Skip to main content
SpringerLink
Log in

Where Do New Ph.D. Economists Go? Recent Evidence from Initial Labor Market

  • Published:
Journal of Labor Research Aims and scope Submit manuscript

Abstract

We collect data on the 2007–2008 Ph.D. economist job market to investigate initial job placement in terms of job location, job type, and job rank. While there is little gender difference in all three dimensions, our results suggest significant source country heterogeneity in placement outcomes. In an analysis linking job location and job type, we find that, among non-U.S. candidates, foreign placements are more likely to be academic relative to U.S. placements. Our analysis contributes to the literature in two aspects: First, compared to existing studies, our sample consists of all job market candidates from 57 top U.S. economics programs and allows us to conduct an analysis more immune to selection bias. Second, with the increasing presence of international students in the U.S. doctoral programs, we examine a new and growing dimension of the labor market – the international perspective of initial job placements for new Ph.D. economists.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

Notes

  1. Source: http://www.nsf.gov/statistics/showpub.cfm?TopID=2&SubID=25.

  2. Coupe (2004) and Ehrenberg (2004) provide excellent surveys of this literature. Other studies focus on the success of completing the Economics Ph.D. (Siegfried and Stock 2001; Stock and Siegfried 2006a; Grove et al. 2007).

  3. For example, Krueger and Wu (2000), Oyer (2006), Athey et al. (2007), and Grove and Wu (2007) use a handful of top programs. McMillen and Singell (1994) and McDowell et al. (1999, 2001) use AEA membership directory data. Most other studies use survey data, including Barbezat (1992), Neumark and Gardecki (1998), Siegfried and Stock (1999, 2004), Stock and Siegfried (2001), and Gallet et al. (2005).

  4. The existing literature studies the role of gender extensively, but not the role of source country heterogeneity. When country heterogeneity is considered, typically a single country dummy (i.e., the U.S.) is used. There are few exceptions (Krueger and Wu 2000; Grove et al. 2007) using a combination of country and continent dummies. Our analysis provides a more careful exploration across many countries.

  5. In existing studies, job location, if considered, usually refers to regions within the U.S.

  6. Note that this gender difference does not necessarily reveal gender discrimination. For example, Barbezat (1992) documents significant gender differences in their preferences for jobs types. On the other hand, McMillen and Singell (1994) find no significant gender difference in sector choice probabilities (employee preference), but the marginal effect of several productive attributes is smaller for women than for men (employer discrimination).

  7. Results in Barbezat (1992) Table 6 indicates significant gender differences in how they value certain workplace characteristics.

  8. Aside from initial job placements, intermediate job search outcomes are also analyzed in this literature (Stock and Alston 2000; Gallet et al. 2005). Another strand of the literature analyzes later career outcomes. Much of the focus is on gender gap, for example, Kahn (1995), McDowell et al. (1999, 2001), Ginther and Hayes (2003), and Kolpin and Singell (1996).

  9. Several studies examine related issues in other academic markets. For example, Zamudio et al. (2011) study matching outcomes between candidates and departments in marketing, and Ginther and Hayes (2003) identify significant gender differences in promotion for humanities professors which result in gender differences in salary.

  10. It is possible that a listed job market candidate may later choose not to proceed and stay in the program one more year. Our data seem to confirm this possibility. It is also possible but less likely that a late candidate may enter into the market after mid-November for jobs starting the following year.

  11. Table 10 in Appendix reports the detailed distribution of country origin, female, and job placement location by country/region.

  12. Refer to Appendix Table 10 for detailed gender composition by country/region.

  13. These two outcomes coincide with each other for U.S. candidates. For candidates from other continents, they are not perfectly correlated since a foreign student can be placed into a country outside U.S. and his/her home continent.

  14. The results are qualitatively the same if we exclude temporary jobs.

  15. Our results suggest that candidates from Tier 4 programs are more likely to stay in the U.S. than those from Tier 2 and 3 programs. That is, the relationship between job location and program ranking is U-shaped. One may speculate that foreign placements are better fits for candidates in the two middle tiers than the top and bottom tiers.

  16. In addition to this baseline specification, we also introduce interaction terms to investigate the gender difference within a specific country and within a program tier. Our main results remain qualitatively unchanged. Results are available from the authors upon request.

  17. There is potential endogeneity problem using stayus as the explanatory variable as unobserved heterogeneity in the candidates may affect both decisions to choose between academic and non-academic and to choose between stay and return. However, with only a single year data, it is difficult to find good instrument variables to account for this endogeneity issue.

  18. This can be easily confirmed by looking at the raw data of Korean candidates. 11 of the 13 who stay in the U.S. find jobs in academia and only 2 go to the private sector. For those who leave the U.S., 3, 8 and 11 are placed to academia, government, and private sector, respectively.

  19. We also run multinomial logit estimation similar to that in Section “Job Type” but with variable stayus and country-stay interaction terms. The results are qualitatively similar to the probit estimation results here. The estimate for stayus is always negative but may be insignificant. The country-stay interaction terms are all negative except for Korea. Details are available upon request.

  20. Our job rank analysis follows Krueger and Wu (2000) closely. Other studies have adopted more crude measure of job rank to compare gender differences. For example, McMillen and Singell (1994) divide placements into top 50 economics department and other departments. Barbezat (1992) divide employment into top 15, 16–30, and other economics departments, respectively.

  21. For example, the previous year’s GDP of each placement country might be appropriate to use as an IV in a multi-year analysis. One would assume that the demand for new Ph.D. economists in a country, to some extent, depends on its GDP from the previous year, especially in the U.S. where these graduates are produced.

  22. The list of top 1000 economists is available upon request (Source: http://student.ulb.ac.be/~tcoupe/update/top1000p.html. Accessed February, 2009).

  23. The full list of the journal ranking is available upon request.

  24. Some institutions offer economics programs both in the business school and the college of arts and sciences. We treat them as different programs but give them the same rank. The list of all programs included in our sample is available upon request.

References

  • Athey S, Katz L, Krueger A, Levitt S, Poterba J (2007) What does performance in graduate school predict? Graduate economics education and student outcomes. Am Econ Rev 97(2):512–518

    Article  Google Scholar 

  • Barbezat D (1992) The market for new PhD economists. J Econ Educ 23(3):262–276

    Article  Google Scholar 

  • Broder I (1993) Professional achievements and gender differences among academic economists. Econ Inq 31(1):116–127

    Article  Google Scholar 

  • Buchmueller T, Dominitz J, Hansen W (1999) Graduate training and the early career productivity of PhD economists. Econ Educ Rev 14:65–77

    Article  Google Scholar 

  • Cawley J (2011) A Guide (and Advice) for economists on the US junior academic job market

  • Ehrenberg R (2004) Econometric studies of higher education. J Econ 121:19–37

    Google Scholar 

  • Coupe T (2004) What do we know about ourselves? On the economics of economics. KYKLOS 57:197–216

    Article  Google Scholar 

  • Gallet C, List J, Orazem P (2005) Cyclicality and the labor market for economists. South Econ J 72(2):284–304

    Article  Google Scholar 

  • Ginther D, Hayes K (2003) Gender differences in salary and promotion for faculty in the humanities 1977–95. J Human Res 38(1):34–73

    Article  Google Scholar 

  • Grove W, Dutkowsky D, Grodner A (2007) Survive then thrive: determinants of success in the economics PhD program. Econ Inq 45(4):864–871

    Article  Google Scholar 

  • Grove W, Wu S (2007) The search for economics talent: doctoral completion and research productivity. Am Econ Rev 97(2):506–511

    Article  Google Scholar 

  • Hilmer C, Hilmer M (2007) Women helping women, men helping women? Same-gender. Mentoring, initial job placements, and early career publishing. Am Econ Rev 97(2):422–426

    Article  Google Scholar 

  • Kahn S (1995) Women in the economics profession. J Econ Perspect 9(4):193–206

    Article  Google Scholar 

  • Kalaitzidakis P, Mamuneas T, Stengos T (2003) Rankings of academic journals and institutions. University of Guelph, Department of Economics

  • Kolpin V, Singell Jr L (1996) The gender composition and scholarly performance of economics departments: a test for employment discrimination. Ind Labor Relat Rev 49(3):408–423

    Article  Google Scholar 

  • Krueger A, Wu S (2000) Forecasting job placements of economics graduate students. J Econ Educ 31(1):81–94

    Google Scholar 

  • McDowell J, Singell Jr L, Ziliak J (1999) Cracks in the glass ceiling: gender and promotion in the economics profession. Am Econ Rev 89(2):392–396

    Article  Google Scholar 

  • McDowell J, Singell Jr L, Ziliak J (2001) Gender and promotion in the economics profession. Ind Labor Relat Rev 54(2):224–244

    Article  Google Scholar 

  • McMillen D, Singell Jr L (1994) Gender differences in first jobs for economists. South Econ J 60(3):701–704

    Article  Google Scholar 

  • Moore WJ, Newman RJ (1977) An analysis of the quality differentials in male-female academic placements. Econ Inq 15(3):413–434

    Article  Google Scholar 

  • Neumark D, Gardecki R (1998) Women helping women? Role model and mentoring effects on female PhD students in economics. J Human Res 33(1):220–246

    Article  Google Scholar 

  • Oyer P (2006) Initial labor market conditions and long-term outcomes for economists. J Econ Perspect 20(3):143–160

    Article  Google Scholar 

  • Siegfried J, Stock W (1999) The labor market for new PhD economists. J Econ Perspect 13(3):115–134

    Article  Google Scholar 

  • Siegfried J, Stock W (2001) So you want to earn a PhD economics: how long do you think it will take? J Hum Resour 36(2):364–378

    Article  Google Scholar 

  • Siegfried J, Stock W (2004) The market for new PhD economists in 2002. Am Econ Rev 94(2):272–285

    Article  Google Scholar 

  • Stock W, Alston R (2000) Effect of graduate-program rank on success in the job market. J Econ Educ 31(4):389–401

    Google Scholar 

  • Stock W, Siegfried J (2001) So you want to earn a PhD economics: How much do you think you’ll make? J Human Res 36(1):320–335

    Google Scholar 

  • Stock W, Siegfried J (2006a) Time-to-degree for the economics PhD class of 2001–2002. Am Econ Rev 92(2):467–474

    Article  Google Scholar 

  • Stock W, Siegfried J (2006b) Where are they now? Tracking the PhD class of 1997. South Econ J 73(2):473–488

    Article  Google Scholar 

  • Zamudio C, Wang Y, Haruvy E (2011) Human brands and mutual choices: An investigation of the Marketing assistant professor job market. Working paper

Download references

Acknowledgments

We thank Leah Brooks, Greg Burge, Jen Graves, John Ham, Cory Koedel, Catherine Mooney, Rati Ram, Adam Rennhoff, and session participants at the 2008 Southern Economic Association Annual Conference and Illinois State University Applied Econometrics Workshop participants for helpful comments and suggestions. We also thank Meredith Willinger, Yoonho Choi, Chris Denly, Hojin Jung, Golaleh Moshrefi, and Collin Phillips for their excellent research assistances, and University of Oklahoma Honors Research Assistant Program for financial support. Finally, we would like to thank placement coordinators and other individuals at various institutions who exchanged correspondences with us during the course of this project. The usual caveat applies.

Author information

Authors and Affiliations

  1. Department of Economics, Illinois State University, Normal, IL, 61790, USA

    Jihui Chen & Sherrilyn Billger

  2. Department of Economics, University of Oklahoma, Norman, OK, 73019, USA

    Qihong Liu

Authors
  1. Jihui Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Qihong Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Sherrilyn Billger
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Jihui Chen.

Appendix

Appendix

Variable Definitions

  1. (1)

    Demographic characteristics.

    • Gender: We define a dummy variable, female, as 1 if the candidate is a female and 0 otherwise.

    • Country/region: We define a set of country/region dummy variables as 1 if the candidate comes from that country/region and 0 otherwise. They include Argentina, China, India, Italy, Japan, Korea, Russia, Taiwan, Turkey, US.

  1. (2)

    Academic characteristics.

    • Additional Master Degree: We define a dummy variable, addmaster, as 1 if the candidate earned a master’s degree prior to entering doctoral training and 0 otherwise.

    • Advisor: To control for the effectiveness of recommendation letters, we refer to Tom Coupe’s index of top economists.Footnote 22 We define a dummy variable, topadvisor, as 1 if the advisor and/or the co-chair is ranked top-50 economists worldwide and 0 otherwise. The underlying assumption is that the more prominent the letter writer is, the recommendation letter will likely yield better job placement (Grove and Wu 2007).

    • Top Publications: We use the journal ranking by Kalaitzidakis et al. (2003) to define two top journal variables.Footnote 23 First, following Grove and Wu (2007), we use top4journal, defined as 1 if the candidate has at least one of the top four economics journals (i.e., American Economic Review (AER) (excluding Papers and Proceedings), Quarterly Journal of Economics (QJE), Journal of Political Economy (JPE), and Econometrica), and 0 otherwise. There are only 8 candidates with at least one top four journal publication at the time of application in the sample. Second, we define top5–50j, which is defined as one if the candidate has at least one publication in a journal ranked between 5 and 50. A total of 29 candidates have at least such a publication.

    • Revise and Resubmit (R&R): To be consistent with the variables used for publications, we define two similar quality indicators for R&Rs, using the Kalaitzidakis et al. (2003) ranking. First, we define a dummy variable, top4rr, as one if the candidate has at least one revision and resubmission in a top four journal, and zero otherwise. Second, we define a dummy variable, top5-50rr, as one if the candidate has at least one R&R in a journal ranked between 5 and 50. At the time of job application in the sample, 3 of them have a top-four R&R, and 18 have a R&R in journals ranked 5 through 50.

    • Teaching Awards: This variable measures the number of teaching awards that the candidate receives while in the Ph.D. program.

    • Program Ranking: Following Buchmueller et al. (1999), we define four dummy variables for graduate program tiers. Tier 1 (tier1), Tier 2 (tier2), Tier 3 (tier3) and Tier 4 (tier4) refer to programs ranked 1 to 6, 7 to 15, 16 to 30, and beyond 30, respectively.Footnote 24 The variable tier4 is omitted as the reference category in the analysis.

    • Program Size: We define size as the total number of candidates in the program on the same job market. This variable indicates competition within the program. For instance, in larger programs, candidates face more competition from their peers with the same degree credentials. On the other hand, a larger program is more likely to offer candidates a wider network in the job market, likely facilitating their job search.

  2. (3)

    Placement Characteristics

    • Job Location: We define two variables regarding job location. Stayus is defined as 1 if the student accepts a job in the U.S. and 0 otherwise. Return is defined as 1 if the student accepts a job in his/her home continent and 0 otherwise. We only include foreign candidates when considering job location.

    • Job Type: We divide all jobs into four types: Academia, Government, Private Sector, and Temporary Job (i.e., a visiting position or a postdoc position). Academia, Gov’t,Private,Temp is defined as one if a person is placed in academia, Government, Private Sector, and Temporary Job, respectively, and zero otherwise.

      In Section 5.3.2, academicjob is defined as 1 if a foreign candidate is placed into academia and 0 if s/he is placed into the government or the private sector.

    • Job Ranking: We use the Kalaitzidakis et al. (2003) ranking to rank academic jobs from 1 to 200 (the lower the number, the better the job is ranked). A placement in a business school is given an extra five ranks, or minus five. Using a methodology similar to the one used in Krueger and Wu (2000), we give the rank of 40 to top government jobs (e.g., IMF, World Bank, and Federal Reserve Banks) and the rank of 120 to top consulting jobs (e.g., Abt, NBER, and Rand). For all the remaining jobs, we give the rank of 300. As discussed in Krueger and Wu (2000), such a ranking system is highly subjective but as we will show later, our findings are robust to different ranking systems. Therefore, candidates with unranked jobs are treated as censored observations, with the censoring point at the rank of 300.

Table 9 List of economics programs in the sample
Full size table
Table 10 Distribution of country origin, female, and placement by country/region
Full size table

Rights and permissions

Reprints and permissions

About this article

Cite this article

Chen, J., Liu, Q. & Billger, S. Where Do New Ph.D. Economists Go? Recent Evidence from Initial Labor Market. J Labor Res 34, 312–338 (2013). https://doi.org/10.1007/s12122-013-9162-4

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s12122-013-9162-4

Keywords

JEL Classification

Search

Navigation