What drives merger decision making behavior? Don’t seek, don’t find, and don’t change your mind

Abstract

Despite the constant and frequent merger activity across various industries in the U.S. and throughout the world, limited evidence of the success of corporate mergers has been documented. The vast body of academic research demonstrates that most mergers add no value or reduce shareholder value for the acquiring firm. Given the failure of so many mergers, the question of why mergers continue to occur in large numbers remains. Overconfidence and optimism have come to the forefront as the most common behavioral explanations for the continued prevalence of ill-advised mergers. This paper investigates a different type of behavioral bias that also may influence merger and acquisition decisions—confirmation bias. Using a unique experimental data set, we provide evidence in support of the existence of confirmation bias in merger decision making behavior, particularly with respect to the behavior of actual corporate executives.

Introduction

In regard to corporate mergers, repeated analyses by academics, management consultants, and investment bankers have reached the same conclusion. In the short-term to medium-term, fewer than half of all mergers add value.¹ Additionally, one study reported that less than 30 percent of companies found that their M&A transactions achieved their stated objective within the planned time frame.² The shareholders whose company is bought end up richer, while the shareholders of the buyer seldom do. Although merging is not always a mistake, it is quite risky. Historically, actual merger activity has been difficult to rationalize in terms of traditional finance theory. However, in the past 20 years, more behavioral explanations have been utilized to provide a theoretical basis for merger activity. Roll (1986) was first to propose a non-rational motivation for corporate merger activity with the “hubris hypothesis.” More recently, overconfidence and optimism have come to the forefront as the most common behavioral explanations for the continued prevalence of ill-advised mergers (Malmendier and Tate, 2005). “Excessively optimistic and overconfident executives press on with an acquisition, even when the reaction in financial markets is negative.”³ However, this paper presents evidence of a different type of behavioral bias that may also influence merger and acquisition decisions—confirmation bias. Using experimental methods, this paper examines merger and acquisition decision making behavior and provides evidence in support of the presence of confirmation bias.

The existence of a confirmation bias in the merger and acquisition decision making process could have implications for many aspects of business management. For example, understanding confirmation bias could be important when developing CEO compensation plans. Typically these executive compensation plans are designed to tie compensation with firm performance for the purpose of eliminating agency problems. However, evidence of a behavioral bias that would interfere with an executive's ability to properly evaluate investment opportunities may or may not be adequately addressed with current compensation schemes. As demonstrated by Bolton et al. (2005) and others, the explanations for the level and structure of CEO compensation have significant policy implications. Correspondingly, the type of biases and behaviors that affect executive decision making should influence the design of compensation plans. The existence of these biases may also have significant implications for the enforcement of the Sarbanes-Oxley Act of 2002 which is used to hold executives accountable for corporate malfeasance.

Financial theory tells us that the value of any asset is equal to the present value of its cash flows. In that context, a publicly held firm is merely a bundle of cash flows expected to be received in the future. Under the standard assumption that investors diversify to hold the market portfolio, merger activity does not necessarily add shareholder value. Mergers simply combine the rights to cash flows that are already held by diversified investors; investors who should be indifferent between receiving future cash flow streams from two separate firms or from one merged firm formed by combining them. Nonetheless, several major, non-mutually exclusive reasons are typically offered to explain merger activity (Bower, 2001, Warshawsky, 1987): limit competition and/or gain market share; extend product line; expand geographically; wrest corporate control from entrenched, inefficient management in order to realize greater profitability; gain tax advantages; exploit inefficiencies in the financial markets that leave corporate equities undervalued relative to their intrinsic worth.

In terms of a more rigorous theoretical basis for merger activity, there are a number of diverse theories. We know that modern finance theory is predicated on several assumptions that hold only as approximations in financial markets. Transaction costs, agency costs, informational asymmetries, taxation, and government regulation are all assumed away in most financial models. The presence of these and other market frictions could create situations in which mergers theoretically have the potential to create shareholder value. These theoretical explanations can be grouped into five major categories: microeconomics, financial distress, capital markets, taxation, economic shocks.

Corporate mergers usually have episodic occurrences across industries within the United States and around the world. Notwithstanding the previous theoretical explanations, the empirical evidence suggests on average that little to no short-term or medium-term benefits and limited long-term benefits are achieved from merging. Hogarty (1970) found that performance of heavily merging firms to be generally worse than the average investment performance of firms in their respective industries. Additionally, he found mergers to have a neutral impact on profitability. Lev and Mandelker (1972) could not point to any clear effect of merging on riskiness of the acquiring firm, growth rate in the post-merger years, financial structure, percentage of income taxes paid, or liquidity position of the acquiring firm. Haugen and Langetieg (1975) also found that mergers fail to produce economically significant changes in the distribution rates of return to the stockholder. Firth (1979) studied mergers and takeovers in the United Kingdom and found that on average there were no gains associated with takeovers and that there were in fact small losses.

Jensen and Ruback (1983) claimed that mergers and acquisitions create social welfare by allowing the most efficient distribution of corporate assets. They reported that successful acquiring firms earned average risk-adjusted excess returns of 3.8 percent with acquisitions and approximately 0 percent with mergers.⁴ However, these results were challenged by a flood of event studies finding negative returns to the shareholders of acquirers during the 1970s and 1980s (Sirower, 1997). Generally, these studies demonstrated that the mean returns to acquirers pursuing acquisition strategies were significantly negative, with only approximately 35 percent of acquisitions being met with positive stock market returns on announcement (Sirower, 1997). Even Jensen and Ruback (1983) revealed that as the event window expanded, the returns to acquiring firms deteriorated significantly.

More recently, Cummins and Weiss (2004) conducted a market model event-study of mergers and acquisitions in the European insurance industry over the period 1990–2002. They found that European mergers and acquisitions created small negative cumulative average abnormal returns for acquirers and substantial positive cumulative average abnormal returns for targets. Additionally, Moeller et al. (2008) examined a sample of 12,023 acquisitions by public firms from 1980 to 2001 and found that shareholders of these firms lost a total of $218 billion when acquisitions were announced. Most merger event studies find that, in the long-term, acquiring firms are found to experience negative abnormal returns (Scherer, 1988).

Lewellen et al. (1989) offered risk reduction as another explanation for merger activity. However, when they empirically tested the hypothesis, they found no evidence in their sample that risk reduction for the acquiring firm is the typical outcome or that when it occurs it is differentially costly for the shareholders. Some recent papers do find some positive effects from mergers (see Pillof, 1996, Rau and Vermaelen, 1998, Gugler et al., 2003, Ramaswamy and Waegelein, 2003). However, overall the empirical results generally show a negative long-term impact on profitability.

The remainder of the paper proceeds as follows. Section 2 establishes the basis of our behavioral explanation and discusses confirmation bias with respect to merger integration costs. Section 3 discusses our experimental study. Section 4 presents our experimental data and results. Section 5 discusses alternative explanations. Section 6 provides concluding remarks.