3. Equity Stakes Acquisition of Foreign Financial Institutions
3.4 Results and Analysis
國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
the purpose of this study, supra-national financial and non-financial institutions, such as International Finance Corporation, Asian Development Bank, and Temasek Holdings, were excluded from the observed samples. To summarize, there were 44 observed samples; 27 foreign banks took equity stakes during the time interval and were regarded as complete data. The remaining 17 foreign banks which did not take the forms of equity shares were regarded as truncated data. In addition, there were 29 Chinese banks which entered into strategic alliances with foreign banks taking the forms of equity investment; 4 were state-owned, 9 were joint-stock commercial banks, 15 were city commercial banks, and 1 was a rural cooperative. Their average and deviation of assets and revenue during the last five years are listed in Table 5.
The variables in the model are categorized as below for the analysis of the foreign financial institution taking equity stake acquisition in Chinese banks:
(1) Macroeconomic variables: GDP, GDP per capita, and GDP growth rate of home countries of foreign financial institutions.
(2) Interaction variables between foreign banks home country and China: bilateral trade, two countries’ capital distance, foreign financial institutions ranked in the
tion Rating System adopted by FFIEC,
e
l component analysis to subtract the reign financial institutions and Chinese banks and rotate them
world’s top 1,000 banks, and interest rate spread between foreign financial institutions home countries and China.
(3) Financial ratios variables of foreign financial institutions: CAMEL standards and indicators in Uniform Financial Institu
including 10 financial ratios as in Table 6: capital adequacy, asset quality, management, earning capability, and liquidity.
(4) Financial ratios variables of invested Chinese banks: the same 10 financial ratios as (3), and taking into consideration of the weights by asset size in observed tim when the Chinese banks received equity stake acquisitions.
3.4 Results and Analysis
In this section, we apply the principa financial ratio variables of fo
as principal component scores to be the alternative variables using the survival model. Thereafter, the maximum likelihood estimator is derived via Cox’s proportional hazard survival model, involving the macroeconomic variables of foreign financial institutions’ home country and interaction between foreign banks’ home country and China for the rotated principal component scores. We derive the optimal
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
probability hazard threshold of equity stake acquisitions for foreign financial institutions by minimizing type I and type II errors in the specific baseline hazard function.
3.4.1 Subtraction of Financial Ratios Using Principal Component Analysis
low expla
ponents subtr
ulative variation reaches more than 80% after subtracting one p
n Equity Acquisition
cquisition, the macro
on th
With the principal component analysis in the subtraction of 10 financial ratios of foreign financial institutions and Chinese banks to eliminate the variables of
nation, we have principal components mutually independent which still keep these alternative variables above a certain power of explanation and apply the principal component score as the financial alternatives in the survival model.
Table 7 presents the results of the principal component analysis regarding the financial ratios of foreign financial institutions. There are four principal com
acted from 10 original financial ratios variables, with 81% power of explanation, and are viewed as alternatives of these financial ratios. Hence, we take the score of principal components 1 to 4 as the alternative variables for financial ratios of foreign financial institutions.
With the same approach in dealing with 10 financial ratios of Chinese banks, Table 8 shows that the cum
rincipal component out and more than 92% after adding one more principal component. Thereafter, we take the score of principal component 1 as the alternative variable for financial ratios of Chinese banks.
3.4.2 Variables and Probability of Intensity i
When applying Cox’s proportional hazard survival model to discuss the motives and timing of foreign financial institutions in taking equity stake a
economic variables of foreign financial institutions’ home country, interaction between foreign bank’s home country and China, and financial ratios alternatives of foreign financial institutions and Chinese banks are taken into consideration. Here we see that the goodness-of-fit of Cox’s survival model is significant in Table 9, and a -2 log likelihood is less than the initial -2 log likelihood, and Cox2 ()2() to represent model is the most fitted.
The results in Table 10 show that the principal components 1 and 2 of the financial ratios of foreign financial institutions have a positive and significant effect
e probability of intensity of equity stake acquisition, which is the same as the
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
principal component 1 of the financial ratios of Chinese banks. This means that the foreign financial institutions with higher earning capability and better asset quality possess higher probability of intensity to improve the performance and earning capability of Chinese banks to make capital gain of equity stake acquisition and provide stability if China’s economy fluctuates. Furthermore, for the Chinese banks with better asset quality like the state-owned, joint-stock city commercial banks, they have higher probability of intensity in general to attract the investment. The lower acquisition cost by that time and subsequent improvement in earning capability has enhanced the value of the tie-ups. In addition, advancing capital adequacy ratio improvement meant that the local bank is closing the gap with international standards.
For the interaction between foreign banks’ home country and China, the results indicate that the ratio of top 1,000 banks ranked has positive and significant effects on the p
re and after Invested
NPL
oreign banks took
robability of the intensity of equity stake acquisition. It is observed that the relative probability of intensity of countries with higher financial development is higher generally than those countries of lower financial development. This echoes the results that higher earning capability and better asset quality of foreign financial institutions possess higher probability of intensity as illustrated previously. In addition, the variable of bilateral trade has a negative impact on the probability of intensity, which means the main purpose of foreign financial institution taking equity stake acquisition is to create the market value of invested Chinese bank for capital gain.
Except that those foreign financial institutions are under higher bilateral-trade, it would adopt the approach of setting up a branch or subsidiary other than equity stakes acquisition in the Chinese market expansion strategy.
3.4.3 Financial Performance of Chinese Banks befo
To further understand whether or not the financial performance of Chinese banks has improved after foreign banks made equity stake acquisition, we apply the
ratio to represent asset quality on the basis of financial ratios as the principal component alternative 1 and observe the average NPL ratio before the year of investment, during the year of investment, and in the year after investment. In addition, we observe the average NPL ratio and its trend by the nature of the Chinese banks: state-owned, city commercial banks, and rural cooperatives.
From the perspective of the overall invested Chinese banks in Figure 1, the average NPL ratio appeared to demonstrate a downward trend after f
equity stake acquisitions; the NPL ratio fell from 5.59% in the year before the investment activity to 4.07% (or decreasing 27.17%) at the year of investment and to
‧
3.25% (or decreasing 20.20%) the year after investment. This explains the events following Chinese banks’ forming strategic partners with foreign banks via equity investment; there was positive improvement on asset quality. For the impacts on the four categories of banks, we find that the extent and impacts are most significant on joint-stock commercial banks (above 30% on average), followed by state-owned, city commercial banks and rural cooperatives. This is because the joint-stock commercial banks and state-owned banks are China’s major banks, and the improvement of the asset quality is relatively important for the purpose of public listing and globalization.
In addition, we observed the capital adequacy and profitability of the Chinese banks before the investment activity took place, during the year of investment, and during the year after investment, using equity-to-assets ratio and return-on-equity ratio as the proxies for capital adequacy and profitability. We can see that both the equity-to-assets ratio and return-on-equity ratio showed upward trends in Chinese banks overall, and the capital adequacy and profitability have improved to a large extent, especially during the year of investment. Furthermore, the increases in equity-to-assets ratio and return-on-equity ratio of state-owned banks are significant during the year the investment took place, whereas during the year after investment, joint-stock commercial banks saw a significant increase in the ratios. This may be due to the simultaneous efforts in write-offs which affect capital adequacy and profitability at the year of investment, but it demonstrated a positive view in terms of more stabilized performance and enhanced capital structure in the future.
3.4.4 Optimal Probability Hazard Threshold of Equity Acquisition
In the Cox proportional hazard survival model, the baseline hazard function )
(t
h is irrelevant to the parameter , so we can assume it to be a specific form to
tion that equity stake acquisition rate is constant during the e the exponential form
0
figure out the probability of intensity of equity stake acquisition as
)) observable period, we us of significant variables x(t).
)
0(t
h as the specific baseline hazard function to derive the optimal probability hazard threshold 0.535 using the approach of minimizing type I and type II errors described in section 2. This means that if the calculated probability of intensity (i.e., default probability) is more than this threshold with current variables in the model, then it is the optimal timing to take equity stake acquisition. According to the macroeconomic variables of of the foreign
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
financial institutions’ home country, interaction with China, the financial ratios of both foreign financial institutions and local banks in China, and the maximum likelihood estimation together calculate the intensity of the share participation rate to be the optimal probability hazard threshold to minimize errors.
We overview the past observable samples with this optimal probability hazard threshold, and find that there are 25 with the relative probability of intensity among the 2
ploys the survival model to discuss the probability of intensity results show that those foreign financial instit
arket value of invested Chinese banks for capital gain,
7 foreign financial institutions taking equity stake acquisitions. This result is more than this optimal threshold (or reaching 92.5%), and 14 have a relative probability of intensity of 17 not taking acquisition are less than the optimal threshold (or reaching 82.3%), and show us that this optimal probability hazard threshold exists as a reliable and significant reference. Meanwhile, the optimal probability hazard threshold could be the initial reference for foreign financial institutions to determine whether or not to take equity stake acquisition with local banks. In other words, when the probability of intensity (same as the default intensity in the model) exceeds this threshold, then we can say it could be the best time for foreign financial instructions to engage in equity stakes acquisition.
3.5 Brief Summary