In this section we report the panel regression results of the relationship between (1) EPS and firm performance. (2) FCF and firm performance. (3) EPS, FCF, growth and firm performance. And then, we use Vuong’s z-statistic to examine whether EPS or FCF has better explanatory power on firm performance. Finally, we use the panel regression to provide results of the relationship between EPS, FCF, corporate governance and firm performance.
3.1 Earnings Per Share and Firm Performance
In Table 4, we investigate the relationship between EPS and firm performance. Both in Panel A and Panel B, the Model (1) to (6) show the results between different countries, including USA、Japan、Germany、UK、France and China. Model (7) is the result of the six countries’ total samples.
Panel A uses the natural log of Tobin’s Q as the dependent variable and wants to find relationship with EPS and other financial statement variables. EPS is positive and significant effect on firm Tobin’s Q in USA, Japan, Germany and UK (Model (1) to (4)), but it not significant in the other countries and total samples. Firm size, which defined as the natural log of total assets, is positive and significant with firm Tobin’s Q in USA, Japan and total samples. But in China, firm size has negative and significant effect on firm Tobin’s Q. The possible reason is that the larger the size of the company, the greater the need of external funding, which shares more dispersed, and china have the lowest shareholders’ right, the agent problem will be more obvious, thus reducing the firm performance. Leverage, which defined as debt to equity ratio is found to have significant effect and negative relationship effect on firm Tobin’s Q at almost all models. When a firm borrows money to finance projects, it faces enormous interest payments, and high debt ratio. Therefore, lower the leverage, the firm’s capital structure may be sounder, the firm can provide greater protection
to creditors and shareholders, and then firm performance will better. But there is a positive and significant relationship with corporate Tobin’s Q in Japan. The possible reason is that firms could eliminate information asymmetry by leverage, so the higher leverage will have higher Tobin’s Q. The results shows that firm size, EPS, leverage have no consistent relationship on market-based firm performance in top world six economic markets.
Panel B also shows a similar relationship between EPS and firm performance, the difference is the dependent variable is using ROA to substitute. EPS have positive and significant relationship in all models, and all models are significant at 1% level. Therefore, EPS has strong and positive relationship with firm performance when dependent variable is ROA. In Panel B, firm size is positive and significant at 1% level in all models, which indicates that larger firm size lead to higher firm performance. Leverage has negative and significant effects on accounting-based firm performance, ROA. The results shows that firms with large scale, higher EPS and lower leverage will have better accounting-based firm performance in top six world economic markets.
[Insert Table 4 here]
3.2 Free Cash Flow Per Share and Firm Performance
The results of the effect of FCF on firm performance are summarized as Table 5. Both in Panel A and Panel B, the Model (1) to (6) show the result between different countries, and the Model (7) is the result of the six countries’ total samples.
Panel A want to find the relationship of FCF on firm performance with control variables by using the natural log of Tobin’s Q as the dependent variable. In USA and UK (Model (1) and (4)), FCF is negative and significant at 5% level on firm Tobin’s Q, positive and significant at 10% level in Japan, but not significant in the other countries and total samples.
Firm size, which defined as the natural log of total assets, has a significant positive
relationship with firm Tobin’s Q in USA, Japan, UK and total samples. But in China, firm size is also negatively significant related to a firm Tobin’s Q. Leverage is defined as debt to equity ratio, and it has significant effect and negative relationship effect on firm Tobin’s Q at almost all models. In Japan, it is also has a positive and significant relationship with corporate Tobin’s Q. The results shows that firm size, FCF, leverage have no consistent relationship on market-based firm performance in top world six economic markets.
In Panel B, this includes a similar relationship between FCF and firm performance, and using ROA as the dependent variable. It shows that FCF has positive and significant relationship in all models, and they are significant at 1% level. That indicated FCF has strong and positive relationship with firm performance when dependent variable is ROA. Firm size is positive and significant at 1% level in all models, and shows that larger firm size lead to higher firm performance. Leverage has negative and significant effects on ROA at almost all models. From the results, we find that firms with large scale, higher FCF and lower leverage will have better accounting-based firm performance in top six world economic markets.
[Insert Table 5 here]
3.3 EPS, FCF, Growth and Firm performance
In Table 6, we investigate the relationship between EPS, FCF, growth and firm performance. We apply the method provided by McConnell & Servaes（1995）to separate low and high-growth opportunities firms, it ranks by P/E ratio. The 1/3 of all sample high P/E firms defined as high-growth opportunities firms; 1/3 of all sample low P/E firms defined as low-growth opportunities firms. Previous studies (Faccio and Lasfer, 1999; Ahn, Denis and Denis , 2006) also use the method to divide the sample firms into low and high-growth groups.
In low-growth opportunity firm, firm size in Model (1) and Model (3) is negatively and
significantly associated with firm Tobin’s Q, which is market-based of firm performance.
This indicated that low-growth opportunity firm with large scale will have lower market-based firm performance. When firm go in to low-growth, it may already be in a mature stage and the opportunity for future growth will relatively lower. Besides, when the firm size larger the demand of external fund may increase. It may cause ownership incline towards dispersion, and the managers may acquire some shares, which will make agency problem appear easier (Lee and Chuang, 2009). Therefore, the larger the firm size may cause lower firm performance in low-growth group. In Model (1) and (3) EPS and FCF is not significant related to firm Tobin’s Q, it indicated that when firm at low-growth state, EPS and FCF does not reveal the market-based firm performance.
In high-growth opportunity firm, firm size, EPS, FCF are positive and significant on firm performance in each model. This indicated that high-growth opportunity firm with large size, higher EPS and higher FCF will have higher firm performance no matter market-based or accounting-based.
In high-growth and low-growth opportunity firm, leverage is negative and significant on firm performance in each model. The results shows that firm with lower leverage may create higher market-based and accounting-based firm performance no matter in high-growth or low-growth opportunity firm.
Comparing Model (2) and (6) to Model (4) and (8), we find both EPS and FCF have significant and positively relationship on ROA. This indicated that no matter in low-growth or high-growth stage, firm with higher EPS and higher FCF will have higher accounting-based firm performance.
When using ROA as accounting-based firm performance, both EPS and FCF are positively significant at 1% level. The adjusted R2 in Table 4 and Table 6 show that EPS are more strongly associated with ROA than FCF. However, only simply to compare R2 does not provide statistically reliable evidence that EPS is superior to FCF. Previous study (Dechow,
1994; Brown and Sivakumar 2003; Shuto, 2007) use the method provided by Vuong (1989) for model selection to test the null hypothesis that the two models are equally close to explaining the "true data generating process" against the alternative that one model is closer.
In order to formally discriminate between the EPS and FCF on ROA, so we use the likelihood test that provided by Vuong (1989) to find that EPS or FCF which has better explanatory power for firm performance.
[Insert Table 6 here]
3.4 Test the explanatory power of EPS and FCF
Table 7 shows the results of Vuong’s test of the two models at each sample group. Panel A provided each Vuong’s Z-statistics at the top six world economic markets and total samples.
All sample groups in Panel A have positive Vunog’s Z-statistics. The positive and significant Z- statistic indicates that the EPS is better to reflect the firm performance, because the residual of the model is smaller than the other. The results in Panel A provided that EPS is a better estimate for valuing firm performance than FCF across the top six world economic markets and the total samples.
Panel B shows each Vuong’s Z-statistics at the low-growth firm and high-growth firm samples. Both low-growth firms and high-growth firms have positive and significant Vuong’s Z-statistics. It means that EPS has a better explanatory power on firm performance than FCF no matter at low-growth firm or high-growth-firm samples.
According to Table 7, we find that EPS has better explanatory power for firm performance across the six countries and the whole samples. Even adding the condition of growth, EPS is still the better estimate for firm performance.
Previous studies find the managers would use their discretion to opportunistically manipulate accruals, earnings will become less reliable and cash flow could be preferable. So that accruals principle gives the managers space to do the earning management (Healy, 1985;
Teoh et al. 1998; Collins and Hribar, 2000).
But our results do not consistent with these study(Healy, 1985; Teoh et al. 1998; Collins and Hribar ,2000), the result in Table 6 shows EPS is still the better estimate for firm performance, the results suggest that accruals are performing a useful role in mitigating timing and matching problems in cash flow, so that EPS better summarizes firm performance.
[Insert Table 7 here]
3.5 EPS, FCF, Corporate Governance and Firm performance
From Table 4 and Table 5, we find the effect on firm performance has different result between countries, so firm performance may cause by the country-level corporate governance factors. And previous studies suggest that corporate valuation may driven by corporate governance (LLSV, 2002; Claessens and
Fan
, 2002; Lemmon and Lins 2003; Brown and Caylor, 2006; Chua, Eun et al., 2007; Bhagat and Bolton, 2008). Therefore, we add the country-level corporate governance factors, like corporate board, corruption, country risk, rule of law and shareholder’s right to discuss.Table 8 shows the effect of country-level corporate governance factors on firm performance. Panel A is regression analysis of top six world economic markets. The corporate board is negative and less significant related to market-based firm performance, Tobin’s Q, at 10% level, but positive and significant on accounting-base firm performance, ROA, at 1%
level. Therefore corporate board may be positively significant on firm performance, that indicated that when the corporate board work more efficient the better of the firm performance. The corruption score has a negative and significant relationship with firm performance, both accounting-based and market-based. This indicated that if the country more corrupt, it will has better firm performance. This is not consistent with previous study provided by Lee and Ng (2003) and Chua, Eun et al. (2007). In Model (1) to (4), country risk
is positively significant with firm performance, both accounting-based and market-based measurement. It is important to note that if the country with lower country risk, it will perform better performance. It is consistent with Chua, Eun et al. (2007) and Morey, Gottesman et al. (2009) that country risk are significantly link with higher valuation.
Rule of law has significant and positive relationship on market-based firm performance at 1% level. It indicated that the more legality country will has greater firm performance.
Shareholder’s right is positive and significant at 1% level on market-based firm performance in Model (1) and Model (2). This shows that firm performance increase as the shareholder’s right increase. It is not only consistent with LLSV (2002) ,and Chua, Eun et al. (2007) that investor protection affect corporate valuation, and poor shareholder protection is penalized with lower valuations, but also consistent with Gompers, Ishii, and Meyrick(GIM,2003) that shareholder’s right is positively with firm performance.
But the result of firms in country which with higher corruption will have higher performance is not consistent with previous literature. We argue that China is different from the rest of the sample because it has the lowest mean corruption index and china has been one of BRICs5 recently, comparing to the other well-developed five countries china is with highly potentiality and growth. Therefore, we exclude china to examine the regression analysis of top five world economic markets in Panel B. From the results, we suggest that the corruption index is not significant with firm performance when drop out the samples of China.
From these results, we find that firms in country which with lower country risk, higher corruption, better quality of law enforcement, and stronger shareholder’s right will lead to have better firm performance. Also, Table 8 shows firm size is positive related to ROA, accounting-based firm performance, it is consistent with Maury (2006) that firm size has
5 In economics, BRIC (typically rendered as "the BRICs" or "the BRIC countries") is a grouping acronym that refers to the related economies of Brazil, Russia, India, and China. Goldman Sachs argued that, since they are developing rapidly, by 2050 the combined economies of the BRICs could eclipse the combined economies of the current richest countries of the world.
positive relationship on firm performance. In Model (1) to (4), leverage is negative and significant on firm performance, both the Tobin’s Q and ROA. It is consistent with Opler and Titman (1994), Majumdar and Chhibber (1999), and Weill (2008) that leverage has negative relationship with firm performance. This indicated that firm with higher leverage will has lower firm performance. From the view of pecking order theory, rich-profit firms could meet the funding requirements through the earnings generated from operations,these companies do not use too much debt, so the stronger the performance, debt levels will be lower.
[Insert Table 8 here]