Do High-Premium Fields Buffer Labor Market Shocks? Evidence from India

Jheelum Sarkar

arxiv: 2508.12471 · v4 · submitted 2025-08-17 · 💰 econ.GN · q-fin.EC

Do High-Premium Fields Buffer Labor Market Shocks? Evidence from India

Jheelum Sarkar This is my paper

Pith reviewed 2026-05-18 23:09 UTC · model grok-4.3

classification 💰 econ.GN q-fin.EC

keywords field premialabor market shocksCOVID-19Indiadifference-in-differencestechnical fieldsemployment resilienceeconomic crises

0 comments

The pith

High-premium technical fields in India delivered labor market resilience during COVID-19 only after gradual adjustment in later phases.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper tests whether fields of study that carried higher earnings before the pandemic also shielded workers from labor market disruptions during the COVID-19 crisis in India. It builds pre-pandemic premia for major technical fields and applies a difference-in-differences design with continuous treatment to track how those premia related to employment and earnings changes. A sympathetic reader would care because the answer shapes advice on which education paths best prepare people for economic shocks. The results indicate that any protective effect from high-premium fields appeared not at the sudden onset but through slower adjustment as the pandemic continued.

Core claim

Using pre-pandemic premia for major technical fields in India and a difference-in-difference with continuous treatment design, the study finds that field-premium advantages in labor market outcomes did not emerge immediately at the onset of the pandemic but materialized through gradual adjustment during later phases.

What carries the argument

Continuous-treatment difference-in-differences design that uses pre-pandemic field premia as the treatment intensity to compare changes in labor market outcomes across fields before and during the COVID-19 period.

If this is right

High-premium fields provide protection that requires time to appear rather than immediate insulation from sudden shocks.
Labor market differentiation by field of study strengthens as the crisis moves from acute disruption into recovery stages.
Technical education choices influence resilience mainly through their capacity to support gradual reallocation during prolonged downturns.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Short-term crisis relief programs could treat workers uniformly across fields while longer-term support targets adjustment capacity.
Similar gradual buffering patterns may appear in other developing economies with comparable technical education systems and labor markets.
The timing result raises the question of which mechanisms, such as skill portability or employer networks, drive the later-phase divergence.

Load-bearing premise

Pre-pandemic field premia are accurately measured and exogenous to pandemic-period outcomes so that the continuous-treatment difference-in-differences design can isolate a causal buffering effect without bias from field selection or other time-varying factors.

What would settle it

Observing no relative improvement in employment or earnings for high-premium field workers compared with low-premium field workers in the later phases of the pandemic, after including individual fixed effects and time trends, would falsify the claim.

Figures

Figures reproduced from arXiv: 2508.12471 by Jheelum Sarkar.

**Figure 2.** Figure 2: Dynamic Difference-in-difference results for intensive margin, with the year 2019 [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗

read the original abstract

Do high-return fields of study provide greater protection in labor market during crises? I construct pre-pandemic premia for major technical fields in India and examine whether workers in higher field-premium fields experience resilient labor market outcomes during COVID-19. Using a difference-in-difference with continuous treatment design, I find that field-premium advantages did not emerge immediately at the onset of the pandemic but through gradual adjustment during later phases.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper finds that high-premium technical fields in India did not buffer labor outcomes immediately at COVID onset but showed advantages emerging gradually later, via continuous-treatment DiD on pre-pandemic premia.

read the letter

The main thing here is that field premia advantages in Indian technical education appeared only after an initial lag during the COVID period rather than right at the shock. The paper uses pre-pandemic premia as a continuous treatment in a DiD setup to track labor outcomes across pandemic phases in this specific context. That timing angle is the clearest addition, and applying the design to Indian technical fields gives some new descriptive evidence on how education choices might relate to resilience in a developing economy setting. The approach is standard and fits the question without overclaiming a big theoretical shift. It does a reasonable job focusing on observable outcomes and gradual adjustment instead of just average effects. The soft spots sit mainly in identification. Field selection could still pick up unobservables like networks or location that also shape pandemic responses, and later phases might involve field-specific demand or policy changes that line up with premia levels. The abstract gives little on data construction, parallel trends checks, or error handling, so the gradual pattern could reflect those factors instead of pure buffering. If the full paper has solid robustness on those points it would help a lot; otherwise the result stays suggestive. This is mainly useful for labor economists working on education returns and shocks in emerging markets. A reader interested in human capital policy or workforce planning for future crises would get some value from the India-specific timing evidence. I would send it for peer review because the question is relevant and the design is workable, even if identification details need tightening.

Referee Report

2 major / 2 minor

Summary. The manuscript claims that high-premium technical fields in India do not immediately buffer workers against labor-market shocks at the onset of the COVID-19 pandemic; instead, field-premium advantages emerge gradually through adjustment in later phases. This conclusion rests on a difference-in-differences design that treats pre-pandemic field premia as a continuous treatment variable.

Significance. If the identification strategy is valid, the timing result would usefully extend the literature on education and crisis resilience by showing that field-specific human capital effects are not instantaneous. The continuous-treatment approach is well-suited to the question and could inform education policy in emerging economies, though the paper would benefit from clearer discussion of mechanisms behind the lag.

major comments (2)

[§4] §4 (Empirical Strategy): The continuous-treatment DiD specification does not include field-specific time trends or controls for differential occupational mobility across fields. Without these, the gradual emergence of premia advantages in later phases could reflect post-shock reallocation correlated with premia rather than a causal buffering effect.
[§3] §3 (Data and Sample Construction): The construction of pre-pandemic field premia is described only at a high level; the manuscript does not report how sensitive the timing result is to alternative definitions of the premium (e.g., different reference groups or wage measures) or to sample restrictions that address selection into fields.

minor comments (2)

The abstract omits any mention of the underlying data source, sample size, or key identification checks; adding one sentence on these would improve transparency.
Figure 1 and Table 3 would benefit from clearer labeling of the exact pandemic phases used for the period-specific interactions.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive comments on our manuscript. We believe the suggested revisions will improve the clarity and robustness of our findings regarding the gradual emergence of field-premium advantages during the COVID-19 labor market shock in India. Below, we provide point-by-point responses to the major comments.

read point-by-point responses

Referee: §4 (Empirical Strategy): The continuous-treatment DiD specification does not include field-specific time trends or controls for differential occupational mobility across fields. Without these, the gradual emergence of premia advantages in later phases could reflect post-shock reallocation correlated with premia rather than a causal buffering effect.

Authors: We agree that this is an important consideration to rule out alternative explanations. In the revised manuscript, we will augment the main specification with field-specific time trends to account for differential pre-existing trajectories across fields. We will also introduce controls for occupational mobility by interacting the continuous treatment with indicators of post-shock job changes, leveraging available data on occupational transitions. These modifications will help attribute the gradual effects more confidently to field-specific human capital rather than reallocation. We plan to present the updated results in the main tables and discuss any changes in interpretation in Section 4. revision: yes
Referee: §3 (Data and Sample Construction): The construction of pre-pandemic field premia is described only at a high level; the manuscript does not report how sensitive the timing result is to alternative definitions of the premium (e.g., different reference groups or wage measures) or to sample restrictions that address selection into fields.

Authors: We appreciate this feedback on improving the transparency of our data construction. In the revision, we will provide a more detailed description in Section 3, including the exact regression used to estimate field premia, the reference group (workers in non-technical fields), and the wage variable (log monthly earnings). To demonstrate robustness, we will add an appendix with sensitivity analyses using alternative reference groups (e.g., all workers or specific low-premium fields), different wage measures (hourly wages where available), and sample restrictions such as limiting to prime-age workers or those with consistent employment history to address selection. These checks will show that the key finding of gradual adjustment remains stable. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical DiD relies on observed pre-pandemic premia and post-shock outcomes

full rationale

The paper constructs field premia from pre-pandemic data and applies a continuous-treatment difference-in-differences design to compare labor-market outcomes across fields during COVID-19 phases. The timing-specific result emerges from differential changes in observed employment or earnings data rather than any definitional equivalence, fitted parameter renamed as prediction, or self-citation chain. The identifying assumptions (exogeneity of field choice, parallel trends conditional on controls) are stated as empirical requirements, not derived from the target result itself. This is a standard self-contained empirical exercise with external data benchmarks.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The analysis depends on the validity of constructed pre-pandemic field premia as a continuous treatment and on standard DiD identification assumptions that are not elaborated beyond the abstract description.

free parameters (1)

field premia
Pre-pandemic premia are constructed from data and serve as the continuous treatment intensity variable.

axioms (1)

domain assumption Parallel trends assumption holds across field groups in the absence of treatment.
Required for causal interpretation of the difference-in-differences estimator.

pith-pipeline@v0.9.0 · 5582 in / 1286 out tokens · 53264 ms · 2026-05-18T23:09:20.275418+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

23 extracted references · 23 canonical work pages

[1]

Amirtham S, N., & Kumar, A. (2023). The underrepresentation of women in STEM disciplines in India: a secondary analysis. International Journal of Science Education, 45(12), 1008–1031.https://doi.org/10.1080/09500693.2023.2179901

work page doi:10.1080/09500693.2023.2179901 2023
[3]

Becker, G. S. (1964). Human capital: A theoretical and empirical analysis, with special reference to education (1st ed.). National Bureau of Economic Research; Columbia University Press. https://www.nber.org/books-and-chapters/human-capital- theoretical-and-empirical-analysis-special-reference-education-first-edition

work page 1964
[4]

Chudnovsky and S

Black, S. E., Muller, C., Spitz-Oener, A., He, Z., Hung, K., & Warren, J. R. (2021). The importance of STEM: High school knowledge, skills and occupations in an era of growing inequality. Research Policy, 50(7), 104249.https://doi.org/10.1016/j. respol.2021.104249

work page doi:10.1016/j 2021
[5]

Chiu, T. K. F., & Li, Y. (2023). How can emerging technologies impact STEM educa- tion? Journal for STEM Education Research, 6, 375–384.https://doi.org/10.1007/ s41979-023-00113-w

work page 2023
[6]

N., Stepner, M., & The Opportunity Insights Team

Chetty, R., Friedman, J. N., Stepner, M., & The Opportunity Insights Team. (2025). The economic impacts of COVID-19: Evidence from a new public database built using private sector data (NBER Working Paper No. 33469). National Bureau of Economic Research.https://doi.org/10.3386/w33469

work page doi:10.3386/w33469 2025
[7]

J., & Noray, K

Deming, D. J., & Noray, K. (2020). Earnings dynamics, changing job skills, and STEM careers. The Quarterly Journal of Economics, 135(4), 1965–2005.https://doi.org/ 10.1093/qje/qjaa021

work page doi:10.1093/qje/qjaa021 2020
[8]

I., & Neiman, B

Dingel, J. I., & Neiman, B. (2020). How many jobs can be done at home? Journal of Public Economics, 189, 104235.https://doi.org/10.1016/j.jpubeco.2020.104235

work page doi:10.1016/j.jpubeco.2020.104235 2020
[9]

(April 15, 2020)

Ministry of Home Affairs. (April 15, 2020). Order No. 40-3/2020-DM-I(A).Government of India.https://www.mha.gov.in/sites/default/files/MHA%20order%20dt% 2015.04.2020,%20with%20Revised%20Consolidated%20Guidelines_compressed% 20(3).pdf 24

work page 2020
[10]

Gunadi, C. (2019). An inquiry on the impact of highly-skilled STEM immigration on the U.S. economy. Labour Economics, 61, 101751.https://doi.org/10.1016/j.labeco. 2019.101751

work page doi:10.1016/j.labeco 2019
[11]

ILO & OECD. (2020). The impact of the COVID-19 pandemic on jobs and incomes in G20 economies. Saudi Arabia’s G20 Presidency

work page 2020
[12]

Khare, M. (2019). Exploring the relationship between economic growth, employment and education in Indian states. In India’s social sector and SDGs (1st ed., Chapter 13). Routledge India.https://doi.org/10.4324/9780367341824

work page doi:10.4324/9780367341824 2019
[13]

Merino, C., Marzabal, A., Miller, B.G., Carrasco, X. (2023). Science Teacher Education in Chile: On the Verge of a Turning Point toward STEM-Oriented Science Education. In: Al-Balushi, S.M., Martin-Hansen, L., Song, Y. (eds) Reforming Science Teacher Education Programs in the STEM Era. Palgrave Studies on Leadership and Learn- ing in Teacher Education. Pa...

work page 2023
[14]

Mehrotra, S., & Parida, J. K. (2019). India’s employment crisis: Rising education levels and falling non-agricultural job growth (CSE Working Pa- per No. 2019-04). Centre for Sustainable Employment, Azim Premji Uni- versity.https://publications.azimpremjiuniversity.edu.in/2119/1/Mehrotra_ Parida_India_Employment_Crisis.pdf

work page 2019
[15]

Ministry of Education. (2020). All India survey on higher education 2019-20. Depart- ment of Higher Education, Government of India.https://www.education.gov.in/ sites/upload_files/mhrd/files/statistics-new/aishe_eng.pdf

work page 2020
[16]

R., & Phelps, E

Nelson, R. R., & Phelps, E. S. (1966). Investment in Humans, Technological Diffusion, and Economic Growth. The American Economic Review, 56(1/2), 69–75.http://www. jstor.org/stable/1821269

work page arXiv 1966
[17]

OECD. (2021). An assessment of the impact of COVID-19 on job and skills demand using online job vacancy data. OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris.https://doi.org/10.1787/20fff09e-en

work page doi:10.1787/20fff09e-en 2021
[18]

conveyor belt

Richiardi, M. G., Westhoff, L., Astarita, C., Ernst, E., Fenwick, C., Khabirpour, N., & Pelizzari, L. (2025). The impact of a decade of digital transformation on employment, wages, and inequality in the EU: A “conveyor belt” hypothesis. Socio-Economic Review. Advance online publication.https://doi.org/10.1093/ser/mwaf011 25

work page doi:10.1093/ser/mwaf011 2025
[19]

Ruˇ zarovsk´ y, R., Holubek, R., Jan´ ıˇ cek, M., Vel´ ıˇ sek, K., & Tirian, G. O. (2021). Analysis of the Industry 4.0 key elements and technologies implementation in the Festo Didactic educational systems MPS 203 I4.0. Journal of Physics: Conference Series, 1781, 012030. https://doi.org/10.1088/1742-6596/1781/1/012030

work page doi:10.1088/1742-6596/1781/1/012030 2021
[20]

Schultz, T. W. (1975). The Value of the Ability to Deal with Disequilibria. Journal of Economic Literature, 13(3), 827–846.http://www.jstor.org/stable/2722032

work page arXiv 1975
[21]

Taubman, P., & Wachter, M. L. (1986). Segmented labor markets. In O. Ashenfelter & R. Layard (Eds.), Handbook of labor economics, (2), 1183–1217. Elsevier.https: //doi.org/10.1016/S1573-4463(86)02011-4

work page doi:10.1016/s1573-4463(86)02011-4 1986
[22]

Primary and Secondary Labor Markets: A Critique of the Dual Approach

Wachter, M.L.(1974). Primary and Secondary Labor Markets: A Critique of the Dual Approach. Brookings Papers on Economic Activity No. 3. Brook- ings Institute.https://www.brookings.edu/wp-content/uploads/1974/12/1974c_ bpea_wachter_gordon_piore_hall.pdf

work page 1974
[23]

Wang, K.-H., & Lu, W.-C. (2025). AI-induced job impact: Complementary or sub- stitution? Empirical insights and sustainable technology considerations. Sustainable Technology and Entrepreneurship, 4(1), 100085.https://doi.org/10.1016/j.stae. 2024.100085

work page doi:10.1016/j.stae 2025
[24]

highly feasible

Widya, R., Rifandi, R., & Rahmi, Y. L. (2019). STEM education to fulfil the 21st century demand: A literature review. Journal of Physics: Conference Series, 1317, 012208.https://doi.org/10.1088/1742-6596/1317/1/012208 26 Appendix Figure A.1: Distribution of jobs by remote work feasibility between STEM vs non-STEM graduates.Notes: Here, jobs are classified...

work page doi:10.1088/1742-6596/1317/1/012208 2019

[1] [1]

Amirtham S, N., & Kumar, A. (2023). The underrepresentation of women in STEM disciplines in India: a secondary analysis. International Journal of Science Education, 45(12), 1008–1031.https://doi.org/10.1080/09500693.2023.2179901

work page doi:10.1080/09500693.2023.2179901 2023

[2] [3]

Becker, G. S. (1964). Human capital: A theoretical and empirical analysis, with special reference to education (1st ed.). National Bureau of Economic Research; Columbia University Press. https://www.nber.org/books-and-chapters/human-capital- theoretical-and-empirical-analysis-special-reference-education-first-edition

work page 1964

[3] [4]

Chudnovsky and S

Black, S. E., Muller, C., Spitz-Oener, A., He, Z., Hung, K., & Warren, J. R. (2021). The importance of STEM: High school knowledge, skills and occupations in an era of growing inequality. Research Policy, 50(7), 104249.https://doi.org/10.1016/j. respol.2021.104249

work page doi:10.1016/j 2021

[4] [5]

Chiu, T. K. F., & Li, Y. (2023). How can emerging technologies impact STEM educa- tion? Journal for STEM Education Research, 6, 375–384.https://doi.org/10.1007/ s41979-023-00113-w

work page 2023

[5] [6]

N., Stepner, M., & The Opportunity Insights Team

Chetty, R., Friedman, J. N., Stepner, M., & The Opportunity Insights Team. (2025). The economic impacts of COVID-19: Evidence from a new public database built using private sector data (NBER Working Paper No. 33469). National Bureau of Economic Research.https://doi.org/10.3386/w33469

work page doi:10.3386/w33469 2025

[6] [7]

J., & Noray, K

Deming, D. J., & Noray, K. (2020). Earnings dynamics, changing job skills, and STEM careers. The Quarterly Journal of Economics, 135(4), 1965–2005.https://doi.org/ 10.1093/qje/qjaa021

work page doi:10.1093/qje/qjaa021 2020

[7] [8]

I., & Neiman, B

Dingel, J. I., & Neiman, B. (2020). How many jobs can be done at home? Journal of Public Economics, 189, 104235.https://doi.org/10.1016/j.jpubeco.2020.104235

work page doi:10.1016/j.jpubeco.2020.104235 2020

[8] [9]

(April 15, 2020)

Ministry of Home Affairs. (April 15, 2020). Order No. 40-3/2020-DM-I(A).Government of India.https://www.mha.gov.in/sites/default/files/MHA%20order%20dt% 2015.04.2020,%20with%20Revised%20Consolidated%20Guidelines_compressed% 20(3).pdf 24

work page 2020

[9] [10]

Gunadi, C. (2019). An inquiry on the impact of highly-skilled STEM immigration on the U.S. economy. Labour Economics, 61, 101751.https://doi.org/10.1016/j.labeco. 2019.101751

work page doi:10.1016/j.labeco 2019

[10] [11]

ILO & OECD. (2020). The impact of the COVID-19 pandemic on jobs and incomes in G20 economies. Saudi Arabia’s G20 Presidency

work page 2020

[11] [12]

Khare, M. (2019). Exploring the relationship between economic growth, employment and education in Indian states. In India’s social sector and SDGs (1st ed., Chapter 13). Routledge India.https://doi.org/10.4324/9780367341824

work page doi:10.4324/9780367341824 2019

[12] [13]

Merino, C., Marzabal, A., Miller, B.G., Carrasco, X. (2023). Science Teacher Education in Chile: On the Verge of a Turning Point toward STEM-Oriented Science Education. In: Al-Balushi, S.M., Martin-Hansen, L., Song, Y. (eds) Reforming Science Teacher Education Programs in the STEM Era. Palgrave Studies on Leadership and Learn- ing in Teacher Education. Pa...

work page 2023

[13] [14]

Mehrotra, S., & Parida, J. K. (2019). India’s employment crisis: Rising education levels and falling non-agricultural job growth (CSE Working Pa- per No. 2019-04). Centre for Sustainable Employment, Azim Premji Uni- versity.https://publications.azimpremjiuniversity.edu.in/2119/1/Mehrotra_ Parida_India_Employment_Crisis.pdf

work page 2019

[14] [15]

Ministry of Education. (2020). All India survey on higher education 2019-20. Depart- ment of Higher Education, Government of India.https://www.education.gov.in/ sites/upload_files/mhrd/files/statistics-new/aishe_eng.pdf

work page 2020

[15] [16]

R., & Phelps, E

Nelson, R. R., & Phelps, E. S. (1966). Investment in Humans, Technological Diffusion, and Economic Growth. The American Economic Review, 56(1/2), 69–75.http://www. jstor.org/stable/1821269

work page arXiv 1966

[16] [17]

OECD. (2021). An assessment of the impact of COVID-19 on job and skills demand using online job vacancy data. OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris.https://doi.org/10.1787/20fff09e-en

work page doi:10.1787/20fff09e-en 2021

[17] [18]

conveyor belt

Richiardi, M. G., Westhoff, L., Astarita, C., Ernst, E., Fenwick, C., Khabirpour, N., & Pelizzari, L. (2025). The impact of a decade of digital transformation on employment, wages, and inequality in the EU: A “conveyor belt” hypothesis. Socio-Economic Review. Advance online publication.https://doi.org/10.1093/ser/mwaf011 25

work page doi:10.1093/ser/mwaf011 2025

[18] [19]

Ruˇ zarovsk´ y, R., Holubek, R., Jan´ ıˇ cek, M., Vel´ ıˇ sek, K., & Tirian, G. O. (2021). Analysis of the Industry 4.0 key elements and technologies implementation in the Festo Didactic educational systems MPS 203 I4.0. Journal of Physics: Conference Series, 1781, 012030. https://doi.org/10.1088/1742-6596/1781/1/012030

work page doi:10.1088/1742-6596/1781/1/012030 2021

[19] [20]

Schultz, T. W. (1975). The Value of the Ability to Deal with Disequilibria. Journal of Economic Literature, 13(3), 827–846.http://www.jstor.org/stable/2722032

work page arXiv 1975

[20] [21]

Taubman, P., & Wachter, M. L. (1986). Segmented labor markets. In O. Ashenfelter & R. Layard (Eds.), Handbook of labor economics, (2), 1183–1217. Elsevier.https: //doi.org/10.1016/S1573-4463(86)02011-4

work page doi:10.1016/s1573-4463(86)02011-4 1986

[21] [22]

Primary and Secondary Labor Markets: A Critique of the Dual Approach

Wachter, M.L.(1974). Primary and Secondary Labor Markets: A Critique of the Dual Approach. Brookings Papers on Economic Activity No. 3. Brook- ings Institute.https://www.brookings.edu/wp-content/uploads/1974/12/1974c_ bpea_wachter_gordon_piore_hall.pdf

work page 1974

[22] [23]

Wang, K.-H., & Lu, W.-C. (2025). AI-induced job impact: Complementary or sub- stitution? Empirical insights and sustainable technology considerations. Sustainable Technology and Entrepreneurship, 4(1), 100085.https://doi.org/10.1016/j.stae. 2024.100085

work page doi:10.1016/j.stae 2025

[23] [24]

highly feasible

Widya, R., Rifandi, R., & Rahmi, Y. L. (2019). STEM education to fulfil the 21st century demand: A literature review. Journal of Physics: Conference Series, 1317, 012208.https://doi.org/10.1088/1742-6596/1317/1/012208 26 Appendix Figure A.1: Distribution of jobs by remote work feasibility between STEM vs non-STEM graduates.Notes: Here, jobs are classified...

work page doi:10.1088/1742-6596/1317/1/012208 2019