《宏观经济学 Macroeconomics》课外读物：Recent Developments in the Theory of Rules Versus Discretion

团购合买资源类别：文库，文档格式：PDF，文档页数：19，文件大小：2.07MB

RECENT DEVELOPMENTS IN THE THEORY OF RULES VERSUS DISCRETION Robert j. Barro . GENERAL FEATURES OF RULES AND DISCRETION The older literature on rules versus discretion focused on the intentions and capability of the policymaker. Arguments for rules were based on imperfed knowledge about the economy and on policymakers' tendencies to furthe inappropriate ends, possibly motivated by interest groups. But an intelligen policymaker could take account of incomplete information about the economy when deciding on the optimal discretionary policy. Then if the policymaker were also well-meaning, there was no obvious defense for using a rule in order to bind his hands in advance. Discretion seemed to be synonymous with flexibility, which one had no reason to deny to a smart, benevolent This perspective on rules versus discretion was changed by Kydland and Prescott(1977), who looked at rules as a form of commitment. A commitment amounts to a binding contract, which specifies in advance the actions that someone will take, possibly contingent on some exogenous variables that everyone can observe. In contrast, under discretion, a person promises only to take those future actions that will best further his objectives later on. (Such promises are easy to keep! Thus, discretion is the special case of a rule or contract in which none of today's provisions restrict a persons future actions In the area of private business dealings, it is natural to think about optimal forms of contracts, which would not usually be pure discretion. Similarly, for public policy, the perspective becomes the optimal form of rules or prior restrictions-even the smart, benevolent policymaker is likely to desire and use an ability to make binding promises Kydland and Prescott discuss various areas of public policy in which commitments are important. One example is patents, which encourage inventions, but also restrict the supply of goods ex post. Under discretion, a policymaker who cares about social welfare would invalidate old patents Conce and for all,), but continue to issue new ones. However, the perception of this policy by potential inventors has adverse effects on new inventions, which soon become old inventions. Hence, the optimal policy contains a mechanism to preclude or at least inhibit the abolition of old patents. Then the details of this policy involve the standard tradeoff between the incentive to invent and the ex post restriction of supply s i have benefited from researd from the National Science Foundation. The present paper is an extel f results contained in a previous paper(Barro, 1984. xample, Friedman( 1960, Chapter 4)

THE ECONOMIC JOURNAL The manner of committing future actions varies with the area of public policy. In some cases, such as the duration and scope of patents, the rules are set out in formal law. Then the costs of changing laws(possibly coming under constitutional restrictions ag x post facto laws)enforces the governments commitments However, in the case of the Gold Standard Act in the United States, the existence of a law proved in 1933 to be inadequate protection for those who held gold or made contracts denominated in gold More often a government's commitments rely on the force of reputation hereby people's expectations of future policy are tied in some fashion to past behaviour. For instance, if a government defaults on its debts then potential bondholders are deterred by the perception that future defaults are more likely If a municipality sharply raises property taxes, and thereby reduces property values, then potential residents are deterred from moving in. But, as a general matter, the linkages between past actions and expectations of future behaviour are difficult to formalise in a model II. MONETARY POLICY UNDER DISCRETION A major contribution of Kydland and Prescott was the recognition that monetary policy involves the same issues about commitments as do such areas accumulated capital(via changes in property taxes or in other taxes that lalo k as patents, default on government debt, and imposition of levies on previousl capital). In the case of patents it is obvious that a policymaker must worry about the link between current actions -such as eliminating past patents or changing the form of patent law-and people's perceptions about the value of presently issued patents. Similarly, the monetary authority must consider the interplay between today's choices- whether to engineer a monetary expansion or to change the 'law' governing monetary policy- and peoples beliefs about future money and prie ces Consider the example about the Phillips curve, as discussed in Kydland and Prescott( 1977)and in Barro and Gordon(1983a, b). These models involve the following main ingredients. First, monetary policy works by affecting the general price level. In the simplest setting the monetary authority can use its instruments in order to achieve perfect control over the price level in each period. Second, unexpected increases in the price level (but not expected changes in prices)expand real economic activity. In other words, there is an expectational Phillips Curve. Third, the representative person, and hence the benevolent policymaker, value these expansions of activity at least over some range(which means that existing distortions make the natural level of output too small). In order to focus on the distinction between rules and discretion, tI models assume unanimity about the public's desires and a willingness of the policymaker to go along with this objective. This is, there are no principal-agent problems. Finally, inflation is itself a bad- people value it only as a device to For a discussion of the abrogation of gold clauses in public and private contracts, see Yeager (1966, p. 305). Additional discussions are in Nussbaum(1950, pp. 283-91)and McCulloch (1980)

THEORY OF RULES VERSUS DISCRETION create unexpected inflation and thereby higher levels of economic activity This setup for inflation is structurally similar to the example about patents. At any point in time the policymaker is motivated to generate unexpected inflation in order to stimulate the economy. The analogue is the expansion of supply via the abolition of past patents. But people understand these incentives in advance and therefore form high expectations of inflation Accordingly, the policymaker must choose a high rate of inflation just to stay even-that is, in order for unexpected inflation to be zero. Finally, this high inflation imposes costs on the economy. ( The parallel is the decrease in inventions because of the expectation that current patents will not be honoured later Barro and Gordon ( 1983a, b) analyse the equilibria for monetary policy and inflation for the Phillips-curve model. In the of pure discretion, the policymaker has no mechanisms for committing the future behaviour of money and prices. Rather, he has a free hand to maximise social welfare at each point in time, while treating past events as givens. In this situation there is an incentive at each point in time to create surprise inflation in order to generate economic boom. But individuals understand this motivation and form their expectations accordingly. Thus, actual inflation cannot end up being ystematically higher or lower than expected inflation Overall, two conditions must be satisfied in equilibrium. First, people xpectations of inflation are correct on average, which is a rational- expectations condition. Second, although the policymaker retains the power in each period to fool people via inflation surprises, he is not motivated to exercise this power. In order for this second condition to hold, the policy-maker's drive to create unexpected inflation must be balanced by the marginal cost of inflation itself. In other words, inflation must be high enough so that the marginal cost of inflation equals the marginal benefit from inflation surprises Only then will the chosen rate of inflation-which ends up equal on average to the rate that people expect- be incentive compatible in the sense of according with the policymakers desire to maximise social welfare at each point in time The important point is that this equilibrium involves inflation that is high, but not surprisingly high. Therefore, the economy bears the costs of high inflation but does not receive the rewards that would arise from unexpected inflation The solution just described rests on the presence of benefits from surprise inflation, but does not depend on the existence of the expectational Phillips curve. An alternative model recognises that surprise inflation amounts to a capital levy on assets, such are denominated in nominal terms. At a point in time, unexpected inflation works like a lump-sum tax as a device for generating government revenue. Given that other taxes are distorting, the policymaker(and the representative person in the economy) would value the use of this lump-sum tax. Therefore, this model parallels the previous one with the Phillips curve, even though the source nded to incorporate the standard inflation tax or other real effects from anticipated inflation. Then the best rate of inflation need not be

THE ECONOMIC JOURNAL of benefit from unexpected inflation is different. There is an analogous discretionary equilibrium with high inflation, but with no tendency for aspected inflation to be positive or negative the example of the Phillips curve, the incentive to create surprise inflation on the desire to expand economic activity. But this depends turn on some distortions that make the natural rate of output too low. The and transfer programmes are possible sources of these distortions. 2 Similarly, in the example where the government values surprise inflation as a lump-sum tax, there must be derlying environment in which alternative taxes are distorting. In both cases, the existence of initial distortions underlies the prediction of high inflation. Calvo (1978)discusses the general role of existing distortions in these types of models The main point is that the bad outcomes under discretion depend on the presence of these distortions Barro and Gordon(1983b)view the discret equilibrium as a positive theory of monetary policy and inflation present-day monetary arrangements. Aside from predicting highaverage inflation and monetary growth, the model indicates the reactions to changes in the benefits from unexpected inflation or in the costs of actual infation. For example, a rise in the natural rate of unemployment can raise the benefits from lowering unemploy ment through surprise inflation. It follows that a secular rise in the natural unemployment rate will lead to a secular rise in the mean rates of monetary growth and inflation. Similarly, the policymaker would particularly value reductions of unemployment during recessions. The implication is that monetary growth will be countercyclical, although such a policy can end up with no effect on the amplitude of business cycles a higher stock of nominally-denominated public debt raises the benefits from capital levies via surprise inflation. The model then implies that more public debt will lead to higher values of nominal interest rates(although not to higher unexpected inflation). In other words, the prediction is that deficits will be partly monetised. A similar analysis suggests that indexation of the public debt for inflation- which removes some of the benefits from surprise inflation -will tend to lower inflation and monetary growth. This prediction comes from the positive theory of the money-supply process, rather than from direct effects of indexation on the economy. Finally, a higher level of government spending tends to raise the benefits from lump-sum taxation, because the deadweight losses from other taxes would be higher. This change leads again to higher rates of inflation and monetary growth. The endo ry growth implies that government expenditures are inflationary The model assumes that actual inflation is costly, but does not explain the sourceof these costs. Two frequently mentioned possibilities are the administra I See Barro(1983)for an elaboration of this model These taxes and transfers may themselves be warranted as neces counterparts of (valuable) Hence, there is no implication that the government is failing to optimise on the

THEORY OF RULES VERSUS DISCRETION tive expenses for changing prices and the transaction costs associated with economising on cash holdings. The positive analysis of monetary policy implies that a downward shift in the costs of inflation will lead to more inflation If people think that inflation is not a serious problem, then the economy will end p with a lot of inflation! The analysis implies also that each flicker in the benefits from inflation surprises or in the costs of inflation will be reflected in variations in inflation. In contrast to an environment in which the government stabilises prices, there will be substantial random fluctuations of inflation and monetary growth. Further, the variances of prices and money will be larger the greater the random fluctuations in the variables that influence the benefits from inflation shocks For example, if there are frequent supply shocks(which alter the natural rate of output), then inflation and monetary growth will be volatile III MONETARY RULES The results under discretion contrast with those under rules, which are regimes where the policymaker can and does make commitments about future monetary growth and inflation. Under discretion, the equilibrium involved high inflation, but no tendency toward surprisingly high inflation. Hence, the economy suffered the costs from high inflation, but secured none of the benefits from inflation surprises. The policymaker can improve on this outcome if he can commit himself ex ante to low inflation. If this commitment is credible which means that some mechanism prevents violations ex post-then people also anticipate low inflation. Therefore, the equilibrium would exhibit low and stable inflation, with the same average amount of surprise inflation(zero)as before. These results support a form of'constant-growth-rate rule, although applied to prices rather than to the quantity of money, per se There is a tension in this type of rules equilibrium because the policymaker may retain the capacity to produce large social gains at any point in time b cheating-that is, by generating surprisingly high inflation. Then there may be a temporary economic boom or at least a substantial amount of government revenue obtained via a distortion-free tax. But, if such cheating were feasible and desirable, then people would understand the situation beforehand. In this case the low-inflation equilibrium would be untenable. (Sometimes people say that this equilibrium is ' time inconsistent, although it is actually not an equilibrium at all. )Rather, there would be a high-inflation, discretionary equilibrium, as described earlier. That is why the enforcement power behind the low-inflation rule is crucial. There must be a mechanism for binding the policymaker's hands in advance, so that(surprisingly) high inflation cannot be generated later, even if such a choice looks good to everyone ex post. Note that the rationale for this " binding of hands'applies even though(or actually lly if)the policymaker is well-meaning. This type of commitment is

THE ECONOMIC JOURNAL necessary in order for low inflation to be incentive-compatible and hence Although the low-inflation, rules equilibrium is superior to the high-inflation discretionary equilibrium, the e rules Librium is still benefits from inflation surprises-for example, from lower unemployment or external effects that have not been eliminated. It is the desire to approach the first-best solution via inflation surprises that threatens the viability of the low inflation equilibrium. The pursuit of the first-best tends to push the econom away from the second best of a rule with low inflation, and toward the third best of discretionary policy with high inflation. Again, this perspective highlights the importance of the enforcement power that makes a rule sustainable IV CONTINGENT RULES More generally, the optimal rule may set money or prices contingent on exogenous events, rather than being non-contingent. In some models, such as those where the monetary authority has superior information about the economy, a contingent reaction to business-cycle variables may help to smooth It business fluctuations. However the direct communication of the government's information may be a substitute for the feedback response of Another example of contingent response is the association of wars with high growth rates of money and prices. High wartime inflation constitutes rprisingly high inflation from the standpoint of earlier times at which low inflation during peacetime. This type of contingent rule may be desirable because it generates lots of easy revenue via the capital levy from unexpected inflation during emergencies. In particular, it is possible to hold down distortions from the income tax at the most important times, such as wars. 1 Although the necessary accompaniment is a loss of revenue during the non mergencies effect of this contingent policy is likely to be beneficial Under the gold standard, governments did in fact tend to go off gold during wars, as in the case of Britain during the Napoleonic period and around World War I, and for the United States during the Civil War. This procedure enables a government to pursue the type of contingent policy for inflation that I sketched above. In this sense a movement off gold during wars is not necessarily a violation of the rules. However the subsequent return to gold at the previous lessens this incentive, but does not el able to trigger the distortion-fre Svensson(1984). These models feature public debt with a contingent real payo, - posited by Persson and during peacetime and low in wartime If government bonds are nominally denominated and no sh contingent(for reasons that escape me), the contingent behaviour of inflation achieves the same end

THEORY OF RULES VERSUS DISCRETION parity-as in Britain in the 1820s and 1920s and the United States in the 1870s vas probably an important part of the enforcement process One difficulty with contingent rules is that they may be difficult to verify. It is easy to confuse contingencies with the type of cheating that I described earlier Further, the policymaker would be inclined to explain away high inflation as he consequence of some emergency, rather than as a failure to conform with the rules. Canzoneri(1984) points out that this situation involves asymmetric nformation whereby the public cannot verify the nature of the policymaker's actions even after the fact These considerations favour a rule that is relativel simple, such as a constant-growth-rate rule for prices or money. In any case the contingencies should be limited to well-defined events, such as major wars Although this limitation may miss some gains from contingent action, the of enforcement makes it less likely that th degenerate into a high-inflation, discretionary equmolun e situation will THE POLICYMAKERS REPUTATION Barro and Gordon(1983a) examine some possibilities for substituting the policymaker's reputation for formalrules. In this setting people' s expectations of futureinflation depend in some way on past performance. Unlike the case of pure discretion, the policymaker's choice of today'sinflation rate assigns some weight to the effect on future inflationary expectations. Such considerations motivate the policymaker to hold down the rates of inflation and monetary growth The example considered in Barro and Gordon(1983a)is an application of repeated games as developed by Friedman(1971). Reputational equilibria emerge in which the rates of inflation are weighted averages of that under discretion and that under a constant-growth-rate rule. The higher the policymaker's discount rate, the greater the weight attached to the discretion ary result From a positive standpoint, the findings are qualitatively in line with those under discretion. The main difference is that the reactions of inflation to various shocks- such as shifts in the natural rate of unemployment or in the size of government, are now smaller in magnitude. Hence, the variances(as well as the means) of inflation and monetary growth are smaller than those under discretion One difficulty with these types of reputational equilibria is that they depend. on an infinite horizon for the game between the policymaker and the public. If there is a known, finite endpoint for the game, then reputational considerations ave no weight in the final period. Anticipating this outcome the force of reputation is also nil in the next to last period, and so on, working backwards p to the current period. In other words, a finite horizon causes reputational equilibria to unravel. However, if the game terminates in any period with a probability less than one- but there is no known finite horizon This result, as applied to the 'prisoner's dilemma'problem, appears in Selten(1978)

THE ECONOMIC JOURNAL then the repeated-games approach goes through. In this case the probability of termination effectively adds to the policymaker's discount rate in the solution In some contexts, such as those where a term of office with fixed length important, the finiteness of the horizon eliminates the type of reputationa equilibria considered in Barro and Gordon(1983a). But, even when an infinite horizon is tenable, there are difficulties with multiple equilibria. This multiplic ity of solutions reflects the bootstrap character of the reputational equilibria. Namely if people base future beliefs on the policymaker's actions in some rbitrary fashion, then the policymaker is motivated in a range of cases to validate these beliefs. Hence, various equilibria conform with rational expect- ations as well as with period-by-period optimisation by the policymaker In Barro and Gordon(1983a) the equilibria can be indexed by the length of Be period over which people expect high future inflation when they observe gh current inflation In Friedman ( 1971)the analogous"punishment interval is infinite, while in the basic model of Barro and Gordon(1983a)it is one period Since the punishment does not arise in equilibrium in these models, the only effect of a longer interval is more deterrence against choosing high inflation. Therefore a longer interval is at least as good as a shorter one Canzoneri(1984) uses the approach of Green and Porter(1984), who treat the length of the punishment interval as a parameter. In Canzoneri's model the policymaker has private information about the economy, which the private gents can never observe directly. This information sometimes gives the policymaker good reason to inflate on a contingent basis. But people cannot tell whether he is instead acting in a discretionary manner so as to exploit low inflationary expectations. It therefore turns out that the punishments, which take the form of high expected inflation, occur from time to time as part of the equilibrium. Since these punishments are undersirable, per se, a longer punishment interval imposes costs, which trade off against a greater deterrent value. Therefore the optimal punishment interval tends to be finite. Neverthe- less it is unclear what process would cause the interval to take on either this optimal value or some other value. Hence, the problem of multiple equilibria emains. The solutions also still unravel if the horizon is finite VI REPUTATION IN MODELS WITH DIFFERENT TYPES OF POLICYMAKERS The basic difficulty in the preceding treatment of reputation is that there is nothing to learn about in the models. The policymaker forms more or less reputationfor high or low inflation; yet people know everything about the policymaker's objectives and abilities from the outset. therefore the link between performance and beliefs has the bootstrap character mentioned before, rather than building directly on the revelation of information. In order to provide a basis for learning it must be that potential policymakers differ in ways that are not immediately observable. These differences could involve preferences for inflation versus unemployment, capacities for making commitments about future monetary growth and inflation, lengths of horizons

THEORY OF RULES VERSUS DISCRETION and so on Observed choices for inflation or monetary growth may alter the probabilities that people rationally attach to the policymaker's being of one type or another. Then this process would determine the connection between performance and expectations of future infation or other variables. The policymaker, who knows his own type, takes this learning and expectational process into account when deciding how to act. In particular, the policymaker may wish to influence people's beliefs in one way or another. A full analysis of reputation considers these incentives, as well as the rationality of the publics thiaectations. Aside from giving content to notions of reputation and learning, approach turns out to have two other advantages. First, the results no longer depend on an infinite horizon, and second, the equilibrium is often Models of this type have been applied to problems in the area of industrial organisation by Kreps and wilson(1982)and Milgrom and roberts(1982) Applications of their method to monetary policy include Backus and driffill (1985), Tabellini(1983), and Barro( 1985). In the following I sketch an example of this application Suppose that there are two types of policymakers. Type I makes a serious commitment to low inflation. Type II is incapable of commitments, but acts in the usual discretionary manner to trade off the costs of inflation against the benefits of inflation surprises. (Results are similar if the type I people worry more than the type II about inflation. People cannot tell directly which policymaker is in office, but must make inferences from the observed choices for monetary growth and inflation. Specifically, there is some prior probability that the policymaker is of type I. Then, as long as good performance(low inflation) is observed, people upgrade this possibility via Bayes' Law. In other words, an additional period of low inflation makes people more confident that the policymaker is committed to low inflation. The type II policymaker takes account of this inference process when deciding how to behave. Notably he may be motivated to choose low inflation for a while in order to acquire some( false)reputation for being type I. Thereby, the type II person can hold down inflationary expectations, which helps to lower the overall costs to the economy. In particular, a low value of expected inflation creates an opportun- ty for large gains through infation surprises once the type II policymaker decides to reveal his true nature e, For typical parameter values the equilibrium involves an interval over which pected inflation is a weighted average of the low committed value and the higher value associated with discretion. The type I policymaker faithfully generates low inflation in each period. But, since people fear that he may be type Il, this low inflation rate is below the expected rate. The implied string of negative inflation surprises imposes costs, which may take the form of recession. In an environment where the policymaker's type is uncertain, these costs are necessary in order to fulfill the type I person's commitment to low inflation If the policymaker turns out to be type Il, then he mimics the low inflation of the type I person for some interval. But the probability rises over time that the

点击进入文档下载页（PDF格式）

共19页，试读已结束，阅读完整版请下载

点击下载（PDF格式）

浏览记录