“There may be times when an adjustment in monetary policy may be appropriate to ameliorate emerging risks to financial stability.” Janet Yellen, July 2, 2014
In June 2015, a committee of Federal Reserve Bank Presidents conducted a “macroprudential tabletop exercise”—a kind of wargame—to determine what tools to use should risks to financial stability arise in an environment when growth and inflation are stable. The conventional wisdom—widely supported in policy pronouncements and in a range of academic studies—is that the appropriate tools are prudential (capital and liquidity requirements, stress tests, margin requirements, supervisory guidance and the like). Yet, in the exercise, the policymakers found these tools more unwieldy and less effective than anticipated. As a result, “monetary policy came more quickly to the fore as a financial stability tool than might have been thought.”
This naturally leads us to ask whether there are circumstances when central bankers should employ monetary policy tools to address financial stability concerns. Are there times when macroprudential tools just don’t do the job and policymakers might need a tool that “gets in all of the cracks” (as former Governor Jeremy Stein famously suggested)? We addressed this issue briefly back in 2014 when Chair Yellen argued that the hurdle for using monetary policy to secure financial stability should be very high (see here). Our concern, evident in the title we chose at the time (“Never Say Never”), was that policymakers ought to be cautious about overcommiting in light of our limited knowledge about macroprudential tools (and the special complexities of applying them in the U.S. financial landscape).
How has our thinking evolved? We continue to believe that macroprudential tools ought to be the first line of defense against financial instability. Indeed, work by a number of researchers has strengthened the case against using monetary policy for this purpose. Yet, not all experts share this view. For example, BIS economists remain advocates of using monetary policy to contain credit growth that may appear excessive. Most recently, a thoughtful analysis by Gourio, Kashyap, and Sim highlights possible benefits from using monetary policy to secure stability.
Making the case for or against use of monetary policy to secure financial stability is usually based on assessing the costs and benefits of a policy that “leans against the wind” (LAW) of financial imbalances (as measured against some benchmark). For example, when either the level of credit or its growth rate exceeds a perceived norm, to counter the risks that might arise one might set the policy interest rate above that needed to stabilize prices and output. The idea behind LAW, buttressed by historical analysis, is that credit-fueled financial bubbles (especially in house prices) eventually lead to disruptions and prolonged recessions when failed credits lead to the impairment of leveraged intermediaries.
Analysts on both sides of the LAW debate generally stipulate that financial crises are extremely costly, so that avoiding them is a key objective of central banks. The disagreements arise from questions about the effectiveness of different tools and the costs of using them.
Let’s start with the case against. Our friend Lars Svensson is the current leading advocate of this view. He maintains that the costs of employing monetary policy to achieve financial stability vastly exceed the benefits. The rationale for his conclusion comes directly from the view that raising the policy rate to counter the credit expansion: (1) has limited impact on credit and (by extension) on the probability of a financial crisis; and (2) results in a persistent welfare loss from failing to meet the traditional objectives of monetary policy (keep inflation at target and unemployment at the natural rate that would prevail in a world of flexible prices).
To illustrate Svensson’s conclusion, consider the consequences of a one-percentage-point increase in the policy rate above the level needed to achieve price and economic stability. Using a Riksbank model, Svensson measures the cost of this LAW strategy as a 0.5-percentage-point increase of the steady-state unemployment rate. And, combining Riksbank measures with the unconditional probability of a crisis of 4 percent estimated by Schularick and Taylor, he calibrates the benefit as the product of a 0.02 percent reduction of the probability of a crisis by a 0.028 percentage-point decline in the crisis unemployment rate. That is, the cost (0.5) is nearly 900 times larger than the benefit (0.02 × 0.028 = 0.00056)!
Calculations like this lead to the strong conclusion that, even if the reduction of the crisis intensity were 100 times larger, the minute impact of monetary policy on reducing the probability of a crisis means that the benefits are overwhelmed by the cost.
Utilizing a similar methodology, a 2015 IMF study illustrates how the attractiveness of the LAW strategy depends on its impact on the probability and severity of a crisis. In scenario 1 (see table below), the reduction in the probability of a crisis—0.02 percent—is (according to the IMF staff) the average reported response in the literature to a 100-basis-point rise in the policy rate to address a credit expansion (see, for example, Ajello et al). Continuing with scenario 1, the crisis lasts 4.5 to 5 years and the unemployment rate climbs 5 percentage points above the natural rate. Using these inputs, the IMF estimates that the costs of a LAW strategy exceed the benefits by a factor greater than 30. In contrast, scenario 2 assumes the crisis-reducing impact of a 100-basis-point policy tightening to be 0.3 percentage point, the maximum that the IMF staff could find in the literature. Scenario 2 also assumes that when the crisis comes, it is far more severe. In this extreme case—where the LAW strategy substantially reduces the probability of an outright depression—the benefits do outweigh the costs by a factor of roughly 18 percent.
Benefits and Costs of a LAW policy strategy: Illustrative scenarios
Unsurprisingly, the IMF concludes that “the case for leaning against the wind is limited” but leaves the door open should other transmission channels for monetary policy boost the strategy’s perceived benefits.
In a very recent paper, Gourio, Kashyap, and Sim open the door a bit further. Using a fairly traditional setup, they compare conventional monetary policy rules (such as a Taylor rule that responds to inflation and output gaps) to a LAW policy rule (that responds to inflation and “inefficient credit”). In their setting, LAW is favored when credit excesses (over-reliance on debt finance) are sufficiently volatile, crises are sufficiently severe, and the probability of a crisis is sufficiently sensitive to credit booms. They emphasize that the key elasticities—including the response of the crisis probability to a policy tightening—are difficult to measure, warranting caution in drawing any firm conclusions about LAW. At the same time, they highlight the tradeoff between stabilizing credit (to limit crises) and stabilizing inflation and growth.
From our perspective, the primary concern about forswearing the use of monetary policy for financial stability arises from the challenges of using macroprudential tools. This is especially so in the United States, where the prevalence of regulation by legal form, the multiplicity of regulators with overlapping jurisdictions, and the existence of a range of regulatory loopholes make it impossible to apply regulation evenly across the financial system. The prevalence of regulation by legal form rather than economic function ultimately shifts risk from those entities facing strict scrutiny (usually banks) to others.
No less important, an increasingly powerful legal and legislative assault is taking shape in the United States. A successful court challenge this year cast doubt on the Financial Stability Oversight Council’s (FSOC) authority to designate nonbanks like MetLife as systemically important financial intermediaries (SIFIs) despite strong evidence of systemic risk-taking (see here). In the House of Representatives, the majority-sponsored CHOICE Act aims to eliminate that authority entirely. It also would repeal the FDIC’s Orderly Liquidation Authority to facilitate the resolution of a failed SIFI. Even the most effective macroprudential tool—stress testing—faces majority-sponsored legislation that, by requiring greater transparency, would reduce its impact (see the FORM Act—H.R. 3189, Sec. 5 (a) (1) (I)). Rather than stipulate that the cost of a financial crisis is high, the FORM Act also would require a new and complex cost-benefit analysis for any new Fed regulation, reducing regulatory credibility by enhancing the potential for future court challenges (see H.R. 3189, Sec. 8).
Even if these attacks fail, there exists no consensus for applying macroprudential tools in an active way to counter an evolving threat to financial stability. To do so, as FRBNY President Dudley has emphasized, policymakers would need a reliable framework for identifying emerging imbalances, selecting and then calibrating the most effective tools. All that requires detailed knowledge not only of the pathologies of systemic risks as they arise, but of each tool’s transmission mechanism as well. Even when a policy response occurs, assessing its effectiveness ex post would require the ability to predict how the financial system and economy would have evolved under the counterfactual assumption of no policy action.
This brings us back to the “wargame” that we highlighted at the start. The Reserve Bank Presidents engaged in the tabletop exercise knew the hypothetical problem—principally a boom in commercial real estate (CRE). They also knew the tools that were available—including leverage ratio requirements, special capital requirements for CRE exposures, loan-to-value requirements, supervisory guidance, and tailored stress tests. Even so, according to President Dudley, “there was no agreement as to what instruments should be emphasized and the ordering in which they should be used relative to monetary policy.”
It’s necessary and healthy for policymakers to acknowledge their limitations—especially when, as in the United States, they result from unavoidable institutional constraints. The fact that we cannot depend on the discretionary actions of central bankers or supervisors to counter emerging financial stability threats is one of the most powerful arguments for making the financial system far more resilient. But, with regard to the use of monetary policy to secure financial stability, the challenges that policymakers still face with macroprudential tools leave us close to our position back in 2014: never say never.