Speech

The Pursuit of Financial Stability and Tasks for Monetary, Regulatory, and Macro-Prudential Policies

March 16, 2018

Good morning ladies and gentlemen. It is a pleasure to be here today.

The focus of my remarks today will be on the role of macro-prudential policies in maintaining financial stability. This is especially germane at this juncture when—as I will argue—monetary policy needs to remain accommodative. But even if inflation were much higher and monetary policy less accommodative, there is still much to be said for a set of tools that can target country specific financial stability concerns, as opposed to monetary policy, which must focus on the region as a whole.

I will draw on the IMF’s bilateral surveillance work in Europe to glean some lessons. In this regard, I will also illustrate the role of a range of other policies – like tax policy, housing finance, and restrictions on land supply – that have a strong bearing on the underlying issues that macro prudential policies typically seek to address.

But first a word on monetary policy, to provide the context for our discussion. We at the Fund are fully supportive of the ECB’s current strong accommodation. Both wages and core inflation are now clearly moving upward. But core inflation remains well below 2 percent.

We have recently looked at the drivers of inflation in the euro area. We find that the Phillips curve is, in fact, alive and well. But our research also shows that inflation in the euro area tends to be sticky and more backward-looking than in the U.S., where long-term inflation expectations play a greater role. This implies that it will take longer for inflation in the euro area to get nearer the ECB’s objective.

Thus, with inflation projected to rise only gradually, we support the ECB's commitment to maintain low policy rates well past the horizon of net asset purchases.

This then leads—as you know—to the question of whether the strongly accommodating monetary policy is causing financial instability. Let me say upfront that we at the Fund do not think so. We see no generalized financial stability concerns at the current juncture.

Why? Overall bank credit growth is still trailing nominal GDP growth in the euro area. We obviously watch continuously for financial stability concerns, looking at a range of indicators—household leverage, housing affordability and corporate sector debt—to assess household and corporate sector vulnerabilities.

To be sure, we find that there are some localized pockets of excesses. For instance, there are two euro area countries where bank lending seems to be over-inflating housing prices. And, in a few countries, corporate debt relative to GDP is also rising fast. But, importantly, these cases are the exception, not the rule. That is, we do not see generalized financial stability concerns at this stage.

Of course, there are many other markets aside from housing and equity, and country-level indicators may be masking localized bubbles.

But there is no doubt that policymakers need to remain vigilant about potential financial sector risks and continue to expand the toolkit. In some cases, they need to move now to counteract isolated pockets of vulnerability. But these are decisions for national authorities—monetary policy settings need to remain focused on euro area-wide inflation.

This brings me to macro prudential policies to reduce systemic risk. Compared to monetary policy, which is only available at the euro area-wide level, macro prudential policies can, in principle closely target risks in specific national markets, thereby contributing to reduce the heterogeneity in financial and business cycles across member states.

This is an area where remarkable progress has been made in just a few years. We have in the European Systemic Risk Board a European institution warning countries and EU institutions if risks are increasing. Moreover, macro prudential authorities are now operational in every member state. And, if deemed necessary, the capital-based macro prudential tools can be “topped-up” by the ECB, which also has macro prudential responsibilities in addition to its micro prudential role.

This being said, our view is that the EU macro prudential framework would benefit from some simplification. Procedures to activate macro prudential instruments are complex, involving many authorities at different levels.

For example, a few countries -- Austria, Belgium, Finland, Luxembourg and the Netherlands -- were alerted by the Risk Board in November 2016 about potential overvaluation in their housing markets and about rising household indebtedness. In response, most of these countries have tightened prudential or borrower-based tools. But in two cases—Belgium and Finland—the activation of capital-based tools took more than six months.

Thus, decision-making processes should be simplified to ensure that national authorities act in a timelier manner. We are developing concrete proposals in this regard.

Let’s move on. As you know, there is a lively debate on the effectiveness of macro prudential measures. Many sceptics argue that they remain untested and that we are in unchartered territory. There is also evidence that macro-prudential measures are subject to leakages—that is, they become less effective—as credit shifts to alternative sources.

What is our experience? Our analysis shows that macro prudential policies targeted toward specific risks work better than instruments that target broad area-wide concerns. But this also means that the toolkit needs to include specific tools—legislated well in advance—so that these can be used when needed.

In view of this, it is somewhat problematic that some euro area countries have not yet legislated borrower-based tools. These are Belgium, Germany, Greece, Italy, Portugal and Spain.

The measures are well-known: borrower-based caps on loan-to-value and debt-to-income ratios. These are best suited to address specific risks for all institutions—domestic banks, foreign branches, nonbank financial institutions—so that the possibility of leakage is low.

Ideally, all countries should legislate borrower-based tools with harmonized definitions. Moreover, macro prudential authorities should be able to tighten these tools for all lending institutions, and they should be applicable to both households and corporates.

Let me comment on both of these two groups of borrowers.

First, let me focus on macro prudential policies to address housing concerns. One of our key findings in this regard that I want to highlight upfront is that the underlying issues fueling housing market booms are typically much wider and cannot be addressed by macro prudential policies on their own.

For instance, supply constraints often play a role in housing cycles, as demographic, urbanization, and income trends outpace construction. Thus, at least in principle, policies to make the supply of housing more elastic could help.

But the case of Spain is cautionary. Here, making more development land available did not contain the property boom.

In these more difficult cases, even more fundamental steps may be needed. They include tax measures that eliminate biases in favor of ownership and debt, as well as measures to deter speculation by investors, such as for example higher transaction taxes for non-primary residences.

This points to the importance of ensuring that macro prudential authorities should be able to coordinate with fiscal and other authorities, not least on tax and zoning restrictions that could be distorting property prices.

Another lesson regarding housing is that macro prudential policies should focus on the resilience of households and banks, rather than on targeting housing prices. Thus, the IMF’s recent research in this regard shows that macro prudential measures usually have a lasting moderating effect on the level of household debt, but only a transitory impact on the level of housing prices.

This is also very much in line with the recent experience of Sweden, where amortization requirements and loan-to-value requirements curbed credit growth but had less of an impact on housing prices.

Finally, on housing, I would also note that this discussion points to the need for national level implementation of at least some macro prudential instruments. This is indeed the current set-up. But some observers have suggested that macro prudential polices are becoming overly fragmented, and need to be consolidated at the central level.

However, we would suggest caution. Thus, there are good reasons for making sure that some controls are local, since they interact in so many ways with various features of the “real economy” such as housing markets, zoning and taxation issues.

And there is also the asymmetric information aspect to consider: local regulators have information about local conditions and complex interactions that the center does not have.

This does of course not mean that the center does not need to play a coordinating role to address inaction biases by local regulators and cross-border spillovers and leakages.

Finally, let me briefly turn to the issue of corporate credit. Tools targeting corporate credit need careful design. In the euro area, only half of corporate loans come from banks. The rest are from nonbank financial sources and from other corporates. Moreover, the corporate sector can access the bond market, and still increase its indebtedness. Thus, there is a considerable potential for leakages that risk rendering tools ineffective.

France is a recent example in this regard. In view of rising corporate debt, the French macro prudential authority is considering tightening the large exposure limit in big French banks for loans to highly indebted large nonfinancial corporations. Such measures will protect the banking sector against corporate defaults, if any.

As with housing, macro prudential tools to target corporate credit need to be supplemented by other measures. For instance, the tax deductibility of interest payments in most corporate income tax systems coupled with no such deductibility for equity financing creates economic distortions and exacerbates leverage. One way to mitigate this debt bias is to provide a deduction for equity costs.

Recent IMF work looked at the effect of the Belgian allowance for corporate equity -- a tax incentive to raise equity finance -- on corporate debt ratios in nonfinancial firms and banks, relative to a control group of similar companies in other countries. It finds that the impact of the Belgium legislation is significant and large: the debt ratio in Belgium is almost 20 percentage points lower than in the control group for nonfinancial firms and almost 14 percentage points lower than in the control group for banks. Of course, while such tax measures can be very effective, they need to be carefully designed to address concerns about revenue cost and potential for tax avoidance.

In some cases, rising corporate indebtedness is accompanied by increasing prices in commercial real estate. While bank loans can fuel such price spirals, like in Ireland before the Global Financial Crisis, tightening loan-to-value or debt-service-to-income ratios for corporate collateralized borrowings from all financial entities may work better. Here too, coordination with fiscal authorities is key, as changes in tax and depreciation rules could spur commercial real estate booms and busts, as the experience in the United States shows in the 1980s.

Let me conclude.

First, in the euro area, a common monetary policy makes macro prudential tools even more important than other jurisdictions. Because of the still significant fragmentation, member states will often be at different stages of economic and financial cycles and the extent of financial excesses will therefore vary across countries. This points to the critical importance of macro prudential policies.

Second, the good news is that the framework that has been set up in the EU is a remarkable achievement. It could benefit from some simplification, but Europe has come a long way. At the same time, all countries have not yet legislated the borrower-based tools with harmonized definitions that are best suited to target specific risks and limit leakage.

Third, our experience at the IMF suggests that the problems that macro prudential policy seeks to address are often caused by other real sector factors, and by “distortions” in other policy areas. Macro prudential policies cannot be a substitute for addressing these underlying problems.