The heterogeneous legacy of rebel and army violence on social capital: evidence from the civil war in Peru (2018, With Eduardo Malasquez) [Submitted]
This paper exploits the progression of the 1980-2000 conflict in Peru over time and space to show that exposure to government violence reduces citizen’s perceptions on the importance of voting and increases the likelihood of voting null, while it reduces participation in civil organizations. The effects vary by the identity of the perpetrator, since the results show that exposure to government violence has negative effects on voting perceptions and behavior, while violence from terrorist groups has a negligible or smaller impact. In addition, our results suggest that the late teens and early twenties is a key period to influence later voting behavior. Finally, results reveal a negative relation between exposure to government violence during childhood and teenage years with the level of participation in political and social organizations, while exposure to violence from terrorist groups leads to higher levels of participation in the same organizations.
Female Genital Cutting and Education: Theory and Causal Evidence from Senegal (2018, with Jorge Garcia Hombrados)
Using data from Senegal we study the causal eect of female genital cutting (FGC) on education. The identication strategy exploits across ethnic-group variation in exposure to a law that in January 1999 banned the practice of FGC. The results show that, in a context where girls are cut during infancy and early childhood, experiencing FGC reduces signicantly the years of education and the probability of ever attending school of the girls. We propose a theoretical model to understand this result and the mechanisms, which suggests that households react to the increase in the cost of FGC caused by the introduction of the anti-FGC legislation through abandoning such practice and compensating uncut girls with larger levels of educational investments to avoid potential losses in the marriage market. On the other hand, the results show that the eect on education is not driven by any lasting health eect of FGC.
The effect of R&D Growth on Employment and Self-Employment in UK Local Labour Markets (2018, with Tommaso Ciarli, Alberto Marzucchi and Maria Savona)
This paper investigates the effects of firms’ investment in Research and Development (R&D) on employment dynamics in the British local labour markets (Travel to Work Areas). We distinguish between local areas characterized by high and low routinisation of the workforce. We implement a robust instrumenting strategy to address endogeneity issues in the relation between innovation and employment. Our results suggest that increases in R&D investments mainly affect routinized areas, where the employment created is low skilled, concentrated in non-tradable sectors (like transport, construction) and services. A significant share of the jobs created is (unincorporated) self-employment, concentrated in the 25-34 age cohort. We qualify the effect of R&D on self-employment by looking at local firms’ dynamics, which suggest that the increase in self-employment is reflected in a higher number of micro-firms. Rather, in non-routinized areas, R&D results in the expected increase in the demand of high-skilled workers and a reduced demand of low-skill employment.
Do Low-WageWorkers Benefit from the Recovery of Productivity Growth? (2018, with Tommaso Ciarli and Maria Savona)[Joseph Rowntree Brief]
This work empirically examines whether productivity gains (loss) are shared with wages and the sources of heterogeneity that characterise this link. We use matched employer-employee data from the UK Annual Survey of Household Earning and the Annual Business Survey from 2011 to 2015. We instrument Labour Productivity (LP) with real Total Factor Productivity, and use matched worker-firm fixed effects (FE) to identify the causal effect of productivity changes on wages. We consider LP growth at the level of firm, industry and city as affecting wages of continuing employees. We look at the effects on wages across different age cohorts (16-24, 25-34, 35-65); wage quintiles; and for unionised workers. We carry out the analysis on the whole UK economy and separately on the manufacturing sector, and several services (construction, trade, transport, financial, business services and creative industries). We find that, overall, only the (nominal) wage elasticity to industry-level productivity is statistically significant, very small and negative: -0.03. The elasticity is nil in manufacturing and negative, albeit very small, in business services. The overall negative wage elasticity to industry productivity is found to be concentrated in the 35-65 age cohort. Unionisation at the industry level is found to reduce the rent sharing elasticity in the UK. These findings open a number of relevant policy issues that can valuably add to the debate in the UK, for instance on the Industrial Strategy, on how to identify policy tools that improve productivity and lift wages at once.
Technical Innovation Shocks, Paid Employment and Self-Employment: An Spatial Equilibrium Setting for Peru (2017)[paper available upon request]
Using data from 167 Peruvian cities from two periods, 1994 and 2007, I find that technological innovation,measured as local TFP shocks, increase employment for medium skilled workers in the paid sector, while it increases employment of high and medium skilled workers in the self-employment sector. In line with standard spatial equilibrium theory, wages for the low skilled workers in the paid sector increase more with respect to high skilled. This implies that a productivity shock closes the skill gap among employees. Surprisingly, the opposite happens in the self-employed sector, where local TFP shocks produce a decline in the earnings of low skilled individuals. I also find that productivity shocks reduce local manufacture employment. A result widely found in industrialized countries, but for the first time in a developing country.
This paper evaluates the effect of conflict over the formation of trust and identity. It finds that Peruvian individuals exposed to violent events during their impressionable years trust less government institutions, and feel less identified with their neighbors, while more identified with religious groups. The effect on identification is heterogeneous by the indigenous origin of the individuals. Individuals who own an agricultural plot embedded in a communal arrangement at the local level exhibit even smaller levels of identification with their local neighbors while higher levels of identification with their ethnic group. In line with recent literature, these findings suggest that conflict has a small but persistent effect on the formation of trust and identity, which is a central feature to understand the interaction between culture and institutions, and ultimately to understand the persistent consequences of wars
Using two rounds of population census for 1043 districts in Peru I document that large-scale mining activity had a positive effect on local employment over 14 years. The effect is differentiated by industry, skill and migration status. Employment grew by 0.04 percentage points faster by one standard deviation increase in the mineral prices. Both high and low skilled workers enjoyed similar employment increase, however only low skilled workers experienced a decline in unemployment. Using data from 10 annual household surveys I find that, consistent with a model of heterogeneous firms and labor, wages for low skilled workers in districts close to the mining activity was 0.05 percentage points higher by every standard deviation increase in the index of mineral prices. Additional evidence with the census data suggests that locals working in the mining or the agricultural sector filled the new employment opportunities. More evidence suggests that mobility costs and not the elasticity of substitution between high and low skilled workers or skill acquisition may explain the outcome. Together these findings suggest that large-scale mining activity increases the demand for mining and agricultural local employment, and the wages in the local economy.
The consolidation of democratic regimes is influenced by the prevalence of positive beliefs on democracy. In this paper I shed light on what determines the democratic beliefs. In particular, I explore the uncertainty experienced during the impressionable years of the individuals as a key factor behind the formation of the beliefs. Results show that this type of uncertainty has no effect on the determination of democratic beliefs. Combining uncertainty with the exposure to authoritarian regimes does not change the result. This result is robust to different definition of rural individuals, the interaction of uncertainty and degree of experienced authoritarianism, and different formative periods. Current uncertainty, on the other hand, is unable to fully explain the formation of democratic beliefs
Regional Financial Development and Firm Growth in Peru (2013, with Eduardo Morón and Cristhian Seminario):
This paper documents the relationship between regional financial development and firm growth in the Peruvian manufacturing sector. In order to control for mutual causality between credit availability and firm growth, industry differences in financial dependence on external funds are exploited. The 1994 and 2008 rounds of the National Economic Census are used, permitting analysis at the firm level as well as the activity level. Results suggest a significant and positive effect of financial deepening on surviving firms` growth. However, this effect is smaller for micro enterprises, suggesting that the cost of external funding decreases with financial development mainly for large firms. The conclusions remain unchanged when entering and exiting firms are included. The paper further finds that credit expansion have encouraged not only firm growth but also firm entry. The results are robust using an alternative measure of financial dependence.
Financial Dependence, Formal Credit and Firm Informality: Evidence from Peruvian Household Data (2012, with Eduardo Morón and Cristhian Seminario):
This paper examines the link between financial deepening and formalization in Peru. Using data from the National Household Survey, Bloomberg and the Central Bank of Peru Central Bank, the Catão, Pagés, and Rosales (2009) model is implemented at activity level (2-digits ISIC), and the Rajan and Zingales (1998) approach of sectors’ dependence on external funds is followed. The sample is divided into three firm size categories, and two formality measures are assessed. Using the accounting books specification, robust results are obtained, supporting a significant and positive effect of credit growth on formalization only for the self-employment firms category. Alternatively, using the pension enrollment specification, the channel is found positively significant only for firms with more than 10 workers; there is a smaller effect for firms with 2-10 workers. There is also a significant between effect, explaining the transition from small firms to larger firms due to greater credit availability.
Productividad, Aglomeración y Asignación de Recursos en el Perú: ¿Qué dice el Censo Económico Nacional 2008? (2008, with Eduardo Morón and Cristhian Seminario):
Con data del Censo Económico 2008 se calcula la productividad a nivel de establecimiento utilizando los enfoques de Foster et al. (2008) y Hsieh y Klenow (2008). Los resultados son distribuciones similares para cada sector. A partir de estadística descriptiva y un análisis de correlaciones se exploran potenciales determinantes (a nivel de planta, distrito y departamento) de la productividad. La dispersión encontrada en manufactura es comparable a otros países de la región y su ineficiente asignación de recursos configura un estimado de potenciales ganancias de productividad de 25 por ciento.
Huelgas en el Perú: Determinantes económicos e institucionales (2006, with Gustavo Yamada):
Este trabajo comprueba el carácter procíclico de las huelgas en el sector privado peruano; es decir, la actividad huelguística se incrementa conforme la economía experimenta fases expansivas y se retrae en épocas de recesión económica. En un contexto expansivo, los trabajadores sentirían un mayor espacio para negociar incrementos en sus beneficios económicos y un menor temor por represalias y pérdidas del puesto de trabajo. Lo contrario ocurriría en momentos de recesión económica. En segundo lugar, el trabajo encuentra que la actividad huelguística es afectada importantemente por los cambios en la legislación laboral peruana. Concretamente, con la reforma laboral de 1992 se redujeron todos los indicadores de huelgas, en todos las fases del ciclo económico. Más aún, este estudio constata que la elasticidad de respuesta de las huelgas al estado de ciclo económico ha caído drásticamente (aunque conservando el signo positivo) luego de la reforma laboral de 1992. En términos técnicos, se halla un cambio tanto en el intercepto como en la pendiente de la regresión estimada. Los resultados del estudio nos permiten simular el nivel de actividad huelguística que hubiese prevalecido en el Perú de no haberse efectuado la reforma laboral de 1992. De esta forma, se estima que en nuestro país dejaron de perderse en promedio más de 15 millones de horas hombre anuales y cerca de 300 mil trabajadores comprendidos, como producto del cambio en la legislación laboral.