François Villeroy de Galhau: Low interest rates and the implications for financial stability

Speech by Mr François Villeroy de Galhau, Governor of the Bank of France, at the first annual conference of the ESRB, Frankfurt am Main, 23 September 2016.

The views expressed in this speech are those of the speaker and not the view of the BIS.

Accompanying slides can be found on the Bank of France's website: Slides (PDF).

The question of the financial stability implications of the low level of interest rates is a complex one. It is all the more difficult as there is growing confusion in the public debate as to the meaning of "low rates". When people talk about low rates, they often confuse them with central banks' negative policy rates or with the flattening of the yield curve. But these are different concepts. The term "low rates" itself is multifaceted and refers to different realities. Before talking about the consequences on financial stability, I'd like to start with these issues: what do we mean by "low rates" and what are their causes?

A. What do we mean by "low rates"?

Let us start with the most common reference: nominal interest rates [Slide1]. The 2007-08 financial crisis clearly caused a break, with short-term nominal rates plummeting first in the United States and then in the euro area, from above 5% to close to 0% in a short period of time; whereas long-term nominal rates decreased more gradually. These developments reflect the responses of central banks to the crisis: they had to bring down interest rates to stimulate the economy and bring inflation back to a more sustainable path. The Eurosystem has been doing so with a comprehensive monetary policy package: the cut in policy rates has reduced money market interest rates; forward guidance has steered expectations of future short-term rates, thereby flattening the whole yield curve; the asset purchase programme has compressed risk premia, which has propagated across asset classes and maturities via portfolio rebalancing effects.

As regards real interest rates [Slide 2], the trend after the financial crisis is less striking, even though both short- and long-term rates in the US and in the euro area have decreased as well. But more importantly, if we go back a little further in time, real rates have been fluctuating significantly over the period - they have already been negative in the past, and they peaked in the 1980s990s. Nevertheless, which period is an exception is still unclear: is it the current low rate situation or the 1980s990s high rates episode? Anyway, the overall trend since the 1980s is of a decrease in real rates, still more when it comes to long-term rates. This is consistent with estimates of the "natural" or "equilibrium" real rate of interest, which can be defined as the real interest rate consistent with the full employment of the factors of production and stable inflation. Although this concept is a matter of debate, there is broad consensus that the natural rate has fallen to very low levels over the past decades in most advanced countries. Given these fundamental developments, it would be unwise to bet on real interest rates rising to the levels of the 1980s990s any time soon.

B. These developments suggest two tentative conclusions.

First, nominal interest rates are now probably close to a low point, which doesn't imply they will rebound soon. The ECB has cut one of its key policy rates to negative territory; negative rates are a useful part of our toolkit, but there are clearly limits to them. We know there is a lower bound, even if we don't know exactly where it is: somewhere slightly below zero. However, the pace of any rise in nominal interest rates will depend on the pace of inflation getting back to the target, which depends, inter alia, on the accommodative stance of our monetary policy and on the broader economic recovery. The ECB was clear in saying that policy rates would remain at current levels or lower for the necessary period of time. It will be in a position to normalise policy rates only if it keeps them low for as long as it takes to push inflation up. This is the apparent paradox highlighted by Mario Draghi: "low interest rates today will lead to higher rates tomorrow."¹

Second, there are more uncertainties over real interest rates because, in the longer-term, they primarily result from non-monetary factors: they reflect the underlying fundamentals of the economy. The superabundance of savings relative to investment has been famously called the "savings glut": but it may be more accurate - and more promising - to call it the "investment dearth". Anyway, it exerts downward pressure on real rates. For some countries in East Asia and in the euro area, the gap between savings and investment is indeed staggeringly large: in 2015, the current account surplus was around 3% of GDP in the euro area, and up to 8.5% of GDP in Germany and 9.1% in the Netherlands. Higher long-term real interest rates require a structural rebalancing of savings and investment, and clearly fostering investment rather than reducing savings. That means structural reforms, as well as coordinated actions at euro area level to boost investment.

In that context, what kind of economic environment can financial institutions expect? In the short run, interest rates will stay low and the yield curve will remain rather flat. In the longer run, as inflation picks up, nominal rates should in all likelihood rise again, more strongly - or less slowly - than real rates. In addition, while inflation will recover, the yield curve should steepen as markets would expect a further rise in future interest rates. This is crucial since what matters for the profitability of banks is nominal rates, not real ones, and the slope of the yield curve.

C. In the meantime, what are the consequences for financial stability?

There are two concerns today, even if the economic environment is to improve later, as I just said:

- First, lower profitability for financial institutions. Insurers and pension funds with a high level of fixed rate guarantees suffer from a widening gap between the high level of interest rates served on liabilities and the low interest rates at which maturing capital and interest are reinvested. As for banks, the cut in interest rates and flattening of the yield curve are squeezing their net margins. However, [Slide 3] monetary policy has also had offsetting effects: lending volumes have picked up again; banks have booked significant capital gains; the cost of risk has fallen as borrower solvency has improved; the cost of funding including market financing has become cheaper; and our TLTRO-II programme has provided them with access to multi-year secured lines of credit at highly favourable rates. The strong take-up in September's TLTRO shows its firepower, which had been underestimated by many analysts. All in all, ECB estimates indicate a net positive impact of recent monetary policy measures on bank profitability for the period 2014-17. Nevertheless, we need to be vigilant going forward as to the impact of low interest rates on bank profitability.

- Second, low interest rates could also lead to excessive risk-taking. We in the ESRB and ECB are closely monitoring the financial system, its participants and its markets, and we are alert to any sign of widespread imbalances in asset prices - notably in equities or in real estate. We do not observe general destabilising trends at present, but we stand ready to act if needed using dedicated macroprudential tools.

Two responses are needed, from financial institutions as well as from supervisors:

- First, financial institutions have to adjust their business models. The challenge for their profitability is probably less a future and very prolonged period of very low nominal rates, than the present accumulation of low rates, digitalisation and regulation. Each of these three evolutions is well-founded and manageable: but their coexistence creates without doubt a demanding challenge. Many financial institutions, especially French ones, are already adapting. For insurance companies, the priorities are to move from guaranteed-return to unit-linked business models, and to gradually lower the returns on risk-free life insurance savings. As for banks, expanding non-interest-based business operations and increasing diversification is one way to reduce their vulnerability to the contraction of the net interest margin. More generally, improvements in terms of efficiency are necessary when cost-income ratios are high. And obviously, a resilient financial structure is of the essence. Here, markets should better differentiate between the global European picture and some specific cases. The global European picture is of substantially enhanced resilience since the crisis. Since 2012 the CET1 ratio of significant institutions in the euro area has risen from 9% to 13%. But there are still some issues with certain banks - regarding non-performing loans in Italy and Portugal for instance. Now these need to be addressed seriously but are manageable if dealt with in a timely manner, as illustrated by the Irish or Spanish experience. Within a more solid European banking system, we should not fear cross-border mergers which are the logical answer to the "overbanking" situation put forward by Mario Draghi.

- And second, supervisors have to adapt the way they control banks and insurers. In France, several steps have already been taken. As regards insurance companies, the ACPR has been putting supervisory pressure individually on insurers to take account of the current environment. With respect to banks, business model analysis and profitability risk has been made a supervisory priority of the Single Supervisory Mechanism (SSM) in 2016 and is an integral part of the annual Supervisory Review and Evaluation Process (SREP). Furthermore, the ACPR regularly assesses within the SSM credit standards to avoid an excessive weakening of the quality of banks' exposures - so far, they have been adequate. The French macroprudential authority, called the HCSF, also plays its full part. For instance, it is closely monitoring the commercial real estate sector; it has recently announced that it was ready to trigger macroprudential instruments if necessary. Last but not least: as regulators, we obviously have to stabilise the rules - at last, 8 years after Lehman. In doing so, we should avoid overburdening European banks, which have significantly improved the quality and quantity of capital they hold and which are currently faced with the challenge of profitability. In this regard, the G20 and GHOS commitment to finalise Basel III without a significant increase in overall capital requirements is of utmost importance. More capital means more financial stability, but only up to a point; if this implies excessive constraints on banks, less efficient transmission of our active monetary policy, and hence less growth, it would then become counterproductive.

To conclude, in the current context, our goal should not be to kill the pain - supposedly the low nominal rate environment - but rather to kill the disease, which is too low inflation. Our commitment as central bankers is to deliver price stability. Our monetary stimulus supports demand so that inflation returns to its target in the medium run and, in turn, policy rates rise back to higher levels. But monetary policy cannot address the structural imbalances that are at the roots of the overall low level of real interest rates: the present combination of a savings glut and an investment dearth. Other policies, and coordinated ones, have to step in to address this challenge, while policies dedicated to financial stability must maintain constant vigilance.