Publications in Peer-Reviewed Academic Journals
Immigration, Search, and Redistribution: A Quantitative Assessment of Native Welfare, with Gabriel Felbermayr, Giovanni Peri and Panu Poutvaara, 2017, Journal of the European Economic Association (forthcoming). NBER version available here. June 2017 version including Online Appendix available here.
We study the effects of immigration on native welfare in a general equilibrium model featuring two skill types, search frictions, wage bargaining, and a redistributive welfare state. Our quantitative analysis suggests that, in all 20 countries studied, immigration attenuates the effects of search frictions. These gains tend to outweigh the welfare costs of redistribution. Immigration has increased native welfare in almost all countries. Both high-skilled and low-skilled natives benefit in two thirds of countries, contrary to what models without search frictions predict. Median total gains from migration are 1.19% and 1.00% for high and low skilled natives, respectively.
High Wage Workers and High Wage Peers, Labour Economics, June 2017, Vol. 46, 47-63. Latest version here.
This paper investigates the effect of co-worker characteristics on wages, measured by the average person effect of coworkers in a wage regression. The effect of interest is identified from within-firm changes in workforce composition, controlling for person effects, firm effects, and sector-specific time trends. My estimates are based on a linked employer employee dataset for the population of workers and firms of the Italian region of Veneto for years 1982-2001. I find that a 0.1 increase in the average labour market value of co-workers’ skills (which is around one within-person standard deviation) is associated with a 3.6 percent wage premium. I also find that a sizeable share of the wage variation previously explained by unobserved individual and firm heterogeneity may be due to variation in co-worker skills. An event-type study, a Placebo exercise and a series of heterogeneity analyses lend credibility to the baseline results. I also evaluate the role of the spillover effects for wage differentials between specific groups of workers. I find that around 12 percent of the gender wage gap and 10 to 16 percent of the immigrant wage gap can be explained by differences in co-worker characteristics.
Individual Wage Growth: the Role of Industry Experience, Industrial Relations, January 2016, Vol. 55(1), 40-70. Latest version here.
This paper focuses on the effect of experience within an industry on wages. I use a correlated random effects simultaneous equation model that allows individual and match heterogeneity to affect wages, job tenure, and industry experience. I estimate my model separately for men and women using a large panel of young Italian workers for the years 1986–2004. Results show that wage returns to industry experience are much higher than wage returns to job seniority. The hypotheses of exogeneity of job seniority and industry experience in the wage equation are rejected: high-wage workers and high-wage matches last longer.
English as a Second Dialect Policy and Achievement of Aboriginal Students in British Columbia, with J. Friesen and B. Krauth, Canadian Public Policy, 40(2), June 2014, 182-192. Latest version available upon request.
Since the 1980s, the BC Ministry of Education has offered funding to support the language development of students who speak non-standard dialects of English. In practice, the students who are supported by this funding are almost exclusively Aboriginal, and English as a Second Dialect (ESD) funding has grown to be an important source of supplemental funding for Aboriginal students in many school districts. We exploit the staggered uptake of ESD funding by school districts to identify its effect on academic achievement. We find a sizable positive effect of ESD on grade seven reading achievement among Aboriginal students.
How Student Disability Classifications and Learning Outcomes Respond to Special Education Funding Rules: Evidence from British Columbia, with Jane Friesen and Ross Hickey, Canadian Public Policy, 38(2), June 2012, pp. 147-166. Latest version available upon request.
In 2002, British Columbia eliminated supplemental grants to school districts for some students with special needs. This study provides estimates of the response of special needs designations and academic performance to this funding change. Using student-level panel data, we find that students were less likely to receive a gifted, moderate behavioural disorder, or mild mental illness designation under the new funding rules. We study standardized test scores in grade 7, finding that the reading scores of gifted students declined substantially among those exposed to the new funding rules for the longest duration.
Non-Standard English Dialects and the Effect of Supplementary Funding on Educational Achievement, with M. Campbell, J. Friesen and B. Krauth, Canadian Journal of Speech-Language Pathology and Audiology, Vol. 35(2), 2011, pp.190-197. Available here.
Abstract: British Columbia provides school districts with supplementary funding to support the language development of students who speak a non-standard English dialect. Many of the students who attract this supplements are Aboriginal. We describe this policy, and record a striking increase in uptake of the funds on behalf of Aboriginal students over the last decade. We describe the results of an evaluation study that measured the effect of supplementary funding on test score gains between grades 4 and 7. The study found that the funding supplement substantially improved the reading scores of the average Aboriginal student.
Labor Supply within the Firm, with Ryan Michaels and Choonsung Park, Ifo Working Paper No. 222, September 2016. Revision Requested from AEJ Macro. Latest version: November 2017 available here.
Estimates of labor supply elasticities can be sensitive to the source of identifying variation. This paper’s model of production complementarities helps to interpret conflicting evidence. Complementarities attenuate working time adjustments to idiosyncratic, or individual-specific, variation in work incentives. Complementarities do not restrict, however, responses to firm-wide shocks; the latter is mediated by preference parameters. Estimating the model using matched firm-worker data, the paper disentangles production from preference parameters. The Frisch elasticity along the intensive margin is found to be around 0.5. A quasi-experimental approach, using idiosyncratic variation in work incentives, would find an elasticity less than half this.
Dynamic Effects of Co-Ethnic Networks on Immigrants’ Economic Success, with Giovanni Peri and Agnese Romiti, NBER Working Paper No. 22389, June 2016. Latest version: November 2017. Submitted.
This paper investigates how the size of co-ethnic networks at arrival affected the economic success of immigrants in Germany. Applying panel analysis with a large set of fixed effects and controls, we isolate the association between initial network size and long-run immigrant outcomes. Focusing on refugees – assigned to an initial location independently of their choice – allows a causal interpretation of the estimated coefficient. We find that immigrants initially located in places with larger co-ethnic networks are more likely to be employed at first, but have a lower probability of investing in human capital. In the long run they are more likely to be mis-matched in their job and to earn a lower wage.
The Labour Market Integration of Refugees in Germany: Evidence from a Field Experiment, with Yvonne Giesing and Nadzeya Laurentsyeva, November 2017. Latest version available here.
We design a field experiment to evaluate whether easing matching frictions affects the labour market integration of recent refugees in Germany. We interview around 400 job-seeking refugees who attend job-counselling sessions of a Munich-based NGO.
The participants are then randomly allocated to the treatment group and the control group. For the treatment group, the NGO identifies potentially suitable employers and, upon agreement of a job-seeker, sends a CV to those employers. This treatment can isolate the effect of frictions concerning the job search process, while it has no effect on the underlying skills of refugees. We track participants over time by conducting follow-up surveys every six months. Preliminary results show positive and significant treatment effects on employment after twelve months, albeit they are currently based on small sample sizes (we have completed about 25% of the second follow-up surveys). Working with the full dataset will allow us to investigate the robustness of these results, the heterogeneity of treatment effects across skill groups and legal status, and the possible tradeoff between obtaining employment earlier after arrival and future match quality.