Factors behind Increased M&A Activity

Mergers and acquisitions (M&As) in Japan have been rapidly increasing since the later half of the 1990s. The number of M&A cases stood at around 500 in the mid 1990s but increased remarkably from 2000 onward. In 2004, the number reached 2,211, roughly four times the number recorded 10 years prior (see figure 1). A look at changes in the number of M&A cases by market reveals three conspicuous trends. First, the number of M&A cases in which a Japanese company acquired a foreign company ("in-out M&As") increased from the later half of the 1980s through the 1990s. The average ratio of in-out M&As to the total number of M&As for the later half of the 1980s stood at 51.5%, compared to 14.6% in the first half of the 2000s. This indicates that Japanese companies' advancement overseas went into full swing in the later half of the 1980s supported by the robust economy at home. Examples of such in-out M&As include Sony Corp.'s acquisitions of CBS Records Inc. and Columbia Pictures Entertainment, Inc. as well as Bridgestone Corp.'s acquisition of Firestone Tire & Rubber Co.

Second, M&A cases in which a Japanese company acquired another Japanese company ("in-in M&As") increased rapidly from the later half of the 1990s through the first half of the 2000s. The number of in-in M&As increased 6.3 fold from 268 in 1990 to 1,677 in 2004. Likewise, the ratio of in-in M&As to overall M&As increased from average 49.8% in the later half of 1980s to average 73.7% in the first half of the 2000s (see figure 2). A major characteristic of in-in M&As in recent years is that they have mostly occurred as industry consolidation amid intensifying international competition. In the steel industry, for instance, two of Japan's major steelmakers - Kawasaki Steel Corp. and NKK Corp. - were consolidated into JFE Holdings, Inc. in 2002 in a move to counteract the fierce competition from their South Korean and Chinese counterparts. Implementing cost reductions through the integrated management, JFE has secured a leading position on the domestic front; JFE, together with Nippon Steel Corp., constitutes the "big two" steelmakers in Japan today. Other driving factors, those apart from industry consolidation, include individual companies' efforts to reorganize group companies and restructure operations. For instance, Matsushita Electric Industrial Co. in March 2004 increased its stake in Matsushita Electric Works, Ltd. from 32.8% to 52.7%, turning it into a consolidated subsidiary. Against this backdrop is the introduction of new accounting rules requiring consolidated financial statements whereby a reporting company is to present its financial position and business performance on a group-wide basis, combining its figures - sales, profits and so forth - with those of its consolidated subsidiaries.

Third, during the period from the later half of the 1990s through the first half of the 2000s, there was a conspicuous increase in the number of M&As in which a Japanese company was acquired by a foreign company ("out-in M&As"). Out-in M&As began to increase sharply from around 1998 with the average number for the first half of the 2000s coming to 165, roughly 6.7 times greater than the average 25 for the first half of the 1990s. The ratio of out-in M&As to overall M&As has also gradually increased from average 4.8% in the first half of the 1990s to 9.2% in the first half of the 2000s.

The primary purpose of this article is to analyze, given these M&A trends involving Japanese companies, the determinants of in-in M&As in Japan focusing on the period from the late 1990s onward when moves for such M&As were activated.^{1, 2}

Why did the number of in-in M&As begin to increase in the late 1990s? Two hypotheses have been presented as factors that contribute to a sharp increase (and a subsequent sharp decrease) in M&As in an economy as a whole during a certain period.

A first hypothesis focuses on industry-level shocks to growth opportunities and profitability. Confirming that M&A activities concentrate in specific industries, Mitchell and Mulherin (1996) and Harford (2005) argue that the occurrence of shocks to growth opportunities and/or profitability that require a large-scale intra- or inter-industry redistribution of resources leads to a boost in the number of M&As as an efficient means of reallocating resources.^{3, 4} Shocks to growth opportunities and/or profitability, as referred to here, include oil shocks and other incidents that cause drastic changes in the prices of input goods, technological advancement and deregulation. Deregulation of an industry, for instance, affects the degree of competition within the industry, thereby changing the incentives for M&As. This view, in which a macro-level increase in the number of M&As is attributed to shocks affecting the relative profit margins between different industries or companies, is often referred to as a neoclassical explanation; wherein industry-level shocks to growth opportunities and/or profitability provide incentives for companies to undertake M&As. However, Harford (2005) emphasizes that it also requires relatively low transaction costs - that is, in addition to shocks - for M&As to become prevalent on a macroeconomic level. More specifically, he suggests that the correlation between the increase of M&As and economic shocks can be observed only when there is sufficient capital liquidity, whereby he points out that an increase in capital liquidity on a macroeconomic level eases financing constraints for M&A activities which, consequently, leads to a general increase in the number of M&As.

The other hypothesis is called "market-driven," with its major characteristic being its assumption of stock market mispricing. Shleifer and Vishny (2003) theoretically showed that managers of bidding companies, when they have the information that the market valuation of their companies is higher than their fundamental value, have an incentive to seek profits by making acquisitions in stock-for-stock deals; and that myopic managers of target companies have an incentive to sell their companies, in disregard of the long-term profit prospects, by agreeing to favorable terms offered by bidders.⁵ Likewise, Rhodes-Kropf and Viswanathan (2004) theoretically demonstrated that companies whose stock is overvalued by the market have an incentive to carry out stock-for-stock M&As.⁶ Their model assumes that managers of bidding companies have the perfect information about their corporate value whereas managers of target companies cannot accurately assess the future synergies derived from M&As. Thus, in cases where stock prices are soaring (market-wide overvaluation), imperfectly-informed managers of target companies are more likely to accept takeover bids in exchange for overvalued stocks of bidding companies based on their overly positive assessment of future synergies.

What is common to these two theoretical models is that managers opt for stock-for-stock deals in carrying out M&As based on the market-driven hypothesis.

Next, we present the results of our analysis of determinants of M&As involving Japanese companies from the 1990s onward, which was conducted based on the aforementioned hypotheses concerning the increase of M&As, using industry-by-industry data. For this analysis, companies listed on the first and second sections of the Tokyo Stock Exchange between 1995 and 2004 are used as samples.

First, in order to grasp the degree of M&A activities by industry, in each industry we calculate the ratio of the number of M&A cases undertaken to the number of listed companies. For the purpose of this article, we extract cases of M&As defined as those amounting to ¥200 million or more in transaction value.

Next, following the hypotheses, explanatory variables perceived to be affecting the number of M&A cases are generated from financial data. First, in order to capture shocks to an industry, we generate variables indicating the profitability and growth opportunities of each industry. We use Tobin's q as a variable measuring the relative size of future growth opportunities. Then as variables for industry-wide profitability at the present, we use return on assets (ROA), sales to cash flow ratio, sales to total assets ratio, employee growth rate and sales growth rate. Since strong correlations are observed among these variables, we conduct principal component analysis (PCA) of them for each industry and the primary components thus extracted are used as a proxy variable for the profitability of the industry.

Meanwhile, as variables designed to take into consideration factors related the stock market valuations as emphasized in the market-driven hypothesis, we use three year average stock returns (ER) and cross-sectional standard deviation of those(σ(ER)). According to the market-driven hypothesis, the higher the ER of an industry, the greater the probability that share prices are overvalued, and accordingly, so is the likelihood of M&As. Lastly, as a control variable measuring the overall liquidity, interest rate spreads (spread) are added. If the greater liquidity on a macroeconomic level is to facilitate financing, thereby increasing the number of M&As, it is then expected that a negative correlation will be observed between the size of spread and the likelihood of M&As.

Using the explanatory variables, we examine factors determining the M&A ratio for each industry. The tobit method is used for estimation. One of the major findings is that the coefficient of Tobin's q remains significantly positive at the level of 1%. This suggests that more M&As are being undertaken in industries that have relatively high growth opportunities as compared to other industries. Meanwhile, it is confirmed that the coefficient of ER is significantly negative and the coefficient of σ(ER) is significantly positive. This indicates that more M&As are being undertaken in industries that have been relatively undervalued in the market as well as in those with greater volatility. The result concerning the coefficient of ER turned out to be exactly opposite of what had been predicted by the market-driven hypothesis that points to overvalued share prices as a primary factor increasing the number of M&As; what is suggested instead is that more M&As are occurring in industries that have been undervalued in the market. Based on all the above, it can be concluded that our findings do not support the market-driven hypothesis, which argues that overvaluation in the stock market is a factor behind the increase in M&As.

Meanwhile, with respect to the impact of an industry's current profitability to M&As, it is confirmed that the greater the profitability of an industry, the greater the M&A activities within the industry. We also find that the coefficient of spread, a variable measuring liquidity on a macroeconomic level, is significantly negative. That is, smaller interest rate spreads tend to be translated into greater M&A ratio. This indicates that M&A activities become more vigorous for the whole economy when the macroeconomic environment is such that companies are less likely to face financial constraints.

Next, we examine how individual companies behave in making M&A decisions. Even if an industry-wide M&A ratio is low, a closer look reveals that certain companies are pursuing M&A opportunities while many others are not. Therefore, the differences in corporate behavior vis-a-vis M&As may not be fully attributable to industry-specific factors. In this section, we examine how the characteristics of individual companies relate to M&As. For the purpose of this analysis, we focus exclusively on the behaviors of acquiring companies in M&A transactions.

We use a dummy variable that takes the value 1 when a company undertook any M&A during a given year and 0 in the case of no M&A. Meanwhile, we use a set of explanatory variables almost identical to those used in the industry-by-industry analysis in the preceding section; the only difference is that these variables are this time created on an individual company level. As a variable measuring the size of growth opportunities, Tobin's q is used, as in the preceding section. Additionally, we use two dummy variables: Hq dummy that takes the value 1 when a company's q is higher than the value at the upper 25% of the overall distribution of q and Lq dummy that takes the value 1 when a company's q is lower than the value at the lower 25%. These two additional variables enable us to see more clearly the relationship between the size of growth opportunities and M&A decisions.

In order to analyze the behaviors of individual companies, we also apply explanatory variables that explicitly represent the financial conditions of each company. Specifically, the debt-to-asset ratio and, for the purpose of explicitly measuring the impact of financial constraints, the cash flow-to-asset ratio are adopted as additional explanatory variables. For an estimation method, we primarily use the logit model.

One of the major findings is that the coefficient of Tobin's q, a variable measuring growth opportunities, remains significantly positive. We also find that while the coefficient of Hq dummy is significantly positive, the coefficient of Lq dummy is significantly negative. Based on these, it is confirmed that companies with higher Tobin's q are more likely to use M&As as a growth strategy whereas companies with lower Tobin's q are less likely become an acquirer in M&A transactions. The coefficient of the debt-to-asset ratio is significantly negative, indicating that the lower the debt-to-asset ratio of a company, the greater the probability that the company will acquire another company in an M&A deal. This can be interpreted that the lower debt-to-asset ratio increases risk-taking capacity, thereby facilitates M&A.

We have examined determinants of M&As in Japan, focusing on the period from the 1990s, when the number of in-in M&As conspicuously increased, to the present. The results of our analysis can be summarized as follows.

First, based on the findings from the analysis of industries with relatively large numbers of in-in M&As, it is confirmed that the greater the relative size of future growth opportunities, the higher the current level of profitability, the lower the past stock market valuation and the higher the risk associated with an industry, the greater the number of M&As in the industry. From this, we can see that the increase in the number of M&As in recent years is basically in line with the neoclassical explanation, that is, M&A activities in a specific industry are boosted as a result of some sort of shock that impacted the growth opportunities and/or profitability of the industry. Second, examination on an individual company level has found that the greater the relative size of growth opportunities, the lower the debt-to-asset ratio and thus the greater the financing capacity, the greater the likelihood for the company to participate in M&A activities as an acquirer.

So then what factors have had the greatest impact on the growth potential and profitability of such industries and companies that have been actively engaged in M&As from the later half of the 1990s onward? Finding the answer is our task for the future.

>> Original text in Japanese