Research
Working Papers
Crowdout and Competition: How non-governmental development programs shape public spending
Demonstrates that non-governmental development programs reduce public service spending, but that electoral competition dampens the incentive to shirk.
Abstract
A large literature explores the effect of aid on public spending, showing that grants substitute for local revenue in some contexts but not others. However, development assistance is frequently provided by aid agencies and non-governmental organizations (NGOs) directly to citizens and alongside public institutions with similar mandates to improve human welfare. I argue that non-governmental development programs also crowd out public spending, but that democratic institutions shape the incentive to shirk. I leverage a randomized evaluation of a World Bank community development program from the Philippines to substantiate these claims and show that crowd-out attenuates returns to human welfare from non-governmental development programs. I also apply supervised machine learning to explore budget reallocation and offer evidence that such programs reduce local tax collection and increase discretionary spending. These results document a significant but unexplored channel in the relationship between aid and political institutions and clarify a long-standing empirical puzzle: aid shapes public spending, but democratic incentives determine whether this complements or corrodes the state.
Suggested citation
Finstein, Blaine. Crowdout and Competition: How non-governmental development programs shape public spending. Working paper.
Working paper
Current Projects
Latent variable measurement under asymmetric observability
Develops a framework for latent variable estimation when labels are observed asymmetrically (positive-only) through a non-random detection process.
Abstract
Many political behaviors of interest shape relational data systematically, are only partially observed, and are detected through political processes correlated with the outcome itself. Examples include corruption, repression, and vote buying. This presents a fundamental inference problem: observed labels are one-sided (positive-only), non-observation is uninformative, the relational structure contains information about the latent trait, and conventional supervised and unsupervised methods fail. Standard supervised learning introduces bias when labelled data are non-random, and unsupervised learning cannot anchor latent features to substantively meaningful characteristics.
I introduce a general framework for latent variable measurement under asymmetric label noise in relational data. I propose a two-layer measurement model with three components: (1) unsupervised relational embedding to capture systematic structure induced by the latent trait in the network, (2) latent variable mapping from embeddings to latent trait probability, and (3) an observation model that assumes detection is determined by a biased process conditional on the latent trait. I show that a joint relational-detection model recovers latent traits more accurately than standard supervised and unsupervised approaches under asymmetric observability. I do so by simulating networks where a latent trait affects tie formation and detection depends on covariates to compare supervised classifiers, unsupervised embeddings, and my proposed estimator. I then apply the method to a 25-year panel of municipal public procurement in the Philippines, a case where corruption shapes contracting but detection is determined politically.
Sweetening the Deal: Colonial sugar cultivation and the historical roots of elite capture
Coauthors: Jose Gloria (ULCA)
Explores how historical economic processes shape modern elite capture through local financial networks.
Abstract
This paper argues that agricultural endowments shaped local financial networks and created enduring opportunities for elite capture. In the Philippines, sugar mills served as economic chokepoints that tied farmers’ livelihoods to local landed elites. This dependence fostered durable patron–client relationships that helped consolidate political and economic power over time. These historical dependencies produced more centralized elite networks in sugar-growing regions, compared to the more fragmented structures found elsewhere. The study combines historical data on sugar mills with municipal-level electoral networks, linking political clans through shared family names. Preliminary results show that municipalities historically organized around sugar chokepoints are more likely to capture disaster aid during crises, suggesting that entrenched elite coordination continues to shape the distribution of public resources in the present day. The analysis uses network measures of centralization and modularity, supported by econometric strategies that leverage variation in exposure to structural and programmatic shocks.
Selected Other Publications
A Comment on “Extraction, Assimilation, and Accommodation: The Historical Foundations of Indigenous-State Relations in Latin America”
Coauthors: Konstantin Ash (University of Central Florida) and Daniel Carnahan (UCLA)
Replicates Carter (2024) as part of the Institute for Replication discussion paper series and meta study.
Abstract
Carter (2024) examines the historical conditions that shape protection versus assimilation for indigenous communities, arguing that state-led conscription programs are one such factor. In a natural experiment leveraging conscription for a 1920s Peruvian highway designed to replicate a pre-colonial road system (Qhapaq Ñan), Carter finds through a geographic regression discontinuity design that eligibility for state conscription increased the likelihood of a municipality having an indigenous movement by about 30 percentage points (approximately .75 standard deviations) and scores on an omnibus accommodation measure by about .3 items (approximately .4 standard deviations). The omnibus measure includes the number of institutions that an indigenous community reports preserving (increased by .3 items on a 7 point scale, or .25 standard deviations), likelihood of having a communal land title (increased by 12 percentage points, or .3 standard deviations), and likelihood of registration with the government (increased by 9 percentage points, or .3 standard deviations). All point estimates are significant at the .1% level.
We successfully computationally reproduce all main claims of the paper but find inconsistencies between the map of the road presented by Carter and that used by Franco et al. (2021) that affect its passage through a small number of municipalities. In order to investigate whether these municipalities drive the main findings without the ability to identify municipalities in the data, we drop municipalities iteratively and re-run the analysis, finding only minor changes in coefficient estimates across subsets. In addition, we explore a number of sensitivity analyses for the regression discontinuity design that vary the functional form, vary the bandwidth window, and use the Rosenbaum method for window selection. While the results remain consistent under all analyses, we recommend for further research to recode treated municipalities on the basis of the alternative road map and explore the as-if random assumption in light of evidence linking proximity to the precolonial road to various economic and political outcomes.
Suggested citation
Finstein, Blaine; Ash, Konstantin; Carnahan, Daniel (2024) : A Comment on “Extraction, Assimilation, and Accommodation: The Historical Foundations of Indigenous-State Relations in Latin America”, I4R Discussion Paper Series, No. 176, Institute for Replication (I4R), s.l.
Paper