Since the global financial crisis, a debate has emerged about the appropriate role of central banks and their associated regulatory authorities in financial stability and resiliency. I believe that there are two basic views of the responsibilities: to act as a fire extinguisher or as a smoke detector.
The classic role of a central bank is to act as a fire extinguisher, focusing on its ability to act as a lender of last resort and to create liquidity in times of financial stress. This role emphasizes that a central bank should stand ready to act as the flames of crisis begin to appear. It can then douse them with liquidity to prevent the flames from spreading from one institution or market to another in order to avoid a system-wide conflagration. In this view, the key responsibility is to minimize damage once the shock hits, not to foresee and prevent crisis, a goal that has been perceived as too difficult and potentially politically charged.
In contrast, the smoke detector, or macroprudential, role emphasizes that a central bank has a fundamental responsibility to act early to prevent the tinder from igniting in the first place. Macroprudential policy focuses on proactive monitoring of individual institutions and interconnected markets for signs of froth and fragility. The smoke detector role certainly does not conflict with the more traditional fire extinguisher role but instead significantly expands the mandate and activities of regulators, supervisors, and central banks.
Policymakers in the G20 and many other countries have been seeking a larger smoke detector role for central banks and regulatory authorities. Rather than take the probability of a crisis as given, or as driven primarily by factors beyond policy control, they have increased the responsibility of central banks and regulatory authorities to try to reduce the likelihood of a crisis, not just reduce the costs once the crisis hits. In this essay, I will provide a brief analysis of some of the costs and benefits of macroprudential policy and key challenges ahead.
Objectives and Trade-Offs in Regulatory Reform
Just as with any regulatory reform, we must articulate the objectives in order to assess the costs and benefits of macroprudential policy. The goal of banking and financial development regulation should be to support and enhance sustainable economic growth, consistent with consumer protection that maintains the integrity of the markets. A large body of research suggests that a deep and developed financial system is a driving force behind economic development and growth (see, e.g., the summaries in Levine 2005 and 2012).
The evidence comes from studies of long-term growth and development across countries, as well as from studies looking across U.S. states with changing regulatory regimes (see Kroszner and Strahan 2014). In the U.S., for example, deregulation of restrictions on bank branching within and across states during the 1970s and 1980s is associated with the development of a more competitive and robust banking system and more access to credit for entrepreneurs, which increased the formation of small businesses. The international research suggests that well-developed financial systems can be particularly helpful for those at the lower end of income distribution. Increasing the efficiency of the allocation of capital to the highest return projects and giving the less affluent access to capital that they would not have in a less developed system appear to be the primary mechanisms for driving economic growth.
From a policy perspective we must ask: Could there be a trade-off between higher growth and higher volatility or potential for crises? (See Kroszner and Strahan 2011.) That is, to obtain a higher growth “return” through financial development, is there a cost in terms of greater “risk” in the system? Theoretically, greater financial depth and development could either increase or decrease stability and resiliency. On the one hand, a larger and more developed financial sector could improve risk sharing and diversification and thereby reduce volatility and increase resiliency. On the other, a larger and more developed financial sector could allow greater concentrations of risk and generate interconnections, thereby potentially making the entire system more fragile and vulnerable to shocks. Research using pre-crisis data suggests that in some circumstances a deeper financial system reduces volatility but in others can contribute to it (see, e.g., Kroszner 2007, Kroszner et al 2007, Morgan, Rime, and Strahan 2007, Arcand et al 2012, and Kroszner and Strahan 2014).
The Role of Macroprudential Policy
Macroprudential policy, thus, should focus on the circumstances in which financial activities may contribute to financial fragility and economic volatility. It is crucial then to assess the potential benefits of greater stability and resiliency against potential costs in terms of lower economic growth. A number of recent papers have attempted to outline such a framework (e.g., Kroszner and Strahan 2011, International Monetary Fund 2013, and Bank of England 2014), and policymakers engaged in financial regulatory reform need to consider both forces.
To be able to assess the costs and benefits of macroprudential policy, it is necessary to define as clearly as possible its scope. The IMF (2013, p. 6), consistent with the Financial Stability Board and the Bank for International Settlements, describes macroprudential policy as “the use of primarily prudential tools to limit systemic risk. A central element in this definition is the notion of systemic risk—the risk of disruptions to the provision of financial services that is caused by an impairment of all or parts of the financial system, and can cause serious negative consequences for the real economy.”
This statement reflects a change from the simple fire extinguisher approach “to clean up the mess afterwards” with liquidity provision. Traditionally, it was seen as difficult to predict disruptions and to find precise tools ex ante to prevent them. This greater sensitivity to and focus on financial stability concerns is valuable, but policymakers and market participants should understand that effective macroprudential policy might not be easy. Though expanding the toolkit is valuable, I caution that there are significant challenges to implementing macroprudential supervision and regulation as an effective smoke detector, as well as political risks for the central bank. Excessive faith in macroprudential policy to stop the buildup of risk concentrations and froth in markets could lead to reduced market discipline and forms of moral hazard, so it is important to be realistic about what macroprudential policy can and cannot do. I will briefly describe three broad challenges for macroprudential policy.
Three Broad Challenges for Macroprudential Policy
The first broad challenge concerns data and measurement. What will be the metrics or indicators used to measure the buildup of system-wide risk and heighted vulnerability to a financial crisis that should trigger preemptive macroprudential action? Following the financial and currency crises in emerging markets in the 1980s and 1990s, academics and researchers at institutions such as the IMF and World Bank tried to develop “early warning flags” to better anticipate where and under what circumstances a crisis might occur. This exercise proved to be extremely difficult and did not produce indicators that would reliably “flash yellow” or “flash red” sufficiently far in advance to allow authorities to act to avoid trouble.
Since the most recent financial crisis, the Basel Committee has emphasized rapid credit growth, either in a specific market or in an economy overall, as a warning signal. Other measures include rapid increases in asset prices, unusually compressed risk spreads, increases in leverage, and reductions in credit underwriting standards (e.g., Borio and Lowe 2002, Drehmann et al 2011, IMF 2013, and Stein 2014). These are important indicators that central banks and supervisory authorities should be monitoring. I applaud the continuing efforts to refine these “flags” but agree with the IMF (2013, p. 18) that the systemic risk monitoring framework should be characterized as a “work in progress.”
Distinguishing between ex ante frothiness that will end in tears and the dynamics of an evolving market economy, for example, is not straightforward. Such a distinction requires two assessments: First, what is the threshold for one or a combination of these metrics to flash yellow or flash red? Much valuable research is being undertaken to try to determine appropriate thresholds, but we are far from knowing when an economy has reached a yellow or red zone. Economies in earlier stages of development, for example, may experience more rapid credit growth than more developed economies as part of a balanced “catch-up” growth path as they converge to the more developed economies. As the IMF (2013, p. 17) has pointed out, “…not all credit booms end in a bust, as they may be justified by better fundamentals, and that loan growth can contribute to a healthy financial deepening” (see Dell’Ariccia et al, 2012).
In addition, what are the system-wide consequences if the bubble does burst? (See, e.g., Mishkin 2009.) Clearly, in a market like housing that involves high leverage and that is widely used as collateral in lending, rapid asset price declines can have significant system-wide consequences, as we saw in the most recent global financial crisis. In contrast, the dramatic declines in asset values and colossal mark-to-market losses with the end of the so-called “dotcom bubble” in the early 2000s left little imprint on the macro-economy and did not trigger a banking or financial crisis. Also, new markets and instruments pose particularly vexing problems because, by their very nature, they have short data trails by which risks could be assessed.
Even if a reasonable set of indicators can be developed, there is a second broad challenge for macroprudential policy that is more theoretical in nature. How strong of a foundation can financial economics provide to supervisors and regulators that an asset is overpriced or a risk premium is too low? As Larry Summers (1985) emphasized many years ago, financial economics and markets are extremely good for ensuring that a 24 ounce bottle of ketchup is priced at twice as much as a 12 ounce bottle but not quite as helpful for determining what an ounce of ketchup should be worth. Assumptions about preferences, risk aversion, discount rates, liquidity, etc. are needed, and reasonable people could disagree about these.
Without a straightforward and theoretically grounded way to argue that market pricing is not properly taking a risk into account, a supervisor or regulator is open to the criticism of being arbitrary and attempting to substitute her judgment for those of the market participants who are putting their own money on the line. It can be difficult for the supervisor or regulator to prove the case. Unfortunately, this also opens the way for political judgments and pressures to determine what is and is not considered “arbitrary.”
The third challenge concerns the political-economy dynamic. Will a central bank’s (or regulatory authority’s) independence be challenged if it is actively engaged in macroprudential policymaking? Many of the instruments of macroprudential policy involve increasing the costs of and/or reducing the availability of credit for particular activities or to certain sectors. This can bring the central bank or supervisor into politically charged areas of credit allocation.
Consider the case of housing. The U.S. and many other countries have numerous government programs and policies that encourage home ownership, ranging from reductions in down payments to subsidies, to securitization (in the U.S., for example, through the government sponsored enterprises). If a central bank becomes concerned about frothiness in housing, how easy would it be to adopt policies that reduce loan to value ratios, restrict securitization, raise capital requirements, or otherwise increase the costs of mortgages? The unelected body of the central bank could be portrayed as trying to overrule public policies explicitly adopted by an elected body. This certainly could put the central bank in the political crosshairs. Effective macroprudential policies thus may involve risks for central bank independence and the independence of supervisors.
With a post-crisis mandate for a broader smoke detector, macroprudential role, central banks and associated regulatory authorities need to analyze carefully the costs and benefits of policy actions. Understanding the sources of vulnerabilities in the system is crucial to evaluating where macroprudential policy can be effective. Assuring sufficient capital through counter-cyclical capital buffers, for example, is quite sensible, but determining the precise level of those buffers is quite difficult. Such policies can be quite valuable but are a work in progress. We should avoid a false sense of confidence that these policies can assure stability or that we have sufficient understanding and experience to fully comprehend the trade-offs involved.