VoxEU Column Financial Regulation and Banking Global crisis

Bank competition and financial stability: The role of financial innovation

An unprecedented process of deregulation took place in the banking sector in the three decades prior to the Global Crisis. This column argues that during periods of intense bank competition, financial innovation can compound the adverse effects of competition on stability. Coupled with strong competition, the significant use of one such innovation – securitisation – in the run-up to the crisis was related to high levels of bank risk.

Although historically the connection between bank competition and financial stability has been weak, the 2007-2009 financial crisis brought this important relationship to the fore. In the five decades following the Great Depression, the banking industry in most developed countries was tightly regulated so that competition was largely contained. In contrast, in the three decades prior to the 2007-2009 financial crisis, an unprecedented process of deregulation took place in the banking sector. During this period, efficiency and competition considerations overshadowed financial stability concerns (Vives 2016).

One strand of the literature argues that bank competition endangers financial stability. In a nutshell, it poses that increased competition for deposits erodes profits thus lowering the market power of banks (Matutes and Vives 2010). This depresses the charter value of banks and encourage them to take on new risks. While the issue is probably more nuanced, and differences in the focus and assumptions of different models matter, there is substantial evidence suggesting that very high levels of bank competition can endanger financial stability (Beck 2008).

A parallel strand of work shows that large increases in securitisation activity in the decades prior to the crisis were a major factor driving incentives in the banking sector in most developed countries (Marques-Ibanez and Scheiche 2010). The use of securitisation has a number of benefits. It allowed banks to turn traditionally illiquid claims – overwhelmingly in the form of bank loans – into marketable securities and to sell them off to investors. In principle, from the perspective of individual banks, securitisation allowed the diversification of risk portfolios both geographically and by sector. At the same time, banks may have also responded to the static reduction in risks due to securitisation by taking on new ones. There is in fact empirical evidence suggesting that banks responded to securitisation by loosening their lending standards (Mian and Sufi 2009).

The effects of innovation and competition on financial stability

In a recent paper, we exploit the 2007-2009 crisis to explore whether competitive conditions prior to the crisis were related to the likelihood of a bank rescue (Altunbas et al. 2016). Our empirical exercise uses securitisation as an illustration to show how financial innovation might augment the negative effect of competition on bank risk. Banks in more competitive markets and with high levels of securitisation activity can be expected to have riskier loan portfolios, and these banks may have fewer incentives to monitor their loan book (Ahn and Breton 2014). The setting also recognises the central position of bank capital as a supervisory tool influencing banks’ behavior.

We use a large sample of listed banks in the EU and the US. For each country, we construct a measure of competition. Based on the ex ante literature, we also add a number of other factors likely to impact on bank risk. More importantly for our purposes, we include the interactions between competition and two bank characteristics: securitisation and bank capital. In this way, we can assess whether banks with different intensities of securitisation use and capital levels adopted riskier strategies, conditional on high competition. 

Using the Boone indicator1 as our main measure of competition, we find that securitisation increases the direct effect of competition on bank risk. In other words, as competition increased, banks with higher levels of securitisation activity tended to have riskier profiles as the crisis erupted. One possible explanation for these results is that enhanced competition before the crisis might have led to greater use of securitisation activities which eventually resulted in an increased risk profile by banks. In other words, it might be that securitisation by itself is not deleterious from a financial stability standpoint, but rather that it is the interaction between competition and securitisation that matters. In periods of high competition, securitisation might offer a way for banks to take on new risks subject to less supervisory scrutiny.

Since our specification uses a non-linear model, our estimates do not necessarily mean that the interaction effects between competition and securitisation are positive and significant across the whole sample. Figure 1 shows the corrected interaction effects for the variable interacting competition and securitisation for the sample. The thrust of the interpretation remains: the interaction effects are negative only for a small percentage of the sample population. 

Figure 1 Interaction effect of competition and securitisation

Panel A: Marginal effects

                                            

Panel B: Z-statistics

Some policy implications

Our results suggest that supervisory, regulatory, and competition authorities would benefit significantly from regularly assessing the combined effect of competition and innovation (in this case, financial securitisation) on financial stability. This assessment would probably also include other considerations, such as efficiency gains derived from financial innovation or competition (Carletti et al. 2002).

However, this may be challenging as it would involve coordination among several institutions also as it might also include other considerations such as efficiency (Carletti et al. 2002). For instance, macro prudential supervisors and other institutions in charge of financial stability would need to cooperate and regularly exchange data with competition authorities. A first step in this direction could be the development of measures of competition that can be integrated in the financial stability framework of these institutions. After all, speaking the same language is a prerequisite for cooperation.

Authors’ note: We are grateful to Philipp Hartmann and Skander Van den Heuvel for very helpful discussions on this theme.  The views expressed here are those of the authors and do not necessarily represent those of the institutions with which they are affiliated.

References

Ahn, J H and R Breton (2014), "Securitisation, competition and monitoring", Journal of Banking and Finance 40(C): 195-210.

Altunbas, Y, D Marques-Ibanez and M Van Leuvensteijn (2016), "Competition and bank risk: The effect of securitisation and bank capital", ECB Working Paper 1678 and International Monetary Fund Working Paper Series, Forthcoming.

Beck, T  (2008), "Bank competition and financial stability: Friends or foes?", World Bank Policy Research Working Paper Series 4656.

Carletti, E, P Hartmann G Spagnolo (2002), "Implications of the bank merger wave for competition and stability", in Risk measurement and systemic Risk, Proceedings of the Third Joint Central Bank Research Conference, 38-50.

Marques-Ibanez, D and M Scheicher (2010), Securitisation: Causes and consequences, in Handbook of banking, Oxford: Oxford University Press.

Mian, A and A Sufi (2009), "The consequences of mortgage credit expansion: Evidence from the U.S. mortgage default crisis”, Quarterly Journal of Economics 124: 1449-1496.

Matutes, C and X Vives (2000), "Imperfect competition, risk taking, and regulation in banking", European Economic Review 44: 1-34.

Van Leuvensteijn, M, A Van Rixtel and B Xu (2016), "Lessons from China on bank competition at the zero lower bound interest rate", VoxEU.org, 12 June. 

Vives, X (2016), Competition and stability in banking: The role of competition policy and regulation, Princeton: Princeton University Press.

Endnotes

[1] For an explanation of the Boone indicator see Van Leuvensteijn et al. (2016).  

7,454 Reads