Obsolescence of Knowledge and Career Publication Profiles: Some Evidence of Differences among Fields in Costs of Interrupted Careers
研究知识过时速度如何影响学术职业中断的成本，发现不同学科知识耐久性差异显著，女性因生育等中断职业的成本更高，从而影响其职业选择。
In recent years, women have received about 45 percent of all baccalaureate degrees, but only approximately 20 percent of all Ph.D.s. Does this imply that academic careers, as compared with careers of the wider market for college graduates, are less conducive to dual career-family objectives? Evidence suggesting the relative tradeoff of these objectives is shown by the fact that over the last thirty years, the (zero-order) correlation between average cohort fertility and the proportion of Ph.D. recipients who are women is 0.939. A female recipient of a doctorate is twice as likely as a recipient to have specialized in the humanities and only one-third as likely to have specialized in a physical science field. Furthermore, of those who receive a doctorate in physical science, a is 60 percent more likely than a female to pursue research as a primary work activity. In the humanities, on the other hand, a woman is at least as likely as a man to have research as an initial postdoctoral work activity. Obviously, many factors affect opportunities and resultant decisions to enter particular careers and to select specific occupational specialties. This study considers only onethe rate of obsolescence or depreciation of knowledge as reflected in the age profile of cited works. This measure of the obsolescence of knowledge refers to the rate of decay in the relative frequency at which citations are made to a particular vintage of research communications as the communications age; in other words, the usefulness of older published research papers as an input to current research depreciates over time. The rate at which this depreciation occurs is interpreted as measuring the durability of a professor's accumulated stock of knowledge. Evidence presented in Section II indicates that the durability of knowledge varies considerably from discipline to discipline. This information raises a number of questions concerning how the durability of knowledge influences career decisions. One influence considered here is that of a temporary career interruption. For example, if a woman experiences a break in her academic career, such as for the purpose of raising children, she will incur an opportunity cost in the sense that her research knowledge will to some extent become obsolete. Consequently, her accumulated stock of knowledge will decline relative to those professors who have continuous careers. The relative cost imposed on a woman who pursues dual family-career objectives is higher the less durable the jobrelated knowledge. This difference in relative costs is expected to alter career incentives. The results presented here suggest that women who pursue academic careers have responded to these incentives. Before discussing the evidence on female careers, I consider a more general model in which a professor is assumed to have a continuous career profile. This model, which is more appropriately viewed as a male model of investment behavior, is used to analyze how the durability of a professor's knowledge affects his research productivity over a career. Interestingly, the evidence suggests that a professor's response to less-durable knowledge is not to compact all learning early in his career, but rather to smooth learning over the career. Perhaps the most *Assistant professor, Arizona State University. I would like to thank an anonymous referee and many friends for helpful comments on earlier drafts. I am especially indebted to Finis Welch for valuable suggestions throughout the development of this paper. The views expressed are my own and I am responsible for any possible errors.