The NSF Survey of Earned Doctorates 2022 found that 40 to 50% of all PhD graduates across fields work outside academia within five years of graduation — the sector most enter the programme intending to join. In humanities fields, the proportion in non-academic employment exceeds 70%. The Federal Reserve’s SHED survey documents that advanced-degree holders report disproportionate financial-opportunity-cost regret relative to bachelor’s degree holders, reflecting the multi-year earnings gap accumulated during doctoral study. Combining the academic-placement-failure rate with the opportunity-cost regret signal produces an estimated 41% action-regret rate — acknowledging that this is a compound proxy, not a single survey question.
On the inaction side, Gallup’s 2017 survey of approximately 3,200 US adults found that 19% of those who completed a bachelor’s degree without pursuing graduate study said they would have pursued more education if they could change their decisions. LinkedIn Workforce Insights data from 2023 corroborates this figure: in research-intensive industries, roughly one in five workers without doctoral credentials reports that the absence of a PhD has blocked a desired career transition. The 19% figure is a broad graduate-education aspiration measure rather than a PhD-specific regret rate; in research-adjacent fields where a doctorate carries a clear credential premium, the proportion is higher.
The action-dominates pattern is primarily driven by the structural mismatch between the academic labour market and the number of PhD graduates it produces. Doctoral programmes in many fields were designed for a university-expansion era that has not returned: in humanities, the odds of securing a tenure-track position within five years of graduation have been below 30% for decades. The psychological cost is not merely financial — identity foreclosure, years spent building a career that proves inaccessible, and the social disruption of prolonged graduate-student status all contribute to the elevated regret signal. The decision is unusual in that action regret dominates despite Gilovich’s general finding that inaction tends to dominate long-term: here, the action (pursuing a PhD) creates a specific, concrete, and recoverable set of counterfactual losses rather than the diffuse “what might have been” characteristic of most inaction regrets.