borrowing capacity relative to the appraisal value of the home. All of these measures were initiated in 2013.
Still, I would like to urge the government not to stop there, and we are not alone. The European Systemic Risk Board recently issued three recommendations to the government, which I wholeheartedly endorse. First of all, a house buyer’s maximum borrowing capacity related to both the property's value and household income remains generous by international comparison. This means mortgage loans end up underwater quickly in the event of a house price correction, resulting in adverse household spending effects. For this reason, I still advocate further reductions in the maximum borrowing capacity relative to the value of a home. I realise this will require supporting policies, to increase the supply of homes in the mid-price segment of the rental market. In addition, the tax relief facility for mortgage interest payments remains generous, even after the planned reduction. With the current low level of interest rates reducing the value of the relief, now is a good time to reduce the relief.
So let me turn to the international environment now. This has deteriorated, with global growth having slowed recently and risks having gone up. In terms of world trade, tensions between the United States and China flare up from time to time, and turbulence can be seen in other parts of the world as well. These rising tensions feed uncertainty and have an impact on confidence, investment decisions and global trade. In Europe, uncertainty is further fuelled by Brexit. With less than two weeks left, three options are still on the table – a further extension, an orderly exit, and a no-deal Brexit. Together, these uncertainties have made growth prospects less favourable compared with a year ago. The ECB estimates that global growth will be around 3% this year, while the IMF – which is due to issue its estimates today – should publish very similar figures.
In addition to growth prospects, the inflation outlook has also worsened. Against this background, monetary policies in the United States and the euro area were recently made more accommodative. However, monetary policy is not the overwhelming factor behind extremely low interest rates. In fact, capital market rates have been declining for several decades, due mainly to structural factors, such as lower potential growth and the ageing population. The point I am trying to make today is that persistently low interest rates and continuing accommodative liquidity conditions are a risky cocktail from the perspective of financial stability.
For example, with interest rates being as low as they are, investors increasingly seek products that still offer some form of return. In particular, institutional investors such as pension funds and insurance firms tend to accept ever higher risks. This is because they are the ones that have difficulty delivering on guarantees issued in the past. After all, returns of 2% to 3% are hard to generate in the current circumstances without taking additional risks. We are also seeing this search for yield among Dutch insurance firms, as their investment portfolios show a greater proportion of lower-rated corporate bonds.
This type of behaviour also has certain effects on financial markets. The search for yield drives up prices of riskier assets, such as equities, high-yield corporate bonds and real estate. Prices sometimes reach levels that are no longer justifiable on economic grounds. For example, price/earnings ratios of US equities are at a high level, and closer to home we are seeing signs of overvaluation in the Dutch housing market. Of course, it is these assets that are vulnerable to price declines. If prices should go down, investors face impairment and potential losses.
The last side-effect I would like to point out is the fact that persistently low interest rates reduce incentives for bringing down debts. After all, rolling over debt is cheap, and so is taking out a new loan. As a result, the combined debts of non-financial firms around the world have gone up from roughly 200% of global GDP in 2008 to around 240% in the past three years. High debt levels pose a risk when unexpected interest rate increases or negative income shocks, or both, occur, making it harder for borrowers to repay their debts. On top of this, there are signs of an increasing proportion of high-risk debt in the combined global mountain of debt. For example, the proportion of BBB-rated bonds on the European corporate bond market has gone up from less than 10% in 2009 to 30% in 2019. Similarly, the leveraged loan market has more than doubled since 2012, and underwriting standards are being increasingly loosened.
So summing up: persistently low interest rates might be desirable from a macro-economic perspective, but they can also lead to financial instability. In addition, instruments designed to safeguard financial stability are not necessarily adequate in preventing risks from accumulating. Most instruments target banks and households, while risks are also emerging among non-banking operators.