Time Consistency: A Primer

“[S]ome useful policy strategies are ‘rule-like’, in that by their forward-looking nature they constrain central banks from systematically engaging in policies with undesirable long-run consequences.” Ben S. Bernanke and Frederic S. Mishkin, “Inflation Targeting: A New Framework for Monetary Policy,” Spring 1997.

The problem of time consistency is one of the most profound in social science. With applications in areas ranging from economic policy to counterterrorism, it arises whenever the effectiveness of a policy today depends on the credibility of the commitment to implement that policy in the future.

For simplicity, we will define a time consistent policy as one where a future policymaker lacks the opportunity or the incentive to renege. Conversely, a policy lacks time consistency when a future policymaker has both the means and the motivation to break the commitment.

In this post, we describe the conceptual origins of time consistency. To emphasize its broad importance, we provide three economic examples—in monetary policy, prudential regulation, and tax policy—where the impact of the idea is especially notable. (For other examples, see here and here.)

Policy as Commitment Strategy. Kydland and Prescott introduced the notion of dynamic inconsistency of sequentially-determined optimal plans in 1977, opening up a rich field of research that eventually earned them the 2004 Nobel Prize in Economic Sciences. They showed that policymakers operating with complete discretion at each moment in time might not obtain the best possible long-term outcome. Rather, in important circumstances, it is possible to improve outcomes by limiting discretion.

This startling conclusion is a result of the fact that, rather than playing a “game against nature,” economic and social policy influences the behavior of thoughtful people. Importantly, what people do today depends on their expectations about the future—including their expectations of future policy. Since behavior today depends on the ability of a policymaker to commit credibly to a future course of action, there are important circumstances where long-run economic outcomes are worse if policymaker has the option to make period-by-period choices without constraint.

Following this line of thinking, an effective policy must be a strategy for the future, including a commitment that conditions expectations and behavior today. A strategy has to be more than a mere “promise,” which will lack credibility unless there is some mechanism to ensure policymakers make good on their word. To make a commitment credible, we need institutions and practices—a policy framework—that make it costly for future policymakers to renege. The more effective the framework, the more likely that policy will prove effective.

The irony of time consistency is that limiting the discretion of policymakers can lead to better outcomes. As a simple example, consider the problem of kidnapping. In any particular instance, paying a large ransom may result in the release of hostages. However, bigger payments can lead to more frequent abductions. Over the long run, it may be useful to limit the discretion of hostage negotiators (as well as to impose harsh penalties on the crime).

Of course, the notions of strategic commitment and constrained discretion pre-date Kydland and Prescott. Nearly three thousand years ago, Homer provided a compelling example. According to his tale of the Greeks’ return from Troy, to avoid the deadly lure of the sirens’ song as his ship sailed by, the hero Odysseus famously compelled his sailors to put wax in their ears. And to prevent Odysseus himself from jumping ship, the sailors bound him to the mast until they had safely passed. (For that reason, Federal Reserve Bank of Chicago President Charles Evans has labeled monetary policy forward guidance accompanied by a commitment to deviate from usual norms as “Odyssean.”)

In the world of economics, the idea of time consistency is most closely linked to the age-old debate over rules versus discretion. As Taylor notes, economists since Adam Smith and David Ricardo have proposed rules for monetary policy. But the work on time consistency established an important intellectual basis for advocating rules, while emphasizing the importance of a broad framework for constraining discretion in a way that makes policy commitments credible.

We now turn to three examples that illustrate the lessons of Kydland and Prescott’s pioneering analysis: inflation policy, regulation of too-big-to-fail financial intermediaries, and capital taxation.  

Monetary policy. The time consistency problem is perhaps nowhere more widely acknowledged than in the case of monetary policy. That is why modern central banks constantly discuss strategy and commitment, and work to design a framework for decision-making and communications that ensures credibility.

The problem is simple. Suppose that the mere promise of price stability persuades price- and wage-setters to set steady prices and wages. In that event, a monetary policymaker with a short horizon has an incentive to ease policy and boost economic activity. Knowing this, however, price- and wage-setters have little reason to believe the initial announcement. Instead, since policymakers’ statements lack credibility, private-sector inflation expectations lead to higher price and wage inflation.

So, how can central bankers make their commitment to price stability credible? One way is to ensure that their horizon is long enough that they will not forsake the long-term goal of low inflation for the short-term benefit of a temporary boom. Over the past quarter century, this compelling logic has led governments in much of the world to delegate monetary policy to independent central banks with legally mandated objectives, overseen by officers with long terms. To enhance policy effectiveness, central bankers themselves developed a regime of inflation targeting based on constrained discretion. At the same time, scholars like Taylor, as well as legislators, have proposed specific policy rules to further limit discretion.

Today, inflation targeting is—de jure or de facto—the prevailing monetary policy regime in countries that produce about two thirds of global GDP. Inflation targeting relies on transparency and communication to raise the cost of reneging on the price stability commitment. Policymakers not only announce a quantitative target for inflation over at least the next several years, but they routinely report publicly on their progress in realizing their objective. And, when they adjust their instruments, central bankers justify their actions in terms of the impact it will have on achieving their commitment.

Indeed, central banks now view transparency as an essential tool for effective policymaking. In contrast to the old central banking world of purposeful concealment to maximize discretion (under the motto “never explain, never excuse”), rule-like behavior (see the opening quote) encourages households, businesses and financial market participants to anticipate central bank actions that are consistent with keeping inflation low and stable.

While the Federal Reserve did not announce a numerical inflation target until 2012, it behaved as if it had an inflation-targeting framework at least since 1990. Over this period, U.S. inflation (measured by the price index of personal consumption expenditures) averaged precisely 2.0 percent (with a standard deviation of 1.0 percent). In contrast, from 1965 to 1989, inflation averaged 5.2 percent (with a standard deviation of 2.6 percent).

Importantly, the Fed’s de facto framework has anchored inflation expectations in the face of temporary price shocks. To see the implications, consider the following chart that shows the real oil price (adjusted for consumer prices) and a survey measure of one-year-ahead inflation expectations. When oil prices spiked in the 1970s, inflation expectations moved accordingly. By contrast, since 1990, inflation expectations show little sensitivity to large swings in the real oil price.

Real Oil Price (Jan 1970=100) and Median One-Year Ahead Inflation Expectations

Sources: FRED and Livingston Survey (Federal Reserve Bank of Philadelphia).

Too big to fail. For several decades, and especially in the aftermath of the financial crisis of 2007-2009, one of the largest problems facing prudential regulators has been that of too big to fail (TBTF). (See for example, Stern and Feldman.) To prevent a financial collapse and another Great Depression, regulators in the United States and elsewhere have repeatedly used public money to bail out the creditors of the largest, most connected, and most complex intermediaries. In addition to popular revulsion, these bailouts have created moral hazard—encouraging financial behemoths to take risks in good times that make the system more vulnerable to crisis.

It is tempting for policymakers merely to pass legislation banning bailouts. The Financial CHOICE Act, approved by the U.S. House last year, takes exactly this tack. However, as we have argued before, this approach is destined to fail. Mere promises lack credibility and will not constrain the risk-taking behavior of TBTF firms. When push comes to shove, in a crisis, policymakers will come under overwhelming pressure to prevent a financial collapse through bailouts. If necessary, future legislators will enact new laws repealing anything in place. This is exactly what the Congress did in September 2008 when they created the Troubled Asset Relief Plan (TARP). Anticipating a bailout should disaster strike again, TBTF firms profit today from systemically risky activities.

In short, TBTF is a problem of time consistency. If, in a future crisis, the perceived alternative is the collapse of the financial system and a depression, policymakers will renege on any “no bailout” promise they could possibly make. Only a regulatory regime that forces financial behemoths to internalize the spillovers of their behavior on to the financial system as a whole will credibly contain the TBTF problem. (See our earlier discussion here.) In our view, such a regime has three components: high capital requirements that force intermediaries to self-insure against losses; stress tests that ensure capital adequacy even under the most severe adverse conditions; and an effective resolution regime that provides for automatic recapitalization of weakened behemoths as well as temporary government provision of resolution funding.

The point is that the only way to prevent a future bailout is to ensure that the alternative is not an economic disaster.

Capital taxation. A basic rule of public finance is that taxing goods that are supplied (or demanded) inelastically minimizes the deadweight losses from taxation. This creates a dangerous temptation for policymakers in search of revenue: once firms have made large investments in plant and equipment, tax their immobile capital stock.

Knowing this, why would firms invest at all? The problem is especially severe for foreign firms who fear becoming easy targets for expropriation. Clearly, a mere promise not to engage in punitive taxation is unlikely to assuage such concerns. Once again, policymakers face a time consistency challenge.

How can national and municipal governments make credible a promise to respect business property rights and to treat all businesses—domestic and foreign—fairly? Since reputation matters, part of the answer is having a long record of honesty and integrity. Steady fiscal behavior across turbulent times—especially during periods when financing is in short supply—earns confidence.

Governments can also act to reduce the risk of future financing strains. One way to do this is simply limit their ability to issue debt—an Odyssean solution followed by most U.S. states. Another is to give debt payment legal priority, motivating other creditors (like workers and pensioners) to support limits on debt. Observing such fiscal rules can make it feasible to borrow in difficult times, and reduce the pressure to impose punitive taxes.

Governments can and do address the special risks to foreign firms through treaty arrangements, including multilateral and bilateral investment regimes that accord “national” treatment to all firms, regardless of their ownership. Reneging on treaty commitments is costly.

At the same time, the free flow of capital across borders makes it difficult for governments to impose credible regimes for taxing mobile forms of capital. For example, in recent decades, some U.S. firms have transferred intellectual property abroad to take advantage of lower tax rates. Similarly, some kept large profits outside the U.S. borders in the hope that the U.S. government would provide a tax holiday (as in 2004) or cut taxes outright (as they did this past December). Put simply, doubts about the credibility of the government’s commitment to its corporate tax regime encourages activities that have no social benefit.

Importantly, doubts about the sustainability of any tax regime limit its effectiveness. Having seen the partisan divide over the recent corporate tax cut, will businesses now move aggressively to expand the capital stock? Or will they anticipate a “repeal and replace” reversal when the U.S. government majority shifts? Leaving aside constitutional obligations, such uncertainties are inevitable because no government has the ability to commit their successors to any particular policy path.

Conclusion. The time consistency problem makes the credibility of policymakers’ commitment paramount. Well-designed policy frameworks constrain discretion without unnecessarily sacrificing flexibility. While there are important examples of success—such as the regime of low and stable inflation that has accompanied the inflation-targeting era—there are also cases where it is difficult, if not impossible, to commit future governments to maintain a strategy (such as a promise not to bail out systemic banks in distress). Ultimately, the path to success is to start by understanding when commitments will be credible, and to design policy frameworks accordingly.