Diego Battiston

About
I am a Lecturer (Assistant Professor) at the University of Edinburgh, an Associate in the Centre for Economic Performance's at LSE and Affiliated to CESifo. I've completed my PhD in Economics at the London School of Economics in 2018.

My research interests lie in the areas of Labor Economics, Organisational Economics and Development Economics. I am also interested in Data Science and Machine Learning applications.

Contact
diego.battiston@ed.ac.uk
School of Economics
University of Edinburgh
30 Buccleuch Place
Edinburgh
EH8 9JT

Research Papers

    “The Persistent Effects of Brief Interactions: Evidence from Immigrant Ships” , download
    Awarded the CESifo Distinguished Affiliated Award 2021
     

    This paper shows that brief social interactions can have a large impact on economic outcomes when they occur in high-stakes decision contexts. I study this question using a high frequency and detailed geolocalized dataset of matched immigrants-ships from the age of mass migration. Individuals exogenously travelling with (previously unrelated) higher quality shipmates end up being employed in higher quality jobs at destination. Several findings suggest that shipmates provide access and/or information about employment opportunities. Firstly, immigrants' sector of employment and place of residence are affected by those of their shipmates' contacts. Secondly, the baseline effects are stronger for individuals travelling alone and with fewer connections at destination. Thirdly, immigrants are affected more strongly by shipmates who share their language. These findings underline the sizeable effects of even brief social connections, provided that they occur during critical life junctures.

    Data Matching Appendix: Python example for Levenshtein Automata Search with Radix Tries.


    "Face-to-Face Communication in Organisations" (with J. Blanes i Vidal and T. Kirchmaier) , download
    Review of Economic Studies Volume 88, Issue 2, March 2021.
    Awarded the CEPR Prize for Outstanding Research in Organisation and Management.
     

    Communication is integral to organisations and yet field evidence on the relation between communication and worker productivity remains scarce. We argue that a core role of communication is to transmit information that helps co-workers do their job better. We build a simple model in which workers choose the amount of communication by trading off this benefit against the time cost incurred by the sender, and use it to derive a set of empirical predictions. We then exploit a natural experiment in an organisation where problems arrive and must be sequentially dealt with by two workers. For exogenous reasons, the first worker can sometimes communicate face-to-face with their colleague. Consistently with the predictions of our model we find that: (a) the second worker works faster (at the cost of the first worker having less time to deal with incoming problems) when face-to-face communication is possible, (b) this effect is stronger when the second worker is busier and for homogenous and closely-located teams, and (c) the (career) incentives of workers determine how much they communicate with their colleagues. We also find that workers partially internalise social outcomes in their communication decisions. Our findings illustrate how workers in teams adjust the amount of mutual communication to its costs and benefits.


    New: "The Effect of Advisors’ Incentives on Clients’ Investments" (with J. Blanes i Vidal, R. Hortala-Vallve and D. Lou) , download
     

    We use granular information from a Spanish investment firm to estimate the causal effect of financial advisors’ compensation contracts on their clients’ investments. Our identification exploits: (a) the fact that, for historical reasons, compensation contracts at our firm differed across mutual funds for the same advisor and across advisors for the same fund, and (b) the overhaul to the firm’s compensation policy triggered by MiFIDII, which resulted in within-advisor-fund plausibly exogenous variation in incentives. We find that clients’ investments react markedly and swiftly to changes in their advisors’ incentives. The effect is larger for new clients, for clients who trust their advisors more, and for clients with lower financial knowledge. We identify a dual mechanism underlying this effect: clients whose advisors experience a change in incentives bring more money into the fund portfolio and then direct this money into their advisors’ preferred funds. We introduce our reduced-form estimates into a portfolio-choice model to quantify investors’ utility loss due to the distortion in advice. We estimate losses ranging between 6% and 9%. The change in compensation policy triggered by MiFIDII reduced these losses significantly..


    New version: "Talent Poaching and Job Rotation" (with M. Espinosa and S. Liu) , download
     

    The value of a firm's service lies both in its workers and its relationship with clients. In this paper, we study the interaction between client-specific experience accumulated by workers, poaching behaviour from clients and strategic rotation of workers by firms. Using detailed personnel data from a security-service firm, we show that an increase in client-specific experience increases both the productivity of workers and their probability of being poached. The firm reacts to this risk by rotating workers across multiple clients, and more frequently so to those workers more likely to be poached. We show that after a policy change that prohibited poaching, the firm sharply decreased the frequency of rotation which in turn increased workers' productivity. We propose a theoretical model that guides the empirical patterns and allows us to argue their external validity beyond our specific empirical setting.


    "The Local Effect of Executions" (with J. Blanes i Vidal) , download
     

    The death penalty is arguably the most controversial criminal justice policy in the US. Despite an extensive body of work, there is no credible causal evidence that it affects crime. In this paper, we show that executions cause a local reduction in serious violent crime (homicides, rapes and assaults with weapon). A simple behavioural model predicts that highly publicised executions reduce crime, albeit only temporarily and as long as capital punishment remains relatively rare. We test these predictions using a panel dataset disaggregated at the county-date level. Controlling for date and county-month fixed effects, we find that serious crime is lower in the days surrounding an execution, in the county where the capital offense was originally committed. An event study analysis and a set of robustness tests reinforce this conclusion. The effect is decreasing in the number of recent executions in the county, and it is higher for executions associated with a lot of media attention. Counties neighbouring the original-crime county also experience a significant, albeit smaller, decrease in crime.


Research in Progress

    New: "Peer Pressure and Manager Pressure in Organisations" (with J. Blanes i Vidal, T. Kirchmaier and K. Szmeredi) , download
     

    We study how peer pressure among workers interacts with the pressure that they receive from their immediate superiors. In our natural experiment at a large organization, individuals work in an open-plan space and, for reasons exogenous to their productivity, their adjacent desks become occupied/unoccupied by co-workers throughout their shift. We identify a causal, sharp and persistent increase in worker's productivity following the occupation of an adjacent desk. We examine how this `peer pressure' effect interacts with `manager pressure', and show two results here: (1) peer pressure is stronger when managers are less able to monitor workers, and (2) peer pressure is stronger among workers who are performance-evaluated by the same manager. These results are consistent with a model in which peer pressure is the mechanism through which managers indirectly exert pressure on their subordinates.

    "Can Productivity Shocks Boost Chain Migration?"
     

    This paper studies to what extent local productivity shocks can trigger immigration of relatives and friends of settled immigrants. I first estimate the geographical distribution of surnames-nationalities in the US in the beginning of 20th Century. Then, using data on local productivity shocks (oil discoveries and weather), I estimate the subsequent change in the inflow of passengers with surnames-nationalities more prevalent in the county which experienced the shock. The high frequency of data allows to observe the leads and lags responses for short intervals of time, which overcomes the difficulties of identifying the effects of local shocks with decennial census data.

    "Congestion in Labor Markets: An Empirical Investigation Based on Daily Arrivals During the Age of Mass Migration"
     

    Using daily variation in passenger arrivals to Ellis Island in the period 1900-1924, this paper shows that individuals arrived in days of high congestion, are employed in worse jobs. The identification strategy follows from the fact that conditional on a one-month time window, the total number of ships arrived (from any port) in a given day are exogenous to individual characteristics. I interpret the results using a search model with high entry costs and asymmetric information.

    "Trade Diversion Effects of the Panama Canal" (with F. Valencia and F. Rossi)
     

    Improvements in communication infrastructure and transportation technologies have been promoted as significant drivers of economic development. However, the negative externalities they create have received less empirical attention, partly because of the challenges involved in their identification. Following the opening of the Panama Canal, the number of ships stopping at intermediate South-American ports suddenly declined. We study how this even affected the decline and structural transformation of west-coastal cities in Chile. Our identification strategy exploits the interaction between the relative geographical position of cities and the post-opening period.

    "The Teacher Labor Market: Allocation Rules, Sorting and Incentives" (with S. Giardili)
     

    Many countries assign teachers to public schools using a centralized matching algorithm based on teachers’ preferences and schools’ priorities. In this paper, we study how the design of the teacher labor market affects teachers’ performance and student achievement. Making use of novel data from Argentina, we first perform a critical assessment of teachers’ preferences in relation to school characteristics and locations. Second, we exploit the discontinuities induced by the clearinghouse mechanism to assess the effect of contract/matching types on teachers’ performance.

Publications (Pre-PhD) [✚]

Policy and Discussion Papers [✚]