Pith. sign in

REVIEW 2 major objections 1 minor 29 references

Temporal Dynamics of Development Aid in Africa: Evidence from a Staggered Difference-in-Differences Study of China and World Bank Projects

T0 review · 2 major / 1 minor · reviewed 2026-06-27 · grok-4.3

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Staggered difference-in-differences analysis shows World Bank and Chinese aid effects on local wealth in Africa are selective by sector rather than uniform.

desk verdict The paper applies dCdH to geocoded aid data and finds sector-specific effects that differ from TWFE, but the central claim still depends on whether the estimator fully removes time-varying selection. read the letter →

arxiv 2606.06651 v2 pith:3WMV26IR submitted 2026-06-04 econ.GN q-fin.EC

Mattias Antar , Adel Daoud , Connor T. Jerzak This is my paper · ORCID
classification econ.GNq-fin.EC
keywords developmentaidAfricadifference-in-differencesWorldBankChinalocalwealthstaggeredtreatmenteffectiveness
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper links geocoded World Bank and Chinese projects to a balanced panel of 2,166 DHS clusters across 35 African countries from 2002 to 2013 and imputes a household wealth index from satellite data. It compares conventional two-way fixed effects event studies against the de Chaisemartin and D'Haultfoeuille switcher-stayer estimator to handle staggered treatment timing and selective project placement. Pre-treatment diagnostics reveal that treated clusters often start from weaker positions, leading standard methods to overstate gains. Under the preferred estimator, positive associations appear mainly for World Bank health projects and Chinese water supply and sanitation projects, with little support for broad donor-wide improvements.

What carries the argument

The de Chaisemartin and D'Haultfoeuille (dCdH) switcher-stayer estimator, which identifies causal effects by comparing newly treated units to untreated units with the same treatment history to avoid contaminated comparisons under staggered timing.

What would settle it

Re-estimating the same models on a direct household asset measure instead of satellite-imputed wealth and recovering uniform positive effects across most donor-sector panels would falsify the selectivity conclusion.

Watch

Extended reading notes

Core claim

The authors establish that once staggered adoption and selective placement are addressed with the switcher-stayer estimator, estimated effects of aid on local wealth become concentrated in a limited set of donor-sector combinations: strongest for World Bank health, positive but less cleanly identified for education, and for China in water supply and sanitation plus other social infrastructure, while Chinese energy projects lose their positive signal. The results reject any claim of uniform improvement from either donor and show that conclusions depend heavily on treatment timing, selection correction, and outcome measurement.

Load-bearing premise

Conditional on the estimator's weighting by treatment history, no remaining time-varying confounders or anticipation effects differ between switchers and stayers.

Editorial extensions

If this is right

  • Aid effectiveness evidence is sector-specific rather than donor-general, so aggregate claims about World Bank or Chinese projects require disaggregation.
  • Conventional two-way fixed effects event studies tend to overstate post-treatment gains when projects are placed selectively into weaker areas.
  • Positive signals concentrate in health for the World Bank and water and sanitation for China, while energy projects show no robust effect under the staggered design.
  • Pre-treatment diagnostics for selective placement are necessary before interpreting any aid impact estimates.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Evaluations of aid should routinely test multiple outcome measures because satellite imputation may emphasize infrastructure-visible dimensions of wealth.
  • The same staggered design could be applied to additional donors or later time periods to check whether the sector patterns generalize beyond 2002-2013.
  • Policymakers might allocate resources differently if health and water projects consistently outperform others once timing and selection are properly handled.
  • The dependence on how treatment timing is modeled implies that aid databases need precise start and end dates for credible subnational studies.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

2 major / 1 minor

Summary. The manuscript claims that a balanced panel of 2,166 DHS clusters across 35 African countries (2002–2013), linked to geocoded World Bank and Chinese AidData projects and satellite-imputed International Wealth Index values, shows selective project placement; conventional TWFE event studies therefore overstate post-treatment gains relative to the dCdH switcher-stayer estimator. Under dCdH the estimated effects become donor- and sector-specific, with the strongest positive associations appearing in World Bank Health and Chinese Water Supply & Sanitation/Other Social Infrastructure, while Chinese Energy Generation & Supply drops to near zero; overall the results reject uniform donor-wide wealth improvements.

Significance. If the dCdH identification assumptions hold after the reported pre-treatment diagnostics, the paper supplies useful subnational evidence on aid heterogeneity that moves beyond nighttime-lights proxies and highlights how staggered-treatment contamination can inflate conventional estimates. The explicit comparison of estimators and the use of a household-centered wealth measure are strengths that would be cited in future work on development assistance.

major comments (2)
  1. [Abstract] Abstract (paragraph on pre-treatment diagnostics and estimator choice): the maintained assumption that dCdH weighting eliminates differential time-varying confounders and anticipation effects between switchers and stayers is load-bearing for the central claim of sector-specific effects; the abstract reports selective placement but supplies no post-weighting balance statistics, explicit anticipation tests, or robustness to alternative weighting schemes, leaving open the possibility that residual confounders drive the concentration in Health (World Bank) and Water/Sanitation (China) panels.
  2. [Abstract] Abstract (final paragraph): the conclusion that results 'depend strongly on how treatment timing, selection, and outcome measurement are handled' is correct in direction but the manuscript does not report sensitivity of the sector-specific point estimates to the satellite imputation procedure or to post-hoc sample restrictions; without these checks the claim that dCdH yields 'more selective' evidence than TWFE cannot be fully evaluated.
minor comments (1)
  1. The abstract would benefit from stating the number of clusters and events per donor–sector panel so readers can assess statistical power behind the reported nulls (e.g., Chinese Energy).

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for these constructive comments highlighting the importance of the dCdH assumptions and sensitivity checks. We respond to each major comment below.

read point-by-point responses

  1. Referee: [Abstract] Abstract (paragraph on pre-treatment diagnostics and estimator choice): the maintained assumption that dCdH weighting eliminates differential time-varying confounders and anticipation effects between switchers and stayers is load-bearing for the central claim of sector-specific effects; the abstract reports selective placement but supplies no post-weighting balance statistics, explicit anticipation tests, or robustness to alternative weighting schemes, leaving open the possibility that residual confounders drive the concentration in Health (World Bank) and Water/Sanitation (China) panels.

    Authors: The dCdH switcher-stayer design is constructed precisely to avoid contaminated comparisons by restricting to units with identical treatment histories up to each period, which directly targets differential time-varying confounders and anticipation under the maintained assumptions. The manuscript already documents selective placement via pre-treatment diagnostics. To strengthen the abstract and address the concern, we will add post-weighting balance statistics, explicit anticipation tests, and results under alternative weighting schemes to the revised appendix. These will show that the sector-specific patterns (e.g., World Bank Health, Chinese Water/Sanitation) are not driven by residual imbalance. revision: yes

  2. Referee: [Abstract] Abstract (final paragraph): the conclusion that results 'depend strongly on how treatment timing, selection, and outcome measurement are handled' is correct in direction but the manuscript does not report sensitivity of the sector-specific point estimates to the satellite imputation procedure or to post-hoc sample restrictions; without these checks the claim that dCdH yields 'more selective' evidence than TWFE cannot be fully evaluated.

    Authors: We agree that explicit sensitivity to the satellite imputation procedure and post-hoc sample restrictions would allow fuller evaluation of the dCdH versus TWFE comparison. The International Wealth Index is satellite-imputed, and while the core results use the primary imputation, we will add robustness checks using alternative imputation specifications and restrictions to clusters with higher-quality direct measures. We will also report results under varied sample restrictions. These will be incorporated in the revision to support the claim that dCdH produces more selective evidence. revision: yes

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: empirical application of external off-the-shelf estimator to independent data

full rationale

The paper applies the published dCdH switcher-stayer estimator (de Chaisemartin & D'Haultfoeuille) and TWFE to geocoded AidData projects linked to DHS clusters and satellite-imputed wealth indices. No equation in the manuscript defines a quantity in terms of itself, renames a fitted parameter as a prediction, or derives the central ATT estimates from a self-citation chain. Pre-treatment diagnostics and estimator choice are described as maintained assumptions whose validity is external to the paper's own algebra. The reported sector-specific patterns are direct outputs of the applied estimators on external data and do not reduce by construction to the paper's inputs.

Assumptions & free parameters 0 free parameters · 1 assumptions · 0 invented entities

The analysis rests on standard econometric identification assumptions for staggered difference-in-differences rather than new free parameters, invented entities, or ad-hoc axioms introduced by the authors.

assumptions (1)
  • domain assumption Standard no-anticipation and conditional parallel trends assumptions required by the dCdH switcher-stayer estimator
    Invoked when the paper prefers dCdH over TWFE to avoid contaminated comparisons under staggered timing.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Temporal Dynamics of Development Aid in Africa: Evidence from a Staggered Difference-in-Differences Study of China and World Bank Projects." pith.science (2026). https://pith.science/paper/3WMV26IR

@misc{pith2026260606651,
  author       = {Pith},
  title        = {Pith review of: Temporal Dynamics of Development Aid in Africa: Evidence from a Staggered Difference-in-Differences Study of China and World Bank Projects},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3WMV26IR}},
  note         = {Machine review of arXiv:2606.06651}
}
read the original abstract

Subnational studies of aid effectiveness often rely on repeated cross-sections or nighttime lights, making it difficult to separate local treatment effects from baseline differences and potentially favoring infrastructure-heavy projects. We address these limitations by studying World Bank and Chinese development projects in Africa with a balanced panel of 2,166 DHS clusters across 35 countries from 2002 to 2013. Geocoded AidData projects are linked to satellite-imputed International Wealth Index estimates, a household-centered measure of material living standards. We compare a conventional two-way fixed effects (TWFE) event-study with the switcher--stayer estimator of de Chaisemartin and D'Haultfoeuille (dCdH), which avoids contaminated comparisons under staggered treatment timing. Pre-treatment diagnostics show that project placement is frequently selective: clusters that later receive projects often begin from weaker relative positions before treatment onset. Consequently, TWFE often implies larger post-treatment gains than the preferred staggered-treatment design supports. Under dCdH, the evidence becomes more selective and sector-specific. For the World Bank, positive evidence is strongest in Health, while Education shows positive but less cleanly identified gains. For China, Water Supply and Sanitation and Other Social Infrastructure and Services show positive associations with local wealth, although residual selection concerns remain. By contrast, Chinese Energy Generation and Supply appears strongly positive under TWFE but falls close to zero under dCdH. Overall, the results do not support a donor-wide claim that either the World Bank or China uniformly improves local wealth. Instead, estimated effects are concentrated in a limited set of donor--sector panels and depend strongly on how treatment timing, selection, and outcome measurement are handled.

Figures

Figures reproduced from arXiv: 2606.06651 by the authors.

Figure 1
Figure 1. Spatial coverage of the balanced sample across the 35-country study area. Darker [PITH_FULL_IMAGE:figures/full_fig_p008_1.png] view at source ↗
Figure 2
Figure 2. Treatment-definition alternatives and outcome-support mapping. The schematic [PITH_FULL_IMAGE:figures/full_fig_p012_2.png] view at source ↗
Figure 3
Figure 3. Last pre-treatment diagnostic contrasts, reported as the pre-treatment diagnos [PITH_FULL_IMAGE:figures/full_fig_p020_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Benchmark TWFE and preferred dCdH event-study estimates for six focal donor– [PITH_FULL_IMAGE:figures/full_fig_p023_4.png]
Figure 5
Figure 5. Figure 5: Cross-panel relationship between the preferred dCdH pre-treatment diagnostic [PITH_FULL_IMAGE:figures/full_fig_p024_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

29 extracted references · 1 canonical work pages

  1. [1]

    World Bank geocoded research release, version 1.4.2 geocoded dataset

    AidData (2017). World Bank geocoded research release, version 1.4.2 geocoded dataset. Technical report, AidData

    2017

  2. [2]

    B., Hansen, H., and Markussen, T

    Andersen, T. B., Hansen, H., and Markussen, T. (2006). US politics and World Bank ida-lending.The Journal of Development Studies, 42(5):772–794

    2006

  3. [3]

    (1988).Spatial Econometrics: Methods and Models

    Anselin, L. (1988).Spatial Econometrics: Methods and Models. Kluwer Academic Publishers

    1988

  4. [4]

    C., Larcker, D

    Baker, A. C., Larcker, D. F., and Wang, C. C. (2022). How much should we trust staggered difference-in-differences estimates?Journal of Financial Economics, 144(2):370–395

    2022

  5. [5]

    Bluhm, R., Dreher, A., Fuchs, A., Parks, B., Strange, A., and Tierney, M. J. (2018). Connective financing: Chinese infrastructure projects and the diffusion of economic activity in developing countries. Technical Report Working Paper No. 64, AidData

    2018

  6. [6]

    Borusyak, K., Jaravel, X., and Spiess, J. (2024). Revisiting event-study designs: Robust and efficient estimation.Review of Economic Studies, 91(6):3253–3285

    2024

  7. [7]

    Briggs, R. C. (2012). Electrifying the base? aid and incumbent advantage in ghana.The Journal of Modern African Studies, 50(4):603–624

    2012

  8. [8]

    and Dollar, D

    Burnside, C. and Dollar, D. (2000). Aid, policies, and growth.American Economic Review, 90(4):847–868

    2000

Show all 29 references
  1. [9]

    and Sant’Anna, P

    Callaway, B. and Sant’Anna, P. H. C. (2021). Difference-in-Differences with multiple time periods.Journal of Econometrics, 225(2):200–230

    2021

  2. [10]

    and Banik, D

    Chasukwa, M. and Banik, D. (2019). Institutional bypass and aid effectiveness in Africa. Technical Report Working Paper 2019/22, UNU-WIDER

    2019

  3. [11]

    Daoud, A., Conlin, C., and Jerzak, C. T. (2026). Chinese vs. world bank development projects: Insights from earth observation and computer vision on wealth gains in africa, 2002-2013.World Development, 202:107328. 33 de Chaisemartin, C. and D’Haultfœuille, X. (2020). Two-way f...

    2026

  4. [12]

    C., Raschky, P

    Dreher, A., Fuchs, A., Hodler, R., Parks, B. C., Raschky, P. A., and Tierney, M. J. (2019). African leaders and the geography of China’s foreign assistance.Journal of Development Economics, 140:44–71

    2019

  5. [13]

    Dreher, A., Fuchs, A., Parks, B., Strange, A., and Tierney, M. J. (2021). Aid, China, and growth: Evidence from a new global development finance dataset.American Economic Journal: Economic Policy, 13(2):135–174

    2021

  6. [14]

    Dreher, A., Fuchs, A., Parks, B., Strange, A., and Tierney, M. J. (2022).Banking on Beijing: The Aims and Impacts of China’s Overseas Development Program. Cambridge University Press

    2022

  7. [15]

    Easterly, W., Levine, R., and Roodman, D. (2004). Aid, policies, and growth: Comment. American Economic Review, 94(3):774–780

    2004

  8. [16]

    C., and Wong, M

    Gehring, K., Kaplan, L. C., and Wong, M. H. L. (2022). China and the World Bank–how contrasting development approaches affect the stability of african states.Journal of Development Economics, 158:102902

    2022

  9. [17]

    Dontamsetti, T., and Croft, T. (2019). Integrating environmental context into dhs analysis while protecting participant confidentiality: A new remote sensing method. Population and Development Review, 45(1):197–218

    2019

  10. [18]

    Hernandez, D. (2017). Are “new” donors challenging World Bank conditionality?World Development, 96:529–549

    2017

  11. [19]

    Jablonski, R. S. (2014). How aid targets votes: The impact of electoral incentives on foreign aid distribution.World Politics, 66(2):293–330

    2014

  12. [20]

    M., Lobell, D

    Jean, N., Burke, M., Xie, M., Alampay Davis, W. M., Lobell, D. B., and Ermon, S. (2016). Combining satellite imagery and machine learning to predict poverty.Science, 353(6301):790–794

    2016

  13. [21]

    T., Johansson, F., and Daoud, A

    Jerzak, C. T., Johansson, F., and Daoud, A. (2023). Integrating earth observation data into causal inference: Challenges and opportunities. arXiv:2301.12985

    2023

  14. [22]

    MacKinnon, J. G. and White, H. (1985). Some heteroskedasticity-consistent covariance matrix estimators with improved finite sample properties.Journal of Econometrics, 29(3):305–325

    1985

  15. [23]

    J., Walsh, K., Solomon, K., Zhang, S., Elston, T., and Goodman, S

    Malik, A., Parks, B., Russell, B., Lin, J. J., Walsh, K., Solomon, K., Zhang, S., Elston, T., and Goodman, S. (2021). Banking on the belt and road: Insights from a new global dataset of 13,427 chinese development projects. Technical report, AidData at William &

    2021

  16. [24]

    and Masi, T

    Marchesi, S. and Masi, T. (2021). Delegation of implementation in project aid.The Review of International Organizations, 16(3):655–687. Öhler, H., Negre, M., Smets, L., Massari, R., and Bogetić, Ž. (2019). Putting your money where your mouth is: Geographic targeting of World B...

    2021

  17. [25]

    B., Kakooei, M., Ortheden, J., Johansson, F

    Pettersson, M. B., Kakooei, M., Ortheden, J., Johansson, F. D., and Daoud, A. (2023). Time series of satellite imagery improve deep learning estimates of neighborhood-level poverty in Africa. InProceedings of the Thirty-Second International Joint Conference on Artificial Intel...

    2023

  18. [26]

    Rajan, R. G. and Subramanian, A. (2008). Aid and growth: What does the cross-country evidence really show?Review of Economics and Statistics, 90(4):643–665

    2008

  19. [27]

    and Steendijk, R

    Smits, J. and Steendijk, R. (2015). The International Wealth Index (IWI).Social Indicators Research, 122(1):65–85

    2015

  20. [28]

    M., Dreher, A., Fuchs, A., Parks, B., and Tierney, M

    Strange, A. M., Dreher, A., Fuchs, A., Parks, B., and Tierney, M. J. (2017). Tracking underreported financial flows: China’s development finance and the aid–conflict nexus revisited.Journal of Conflict Resolution, 61(5):935—-963

    2017

  21. [29]

    and Hulme, D

    Yanguas, P. and Hulme, D. (2015). Barriers to political analysis in aid bureaucracies: From principle to practice in dfid and the World Bank.World Development, 74:209–219. 35 8 Appendix 8.1 Primary treatment-definition robustness Table A.1: Treatment-definition alternatives Ch...

    2015

Pith tools

Reviewed June 27, 2026 · model on record in the stance chip above.