What Is Actually Changing?
The subject is the changing first-placement market for economics PhDs, and the question is whether fewer academic placements mean a more open profession or merely a different distribution of advantage. Taken together, the evidence supports a conditional answer. The map of early-career destinations is widening across sectors and regions, but access to the most prestigious academic positions remains stratified, and one of the most visible new routes is concentrated in a single employer.
The answer is a reallocation, not a retreat
The shortest accurate answer to the title is that economics PhDs are not simply leaving academia. In the IZA study’s selected group of leading U.S. programs, academia’s share of first placements rises to about 67% in 2009 and falls to 52% in 2021, while an average of 42% of graduates across the period take jobs outside academia. That is a substantial reallocation of early-career destinations. It is not evidence that academic work has disappeared, that every graduate prefers a company job, or that a nonacademic first placement ends a research career.
The reallocation is meaningful because the nonacademic category contains several routes. Public-sector, consulting, finance-and-industry, and technology placements have different organizational settings and different trade-offs involving research autonomy, compensation, stability, location, and later mobility. Technology is the clearest new route: its share rises from about 1.3% in 2010 to 12% by the end of the sample. That growth changes what counts as a credible first destination for an economics PhD, even though the study does not identify whether employer demand, candidate preferences, academic conditions, or broader economic forces drove it.
The technology evidence is especially useful because it separates two meanings of openness. In the candidate-level model, technology is the only destination for which estimated effects of the measured candidate characteristics are consistently insignificant and close to zero. Candidates of different genders and ethnic backgrounds, from different programs and advisor types, enter the sector. In that limited sense, technology is broader than access to top academic placement. But more than 42% of the technology placements go to Amazon. The sector is therefore broad in observed candidate composition and narrow in employer concentration.
That combination is not a contradiction. Candidate breadth asks who appears among the people hired; employer breadth asks how many organizations provide the jobs. A sector can become more accessible across measured candidate groups while remaining dependent on one firm’s hiring strategy. This makes technology a genuine opening but not a secure one. A department may celebrate the rise of technology placements as diversification, yet the pathway could weaken quickly if the dominant employer reduces recruitment. A graduate may gain a new option without gaining a broad market of interchangeable employers.
The geographic evidence follows the same pattern. The North American share of academic first placements falls from about 75% in the early 2000s to about 50% in the most recent period. Europe and East Asia gain share, with different paths by program tier: top-10 graduates see a larger increase in European placements, while graduates from programs ranked 11–30 see a particularly pronounced rise in East Asian placements, reaching about 30% in 2021. The academic labor market is more international in the observed sample, but it is not evenly global. Latin American institutions hire 164 candidates in the study’s counts, compared with 34 in India and 7 across Africa.
The regional result is therefore an expansion of the map, not the disappearance of borders. It shows that more first academic jobs are located outside North America and that China and Europe matter more than before.