سایه روشن TARP

This paper empirically investigates the impact of the first announcement of TARP, the announcement of revised TARP, respective capital infusions under TARP–CPP and capital repayments on changes in shareholder value and the risk exposure of supported US banks. Our analysis reveals a light and a dark side of TARP. While announcements as well as capital repayments may provoke positive wealth effects and a decrease in bank risk, equity capital injections to banks are observed to be a severe impediment to restore market confidence and financial stability. Furthermore, while TARP announcements and capital injections may increase systemic risk, no significant effect on systemic risk is found for capital repayments.

The 2007–2009 global financial crisis triggered a unique liquidity shock affecting a number of banks worldwide. As a response, comprehensive governmental capital assistance programs have been introduced in many countries. As regards the US, under the “Troubled Asset Relief Program” (TARP) the Department of the Treasury provided USD 204.9 billion in capital to 707 institutions in 48 states helping banks to absorb losses from toxic and illiquid assets ( U.S. Treasury, 2010). Similarly, in Europe 20 bank debt guarantees and 15 bank recapitalization schemes as well as 44 cases of individual bank aid were dealt with by the European Commission under the state aid rules during the crisis period. At the height of the crisis, the total of aid effectively committed amounted to 13% of the GDP of the EU ( CEPS, 2010). In most cases capital assistance programs were politically justified by the objective of liquidity creation to restore confidence in the banking industry, with the ultimate goal of overcoming the “loan freeze problem” by stimulating the banks’ lending activities and promoting financial stability for the economy as a whole. Nevertheless, partial nationalization of large banking groups revived the debate concerning the benefits and costs of providing a lender of last resort and government ownership of banks. In this context, in particular capital injections into banks are questioned for the following aspects. It is argued that the effectiveness of bailout assistance may be challenged by an insufficient monitoring of supported banks in combination with inadequate reporting requirements as regards the supported banks’ reinvestment strategies after having received capital injections. Accordingly, although capital assistance is given to increase bank stability and to reduce incentives to take excessive risks, it was also given with the understanding that the injected capital would be used to expand lending during a period of financial crisis. Taking this into account capital assistance may plant the seed of future distress by exacerbating moral hazard problems triggering excessive bank risk-taking through “zombie lending”. As a consequence, revitalizing bank lending activities and promoting financial stability may not work when market confidence is still weak (Beck et al., 2010). Against this background and since no empirical consensus exists yet on whether capital assistance programs are reliable instruments, the analysis at hand empirically investigates if the U.S. “Troubled Asset Relief Program” may have contributed to restore market confidence and to promote financial stability. In particular, employing data on supported US banks we analyze the impact of the first announcement of TARP, the announcement of revised TARP, respective capital infusions and capital repayments on changes in bank shareholder value and bank risk as perceived by the capital market through share price reactions. Empirical results reveal a light and a dark side of TARP. While the first and revised announcement of TARP as well as capital repayments may provoke an increase in bank shareholder value and a decrease in systematic risk, equity capital injections are observed to be a severe impediment to restore market confidence and to promote financial stability. Furthermore, while TARP announcements and capital injections may increase systemic risk, no significant effect on systemic risk is found for capital repayments. Baseline regressions results hold even when performing a large variety of robustness checks while subsample regressions reveal further important insights into the impact of the four TARP events on bank shareholder value and risk. Our analysis complements and extends previous event studies on TARP (Kim and Stock, 2012, Elyasiani et al., 2011, Veronesi and Zingales, 2010 and King, 2009) for several aspects. First, to the best of our knowledge this is the first comprehensive study that empirically investigates the impact of the entire set of four TARP events on wealth effects and systematic risk. Second, additionally analyzing the decomposition of the beta factor (idiosyncratic risk and systemic risk) allows a deeper insight into the drivers of the change in bank systematic risk due to the four TARP events. Third, as regards the variability of systematic and systemic risk during respective event windows we allow for (a) different model parameters of systematic and systemic risk before, during and after the event window and (b) gradually changing systematic and systemic risk within the event window. Consequently, our analysis reveals a significant change in systematic and systemic risk during the event window period which has remained undetected by related event studies on this issue so far. Finally, while previous studies have not accounted for conditional variance at all, we employ a GARCH structure throughout all regressions in order to address volatility clustering in our time series of bank stock returns which is even more important during periods of financial stress. The remainder of the paper is organized as follows. Section 2 presents a brief description of TARP and introduces the theoretical background and related empirical studies. While Section 3.1 describes data and sources, the econometric approach is presented in Section 3.2 and elaborated in more detail in the Technical Appendix. Baseline regressions, robustness checks and subsample regressions are discussed in Section 4. Finally, Section 5 concludes.

Employing four event dates of the US “Troubled Asset Relief Program” (TARP) this paper empirically investigates the impact of the first announcement of TARP (September 19, 2008), the announcement of revised TARP (October 14, 2008), respective capital infusions and capital repayments on changes in shareholder value and risk exposure of 125 supported US banks as perceived by the capital market through share price reactions for an entire sample period from September 19, 2008 to June 16, 2010. The empirical analysis at hand suggests that both announcements of TARP and respective capital repayments by banks may provoke an increase in bank shareholder value and a decrease in systematic risk whereas equity capital injections are observed to be an impediment to restore market confidence and to promote financial stability. In addition, while announcements and capital injections may increase systemic risk, no significant effect on systemic risk is found for capital repayments. Subsample analyses further reveal that TARP may fail to encourage a decrease in market uncertainty and bank systemic risk even when forcing the largest US commercial banks to accept capital assistance under the TARP–CPP. In contrast, empirical results rather indicate that in particular capital-aid towards smaller US banks may be an impediment to restore market confidence and promote financial stability. Finally, capital injections into “qualifying”, unforced banks may be anticipated by capital market investors as a signal of higher financial distress whereas evidence suggests that repayments by forced banks are perceived as a signal of stronger financial recovery during a period of recovering capital market conditions. Empirical findings remain robust even when performing a large variety of robustness checks, especially when employing control group regressions. However, as we cannot completely rule out that the statistical impact of capital injections and repayments on bank risk may be slightly biased by a general financial trend and resulting reverse causality effects, findings must be observed with caution. Nevertheless, the analysis at hand generally reveals a light and a dark side of TARP. While the announcement of TARP and revised TARP including the TARP–CPP may help restore confidence and promote financial stability in the US banking sector, respective capital infusions seem to harm bank financial soundness and confidence among capital market investors. If this is true, an increasing uncertainty due to capital assistance could be the result of an insufficient monitoring of supported banks in combination with inadequate reporting requirements as regards the supported banks’ reinvestment strategies after having received TARP capital injections. Therefore, future financial rescue programs should account for a consistent framework setting clear standards concerning (a) the selection of viable “qualifying” banks, (b) the instruments of intervention and (c) an effective ex post-monitoring of supported banks in order to avoid adverse effects from capital assistance.