Study with the several resources on Docsity
Earn points by helping other students or get them with a premium plan
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
Community
Ask the community for help and clear up your study doubts
Discover the best universities in your country according to Docsity users
Free resources
Download our free guides on studying techniques, anxiety management strategies, and thesis advice from Docsity tutors
The document investigates the hypothesis that the less transparent the target's assets are, the more likely it is that the acquiring firm can obtain higher short- and long-term returns. The authors analyze a sample of 1,538 friendly acquisitions partitioned in two separate dimensions: acquisitions of public versus private firms, and acquisitions of a firm's assets versus acquisitions of a firm's assets and its management. The conclusions are robust to the type of financing used in the acquisition.
Typology: Study notes
1 / 50
This page cannot be seen from the preview
Don't miss anything!
Mergers and Target Transparency
Iftekhar Hasan Fordham University Jarl G. Kallberg Washington State University Carson College of Business Crocker H. Liu Cornell University School of Hotel Administration Xian Sun Johns Hopkins Carey Business School
Chapter prepared for Corporate Governance in the US and Global Settings (Advances in Financial Economics, Volume 17) Edited by: John, K., Makhija, A. K., & Ferris, S. P.
Abstract
We empirically investigate the hypothesis that the less transparent (more difficult to value) the target’s assets are the more likely it is that the acquiring firm can obtain higher short- and long- term returns. We analyze a sample of 1,538 friendly acquisitions partitioned in two separate dimensions: acquisitions of public versus private firms, and acquisitions of a firm’s assets versus acquisitions of a firm’s assets and its management. Using a sample of (nondiversifying) real estate transactions with a public REIT as the acquirer, we find that acquisitions of public firms have insignificant short-term abnormal returns. Acquisitions of private targets have positive and significant short-term abnormal returns. The acquirer’s abnormal returns are higher in both cases when the transactions involve acquisition of the target firm’s management. We find parallel results when analyzing the acquirer’s Q over the merger year and the three following years. Our conclusions are robust to the type of financing (cash, stock, or a combination) used in the acquisition.
Keywords: mergers and acquisitions; information asymmetry; value of firms
good investment because they offer additional benefits to already diversified shareholders through helping internalize externalities (Hansen & Lott, 1996). Hansen and Lott (1996) argue that if shareholders are diversified and own both shares of acquirers and targets, they do not care whether acquirers overpay for shares of listed companies but would demand that managers not overpay for unlisted targets. The main purpose of this study is to analyze the short- and long-term performance of firms acquiring targets with different degrees of transparency (difficulty in valuation). The possibility of achieving success in a merger depends in large part on the acquirer’s ability to identify attractive targets. Targets are attractive either because they are undervalued by other potential acquirers, or because they might have greater synergies with a specific acquirer. In addition, information costs and valuation uncertainty vary by target type and are generally higher for acquiring private targets or for acquiring intangible rather than physical assets. We hypothesize that acquiring assets that are less transparent (more difficult to value) will lead to higher returns in the long and short run than acquiring assets that are relatively easy to value. That is, we assume that the possibility of the acquirer having an informational advantage over other potential acquirers will be higher when, for example, the asset is private or its assets are less tangible. This informational advantage is reflected in that such hypothesis will hold regardless of the payment method. Several studies show that taking over private targets with stock yield particularly higher returns because: (i) the risk sharing hypothesis (Fuller et al., 2002 among others) (ii) the block shareholder hypothesis (Chang, 1998). To distinguish our hypothesis from theirs, we argue that acquirers having informational advantage enjoy higher returns regardless of the payment method.
Specifically, we analyze acquisitions in two separate dimensions: acquisitions of public versus private firms, and acquisitions of a firm’s assets versus acquisitions of a firm’s assets and its management. Our hypothesis implies that it should be very difficult to achieve abnormal returns or to have a viable informational advantage when the target is publicly traded. Conversely, acquiring private firms should offer the acquirer more scope to exploit an informational advantage. This potential should be even greater if the target’s assets are more difficult to value, which would be the case if the acquisition involved the retention of the target’s management. We use the retention of management because it allows us to partition our sample by differing degrees of transparency. Our focus is not on the retention of management per se. Thus, our hypothesis suggests the following partial ordering of our four categories of transactions: the lowest returns should result from acquisitions of a public firm’s physical assets only; the highest returns should come from acquiring the management and assets of a private company. Returns to acquiring both the management and assets of a public firm, or to acquiring the physical assets of a private firm, will fall between these two extremes. We test our hypothesis by examining 1,538 acquisitions from January 1990 to December
acquired. Combining the 20 or more years of market experience and relationship building of these teams with the efficiency of the Duke delivery system creates a situation where the whole truly exceeds the sum of the parts.
(iii) Public mergers of REITs are typically mergers of equals rather than a large firm acquiring a smaller target.^2 (iv) It is unlikely that a REIT merger creates any monopolistic power since 75 percent of REIT income must come from “passive” real estate investments such as rents.^3 Perhaps the most significant of these advantages is the absence of hostile mergers in our sample. This characteristic is important because it eliminates the variations in performance arising from the differential pricing of hostile versus friendly mergers. 4 It is clear that friendly mergers and mergers where target management is retained imply a less intensive level of bargaining and a more likely completion of the transaction than a hostile merger would. Removing hostile mergers reduces other possible complications. As the theoretical model of Schnitzer (1996) shows, the probability of the bidder adopting a hostile versus a friendly takeover increases as the synergy gains become less certain, which would be the case with the acquisition of management versus the acquisition of physical assets. Our study supports the hypothesis that acquisition performance is linked to the difficulty in assessing the target’s value. Acquisitions of public firms have insignificant short-term abnormal returns. Acquisitions of private targets have positive and significant short-term abnormal returns. Furthermore, these returns are higher when the acquisitions include acquisition of the firm’s management. As such, our work extends the short-term performance findings of Matsusaka (1993) of the positive benefits of retaining target management to cases where intra-
industry mergers or public acquisitions are involved. An analysis of Tobin’s Q over the three years following the acquisition yields parallel results. In contrast to the findings of Chang (1998), Francis, Hasan, and Sun (2008), and Fuller et al. (2002), we find that our conclusions are robust to the type of financing (cash, stock, or a combination) used in the acquisition. Furthermore, it appears that disciplining managers is not as important a motive for acquisitions in our sample as it is in broader industry samples. This study is organized as follows: the section “Background literature” overviews the related literature. The section “Data” describes our data set and statistical approach. The section “Empirical Results” presents our empirical analysis. The section “Conclusions” concludes.
Background Literature One important fact has emerged from the extensive literature on mergers and restructurings: successful mergers are relatively rare events with the acquirer usually suffering poor returns, at the time of the announcement as well as in the short and the long-run. Early examples include Dodd (1980), Asquith (1983), and Eckbo (1983). Franks, Harris, and Titman (1991) find no significant abnormal returns in the three-year period following the completion of the merger. In a complementary study, Agrawal, Jaffe, and Mandelker (1996) find that neither firm size, nor beta, nor a slow adjustment of the market to the merger event explains this long- run, post-merger underperformance of acquirers. They also find that underperformance of acquirers is worse in non-conglomerate mergers than in conglomerate mergers. Loughran and Vijh (1997) show that the long-term returns of acquisitions financed by stock are significantly negative, but cash-financed transactions’ returns are significantly positive. 5 Healy, Palepu, and Ruback (1992), show that the acquirer’s higher productivity leads to improvements in operating
development. Coff (1999) shows empirically that buyers are more likely to seek information from targets when the target is in a more information-intensive industry. Retaining the target’s management makes it much more likely that this information is truthfully revealed. Acquisitions of Public versus Private Targets The analysis of acquisitions of private targets has been a recent area of research. However, comparisons between the performance of public and private acquisitions are complicated by the fact that analysts and investors more closely scrutinize public targets. Thus, they may be subject to takeover speculation, which could cause a run-up in the stock price and a higher final takeover premium.^7 In addition, analysts have considerable data available when offering their opinions of the acquisition of a public firm. For private targets, in almost all cases the analyst is forced to rely on management’s information, which makes it highly unlikely that the analyst is able to render a negative opinion on the proposed takeover. Chang (1998) confirms the usual stock versus cash finding for acquisitions of publicly traded firms; i.e., a negative and significant price reaction for stock offers and a negative and insignificant price reaction for cash offers. For acquisitions of private firms however, he finds that, although the market reaction to the cash offers is insignificant, the market reaction to stock offers is positive and highly significant. He attributes these positive returns to a monitoring story; the emergence of a large blockholder improves the target’s efficiency. Fuller et al. (2002) using a time frame matching ours, find that the returns to acquisitions of private targets are significantly positive. Returns to acquirers of public targets, in contrast are negative and significant, especially when the transaction is financed by stock. Moeller et al. (2004), using dollar abnormal returns, 8 also show that acquisitions of private firms have higher returns than public acquisitions.
When the target has a significant non-tangible component (such as management), there can be substantial disagreement about target value. Kohers and Ang (2000) discuss the use of earnouts 9 made by bidders to target shareholders as a way of binding target management. They estimate the announcement effect of an earnout of 1.36 percent. They find no significant differences arising from the acquirer’s choice of financing mechanism.^10 Retention of Management Human capital (target management) is naturally more difficult to value^11 than are the physical assets of the firm and the possibility of employee turn- over^12 presents a major risk to the acquirer. The studies that analyze the issue of retained management do not reach a clear consensus on the value of retaining management. Matsusaka (1993) focuses on the motivation for mergers during and after the conglomerate merger wave of the late 1960s. The majority of the mergers examined involve privately owned targets and all mergers are in part stock-financed. He finds that acquirers realize positive announcement-period returns from inter-industry (diversifying) mergers and negative returns from intra-industry (related) mergers. 13 He further finds that target management could be a key asset in a diversification acquisition, since the market reacts positively (negatively) to bidders that retain (remove) target management. This result suggests that acquirers who replace target management may either overpay, or overestimate their own abilities. Coff (2002) makes the point that human capital is a key aspect of many acquisitions and empirically demonstrates that it materially affects the probability of a transaction being completed. In contrast to Matsusaka (1993), Hubbard and Palia (1999) do not find a significant impact of management retention on acquirer returns in their analysis of the conglomerate mergers wave of the 1960s.^14
relation between returns and the target lockout period, consistent with an information asymmetry story. Ambrose, Ehrlich, Hughes, and Wachter (2000) study whether efficiency and informational gains result from increased size, brand imaging, or from geographic specialization. The authors find no benefits associated with REIT consolidation.^15 Hardin and Wolverton (1999) address the issue of takeovers of private firms and conjecture that the rapid growth of REITs during the 1990s led them to overpay for their acquisitions. They find that acquisition of private firms yield poor results.^16
Data Sample Design and Data We use Lexis-Nexis, NAREIT Statistical Digest, Securities Data Corporation (SDC), and SNL REIT databases to assemble a list of all REITs that are either the acquirer or the target in a merger from 1990 to 1999. We obtain information on the price, return, and the number of shares outstanding from the Center for Research in Securities Prices (CRSP) database. Balance sheet and income statement data for each firm are obtained from Compustat and SNL REIT databases. We gather information on the merger characteristics from the proxy statements, 10Ks, and news articles from the Dow Jones News Retrieval and Lexis-Nexis. We verified each announcement date taken from SDC with the information in Lexis-Nexis and SNL REIT databases. We determine the financing structure of the merger by estimating the percentages of each financing mechanism used in each transaction. We exclude REITs from our study if any of the following conditions apply: (i) the publicly traded company is not in the CRSP or Compustat database, (ii) the transaction value of
the merger is below $2 million, (iii) the acquirer or target is a foreign company, (iv) the exact announcement date could not be determined, or (v) the transaction was hostile. 17 The resulting database of REIT mergers also captures the retention of target management. We classified a transaction as retaining management if the majority of the target’s senior management takes on significant operating positions in the merged company. For example, we do not consider transactions in which the target’s management is offered a seat on the board of directors as, in and of itself, signifying the retention of management. These awards are often given as compensation for being removed from an operating position.
Estimation Approaches Our initial estimations focus on the market reaction and abnormal stock price performance when there is a merger announcement. In the second part, we calculate two performance measures. The first is Tobin’s Q , which we define as
𝑄𝑄 = Share price × shares outstanging + total assetsTotal assets^ −^ book equity The second measure is 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑡𝑡, which is the holding-period return over t trading days less the return over the same period on a value-weighted index of seasoned REITs. We designate the announcement date as 𝐴𝐴 = − 1 in event time. We denote the announcement period of interest, the two-day period that includes the announcement day and the day following as 𝐴𝐴 = (−1,0). We use standard event study methodology (following Brown & Warner, 1985) and assume security returns are driven by a single-index market model, 18 given by 𝐴𝐴𝑖𝑖𝑡𝑡 = 𝛼𝛼𝑖𝑖𝑡𝑡 + 𝛽𝛽𝑖𝑖𝐴𝐴𝑚𝑚𝑡𝑡 + 𝜀𝜀
where 𝐴𝐴𝑖𝑖^2 is the residual variance for firm i from the regression in Eq. (2); 𝐿𝐿𝑖𝑖 is the number of observations for firm i in the comparison period; 𝐴𝐴𝑚𝑚𝑡𝑡 is the real estate index return on day t of the event period; 𝐴𝐴𝑚𝑚 is the mean real estate index return for the comparison period; 𝐴𝐴𝑚𝑚𝑘𝑘 is the index return for day k of the comparison period. Assuming cross-sectional independence, 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴, is asymptotically normal with mean zero and variance 1/𝑁𝑁. The statistics 𝑍𝑍𝑡𝑡 and 𝑍𝑍𝑎𝑎,𝑏𝑏 are unit normal. We use these statistics to test the null hypothesis that 𝐴𝐴𝐴𝐴𝐴𝐴𝑡𝑡 and 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑎𝑎,𝑏𝑏 equal zero, where
𝑍𝑍𝑡𝑡 = (^) √𝑁𝑁 × 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑡𝑡 and
𝑍𝑍𝑎𝑎,𝑏𝑏 = √𝑁𝑁 � 𝑏𝑏𝑡𝑡=𝑎𝑎√𝑏𝑏 − 𝑎𝑎 −𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑡𝑡 1
We check the robustness of the event study results by using different windows for estimating market model parameters, and by adjusting for nonsynchronous trading using the estimation procedure described in Scholes and Williams (1977). Results in all cases are similar and are available on request. Empirical Results Sample Characteristics Table 1 shows the distribution of acquisitions in our sample over the 11- year observation period. There is a substantial amount of clustering; of the 763 observations that have data available on the transaction size, 424 occur in the 1997-1998 window. Of the total sample, 156 involve the acquisition of property and management. These acquisitions are, on average, considerably smaller than the acquisitions of property only. In each of the years in which there are sufficient data, this difference is statistically significant. The distribution of transactions between acquisitions of property only and property and management is relatively stable over
time. Transaction values increase quickly over the decade, beginning with an average target size of $24.3 million in 1990 and ending with an average target size of $356.3 million in 2000.
Insert Table 1 Table 2 presents the financial characteristics of the acquiring REITs. The table shows that there are no important differences among the acquirers classified by acquisition type. This conclusion also holds if we break the sample into acquisitions of property versus property and management, or if we partition by financing mechanism. For example, the average value of acquirer assets ranges only from $879.7 million in the cash — acquire property only category to $1,092 million in the stock — acquire property only category. This similarity in asset values suggests that the following results will not be significantly influenced by differences in acquirer size or profitability.
Insert Table 2 Event Study Results We find a significant positive announcement effect consistent with that of Allen and Sirmans (1987). Columns 2 and 3 of Table 3 show the CAAR values and associated significance statistics for various event time windows for the entire sample of 1,538 companies. The acquirer’s CAAR for the two-day announcement period 𝐴𝐴 = (−1,0) is 1.48 percent, which is statistically significant at the 1 percent level. This figure implies that the announcement of a merger conveys positive information to the market. The CAAR for the period 𝐴𝐴 = (−1,1) is 1. percent, which is again significant at the 1 percent level. Consistent with market efficiency, the acquirer’s CAAR for the period 𝐴𝐴 = (−5, −2) is 0.06 percent, which is not statistically significant (𝑍𝑍-value = 1.0). For the five days after the announcement 𝐴𝐴 = (1,5), the 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 is 0. percent and again is insignificant.
targets the three groups contain 24, 36, and 34 observations, respectively. For the private acquisitions, the three groups contain 150, 142, and 132 observations, respectively. The results summarized in Table 4 are quite different than those obtained by Chang (1998) or Fuller et al. (2002) and Burns, Francis, and Hasan (2007) for regular corporate mergers and Campbell et al. (2001) for REIT mergers. Rather than finding that stock transactions have a more positive announcement effect, we discover that the financing mechanism has little effect on the results reported in Table 3. For public targets, again the results are all insignificant. For private targets, the results are all positive and significant for time periods that include the announcement date. Furthermore, the magnitudes of the announcement effects are similar across each of the three financing mechanisms.
Insert Table 4
Insert Table 5 Table 5 presents the most important results in this study. It displays the announcement effects of acquirers given different degrees of transparency of the target firm: acquire the property only of a public target, acquire property and management of a public target, acquire property only of a private target, and finally acquire property and management of a private target. It also focuses more precisely on acquisitions of property only (607 transactions, 32 involving public targets) versus acquisitions of both property and management (156 transactions, 72 public). Our results for public versus private acquisitions are comparable to those obtained previously for more general samples of firms. In the short term, private acquisitions produce positive and significant returns, but public acquisitions produce negative and insignificant returns. For private acquisitions involving only property, we find that the announcement effects
are positive and significant at the .10 level for the (−1,0) and (−1,1) performance windows. For public acquisitions involving only property, we find that for all performance windows the abnormal returns are negative and insignificant. When a firm acquires both the property and the management of private targets, the performance windows containing the announcement date have positive and significant returns. For firms that acquire the property and the management of public firms, the returns are similar, but significant only for 𝐴𝐴 = (−1,0) and 𝐴𝐴 = (−1,1). These abnormal returns suggest that acquisitions that include management earn more positive abnormal returns for both private and public acquisitions. Thus, our data support our basic hypothesis on the ordering of these four classes of transactions. Our results differ from Matsusaka (1993) since we find that a premium can exist for mergers within the same industry (“related” acquisitions) if target management is retained. However, our finding that including target management can yield positive abnormal returns for public acquisitions is in agreement with Matsusaka (1993) since his study deals primarily with private acquisitions. To verify that our results in Table 5 do not depend on the financing mechanism we look at acquisitions of property versus property and management by financing mechanism in Table 6. As noted in the first section, previous research concerning the value of stock versus cash financing is inconclusive. For each of the financing mechanism we find results consistent with Table 5. Transactions involving management have higher returns, in both magnitude and statistical significance, than do transactions involving only property. The abnormal returns for acquisitions of property only are weakly significant, although almost all are positive. For the observation windows 𝐴𝐴 = (−1,0) and 𝐴𝐴 = (−1,1), the announcement effects are positive and significant at the .05 level for acquisitions of property and management. These announcement
