Brian Cheffins is a Professor of Corporate Law at the University of Cambridge. This post is based on a paper co-authored by Professor Cheffins, Steven A. Bank, Paul Hastings Professor of Business Law at UCLA School of Law, and Harwell Wells of Temple University Beasley School of Law.
“Leximetrics,” which involves quantitative measurement of law, has become a prominent feature in empirical work done on comparative corporate governance, with particular emphasis being placed on the contribution that robust shareholder protection can make to a nation’s financial and economic development. Using this literature as our departure point, we are currently engaging in a leximetric analysis of the historical development of U.S. corporate law. Our paper, Law and History by Numbers: Use, But With Care, prepared for a University of Illinois College of Law symposium honoring Prof. Larry Ribstein, is part of this project. We identify in this paper various reasons for undertaking a quantitative, historically-oriented analysis of U.S. corporate law. The paper focuses primarily, however, on the logistical challenges associated with such an inquiry.
In the corporate law context, leximetrics has typically been deployed in cross-country studies that can be termed “comparative law and finance.” It is also possible, however, to focus on one country by coding the law across time. Correspondingly, in Questioning “Law and Finance”: US Stock Market Development, 1930-70 we deployed the “anti-director rights index” (ADRI), a well-known mechanism for quantifying the protection various nations’ corporate laws offer investors, to “score” Delaware corporate law from the turn of the 20th century to the present day. We are currently expanding our research by taking into account two additional bodies of corporate law, Illinois’s and the Model Business Corporations Act (MBCA), the model set of laws promulgated by the Committee on Corporate Laws of the Section of Business Law of the American Bar Association. We are also measuring all three bodies of corporate law (Delaware, Illinois and the MBCA) by reference to a second well known measure of corporate law, an “anti-self-dealing index” (ASDI) that focuses on regulation of transactions between a company and those who control it.
While engaging in historically oriented leximetric research can be analytically fruitful it is not a straightforward task to find the law and translate it into numbers, particularly as one goes back through time. Correspondingly, the admonition “use with care” is one that is apt here. In “Law and History by Numbers” we use a number of examples to make our point that scoring corporate law historically using indices such as the ADRI and the ASDI can be a challenging exercise, two of which we will focus on here.
First, we draw attention to difficulties associated with scoring the MBCA over time with respect to cumulative voting, the treatment of which is one of six elements of the ADRI. La Porta, López-de-Silanes, and Shleifer (with Djankov) revised in a 2008 article key ADRI definitions and concepts they (with Vishny) developed initially in the mid-1990s. One change was to code enabling (“opt-in”) provisions—rules that a corporation can choose to adopt as opposed to mandatory or default terms—as “0” rather than “1”. Under this revised protocol, coding the MBCA in relation to the cumulative voting variable becomes a complicated affair. From 1950, which is when the original MBCA was drafted, to 1955, the MBCA urged states to provide for mandatory cumulative voting, which would be appropriately scored as a “1”. From 1984 to the present day the MBCA has recommended that states adopt an enabling provision to deal with cumulative voting, meaning a “0” is appropriate. From 1955 to 1984, however, scoring the MBCA with respect to cumulative voting is anything but straightforward because the MBCA drafters indicated that states should feel free to choose between adopting a presumptive statutory rule (appropriate for a “1”) and a permissive scheme (appropriate for a “0”).
Second, we highlight by reference to the ASDI the fact that coding corporate law on the basis of case law is potentially fraught with difficulty. The ASDI in fact comprises two anti-self-dealing indices, one measuring public enforcement (fines and other criminal sanctions) and the other private enforcement (civil remedies). The private enforcement index is in turn composed of two sub-indices, focusing respectively on ex ante and ex post regulation. One element of the ex post private enforcement ASDI relates to whether a minority shareholder has standing to sue by way of a derivative suit to challenge a potentially problematic related party transaction. There is turn of the 20th century case law authority from the highly influential New York courts indicating that a minority shareholder lacked such standing. When we sought to ascertain whether this was the law in Delaware, which would mean Delaware should be coded as “0” for the standing to sue variable during the opening decades of the 20th century, we came upon an unexpected, stumbling block: for the first part of the 20th century, there is remarkably little Delaware case law to draw upon. The paucity of jurisprudence means that it is difficult to say with certainty whether Delaware should be awarded a “0” with the standing to sue variable for the opening decades of the 20th century and, if so, when Delaware’s standing to sue score would have increased to its present-day “1.”
While we use the ADRI standing to sue variable to illustrate the challenges associated with tracking changes to the law by way of leximetrics, we conclude by drawing upon this ADRI element to illustrate the insights that can be derived from analyzing law historically by numbers. The conventional wisdom with the development of the law governing related party transactions in the U.S. is that fiduciary standards were watered down over time. Partly due, however, to Delaware’s standing to sue score possibly changing from “0” to “1”, it seems likely Delaware’s private enforcement ASDI score increased as the 20th century progressed, implying that regulation of self-dealing became more robust when a relaxation of the rules would have been anticipated. Leximetric analysis correspondingly might prompt a rethink of the chronology of regulation of related party transactions, which is the sort of insight that suggests the historically-oriented quantification exercise in which we are currently engaging is likely to prove to be worthwhile.
The full paper is available for download here.