A perennial topic in academic corporate law is whether the evolution of state corporate law is better described as a race to the top or a race to the bottom. According to the "race to the bottom" hypothesis, states compete in granting corporate charters. After all, the more charters (certificates of incorporation) the state grants, the more franchise and other taxes it collects. Because it is corporate managers who decide on the state of incorporation, states compete by adopting statutes allowing corporate managers to exploit shareholders.
Many legal scholars reject the race to the bottom hypothesis. According to a standard account, investors will not purchase, or at least not pay as much for, securities of firms incorporated in states that cater too excessively to management. Lenders will not make loans to such firms without compensation for the risks posed by management's lack of accountability. As a result, those firms' cost of capital will rise, while their earnings will fall. Among other things, such firms thereby become more vulnerable to a hostile takeover and subsequent management purges. Corporate managers therefore have strong incentives to incorporate the business in a state offering rules preferred by investors. Competition for corporate charters thus should deter states from adopting excessively pro management statutes. The bulk of the empirical research appears to bear out this view of state competition, suggesting that efficient solutions to corporate law problems win out over time.
Lucian Bebchuk and Alma Cohen have made an important recent contribution to the debate in an empirical paper entitled Firms’ Decisions Where to Incorporate. Bebchuk and Cohen claim two key empirical findings:
[First, their] study has … produced findings that cast substantial doubt on the proposition that there is a vigorous competition among states over corporate charters. ... The overwhelming majority of non-Delaware companies are simply incorporated in the states where their headquarters are based.
[Second, their] findings indicate that the incorporation market rewards states that amass antitakeover statutes. ... At one end of the spectrum, states with no antitakeover statutes, such as California, do poorly and retain a relatively small fraction of the companies located in them. At the other end of the spectrum, states that amass most or all standard antitakeover statutes are the ones most successful both in retaining in-state companies and in attracting out-of-state companies.
In other words, the market for state competition in corporate charters is not perfect, and (2) agency costs affect management decisionmaking. Neither result will surprise any informed observer of the corporation law scene. The question is what, if anything, does confirming those points prove? Bebchuk clearly intends it to be one more nail in Delaware’s coffin. Does it succeed in slaying the race to the top beast?
Bebchuk and Cohen construct the following version of the race to the top hypothesis as a foil against which to argue:
[The] conventional view regards incorporation choices as a “pure” choice of a legal regime, based only on a comparison of states’ corporate law systems and a judgment on which of those systems would be best for the firm. … On this view, all states are viewed as “selling” their corporate law system to all publicly traded firms, and not especially to the firms located in them.
They then assert that their data disproves that conventional view.
Their chosen foil, however, is a caricature of the race to the top hypothesis. Did Ralph Winter or Roberta Romano, for example, ever claim that a Los Angeles-based lawyer sits down and thumbs through all 50 state statutes before deciding where to incorporate a client? If so, I missed it. Tellingly, there is no citation of any source setting out the purported “conventional view.” In my view, a fairer picture of the conventional view is that each state views itself as competing with Delaware, not with the other 48.
Why isn’t there greater state-to-state competition in the market for corporate charters? We all know that lawyers play a big role in the decision of where to incorporate. Lawyers are subject to the same bounded rationality constraints everybody else is, as well as the familiar incentives of agency cost economics. Under such conditions, lawyers naturally will adopt a heuristic; and, home state versus Delaware is far and away the most sensible heuristic.1 Bebchuk and Cohen’s data are consistent with this hypothesis, but I doubt either the hypothesis or the data comes as a surprise to even the most ardent race to the top theorist. In other words, their data only helps disprove the race to the top hypothesis if we accept Bebchuk and Cohen’s characterization of that hypothesis, which I am not prepared to do.
Let us turn then to their second major finding, which you will recall is that the incorporation market rewards states that amass antitakeover statutes. Again, the result is not particularly surprising, although it is very helpful to have data showing the magnitude of the trend.
The causal relationship, however, is a bit harder to sort out. I said a moment ago that each state views itself as competing with Delaware, not with the other 48. One contribution of the Bebchuk and Cohen paper is to confirm that content matters, at least on the margins, in incorporation decisions. Curiously, however, their paper also provides further inferential support for the proposition that not all states are running in the race – i.e., not all states are trying to maximize the number of incorporations. Again, however, this is not a particularly surprising result. I suspect that the California legislature is annoyed that so many California-based firms incorporate in Delaware, but I don’t see any evidence that California wants to race Delaware for those incorporations. Why doesn’t California compete? California gets a very small share of state revenues from incorporation-related taxes. A key tenet of the race to the top hypothesis is that states compete to attract incorporations so as to collect the franchise taxes firms pay for the privilege of incorporating within that state. Delaware profits hugely from this competition, as Bebchuk and Cohen elsewhere observe: "Delaware's annual revenues from franchise taxes are on the order of $3,000 for each household of four in the state, and they constitute about 30% of the state's budget." This gives Delaware a strong incentive to offer corporation laws that attract entrepreneurs looking for a place to incorporate their business. In contrast, California generates such a small percentage of its revenues from incorporation-related fees, that California has very little incentive to draft corporation laws attractive to entrepreneurs and managers. Instead, California can indulge the luxury of allowing its corporation laws to reflect the overall anti-business/anti-capitalist leanings of the state's dominant left-liberal political elites. This is significant because it relates to Bebchuk and Cohen’s second finding. They view a particular type of content – antitakeover – statutes as being a driving engine of the incorporation decision. Arguably, however, causation runs in the other direction. In other words, California doesn't retain so few incorporations because it has no antitakeover statute; California has no antitakeover statute because it retains so few incorporations ... and doesn't care.2
A second reason to doubt the causal inferences drawn in the paper is suggested by their finding that large firms based in small states tend to incorporate in their home state. Bebchuk and Cohen posit, correctly I think, that this result reflects the ability of such firms to exercise clout through local favoritism. Many – probably most – state antitakeover laws are adopted at the behest of a large firm that is currently the subject of a takeover bid. At least two of the extreme statutes they discuss, Massachusetts and Pennsylvania, fall into that category. (The takeover of Armstrong World occasioned Pennsylvania’s disgorgement statute, while the bid for Norton was the motivation behind the Massachusetts classified board statute.) Hence, antitakeover statutes may be a symptom of local clout rather than an engine of incorporation decisions. The fact that the extreme statutes have not helped Pennsylvania or Massachusetts in attracting incorporations tends to confirm that interpretation.3
Granted, some large states have antitakeover statutes. All I am trying to suggest is that the causation analysis may be more complex than Bebchuk and Cohen acknowledge.
Having said that, I am persuaded by their finding to the extent it demonstrates that the content of state law matters, at least at the margins. Amassing state takeover laws really does seem to help states attract and retain incorporations. Does that disprove the race to the top hypothesis? As I read the pro-race to the top literature, proponents of the race to the top hypothesis concede that state regulation of corporate takeovers is an exception to the rule that efficient solutions tend to win out when states compete. And a very important exception, at that.
Bebchuk and Cohen therefore again construct a foil against which to argue:
The proliferation of antitakeover statutes is consistent with the view that state competition provides adverse incentives with respect to issues, such as the level of antitakeover protections, that have a substantial effect on the private benefits of managers. … However, the proliferation of state antitakeover statutes might present a problem for those holding the dominant view that state competition is generally beneficial. Supporters of this view have sought to reconcile it with their belief that state antitakeover statutes do not serve shareholders by arguing that state competition does not encourage, and is thus not responsible for, the adoption of antitakeover statutes.
Granted, there are passages in the pro-race to the top literature than can be read as making the point Bebchuk and Cohen claim for it. Yet, I do not think it is the only arrow in the race to the top quiver – or even the most significant one. As I read Roberta Romano’s The Genius of American Corporate Law, for example, her principal explanations for Delaware’s failure to adopt a strong antitakeover statute are three-fold: 4 (1) Delaware’s diverse corporate constituency, which includes both potential targets and potential bidders. (2) The lack of dominance by any one firm, which makes it easier for pro-takeover forces to have an influential lobby. (3) The importance of the Delaware bar, which benefits from takeover activity. (Pp. 59-60) This last point is confirmed by Larry Ribstein’s recent paper, Lawyers as Lawmakers paper, which argues that states where transactional lawyers are powerful (e.g., Delaware) are likely to have efficient laws in areas where ex ante choice of law is viable, as with corporate governance. So the problem in California, for example, is that the trial lawyers are more powerful than the transactional lawyers.
Hence, contrary what the paper argues, it is possible to simultaneously believe that antitakeover laws are bad public policy and that state competition is a race to the top. Markets are not perfect. Efficient solutions usually win out, but not always, and antitakeover statutes are a readily explicable exception to the general rule.
So what are we to conclude about this second finding? There is an agency cost problem in the decision of where to incorporate, with some managers opting for the protections provided by state anti-takeover laws. Coupled with the home state versus Delaware heuristic I posted earlier, this explains Bebchuk and Cohen’s result that adoption of antitakeover statutes helps a state retain or even attract incorporations. Again, however, I think is neither surprising nor fatal for the race to the top project.
Let us assume, however, that I am wrong about that – i.e., assume Bebchuk and Cohen have finally slain the race to the top. Does that validate their normative agenda? As they explained in a recent Daily Deal article, their normative agenda has two pieces:
First, to invigorate competition a federal incorporation option should be provided. Such an option has been offered in Canada with considerable success.
Second, it would be desirable for the federal government to provide a mandatory switching rule under which shareholders would be able to initiate and approve, even over the objection of management, a reincorporation to another jurisdiction. Such a measure would tie success in the incorporation market more closely to the provision of rules that serve shareholders.
I have little sympathy for either goal.
Lucian would have made my life easier, of course, if he proposed mandatory federal incorporation. Instead, he proposes only an option. As a good contractarian, how can I argue against giving people more choices? Suffice it to say that Bebchuk and Cohen’s paper does demonstrate that the content of corporate matters, at least at the margins.5 If so, I doubt whether a federal incorporation option would keep Delaware up nights. Anybody who watched Congress make a hash out of Sarbanes-Oxley or the SEC muck up shareholder access and still thinks the federal government can do a better job than Delaware is beyond my powers to help.
Turning to Bebchuk and Cohen’s second idea – a shareholder power to initiate reincorporation – why is it necessary? When a corporation goes public, shareholders in a sense approve the choice of state of incorporation by buying shares. Shareholders who don’t like Pennsylvania’s antitakeover laws can decline to buy those shares unless the startup entrepreneurs opt-out of the statute. (Granted, there are imperfections in the IPO market, as result of which the impact of the incorporation choice may not be fully priced, but are we to think there is no effect?) When a public corporation changes its state of incorporation, it does so through a merger that typically requires shareholder approval. When Adolph Coors recently asked shareholders to approve a merger to change its corporate domicile to Delaware, by the way, 80% of the shareholders approved. Given that shareholders already have a voice in selecting the state of incorporation, I don’t see what this proposal adds. On top of which, as those of you who have been following my work on director primacy know, I am no fan of proposals that shift power from boards to shareholders. But that is a topic for another day.
1In California we also think about Nevada, which is consistent with their finding that Nevada is coming in (a very distant) second in the race.
2The preceding paragraph will be familiar to regular readers. I extracted it from an earlier post. It’s encouraging that blogging and scholarship are proving to have at least some synergies.
3Bebchuk and Cohen conclude, by the way, that Pennsylvania’s statute has not hurt that state in the incorporation market. Why should it? It is very easy to opt out of the statute and the vast majority of firms have in fact opted out. Given an easy opt-out, why should the statute discourage Pennsylvania incorporations. The much more significant finding is that the statute has not helped Pennsylvania attract incorporations.
4A word perhaps should be said at this point about Bebchuk and Cohen’s critique of the standard argument that Delaware has weak anti-takeover protections. They acknowledge, as they must, that Delaware has no antitakeover statute. Instead, they argue that Delaware has a “large body” of case law validating the just say no defense. As our exchange in the Stanford colloquium last year illustrated, Lucian and I read the Delaware cases quite differently. As I read the Delaware cases, especially recent decisions like Omnicare and Quickturn, I come away doubting that “just say no” is good law in Delaware today, if it ever was.
5Content isn’t the only thing that matters, of course. Delaware’s other advantages include having a separate court, the Court of Chancery, devoted largely to corporate law cases. The Chancellors have great expertise in corporate law matters, making their court a highly sophisticated forum for resolving disputes. They also tend to render decisions quite quickly, facilitating transactions that are often time sensitive.
