The concept of voluntary unemployment as applied by new classicals provides us with important insights as to the practice economists by highlighting certain mechanisms and neglecting others make efforts to explore the underlying causal structure. On the paper co-authored by Rapping (Lucas – Rapping 1969), Lucas had a serious debate with Albert Rees (see Rees 1970). In this debate Rees, who was shocked by the fact that Lucas and Rapping tried to interpret an evidently involuntary unemployment situation on the basis of voluntary unemployment, voiced the usual arguments. These concerns were responded to by Lucas with a brief paper (Lucas 16972). Assuming labour market equilibrium did not imply their neglecting or rejecting unemployment as a real social phenomenon and, moreover, the equilibrium approach was definitely in tune with the perceived fluctuations in employment. Lucas and Rapping detailed an employment equation which by Rees was erroneously said to be independent of real GNP. Rees interpreted this step of theirs as omitting aggregate demand as an explanatory variable. This mistake of Rees can be traced back to a simple methodological error: before econometric estimation a simultaneous equation system has to be re-specified in a reduced form instead of the original structural form in order that the effects of exogenous variables could be evaluated. In the Lucas-Rapping model real GNP is an exogenous variable, so eliminating the endogenous variables must not be skipped before the stage of estimations (this is what is meant by specifying a reduced form). By taking this fact into consideration one can clearly understand the relationship: the rate of unemployment is presumed to be dependent on the real wage on the one hand, whereas real wage is assumed to be dependent on real GNP. Bearing this in mind, at the end of a correct mathematical deduction one may get to an equation in which the rate of unemployment can be estimated as a dependent variable of real GNP per household (among other explanatory variables).
All in all, the interpretation in which Lucas and Rapping were said to make unemployment independent of aggregate demand is fundamentally erroneous. Moreover, this model was fairly in accordance with the relationship between real output and unemployment dynamics.
The concept of voluntary unemployment means that households reduce their labour supply when realizing that current wages are below the level of real wages they expected. The crucial problem here is that how quickly households would modify their expectations. According to the critique this modification should occur in quite a short time when the facts once are realized. However, Lucas and Rapping called attention to the fact that “facts” are rather dubious on the one hand, and on the other hand expected wages and prices definitely dropped during the Great Depression, even though slower than the factual data. Actually, these two explanations are interrelated, since there are only few people among the unemployed who are clearly cognizant of their normal and expectable wage level and who in the hope of higher wages tend to persist in looking for new jobs before they may feel tempted to cut their wage claims. On second thoughts this mechanism involves the signal processing problem (according to which information deficiencies bear a part in triggering large scale fluctuations) of the island models appeared soon afterwards.
It stands to reason that unemployment dynamics under the Great Depression could not be explained on information deficiencies and a conception built on voluntary unemployment alone. The persistence of unemployment must have had further causes–causes that were neglected by Lucas and Rapping. But anyway, Lucas and Rapping succeeded in highlighting the fact that until the revision took place unemployment was voluntary. In other words, before unemployment turned into involuntary there was a stage in which it was voluntary. There are multiple, interconnected explanations. As Lucas (1972) puts it:
“It would also be misleading to conclude this paper without discussing the possibility that »traditional« theory can account for the residual rigidities unexplained by our model.”
Yes, this is how Lucas phrased. A careful analysis of the texts sheds light on the fact that Lucas has always been accurate to highlight that the mechanisms he postulated can hardly account on their own for the fluctuations under study.
Lucas, R.E. (1972): Unemployment in the Great Depression–Is There a Full Explanation? Journal of Political Economy, 80(1): 186–191.
Lucas, R.E. – Rapping, L.A. (1969): Real Wages, Employment, and Inflation. Journal of Political Economy, 77(5): 721–754.
Rees, A. (1970): On Equilibrium in Labor Markets. Journal of Political Economy, 78(2): 306–310.