از بحران اعتباری به بحران بدهی: عوامل مؤثر بر عملکرد قیمت سهام بانک های جهانی

We compare and contrast the determinants of the share price performance of global banks in the credit crisis and the sovereign debt crisis. Higher loans and funding fragility, as measured by short-term funding, explain performance in the credit crisis, as banks could obtain short-term finance and hence take risks by lending more. In contrast, in the sovereign debt crisis banks with higher capital, tangible equity, deposit, lower agency problem, and smaller boards performed well. The banks that increased regulatory capital as per the policy prescription, relied more on deposit financing, and decreased board size performed well in the sovereign debt crisis. Interestingly, deposit insurance is negatively related to the performance in the sovereign debt crisis, as the governments were closer to default. We find some similarities in the share price performances of banks across these two crises. Beta and idiosyncratic risk explain the share price performances of banks in both the crises. We further examine the effect of regulations on risk, as returns should be compensation for taking risks. We find that banks in countries with higher restrictions and higher tier I capital are less risky, while countries with deposit insurance are more risky.

In this paper, we analyse the share price performance and riskiness of the 378 largest global banks during the credit crisis of 2007–08 and the sovereign debt crisis of 2010–11. After a credit crisis, the natural expectation is that banks will take measures to safeguard themselves against future financial turmoil; however, banks performed poorly again in the stock market during the sovereign debt crisis.1 We examine the impact of funding, loans, equity, regulations, and risks on the share price performance of global banks in both periods of stock market turmoil. We compare and contrast the cross-sectional determinants of the share price performance and riskiness of global banks in the credit crisis versus the sovereign debt crisis. Both of these crises had severe impacts on global banks (Cetorelli & Goldberg, 2012).2 As a result, in November 2011, the Financial Stability Board announced the systemically important financial institutions that are vital for the global economy to function well. This paper contributes to the literature by testing whether global banks strengthened their balance sheets in the aftermath of the credit crisis in order to defend against financial turmoil. Since there was another global financial crisis in 2010–11, this provides a good testing ground for our conjectures. Although some prior studies have examined bank performances in the credit crisis (e.g. Beltratti and Stulz, 2012, Demirgüç-Kunt et al., 2010 and Fahlenbrach et al., 2012), we are the first to study bank performances in the sovereign debt crisis. Demirgüç-Kunt et al. (2010) studied banks from a number of countries and used bank-fixed effects to examine the effect of bank capital on the share price performance of banks during the credit crisis. Regulations and country-level variables were not included in their models. Fahlenbrach et al. (2012) analysed US banks to examine their performance in the credit crisis. The findings of their study are limited to the US only. In contrast, we analyse global banks by using bank level, country-level regulations, and governance variables to test our hypothesis. Beltratti and Stulz (2012) analysed the performance of banks in the credit crisis; we analyse the performance of banks in the sovereign debt crisis. We extend previous evidence by performing an out of sample test on the performance of banks in the aftermath of the credit crisis by using data during the sovereign debt crisis. By comparing the share price performances of banks in the credit crisis and the sovereign debt crisis, we are able to understand the underlying dynamics of how different types of crises affect the share price performance of banks. We expect that some factors are important in the credit crisis, while others are important in the sovereign debt crisis. Particularly, we expect that funding fragility (Adrian and Shin, 2008 and Gorton, 2010) and loan (Ivashina & Scharfstein, 2010) are important determinants of share price performance in the credit crisis. In contrast, deposit (Beltratti & Stulz, 2012), core capital (Demirgüç-Kunt et al., 2010), and agency problem (Kashyap, Rajan, & Stein, 2008) should be important in the sovereign debt crisis. However, beta (Acharya, Pedersen, Philippon, & Richardson, 2010) and idiosyncratic risk should have similar effects in both crises. Most of the results are in line with our expectations. We find that too much reliance on short-term funding led to poor performance during the credit crisis, which is consistent with Gorton (2010). However, funding fragility was not important during the sovereign debt crisis, as the short-term funding market had already dried up as a result of the credit crisis (Adrian & Shin, 2008). At the same time, higher deposits led to better performance in the sovereign debt crisis, as banks could not obtain short-term funding (Beltratti & Stulz, 2012). We find that loans are negatively (significantly) related to performance in the credit crisis; however, the relationship is not economically significant in the sovereign debt crisis as the banks suffered liquidity problems and thus were unable to lend (Lane, 2012). We find that tangible equity and tier I capital were not significant determinants of bank performance during the credit crisis. However, they were significant during the sovereign debt crisis. This is possibly because banks strengthened their equity capital in the aftermath of the credit crisis; as a result, banks with higher tangible equity and tier I capital performed well during the sovereign debt crisis. Our results are consistent with the increase in regulatory capital, which was prescribed in Basel III (BIS, 2011). Agency problems explain returns in the sovereign debt crisis, but not in the credit crisis. We find that some aspects of bank performances are similar in both the crises. Banks are generally associated with higher market risks. We find that beta, idiosyncratic risk, and log z are negatively related to stock market performance in both crisis periods. We also show that stock returns and risks are positively correlated in the credit crisis and the sovereign debt crisis. This is consistent with Fahlenbrach et al.'s (2012) evidence that 1997–98 returns are a good predictor of the credit crisis returns for US banks. We extend their evidence by using global banks and showing that the returns and risks in the credit crisis and the sovereign debt crisis are related. Since the nature and origin of the credit crisis and the sovereign debt crisis are different, we find that pre-credit crisis data do not explain returns in the sovereign debt crisis. We contribute to the literature by examining whether lenient regulations were similarly responsible in the credit crisis and the sovereign debt crisis (as stated by Stiglitz, 2010) by using the 2008 World Bank survey on bank regulation data. We test whether the power of regulators, concentration, deposit insurance, and anti-director Index (ADRI) are related to performance. We find that the banks in countries with lower concentration, better institutions, and higher official scores performed better during both the crises. Deposit insurance is positively related to the returns in the credit crisis, while it is negatively related to the returns in the sovereign debt crisis. This is consistent with the notion that having deposit insurance in place worked fine in the credit crisis, but did not work well in the sovereign debt crisis when the sovereigns were affected. There are no studies that we know of that test whether board changes after the credit crisis helped banks perform better in the sovereign debt crisis. We test whether board changes made banks perform any better during the sovereign debt crisis.3 We find some support in favour of board changes. The smaller the board size was, the better the bank performed in the sovereign debt crisis. We find that bank performance was better during the sovereign debt crisis when the bank directors held fewer external board positions. This implies that fewer interconnections among boards are better, so that directors can focus more on a particular board. These findings are consistent with Guner, Malmendier, and Tate's (2008) study of network connections, and Hermalin and Weisbach's (2003) survey evidence on board size in normal market conditions. We extend their evidence for board size and network connections in the situation of crisis. Since returns should be a compensation for taking risks, we analyse determinants of risks in both the crises. Our main objective here is how regulations might have made the banks less risky. We find evidence that bank restrictions are negatively related to risk in both crises, which implies that the higher the restrictions, the lower the riskiness of the banks. Deposit insurance is positively related to the idiosyncratic risks of the banks, which is consistent with the moral hazard potentials created by deposit insurance (Merton, 1977). The higher the tier I capital, the lower the idiosyncratic risk, which means that better-capitalised banks could absorb more losses in times of crisis, and were perceived as less risky (Demirgüç-Kunt et al., 2010). The rest of the paper is organised as follows: the next section describes the origin and nature of the credit crisis and the sovereign debt crisis, and their effects on global banks. Section 3 develops a review of the literature and identifies the gaps in the literature. Section 4 discusses the methodology and provides data sources and descriptive statistics. Section 5 provides results on multivariate analyses. In Section 6, we carry out a few robustness checks. Finally, Section 7 discusses the results and provides the conclusion.

In this paper, we compare and contrast the determinants of bank performances across the credit crisis and the sovereign debt crisis. We examine in detail how various factors affected banks in these two crises. We find that banks with higher regulatory capital and tangible equity did not perform well during the credit crisis; however, during the sovereign debt crisis banks with higher capital and tangible equity performed well. Since banks lost a huge amount of equity in the crisis and, as a result, they had to be injected with equity capital, core capital is important in the sovereign debt crisis (Demirgüç-Kunt et al., 2010). While loans explain performance in the credit crisis, they do not explain performance during the sovereign debt crisis. This is due to problems in obtaining funds after the credit crisis, and, as a result, bank lending decreased during the sovereign debt crisis. This is consistent with Ivashina and Scharfstein (2010) who assert that loans are a significant determinant of performance in the credit crisis; but loans are unlikely to play any important role in explaining performance during the sovereign debt crisis. Similarly, funding fragility explains bank performances in the credit crisis, but not in the sovereign debt crisis. This is related to the fact that short-term funding sources were working properly before the credit crisis (Adrian and Shin, 2010), and as a result of the credit crisis, they dried up before the sovereign debt crisis (Acharya et al., 2011). Related to this, we show that banks that had higher customer deposits performed better during the sovereign debt crisis. Closely held shares did not explain the performance of banks during the credit crisis, but they explain the performance during the sovereign debt crisis. Closely held shares increased significantly in the aftermath of the credit crisis as policy interventions included the acquisition of significant stakes of several global banks. All the main results are summarised in Table 12.We find that returns during the credit crisis are a good predictor of returns during the sovereign debt crisis. When we estimate the pooled model, beta and idiosyncratic risks stand-out as significant factors that explain returns during both the crises. These two sources of risks are fundamental for understanding the risk exposures of the global banks during crisis periods. Our results show that banks are more sensitive to financial crisis than other institutions. This is very typical for the banks who deal with money. We show that pre-credit crisis data cannot predict the sovereign debt crisis, except beta. The results also imply that the nature and origin of the credit crisis is different from the sovereign debt crisis. While our research is not about why the crisis happened, it sheds some light on the changes in regulations and the changes in boards after the credit crisis. Because banks performed poorly in the sovereign debt crisis, it gives a testing ground for the effectiveness of the regulations and board changes. The results in this paper suggest that regulations of capital were effective in the aftermath of the credit crisis. The policy response in Basel III to increase tier I capital and tangible equity seems to be effective in the sense that higher core capital helped banks to perform better during the sovereign debt crisis. We show that board size and board connectedness were negatively related to performance in the sovereign debt crisis. This is in line with the policy response in several countries including the US and the UK to reduce the board size to a manageable number. However, we do not have the qualifications and expertise of the board members to further test whether board changes are effective. As returns should be commensurate to risks, we analyse the determinants of risks in both the crises. When we analyse idiosyncratic risks, we find that the higher the tier I capital, the higher the risk is in the credit crisis. Tier I capital was negatively related to idiosyncratic risk in the sovereign debt crisis. This shows that, in the aftermath of the credit crisis, banks became more careful and they took risks sensibly. Countries with better institutions and higher restrictions had lower idiosyncratic risk in both the crisis. In the presence of deposit insurance, banks took more risks, and thus we find that banks in countries with deposit insurance were more risky during the sovereign debt crisis as in the credit crisis.