Question: What is the role of menu costs in New Keynesian models? Is the importance placed on menu costs in these models justified by the empirical evidence?
Most economists believe that short-run fluctuations in output and employment represent deviations from the economy’s natural rate. They think these deviations occur because nominal wages and prices are slow to adjust to changing economic conditions. This stickiness makes the short-run aggregate supply curve upward sloping rather than vertical. For decades, many New Keynesians worked on possible reasons and relevant evidence for the price stickiness by putting the traditional theories of short-run fluctuations on a firmer theoretical foundation, examining the microeconomics behind. Among these well-known economic theories, menu cost is the one I will talk about here. The aim of this essay is to demonstrate the role of menu costs in New Keynesian economic models and find out whether it was justified by the empirical evidence.
Before talking about the role of menu costs, I have to point out that the fundamental ‘new’ idea behind New Keynesian models is that of imperfect competition. All the major innovations of the New Keynesian school are made possible or worthwhile only because of imperfect competition. By this I mean that all the firms set their own profit-maximizing prices rather than take from the market.
So what are exactly menu costs? Menu costs are costs of changing nominal price. The classic example of menu costs is the costs that a restaurant faces when it has to reprint its menu to show changes in the prices of its offerings. More general examples of menu costs include costs of remarking merchandise, reprinting price lists and catalogues, and informing potential customers. The New Keynesian approach assumes that small menu costs will deter imperfectly competitive firms from constantly changing their prices.
After the definition let us turn to find out the role of menu costs. Suppose there is an increase in aggregate demand (See figure 1 in appendix). This increase will lead to an increase in demand at firm level from D0 to D1; the new profit-maximizing price is P1 at output level Y1. Assume that the nominal wages are sticky so the marginal cost schedule is unchanged. Firm will incur menu cost z if it increases price to P1. The gain from changing price must be larger than the gain from not changing price, as P1 is the new profit-maximizing price. So the change in profit from not adjusting price must be negative because area A is greater than area B. That is ??= B-A < 0. It is obvious that the firm will not adjust its price if and only if menu cost incurred is larger than the profit gain from adjusting price, i.e. z > -(B-A) = A-B.
On the other hand, the change in social welfare from firm not adjusting price equals the change in consumer surplus plus the change in firm’s profit. The change in consumer surplus is area A+C so that ?SW = ?CS +??= (A+C)+(B-A)=C+B.|C+B|>|B-A|so the gain in social welfare is larger than the firm’s profit loss. The firm has a private incentive to increase price if A-B>z but it is socially desirable that the firm does not adjust its price if C+B>z.
This is an example of the Keynesian notion of macroeconomic externality: a cost incurred by society as a result of a decision by an individual economic agent. If the change in demand is small, the profit loss from not adjusting price is likely to be smaller than menu costs so the firm will not change price. Even quite small menu costs can lead to significant price rigidity. However, whilst the menu costs might be quite small, the welfare benefits of an increase in demand can be large. If the increase in demand is due to a tax cut or an increase in the nominal money supply, then when there are menu costs, this sort of policy can lead to ‘Keynesian’ multiplier effects and also ‘Keynesian’ welfare effects, since there is a Pareto improvement – consumers and the owners of firms are both made better off.
The case of a fall in aggregate demand could be analysed in the similar way (See figure 2 in appendix). The firm ignores the macroeconomic externality when making its decision, so it may decide not to pay the menu cost and cut its price even though the price cut is socially desirable. The resulting profit gain for the firm from not to adjust price will be small in comparison to the large welfare loss for society. Hence, sticky prices may be optimal for those setting prices, even though they are undesirable for the economy as a whole.
Now the role of menu costs is clear. For New Keynesians, the importance of price stickiness is that it helps explain monetary nonneutrality. Menu costs hence have important macroeconomic implications. First, menu costs can be a source of price rigidity and thus provide a micro-based explanation for monetary nonneutrality. Second, even small menu costs may be sufficient to generate substantial aggregate nominal rigidity and large business cycles.
However, despite the important role of menu costs in New Keynesian models, economists disagree about whether menu costs explain the short-run stickiness of prices. What will the empirical evidence tell us? Is there evidence to support the menu costs theory?
The behaviour of L.L.Bean (a US mail-order clothing company) catalog prices, documented by Kashyap in 1991 illustrated little evidence of menu costs on the nature of firms’ pricing policies. Bean issues over 20 catalogs every year, changing the prices listed in a new catalogue is virtually costless. Yet Kashyap found that the nominal prices of many goods remained fixed in successive issues of the catalogue, only changed in two of the catalogs (Fall and Spring).
Neither fact supported the view that the cost of printing and posting a new price is the barrier to price adjustment. When nominal prices were changed, Kashyap found both large and small changes. He interpreted the combination of small price changes and long periods of unchanged prices as evidence against menu costs. If menu costs are the reason that prices are not changed frequently, prices should be changed only when they are relatively far out of line, and the price changes should be large; small changes seem to contradict this implication of menu costs.
Also in 1991, Alan Blinder attacked price stickiness directly by surveying firms about their price-adjustment decisions. Blinder began by asking firm managers how often they change prices and the answers were summarized in table 1.
The votes pointed overwhelmingly toward menu costs. The impression that menu costs are the dominant form of adjustment costs was strongly confirmed.
Blinder concluded that first, about 70 percent of all firms reported that they have special costs of price adjustment. Second, there were strong indications that adjustment costs for changing prices are more often menu costs than convex. Empirical findings that some firms make very small price changes should not be construed as rejection of the menu-cost theory. Third, fewer than half thought adjustment costs an important factor in slowing down price responses.
Another study was carried out by Daniel Levy, Mark Bergen, Shantanu Dutta and Robert Venable in 1997. They measured menu costs of five large US retail supermarket chains and provided evidence that menu costs may form a barrier to price changes. First, they contrasted the price change activity of a chain that operates in a state with an item pricing law with another four chains that operate in states not subject to such laws. They showed that the average menu cost per price change for the chain subject to the item pricing law was $1.33, over two and a half times the cost for the other four, $0.52. Hence supermarket chains not subject to item pricing law changed prices two and a half times more frequently than the chain that is subject to the law.
Second, within the chain facing the item pricing law, there are 400 products that are exempt from this law so they faced lower menu costs. Levy et al showed that for these products the chain each week changed the prices of 21 percent of the products on average, which was three times more frequently than for products subject to the law. Third, they found that the chains not subject to item pricing laws every week experienced cost increases on about 800-1000 products they sold. Yet, they adjusted prices of only about 70-80 percent of these products. The remaining 20-30 percent of the prices were not adjusted immediately because the existing menu costs made the necessary price adjustment unprofitable. Levy et al’s findings offered direct support for the relationship between menu costs and store-level individual price rigidity.
To conclude, menu cost is important in New Keynesian economics as it helps to explain the price stickiness which is in turn the explanation of monetary nonneutrality. In addition, it is a classic example of macroeconomic externality and has big macroeconomic implication – small menu costs but large business cycles. The empirical evidence told us that menu costs did appear to play a role for many firms. However, menu costs by themselves are unlikely to generate substantial nominal price rigidity. But they may have the effect of keeping nominal prices unchanged to become equivalent to “doing nothing,” and thereby generate considerable rigidity. The end result is likely to be that menu costs are a significant barrier to price changes only if they are reinforced by other rigidities such as nominal wage rigidity. Nominal price stickiness may be both stronger and more complicated than it would be if menu costs alone were the only frictions.
R. J. Gordon (2000) Macroeconomics (8th edition) Chapter 17, sections 17.6-17.11
G. N. Mankiw (1985) “Small menu costs and large business cycles: a macroeconomic model of monopoly” Quarterly Journal of Economics 100, P529-537
D. Romer (2000) Advanced Macroeconomics (2nd edition) Chapter 6 Part C
N. G. Mankiw (2000) Macroeconomics (4th edition) Chapter 19, P515-525
A. Abel, B. Bernanke, R McNabb (1998) Macroeconomics Chapter 12
A. S. Blinder (1994) “On sticky prices: academic theories meet the real world” Chapter 4 in N. G. Mankiw (ed) Monetary Policy University of Chicago Press
H. D. Dixon (1997) “The role of imperfect competition in new Keynesian economics”
Chapter 7 in B. Snowden and H. R. Vane (eds) Reflections on the development of modern macroeconomics, Edward Elgar
A. K. Kashyap (1991) “Sticky prices: new evidence from retail catalogs” Quarterly Journal of Economics, P245-274
D. Romer (1993) “The New Keynesian Synthesis”, Journal of Economic Perspectives, P6-22
D. Levy et al (1997) “The magnitude of menu costs: direct evidence from large US supermarket chains” Quarterly Journal of Economics P791-825