5. Methodology
6.2. Means of payment
6.2.1. Bidder analysis
6.2.1.1. Introduction and expectations
The method of payment is a thoroughly researched topic in corporate finance literature, with the majority of research presenting evidence that cash offers provide bidders with higher returns than other financing methods. This result is often supported by the signaling hypothesis and the pecking order theory. Implicitly, the choice of payment method may signal valuable information about the company to the market. Scholars especially refer to stock bids when explaining the effect of signaling, as such decisions often indicate overvalued bidder equity (Shleifer & Vishny, 2003). Additionally, cash offers may signal a strong financial situation for the bidding firm, as it implies that the company has the ability to take on more debt (Faccio & Masulis, 2005). Contradictorily, such predictions can be argued to change during downturns. Firstly, issuance of debt will not necessarily have a positive signaling effect during downturns, as high leverage may not be an attractive feature when markets are weak. Secondly, the structure of capital markets may change, potentially making it more difficult to obtain external M&A-financing all together (Harford, 2005). Thus, while issuance of debt may be difficult during downturns, the attractiveness of paying with equity may also be low, as companies are potentially undervalued. Overall, potential changes in management behavior make the effects of means of payment hard to predict in downturns. Nevertheless, the hypothesis follows the majority of historic evidence, thus seeking to test if cash offers outperform those of stock offers.
Hypothesis 2.1: Cash offers generate higher bidder returns than stock offers
The difference in returns generated by cash, stock and mixed bids are determined using dummy variables for stock and cash. Mixed bids are not included as a dummy, as it will act as the benchmark in regression, representing the average CAR when the mean of payment is mixed (Stock & Watson, 2012). This data has been gathered from Zephyr, but has also been reviewed in Bloomberg to ensure validity.
6.2.1.2. Results and interpretation
The regressions that has been conducted to test the effect of different means of payment is disclosed below. Furthermore, we include a set of control variables in order to increase the robustness of the regression model. Following the methodology for selection of control variables as presented in section 5.2., we control for the geographic location of the acquirer, as well as for differences in bidder’s initial ownership.
77
𝐶𝐴𝑅[−𝑡; 𝑡]𝑖 = 𝛽0+ 𝛽1𝐷𝐶𝑎𝑠ℎ𝑖+ 𝛽2𝐷𝑆𝑡𝑜𝑐𝑘𝑖+ 𝛽3𝐷𝐺𝑒𝑜𝑔𝑟𝑎𝑝ℎ𝑦𝑖+ 𝛽4𝐷𝑇𝑜𝑒ℎ𝑜𝑙𝑑𝑖+ 𝑢𝑖
where 𝐸(𝑢𝑖) = 0
Table 6.9: Regression results from hypothesis 2.1
[-1; +1] [-5; +5]
Independent variables Coefficient t-statistic p-value Coefficient t-statistic p-value
Cash 0,039 1,79* 0,078 0,015 0,58 0,209
Stock 0,065 1,95* 0,054 0,029 0,88 0,373
Geography 0,011 0,53 0,599 -0,006 -0,25 0,043
Toehold -0,019 -0,87 0,389 -0,023 -0,71 0,746
Constant -0,021 -0,95 0,344 -0,005 -0,22 0,657
Number of observations 86 86
F-Statistic 1,28 0,51
R-squared (Adjusted) 0,066 0,021
*, **,*** indicate significance level of 10%, 5% and 1% respectively.
The table above shows results from two OLS regressions specified to measure the average difference in bidders’
cumulative abnormal returns (CARs), given different payment methods. Hence, the dependent variable is CAR for each M&A transaction over the estimation windows of [-1; +1] and [-5; +5]. CAR is measured using the market model, as presented in section 5. The independent variables are defined as follows: deals with cash and stock bids are dummy variables that take a value of 1, if the bids are proposed with cash and stock respectively. Implicitly, mixed bids serve as baseline. The control variables included are the geographic location of acquirer and the bidder’s toehold. All control variables are described in detail in appendix 8. T-statistics are calculated using
heteroscedasticity-robust standard errors. The statistical fit of the model is implied by the (adjusted) determination coefficient 𝑅2 and the F-statistic.
Source: Zephyr M&A database and Bloomberg Terminal
The findings from the regression analysis are presented above. The output portrays positive coefficients for cash and stock, thus indicating higher average CARs from stock and cash offers, relative to mixed bids. Furthermore, stock bids generate significantly higher returns than cash bids, more specifically 6,5% against 3,9% respectively, based on the three-day event window. The same is observed in the eleven-day event window, reporting 2,9% and 1,5% respectively. However, neither stock nor cash offers have significant coefficients in this regression. The overall explanatory power constitutes an R-squared of 0,066 and 0,021 for the two regression models.
78
Figure 6.3: Average abnormal returns (indexed)
The figure illustrates the development of average abnormal returns (AAR) for cash and stock bids, for individual event days by indexing AAR to 100 at the event day -5.
Source: Authors
The indexed AARs for cash, stock and mixed bids are illustrated above. This figure proves the results from the regression, as stock bids portray higher returns around the day of announcement than cash bids, and both are superior to mixed bids. Thus, this contributes to superior CARs for cash bids in both the three and eleven-day event windows.
6.2.1.3. Discussion of result
Previous research investigating the effect of different means of payment on bidder returns in normal times often determines that cash bids outperform stock bids. However, the results from this study suggest the opposite effect, with findings being significant in the three-day event window. In general, former research explains the superior cash offer performance with the previously elaborated signaling effect. As the results of our study deviate from this prediction, the signaling effect is arguably not as strong during downturns. Following the initial expectations, this may come as a result of different market dynamics during downturns. In particular, the lower capital liquidity that reduces the availability of M&A-financing through debt, implicitly also cash payments, thus limiting this as an option for bidders. Additionally, paying with cash may be problematic in terms of maintaining a sustainable leverage ratio, which may be especially important during downturns. Conclusively, our results therefore lead to a rejection of the hypothesis, as both empirical evidence and downturn rationales are found to support the possibility of stock offers outperforming cash and mixed offers.
97 98 99 100 101 102 103 104
-5 -4 -3 -2 -1 0 +1 +2 +3 +4 +5
Cash offer Stock offer Mixed offer
79
The findings from this analysis suggest that the bidding managers seeking to conduct M&As during downturns should prioritize stock offers, rather than cash or mixed structures.