Sorting, Matching and Economic Complexity

Assignment models in trade predict that countries with higher productivity levels are assortatively matched to industries that make better use of these higher levels. Here, we assume that the driver of productivity differences is the differential distribution of factors among countries. Utilizing such a structure, we define and estimate the average factor level (AFL) for countries and products using only the information about the production patterns. Interestingly, our estimates coincide with the complexity variables of (Hidalgo and Hausmann, 2009), providing an underlying economic rationale. We show that AFL is highly correlated with country-level characteristics and predictive of future economic growth.

Lockdown Fatigue: The Diminishing Effects of Quarantines on the Spread of COVID-19

Non-Pharmaceutical Interventions (NPIs) have been for most countries the key policy instrument utilized to contain the impact of the COVID-19 pandemic. In this article, we conduct an empirical analysis of the impact of these policies on the virus’ transmission and death toll, for a panel of 152 countries, from the start of the pandemic through December 31, 2020. We find that lockdowns tend to significantly reduce the spread of the virus and the number of related deaths. We also show that this benign impact declines over time: after four months of strict lockdown, NPIs have a significantly weaker contribution in terms of their effect in reducing COVID-19 related fatalities. Part of the fading effect of quarantines could be attributed to an increasing non-compliance with mobility restrictions, as reflected in our estimates of a declining effect of lockdowns on measures of actual mobility. However, we additionally find that a reduction in de facto mobility also exhibits a diminishing effect on health outcomes, which suggests that lockdown fatigues may have introduce broader hurdles to containment policies.

Podcast: Do Lockdowns Work? Eduardo Levy Yeyati discusses the research with Sam Munson of the Octavian Report.

Inequality, Openness, and Growth through Creative Destruction

We examine how inequality and openness interact in shaping the long-run growth prospects of developing countries. To this end, we develop a Schumpeterian growth model with heterogeneous households and non-homothetic preferences for quality. We show that inequality affects growth very differently in an open economy as opposed to a closed economy: If the economy is close to the technological frontier, the positive demand effect of inequality on growth found in closed-economy models may be amplified by international competition. In countries with a larger distance to the technology frontier, however, rich households satisfy their demand for high quality via importing, and the effect of inequality on growth is smaller than in a closed economy and may even be negative. We show that this theoretical prediction holds up in the data, both when considering growth in export quality at the industry level and when considering growth in GDP per capita.

The Economic Case for Global Vaccinations: An Epidemiological Model with International Production Networks

COVID-19 pandemic had a devastating effect on both lives and livelihoods in 2020. The arrival of effective vaccines can be a major game changer. However, vaccines are in short supply as of early 2021 and most of them are reserved for the advanced economies. We show that the global GDP loss of not inoculating all the countries, relative to a counterfactual of global vaccinations, is higher than the cost of manufacturing and distributing vaccines globally. We use an economic-epidemiological framework that combines a SIR model with international production and trade networks. Based on this framework, we estimate the costs for 65 countries and 35 sectors. Our estimates suggest that up to 49 percent of the global economic costs of the pandemic in 2021 are borne by the advanced economies even if they achieve universal vaccination in their own countries.

Uncertainty in the Search for New Exports

This paper explores the role that uncertainty plays in the emergence of new products or services for export in developing countries. Using a comparative case study method, I explore the degree to which those entrepreneurs who discovered new export activities faced uncertainty, and what the nature of this uncertainty was. I then document how this uncertainty, when present, was resolved, and how this affected subsequent diffusion of the newly discovered activity. The cases suggest two important dimensions of uncertainty in the emergence of new export activities: productivity characteristics and demand characteristics. A new activity could feature one, both, or neither types of uncertainty. The reasons for lower inherent uncertainty in these cases suggest a new theory of product similarity that is heterogeneous, multi-dimensional, and operating at a highly disaggregated level. Furthermore, the degree of uncertainty has implications for the expected ‘triggers’ of discovery, and these are born out in the cases. Finally, when uncertainty was present, its resolution often provided significant benefits to subsequent entrants, and the manner in which high uncertainty was overcome suggests potential avenues for policy.

Finance as the Binding Constraint to Growth

Finance forms a necessary input for production, one so central that it was placed atop the decision tree in the original Growth Diagnostics framework. As we argue, one of the thorniest findings from more than a decade of practice in conducting Growth Diagnostics has been that it is often more difficult to disprove a finance constraint than it is to prove one. Finance has often earned more attention than deserved when considering the many complementary inputs that must be present for production to take place and investments to be profitable. The challenge is in getting the diagnostic right, starting with the use of sound evidence to test for signals.

This paper revisits the starting question of the Growth Diagnostic framework: what does it mean for finance to be a constraint to economic growth? We provide an updated, detailed decision tree for finance, including a rethink of other sources of finance constraints, such as insufficient equity, that were not fully considered in the original decision tree. Our starting point to test for the presence of a finance constraint is to recognize that every financial system suffers from asymmetric information. While information is important for almost all assets in economic transactions, in financial markets, information is the asset. The inherent nature of information asymmetries to financial markets is, in part, what makes finance a focal point for constraint analysis, as greater size and sophistication of financial systems do not make a country immune to finance constraints.

We present three reasons that finance may be constrained: a) insufficient aggregate savings, due to a both inadequate domestic savings and restricted access to foreign borrowing, resulting in not enough loanable funds to finance good projects; b) inadequate institutions and tools for assessing and mitigating risk, that are unable to resolve information asymmetries, preventing markets’ access to savings; and c) problems in financial intermediation, where intermediation itself may be high-risk, monopolistic, or otherwise inefficient to result in insufficient bank lending, or may face borrowers who lack sufficient equity. The paper aims to share lessons learned in testing whether finance is constrained – or not, as well as the policy space to address a finance constraint. The policy discussion emphasizes the risk of misclassifying finance as a constraint when it is not binding on production, as the alternate response of overregulating financial markets can create new intermediation failures to the trust between savers and borrowers. Ultimately, we conclude that policy responses to a finance constraint must be as context specific as the syndrome presented by the diagnosis, where creating functional financial markets lies in preserving the delicate balance of trust between savers and borrowers.

This publication is part of the Mindbook Paper Series.

Last updated on 06/12/2025

Coordination Failures in Adopting New Technological Capabilities as the Binding Constraint to Growth

The process of structural transformation that has accompanied economic success stories requires an expansion of the technological capabilities held in society. Adding new technological capabilities faces several constraints related to coordination failures, information failures, and the asymmetric nature of technology itself. Although these coordination failures were included in the original Growth Diagnostics framework, practitioners have often found them challenging to analyze. This paper aims to provide a systematic framework and analytic techniques that bring clarity and rigor to the examination of potential constraints in this branch. We posit four different strategic approaches that countries face in the process of structural transformation, centering on two factors: are existing technological capabilities sufficient for growth? And: how easy is it to add the new technological capabilities required to develop new productive sectors? Countries that lack sufficient existing capabilities and must add several capabilities at once in order to enter higher-productivity sectors may be constrained by a capability trap resulting from coordination failures. Even for places where promising opportunities exist, they may be constrained by “low jumpiness” related to information externalities in the process of self-discovery. Diagnostic tests are detailed that can identify the necessary strategic approach. The paper also reflects on the policy space for addressing coordination constraints and outlines the central role of the public sector in enabling coordination of technological capabilities. Both public-private coordination as well as underappreciated elements of public-public coordination in the provision of public goods are addressed. 

This publication is part of the Mindbook Paper Series.

Last updated on 05/30/2025

Buscando virtudes en la lejanía: Recomendaciones de política para promover el crecimiento inclusivo y sostenible en Loreto, Peru

Loreto es un lugar de contrastes. Es el departamento más grande del Perú, pero se encuentra entre los de menor densidad poblacional. Su capital, Iquitos, está más cerca de los estados fronterizos de Brasil y Colombia que de las capitales de sus regiones vecinas en el Perú – San Martín y Ucayali. Sólo se puede llegar a Iquitos por vía aérea o fluvial, lo que la convierte en una de las mayores ciudades del mundo sin acceso por carretera. Desde la fundación del departamento, la economía de Loreto ha dependido de la explotación de recursos naturales, desde el boom del caucho a finales del siglo XIX y principios del XX hasta la extracción petrolera y explotación de recursos forestales que predomina en nuestros días. Este modelo ha traído consigo daños ambientales significativos y ha producido un patrón de crecimiento lento y volátil, que ha abierto una brecha cada vez más amplia entre la economía de la región y la del resto del país. Entre 1980 y 2018, Loreto creció a una tasa promedio compuesta anual cuatro veces menor a la del resto del Perú. Es decir, mientras el resto del Perú triplicó el tamaño de su economía, la de Loreto creció algo menos que un tercio.

En la última década (2008-2018), la región también se ha venido distanciando de sus pares amazónicos en el país (Ucayali, San Martín y Madre de Dios), que han crecido a una tasa promedio anual cinco veces mayor. En este período, el ingreso promedio por habitante en Loreto ha pasado de ser tres cuartas partes del promedio nacional en 2008 a menos de la mitad para 2018. Además del rezago económico – o quizás como consecuencia de él -, Loreto también se ubica entre los departamentos con peores indicadores de desarrollo social, anemia y desnutrición infantil del Perú.

En este contexto, el Laboratorio de Crecimiento de la Universidad de Harvard se asoció con la Fundación Gordon and Betty Moore para desarrollar una investigación que proporcionara insumos y recomendaciones de política para acelerar el desarrollo de la región y generar prosperidad de forma sostenible.

Diagnóstico de Crecimiento de Loreto: Principales Restricciones al Desarrollo Sostenible

Sembrado en el flanco oeste de la selva amazónica, Loreto se encuentra entre los departamentos más pobres y con peores indicadores sociales del Perú. El desarrollo enfrenta allí un sinfín de barreras, pero no todas son igualmente limitantes y tampoco hay recursos para atender todos los problemas a la vez. El Laboratorio de Crecimiento de la Universidad de Harvard, bajo el auspicio de la Fundación Gordon and Betty Moore, ha desarrollado un Diagnóstico de Crecimiento que buscar identificar las restricciones más limitantes, y priorizar las intervenciones de políticas públicas alrededor de un número reducido de factores con el mayor impacto. La investigación, que se fundamenta en análisis de bases de datos nacionales e internacionales, e incluye factores cuantitativos y cualitativos derivados de las visitas de campo, identifica a la conectividad de transporte, los problemas de coordinación asociados al autodescubrimiento, y la energía eléctrica, como las restricciones más vinculantes para el desarrollo de Loreto. De acuerdo con nuestras conclusiones, mejoras en la provisión de estos tres factores tendrían un mayor impacto sobre el desarrollo sostenible de la región que mejores en la educación y los niveles de capital humano, el acceso a financiamiento, y otros sospechosos habituales. Este reporte es el segundo de una investigación más amplia – Transformación estructural y restricciones limitantes a la prosperidad en Loreto, Perú – que busca aportar insumos para el desarrollo de políticas públicas a escala nacional y regional que contribuyan a promover el desarrollo productivo y la prosperidad de la región.

On Globalization and the Concentration of Talent: A General Result on Superstar Effects and Matching

We analyze how globalization affects the allocation of talent across competing teams in large matching markets. Focusing on amplified superstar effects, we show that a convex transformation of payoffs promotes positive assortative matching. This result holds under minimal assumptions on how skills translate into competition outcomes and how competition outcomes translate into payoffs. Our analysis covers many interesting special cases, including simple extensions of Rosen (1981) and Melitz (2003) with competing teams. It also provides new insights on the distributional consequences of globalization, and on the role of technological change, urban agglomeration, and taxation for the composition of teams.