# Reference points and method of payment in mergers

Inga Chira (Northridge College of Business and Economics, California State University, Nothridge, California, USA)
Jeff Madura (Department of Finance, Florida Atlantic University, Boca Raton, Florida, USA)

ISSN: 0307-4358

Publication date: 12 February 2018

## Abstract

### Purpose

The purpose of this paper is to examine the impact of the target and bidder reference points on the method of payment in mergers. When considering initial and final results for target and bidder, the target appears to have more negotiating power than the bidder in achieving the financial mix preference that was initially articulated.

### Design/methodology/approach

The authors examine the impact of target and bidder reference points on the consideration sought by the target and the consideration offered by the bidder. The authors test whether the target reference points has an impact on the final method of payment agreed upon by the target and the bidder.

### Findings

The authors find that targets with a longer distance to their respective target reference points prefer to receive cash financing in the consideration sought, while bidders with a longer distance to their respective reference points prefer stock financing in consideration offered. The authors also find evidence that target’s longer distance to its reference point is associated with the use of cash over stock, while the bidder reference point has no impact on the final method of payment used in the merger.

### Practical implications

These insights may be used by the management to formulate the optimal mix of financing in M&A transactions.

### Originality/value

This is an original paper exploring the effect of behavioral finance on corporate decision making.

## Keywords

#### Citation

Chira, I. and Madura, J. (2018), "Reference points and method of payment in mergers", Managerial Finance, Vol. 44 No. 2, pp. 278-290. https://doi.org/10.1108/MF-04-2017-0109

### Publisher

:

Emerald Publishing Limited

## 1. Introduction and motivation

A number of studies in the finance literature attempted to explain the method of payment chosen to finance mergers. Hansen (1987) argues that the target will only accept cash when it believes that its value is less than the offer. Rappaport and Sirower (1999) trace the change in the financing used during the 1990s from cash to stock, emphasizing that when a bidder offers cash, it sends a strong positive signal to the market that it is confident about the outcome of the deal. Shleifer and Vishny (2003) argue that target managers who accept stock are “likely to have relatively short horizons” when agreeing to the deal.

Recently, a different, behavioral strand of literature has also been used to explain merger financing. Harford (2005) suggests that managers tend to use overvalued stock to buy less overvalued firms. However, Malmendier and Tate (2008) argue that overconfident managers are more likely to finance mergers internally, which results in the use of more cash. Billett and Qian (2008) point to the self-attribution bias[1] as a reason for overconfidence. They predict that managers who become overconfident based on a prior successful experience will “exhibit greater optimism regarding firm prospects” and will pursue mergers and share repurchases. Confident managers put too much emphasis on their skills and too much faith in the positive outcome of the merger, choosing to finance mergers with stock rather than cash as they believe the value of the stock will increase following their financial decisions. Baker et al. (2012) use the reference point theory, defined as the high price the target company recorded within the one year (52 weeks) before the merger announcement date, scaled by the base price to explain the variation in premiums paid for targets.

We build on prior behavioral studies by investigating an alternative explanation for the mix of financing used in mergers. To the extent that the reference point can influence the perception of valuation necessary to derive the proper premium to be paid, it can also influence the decision of using cash or stock financing in a merger. While Baker et al. (2012) focus on how the target reference point can influence the appeal of a bid, we build on their work by showing how the reference point from the target’s and bidder’s perspective can influence the proposed and final method of payment. Similar to the definition of the target reference point, we define the bidder reference point as the percentage change based on the 52-week bidder high price over the 365 calendar days ending 30 days before the announcement date. Our findings and implications offer new insight into the relative power and the negotiation process between the two parties. Furthermore, our analysis focuses not only on the final mix that is usually examined in literature, but also on the initially proposed mix that the target and bidder articulate as their preference when they start the negotiating process. These new insights can be used by the strategic management of the firm to articulate the optimal mix of financing in M&A transactions.

Our contribution is threefold. First, we examine the extent to which target and bidder reference points affect the mix of financing sought by the target and offered by the bidder. Second, we test whether the target reference points has an impact on the final method of payment agreed upon by the target and the bidder. Third, we also apply the reference point theory to the perspective of the bidder, and analyze the impact of the bidder reference point on the final mix of financing used in mergers. Previous studies address the method of payment question almost exclusively from the perspective of the target. By comparison, we study the behavior of both the target and the bidder. To our knowledge, this is the first study to investigate the influence of target and bidder reference points on the method of payment used in mergers.

Based on our analysis, targets with a longer distance to their respective target reference points prefer to receive cash financing in the consideration sought, while bidders with a longer distance to their respective reference points prefer stock financing in consideration offered. We also find strong evidence that target’s distance to its reference point is associated with the use of cash over stock, while the bidder reference point has no impact on the final method of payment used in the merger. Thus, while the target’s reference point is associated with its consideration sought and the final financing outcome, the bidder’s reference point is associated with consideration offered, but not with the final financing outcome. These results are of particular interest because they shed some light on the intent and final negotiation regarding method of payment between the two parties.

## 2. Hypotheses development

### Target perspective

Baker et al. (2012) offer strong evidence to suggest that the premium offered by bidders for targets is influenced by the target’s prevailing stock price relative to its respective reference point (as measured by its historical high stock price in the last year). They suggest that targets negotiate for higher premiums when their distance to their respective reference point is larger. Under these conditions, the target maintains a stronger negotiating stance as it may believe that its stock is undervalued, and its price will ultimately revert toward its 52-week high if it is not acquired. We hypothesize that the target’s long distance from its prevailing stock price to its reference point may also enhance its negotiating power in the method of payment, so that it may insist on cash rather than stock[2]. Conversely, when the target’s distance from its prevailing stock price to its reference point is short, the target has less negotiating power and is more willing to accept bidder stock:

H1a.

The smaller the distance between the target current market price and its reference point, the more likely that the target asks for stock in the acquisition process (the higher the proportion of stock in the consideration sought).

H1b.

The smaller the distance between the target current market price and its reference point, the more likely that stock is used in the acquisition.

### Bidder perspective

Bidders may also attempt to negotiate the method of payment based on their own characteristics. Hansen (1987) uses a two-agent bargaining game under imperfect information to argue that when a bidder does not know the value of the target, it prefers to offer stock rather than cash. Martin (1996) suggests that growth opportunities of the bidder determine the method of payment used. When the bidder has high growth opportunities, it is more able to use stock to finance the merger. Several studies, such as Myers and Majluf (1984), Hansen (1987), Shleifer and Vishny (2003) and Rhodes-Kroph and Viswanatghan (2004), use the overvaluation hypothesis to explain the mix of financing. Bidders are more likely to use stock when their stock is overvalued, and they are more likely to use cash when their stock is undervalued.

When the bidder’s stock price is high, it needs fewer shares to buy a target. However, this is not necessarily true if the target stock is also rising. Just as target managers may rely on their reference point when making merger decisions, bidders may rely on their own reference point. We hypothesize that bidders may prefer to use stock to finance mergers when their stock price is near their reference point. If there is a large distance between the bidder stock price and its reference level and if the bidder expects its stock price to revert to the 52-week high, it will be less willing to use stock because it may think it is overpaying for the target with its undervalued stock. In this situation, the bidder would prefer to use cash.

Even if the bidder’s price is close to its reference point and the bidder would prefer to use stock financing, the target may request cash if its own stock price is low compared to its respective reference point. Under these conditions, the target perceives that its valuation is much higher than its prevailing stock price because its reference point is much higher. It is less willing to accept the bidder’s stock to replace its own stock when its prevailing stock price is weak; there is uncertainty regarding the future value of the bidder stock that it receives. While the target managers may expect that their own stock price will revert toward their historical 52-week high, they do not necessarily apply the same logic when assessing the stock valuation of the bidder:

H2a.

The smaller the distance between the bidder current market price and its reference point, the more likely that the bidder uses stock in the acquisition (the higher the proportion of stock in the consideration offered).

H2b.

The smaller the distance between the bidder current market price and its reference point, the more likely that stock financing is used in the acquisition.

### Reference points, financing, and the economy

The sensitivity of the method of payment to the target and bidder reference points may vary with the overall performance of the economy. Harford (2005) observes that bidders use more stock and targets accept more stock in strong economic periods. We argue that it is the relative value of the bidder and target stock to their respective reference levels that contributes to the choice of stock/cash during mergers. However, the sensitivity of the method of payment to the reference points may be conditioned on the performance of the economy. In a weak economy, the target stock price is typically low compared to its reference point. We expect that a longer distance to the target’s reference point should have a more pronounced influence on the target’s preference to receive a cash payment when economic conditions are weak, because the target may blame its prevailing stock price on the weak economy. Thus, it should have more confidence that its stock price would naturally revert toward its 52-week high once the economy improves, and therefore has more bargaining power in negotiating the method of payment.

During a strong economy, the target’s stock price should typically be close to its reference point; there is no need to use the reference point argument. If, however, the target’s prevailing stock price is far from its respective reference point when the economy is strong, target shareholders cannot blame the economy for the weak performance of the stock. They may still argue that their stock price will eventually recover but it is harder to justify an unconditional recovery. Thus, a target with a long distance to its reference point may have less negotiating power if the economy is strong, and therefore may accept more stock:

H3a.

The worse the overall performance of the economy, the stronger is the impact of the target reference point on the method of payment decision, leading to a greater use of cash compared to stock (less stock used).

H3b.

The worse the overall performance of the economy, the stronger the impact of the bidder reference point on the method of payment decision, leading to a greater use of cash compared to stock (less stock used).

## 3. Variables used

### Measurement of the method of payment

We initially test our hypotheses using data representing consideration offered by the bidder and sought by the target. In essence, these tests apply to the initial preferences of the bidder and target. The final method of payment, however, may not necessarily be what either the target initially requested or what the bidder initially offered. Merger negotiations might cause the final financing outcome to differ from consideration offered and sought, especially if one party has stronger negotiating power than the other. Therefore, we also consider the final cash/stock mix used to finance the merger as an alternative (and the most commonly used) proxy for the method of payment. Specifically, we consider the proportion of the payment made with stock, as well as the predominant method (cash/mixed/stock) used as the final method of payment. By comparing the initial and final methods of payment used, we can draw further conclusions into the negotiating power of the two parties.

### Measurement of the target and bidder reference points

Following Baker et al. (2012), we define the target reference point as the high price the target company recorded within the one year (52 weeks) before the merger announcement date, scaled by the base price. We use the 52-week target high price over the 365 calendar days ending 30 days prior to the announcement date obtained from CRSP. Studies, such as Keown and Pinkerton (1981), McPhee and Heckler (2007), and Agrawal and Nasser (2012), find that there is significant leakage in merger announcements and the actual price of the target stock increases in advance of the announcement date. If insiders take advantage of their private information before the announcement, the 52-week high will most likely be around the announcement date. To avoid this bias, we use the 52-week high that ends 30 days before the announcement as the measure of the target reference point.

For the bidder reference point, we use a measure similar to the target reference point. It is percentage change based on the 52-week bidder high price over the 365 calendar days ending 30 days before the announcement date.

### Identification of control variables

A number of factors have previously been cited in literature that may influence the method of payment used in mergers. These factors serve as control variables in our analysis and are defined in Tables I-IV[3].

## 4. Data and method

From SDC, we identify all mergers with a merger announcement date between January 1992 and December 2011, for US targets, excluding the regulated industries. We only include transactions in excess of ten million dollars. We include mergers (M), acquisitions (A), and acquisitions of majority interest (MA). For the control variables that are not available in SDC, we match the sample with Compustat Fundamentals Annual for the date closest (before) the merger announcement date. To reduce the impact of extreme values in the sample, we exclude any observations for which the main variables, TDISTANCE and BDISTANCE, are outside three standard deviations from the mean.[4] We also exclude deals where the stock price of either party is below $1[5]. Our sample is further reduced after we collect data for consideration offered and consideration sought. We identify 589 firms for which we can calculate the target reference point (along with all control variables and consideration sought) and 582 firms for which we can calculate the bidder reference point and consideration offered. The sample by year is presented in Table I. It ranges from five firms (0.86 percent of the sample) in 1992 to 60 firms (10.31 percent) in 1997. The majority of mergers occurred in the late 1990s, declined by 2002 and increased again from 2005 to 2008. Table I, panel B, presents the descriptive statistics for our main variables. The target’s reference point was 30.28 percent higher[6] on average than the stock price before announcement. By comparison, the bidder’s own stock price was 17.98 percent higher than the price at announcement, on average. ### Accounting for economic conditions In addition to analyzing the entire data set, we split the sample into sub-periods based on the performance of the economy. The influence of the economy on the means by which the reference point of the target or bidder could affect the method of payment could be dependent on the proxy used to measure economic conditions. Four different proxies[7] are used to measure economic conditions: 1. Dummy variable based on official NBER data where 1= recession and 0 otherwise (NBER recognizes two recessionary periods during this sample period). 2. The performance of the SPY index by month. The top 25 percent worst performing months are assigned a 1 (bad economy) and the rest a 0 (good economy). 3. VIX level by month. We use the quarterly volatility index on the S&P 500 (VIX) to parse the period into periods of high and low volatility. The top 25 percent highest months are assigned a 1 (high volatility=bad economy) and the rest are assigned 0 (low volatility = good economy). 4. Harford’s liquidity measure by quarter used as a continuous variable. It is defined as the spread between the average interest rate on commercial and industrial loans and the Federal Funds rate (which can serve as a measure of credit risk). ### Model To test whether the consideration sought by the target or offered by the bidder is associated with their respective reference points, we apply a logistic regression model in which the dependent variable (y) is set equal to 1 if the consideration sought or offered is equity, and 0 otherwise. The model is specified as: (1) Equity = β 0 + β 1 TDISTANCE + β 2 BDISTANCE + β 3 TSIZE + β 4 BTANAS + β 5 BLEV + β 6 BSIZE + β 7 MTB + β 8 BLIQ + β 9 BTECH + β 10 BSTPERF + β 11 BCONF + β 12 NUMBIDS + β 13 INDPERF + β 14 RELATED + u i where ui is an independently distributed error term assumed to be normal with zero mean and constant variance. To test whether the final financing mix is associated with the target and bidder distance to their respective reference points, we follow Faccio and Masulis (2005) and apply Tobit multivariate models to investigate the weight of the main and control variables on the method of payment. In the Tobit regression models, the dependent variable is the equity proportion of the payment for the merger transaction, which is in the interval [0, 100]. In addition, we apply an alternative model as a robustness check. We use a multinomial logistic regression in which the dependent variable is assigned a 0 for cash, 1 for mixed (cash and stock) and 2 for all stock deals. When testing the effects of target and bidder distance to their reference points on the financing mix, we initially apply our models without accounting for economic conditions, and then replicate the models while accounting for economic conditions. ## 5. Results First, we test the impact of the target and bidder reference point on the consideration offered (by the bidder) and consideration sought (by the target). The results are presented in Table II[8]. Results in the first two columns show that the target’s reference point has a strong relationship on the consideration sought, supporting H1a. As we have predicted, when the target’s distance from its prevailing stock price to its reference point is long, it is less likely to seek stock, and more likely to seek cash as payment[9]. Results for consideration of stock offered by the bidder are shown in the last two columns of Table II. As predicted by H2a, the bidder’s reference point is inversely related to its likelihood of requesting the use of stock in the consideration offered. This supports our H2a, which implies that a shorter distance between the bidder current stock price and its reference point is associated with a higher likelihood that the bidder will offer stock as a method of payment. Results from testing whether the actual (final) use of stock in mergers is negatively influenced by the target and bidder reference point are presented in Table III. Our findings partially support our hypotheses. As predicted in H1b, TDISTANCE is negative and significant in all three models, which implies that a large distance to the target’s reference point is associated with a higher likelihood of cash as the method of payment. This result supports the hypothesis that targets have more bargaining power when their prevailing stock valuation is distant from the reference point. Put another way, when the target’s valuation is relatively high (close to its reference point), the target is more willing to end up with bidder stock. Unlike the target reference point, H2b is not supported. We find that the bidder reference point has no impact on the final method of payment. Even if the bidder wants to use stock when its stock price is close to its reference point (as was shown in earlier results explaining consideration offered), the financing that is ultimately used for the negotiated deal does not appear to be influenced by the bidder reference point. One possible explanation is that the target is in a better bargaining position when negotiating the method of payment, despite the bidders’ desire to do so. Regarding the control variables, the relationships closely match those disclosed when testing the effects of target reference point. Next, we combine the target and bidder reference points into one model. Results are disclosed in last model of Table III. The target reference point maintains its strong relationship with the likelihood that stock is used, or with the proportion of stock used. The results are confirmed by the multinomial logistic model. Most of the control variables (such as TSIZE, BTANAS, BSIZE, and BMTB) retain their significance status in the models that combine the target and bidder reference points. An alternative explanation to the undervaluation hypothesis and the effect observed in our results is the degree of information asymmetry between the target and the bidder. Given that the insiders (managers) of the firm are more familiar with the direction and the inner workings of the company, at times, the lower valuation relative to the reference point may be an indication of true value. Small targets may be especially sensitive to their reference point because the information asymmetry is higher for those firms. Chira and Madura (2015) argue that with little information available to derive a true value, small firms might be more inclined to use their reference point to arrive at their selling price. By comparison, larger targets that have more information available on the market would allow the prospective bidders to assess the value of the firms they are buying in a more objective way, relying less on reference points. Thus, the size of the firm can be used as a proxy for asymmetric information. In additional specifications not reported in Table III, our results do not support this idea. We do not find any evidence that targets that may be less susceptible to information asymmetry are less reliant on their reference points. Looking at the Tobit model, it appears that the bidder reference point economically has a stronger impact on the outcome of the deal (the coefficient on the bidder appears larger). However, as the statistical significance is lacking and as the multinomial logit model (which we use as a primary source of data analysis) point to the relative strength and significance of the target (and not the bidder) reference point, we drive our conversation from the later model. It is possible that the relationship between the bidder/target reference point and the mix of financing is not linear. Baker et al. raise the same concern in their study on the initial offer price. Following their model, we use piecewise regressions to examine how the target and bidder reference points affect the mix of financing, by quartile. Our results are somehow inconsistent but they generally point to a potentially non-linear relationship, especially when it comes to the target reference point. For example, we find that the consideration offered by the target (H1a) and used in the deal (H1b) influences the mix of financing and the results are strong in the lower quartile. The relationship tappers off somewhere between quartiles 2 and 3. When we perform the same test on the bidder side, we do not find a variability in sensitivity to bidder reference point (H1b) or the final mix of financing accepted by the bidder (H2b). The relationship between the bidder consideration offered and the final stock/cash mix seems to be more linear than the mix on the target side. Results from testing whether the influence of reference points on the proportion or likelihood of stock as method of payment is conditioned on the economy are disclosed in Table IV. Results from applying the Tobit model are provided in Panel A of this table, while results from applying the multinomial logit model are disclosed in Panel B. The main variables of interest to test the influence of the economic conditions are the performance of the economy (denoted as ECONOMY) and the interaction between the target reference point and the performance of the economy (TDIST × ECON)[10] and between the bidder reference point and the performance of the economy (BDIST × ECON). In general, we do not find support for H3a; we find no evidence that the relationship between the target’s distance to its reference point and the method of payment is associated with economic conditions. Regarding the bidder reference point, we find some weak evidence to support H3b. Specifically, we find that for a given distance to its reference point, the bidder may be more willing to consider stock as payment when economic conditions are weak. This result may be because the bidder is less confident that its stock price will revert toward its 52-week high under these weak conditions. Although there is some indication that the impact of bidder reference point varies with the performance of the economy, our results vary based on the definition of weak/strong economy. Some of the measures we use for defining the performance of the economy may have a greater impact on the stock price, and consequently, on the reference point of both bidders and targets. For example, the performance of the S&P 500 may be a more appropriate short-term measure of the economy than the overall level of liquidity. Overall, our combined results support H1a and H2a, suggesting strong evidence that the target reference point has a significant impact on the mix of financing, and the relationship is not altered by changing economic conditions. By comparison, we do not find the same results for the final mix of financing on the bidder side. The bidder reference point does not seem to affect the final mix of financing, unlike our prediction in H2b. This represents a major difference from the preference articulated in the original mix of financing in H1a though consideration offered. ## 6. Conclusions and implications Our objective is to determine whether target and bidder reference points influence the consideration sought by the target and offered by the bidder, and the actual financing mix used for mergers. Identifying these relationships may provide management with an insight into the mix of financing that will be optimal for a proposed M&A transaction. We find evidence that both the target and bidder use their respective reference points when considering the desired method of payment. We also find that the actual use of stock to finance mergers is negatively influenced by the target reference point. The longer is the distance between the 52-week target stock price and the price 30 days before announcement, the greater the likelihood that cash is used to finance the merger. At the same time, we find no evidence that the bidder reference point has an impact on the final method of payment used in the merger. These results are of particular interest because they shed some light over the intent and the final negotiation between the two parties as they relate to the method of payment. Overall, the relationship between target’s reference point and target’s consideration sought is consistent with the actual financing mix, while the relationship between bidder’s reference point and bidder consideration offered is not consistent with the actual financing mix. This outcome may be attributed to the relative bargaining positions of the target and bidder as they relate to the respective reference points of the two parties. Target shareholders may be in a better position to demand a particular financing mix in order to engage in a merger transaction, and their bargaining power may be dependent on the relative level of their firm’s stock price relative to their respective reference point. Conversely, bidders may have their preferred method of financing that is conditioned on their reference point, but may be more flexible on the final financing mix if that is necessary to complete the deal. ## Table I Descriptive Statistics  Panel A: sample by year Year Number of firms in sample Percentage 1992 5 0.86 1993 12 2.06 1994 34 5.84 1995 49 8.42 1996 54 9.28 1997 60 10.31 1998 50 8.59 1999 44 7.56 2000 40 6.87 2001 20 3.44 2002 18 3.09 2003 19 3.26 2004 24 4.12 2005 17 2.92 2006 20 3.44 2007 31 5.33 2008 27 4.64 2009 14 2.41 2010 21 3.61 2011 23 3.95 Total 582 100.00 Panel B: descriptive summaries of independent variables used Variable Mean Median SD Low High TDISTANCE % 30.28 23.33 52.79 0 292.34 BDISTANCE % 17.98 16.72 28.22 0 179.02 Ln (TSIZE) 5.51 5.32 1.93 1.13 12.7 BTANAS % 86.11 96.03 18.83 15.18 99.03 BLEV % 22.28 13.08 16.82 0 78.14 Ln (BSIZE) 7.21 7.33 2.35 0.53 13.91 BMTB (times) 2.76 7.25 34.82% −4.9 72.6 BLIQ % 12.81 5 18.18 0 83 BTECH (dummy) 0 0.22 0.42 0 1 BSTPERF % 30.06 26.92 23.17 −44.23 296.72 BCONF 0 0.29 0.45 0 1 NUMBIDS (dummy) 0 0.06 24.63% 0 1 INDPERF % 1.45 1.9 2.96 −24.56 23.85 RELATED (dummy) 0 0.31 46.07% 0 1 n 582 582 582 582 582 ## Table II Impact of target and bidder reference points on consideration sought/offered Variable Coeff (p-value) TDISTANCE Coeff (p-value) TDISTANCE Coeff (p-value) BDISTANCE Coeff (p-value) BDISTANCE Constant 1.5675 (0.000)*** 1.8930 (0.002)*** −0.1284 (0.178) −1.3107 (0.037)** TDISTANCE −0.0390 (0.001)*** −0.0319 (0.003)*** BDISTANCE −0.0271 (0.002)*** −0.030 (0.003)*** TSIZE 0.2102 (0.015)** 0.2141 (0.003)*** BTANAS 0.1253 (0.802) 1.2219 (0.026)** BLEV −1.1983 (0.094)* −1.5751 (0.008)*** BSIZE −0.1171 (0.082)* −0.2764 (0.000)*** BMTB −0.0018 (0.505) −0.0003 (0.921) BLIQ 0.6917 (0.391) −0.1081 (0.854) BTECH −0.3340 (0.296) −0.2809 (0.297) BSTPERF 0.0018 (0.720) 0.0115 (0.018)** BCONF −0.2285 (0.318) 0.0128 (0.950) NUMBIDS 0.5044 (0.302) −0.2641 (0.452 ) INDPERF −0.0096 (0.587) 0.0069 (0.631) RELATED 0.0197 (0.927) 0.0001 (0.002)*** n 695 589 682 582 p-value 0.0006 0.0162 0.0024 0.0000 Pseudo R2 0.0164 0.0415 0.0119 0.0700 Notes: Dependent variable: a Logit model is applied to categorize targets (consideration sought) and bidders (consideration offered). The dependent variable is assigned a value of 1 if the consideration sought/offered is stock, and 0 otherwise. TDISTANCE is the key independent variable and is calculated as the distance between the 52-week target high price over the 365 calendar days ending 30 days prior to the announcement date obtained from CRSP and the stock price 30 days before the merger announcement. The following control variables are used in the model: TSIZE (target size is proxied by ln (TA)); BTANAS (bidder tangible assets is the ln of the ratio of tangible assets to total assets; BLEV (bidder leverage is calculated as the total debt to total assets ratio); BSIZE (bidder size is proxied by ln (TA)); BMTV (bidder market to book is calculated as the market price 30 days before the announcement to the latest book value of the target before the merger announcement); BLIQ (bidder liquidity is the ratio of cash and marketable securities to total assets); BTECH (bidder high tech firm, 1 if the bidder is a technology firm and 0 otherwise); BSTPERF (the average stock return of the bidder over the 52-weeks before the merger announcement date); BCONF (1 if the bidder has been involved in any mergers since 1982 and 0 otherwise); NUMBIDS (1 is assigned if multiple bids are recorded in SDC and 0 otherwise); INDPERF (average of the 48 Fama-French average industry performance industry average); RELATED (if the bidder and the seller are in the same four digits SIC code and 0 otherwise). ***,**,*Significance at 1, 5 and 10 percent respectively ## Table III Impact of target and bidder reference point on the method of payment  Variable Coeff (p-value) target distance Coeff (p-value) bidder distance Coeff (p-value) target and bidder distance Panel A: Tobit model Constant −0.0772 (0.870) −0.7662 (0.873) 0.1560 (0.354) TDISTANCE −0.0204 (0.060)* −0.0188 (0.000)*** BDISTANCE 0.0111 (0.199) 0.0379 (0.194) TSIZE 0.3332 (0.000)*** 0.2973 (0.000)*** 0.1146 (0.000)*** BTANAS 1.0766 (0.014)** 1.1597 (0.009)*** 3.7456 (0.023)** BLEV −1.7508 (0.000)*** −1.7558 (0.000)*** −1.5028 (0.002)*** BSIZE −0.2508 (0.000)*** −0.2112 (0.000)*** −0.8463 (0.000)*** BMTB 0.0244 (0.002)*** 0.0270 (0.001)*** 0.0322 (0.017)** BLIQ −0.8830 (0.084)* −1.0994 (0.035)** −2.8770 (0.113) BTECH 0.1763 (0.433) 0.0814 (0.725) 0.0343 (0.673) BSTPERF 0.0044 (0.291) −0.0031 (0.488) 0.0634 (0.692) BCONF −0.0879 (0.605) −0.1317 (0.447) −0.1239 (0.835) NUMBIDS −0.5875 (0.057)* −0.4775 (0.122) −0.1698 (0.115) INDPERF 0.0075 (0.524) 0.0276 (0.818) 0.0452 (0.256) RELATED 0.2672 (0.102) 0.1476 (0.371) 0.3201 (0.571) n 589 582 582 p-value 0.0000 0.0000 0.0000 Pseudo R2 0.0310 0.0284 0.0473 Variable Coeff (p-value) multinomial logit target distance Coeff (p-value) multinomial logit bidder distance Coeff (p-value) multinomial logit target and bidder distance Panel B: multinomial logit model Constant −0.6227 (0.351) −0.6970 (0.299) −0.6480 (0.321) TDISTANCE −0.0378 (0.022)** −0.0423 (0.012)** BDISTANCE 0.0162 (0.207) 0.0169 (0.165) TSIZE 0.5516(0.000)*** 0.4932 (0.000)*** 0.4570 (0.000)*** BTANAS 1.0847 (0.079)* 1.1623 (0.061)* 1.1054 (0.068)* BLEV −1.2654 (0.132) −1.3469 (0.106) −0.868 (0.263) BSIZE −0.3895 (0.000)*** −0.3477 (0.000)*** −0.3086 (0.000)*** BMTB 0.0373 (0.010)** 0.0342 (0.014)** 0.0202 (0.066)* BLIQ −1.3282 (0.073)* −1.4298 (0.053)* −1.1475 (0.107) BTECH 0.0817 (0.791) −0.0357 (0.910) −0.0818 (0.789) BSTPERF 0.0098 (0.168) 0.0030 (0.697) 0.0044 (0.562) BCONF −0.2936 (0.250) −0.2067 (0.422) −0.1277 (0.612) NUMBIDS −0.7143 (0.126) −0.6874 (0.137) −0.3168 (0.484) INDPERF 0.0238 (0.171) 0.0208 (0.228) 0.0296 (0.078)* RELATED 0.5778 (0.020)** 0.5064 (0.042)** 0.3338 (0.162) n 589 582 582 p-value 0.0000 0.0000 0.0000 Pseudo R2 0.1212 0.1123 0.1413 Notes: Panel A – dependent variable: the equity proportion of the payment for the merger transaction between [0, 100]. TDISTANCE: the distance between the 52-week target high price over the 365 calendar days ending 30 days prior to the announcement date obtained from CRSP and the stock price 30 days before the merger announcement; BDISTANCE is calculated based as the 52-week bidder high price over the 365 calendar days ending 30 days prior to the announcement date obtained from CRSP. All independent variables are the same as defined in Table II. Panel B – dependent variable: 0 is assigned to cash, 1 for mixed (cash and stock) and 2 for all stock deals ## Table IV Impact of target and bidder reference points on method of payment, conditioned by economy Variable Coeff (p-value) of NBER Coeff (p-value) of SPY dummy Coeff (p-value) of VIX dummy Coeff (p-value) of market liquidity Panel A: results from applying Tobit model Constant 0.1675 (0.323) 0.1789 (0.290) 0.1295 (0.445) 0.1859 (0.322) TDISTANCE −0.0189 (0.000)*** −0.0214 (0.000)*** −0.0151 (0.008)*** −0.0214 (0.019)** BDISTANCE 0.0248 (0.407) 0.0561 (0.084)* 0.0154 (0.636) 0.0636 (0.434) ECONOMY −0.1375 (0.199) −0.0833 (0.919) 0.0561 (0.493) −0.0172 (0.724) TDIST × ECON −0.0732 (0.950) 0.0240 (0.092)* −0.1235 (0.224) 0.1546 (0.862) BDIST × ECON 0.2234 (0.053)* 0.7446 (0.211) 0.7921 (0.019)** −0.1616 (0.727) TSIZE 0.1140 (0.000)*** 0.1117 (0.000)*** 0.1142 (0.000)*** 0.1152 (0.000)*** BTANAS 0.3524 (0.033)** 0.3620 (0.028)** 0.3908 (0.021)** 0.3727 (0.024)** BLEV −0.5102 (0.002)*** −0.5004 (0.002)*** −0.5480 (0.001)*** −0.4979 (0.003)*** BSIZE −0.0848 (0.000)*** −0.0817 (0.000)*** −0.0826 (0.000)*** −0.0853 (0.000)*** BMTB 0.0315 (0.019)** 0.0331 (0.014)** 0.0337 (0.014)** 0.0321 (0.018)** BLIQ −0.2577 (0.157) −0.2876 (0.113) −0.2881 (0.112) −0.2871 (0.114) BTECH 0.2649 (0.745) 0.2131 (0.794) 0.2988 (0.714) 0.3548 (0.663) BSTPERF 0.0890 (0.585) 0.0213 (0.794) 0.0634 (0.693) 0.0785 (0.649) BCONF −0.1051 (0.0859) −0.1189 (0.842) −0.0787 (0.894) −0.1644 (0.783) NUMBIDS −0.1677 (0.119) −0.1943 (0.073)* −0.1599 (0.137) −0.1696 (0.117) INDPERF 0.0501 (0.209) 0.0404 (0.310) 0.0432 (0.278) 0.0451 (0.258) RELATED 0.3475 (0.537) 0.3046 (0.589) 0.3788 (0.502) 0.2909 (0.609) n 582 582 582 582 p-value 0.0000 0.0000 0.0000 0.0000 Pseudo R2 0.0181 0.0181 0.0181 0.0174 Panel B: results from applying multinomial logit model Constant −0.6783 (0.308) −0.6392 (0.336) −0.7347 (0.264) 0.0938 (0.900) TDISTANCE −0.0382 (0.047)** −0.0486 (0.006)*** −0.0312 (0.014)** −0.0766 (0.021)** BDISTANCE −0.0138 (0.270) −0.0251 (0.073)* 0.0093 (0.474) 0.0583 (0.140) ECONOMY −0.1628 (0.775) 0.0963 (0.816) 0.3604 (0.302) −0.4479 (0.028)** TDIST × ECON −0.0318 (0.451) 0.1283 (0.107) −0.0424 (0.226) 0.0188 (0.581) BDIST × ECON 0.0544 (0.305) 0.0980 (0.017)** 0.0387 (0.207) −0.0259 (0.245) TSIZE 0.4615 (0.000)*** 0.4433 (0.000)*** 0.4611 (0.000)*** 0.4918 (0.000)*** BTANAS 1.0932 (0.074)* 1.0613 (0.082)* 1.0821 (0.079)* 1.0329 (0.090)* BLEV −0.9776 (0.209) −0.8399 (0.282) −1.1826 (0.135) −0.8845 (0.258) BSIZE −0.3076 (0.000)*** −0.3043 (0.000)*** −0.3050 (0.000)*** −0.3113 (0.000)*** BMTB 0.0204 (0.063)* 0.0186 (0.103) 0.0210 (0.054)* 0.0218 (0.040)** BLIQ −1.1267 (0.115) −1.1051 (0.123) −1.1754 (0.101) −1.0836 (0.132) BTECH −0.0803 (0.795) −0.1748 (0.578) −0.1414 (0.648) −0.0896 (0.771) BSTPERF 0.0047 (0.537) 0.0044 (0.582) 0.0042 (0.582) 0.0048 (0.516) BCONF −0.1434 (0.571) −0.0853 (0.738) −0.1188 (0.639) −0.2083 (0.419) NUMBIDS −0.3172 (0.485) −0.4137 (0.366) −0.2709 (0.551) −0.3335 (0.469) INDPERF 0.0298 (0.078)* 0.0378 (0.074)* 0.0375 (0.030)** 0.0258 (0.132) RELATED 0.3258 (0.173) 0.3149 (0.189) 0.3489 (0.147) 0.3184 (0.191) n 582 582 582 582 p-value 0.0000 0.0000 0.0000 0.0000 Pseudo R2 0.1064 0.1064 0.1069 0.1101 Notes: Dependent variable: in Panel A the Tobit model is applied, and the dependent variable is the equity proportion of the payment for the merger transaction between [0, 100]. In Panel B, the multinomial logit model is applied, and the dependent variable is set equal to 0 for cash payment, equal to 1 for mixed (cash and stock), and equal to 2 for all stock deals. TDISTANCE is calculated as the distance between the 52-week target high price over the 365 calendar days ending 30 days prior to the announcement date obtained from CRSP and the stock price 30 days before the merger announcement; BDISTANCE is calculated based as the 52-week bidder high price over the 365 calendar days ending 30 days prior to the announcement date obtained from CRSP. All independent variables are the same as defined in Table II. Four different measures of the economy are used: first, official NBER data with results in columns I where 1= recession and 0 otherwise; second, the performance of the SPY index by month. The top 25 percent worst performing months are assigned a 1 (bad economy) and the rest a 0 (good economy); third, VIX level by month. The top 25 percent highest months are assigned a 1 (high volatility=bad economy) and the rest a 0 (low volatility=good economy); fourth, Harford’s market liquidity measure by quarter used as a continuous variable. It is defined as the spread between the average interest rate on commercial and industrial loans and the Federal Funds rate ## Notes 1. Individuals overemphasize their role in good outcomes and deemphasize their role, blaming external factors in bad outcomes. 2. We do not argue that cash is better than stock as some targets have reasons to prefer stock. But all else equal, they may be able to obtain more cash when their reference point is further from the prevailing stock price. 3. A detailed appendix of all control variables is available upon request. 4. This eliminated 12 observations with extreme outlier values for which we could not verify the reference points from public sources. 5. We performed robustness checks for the$2 and \$5 bidder and target price thresholds.

6.

In a much larger sample, Baker et al. (2013) find a mean 52-week target high price of 36.73 percent over the period 1984-2007.

7.

Alternatively, we also use the performance of the SPY index by month and VIX level by month used as continuous variables with no major difference in results.

8.

All analysis uses White’s method to correct for potential heteroskedasticity in estimation.

9.

The highest coefficient of correlation between control variables is 0.6688 (between the target and bidder size) with VIFs ranging from 1.05 to 2.44.

10.

First, we analyze the individual impact of target and bidder reference points in the context of the economy and then we combine the two variables of interest into one model. For conciseness, we only present the combined method.

## References

Agrawal, A. and Nasser, T. (2012), “Insider trading in takeover targets”, Journal of Corporate Finance, Vol. 18 No. 3, pp. 598-625.

Baker, M., Pan, X. and Wurgler, J. (2012), “The effect of reference point prices on mergers and acquisitions”, Journal of Financial Economics, Vol. 106 No. 1, October, pp. 49-71.

Billett, M. and Qian, Y. (2008), “Are overconfident CEOs born or made? Evidence of self-attribution bias from frequent acquirers”, Management Science, Vol. 54 No. 6, pp. 1037-1051.

Chira, I. and Madura, J. (2015), “Reference points and the pursuit of deals”, Financial Review, Vol. 50 No. 3, pp. 275-300.

Faccio, M. and Masulis, R.W. (2005), “The choice of payment in European mergers and acquisitions”, Journal of Finance, Vol. 60 No. 3, pp. 1345-1388.

Hansen, R. (1987), “A theory for the choice of exchange medium in mergers in acquisitions”, The Journal of Business, Vol. 60 No. 1, pp. 75-95.

Harford, J. (2005), “What drives merger waves?”, Journal of Financial Economics, Vol. 77 No. 3, pp. 520-560.

Keown, A. and Pinkerton, J. (1981), “Merger announcements and insider trading activity: an empirical investigation”, The Journal of Finance, Vol. 36 No. 4, pp. 855-869.

McPhee, J. and Heckler, B. (2007), “Avoiding value leakage”, Mergers and Acquisitions, Vol. 42 No. 7, pp. 53-57.

Malmendier, U. and Tate, G. (2008), “Who makes acquisitions? CEO overconfidence and the market’s reaction”, Journal of Financial Economics, Vol. 89 No. 1, pp. 200-243.

Martin, K. (1996), “The method of payment in corporate acquisitions, investment opportunities, and management ownership”, The Journal of Finance, Vol. 51 No. 4, pp. 1227-1246.

Myers, S.C. and Majluf, N.S. (1984), “Corporate financing and investment decisions when firms have information that investors do not have”, Journal of Financial Economics, Vol. 13 No. 2, pp. 187-221.

Rappaport, A. and Sirower, M.L. (1999), “Stock or cash?”, Harvard Business Review, Vol. 77 No. 6, pp. 147-158.

Rhodes-Kroph, M. and Viswanatghan, S. (2004), “Market valuation and merger waves”, The Journal of Finance, Vol. 59 No. 6, pp. 2685-2718.

Shleifer, A. and Vishny, R.W. (2003), “Stock market driven acquisitions”, Journal of Financial Economics, Vol. 70 No. 3, pp. 295-311.