Working Paper: CEPR ID: DP12526
Authors: Olivier Dessaint; Jacques Olivier; Clemens Otto; David Thesmar
Abstract: There is a discrepancy between CAPM-implied and realized returns. As a result, using the CAPM in capital budgeting decisions { as is recommended in fi nance textbooks should have valuation e ffects. For instance, low beta projects are expected to be valued more by CAPM-using managers than by the market. This paper empirically tests this hypothesis using publicly announced M&A decisions. We show that takeovers oflower beta targets are accompanied by lower CARs for the bidder. Consistent with our hypothesis, the e ffect is more pronounced for larger acquisitions, higher growth targets, and private targets. Furthermore, low beta bidders are more likely to use their own stock to finance the deal. More generally, low beta firms are less likely to issue equity, and more likely to repurchase shares. These eff ects are not reversed in the long-run, suggesting that CAPM-using managers may be irrational, though this last test lacks power.
Keywords: capital budgeting; valuation; mergers and acquisitions; capital asset pricing model
JEL Codes: G31; G34; G41
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| CAPM-based company misvaluations (G12) | market reactions to M&A announcements (G34) |
| takeovers of lower beta targets (G34) | lower CARs for bidders (D44) |
| one interquartile range difference in target betas (C46) | 0.5% to 1.2% difference in bidder CARs (D44) |
| target's beta (C46) | bidder's market reaction (D44) |
| low beta bidders (D44) | use stock to finance deals (G32) |
| reliance on CAPM (G12) | average loss of USD 37 million per deal for acquirers (G34) |
| reliance on CAPM (G12) | aggregate losses exceeding USD 10 billion annually (F65) |