| Journal of Institutional and Theoretical Economics |
Vol. 157, No. 3, September 2001 |
Book Review by Simon Gächter
Why are wages rigid? This question certainly is a fundamental one in economics. It is fundamental because it is at odds with two often invoked assumptions in economics: that people are rational and selfish maximisers and that (labour) markets are competitive. In particular, in recessions, where labour is in excess supply, wages should fall to clear the labour market. However, wage cuts are rare and do not even occur in unregulated nonunionised markets. The question of wage rigidity has attracted the interest of many economists ever since KEYNES [1936], who suggested that rigid wages may lead to involuntary unemployment. Others, like LUCAS AND RAPPING [1969], argue that labour markets are competitive and that wage rigidity is an illusion. Unemployment is voluntary and occurs because in a recession wages fall below reservation wages. Workers, therefore, prefer leisure to employment. In contrast to this explanation, efficiency wage theories provide theoretical arguments why it may not be in the interest of a profit-maximising firm to lower wages. To explain wage rigidity, still others have developed theories of implicit contracts, bargaining, and "insider-outsider" relationships. Thus, theories abound, and evidence is needed. But what counts as evidence? Typically, empirical economists very often only have highly aggregated data at their disposal; if they are lucky, they can analyse payroll data at the plant level. However, while such data are clearly welcome, they do not provide insights into the decision mechanisms or motivations that have generated these data. Experimental methods can circumvent the latter problem, because a controlled environment allows studying motivations that may generate decisions. Indeed, it has been possible to find wage rigidity even in highly competitive laboratory markets. In these experiments rigid wages were a consequence of workers' reciprocal behaviour, which prevented profit-maximising firms from lowering wages (see FEHR AND FALK [1999]). The drawback of experiments is that they only suggest possible mechanisms for wage rigidity that may be important in real-world labour markets. But what is important for real-world wage setters? What do they think are crucial determinants of wage policy? How does the labour market function in the perception of personnel managers who deal with employees on a daily basis? To answer these questions, Truman Bewley took a novel and highly original approach: he interviewed personnel managers, union representatives, labour lawyers, and job service counsellors about their perception of important features of labour markets and determinants of wage policies. While such an interview approach is common in the other social sciences, it is certainly unusual for economists (but see, e.g., BLINDER AND CHOI [1990]; AGELL AND LUNDBORG [1995]; CAMPBELL AND KAMLANI [1997]). Even more unusual is the style of his survey methods: His aim was not to produce representative questionnaire data that are easily quantifiable and amenable to rigorous statistical analyses. Bewley's approach was to use unstructured interviews in which there was no prespecified questionnaire according to which people gave their answers. His interview partners just told him how they see things. Bewley's style of analysis is qualitative. The answers are grouped according to recurring arguments, and absolute numbers (and percentages) of certain arguments are given. He illustrates important arguments by quoting at length what his interview partners actually said. Bewley conducted 336 interviews, most of them in person but some by telephone. The interviews took place in 1992, i.e., at a time where the U.S. was recovering from a recession. His sample is not representative; many of the interviews were arranged by other interviewees, by private connections, by reference, or by just calling. The addressed topics comprise morale, internal and external pay structure, the shirking variant of efficiency wage theory, hiring, the experiences with pay reductions, layoffs, unemployment, and the role of information and negotiations. Given Bewley's almost ethnographic approach, the resulting picture of how actual labour markets work is extremely rich and detailed. Details notwithstanding, there are some strong regularities. One thread of the amassed evidence is the importance of morale: Firms abstain from wage reductions because they would adversely affect morale ("cooperativeness," "loyalty," "reciprocity," etc.), which, at least in the primary sector, is crucial according to most interviewees. Wage reductions that just take advantage of labour market conditions would almost certainly be counterproductive (and therefore they rarely occur). Only in cases where a trusted management can communicate that the company itself is in serious danger of going out of business are wage reductions acceptable. Morale is also at the heart of internal wage structures, a main function of which is to ensure pay equity. Internal pay equity also determines the pay of new hires (although their pay is more downwardly flexible than those of incumbents, particularly in the secondary sector). The external labour market turned out to be much less important for morale than the internal pay structure. The importance of morale is also a reason why the shirking variant of efficiency wage theory is not well received by the managers. Management by threat of dismissal is viewed as being detrimental to morale and therefore inappropriate. An interesting finding is that firms prefer layoffs to wage reductions. Although layoffs also hurt morale, their effects are less severe and long-lasting than those of wage reductions, which particularly upset employees. Morale is particularly important in the primary labour market, and less so in the secondary sector, which is characterized by simple part-time jobs with no long-term perspectives and high turnover. Labour contracts are less incomplete than in the primary sector, and morale is less important to ensure compliance. Moreover, since turnover is high and many employees have short-term perspectives, equity concerns are negligible. Consequently, wages are expected to be much more flexible than in the primary sector, and indeed they are. Bewley documents many further interesting business practices. For example, managers in the primary sector are generally reluctant to hire overqualified people, who very often are desperate enough to be willing to work for very low wages. Many employers refuse to hire the overqualified because managers believe them to be disgruntled and ready to leave whenever a better job shows up. Bewley's interview evidence also sheds light on the important question of involuntary unemployment. If employees are taking some time off because they are not willing to work for the going wage, as is assumed in the LUCAS-RAPPING [1969] theory, then the unemployed should not be too unhappy about the situation. However, all interviewees agree that becoming unemployed is almost certainly not a voluntary choice. It is a serious blow that undermines self-esteem and leads to severe economic and psychic hardship. Thus, Bewley's evidence is consistent with findings from happiness studies, which also consistently find that being unemployed is one of the most important factors for unhappiness (see, e.g., CLARK AND OSWALD [1994]). What do these observations mean for the received theories of wage determination? In one of the last chapters of the book Bewley addresses existing theories of wage determination and confronts them with his findings. His conclusion (on p. 423) is short, and extremely sharp: "The only one of the many theories of wage rigidity that seems reasonable is the morale theory of Solow and Akerlof. The others fail because of the lack of realism of their basic assumptions." This certainly is a provocative statement that should encourage further investigations. The book concludes with suggestions for the directions of future work. Of course, like any empirical method, surveys do have their problems, and Bewley addresses them right from the start. Surely, the findings are not representative in a statistical sense. Also, many readers will have to get accustomed to the often bewildering details that cannot easily been subjected to known statistical tests. However, there are threads and regularities, like the importance of morale and nonselfish behaviour, that can be investigated with more conventional methods, like laboratory experiments. Indeed, some of the important regularities reported by Bewley can be detected (and replicated!) in the lab as well (see, e.g., FEHR AND FALK [1999]). Morale effects can also be rigorously modelled as Bewley himself does in the second to last chapter of his book. The evidence provided by Bewley and others has already stimulated new theoretical developments that can be fruitfully applied to the issue of wage rigidity (see, e.g., DUFWENBERG AND KIRCHSTEIGER [2000]). This is an amazing book, which presents many findings that are only surprising to those economists who have never studied economic behaviour as exhibited by humans. To most practitioners and other social scientists many of Bewley's observations just record the obvious. Does this mean that this book is not worth reading? Quite the contrary! I think any scholar interested in labour markets and wage determination should read this well written, lively, and highly stimulating book. A very valuable feature for those who want to dig deeper is the extensive bibliography (drawn both from economics and from other social sciences) that is provided on each topic. In summary, this book provides a fresh view and a lot of complementary background knowledge about how experienced people in the field see the employment relationship and what is actually crucial. Knowledge of this sort is all too rare in economics, and Truman Bewley's truly impressive study can serve as a role model for future investigations. References AGELL, J., AND P. LUNDBORG [19951, "Theories of Pay and Unemployment: Survey Evidence from Swedish Manufacturing Firms," Scandinavian Journal of Economics, 97, 295-307. BLINDER, A., AND D. CHOI [1990], "A Shred of Evidence on Theories of Wage Stickiness," Quarterly Journal of Economics, 105, 1003-1015. CAMPBELL, C., AND K. KAMLANI [1997], "The Reasons for Wage Rigidity: Evidence from a Survey of Firms," Quarterly Journal of Economics, 112, 759-789. CLARK, A., AND A. OSWALD [1994], "Unhappiness and Unemployment," Economic Journal, 104, 648-659. DUFWENBERG, M., AND G. KIRCHSTEIGER [2000], "Reciprocity and Wage Undercutting," European Economic Review, 44, 1069-1978. FEHR, E., AND A. FALK [1999], "Wage Rigidity in a Competitive Incomplete Contract Market," Journal of Political Economy, 107, 106-134. KEYNES, J. M. [1936], The General Theory of Employment, Interest, and Money, Macmillan: London. LUCAS, R., and L. RAPPING [1969], "Real Wages, Employment, and Inflation," Journal of Political Economy, 77, 721-754. |