Blockholder Heterogeneity, Multiple Blocks, and the Dance Between Blockholders

Charles J. Hadlock is the Frederick S. Addy Distinguished Chair in Finance at Michigan State University and Miriam Schwartz-Ziv is Assistant Professor of Finance at Michigan State University. This post is based on a recent article authored by Professor Hadlock and Professor Schwartz-Ziv, forthcoming in the Review of Financial Studies.

In our recent article titled Blockholder Heterogeneity, Multiple Blocks, and the Dance between Blockholders, forthcoming in the Review of Financial Studies, we consider issues related to blockholder heterogeneity and coexistence. We collect data on the blockholders of approximately 3,000 companies during the 2001–2014 period. We document substantial heterogeneity in holding periods, position sizes, and positions taken across blockholder types. In addition, we find that nonfinancial blocks are more likely to be observed in smaller, riskier, younger, and less liquid firms; these patterns are either not evident, or are reversed, for financial blocks.

Quite uniquely, we analyze the dynamics among multiple blocks that coexist in the same firm. We show that large and non-financial blocks crowd out other blocks, a behavior that appears causal. Small financial blocks often coexist in the same firm, but this outcome appears to be driven by correlated investment styles.

More specifically, we first examine the factors that predict blockholder presence for different types of blockholders. We interpret this evidence in light of existing theories of blockholder motivations, thus offering insights into the varying motivations and roles of the different flavors of blockholders that appear in public corporations. After establishing this initial picture of blockholder presence, we directly examine the relation between different blockholders’ investment decisions. This evidence allows us to provide evidence on potential blockholder interactions, a theme that is emphasized in many recent discussions of multiple blocks coexisting at the same firm.

The data indicate that blockholders are a heterogeneous group, with systematic variation across blockholder types in holding periods, position sizes, number of positions taken, and types of firms selected for a position. A useful dichotomy is to compare nonfinancial blockholders (e.g., individuals, corporations, strategic investors) with generic financial blockholders (e.g., mutual funds). Both of these groups are quite common (frequencies of 58.9% and 73.6% respectively). Comparing the two, nonfinancial blockholders tend to have larger and longer-lived block positions, and they are much less likely to invest in a large number of firms. When we model the factors that predict blockholder presence, we find that nonfinancial blocks are more likely to be observed in smaller, riskier, younger, and less liquid firms. These patterns are either not evident, or are opposite to what we observe, for financial blocks.

Holding constant these identified factors associated with blockholder presence, we focus our attention on the interdependence of blockholder investment decisions. This allows us to test theories that emphasize potential blockholder interactions that may lead to negative or positive externalities within a blockholder group and any consequent effect on blockholder coalition formation (e.g., Zwiebel (1995), Edmans and Manso (2011)). Given that prior studies have reported that the median public U.S. firm has multiple blockholders, this would appear to be a particularly important issue to investigate. While prior empirical evidence suggests that the structure of a firm’s set of blockholders may affect firm outcomes, little evidence exists regarding what situations give rise to multiple block formations in the first place.

When we investigate this issue in the context of models predicting blockholder presence, we detect compelling evidence of negative blockholder interdependence in the case of large (10% or greater) block positions of any type, and of nonfinancial blocks regardless of position size. Thus, except for small financial blocks, the data appear consistent with the models of Zwiebel (1995) and others that predict a negative influence of the presence of an incumbent blockholder on the decision of others to establish or maintain a block position in a firm. The estimated magnitude of this relation is quite strong, with the presence of a blockholder in some cases decreasing the likelihood of observing another blockholder by a factor of more than one third. This evidence on negative interdependence, which we regard as our most important finding, is compelling, as any inadequately controlled for positive correlation in investing styles will tend to bias us against detecting this behavior. However, to further consider causality issues, we examine cases in which an individual blockholder departs from the firm for likely exogenous reasons associated with death, illness, or advanced age. We find that after these exit events, firms experience abnormally high net blockholder entry, consistent with a causal negative blockholder interdependence relation.

In contrast to nonfinancial blocks, we do detect some evidence of a positive correlation in the appearance of small financial blocks in firms. This could reflect a causal positive interdependence relation, or it may reflect small financial blockholders’ common attractions to similar firms on unobserved dimensions. To investigate, we consider exogenous financial blockholder departures associated with the 2003 mutual fund trading scandal. After these exogenous events, we do not detect any abnormal net changes in the presence of other financial blockholders. This suggests that the positive correlation we observe in the appearance of small financial blocks likely reflects correlated investment styles rather than a causal relation.

In sum, our evidence suggests substantial heterogeneity in blockholder motivations for establishing positions. Many of the investment patterns we detect can be interpreted as consistent with a governance role primarily through monitoring/voice by nonfinancial blocks, and through trading/exit for financial blocks, although this interpretation is admittedly speculative. Holding constant factors that appear to govern blockholder presence viewed in isolation, we detect compelling evidence that blockholders do condition their participation decisions on the presence or absence of other blocks at a firm. In general, the presence of one blockholder appears to inhibit others from establishing block positions at the same firm. This negative interdependence is more pronounced for larger blocks and nonfinancial blocks, and collectively our evidence suggests that the negative relation is causal.

In addition to our main findings, we fill in some important empirical details regarding blockholder behavior. In particular, we document a substantive trend towards more blockholdings and more cases of multiple blockholders at the same firm. These trends reflect a sharp increase over time in the presence of financial and strategic investor blocks, a pattern that is only slightly offset by a moderate decline in other blocks. We detect substantial differences in the median size of block positions, with a high of 13.0% for corporate blockholders and a low of 7.1% for generic financial blockholders. We also report that blockholder positions are moderately durable, with implied expected durations of 4.29 years for nonfinancial blocks, 3.29 years for financial blocks, and richer variation when considered at a less aggregated level.

The complete article is available for download here.

Both comments and trackbacks are currently closed.