top of page
Matt Marx
Working papers
“Firm Failure, Scandal, and Employee Exit From Innovation-Driven Industries." (w/Tristan Botelho)
Almost 4 million US workers lose their jobs each year after their employers fail, but organizational research only addresses the career conse-
quences of failure for a fraction of these ex-employees: executives. One might assume that the deleterious effects of failure for the executives who led these organizations would generalize to non-executives, but there are reasons to question this inferential leap. Bringing together organizational research on careers, evaluations, and labor markets we theorize about the conditions under which employees may be affected by their employer’s failure. We test these expectations using two complementary data sets (i) confidential US Census data for all workers in 26 states from 1991-2014 and (ii) a hand-collected sample of nearly 14,000 workers in the innovation-driven industry of Automatic Speech Recognition (ASR). In the Census, we replicate the negative impact of employer failure for executives from prior research. However, we find the opposite for non-executives, those affiliated with a failed firm earn higher wages than those who voluntarily leave ongoing firms. The detailed data available in ASR further demonstrates that non-executive employees go to more innovative firms after failure.
"Startups, Unicorns, and the Local Supply of Inventors." (w/Ben Balsmeier, Lee Fleming, and Ryan Shin)
We provide causal evidence that the local supply of technical human capital improves regional entrepreneurship, both by increasing entry and by reducing failure rates of new ventures. The arrival of inventors in a county shifts the locus of venture capital investment away from low-tech startups to high-tech startups and moreover towards new ventures in the same sector as those inventors’ skills. Identification is provided by a shift-share instrument combining the entire spatial distribution of surnames in the 1940 U.S. Census with thousands of surname-specific shocks based on modern inventor mobility.
"(When) does patent protection spur cumulative research within firms?" (w/Ashish Arora, Sharon Belezon, and Dror Shvadron)
We estimate the effect of patent protection on follow-on investments in corporate scientific research. We exploit a new method for identifying an exogenous reduction in the protection a granted patent provides. Using data on public, research-active firms between 1990 and 2015, we find that firms respond to a decrease in patent protection by reducing follow-on research, as measured by a drop in internal citations to an associated scientific article. We study this effect across firms with varying commercialization capabilities and across varying thickness levels of markets for technology. We find that the effect is stronger in technologies where patents are traded less frequently. Our findings are consistent with a stylized model whereby patent protection is a strategic substitute for commercialization capability. Our results imply that stronger patents encourage follow-on research, but also shift the locus of research from big firms toward smaller firms and startups. As patent protection has strengthened since the mid-1980s, our results help explain why the American innovation ecosystem has undergone a growing division of innovative labor, where startups become important sources of new ideas.
"Field-Specific Investments in Human Capital & Path Dependence in Early Career Choices: Field Experimental Evidence from Engineering." (w/ Kevin Boudreau)
Young people often make field-specific investments in education, certification, and credentials before entering into practice and more fully understanding their "fit" with their chosen field. In this paper, we investigate whether these sunk investments and uncertainty in making these choices end up shaping career choices to a significant degree. We report on a large-scale field experiment involving 2,072 Engineering students, across multiple cohorts, at a large U.S.
university. Individuals were randomly assigned to receive exposure to a 6-month professional work experience in their chosen field. Individual were randomly assigned to receive this exposure either in the second or third year of the program. Consistent with high degrees of career uncertainty, we find that those doubled their rate of changing careers to about 9 percentage points immediately after being exposed to professional work, relative to a control group
who had not yet received the exposure. Consistent with prior sunk investments discouraging changes, those given the later exposure, after investing in greater prior course work, were 3 percentage points less likely to change careers after the exposure, or a full third less. Overall, changing to fields with higher wages and towards areas where individuals have made greater prior investments in relevant courses. The results are consistent with (very) large information frictions shaping career choices among young people in higher education.
bottom of page