1、SMALL MENU COSTS AND LARGE BUSINESS CYCLES: A MACROECONOMIC MODEL OF MONOPOLYX The conflict between modern neoclassical and traditional Keynesian theories of the business cycle centers upon the pricing mechanism. In neoclassical models, prices are fully flexible. They represent the continuous optimi
2、zation of economic agents and the continuous intersection of supply and demand. In Keynesian models, prices are often assumed to be sticky. They do not necessarily equilibrate all markets at all times. One of the reasons for the resurgence of the equilibrium approach to macroeconomics has been the a
3、bsence of a theoretical underpinning for this Keynesian price stickiness. This note shows that sticky prices can be both privately ef- ficient and socially inefficient. The business cycle results from the suboptimal adjustment of prices in response to a demand shock. To the extent that policy can st
4、abilize aggregate demand, it can mitigate the social loss due to this suboptimal adjustment. In some Keynesian models, prices are simply exogenously fixed.2 In others, agents must set their prices in advance of the transaction date.3 The act of altering a posted price is certainly costly. These cost
5、s include such items as printing new catalogs and informing salesmen of the new price. Yet these “menu“ costs are small and, therefore, generally perceived as providing only a weak foundation for these fixed-price models. However, this in- ference is flawed. Small menu costs can cause large welfare
6、losses. The claim that price adjustment costs are small does not rebut the claim that they are central to understanding economic fluc- tuations. *I am grateful to Olivier Blanchard, Alan Blinder, Avery Katz, Eric Ras- musen, James Rauch, Deborah Roloff, David Romer, Julio Rotemberg, Robert Solow, La
7、wrence Summers, and two anonymous referees for helpful comments, and to the National Science Foundation for financial support. 1. See Gordon I9811 for a general discussion of the role of price adjustment in macroeconomic debate. 2. See, for example, Malinvaud 19771. Gordon I9811 references other fix
8、ed- price models. 3. See, for example, Fischer I9771 and Blinder and Mankiw, 19841. The economy I describe in this paper is a relative of the Type 3 (nominal price contracts) economy in Blinder and Mankiw. O 1985 by the President and Fellows ofHarvard College. Published by John Wiley namely, that wh
9、ile aggregate demand contractions are as- sociated with grotesquely inefficient underproduction, aggregate demand expansions are not associated with similarly inefficient overproduction. There is no obverse to the Great Depression. In- stead, periods of expansion, such as the late 1960s in the Unite
10、d States, are often considered periods of economic prosperity.1 The analysis presented here is all in the context of partial equilibrium. It is, however, possible to construct simple general equilibrium examples that encompass exactly these partial-equi- librium results.ll I suspect that a more comp
11、lete general equilib- rium model would exhibit more pronounced price stickiness. In particular, the introduction of interfirm purchases would exac- erbate price rigidity. In such a model the failure of one firm to reduce its price following a contraction in demand would prevent the costs of other fi
12、rms from falling, thereby reducing those firms incentive to cut prices. The primary qualitative conclusion-that trivial menu costs can have important efficiency effects-would certainly remain true in the context of general equilibrium.12 The theme of this paper appears robust: In almost all eco- nom
13、ic models, agents who have the power to affect prices, exert that power by restricting output. The economys equilibrium, or 9. The price adjustment rule followed by the firm is not itself asymmetric. Instead, the welfare properties of the adjustment process exhibit asymmetry. Studies that concentrat
14、e upon the positive aspects of the adjustment process (e.g., Barro 19721), rather than its normative aspects, thus report no asymmetry. 10.These periods are often considered times of excessive inflationary pressure. Yet there is little concern about the level of output per se. 11.An earlier version
15、of this paper contained such an example. The simplest general equilibrium example contiins n yeoman farmers, each ihoosing between leisure and production of his uniauelv differentiated outut. In eauilibrium. each produces too little. The price adhstkent rule each folfows is s;boptimal,as in the part
16、ial-equilibrium analysis presented in this note. 12. The dynamic nature of the price-setting process, which is undoubtedly important, is ignored in this paper. Whether a contraction is viewed as temporary or permanent probably affects the reaction of firms. Rotemberg 19821, in the context of a somew
17、hat different model, considers the dynamics of price adjustment in more detail. Another important aspect of the problem ignored here is the effects of price desynchronization. Blanchard I9821 explicitly examines this issue. SMALL MENU COSTS AND LARGE BUSINESS CYCLES 537 natural rate, is thus below t
18、he social optimum.13 Because of this, deviations below the natural rate impose greater costs on society than on the price-setting agents. These agents, therefore, have inadequate incentive to return the economy to its equilibrium. An economy of this sort does not recommend passive mone- tary policy.
19、 As long as new information about exogenous demand factors (e.g., velocity) is made available to the monetary authority after private agents set their prices, systematic feedback rules can stabilize output.14 These exogenous demand shocks cause sub- stantial and inefficient fluctuations in output an
20、d employment if the monetary authority does not react. Although firms optimize, their prices are not socially optimal, and in particular, respond too little to adverse demand shocks. This inefficiency appears to be the target of policies that aim directly at the pricing mecha- nism, such as wage-pri
21、ce controls and tax-based incomes policy. Barro, Robert J.,“A Theory of Monopolistic Price Adjustment,“ Review ofEco- nomic Studies, XXXIX (19721, 17-26. -, and Mark Rush, “Unanticipated Money and Economic Activity,“ Rational Expectations and Economic P01ic.y Stanley Fischer, ed. (Chicago, IL: Uni-
22、-versity of Chicago Press, 1980). Blanchard, Olivier J.,“Price Desynchronization and Price Level Inertia,“ National ureauof conomic eseakh Working Paper No. 900, 1982. Blinder, Alan S., and N. Gregory Mankiw, “Aggregation and Stabilization Policy in a Multi-Contract Economy,“ Journal of Monetary Eco
23、nomics, XI11 (19841, 67-86-. Fischer, Stanley, “Long-Term Contracts, Rational Expectations, and the Optimal Money Supply Rule,“ Journal ofPolitical Economy, LXXXV (1977i, 191-206. Gordon, Robert J., “Output Fluctuations and Gradual Price Adjustment,“ Journal ofEconomic Literature. XIX (1981). 493-53
24、0. ray;Jo Anna, “Wage Indexation. A acroeconomic Approach,“ Journal of Mon- etary Economics, I1 (19761, 221-35. Malinvaud, Edmond, The Theory of Unemployment Reconsidered (Oxford: Black- . . - -well, 1977). Okun, Arthur M., Prices and Quantities: A Macroeconomic Analysis (Washington, D.C.: The Brook
25、ines Institution. 1981). Rotemberg, Julio, “ooolistic price Adjustment and Aggregate Output,“ Review of Economic Studies, XLIV (19821, 517-31. 13. Okun 1981, p. 2671, in his already well-known Prices and Quantities, writes that “there are strong grounds for the presumption that in macro equilib- rium the output of the price-tag economy is below a social optimum, and that the extra output generated by a strengthening of aggregate demand augments social welfare.“ 14. See Blinder and Mankiw I9841 for a more complete exposition of this issue.
