Vladimir A. Atanasov

Shock-Based Causal Inference in Corporate Finance and Accounting Research

We study shock-based methods for credible causal inference in corporate finance research. We focus on corporate governance research, survey 13,461 papers published between 2001 and 2011 in 22 major accounting, economics, finance, law, and management journals; and identify 863 empirical studies in which corporate governance is associated with firm value or other characteristics. We classify the methods used in these studies and assess whether they support a causal link between corporate governance and firm value or another outcome. Only a stall minority of studies have convincing causal inference strategies. The convincing strategies largely rely on external shocks – usually from legal rules – often called “natural experiments†. We examine the 74 shock-based papers and provide a guide to shock-based research design, which stresses the common features across different designs and the value of using combined designs.

Is There Shareholder Expropriation in the United States? An Analysis of Publicly Traded Subsidiaries

This paper examines the relation between the performance and valuations of publicly traded subsidiaries in the United States and the ownership stake of their parent companies. Cross-sectional and time-series tests demonstrate that subsidiaries of parents that own a substantial minority stake exhibit negative peer-adjusted operating performance and are valued at a 23% median discount relative to peers. In contrast, majority-owned and fully divested subsidiaries show no abnormal performance or valuations. The results of our study indicate that the association between parent ownership and subsidiary performance is nonlinear and that some parents behave opportunistically toward their publicly traded subsidiaries.

Shock-Based Causal Inference in Corporate Finance Research

We study shock-based methods for credible causal inference in corporate finance research. We focus on corporate governance research, survey 13,461 papers published between 2001 and 2011 in 22 major accounting, economics, finance, law, and management journals; and identify 863 empirical studies in which corporate governance is associated with firm value or other characteristics. We classify the methods used in these studies and assess whether they support a causal link between corporate governance and firm value or another outcome. Only a small minority have convincing causal inference strategies. The convincing strategies largely rely on external shocks – usually from legal rules – to generate natural experiments. We examine the 75 shock-based papers and provide a guide to shock-based research design, which stresses the common features across different designs and the value of using combined designs.

Financial intermediaries in the midst of market manipulation: Did they protect the fool or help the knave?

We examine a fund managers alleged manipulation of platinum and palladium futures settlement prices. Using benchmarks from parallel electronic markets, we find that the managers market-on-close trading causes significant settlement price artificiality. Defying predictions that competition among floor traders should limit any artificiality, the artificiality increases in the second half of the alleged manipulation period. Between 35% and 52% of the latter-period artificiality is directly attributable to noncompetitive floor prices. Inflated floor volume contributes a similar proportion to artificiality via the exchanges trade-weighted settlement price formula. We estimate that floor counterparties reaped more than

The Trouble with Instruments: Re-Examining Shock-Based IV Designs

6.0 million in excess profits.

Is There Shareholder Expropriation in the U.S.? An Analysis of Publicly-Traded Subsidiaries

Credible causal inference in accounting and finance research often comes from “natural�? experiments. These experiments can be exploited using several “shock-based�? research designs, including difference-in-differences (DiD), instrumental variables based on the shock (shock-IV), and regression discontinuity (RD). We study here shock-IV designs using panel data. Shock-IVs are potentially more credible than non-shock IVs, but can be problematic unless carefully handled. We identify all shock-IV papers in two broad datasets; and re-examine three of the apparently strongest papers – Duchin, Matsusaka and Ozbas (“DMO,�? JFE 2010); Iliev (JF 2010); and Desai and Dharmapala (REStat 2009). We show that the IVs in all three papers are unusably weak – once we enforce covariate balance and common support for treated and control firms, the instruments are no longer significant in the first stage. All three papers also show non-parallel pre-treatment trends on outcomes or core covariates. The problems with these papers generalize to our full sample, and to other papers exploiting the same shocks as DMO. A core conclusion of our reexamination is that “pre-treatment balance�? (common support, covariate balance, and parallel pre-treatment trends) is necessary for credible shock-IV designs. We provide a good practice checklist for shock-IV design with panel data. Finding valid, usable shock-IVs is more challenging than prior research suggests. The Online Appendix is available at: http://ssrn.com/abstract=2859113.

Self-Dealing by Corporate Insiders: Legal Constraints and Loopholes

This paper examines the relation between the performance and valuations of publicly-traded subsidiaries in the United States and the ownership stake of their parent companies. Cross-sectional and time-series tests demonstrate that subsidiaries in which the parent owns a substantial minority stake exhibit negative peer-adjusted operating performance and are valued at a 23% median discount relative to peers. In contrast, majority-owned and fully divested subsidiaries show no abnormal performance or valuations. The results of our study indicate that the association between parent ownership and subsidiary performance is nonlinear and that some parents do, in fact, behave opportunistically toward their publicly traded subsidiaries.

Optimal Portfolios with Monitoring, Private Benefits of Control, and Budget Constraints

Insiders (managers and controlling shareholders) can extract (tunnel) wealth from firms using a variety of methods. This article examines the different ways in which U.S. law limits, or fails to limit, three types of self-dealing transactions – cash flow tunneling, asset tunneling, and equity tunneling. We examine how U.S. corporate, securities, bankruptcy, and tax law, accounting rules, and stock exchange rules impact each form of self-dealing, and identify weaknesses in these rules. We argue that a variety of complex asset and equity transactions, as well as equity-based executive compensation, can escape legal constraints. We propose changes in corporate, disclosure, and shareholder approval rules to address the principal gaps that emerge from our analysis. For an extended version of this article, including case studies illustrating how these loopholes are exploited, see Atanasov, Black, and Ciccotello, Law and Tunneling (2011), 37 Journal of Corporation Law 1-49, available at http://ssrn.com/abstract=1444414.

Online Appendix for the Trouble with Instruments: Re-Examining Shock-Based IV Designs

This paper develops a model that combines features of portfolio theory and corporate governance research. The model analyzes the problem of how large investors form optimal equity portfolios when the return on each investment depends on the size the investment. The optimal solution counterweighs the benefits of higher returns through private benefits of control and monitoring of firm management with the costs of higher transaction prices and bearing diversifiable risk. The general model predicts that controlling investors will monitor more the firms in which they can appropriate more private benefits of control. Large investors are more likely to buy controlling blocks in smaller firms, firms with higher private benefits of control, and firms where the trading costs of buying large blocks are smaller. A special case is numerically solved using an integer-programming algorithm. The numerical analysis shows that investor utility is strictly increasing in available capital for purchasing controlling blocks. The optimal portfolios of large investors dominate the portfolios of small investors both in terms of expected return and risk. The classical result that the efficient frontier is the same for all investors regardless of their wealth does not hold in the setting of this paper.

Which Limited Partners Limit VC Opportunism

This online appendix contains additional results for Atanasov and Black, 2016, The Trouble with Instruments: Re-Examining Shock-Based IV Designs (AB-2016). Part 1 of the Appendix contains further tests of Duchin, Ran, John Matsusaka, and Oguzhan Ozbas, 2010, When Are Outside Directors Effective? Journal of Financial Economics, 95, 195-214 in response to Duchin, Ran, John Matsusaka, and Oguzhan Ozbas, 2015, Comments on “The Trouble with Instruments: Re-examining Shock-Based IV Designs” by Atanasov and Black, available here:http://ssrn.com/abstract=2697098.Part 2 contains brief discussions of selected shock-based IV papers that are identified in AB-2016 but not re-examined in detail. The underlying article is available here: http://ssrn.com/abstract=2417689.This online appendix contains additional results for Atanasov and Black, 2016, The Trouble with Instruments: Re-Examining Shock-Based IV Designs (AB-2016). Part 1 of the Appendix contains further tests of Duchin, Ran, John Matsusaka, and Oguzhan Ozbas, 2010, When Are Outside Directors Effective? Journal of Financial Economics, 95, 195-214 in response to Duchin, Ran, John Matsusaka, and Oguzhan Ozbas, 2015, Comments on “The Trouble with Instruments: Re-examining Shock-Based IV Designs” by Atanasov and Black, available here:http://ssrn.com/abstract=2697098.Part 2 contains brief discussions of selected shock-based IV papers that are identified in AB-2016 but not re-examined in detail. The underlying article is available here: http://ssrn.com/abstract=2417689.