Expanding opportunities: Economics PhD placement in the 21st century

Since 2010, a growing proportion of doctoral graduates have been taking jobs outside academia. This column explores where elite early-career economists work, using a dataset covering all PhD graduates from US economics programmes. The findings revealt that recent elite PhD graduates are more likely to leave academia, to take jobs outside the US, and to be absorbed by a tech sector that draws broadly across gender, ethnicity, and programme prestige. However, East Asian candidates continue to face a substantial penalty in access to top academic positions. Within academia, elite advisor networks continue to confer advantages, reinforcing stratification at the top.

In most scientific disciplines, award-winning scholars work in an expanding number of universities and research centres, leading to broader diffusion of knowledge, greater diversity of perspectives, and more competition among institutions. At the same time, Freeman et al. (2024) have documented that elite researchers in economics are increasingly concentrated in a small number of institutions and trained in a narrow set of elite PhD programmes. Deming (2024) argues that these findings reflect the growing influence of institutional prestige in research evaluation, with “the marketplace of ideas in economics… becoming less efficient and fair” and the profession becoming “insular and status-obsessed”. 

While existing research has focused on established, award-winning academics, much less is known about where elite early-career economists work, particularly outside academia. If talent is increasingly allocated based on institutional prestige, is the profession becoming concentrated within a small network of institutions that both employ leading scholars and recruit each other’s graduates? If so, are economists becoming increasingly disconnected from the real world?

To answer these questions, in Eberhardt et al. (2026), we assemble a novel dataset covering all PhD graduates from the top-30 US economics programmes between the late 1990s and the early 2020s, combining detailed information on both students and their advisors. They show that career paths have become more diverse over time, although institutional prestige and professional networks continue to play an important role in shaping the careers of young economists.

Top PhD graduates are less likely to take jobs in universities and research institutions

Figure 1 plots the evolution of academic placement, documenting an inverted U-shape pattern. At its peak in 2009, around 2/3 of fresh graduates took a job at a university. Since then, the share has decreased significantly, dropping to 52% in 2021 (all figures are 3-year moving averages). With the exception of 2007–2012, the patterns are very similar for male and female candidates (panel A). This decline confirms trends documented for all PhD programmes up to the mid-2010s by Foster et al. (2023).

Figure 1 Academic placement of economics PhDs, by gender and school rank

Splitting the sample by rank of the PhD-granting institution (panel B) reveals starker differences: academic placement is significantly higher for graduates from top-10 institutions (by 3 to 15 percentage points); in addition, while for top-10 programmes, graduate placement in academia increased until 2018, it already started to decline from the early 2010s for institutions ranked 11–30. 

Academic placements are becoming more international

Graduates of top US institutions are increasingly seeking academic employment overseas (Figure 2, panel A). From a peak of 75% in the early 2000s, placement in North American institutions has decreased to 50% by the early 2020s. European and East Asian institutions have become increasingly important destinations. Additional analysis shows that the decline in North American placement is especially pronounced for graduates from institutions ranked 11–30. Furthermore, while graduates from top-10 programmes experience a larger increase in European placements than to East Asia, the reverse is true for those from institutions ranked 11–30. Importantly, notwithstanding increased geographic diversification, large swathes of the Emerging South, such as India and parts of Africa, remain on the sidelines of the job market. 

Figure 2 Academic placement of economics PhDs, by region and sector

Tech fuels growth in private sector placement

The share of graduates taking jobs outside academia has increased since 2010 (Figure 1). We study four main non-academic destinations: the public sector, consulting, finance and industry, and tech (Figure 2, panel B). Although the public sector remains the largest employer, its share fell from about 25% in the early 2000s to below 20% by the early 2020s. Consulting and finance have remained broadly stable, accounting for around 10% and 6%–7% of placements, respectively. By contrast, the tech sector has expanded rapidly, growing from just 1.3% of placements in 2010 to 12% by the early 2020s.

Figure 3 uses word clouds to provide a more granular view of graduate placement outside academia, with the vertical axis capturing the employment share of each company or institution within that broad sector. Finance is very competitive, with many equally sized recruiters. There is more concentration in the public sector and in consulting. The IMF, the World Bank, and the US Federal Reserve each capture a significant share of the former market. Similarly, Bates White, Charles River Associates, and particularly Analysis Group and Cornerstone dominate placement in consultancies. These levels of concentration are however dwarfed by those in the tech sector, where a single company – Amazon – accounts for over 42% of all PhD placements.

Figure 3 Graduate placement outside academia, by sector

Drivers of placement inside and outside of academia

Finally, we investigate the correlates of placement outcomes in and outside academia. The results are presented in Figure 4 – with column (A) considering the entire sample and column (B) focusing on the most recent decade. Using a multinomial logit model, we analyse characteristics of the candidates (top panels), as well as of their advisor team (bottom panels). The markers in each row report the marginal effects of each characteristic – relative to the comparison group – on the probability of placement in and out of academia, with transparent coefficients indicating lack of statistical significance.

Figure 4 Candidate and advisor characteristics and placement

Among candidate characteristics, gender has little effect on placement. Comparing women to men, small differences appear only in consulting (more women) and in the more recent sample in finance and industry (fewer women). Larger differences emerge by ethnicity. Relative to whites, East Asians are less likely to place in top-50 institutions and more likely to end up in lower-ranked ones. These gaps have become larger in the recent period (see panel B). Academic placement gaps for South Asians and Hispanics are smaller. Relative to whites, Hispanics are the only ethnic group with a strong propensity for employment in the public sector. Symmetrically, Middle Easterners are the least likely to find employment in that sector, both in the entire sample and in the recent period. A striking feature that emerges from this analysis is that tech is the only sector entirely unaffected by any candidate characteristics. 

Advisor characteristics matter much more for top academic placements. Students advised by Nobel laureates or top-five journal editors are more likely to join leading universities. Similarly, candidates whose advisors have stronger academic networks, measured by top-10 institutional backgrounds and co-authorship connections, are more likely to secure top academic positions and less likely to join lower-ranked institutions. Since these effects remain after controlling for advisor prestige, they point to an independent role for academic networks. Taken together, these results point to advisor teams as a meaningful determinant of top academic placement, operating through both candidate quality and network reach.

Advisor quality and networks have little effect on most non-academic careers. They are largely unrelated to public-sector placements and are weakly associated with consulting and finance. The main exception is tech: students advised by Nobel laureates are significantly more likely to enter that sector, especially in recent years. This suggests that tech has become an increasingly attractive destination for top young economists, the same who are likely to secure jobs in leading academic institutions. 

Conclusions

Our analysis offers a qualified reassessment of Deming’s insularity hypothesis. The aggregate trends point toward decreasing insularity: recent cohorts of elite PhD graduates are more likely to leave academia, more likely to take jobs outside the US, and increasingly to be absorbed by a tech sector that draws broadly across gender, ethnicity, and programme prestige. In this sense, the ‘exclusive circle’ of elite academic institutions is not fully closed and might even be opening up. 

Yet, the disaggregated picture is less reassuring. East Asian candidates continue to face a substantial penalty in access to top academic positions. In addition, within academia, elite advisor networks continue to confer advantages, reinforcing stratification at the top of the profession. The marketplace of ideas may be widening at the edges; at its centre, prestige and networks continue to shape careers.

Source : VOXeu