We analyze how firms manage their financial flexibility conditional on the expected probability of recession. Using an ex ante measure of future recession, we find that, in the aggregate, firms do not appear to prepare. However, a closer analysis reveals a more nuanced relation. The lack of preparation in aggregate is driven by firms that may be unable to prepare: financially constrained and cash poor firms. We find some evidence that firms able to prepare, unconstrained and cash rich firms, may prepare for future recessions.

A two-stage stock-financed merger occurs when an acquiring firm first issues shares, and then engages in a cash acquisition shortly afterward. Such deals allow us to test two important hypotheses derived from decoupling: by clienteles via segmentation and by time. The acquirer's value is maximized by selling shares to investors preferring to hold them, and use the raised cash to pay the target shareholders (the decoupling by clienteles hypothesis). Two-stage deals also provide an option to the acquirers by allowing them to decouple their own shares from the correlated target's shares by issuing at an earlier date and wait for good acquisition opportunities (the time decoupling hypothesis). We find empirical evidence in support of both hypotheses.

Managers prefer investors who share similar expectations of their firms' prospects. Instead of taking the distribution of investor types as given, we investigate the question of how the managers may be able to effect a change in the pattern of ownership in a world where outside shareholders hold heterogeneous expectations. Under the requirements that the mechanism is costless to the firm and involves no initial cash transfer among the shareholders, the solution is a menu of securities in the form of sidebets among the shareholders. Ex post, the mechanism allows the high-valuation investors to own a greater proportion of the firm and be rewarded with a greater share of the firm's wealth gains.

We find, in a sample of 7581 merger offer announcements from 1990 to 2013, shareholders of 1283 (or 17%) target firms responded to the offer with negative market returns. These investors were disappointed at the offer, despite the price premium. To explain their disappointment, one must understand how target shareholders form expectations of premium to be received. We use a novel empirical design to find the relative weights of the rational vs. behavioral factors underlying the process of expectation formation. The estimated expected premiums are shown to have predictive power in the subsamples of both the positive and negative market responses. We also compare how the weights of the expectation factors change under different market conditions: hot vs. cold M&A regimes, bull vs. bear stock market, financial crisis vs. non-crisis periods, and dotcom bubble vs. no bubble.

We examine the impact of familiarity with business segments on CEOs' divestment decisions. We find CEOs are less likely to divest assets from familiar than from non-familiar segments. We attribute this effect to CEOs' comparative information advantage with respect to familiar segments. Consistent with this information advantage, we document that the familiarity effect is particularly strong in R&D intensive industries. We further find the familiarity effect to be most pronounced for longer-tenured CEOs who have built up sufficient political power over the course of several years in office to enable implementation of their preferred divestment choices. We also document the value effects of divestments and show that familiarity affects returns on divestment announcements.