خشم مشتریان در افزایش قیمت ها، تغییر در فرکانس تعدیل قیمت و سیاست های پولی

While firms claim to be concerned with consumer reactions to price increases, these often do not cause large reductions in purchases. The model developed here fits this by letting consumers react negatively only when they become convinced that prices are unfair. This can explain price rigidity, though its implications are not identical to those of existing models of costly price adjustment. In particular, the frequency of price adjustment can depend on economy-wide variables observed by consumers. This has implications for the effects of monetary policy and can explain why inflation does not fall immediately after a monetary tightening.

Price setters have been asked on repeated occasions to explain why their prices stay constant in nominal terms for periods of time that are vastly longer than the period over which the opportunity cost of production stays constant. The two most common answers received by Hall and Hitch (1939) involved the psychology of customers. They were, in particular, that “conventional price [is] in [the] minds of buyers” and that “Price changes [are] disliked by buyers.” Blinder et al. (1998) asked price setters about the validity of various theories of sticky prices developed by economists, so they did not ask directly whether price changes were disliked by customers. Nonetheless, a majority of their respondents volunteered that changing prices would “antagonize” or “cause difficulties” with their customers (p. 308). The simplest model of such negative reactions would posit that the quantity demanded falls dramatically every time a price is increased, perhaps because price increases trigger search as suggested by Stiglitz (1984). However, many price increases are associated with only trivial instantaneous reductions in the quantity demanded. A second, and more standard explanation for the existence of periods where prices are fixed, involves the assumption that there are non-convex costs of changing prices. These costs are usually motivated by the observation that posting prices involves physical resources so that resources must be spent to change prices by, for example, printing new catalogues (Sheshinski and Weiss, 1977). While Levy et al. (1997) show that the resource costs of changing prices in supermarkets are nontrivial, these administrative costs simply cannot be the whole story. To see this, Fig. 1 shows the prices charged by a supermarket chain for a particular product. A striking feature of this series is the recurrence of downwards spikes, short periods where this particular item is “on special”. In addition to new price labels, this often involves changes in the item's physical display. The price changes again when the promotion is over though the supermarket often chooses exactly the same price as prevailed before the special. Thus, firms appear to have a preference for prices they have charged before even when the resource costs needed to post such prices are the same as those that would be needed to display any other price.Another common approach to modelling price rigidity is to suppose that price setters have imperfect information (see Lucas, 1972). However, the prices of many raw materials are well publicized and move minute by minute while, at the same time, finished goods prices accounting for a large fraction of sales are set by a small number of individuals. It seems hard to believe that these individuals, who are presumably selected for their ability, fail to update their beliefs regarding optimal prices for months at a time. On the other hand, it seems quite plausible for customers to have poor information about the costs of producers, and this plays a key role in the model. One attraction of focusing on how consumers perceive prices is that firms routinely say they want their prices to be “fair.” As discussed by Hall and Hitch (1939), many firms set prices using the “cost-plus” method which involves starting with variable unit cost, adding the average overhead cost per unit under that assumption that firms produce at “capacity” and, lastly, adding a margin of profit. Oxenfeldt (1951, p. 158) reports: “Questionnaire and field surveys indicate the particular acceptance of a margin as ‘fair’ to be the most important reason for the widespread use of that margin.” As emphasized in the questionnaire study of Kahneman et al. (1986), consumers also have opinions about the fairness of price changes. My theory of price rigidity hinges on the assumption that consumers use nominal price changes as a trigger for reflection about the whether producers are fair or not. Because price increases lead consumers to re-think the extent to which firms are acting fairly, firm will be reluctant to change their prices often. This fits with the suggestion that firms ought to try to “improve” their product when they raise prices. Miller (1976, p. 23) makes this suggestion to restaurants as a way to overcome customer resistance when printing a menu with higher prices. He recommends adding something (like potato chips) “to a standard item and creat[ing] a new package that includes the standard but can be sold for a slightly higher price”. Consumers’ evaluation of the fairness of a price-changing firm depends on their information. As their information changes, their resistance to price increases should change as well. A firm that knows its customers would obviously time its price increases so they occur when resistance is relatively low. Thus, the random receipt of information by consumers can rationalize the random price changes assumed by Calvo (1983). A difference with the Calvo (1983) model, however, is that price increases ought to be more frequent when the macroeconomic environment suggests that such increases are fair. Thus, for example, high inflation in the past might convince consumers that costs are likely to have increased and thus may make price increases easier to sustain. This means that lagged inflation can have an effect on current inflation even if price-setters are purely forward-looking. Considerations of this type may help resolve some of the empirical difficulties faced by the unmodified Calvo (1983) model. One well known property of this model is that it tends to predict that realistic monetary contractions have their largest impact on the inflation rate in the quarter in which interest rates are unexpectedly increased. Empirically, by contrast, unexpected increases in interest rates seem to reduce inflation only with a lag. While several modifications of the Calvo (1983) model help resolve this inconsistency (see Rotemberg and Woodford (1997) and Christiano et al. (2005)) it is worth understanding how customer anger can do so as well. Consider an environment with positive steady state inflation so that the typical firm that changes its price raises this price. A contractionary monetary policy typically reduces the size of this price increase because the reduction in demand tends to lower marginal cost. In the case where price rigidity is caused by the fear of consumer anger this can, paradoxically, increase the fraction of firms that raise their price. The reason is that the price increase that firms now desire has just become more palatable. This increased frequency of price increases can offset the decline in the price chosen by those firms that change their prices. There is also a second reason for an increase in the frequency of price adjustment in the immediate aftermath of a monetary contraction, though I do not model this explicitly. This is that such a shock makes firms realize that price increases will become less palatable once inflation slows down. This acts as an incentive to raise price before inflation falls, and thereby postpones the onset of low inflation. The paper proceeds as follows. The next section presents some evidence on the effects of price increases. It considers whether consumer “resistance” to price increases boils down to a precipitous fall in demand at the moment that prices are increased. Insofar as sales do not fall sharply every time prices are raised, the model proposed in the subsequent section is more attractive. The reason is that this model involves the possibility of sharp drops in sales if consumers feel the price increase is unfair but, most of the time, firms will choose price changes that pass muster with their customers. Section 3 presents the model of consumer reactions to price increases. Section 4 turns to macroeconomic considerations by discussing a general equilibrium model with random price changes where the frequency of price changes can depend on observable economic variables. Section 5 then analyzes how this model behaves in response to monetary policy shocks and Section 6 concludes.

This paper has shown that the threat of consumer anger can account for the constancy of prices from one period to the next while also having the potential to explain some of the dynamic responses of the economy to monetary policy shocks. The consumer reactions in the model are “irrational” in the sense that consumers are maximizing something other than a utility function that depends only on their own material payoffs. Rather, they also wish to harm (or at least not to help) firms that they see as having given them a bad deal. Understandably, this leads firms to be careful not to induce these emotional reactions. One attraction of modelling price rigidity as stemming from consumer reactions is that this provides a new mechanism through which lack of information about economic conditions translates into muted price responses. It is easy to believe that consumers are poorly informed about cost changes and this may lead firms that are concerned with consumer reactions to make their prices less sensitive to costs. By contrast, more standard models in which poor information leads to sluggish price adjustment suppose directly that producers are imperfectly informed about either costs or demand. Given the huge incentives for producers to acquire all relevant information and given that a small number of sophisticated individuals makes the bulk of the economy's pricing decisions, this approach seems less attractive. Heterogeneity in information sets ought not to be confined to differences between producers and consumers. Consumers, in particular, are likely to differ a great deal from each other in both their information and their attitude towards suppliers. Nonetheless, anger at producers appears to be communicable and this seems capable of leading to the sort of discontinuous change in purchases considered in this paper. Thus, information transmission from one set of consumers to another, particularly in situations where some consumers feel that the firm has stepped over the line, seems to be important in practice. Modelling this information transmission thus remains an important topic for future research. Another attractive modification would involve allowing consumers to be more aware of the actual problems faced by producers. In the current model, consumers have a theory about how “fair” producers behave and producers take this theory into account when setting prices. While inherently plausible, the theory that consumers have about producer behavior does not correspond as closely as seems desirable to the way the producers actually behave in the model. In particular, it would be attractive to let consumers recognize that producers solve a dynamic model when they set prices.