Measuring the Cyclicality of Real Wages: How Important is the Composition Bias?
"Measuring the Cyclicality of Real Wages: How Important is the Composition Bias?" with Gary Solon and Robert Barsky
Quarterly Journal of Economics, Vol 109 No 1, (February 1994), 1-25.
Abstract
In the period since the 1960s, as in other periods, aggregate time series on real wages have displayed only modest cyclicality (blue starred line above). Macroeconomists therefore have described weak cyclicality of real wages as a salient feature of the business cycle. Contrary to this conventional wisdom, our analysis of longitudinal microdata indicates that real wages have been substantially procyclical since the 1960s (red dash line above). We show that the true procyclicality of real wages is obscured in aggregate time series because of a composition bias: the aggregate statistics are constructed in a way that gives more weight to low-skill workers during expansions than during recessions.
The article on JSTOR, or a slightly older NBER WP version.
The OSIRIS programs for data construction and regressions.
Average real wage growth for the balanced panel of men is reported in NBER Working Paper 4202 (and can be calculated from the balanced panel available above). The time dummies used to construct the cyclicality of the real wage for the unbalanced panel (see p.13 and Table II) are:
Avg PSID Real Wage Changes (1) (2) (3) (4) (5) Unemployment Men Men Women Women Every1 Ch Log Year Rate Change Unwghtd Wghtd Unwgtd Wghtd Wghtd BEA Wage 1968 0.036 -0.002 0.10317 0.08416 0.09705 0.07486 0.04698 0.01480 1969 0.035 -0.001 0.08546 0.07610 0.09111 0.08776 0.05281 0.00556 1970 0.049 0.014 0.07924 0.06142 0.04743 0.04317 0.04501 0.00604 1971 0.059 0.010 0.06820 0.05035 0.06136 0.06061 0.03019 0.00863 1972 0.056 -0.003 0.08741 0.08004 0.06191 0.05594 0.06469 0.01315 1973 0.049 -0.007 0.07308 0.06733 0.06478 0.07892 0.03482 0.00771 1974 0.056 0.007 0.07035 0.05688 0.06722 0.05296 0.02019 -0.00461 1975 0.085 0.029 0.01755 0.01112 0.03955 0.05480 -0.00687 0.00410 1976 0.077 -0.008 0.06820 0.06728 0.05524 0.05601 0.04071 0.01157 1977 0.071 -0.006 0.07290 0.06343 0.04437 0.05064 0.03954 0.00640 1978 0.061 -0.010 0.07335 0.06556 0.06284 0.06267 0.00480 0.00824 1979 0.058 -0.003 0.06738 0.05335 0.07603 0.07203 0.03933 0.00048 1980 0.071 0.013 0.04567 0.05153 0.03535 0.03896 0.02264 -0.00144 1981 0.076 0.005 0.02847 0.02628 0.02723 0.02945 0.02263 0.00144 1982 0.097 0.021 0.02531 0.01882 0.05331 0.05432 0.01431 0.00571 1983 0.096 -0.001 0.05539 0.06441 0.06645 0.07673 0.05374 0.00047 1984 0.075 -0.021 0.07858 0.07981 0.03092 0.03628 0.05667 0.00094 1985 0.072 -0.003 0.05323 0.05316 0.09000 0.08601 0.03694 0.00516 1986 0.070 -0.002 0.06091 0.06990 0.06311 0.06574 0.01448 0.01421 1987 0.062 -0.008 0.05155 0.05023 0.08302 0.09988 0.03882 0.00135 Regress chlnW(n) = b(0) + b(1) trend + b(2) ChUnemp + e b(2)-hat recovers resutlts for Tab II col 4 using chlnW(1) and Table II col 6 using chlnW(3). These series (de-meaned) are graphed below. The graph at the top of the page displays the series chlnW(5).
