Student or educator? Sign up with your school email and get 6 months of Pro free.

The DOI matches its Crossref record; 67 of the 67 references it cites were found in a registry or an index; none of the 14 sources its claims cite is checked for support yet.

Taxation, Corruption, and Growth

Open access
Aghion, Philippe, Akcigit, Ufuk, Cagé, Julia, Kerr, William · 2016 · National Bureau of Economic Research

Abstract

We build an endogenous growth model to analyze the relationships between taxation, corruption, and economic growth.Entrepreneurs lie at the center of the model and face disincentive effects from taxation but acquire positive benefits from public infrastructure.Political corruption governs the efficiency with which tax revenues are translated into infrastructure.The model predicts an inverted-U relationship between taxation and growth, with corruption reducing the optimal taxation level.We find evidence consistent with these predictions and the entrepreneurial channel using data from the Longitudinal Business Database of the US Census Bureau.The marginal effect of taxation for growth for a state at the 10th or 25th percentile of corruption is significantly positive; on the other hand, the marginal effects of taxation for growth for a state at the 90th percentile of corruption are much lower across the board.We make progress towards causality through Granger-style tests and by considering periphery counties where effective tax policy is largely driven by bordering states.Finally, we calibrate our model and find that the calibrated taxation rate of 37% is fairly close to the model's estimated welfare maximizing taxation rate of 42%.Reducing corruption provides the largest potential impact for welfare gain through its impact on the uses of tax revenues.

Overview

CoChat Summary

TL;DR

The paper builds an endogenous Schumpeterian growth model in which taxation funds public infrastructure but corruption reduces how efficiently tax revenue becomes infrastructure, predicting an inverted-U relationship between taxation and growth that shifts down with corruption. Using US state and county panel data from the Census Bureau's Longitudinal Business Database, the authors find the marginal growth effect of taxation is positive in low-corruption states but much weaker or zero in high-corruption states, and calibration suggests eliminating corruption yields a consumption-equivalent welfare gain of over 20%.

Key findings

  • The model predicts an inverted-U effect of the tax rate on equilibrium growth, with the positive impact of tax on growth smaller when government is less effective (`∂²g/∂τ∂φ < 0` for τ below the peak).
  • Empirically, the marginal effect of taxation on growth is significantly positive for a state at the 10th or 25th percentile of corruption, while marginal effects for a state at the 90th percentile of corruption are "much lower across the board" and rarely statistically different from zero except at very low initial tax levels.
  • The interaction of taxation and corruption is negative and statistically significant in state GDP levels regressions (e.g., −0.060, SE 0.023 in the baseline column) and in growth regressions; the interaction is weak and insignificant in Granger-style reverse regressions.
  • Calibration yields an empirical tax rate of 37%, close to the growth-maximizing rate of 42.47% and welfare-maximizing rate of 42.3%; removing corruption entirely (from φ = 0.243 to 0) raises growth to 3.23% and produces a 22% consumption-equivalent welfare gain, versus only a 1.1% welfare gain from setting the optimal tax rate at current corruption.

Methods

  • Theoretical: a continuous-time Schumpeterian growth model building on Klette and Kortum (2004), adding government-provided infrastructure financed by a tax on firm operating profits, with a fraction φ of tax revenue lost to corruption; later generalized to include capital and calibrated.
  • Empirical: state-level panel of 47 states (46 states plus DC with non-zero income taxes; Nevada, Texas, Washington, Wyoming excluded) over 1983–2007, structured into four five-year periods (188 observations), with state and period fixed effects, standard errors clustered by state; dependent variables include log state GDP, GDP per worker, employment, and establishment counts, with lagged tax revenues and corruption (federal convictions of public officials, averaging 18 per annum) normalized by initial government expenditures and demeaned before interaction.
  • County-level border analysis: for each of 2,795 counties, a 100-mile-radius ring around the county centroid is used to construct employment-weighted localized tax and corruption measures from home and neighboring states; 38% of counties are on a state border and 26% draw more than half their effective influence from other states. Calibration uses moments M1 entrant employment share 0.058, M2 growth rate 0.022, M3 average profitability before tax 0.091, M4 corporate tax revenue over GDP 0.034.

Limitations

  • The paper states that evidence for the predicted inverted-U is modest: the squared taxation coefficient is negative but statistically insignificant for most growth measures, which the authors attribute to US variation lying mostly on the left-hand side of the curve.
  • The paper states that the negative main effect of corruption is the largest gap between theory and empirics — point estimates are mostly negative but only powerful in growth regressions lacking state fixed effects.
  • The paper states it cannot use instrumental variables and remains "far from a complete proof" on causality, relying instead on Granger-style tests and state-border analysis; it also notes that convictions of public officials are an imperfect corruption measure that misses legal but wasteful diversion of funds, and that the empirical work covers only US regional variation rather than cross-country settings with extreme corruption or tax rates.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3386/w21928

Cites

67 of 67 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

0 Valid · 1 Not valid · 9 Not checked · 172 Own results and interpretations

97 more claims

Create a free account to see all 182 claims and to verify the 6 references still open.

See all claims — free account
  • It may state two separate claims as one. 57
  • Probably not a substantial claim. 32

Create a free account to see these claims.

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

1 of 1

Citations

Cites

Research works printed but not linked: 10

  1. Aghion, P, and P. Howitt (2009), The Economics of Growth, MIT Press.
  2. Barro, R, and X. Sala-i-Martin (1995), Economic Growth, Cambridge: MIT Press.
  3. Glaeser, E, Kerr, S, and W. Kerr (2015), "Entrepreneurship, Mines, and Urban Growth", Review of Economics and Statistics, 92:7, 498-520.
  4. Helms, L (1985), "The E¤ect of State and Local Taxes on Economic Growth, A Time Series-Cross Section Approach", Review of Economics and Statistics, 67, 574-582.
  5. Holmes, T (1998), "The E¤ects of State Policies on the Location of Manufacturing: Evi- dence from State Borders", Journal of Political Economy, 106, 667-705.
  6. Kerr, W, and S. Kominers (2015), "Agglomeration Forces and Cluster Shapes", Review of Economics and Statistics, 97:4, 877-899.
  7. Mo…di, A, and J. A. Stone (1990), "Do State and Local Taxes A¤ect Economic Growth?", Review of Economics and Statistics, 72:4, 686-691.
  8. Paserman, D, Neeman, Z, and A. Simhon (2008), "Corruption and Openness", Berkeley Electronic Journal for Economic Analysis and Policy, 8:1, Article 50.
  9. Saiz, A (2010), "The Geographical Determinants of Housing Supply", Quarterly Journal of Economics, 125:3, 1253-1296.
  10. Wilson, D (2009), "Beggar Thy Neighbor? The In-State, Out-of-State, and Aggregate E¤ects of R&D Tax Credits", Review of Economics and Statistics, 91:2, 431-436.

Cited by

See all 29

Metadata

Authors
Aghion, P., Akcigit, U., Cagé, J., & Kerr, W.
Affiliations
  • Harvard University (US)
  • University of Chicago (US)
  • Institut d'Etudes Politiques de Paris (FR)
Venue
National Bureau of Economic Research
Published
2016-01
OpenAlex
W2891209103
Cited by
28
References
65
Citations by year
2026: 1 · 2025: 1 · 2024: 3 · 2023: 3 · 2022: 4 · 2021: 2 · 2020: 6 · 2019: 2 · 2018: 1 · 2017: 4
Type
report
Language
en
Open access
Locations
  • National Bureau of Economic Research (publishedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
Funders
  • National Science Foundation (0427889)
  • Ewing Marion Kauffman Foundation
  • University of Pennsylvania
  • Harvard Business School
Topics
  • Social Sciences › Economics, Econometrics and Finance › Economics and Econometrics › Fiscal Policy and Economic Growth
  • Social Sciences › Sociology and Political Science › Corruption and Economic Development
  • Social Sciences › Economics, Econometrics and Finance › Economics and Econometrics › Taxation and Compliance Studies
Keywords
taxation, political corruption, economic growth, endogenous growth model, entrepreneurs, public infrastructure, tax revenue use, inverted U-shaped relationship, optimal taxation, social welfare maximization, Longitudinal Business Database

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A language model drew the claims from the paper on 2026-11-16, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-11-16 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.