This paper models growth via on-the-job learning when firms and workers are heterogeneous. It is an overlapping generations model in which young agents match with the old. More efficient assignments lead to faster long-run growth, more inequality, and less turnover in the distribution of human capital. Constant-growth paths are characterized for general functional forms and then, for the Cobb-Douglas case, the transition dynamics are solved analytically when the skill of the young is log-normally distributed and the initial human capital of the old generation is also log-normal. Growth and inequality move together on the transition to the balanced growth path.

This paper endogenizes the frequency of major discoveries and the extent of their refinement.Four axioms deliver a one-parameter family of beliefs that guide exploratory effort.Such effort trades off the prospect of major new discovery against the chance of successfully refining discoveries made in the past.The only other parameter is the cost of making new discoveries relative to the cost of refining old ones.The paper derives time-series properties of inventive activity as they relate to the two parameters, and it discusses several specific inventions and their subsequent refinement.In doing so, the paper arguably enhances our understanding of the process of discovery.1 We thank the C. V. Starr Center for Applied Economics for technical and financial assistance.The second

Investment of US firms responds asymmetrically to Tobin’s Q: investment of established firms—“intensive” investment—reacts negatively to Q whereas investment of new firms—“extensive” investment—responds positively and elastically to Q. This asymmetry, we argue, reflects a difference between established and new firms in the cost of adopting new technologies. A fall in the compatibility of new capital with old capital raises measured Q and reduces the incentive of established firms to invest. New firms do not face such compatibility costs and step up their investment in response to the rise in Q.

Firm numbers first rise, then later fall, as an industry evolves. This nonmonotonicity is explained using a competitive model in which innovation opportunities fuel entry and relative failure to innovate prompts exit; equilibrium time paths for price and quantity also share features of the data. The model is estimated using data from the U.S. automobile tire industry, a particularly dramatic example of the nonmonotonicity in firm numbers.

This paper develops a model of sectoral labor mobility and tests its main implications. The model nests two distinct hypotheses on the origin of mobility: (a) sectoral shocks and (b) worker-employer mismatch. We estimate the relative importance of each hypothesis and find that the bulk of labor mobility is caused by mismatch rather than by sectoral shift. We then try to put a value on society's match-specific information. That is, we ask to what extent the availability of the option to change jobs raises GNP. We find that the mobility option raises expected earnings by roughly between 8.5 percent and 13 percent of labor earnings, which translates to an increase in GNP of between 6 percent and 9 percent.