Businesspeople in Elected Office: Identifying Private Benefits from Firm-Level Returns

Abstract

Do businesspeople who win elected office use their positions to help their firms? Business leaders become politicians around the world, yet we know little about whether their commitment to public service trumps their own private interests. Using an original dataset of 2,703 firms in Russia, I employ a regression discontinuity design to identify the causal effect of firm directors winning seats in subnational legislatures from 2004 to 2013. First, having a connection to a winning politician increases a firm’s revenue by 60% and profitability by 15% over a term in office. I then test between different mechanisms, finding that connected firms improve their performance by gaining access to bureaucrats and not by signaling legitimacy to financiers. The value of winning a seat increases in more politically competitive regions but falls markedly when more businesspeople win office in a convocation. Politically connected firms extract fewer benefits when faced with greater competition from other rent-seekers.

Principled Estimation of Regression Discontinuity Designs With Covariates: A Machine Learning Approach

“…In these circumstances, the covariates included are often characteristics of the candidate (age, sex, etc) and characteristics of an electoral unit that they represent (pre-treatment demographics etc). Szakonyi (2018), for instance includes candidate controls such as age, gender, incumbency, ruling party membership, state ownership, foreign ownership, and logged total assets in the pre-election year in his estimates. As I mentioned above, selection of this initial set of covariates should always be dictated by a substantive understanding of the problem at hand.…”

“…Earlier work using RDDs to estimate the returns to office found large lifetime earnings effects by barely (initially) elected members of the British Parilament (Eggers and Hainmueller 2009 (2018) adds an interesting dimension to this literature by using a close election RDD to explore whether office-holding affects the profits of firms whose board members held political office in Russia. Using a close election RDD, Szakonyi (2018) finds that office holding positively affects both firm profitability and firm revenue. In the empirical illustration below, I apply both the adaptive lasso algorithim described above to optimize covariate adjusted treatment effects.…”

“…In the following analysis, I replicate the local linear regression in Table 2 of Szakonyi (2018). In this table, the author uses a close election RDD to estimate the causal effect of holding political office on firm profitability with and without covariates using a recommended 5% bandwidth and the I-K optimal bandwidth estimated without covariates.…”

“…In order to explore the bias and coverage properties of the adaptive lasso method in as realistic a scenario as possible, I perform a series of simulations using the same election and profit data from Szakonyi (2018) described above but with a true simulated treatment effect. To accomplish this, I use the correlation matrix of the covariates and vote margin used by Szakonyi (2018) to construct 2,000 simulated data sets which have the same covariance structure and mean of the original dataset and set the true treatment effect τ RDD to 0.30.…”

“…The remainder of this paper is as follows. Section 1 provides a brief introduction to LATE estimation for sharp RDDs with local linear regression (LLR), the focus of this paper; Section 2 introduces the adaptive lasso and accompanying implementation algorithm along with relevant theoretical derivations; Section 3 provides an applied example of enhanced LATE estimation using a close election RDD study of the effect of holding political office on profit margins in Russian firms published in the American Political Science Review by (Szakonyi 2018). Section 4 provides empirical evidence of the bias reduction and efficiency gains of this method using a series of simulated close election RDDs with covariates and finally, Section 5 concludes with a discussion of future research in this area.…”