Conrad Miller, University of California, Berkeley and NBER; Jennifer Peck, Swarthmore College; and Mehmet Seflek, University of California, Berkeley

Big Push Policies and Firm-Level Barriers to Employing Women: Evidence from Saudi Arabia

Miller, Peck, and Seflek study firm demand for female labor in Saudi Arabia, a country where laws and norms regulate interactions between men and women and female labor force participation is among the world's lowest. In 2011, Saudi Arabia imposed gender-neutral quotas for Saudi employment at private sector firms. The researchers find this policy more than tripled the female share of Saudis working in the private sector within four years, with this increase concentrated at firms that were previously all-male. Motivated by these findings, the researchers examine whether the potential fixed costs that Saudi firms must pay to employ both men and women affect firm personnel decisions. Fixed integration costs can potentially lead to multiple equilibria for female employment: firms only integrate if enough women are in the labor force, and women only join the labor force if enough firms can employ women. Consistent with fixed integration costs, the researchers find that firms are highly segregated with few firms employing small numbers of women, the distribution of female hiring across firms is similarly 'lumpy', and this pattern is concentrated at previously all-male firms, and that employment quotas lead to a disproportionate and persistent increase in female hiring, and this increase is concentrated at previously all-male firms. The researchers also find that the quota policy induces integration at firms where quotas were not binding, consistent with demand externalities or changing norms. The results suggest an important role for 'big push' demand-side policies in increasing female employment.

---

Randall Akee, University of California, Los Angeles and NBER, and Maggie R. Jones, U.S. Census Bureau

Immigrants' Earnings Growth and Return Migration from the U.S.: Examining their Determinants using Linked Survey and Administrative Data

Using a novel panel data set of recent immigrants to the U.S. (2005-2007) from individual-level linked U.S. Census Bureau survey data and Internal Revenue Service (IRS) administrative records, Akee and Jones identify the determinants of return migration and earnings growth for this immigrant arrival cohort. They show that by 10 years after arrival almost 40 percent have return migrated. The researchers' analysis examines these flows by educational attainment, country of birth, and English language ability separately for each gender. They show, for the first time, that return migrants experience downward earnings mobility over two to three years prior to their return migration. This suggests that economic shocks are closely related to emigration decisions -- time-variant unobserved characteristics may be more important in determining out-migration than previously known. The researchers also show that wage assimilation with native-born populations occurs fairly quickly -- after 10 years there is strong convergence in earnings by several characteristics. Finally, the researchers confirm that the use of stock-based panel data lead to estimates of slower earnings growth than is found using repeated cross-section data. However, they also show, using selection-correction methods in their panel data, that stock-based panel data may understate the rate of earnings growth for the initial immigrant arrival cohort when emigration is not accounted for.

---

François Gerard, Columbia University and NBER; Lorenzo Lagos, Columbia University; Edson R. Severnini, Carnegie Mellon University; and David Card, University of California, Berkeley and NBER

Assortative Matching or Exclusionary Hiring? The Impact of Firm Policies on Racial Wage Differences in Brazil

A growing body of research shows that firms' employment and wage-setting policies contribute to wage inequality and pay disparities between groups. Gerard, Lagos, Severnini, and Card measure the effects of these policies on racial pay differences in Brazil. They find that nonwhites are less likely to work at establishments that pay more to all race groups, a pattern that explains about 20% of the white-nonwhite wage gap for both genders. The pay premiums offered by different employers are also compressed for nonwhites relative to whites, contributing another 5% of the overall gap. The researchers then ask how much of the under-representation of nonwhites at higher-paying workplaces is due to the selective skill mix at these establishments. Using a counterfactual based on the observed skill distribution at each establishment and the nonwhite shares in different skill groups in the local labor market, the researchers conclude that assortative matching accounts for about two-thirds of the under-representation gap for both men and women. The remainder reflects an unexplained preference for white workers at higher-paying establishments. The wage losses associated with unexplained sorting and differential wage setting are largest for nonwhites with the highest levels of general skills, suggesting that the allocative costs of race-based preferences may be relatively large in Brazil.

---

Shai Bernstein and Rebecca Diamond, Stanford University and NBER, and Timothy McQuade and Beatriz Pousada, Stanford University

The Contribution of High-Skilled Immigrants to Innovation in the United States

Bernstein, Diamond, McQuade, and Pousada characterize the contribution of immigrants to US innovation, both through their direct productivity as well as through their indirect spillover effects on their native collaborators. To do so, the researchers link patent records to a database containing the first five digits of 160 million of Social Security Numbers (SSN). By combining this part of the SSN together with year of birth, the researchers identify whether individuals are immigrants based on the age at which their Social Security Number is assigned. The researchers find that over the course of their careers, immigrants are more productive than natives, as measured by number of patents, patent citations, and the economic value of these patents. Immigrant inventors are more likely to rely on foreign technologies, to collaborate with foreign inventors, and to be cited in foreign markets, thus contributing to the importation and diffusion of ideas across borders. Using an identification strategy that exploits premature inventor deaths, the researchers find that immigrant collaborators create especially strong positive externalities on the innovation production of natives, while natives create especially large positive externalities on immigrant innovation production, suggesting that combining these different knowledge pools into inventor teams is important for innovation. A simple decomposition suggests that despite immigrants only making up 16% of inventors, they are responsible for 30% of aggregate US innovation since 1976, with their indirect spillover effects accounting for more than twice their direct productivity contribution.

---

David J. Deming, Harvard University and NBER, and Kadeem L. Noray, Harvard University

STEM Careers and Technological Change (NBER Working Paper No. 25065)

Science, Technology, Engineering, and Math (STEM) jobs are a key contributor to economic growth and national competitiveness. Yet STEM workers are perceived to be in short supply. This paper shows that the "STEM shortage" phenomenon is explained by technological change, which introduces new job tasks and makes old ones obsolete. Deming and Noray find that the initially high economic return to applied STEM degrees declines by more than 50 percent in the first decade of working life. This coincides with a rapid exit of college graduates from STEM occupations. Using detailed job vacancy data, the researchers show that STEM jobs changed especially quickly over the last decade, leading to flatter age-earnings profiles as the skills of older cohorts became obsolete. The findings highlight the importance of technology-specific skills in explaining life-cycle returns to education, and show that STEM jobs are the leading edge of technology diffusion in the labor market.

---

Luigi Pistaferri, Stanford University and NBER, and Hamish Low, University of Cambridge

Disability Insurance and Gender Differences: Evidence from Merged Survey-Administrative Data

Using HRS data matched with Social Security administrative data, Pistaferri and Low document large gender differences in disability insurance programs admission rates and type I error rates. In particular, women who apply for DI/SSI are 13 percentage point less likely to be awarded benefits than men, controlling for health, occupation and a host of demographic characteristics. Moreover, women who self-report to be disabled are 20 percentage points more likely to be rejected than observationally similar men. The researchers investigate whether these gender differences can be explained by heterogeneity in underlying unobserved health, differences in disability perceptions, higher noise-to-signal ratios, or SSA evaluators' assessment bias. They find little support for the first three explanations, and some indirect support for the latter.

---

Alisa Tazhitdinova, University of California, Santa Barbara

Increasing Hours Worked: Moonlighting Responses to a Large Tax Reform

Holding multiple jobs -- or moonlighting -- is increasingly popular in OECD countries, with 5 to 10% of workers holding two or more jobs. Yet, little is known about the determinants of moonlighting and its responsiveness to financial incentives: research has been held back by the lack of identifying variation as most policies treat primary and secondary employments equally. Tazhitdinova circumvents these limitations by studying a unique reform in Germany that allowed workers to hold small secondary jobs tax-free, thus decreasing the tax rate on secondary earnings by 21-74 percentage points. Using a difference-in-differences framework, Tazhitdinova documents three findings. First, Tazhitdinova estimates large participation elasticities that are several times larger than participation elasticities in primary employment. Second, Tazhitdinova shows that individuals did not substitute primary earnings with secondary jobs, despite large potential savings. Third, the number of small-earnings jobs increased rapidly after reform, and did not result in decreased labor supply among low-income individuals. Finally, Tazhitdinova explores mechanisms behind the varying rates of response, and find that job availabilities and hour constraints are key determinants of moonlighting.

---

Brigham Frandsen and Emily C. Leslie, Brigham Young University, and Lars Lefgren, Brigham Young University and NBER

Judging Judge Fixed Effects

Frandsen, Lefgren, and Leslie propose a test for the identifying assumptions invoked in designs based on random assignment to one of many "judges." They show that standard identifying assumptions imply that the conditional expectation of the outcome given judge assignment is a continuous function with bounded slope of the judge propensity to treat. The implication leads to a two-part test that generalizes the Sargan-Hansen overidentification test and assesses whether implied treatment effects across the range of judge propensities are possible given the domain of the outcome. The researchers show the asymptotic validity of the testing procedure, demonstrate its finite-sample performance in simulations, and apply the test in an empirical setting examining the effects of pre-trial release on defendant outcomes in Miami. When the assumptions are not satisfied, the researchers propose a weaker average monotonicity assumption under which IV still converges to a proper weighted average of treatment effects.