Skip to main content
On the Record

Monetary policy and financial stability

Donald Kohn

Editor’s note: The following speech was prepared for Tsinghua University-St. Louis Fed Monetary Policy and Financial Stability Conference at PBC School of Finance, Beijing on May 21, 2016. Not delivered.

I’m glad to have this opportunity to contribute to another conference organized by PBC School of Finance at Tsinghua University on critical economic challenges facing policymakers around the world. When I participated at the inaugural conference two years ago, I addressed the organization and use of macroprudential policy to protect financial stability—employing regulatory tools to make the financial system safer and more resilient to systemic threats. This year the topic is monetary policy and financial stability.

Financial stability is the common theme, and in my view it is the right focus at the right place at the right time. The global economy is still suffering from the after effects of the financial instability that marked the Global Financial Crisis of 2007-09; many economies are not back to full employment and inflation is almost everywhere running below the targets set by central banks and governments. While much has been done to strengthen financial systems around the world, we are still learning the lessons of the GFC—what went wrong and how to avoid a repeat.

One aspect of the answer has been the macroprudential policy I spoke about two years ago. Of course, microprudential regulation—on an institution by institution basis–was universal before the GFC. And authorities, including central banks, worked at identifying broader financial risks in the years leading up to the crisis, but the assignment of responsibility for identifying and taking actions to address systemic risks—externalities for the economy associated with financial decisions– was either ambiguous or missing altogether in many jurisdictions, as were the tools to deal with those risks. The development of decision-making bodies for macroprudential policy with tools to address risks identified has been an encouraging and promising response to the GFC.

Importantly for the subject of this conference, however, questions persist about the contribution of monetary policy to the buildup of risks in mid 2000s. In my view that contribution was minor; the conditions leading to the crisis mostly reflected failures by both the public and private sectors to appreciate and manage growing risks, especially in mortgage markets in the US and elsewhere, born out of complacency after years of good growth and low inflation.

Still, the intersection of monetary policy and financial stability merits further research. Among other things, concerns over the possible effects of monetary policy on financial stability have re-emerged since the GFC in the current context of very accommodative monetary policies over prolonged periods that many advanced economy central banks have undertaken to bolster growth. These policies have been in fact designed to encourage borrowing and spending by households and businesses—the opposite of the deleveraging that might make systems safer. They have entailed promises to keep rates very low for long periods to hold back expectations of rate increases and asset purchases that have driven down term premiums on longer term securities.

For China, recent data have also highlighted the potential intersection of monetary policy and financial stability. These data show very strong credit growth in late 2015 and early 2016 reflecting monetary policy efforts to encourage borrowing to support continued solid economic growth. Similar surges in credit in other jurisdictions and at other times have not been consistent with sustaining growth and avoiding financial instability over time.

I understand that China is making a number of difficult economic transitions to increase welfare of its citizens: to a more consumption-based growth model, especially consumption of services; to a more market based financial system backing more efficient and more market based resource allocation in the real economy. The challenges are how to make these transitions without putting financial stability at risk, and how to design a policy system that will be able to sustain financial stability as markets and institutions are deregulated and play an increasing role in the allocation of finance and resources—in the context of the current conference, how much weight to put on using monetary policy for financial stability purposes.

We all have an interest in Chinese success. China is an increasingly key player in the global economy and financial markets. We saw last summer how actions here can resonate around global financial markets, especially when they are not well understood.

Those are the reasons I say the topic is the right focus at the right time and right place. I am not an expert on China. My reflections on monetary and macroprudential policies to protect financial stability are drawn from experience and knowledge of US, UK, and other industrial economies, but I hope they are useful as China decides how to move forward.

To preview my conclusions: Monetary policy can have important effects on financial stability risks, but, for the most part, it is not the right policy to address those risks. I am concerned about burdening monetary policy with too much to do; putting weight on financial stability in monetary policy decisions implies less weight on economic and price stability in the conduct of policy, and that can have substantial costs in terms of economic welfare. Financial stability is a prerequisite for price and economic stability, so we cannot rule out adjusting monetary policy for financial stability purposes under some, hopefully rare, circumstances; but authorities should develop other tools and other decision processes to rely on first—macroprudential policies—and the more fully developed are these alternatives to monetary policy, the less monetary policy itself might need to be used to defend financial stability.

Monetary policy and financial stability

Monetary policy can affect risks to financial stability. Of particular concern in recent years has been the possibility that the prolonged period of very easy policy might be inducing behaviors that raise these risks, or could do so in the future. The very accommodative policies that many central banks have felt compelled to run are designed to discourage saving and encourage households and businesses to bring consumption and investment forward from the future to the present, financed, at least in part, by borrowing—increasing leverage or at least reducing the pace of deleveraging. These policies are intended to offset other forces that temporarily appear to be holding back economic growth—for example tight credit from banking systems repairing balance sheets post crisis, and restrictive fiscal policies from countries worried about long-term debt trajectories.

Low rates may affect the behavior not only of savers and borrowers in the business and household sectors, but also raise risks inside the financial sector. The drop in interest rates raises asset prices, creating capital gains for lenders. But intermediaries can feel market pressures to re-leverage those gains— or not permit them to reduce leverage—in what appears to be a favorable financial environment. The resulting outward shifts in credit supply curves can cause asset prices to overshoot, along with investment in credit-sensitive sectors.

Very low rates over long periods can induce a reach for yield among investors who, who for contractual or behavioral reasons haven’t fully adapted their desired or expected returns to the new low-rate environment. Low short-term rates and upward sloping yield curves encourage increased maturity mismatch—a borrowing short to lend long carry trade—with associated vulnerability to runs and other financial stability risks.

In my view, the period of very low rates has not in fact created the financial stability risks that people worry about—at least not to the extent sometimes claimed– but the potential causality has led some to advocate making financial stability considerations an integral and persistent factor in monetary policy decision-making. In effect, monetary policy would target the financial cycle alongside the business cycle. When financial stability is undermined, so is economic stability, and the effects of instability can be severe, long lasting, and not easily corrected by easier monetary policy, as we have learned in recent years; it is better in this view to act pre-emptively to avoid financial instability.

Moreover, in this view, regulation—microprudential or macroprudential—could well prove inadequate to maintaining financial stability, especially when the risks are in securities markets and not concentrated in the most regulated financial intermediaries. Monetary policy –actual and expected interest rates—“gets in all the cracks” in Jeremy Stein’s memorable phrase; it’s a pervasive instrument from which few financial decisions can escape. Running a bit tighter policy than otherwise would discourage excessive risk taking across a broad swath of the financial markets.

But what this argument sometimes misses, I believe, is that the costs of such a policy may be considerable and the benefits limited. A blunt and pervasive instrument has its disadvantages as well as its advantages. Higher interest rates than otherwise will affect sectors and transactions that do not threaten financial stability; collateral damage may be substantial and unnecessary. In particular, weighting financial stability in monetary policy implies a temporary steering away from, or delaying of the approach to, the primary monetary policy objectives of price and economic stability.

As compared with a policy path that aims to achieve these goals as expeditiously as possible, output and employment will be lower, as will inflation. Persistent undershooting of the inflation goal can undermine the credibility of that goal and expectations that it will be reached. We are seeing in Japan today the great difficulty of lifting inflation to a two percent goal when expectations have adjusted to a long period of very low inflation or deflation. Low inflation expectations contribute to low nominal interest rates and greater odds of hitting the effective lower bound on interest rates in a policy response to a negative economic shock.


Moreover, to be effective, a monetary policy adjustment might need to be large and could risk being ill-timed. Small adjustments in rates may well have little effect on asset price misalignments or high levels of leverage and maturity mismatch that are driven by a wave of optimism and complacency—so large adjustments in rates and deviations from medium-term goals might be necessary. And rate adjustments could come at just the wrong time—weakening the economy when an imbalance is about to correct anyway, deepening the ensuing recession.

So in my view the use of monetary policy for financial stability purposes has a very demanding cost-benefit hurdle to overcome. That’s not to argue that monetary policy should never be adjusted to take account of financial stability risks. In situations in which regulation is likely to prove ineffective, and the odds on an episode of financial instability are rising, a tightening of monetary policy might well pass the cost-benefit test by heading off a crisis that had far more serious costs than those resulting from the temporary steering away from price and economic stability. But we do need to recognize the costs of using monetary policy in this way—of adding this burden to monetary policy making.

From my limited knowledge, I would think that adding financial stability to the objectives of monetary policy would be especially a concern in China at this time. Monetary policy in China already is trying to accomplish a number of things. Of course, its main task is to encourage solid growth in output and employment and low, stable inflation. It appears to be doing this while simultaneously operating within the constraint of damping sharp and potentially disruptive movements in exchange rates when capital controls are, deliberately, only partly effective and gradually being lifted. And it is adapting its policies and tools to encourage the transition to a more market-based allocation of savings and more efficient allocation of resources.

Now, as I understand it, because the transition to market based finance is incomplete, some monetary policy tools work directly to encourage or restrict lending by banks—operating on the price of lending and borrowing. through quantitative restrictions. So, for a time, the mix of monetary policy tools can be adjusted to take account of financial stability as well as growth; in effect the PBOC has more than one tool to accomplish more than one objective.

In this regard, monetary policy in China today has some similarities to monetary and financial policy in the US in the 1960s, 1970s, and into the early 1980s. Through this period, the Federal Reserve changed its interest rate target frequently to balance demand with potential supply and control inflation. But it also supplemented this usual policy tool in a number of ways: It varied reserve requirements for banks to influence their lending; it changed margin requirements on loans to purchase equities with a view to fighting asset price misalignments; it adjusted ceilings on deposit rates (Regulation Q) to influence credit availability to the housing sector; and in the early 1980s, at the direction of the Carter administration, it imposed limits on some types of credit in order to slow borrowing. These types of actions were aimed at reinforcing the effects of interest rate policy, but they also were aimed at particular aspects of the financial cycle, like bank lending, mortgage availability, and equity prices.

However, it found that these tools became less effective and even counterproductive as markets became more complete and avoidance that much easier. Futures and derivatives enabled highly leveraged positions to be built in equity and other markets; nonbank intermediation replaced some types of bank credit; money market funds grew in response to controls on deposits rates; etc. So the Federal Reserve gradually reduced its reliance on price and quantity restrictions and evolved to rely only on its federal funds rate target for monetary policy.

But without these tools to manipulate, and with the success of monetary policy over the 1980s and 1990s and decades of the great moderation, the Federal Reserve perhaps reduced its focus on financial stability risks. Even if it had been more focused on such risks, it didn’t have the right tools to deal with them in the evolving financial system.

An important lesson of the GFC is that the transition to more market-based allocation of savings must be accompanied by consideration of the implications for financial stability and by a new set of structures and tools to preserve that stability—to build resilience against risks. Good design of such a system in China will itself facilitate the transition to more reliance on market prices to allocate credit because it will build in the guard rails as the system is evolving and it will give the authorities greater confidence that stability will not be sacrificed as allocative efficiency is improved. And, importantly for the topic of this conference, it will reduce the pressure for monetary policy to be deflected to protecting financial stability.

Structures to protect financial stability

The key to allowing monetary policy maximum scope to focus on the business cycle is to have well-developed structures in terms of decision-making, tools, and accountability for micro-and macroprudential policy. I am going to focus on macroprudential policy, because that is the area of my activity on the Financial Policy Committee at the Bank of England and the aspect of regulation that had fallen into disuse in advanced economies until the GFC and is now getting renewed attention.

Protecting financial stability efficiently and effectively requires a different focus and set of tools than monetary policy does. Attention is most often centered on tail risks, rather than on the most likely outcomes that are often at the heart of monetary policy decisions. It’s those tail risks—the unlikely event not fully priced into the market—that can have the most severe implications for financial stability. Macroprudential policy action is called for when those tail events are expected to have externalities—ramifications for the whole economy and not just for the parties involved in the transactions. Those ramifications often play out through the response of highly leveraged lenders or borrowers to the unexpected event—or the actions of the counterparties to these lenders or borrowers, as they fear for the safety of the lending or funding they have supplied.

The instruments utilized in macroprudential policy are often the microprudential tools of capital or liquidity requirements for financial institutions, or constraints on terms of lending—like limits on LTV or LTIs for residential real estate loans. The calibration of those tools has a “macroprudential finish” on them to take account of systemic risks.

Having a variety of tools is important to better focus on the source of the risk; the more the remedy can be targeted to the problem, the less likely are unintended consequences and the better the anticipated cost-benefit calculus. So, sectoral capital requirements for say real estate lending should be in the tool kit, as should the ability to set limits on the degree to which the terms of particular types of lending can be eased when the extra supply of credit can threaten stability. Real estate lending is often the force behind episodes of financial instability and the ability to limit risky practices in these credits would seem to be essential. In addition, the US authorities have been concerned about rising leverage and easier covenants in some kinds of business loans, and, judging from commentary like the IMF’s Global Financial Stability Report, some types of business loans in China may be growing as a threat to stability; perhaps an ability to directly constrain risky practices in that area would also be useful.

The macroprudential authorities need processes and procedures for spotting and addressing possible problems outside already highly regulated sectors should new technology or regulatory arbitrage contribute to risk migration. And they should have the willingness and ability to use their tools countercyclically in order to push back the frontier where monetary policy action would be called for to protect financial stability.

Because macroprudential policy requires somewhat different expertise and focus than does monetary policy, I believe it should be carried out in a separate committee from those responsible for microprudential and for monetary policies. Such a committee should be clearly tasked with using its analytic and information gathering powers to identify risks to financial stability and with utilizing the tools it has been given to build resilience in the financial system so that if those risks materialize they would not impair the critical functions of the system to intermediate savings and investment and allow users of that system to manage risk.

The macroprudential authorities will need to take a long view of risks to financial stability, tightening policies when times are good and risks are building and loosening them when the cycle turns and maintaining the supply of credit will contribute to economic stability. Those forward-looking policies will require a good deal of clear communication about their intended effects and their contribution to economic welfare. And their effective implementation is also likely to require a degree of independence from short-term political pressures, especially those that might be brought to bear by lenders and borrowers whose actions are being constrained.

Good macroprudential decision-making will require a major role for the central bank. Central banks bring expertise in financial markets and in the intersection of those markets with the real economy. Their macroeconomic responsibilities give them a broader focus and expertise than the microprudential authorities, which focus on individual institutions.

As I did two years ago, I would suggest that the UK model for implementing macroprudential, microprudential, and monetary policies is quite likely to prove more successful than that prevailing in the US. The US is working with a fragmented regulatory system in which agency responsibilities often overlap or intersect. This system requires considerable coordination across many agencies, each operating with a different focus and mandate. Such cooperation takes considerable time, making many types of countercyclical policies largely impractical.

Although the formation of the Financial Stability Oversight Council in the Dodd-Frank legislation was a step forward, coordination remains difficult and time consuming, and responsibility and accountability for financial stability is not clearly spelled out or aligned with the available tools. And the tool kit itself does not seem adequate; in particular it lacks the authority to limit cyclical deterioration in the terms of residential real estate lending, the source of several episode of financial instability in the past.

In the UK macroprudential policy is made by a committee in the Bank of England. There are separate committees for microprudential and monetary policies, also in the Bank, and these are represented on the macroprudential policy committee through overlapping membership facilitating mutual understanding and collaboration. The committee also includes four external members who are appointed for up to two three year terms. The head of the financial conduct regulator is also automatically a member of the committee and provides another external perspective by being independent to the Bank of England. These external members bring specialized knowledge and perspective from the financial sector and help challenge potential groupthink. The macroprudential committee has a clear financial stability objective, under the Bank’s financial stability mandate, and is held accountable for its actions to further achieving its objective by the UK Parliament. It has a variety of tools at its disposal and has spelled out in public documents how it intends to use them.

To be sure, the UK system, as the US system, is only a few years old, and success at protecting financial stability ultimately must be assessed over decades, not years. But, if the Chinese authorities are looking to foreign experience to guide their own efforts, I see the UK set up as considerably more promising in many dimensions than that of the US.


It is my understanding that China is indeed working on its framework for macroprudential regulation. I applaud this effort. The thrust of my presentation is that its favorable execution will yield payoffs, not only in preserving financial stability, but also in freeing up monetary policy to pursue its goals for growth and price stability consistently and aggressively. The global economy has a stake in its success.


Get daily updates from Brookings