Unformatted text preview: 151 lectures Rising Wage Inequality: Institutional Explanations SBTC explanation suggests to some that rising inequality is inevitable. We can suspect this conclusion because the technology is available and adopted in all countries, but only a few (the U.S. and the U.K.) have experienced substantial increases in inequality. A. Problems with the SBTC explanation 1. Remember that what we observe regarding rising returns to a college degree already reflects public policy, not just market and technological forces. --An increase in the relative demand for skilled labor was accompanied by a slowdown in the increase in the number of college graduates. It could have been accompanied by a much greater increase in the relative supply skilled labor. If that had occurred, relative wages would have remained unchanged. --That is what happened during the 1950s and 1960s with the expansion of college capacity, as well as during the period of electrification in the 1910s and 1920s, when the expansion of high school capacity and graduation rates meant that supply of skilled labor grew as fast as the demand. 2. CPS earnings inequality, when measured by the monthly ORG data, show that most of the overall growth in inequality took place in the early 1980s and that inequality in the lower half of the distribution did not increase in the 1990s, even falling slightly in the late 1990s. --This time pattern is inconsistent with the time pattern of the growth of computer use and IT. --Recall that the rising return to college graduates for males was more the result of declines in pay for HS grads, while the level of real pay for BAs did not change. Yet SBTC should have increased productivity of BAs. 1 3. Re-analyses of Krueger’s study showed that workers who were higher paid were the ones who more likely to be given computers first. --With more complex causality estimations, the computer premium fell to 2 percent. Krueger himself noted that many workers getting the premium had a computer but did not know how to use it. --A study using similar data for Germany found that workers who used pencils at work, or who sat down at work, also received more pay and the computer premium disappeared. 4. Much of earnings inequality is not between education-age groups, as SBTC would predict, but within these groups—at least three-fourths (using variance of log wages decompositions). --This proportion has not changed much over the past few decades. 5. SBTC is not entirely wrong as an explanation, but it is highly incomplete, not the sole answer. B. Institutional explanations Three prime suspects because their effects were concentrated in the 1980s. 1. Deregulation of product markets. Not a substantial source of growth of wage inequality by itself, but perhaps indirectly through de-unionization. 2. De-unionization. The best study is again by Card. He finds that about one-fourth of the rise in male wage inequality is attributable to the decline in unionization. 3. Falling minimum wage. This can explain over half of the changes in wage in wage inequality. C. CEO pay and changing social norms: going beyond CPS data 2 A surprisingly large proportion of the increase in pay inequality has recently been shown with IRS datasets to be concentrated among the top one percent and even the top one-tenth of one percent of earners. --These increases are not due to SBTC but to changes in social norms regarding compensation for CEOs. C. International comparisons Two policy and institutional indicators-- extent of coordinated bargaining and extent of training for non-college graduates—are highly correlated with cross-country measures of wage inequality. D. Consequences of growing inequality: positive or negative trade-offs with efficiency, unemployment and growth? 3 ...
View Full Document
This note was uploaded on 06/06/2010 for the course ECON 151 taught by Professor Staff during the Spring '08 term at Berkeley.
- Spring '08