Hypothesis

國

立 政 治 大 學

‧

N a tio na

l C h engchi U ni ve rs it y

7

recessions would reduce M&A activity. According to Braun and Larrain (2005) and Campello, Gramham, and Harvey (2010), firms with financial constraints cut more on capital expenditure during recessions. Without the ability to access external finance, many firms bypass attractive investment opportunities.

Third, it is decrease in investment opportunities and decline in synergies that prohibit firms from undertaking M&As in recessions. According to Kahle and Stulz (2013), during recessions, household’ wealth (real estate, stocks) would decrease, which induce them to reduce their consumption, and as a result, reduce the overall market demand. With decrease in demand, there would be fewer investment opportunities. In addition, Lambrecht (2004) proposes that the timing of mergers is linked to economies of scale and merger synergies increase with product market demand, which would increase during expansions and decrease during recessions.

2.2 Hypothesis

2.2.1 Probability of Bidding

Previous literature have emphasized the impact of uncertainty on investment (McDonald and Siegel, 1986; Dixit and Pindyck, 1994; Bloom, 2009). For example, Bhagwat et al. (2016) show that macroeconomic shocks play an important role when determining M&A activities. In the period of high market uncertainty measured by VIX, firms are less likely to undertake M&As for the interim risk of completing a deal would increase sharply, which makes the deal less profitable. Also, Bernanke (1983) propose that investment projects with irreversibility have the value of waiting. As market uncertainty increases, it is more valuable for firms to postpone their investments and wait until the uncertainty is resolved. While Bloom (2012) and Panousi (2012) provide evidence that uncertainty rise significantly during recessions, it is less desirable for firms to undertake M&As.

Another determination of whether a firm decide to undertake M&A or not is the synergy that the deal would bring to the combined firm in the future. According to Lambrecht (2004), synergy comes from economic of scale created by the combined firm and is correlated to industry demand. It rises in economic booms and falls in economic downturns.

Despite increase in uncertainty and decrease in synergies, however, overconfident CEOs may still undertake more M&As because they are biased by their cognitive. As mentioned in

‧ 國

立 政 治 大 學

‧

N a tio na

l C h engchi U ni ve rs it y

8

Malmendier and Tate (2005), the cognitive bias of overconfident CEOs include the illusion of control, high degree of commitment to good outcome, and abstract reference point. All of them induce CEOs to believe he can control the outcome and underestimate the possibility of failure when it comes to picking up investment project. Moreover, Doukas and Petmezas (2007) and Malmendier and Tate (2008) mention that overconfident CEO believe that they have superior skills as well as underestimate the risk and overestimate the synergy gains from the bid when evaluating M&A opportunities. As a result, there would be more available M&As opportunities in overconfident CEOs’ cognitive than in non-overconfident CEOs’ during recessions. We predict that overconfident CEOs would undertake more M&As than non-overconfident CEOs during recessions.

Hypothesis 1.a. Overconfident CEOs are more likely to undertake M&As than non-overconfident CEOs during or after recessions due to their underestimation of uncertainty and overestimation of synergies.

Although overestimation of synergies and underestimation of risk would stimulate overconfident CEOs to undertake more M&As than non-overconfident CEOs. However, the level of internal resource can determine if overconfident CEO would overinvest or underinvest.

According to Heaton (2002) and Malmendier and Tate (2005), overconfident CEOs with insufficient internal fund or are highly financial constrained would undertake much less M&As than non-overconfident CEOs. Because overconfident CEOs overestimate the equity value of the firms, which they’re managing, they believe the capital market undervalue their stocks.

This would cause overconfident CEOs to be less willing to issue equity or risky debt than non-overconfident CEOs. In cases when non-overconfident CEOs do not have sufficient internal fund to finance investment project that would create value for firms, they would rather pass the project than accessing external fund from the capital market, for they find it too costly finance externally. During recessions, the internal fund of firms would reduce significantly due to decrease in market demand (Braun and Larrain, 2005; Campello, Gramham, and Harvey, 2010).

Thus, overconfident CEOs who originally have just sufficient internal fund to finance M&As would find it unable to do so. Hence, they are prone to undertake less M&As than non-overconfident CEOs.

Hypothesis 1.b. Overconfident CEOs would undertake less M&As than

overconfident CEOs during or after recession, for they tend to underinvest relative to

non-overconfident CEOs when they do not have enough internal funds.

‧ 國

立 政 治 大 學

‧

N a tio na

l C h engchi U ni ve rs it y

9

2.2.2 Announcement Effect and Long-term Performance

Roll (1986) proposed that overconfident (“hubris” in his literature) CEOs would overpay on M&As, which result in value-decreasing consequences. Also, Hayward and Hambrick (1997) show that premiums paid by acquiring firms are highly correlated with the degree of CEOs’

hubris, and greater the hubris endowed with CEOs the higher the premiums the firm pay.

Similarly, Billett and Qian (2008) provide evidence that overconfident CEOs, measured by their acquisitiveness, face negative announcement return when announcing high-order deals.

Malmendier and Tate (2008) provide evidence that announcement effects of M&As undertaken by overconfident CEOs are lower than those undertaken by non-overconfident CEOs for they might buy targets which bring negative NPV.

On the other hand, the reduced investment opportunities during recession make the problem more severe. According to Kahle and Stulz (2013), decrease in demand during financial crisis would leave many investment opportunities no longer as valuable. Maksimovic (2013) and Harford (2005) show that merger waves are accompanied by abundant investment opportunities and end with recessions when there are few investment opportunities. Thus, overconfident CEOs have higher probability of purchasing targets which would bring value destruction to the firm during recessions.

Hypothesis 2. The announcement effect and long-term performance of M&A undertaken by

overconfident CEOs are lower than those undertaken by non-overconfident CEOs during

recessions.

‧ 國

立 政 治 大 學

‧

N a tio na

l C h engchi U ni ve rs it y

10