February 2002, Vol. 125, No. 2
Précis
Stock options and wage puzzles
Technology, work, and wages
Stock options and wage puzzles
The 1990s, according to Hamid Mehran and Joseph Tracy’s report in the December 2001 Economic Policy Review from the New York Federal Reserve, provided economists with two "wage puzzles." First, why did both nominal compensation growth and the unemployment rate fall in the 1992–95 period? And, second, why did the compensation growth rate drop again in 1999 even as the labor market continued to tighten?
Mehran and Tracy seek answers to these conundrums in the changing structure of compensation. A growing share of compensation for workers at all levels has been paid as profit sharing or in stock options. While the data captures straight profit-sharing payments well enough, stock options are captured when they are realized, not when they are granted. Mehran and Tracy use data available from several sources to estimate the impact changing that timing would have on the total compensation series.
First, they model some of the factors that help determine how many stock options firms will grant in terms of the standard Black-Scholes value of those options. Some significant factors are the firm’s return on assets (higher returns are associated with fewer new stock options), firm size (bigger firms grant more options), firm age (older firms grant fewer), and profitability (firms suffering losses often grant more stock options than similar firms with operating profits).
Mehran and Tracy then model the determinants of stock option realizations, both in terms of the likelihood a firm will experience any realizations of its options (about a third of firms that have granted options experience no realizations in a year) and in terms of the magnitude of those realizations. The number of previous grants, the return on the underlying stocks over the previous 2 years, and the market-to-book value of the shares have strong effects on both the incidence and magnitude of realizations. Firm size and higher stock risks have effects on the likelihood options will be realized, but not on the magnitude of the realizations.
Once these model-based results are taken into account, Mehran and Tracy attempt to adjust estimates of growth in compensation hour that include estimates of option realizations to an estimate of the growth rate in compensation that includes option grants. Data constrains limited this analysis to 1995–99, more or less the time period of the second wage puzzle. They found that a measure of compensation that included their estimate of grants and excluded their estimates of realizations grew in every year they had data for. In contrast, the unadjusted series had that puzzling slowdown in 1999.
The impact of technological change on workplace practices, productivity, and, most especially, wages and wage structures has been a frustrating and often controversial research field. Indeed, Clair Brown and Benjamin A. Campbell begin their recent Industrial Relations review article on these topics with two paradoxes. The "paradox of productivity" is that the impact of new technology on productivity is more evident at the individual units of the economy than at the national level. That is, individual firms adopting new technologies often experience significant improvements in productivity, but these impacts do not seem to percolate up to the aggregate statistics.
The "paradox of wage inequality" is that wage premiums for workers using advanced technology appear to be much more significant in national statistics than they do at the firm level. In other words, while measure of aggregate inequality move in concert with measures of technology, individual units that adopt new technology do not report much widening of their wage distributions.
After reviewing the literature that has led us to these paradoxes, Brown and Campbell suggest three critical topics for future research. First, there must be additional improvements in the measurement of technology and productivity. They state strongly, "Our current measures of technology usage and costs are woefully inadequate, and our measures of output, and hence productivity, do not reflect many of the improvements provided by the new technology."
Second, analysts must further unravel the impact of new technology on wages. Brown and Campbell assert, "If wage premiums are not directly related to skills and other observable productivity-enhancing characteristics, then we must ask how institutions governing the rationing of jobs into higher-paying firms … are affected by technological change as well as other phenomena."
Third, there should be more studies of the impact of technical change on workplace practices and employment systems. "In particular, say Brown and Campbell, "we need to study what creates market rent for a firm, how these rents are divided among the owners and the workers, and how new technology changes these relationships.
We are interested in your feedback on this column. Please let us know what you have found most interesting and what essential reading we may have missed. Write to: Executive Editor, Monthly Labor Review, Bureau of Labor Statistics, Washington, DC. 20212, or e-mail MLR@bls.gov |
Within Monthly Labor Review Online:
Welcome | Current
Issue | Index | Subscribe
| Archives
Exit Monthly Labor Review Online:
BLS Home | Publications & Research
Papers
