Quasi-Indexer Ownership and Conditional Conservatism: Evidence from Russell Index Reconstitutions

Abstract

We examine the effects of quasi-indexer ownership on conditional conservatism using the plausibly exogenous discontinuity in quasi-indexer ownership around the Russell 1000/2000 index cutoff. Using both regression discontinuity and instrumental variable research designs, we find higher quasi-indexer ownership leads to higher levels of conditional conservatism. We also find the magnitude of this conservatism effect is larger when the expected benefits of conservatism are greater and in the presence of activist shareholders. Importantly, we examine the mechanisms through which the conservatism effect occurs. We find the conservatism effect is completely explained by the influence quasi-indexer ownership has on CEOs’ equity-based incentives. Differences in quasi-indexer ownership around the threshold cause changes in CEO incentives, which in turn result in the firm choosing more conservative accounting policies. Our evidence provides new insights into the role of quasi-indexers as external monitors and to what extent they directly or indirectly influence firm policies.

Keywords: quasi-indexer institutional owners, conditional conservatism, monitoring, causal effect, regression discontinuity

JEL Classification: G23, G34