Commitment Versus Discretion In * BY MICHAEL DOTSEY

hether policymakers should commit to a event comes to pass. The absence of this ability is called discretion. W certain course of action or have the flexibility Under discretion, a policymaker is to approach each situation as it arises allowed to change policy depending on current circumstances and to continues to be a central question in the disregard any past promises. Because design of monetary policy. A seminal article written by the discretionary planner does not make any binding commitments, it two prominent economists in 1977 analyzed the benefits would appear that discretion offers of carrying out plans based on commitment rather than more flexibility and it would seem to be preferable to a policy whereby the discretion. Since then, others have joined the debate. policymaker must honor past promises. In this article, Mike Dotsey elaborates on the merits of The idea that it is better for commitment versus discretion in setting monetary policy. a to follow through on policies promised in the past, rather than being free to respond to conditions as they evolve, is a subtle and perhaps surprising one. Not The debate over whether it is the New York Association for Business only are better long-run outcomes better for a policymaker to commit Economics, Philadelphia Fed President achieved under commitment, but to a particular course of action or to Charles Plosser explained his views monetary policy is also better able to approach each situation with perfect on credibility and commitment in respond to shocks if the central bank flexibility has been and continues to monetary policymaking. This article is constrained to honor past promises be a central question in the design of elaborates and expands on some of concerning its future behavior. As monetary policy. In 1977, economists these ideas. I’ll discuss below, lower inflation, Finn Kydland and Edward Prescott To start with, let me first define with no adverse effects to economic wrote the seminal article analyzing what we mean by commitment activity, is obtained under a policy the benefits of carrying out plans versus discretion. Commitment is the of commitment, and such a policy based on commitment as opposed to ability to deliver on past promises no can achieve less volatility in both discretion. Since then, the benefits of matter what the particular current inflation and output as well. Indeed, commitment have been analyzed in situation is. I should stress that, under the inability to commit often leads to many settings and in many economic commitment, promised behavior problems for policymakers. models. Indeed, in a 2007 speech to is generally contingent on future Comparing policymaking under events. Promises are not typically discretion and under commitment blanket commitments to be fulfilled Mike Dotsey is an analysis of two polar cases. It is a irrespective of future situations. The vice president and sidesteps the question of how a central senior economic key aspect of commitment is that the bank can act in a committed fashion policy advisor in policymaker keeps his promise to act in even if it desires to do so. Also, how the Philadelphia a certain way when a particular future Fed’s Research could a central bank convince the Department. public that it is operating in a manner This article is consistent with commitment when available free of *The views expressed here are those of the charge at www. the institutional setting places little author and do not necessarily represent philadelphiafed. the views of the Bank of restriction on future policies? For org/research-and-data/publications/business- Philadelphia or the Federal Reserve System. instance, the members of the policy- review/. www.philadelphiafed.org Business Review Q4 2008 1 making boards change over time by Kydland and Prescott to illustrate and in areas not subject to flooding. as do the legislators that monitor the benefits of commitment over Thus, the best outcome is for people the behavior of monetary policy. discretion is that of the flood plain. to decide not to build houses in Commitment requires tying the hands Recently, Robert King provided a areas subject to floods and for the of future policymakers, and in reality, detailed description of this example, government to choose not to build we don’t even know who they will be. which highlights the importance of dams. If the government can commit Research analyzing ways that expectations and the role they play to never building a dam, this will be policy can come close to the ideal in economic outcomes.1 The role of the outcome. Everyone will believe of full commitment has generally expectations will also be a central that the government will not build proceeded along two lines. One is aspect in the analysis of monetary a dam and no one wants a flooded institutional design. How does one policy. house. As a result, no one chooses to set up institutions that will improve on discretionary outcomes? The other is the role of reputation and the Economists refer to the desire to alter credibility an institution can achieve previously made plans as the time-consistency by behaving like a committed planner over time. While of tremendous problem because, at each date, an individual interest, investigations into these areas or policymaker finds it tempting to deviate from are beyond the scope of this article. But we cannot hope to understand what an earlier plan dictated. these more advanced investigations without first understanding the In this example, people make a build near the water. The individual’s different nature of policy under single decision: whether they wish to decision about where to build a house commitment and under discretion. live near the water. Unfortunately, is a relatively simple one and does not Economists refer to the desire areas near the water are subject to depend on where other individuals to alter previously made plans flooding. The government can prevent decide to build their houses. If you as the time-consistency problem flooding by building dams, but doing want to avoid flooding, stay away from because, at each date, an individual so is expensive. The government also the water. or policymaker finds it tempting to has a single decision: whether to build Under discretion the government deviate from what an earlier plan a dam. Furthermore, the government cannot commit to not building a dictated. The temptation to alter wants its policies to conform to dam. As King explains, this inability strategies affects how others view your individual preferences. It wants to do complicates the problem considerably. proposed plan, and it is the interaction what makes society as a whole better The government’s decision is now between the public’s expectations off. There is no conflict between what based on how many people live near and the policymaker’s decisions that individuals think is best and what the water. If a sufficient number leads to problems for a policymaker the government thinks is best. The decide to live near the water, it is who cannot commit. Economics has problem is determining what the best now better to build a dam than to many examples of the time-consistency outcome will be, given that people subject many people to floods. Now, problem, and although I will primarily prefer living near the water and the an individual’s decision about where focus on monetary policy, I will start fact that building dams is costly. Of to build is complicated. If he thinks with a simpler setting that lays out the course, the best outcome will depend a lot of people will build houses near basic issues in a fairly transparent way. on how costly dams are relative to the the water, he should too because a pleasures of living near the water. dam will be built, and he will have to THE EXAMPLE OF THE The problem is interesting only if pay his share of the dam’s cost. If he FLOOD PLAIN we assume that, all things considered, anticipates that only a few people may Before we delve into monetary dams are prohibitively expensive, build houses near the water, he should policy, it will be helpful to look at the and therefore, the best outcome is not follow their example because he difference between commitment and for people to live away from water will be subject to the risk of floods. discretion in a simpler setting. One In either instance, if he anticipates correctly, he either lives near the water of the more famous examples used 1 See the article by Robert King.

2 Q4 2008 Business Review www.philadelphiafed.org protected by a dam or he lives in safety going to increase the , this rate is viewed as undesirable. away from the water. If incorrect, he they respond by increasing prices. To A second feature of the economy is lives near the water and his house is be concrete, consider the case where that the central bank and the public periodically flooded, or he pays for individuals anticipate a doubling of desire output to be somewhat greater a dam and lives in a less desirable the money supply. In this case firms than potential.3 The justification for location. respond by doubling their prices and this assumption is that other features If we focus on situations where workers similarly respond by doubling of the economy, such as the lack of everyone behaves in a similar fashion, their wage demands. Workers would perfect competition or the presence there are two potential outcomes. like to be able to purchase the same of distortionary taxes, prevent the Everyone believes that no one else will number of goods for a given number economy from operating efficiently, build near the water; no one does; and of hours worked and firms are willing and to some extent, it is desirable for no dam is built. That is the optimal to pay the higher wages because, the monetary authority to offset these outcome and the one obtained under in the end, they are paying workers features. commitment. If, however, everyone the same amount in terms of goods Under these assumptions the believes that others will build near produced. Thus, a doubling of money central bank can move output above the water, everyone does build near and a doubling of prices and wages potential only if it surprises individuals the water, and a dam is built — a less leaves everyone in the same position by inflating at a rate greater than pi- than desirable outcome. An important as before. Therefore, anticipated star. Under commitment, the central thing to note is how complicated an changes in money affect only prices, bank would inform the public that it individual’s decision-making process and this is a long-run attribute of will keep inflation at pi-star. Knowing is. He must factor in not only what every established model in monetary that the central bank is capable of he believes the government will do economics. honoring its promises, the public will but what everyone else will do as However, unanticipated changes believe the central bank and expect well. It is precisely this feature of how in money do affect output. For the inflation rate to be pi-star. With no expectations affect an individual’s example, if the central bank adopts an surprises, inflation will be pi-star, and decision that leads to the less desirable expansionary policy by unexpectedly output will attain its potential level, results under discretion. I will return increasing the money stock, output which is somewhat below its desired to this aspect of behavior when I expands and inflation increases. Firms level. discuss monetary policy. and workers are both surprised by the The question then is: Can a increase in money and initially do not policymaker who cannot commit to THE LONG-RUN BENEFITS OF demand higher prices or higher wages. achieving an inflation rate of pi-star do COMMITMENT IN MONETARY The increased money stock, which is better? Can that policymaker increase POLICY held by the public, can now be used output enough at the expense of some Now let’s analyze the benefits to purchase more goods and aggregate surprise inflation to make everyone that commitment confers on average demand subsequently increases. As better off? Perhaps surprisingly, the inflation and average output. As firms and workers catch on to what has answer is no. Suppose the public in the example just considered, a happened, prices and wages increase, initially thought that the central bank key ingredient in the analysis is resulting in inflation. Symmetrically, would target inflation at the rate of the forward-looking behavior of unexpectedly tight monetary policy pi-star. Knowing this, the central bank individuals. It is people’s ability to plan lowers inflation and causes output to is now faced with the opportunity to ahead and anticipate the policymaker’s contract. increase the level of output by creating actions that makes outcomes under Furthermore, there is a rate of a bit of surprise inflation through discretion sub-optimal. inflation that everyone prefers, which, expansionary policy. In other words, In particular, we will analyze the for the purposes of this article, need issue using a classical framework in not be specified. I will refer to this rate which prices and wages are perfectly as pi-star.2 Inflation above or below 3 An economy’s potential output is the level that flexible. In such a setting, anticipated would occur in the absence of any economic changes to the money supply have no distortions. Basically, it is the level that would 2 Depending on one’s view of the structure of effect on output. In this environment, obtain if prices were free to vary, markets were the economy, the optimal rate could be slightly competitive, and there were no distortionary if firms believe the central bank is negative, zero, or even perhaps slightly positive. taxes.

www.philadelphiafed.org Business Review Q4 2008 3 a discretionary policymaker has an policy. Some economists have argued policymaker from reacting optimally to incentive to deviate from the policy that the gold standard was such an economic shocks is actually mistaken. that would occur under commitment. arrangement or that currency boards The ability to keep promises allows A small increase in the inflation rate help achieve commitment. Others, a central bank operating under a would not be very costly, and the such as Kenneth Rogoff, have argued policy of commitment to influence benefit would entail more output. for the benefits of appointing central expectations in a way that the Facing this tradeoff, the central bank bankers who have a strong aversion discretionary planner cannot. In a will generally find it desirable to initiate additional inflation. For concreteness, assume it is Under both commitment and discretion, output desirable to generate a 1 percent surprise increase in inflation. In this remains at potential, but commitment achieves case, it would be foolish for the public this result with lower inflation. to expect the inflation rate to be pi-star. They should expect it to be pi-star plus 1 percent. If the central to inflation. Carl Walsh has suggested sense, this gives the policymaker who bank does not revise its strategy, the contracts that penalize central bankers can commit another tool to work with. outcome will be no surprise: Inflation if inflation deviates too much from its In fact, a policy under commitment will be pi-star plus 1 percent, and target. Currently, there is a good deal can achieve all of the outcomes of output will be at its potential level. At of interest in whether explicit forms a policy under discretion and can this stage we can repeat the reasoning of help to achieve also achieve outcomes unobtainable in the previous paragraph. A further the better outcomes associated with under discretion. The committed attempt to surprise the public, with commitment.4 policymaker cannot do worse than the say 0.5 percent additional inflation, discretionary planner. will generate increased output that, in THE RESPONSE TO SHOCKS It is precisely because a the end, may be worthwhile. If that is UNDER COMMITMENT AND policymaker who can commit has the case, the public should anticipate DISCRETION the ability to follow through on an inflation rate of pi-star plus 1.5 The ability to respond to promised actions that he can influence percent. economic shocks, such as oil-price expectations in a desirable way. The Again, all that occurs in the end shocks or changes in productivity, so discretionary planner, because he is more inflation and no additional as to limit their effects on economic makes decisions period by period, output. At some point additional volatility is one of the supreme makes no promises and, as a result, inflation will be too costly, and the challenges confronting central banks. does not have a similar ability to central bank will no longer try to fool It is this aspect of monetary policy influence expectations. A planner the public. The public will expect the that most often elicits arguments who can commit to future actions higher inflation rate, and output will touting the benefits of discretion. It is in various situations can affect what remain at potential. Forward-looking argued by those in favor of discretion people expect will happen in these individuals will not be fooled, but that monetary policymakers must situations, and these expectations under discretion, there is a temptation be allowed a free hand to respond to influence current behavior. By making to try to fool them. The result is each situation as it arises and not be well-designed promises, policymakers just more inflation. So, under both constrained, for example, by promises can influence expectations in ways commitment and discretion, output to keep inflation at some targeted rate. that elicit better economic outcomes. remains at potential, but commitment Discretion is needed to adequately However, along with these promises achieves this result with lower guide the economy through turbulent comes the constraint to honor them. inflation. times. Thus, actions today, which affect the The example above makes clear However, the notion that future, also affect future policy, and the long-run benefits of commitment commitment unduly constrains the this in turn implies that the history and of devising institutional of actions taken will affect current arrangements that prevent the 4 For a survey of inflation targeting and its policy. In this sense, the committed central bank from using discretionary effects, see my 2006 Business Review article. policymaker is not free to base today’s

4 Q4 2008 Business Review www.philadelphiafed.org policy only on current economic feature to note is that higher interest policymaker or a committed conditions. rates reduce aggregate demand and policymaker? But having policy constrained lower output. To answer this question, in this way should not be viewed as The model’s other component is I examine how both types of a negative attribute of commitment. a Phillips curve that relates current policymakers and the economy These constraints, if designed inflation to future expected inflation respond to an aggregate supply shock. appropriately, can actually lead and to the level of output. This is the Figure 1 displays the model economic to better outcomes through their supply part of the model. If future responses to a 1 percent shock to the influence on expectations that allow inflation is expected to be high, inflation rate.7 Because the public does for better economic decisions. These firms will want to raise prices more not like inflation above target and the last points have been persuasively aggressively today so that their prices central bank is trying to maximize illustrated by economists Richard do not get too far out of line with the the public’s welfare, policy responds by Clarida, Jordi Gali, and Mark Gertler behavior of prices in general. This tightening: The central bank raises the and by economist Michael Woodford. leads to greater inflation today. Thus, nominal interest rate (panel a). Note To make these points more higher expected future inflation leads that, under discretion, the interest concretely, I will use a simple to higher inflation today. Also, when rate must be raised by approximately benchmark New Keynesian model the level of output is high, firms’ costs 50 basis points more than under of the economy.5 That model has of production rise, and as a result, commitment. As a result, output two basic components. One is a firms pass on some of these additional declines by more under discretionary description of aggregate demand, costs to consumers. The result is policy (panel b), but the effect of this commonly referred to as an IS higher inflation. The economy will be more aggressive tightening under curve, that essentially relates current in equilibrium when the level of the discretion has less of an effect on output demand to the level of the real interest rate and inflation implies inflation (panel c). Inflation moves real interest rate (the interest rate that output demand is equal to output up more in response to the shock to adjusted for inflation) and to future supply. inflation and declines more slowly. levels of output.6 Basically, high real Importantly, in the model, Policy under commitment experiences interest rates imply lower demand for monetary policy can affect the level a smaller rise in inflation and a more consumption and investment. A high of output. Underpinning this model rapid return of inflation to target, with real interest rate implies a greater of the economy is the feature that less loss of output. Policy also does return to saving and induces people prices and wages are costly to adjust. not need to be as aggressive because to consume less and save more. It also These costs may involve the resources inflation doesn’t rise as much. means that firms must earn a higher used in acquiring information, the How does the committed rate of return on each project in order resources employed in figuring out policymaker achieve the best of both for those projects to be cost effective. exactly what the correct price or wage worlds: less inflation as a result of Thus, only relatively profitable projects is, and the resources needed to change the shock and less loss of output will be undertaken, and consequently, prices. These costs imply that firms while at the same time acting less there will be less investment. and workers will not immediately and aggressively? The answer is that Alternatively, greater future fully react to changes in monetary expectations of future inflation economic activity implies both an policy. As in our previous example, in affect current inflation. Under increase in current consumption which unanticipated changes in policy commitment, individuals take into through a wealth effect and more affected the economy, here anticipated account the policymaker’s promise investment because future economic changes in policy affect the economy to bring inflation down and not prospects appear rosy. The important as well. They do so because it takes exploit the output gains arising from time for the price system to fully inflation. As a result, expectations 5 For a more detailed description of the model respond to changes in policy. Thus, of inflation do not increase as much economy used in this section, see the article by the central bank can move output and under commitment (panel d), implying Richard Clarida, Jordi Gali, and Mark Gertler. inflation around in response to an 6 It is also common to describe the IS and economic shock. Phillips curves in terms of output relative to its 7 In these simulations the monetary authority potential level, which is referred to as an output The question I now address is: places only half as much weight on output gap. Who does it better — a discretionary fluctuations as it does on inflation fluctuations.

www.philadelphiafed.org Business Review Q4 2008 5 FIGURE 1 Economic Responses Under Commitment and Discretion

1.4 (a) interest rates (b) output

-0.1

1.0 discretion discretion -0.3 commitment commitment

0.6

percent -0.5 percent

0.2 -0.7

-0.2 -0.9 0 2 4 6 8 10 12 14 16 0 2 4 6 8 10 12 14 16 quarters quarters (c) inflation (d) expected inflation 1.4 0.8

discretion discretion 0.6 1.0 commitment commitment

0.4 0.6

0.2 percent percent 0.2 0.0

-0.2 -0.2

-0.6 -0.4 0 2 4 6 8 10 12 14 16 0 2 4 6 8 10 12 14 16 quarters quarters

that firms do not raise their current policymaker to achieve superior of inflation, the greater the volatility prices as aggressively as they would economic outcomes in response to of output will be. Symmetrically, the in an environment characterized by economic disturbances.8 policymaker can lower the volatility discretion. The stability of inflation Commitment’s superiority to of output only by accepting more expectations under commitment discretion can be further characterized volatility in inflation. Thus, the implies that policy does not have by investigating what kind of inflation policymaker will have a whole menu to be as aggressive in order to bring and output tradeoffs confront the of attainable combinations of output down inflation, and as a result, output economy under the two different types variability and inflation variability to does not have to decline by as much. of policy. In this model of the economy, choose from. The particular choice Contrary to intuition, the constraint decreasing the variability of inflation will depend on the public’s preferences. of having to abide by past promises can be achieved only by allowing Figure 2 graphs the choices actually allows the committed output to be more variable. If, in order available to each type of policymaker. to combat inflation or disinflation, Because people dislike volatility in the policymaker responds more both output and inflation, points that aggressively to inflationary shocks, lie closer to the origin are preferred. It

8 In this simple model, the committed and output will end up varying more is obvious that under commitment the discretionary policymakers achieve the same because of the more aggressive policy economy can achieve better outcomes outcomes in response to a shock to aggregate response. Therefore, the more the than under discretion because the demand. However, this is not generally true in more sophisticated models. policymaker tries to limit the volatility curve depicting the tradeoff under

6 Q4 2008 Business Review www.philadelphiafed.org effects of the oil shocks on economic FIGURE 2 activity and inflation could be different under these two different Output and Inflation Tradeoffs settings. In actuality, they were quite standard deviation of output gap different. 3.6 The two episodes are the oil- discretion price shock of the late 1970s and a commitment 3.2 more recent oil-price shock in the 10 2.8 early 2000s. By the time the oil- price shock of 1979 hit, more than 2.4 doubling oil prices over the course of 2.0 the year, inflation had already reached 9 percent. These historically high 1.6 inflation rates were caused by overly 1.2 easy monetary policy. It is fair to say the Federal Reserve had, by the time 0.8 of the oil shock, lost credibility for 0.4 maintaining low inflation. The rise in oil prices further ignited inflationary 0.0 0.0 0.4 0.8 1.2 1.6 2.0 2.4 pressures, and without credibility for maintaining low inflation, the Fed standard deviation of inflation was put in the situation of ratifying the higher expected inflation or trying to contain inflation with a large subsequent loss of output. It chose the first option, and by the first quarter of commitment lies everywhere below monetary policy, at least examples in 1980, inflation had increased to more the curve under discretion. This which the Federal Reserve has had than 15 percent. means that for any given level of sufficient credibility that the public In contrast, from the end of variability in inflation, the committed believed that monetary policy would 2003 to the end of 2005 oil prices policymaker can obtain less variability behave in a manner that approximates more than doubled, yet inflation in output than the discretionary commitment. I will contrast two remained contained without any planner. Similarly, for any degree of episodes, both involving oil-price significant adverse effect on output. volatility in output, the committed shocks. The main difference between these planner can generate less volatility Although I cannot give definitive two episodes is the credibility that in inflation. Thus, not only will the proof for the following argument, one the Federal Reserve now enjoys for economy achieve a lower average can view the differential economic maintaining low and stable inflation. rate of inflation under commitment, impact of oil-price shocks in the This credibility is portrayed by it will also experience less volatile 1970s and 2000s through the lens of the stability of various measures of inflation. This depiction along with commitment.9 In one instance, the inflation expectations over this period. the discussion in the previous section Fed lacked credibility for maintaining For example, the 10-year expected highlights the observation made low inflation and in the other the Fed earlier: Under commitment, policy had that credibility. The theoretical can achieve outcomes that cannot be material covered suggests that the achieved under discretion. 10 There are many other documented episodes. Some are discussed in President Plosser’s AN EXAMPLE: speech, and the history of inflation scares is documented in the article by Marvin OIL-PRICE SHOCKS 9 Recent evidence outlined in the article by Goodfriend. Also, for a more detailed analysis There are many examples of the Sylvain Leduc, Keith Sill, and Tom Stark is of appropriate monetary policy in the face of consistent with the interpretation of events shocks to oil prices, see the article by Sylvain benefits of commitment — or, in U.S. described here. Leduc and Keith Sill.

www.philadelphiafed.org Business Review Q4 2008 7 inflation rate in the Philadelphia Fed’s SUMMARY central bank to follow policies consis- Survey of Professional Forecasters This article has explored the tent with the assumption of full com- hardly moves over this period, and benefits of policy under commitment mitment is not easy or straightforward. expected inflation as represented by versus discretion. In particular, it The credibility the Fed has achieved is the difference between the yield on has discussed the added benefits due, in no small part, to the leadership 10-year nominal and indexed Treasury policymakers derive from fulfilling past of the two previous Fed Chairmen, bonds is quite stable. Therefore, as promises. Rather than constraining Paul Volcker and Alan Greenspan. in Figure 1, the more recent oil- policy, adhering to honoring policy The current Chairman, Ben Bernanke, price shock had very little impact on promises enables monetary policy is maintaining their example of com- inflation expectations, and as a result, to attain outcomes that cannot be mitment to low and stable inflation. there has been no need for exceedingly attained by a policy arrived at anew at The benefits of following a commit- aggressive policy. In turn, there has each point in time. Committed policy ted plan are now so entrenched in been very little impact on output. The generates lower long-run inflation policy-making circles that most central current FOMC is committed to low without any adverse effects on banks aggressively strive to maintain and stable inflation and is perceived economic activity and ameliorates the their credibility. The loss of credibility in that light. Acting as a committed effects of economic disturbances. presents grave problems for monetary policymaker has its benefits both in In practice, achieving and main- policymakers, problems that have been theory and in practice. taining the credibility that allows a highlighted in this article. BR

REFERENCES

Clarida, Richard, Jordi Gali, and Mark King, Robert G. “Discretionary Policy and Plosser, Charles I. “Credibility and Gertler. “The Science of Monetary Policy: Multiple Equilibria,” Federal Reserve Bank Commitment,” speech delivered to A New Keynesian Perspective,” Journal of of Richmond Economic Quarterly, 92:1 the New York Association of Business Economic Literature, 37 (December 1999), (Winter 2006), pp. 1-5. Economists, New York, March 6, 2007. pp. 1661-1707. Kydland, Finn E., and Edward C. Prescott. Rogoff, Kenneth. “The Optimal Degree of Dotsey, Michael. “A Review of Inflation “Rules Rather Than Discretion: The Commitment to an Intermediate Monetary Targeting in Developed Countries,” Inconsistency of Optimal Plans,” Journal of Target,” Quarterly Journal of Economics, Federal Reserve Bank of Philadelphia Political Economy, 85 (1977), pp. 473-91. 100:4 (1985), pp. 1169-89. Business Review (Third Quarter 2006), pp. Leduc, Sylvain, and Keith Sill. “A Walsh, Carl. “Monetary Policy Design: 10-20. Quantitative Analysis of Oil-Price Shocks, Institutional Developments from a Systematic Monetary Policy, and Economic Contractual Perspective,” International Goodfriend, Marvin S. “Interest Rate Downturns,” Journal of Monetary Finance, 3:3 (November 2000), pp. 375-89. Policy and the Inflation Scare Problem: Economics, 51 (2004). 1979-1992,” Federal Reserve Bank of Woodford, Michael M. “Optimal Monetary Economic Quarterly Leduc, Sylvain, Keith Sill, and Tom Stark. Richmond , 79/1 Policy Inertia,” manuscript (May 1999). (Winter 1993), pp. 1-23. “Self-Fulfilling Expectations and the Inflation of the 1970s: Evidence from the Livingston Survey,” Journal of , 54:2 (March 2007), pp. 433-59.

8 Q4 2008 Business Review www.philadelphiafed.org