I explained in my first post on Earl Thompson’s reformulation of macroeconomics that Thompson posited a model consisting of a single output serving as both a consumption good and as a second factor of production cooperating with labor to produce the output. The single output is traded in two markets: a market for sale to be consumed and a market for hire as a factor of production. The ratio of the rental price to the purchase price determines a real interest rate, and adding the expected rate of change in the purchase price from period to period to the real interest rate determines the nominal interest rate. The money wage is determined in a labor market, and the absolute price level is determined in the money market. A market for bonds exists, but the nominal interest rate determined by the ratio of the rental price of the output to its purchase price plus the expected rate of change in the purchase price from period to period governs the interest rate on bonds, conveniently allowing the bond market to be excluded from the analysis.
The typical IS-LM modeling approach is to posit a sticky wage that prevents equilibrium at full employment from being achieved except via an increase in aggregate demand. Wage rigidity is thus introduced as an ad hoc assumption to explain how an unemployment “equilibrium” is possible. However, by extending the model to encompass a second period, Thompson was able to derive wage stickiness in the context of a temporary equilibrium construct that does not rely on an arbitrary assumption of wage stickiness, but derives wage stickiness as an implication of incorrect expectations, in particular from overly optimistic wage expectations by workers who, upon observing unexpectedly low wage offers, choose to remain unemployed, preferring instead to engage in job search, leisure, or non-market labor activity. The model assumptions are basically those of Lucas, and Thompson provides some commentary on the rationale for his assumptions.
One might, however, reasonably doubt that government policy makers have systematically better information than private decision makers regarding future prices. Such doubting would be particularly strong for commodity markets, where, in the real world, market specialists normally arbitrage between present and future markets. . . . But laws prohibiting long-term labor contracts have effectively prevented human capital from coming under the control of market specialists. As a consequence, the typical laborer, who is not naturally an expert in the market for his kind of service, makes his own employment decisions despite relative ignorance about the market. (p. 6)
I will just note parenthetically that my own view is that the information problem is exacerbated in the real world by the existence of many products and many different kinds of services. Shocks are transmitted from sector to sector via complicated and indirect interrelationships between markets and sectors. In the process of transmission, initial shocks are magnified, some sectors being affected more than others in unpredictable, or at least unpredicted, ways causing sector-specific shocks that, in turn, get transmitted to other sectors. These interactions are analogous to the Cantillon effects associated with sector-specific variations in the rate of additional spending caused by monetary expansion. Austrian economists tend to wring their hands and shake their heads in despair about the terrible distortions associated with Cantillon effects caused by monetary expansion, but seem to regard the Cantillon effects associated with monetary contraction as benign and remedial. Highly aggregated models don’t capture these interactions and thus leave out an important feature of business-cycle contractions.
Starting from a position of full equilibrium, an exogenous shift creates a temporary equilibrium with Keynesian unemployment when there is an overall excess supply of labor at the original wage rates and some laborers mistakenly believe that the resulting lower wage offers from their present employers may be a result of a shift which lowers the value of their products in their present firms relative to other firms who hire workers in their occupations. As a consequence, some of these laborers refuse the lower wage offers from their present employers and spend their present labor service inefficiently searching for higher-wage jobs in their present occupation or resting in wait for what they expect to be the higher future wages.
Since monetary shifts, which are apparently observed to induce inefficient adjustments in employment, also change the temporary equilibrium level of prices of current outputs, we must assume that some workers do not know of the present change in the price level. Otherwise, all workers, in responding to a monetary shift, would be able to observe the price level change which accompanied the change in their wage offers and would not make the mistake of assuming that wage offers elsewhere have not similarly changed. . . . (p. 7)
The price level of current outputs is only an expectation function for these laborers, as they cannot be assumed to know the actual price level in the current period. This is represented . . . by allowing labor’s perception of current non-labor prices to depend only on last period’s prices, which are parameters rather than variables to be determined, and on current wage offers. (p. 8)
Because workers may construe an overall shift in the demand for labor as a relative shift in demand for their own type of labor, it follows that future wage and price expectations are inelastic with respect to observed increases in wage offers. Thus, a change in observed wages does not cause a corresponding revision of expected future wages and prices, so the supply of labor does not shift significantly when observed wages are higher or lower than expected. When wages change because of an overall reduction in the demand for labor destined to cause future wages and prices to fall, workers with slowly adjusting expectations inefficiently supply services to employers on the basis of incorrect expectations. The temporary equilibrium corresponds to the intersection of a demand curve and a supply curve. This is a type of wage rigidity different from that associated with the conventional Keynesian model. The labor market is in equilibrium in the sense that current plans are being executed. However, current plans are conditional on incorrect expectations. There is an inefficiency associated with incorrect expectations. But it is an inefficiency that countercyclical policy can overcome, and that is why there is potentially a multiplier effect associated with an increase in aggregate demand.
Uneasy Money 
David
Sorry to be uncharitable, but this looks like a lot of words for what should be a relatively simple question to solve empirically using a survey sent out to hiring managers other sundry wage setters :
1) When you expect business to be bad, do you lower wages and retain workers if they accept, only firing those that decline?
LikeLike
David,
The Cantillon effect seems to allow for patterns that emerge over time from the complex interaction of heterogeneous agents. Viewing the economy as a complex adaptive system makes it harder to say that any set of agent expectations set is “incorrect”. How new monetary policy information propagates through the economy depends on multiple factors that change in importance and relationship (i.e. feedback loops) over time. For instance, one can describe a dynamic in which the dominance of the service sector steepens the SRAS curve. Within this dynamic, easing results in large relative price changes that shock middle-class real income expectations and dampen, rather than amplify, RGDP growth. All this stems from a set of initial conditions and relationships that were quite different from that existing during the GD.
LikeLike
Do any of these wage models take into account the fact that future wages depend on past wages? Eg, if I have a choice between a job paying 80K and one paying 100K, I not only need to take into account the loss of 20K for the next year, but _also_ the fact that any new employer 2 years from now will base their wage offers on my previous salary…
LikeLike
Ritwik, Fair enough, but surely you know that economic theorists are allergic to taking surveys.
David, Actually, I agree with you. The high level of aggregation in almost all macro-models necessarily abstracts from the kind of complex interactions that you talking about. And I do think that those complex interactions are an important part of what happens in business cycles. But if we are going to go with a highly aggregated model, it is worthwhile trying to think through a logical structure of the model that is consistent with the rest of our theory. That’s not necessarily where we should wind up, but that should be a stop we make in our journey.
MMM, They might or might not.
LikeLike