Research Questions and Answers

Raising Academic Achievement in Under-Performing Schools through Improved Management:A Research-Practice Partnership

Winner of the William T. Grant Foundation’s Institutional Challenge Grant co-sponsored by the Spencer Foundation.

Principal Investigator: Gustavo J. Bobonis | Department of Economics | University of Toronto


Marco Gonzalez-Navarro | Department of Agricultural and Resource Economics | University of California-Berkeley

Daniela Scur | Dyson School of Applied Economics and Management | Cornell University

Orlando J. Sotomayor | Department of Economics | University of Puerto Rico-Mayagüez


Improvements in management practices have been credited with generating striking productivity improvement in the private sector. Although their adoption in the education sector has been slow, management-level interventions have been shown to lead to significant improvements in school quality. Promoting improvements in school management practices is thus fitting for policy, given the need to identify ways to achieve this sustainably and at scale. Drs. Bobonis, Gonzalez-Navarro, Scur, and Sotomayor are partnering with the Puerto Rico Department of Education to conduct policy research into the role that improved management practices and organizational structures have in promoting educational quality. They are conducting a prospective evaluation of a large-scale principal training program being implemented over the 2019-2022 academic years, collecting detailed longitudinal data to measure short- and long-term adoption, as well as using standardized test scores data to assess the impact on student achievement. A second project is the evaluation of the consequences of past school consolidations on student achievement and a range of other outcomes, taking advantage of newly available administrative longitudinal data linking information on students, teachers, and schools. Finally, the partnership is establishing a systematic approach for the development of additional joint research-practice projects to promote the use of research and evidence in public policy in Puerto Rico.

Can poor uterine environments cause diabetes, high cholesterol, hypertension, and cardiovascular disease in adult life?

Strong identification strategy establishes causal link between poor uterine environment and all three Syndrome X conditions, and doing so using two natural experiments.


Shocks generated by two powerful tropical storms striking Puerto Rico during the late 1920s and early 1930s had long-term consequences consistent with the fetal origins hypothesis. Individuals in the womb or early infancy in the aftermath of each of the storms are more likely to report a diagnosis of hypertension, high cholesterol, diabetes, and are considerably more likely to have no formal schooling.

Was the extraordinary beginning-of-the century decline in Brazilian poverty and inequality associated with public policy or with economic growth?

Twenty years of thinking about Brazilian poverty, wage, and income distribution.


A comprehensive decomposition approach is applied to identify the factors that generated one of the most remarkable episodes of redistribution in recent memory, whereby in a span of a decade Brazilian inequality fell by one-fifth and the incidence of poverty declined by two-thirds. It is argued that important elements related to these developments had been in place for some time. These included macroeconomic stability, long-standing increases in schooling attainment, and favorable demographics in the form of a declining dependency ratio that proved to be both poverty- and inequality-reducing. Robust economic growth during the 2000s and the early 2010s resulted in across-the-board income gains that were widely shared, owing to mechanisms that favored advances at the lower end of the distribution of earnings. Demographics, as well as changes in educational attainment, labor force participation, and lower-skill prices explain most of the significant drop in the headcount ratio, with education and interaction between economic growth and a rising wage floor accounting for the bulk of the change. The same factors explain up to 85% of the decline in income dispersion that was especially driven by markedly compressed earnings differentials. Human capital accumulation and strong labor market institutions thus stand out as key mechanisms linking economic growth to redistribution.

Is the minimum wage an effective poverty- and inequality-reducing instrument?

Developed country evidence is not supportive. However, …


Even though there is growing social support for higher minimum wages as anti-poverty policy tools, very little is known about their effectiveness in reducing poverty or inequality in the developing world. Latin America’s largest economy offers a fertile setting for shedding light on the issue, in being a large and data-rich country where frequent increases in the minimum wage can allow for direct estimation of influence on the distribution of income. Using a difference-in-difference estimator that takes advantage of substantial regional income variation and 21 increases in the Brazilian national wage floor, the study finds that within three months of these minimum wage hikes, poverty and inequality declined by 2.8% and 2.4%, respectively. Influence waned over time, particularly with respect to bottom-sensitive distribution measures, a development that is consistent with resulting job loses that fell more heavily among poorer households. The fact that the following annual hike in the minimum wage led to a renewed decline in poverty and inequality, suggests that potential unemployment costs were again overwhelmed by benefits in the form of higher wages among working individuals. However, evidence also establishes an inelastic relationship between wage floor hikes and changes in the incidence of poverty, as well as diminishing returns to the strategy when the legal minimum is high relative to median earnings.

How does health insurance improve health?

The endogenous nature of health insurance coverage makes it difficult to establish causal effects on health, but quasi-experimental evidence establishes widespread and lasting effects on the near-elderly population.


An eight percentage point increase in the share of the near-elderly covered by insurance leads to almost immediate declines in self assessments of poor health that range from 7% among the population diagnosed with cancer to 25\% among that diagnosed with angina pectoris. Increased coverage is tied to a reduction in activity limitations related to a broad set of medical conditions, with the largest effect being a 40% reduction in the population limited by depression. Preventive and diagnostic procedures rise, and risky behaviors such as smoking and being overweight decline. Enhanced chronic disease management is also manifested through improved blood sugar control among diabetics.

Does increased importance of women in the labor market make for a more equal or a more unequal distribution of income?

Research did not get a lot of attention but to this day, goes a long way towards explaining the dynamics of income distribution in Brazil.


Substantial changes in the structure of female earnings reflecting rising wages and participation rates had equalizing and considerable poverty-reducing consequences that counterbalanced unequalizing and poverty-increasing changes in the structure of male earnings. In the context of a stable macroeconomic environment inequality and particularly poverty fell as a result of rising earnings among women at the lower end of the household income distribution.

Does growth alleviate poverty?

Once, the relationship was very much in doubt.


The study finds an inelastic link between poverty and wages over the macroeconomic cycle that is devoid of the asymmetric effects that characterise the phenomenon’s response to changes in the rate of unemployment. Deepening of structural reforms in the second half of the 1990s had no effect on the aforementioned relationships, which also varied little by regional level of development. Finally, the connection between poverty and long-term income changes was more elastic, suggesting an association between the economy’s recent disappointing performance in alleviating poverty and its lacklustre growth record over the past two decades.

Can education reduce inequality?

Most people care about inequality but the central issue is what can be done about it.


In countries with high levels of inequality, progress in education has often been placed high in the list of policy proposals designed to change the unequal state of affairs. This study uses Brazilian annual data to chart the trends in wage inequality and a decomposition procedure to ascertain how wage distribution was affected by advances in education that doubled the median schooling level of prime-age men during the sample period. The results show that while falling returns to education were an equalizing factor across the wage distribution, changes in the educational composition of the work force exerted a disequalizing influence that was strongest at the top of the distribution. The net result was a fall in wage dispersion that was not as dramatic as might have been hoped.

What explains the failure of Puerto Rican economic development?

Any hope of future growth hinges on either increasing labor productivity or reforming institutions inconsistent with a low-productivity environment.


Abstract Puerto Rico’s long running experiment with active wage floor and public transfer policies manifests substantial program benefits at great efficiency costs. Transfers have sheltered large segments of the population from a deep and prolonged recession and in past decades been important instruments of positive distributive change. They have also curtailed growth potential, reducing the effective labor force through retirements that occur at very early ages and through limited market entry among the low-skilled. A high wage floor has in the past driven large increases in productivity and real wages, but has also resulted in structural unemployment and appears behind a recent marked loss in competitiveness concentrated on the lower half of the earnings distribution. Resulting policy design interaction has contributed to an economy with scarce job opportunities and one of the lowest workforce participation rates in the world.

Why is working the exception rather than the rule in Puerto Rico?

The Anne Krueger report and two New York Fed economic conclusions are partly based on this research.


A large part of income gap difference between Puerto Rico and the United States can be traced to employment rate differences between the two regions. Although participation rates did not differ much in the early 1950s, by the early twenty-first century the difference was large. The rapid expansion of government transfers in the 1970s and early 1980s produced these effects in Puerto Rico.

Has economic growth reduced poverty and inequality in Puerto Rico?

Low labor force participation is the root cause of inequality that is a high as in the rest of Latin America.


A remarkable spell of economic development in the 1950s and 1960s generated an ambiguous long-term change in inequality but a drastic reduction in poverty. Thereafter, public transfers played a decisive role in restraining growth in inequality and in reducing poverty levels that were also affected favorably by changes in the distribution of female earned income. Although rising schooling levels and their absolute returns increased households’ income generating capacities, the realization of the potential was hindered by increasing levels of unemployment and economic inactivity.

Has there ever been a brain drain from Puerto Rico?

An empirical test of the Borjas migration model assumptions.


Census data and a semi-parametric procedure for creating counterfactual distributions are used for analyzing migration patterns between two economies with large differences in average wages, wage dispersion, and returns to education. Results show that migration flows to the United States have been composed of individuals from the middle of the distribution of skill, thus supporting a process of intermediate selection. In the other direction, higher inequality and returns to education in the island attract migrants drawn from the top of the distribution. Findings are consistent with a selectivity model where time-equivalent migration costs are large and declining in skill.