Calendrier du 13 décembre 2022
PSI-PSE (Petit Séminaire Informel de la Paris School of Economics) Seminar
Du 13/12/2022 de 17:00 à 18:00
R1-13
ABOYA Nakita ()
E-filling, corruption and taxation in developing countries : case study of Cameroon
Paris Trade Seminar
Du 13/12/2022 de 14:45 à 16:15
SANTAMARIA Marta (U. Warwick)
CANCELLED - Understanding Home Bias in Procurement: Evidence from National and Subnational Governments
écrit avec Manuel García-Santana (World Bank)
Are governments locally biased when buying goods and services from private firms? And if so, to what extent can this home bias explain the lack of integration in procurement markets across regions and countries? Using more than one million public procurement contracts awarded by 30,000 government agencies in French and Spanish regions, we explore the hypothesis that a government’s home bias depends on the geographical level at which the government operates. To test our hypothesis, we classify agencies into national agencies (i.e., the central or federal administration) and subnational agencies (i.e., an individual region or territory within the country). We then identify the relative home bias across governments with a novel identification strategy that relies on observing the same establishment selling the same product to national and subnational governments located in the same destination, controlling for firm and origin-destination characteristics. Our results show that the government’s home bias, especially that of subnational agencies, explains a big part of the high levels of local concentration in government procurement across regions and countries. Our findings point towards significant inefficiencies in the allocation of government procurement expenditure across firms, regions, and countries within the European Union.
Applied Economics Lunch Seminar
Du 13/12/2022 de 12:30 à 13:30
Salle R2.21, Campus Jourdan
HARHOFF ANDERSEN Lars (University of Copenhagen)
Collective memory: Evidence from first names
écrit avec joint with Tom Raster (PSE)
How do historical events change individuals' values, and when do they enter collective memory? A fundamental problem is measuring values when surveys are lacking. To tackle this problem, a promising recent literature uses first names as proxies for values, however, we argue and demonstrate that the scope and systematization of this approach can be broadened. To this end, we develop new methods to derive a broad set of values from first names. Our methods leverage parents’ tendency to name their children for groups in society that they sympathize with because those groups represent their values. Drawing on all available US full-count censuses, we use professions (e.g., military personnel, entrepreneurs, or scientists) as indicators for different values (e.g., militarism, entrepreneurialism, or scientific values). For each first name in the censuses, we measure how associated it is with a given profession and, thus, which values it represents. We then merge the value measure of first names with the names parents give their children. This provides us with proxies of values for millions of US parents from 1790 to 1940. Furthermore, the rich information in the censuses allows us to differentiate the effect of values we measure from those of occupation, geography, and demography. We then turn to applying our first-name based value measure. First, we show that the effect of early Irish and Scottish migration on today’s crime rates, as described in Grosjean (2014), disappears once we control for differences in military values. In a second application, we demonstrate that the experience and collective memory of fighting in the American Civil War affects military values in the following decades.