Microsoft Word 2007 ichrie conference Proceedings Final-Final 06-06-07. doc

Table2.
Binary Logistic Regression with Full Sample
Dependent Variable: 1 = cash, 0 = stock 
Model 1 
 
 
Model 2 
 
 
number of observation 
cash = 207 , stock = 51 
cash = 207, stock = 51 
coefficient 
p-value 
Z 
coefficient p-value 
Z 
Constant -1.34 
0.02 
-2.42 
-0.78 
0.24 
-1.19 
Cash & short-term 
investment-to-sales 0.18 0.51 
0.66 
0.03 0.62 
0.50 
Capital expenditure-to-
sales 
-1.34 0.16 
-1.40
Capital expenditure-to-
assets
* 
-2.46 
0.09 
-1.72 
Size 
*** 1.02 
0.00 
4.03 
*** 0.88 
0.00 
3.56 
Debt ratio 
*** 3.06 
0.00 
3.46 
*** 2.77 
0.00 
3.04 
Goodness-of-fit test 
chi-sq 
p-value Chi-sq p-value
Deviance 198.45 
0.42 
201.29 
0.36 
% correctly classified 
78.8 
78.5 
*, **, ***: significant at 10%, 5% and 1% level 
Second, we found positive significant coefficients for the debt ratio in Model (1) and (2). However, the 
sign of the coefficient of the debt ratio variable is opposite to the predicted direction. Thus, hospitality acquiring 
firms with larger debt ratios are more likely to use cash rather than stock payments in their acquisitions. 
Third, the coefficient of the capital expenditure to assets ratio is negative and significant in Models 2 
(Table 2) at a level 0.09. This result supports hypothesis 3 that hospitality acquiring firms with low growth 
opportunities are more likely to pay cash for the target acquisition (as shown in model 1, the capital expenditures-to-
sales ratio was not significant). As previous literature (Kim & Olsen, 1999; Kwansa, 1994; Harford, forthcoming) 
has pointed out, it seems that a lack of growth opportunities in the hospitality industry motivates hospitality 
acquisitions. Our coefficient estimates have the same sign as previously found in the general finance literature 
indicating that acquiring firms with low (high) growth opportunities tend to pay with cash (stock) (Martin, 1996).
Last, the coefficient on the size term is positive and significant in Model 1 and 2 (Table 2). The sign of the 
size coefficient is opposite to the predicted direction. Thus, hospitality acquiring firms that are larger are more 
likely to use cash rather than stock in an acquisition payment. It is possible that managers of larger firms 
accumulate cash to give the firm financial slack prior to undertaking an acquisition. Managers may then utilize their 
financial slack to make cash financed acquisitions. As a firm’s size becomes larger, the firm has more potential to 
generate and accumulate cash flow.
Models 1 and 2 (Table 2) have large p-values for the Deviance test. In Deviance test, the higher the p-
value, the better the logistic regression model fits the data. A low p-value implies that the predicted probabilities 

---