Financial structure: lessons from the crisis: introduction.

Armstrong, Angus ; Davis, E. Philip

The global financial crisis is the most important economic event in the West for two generations. The challenge this presents to our thinking in economics and finance is only just beginning to be addressed. The foundations of much of our current theory, efficient markets and rational expectations, have been shown to be inappropriate. Even behavioural finance, the so-called 'radical alternative', while showing some important insights, was unable to make useful predictions ahead of the crisis. This matters because how we think about finance determines how we approach regulating institutions and markets.

The following set of articles in this Review can be seen as our first step in contributing to this fundamental rethink on finance and consider what the financial system should look like to best serve the interests of citizens. We have invited world renowned experts to reassess the received wisdom on topics in financial structure in the light of the crisis and its ongoing repercussions. Have some previously perceived strengths turned out to be weaknesses? Is there an 'ideal' financial structure or must this necessarily be nation specific? And without efficient markets what framework is required for regulation to be developed?

A traditional definition of financial structure is 'the mixture of financial markets and institutions operating in an economy', focused on whether the economy grows better with mainly banks or markets as sources of private sector finance. This had been expanded to include the 'law and finance view' (La Porta et al., 1998)--that the basis of financial efficiency, and a key aspect of financial structure, is the form of the legal system; and the 'financial services view' (Levine, 1997)--that in fact that structure per se is much less important than overall financial development. Our current papers broaden the scope further to include aspects of regulation, trust, management strategies and institutional structure.

Setting the scene is a paper by Angus Armstrong (NIESR) which draws on institutional economics (North, 1994) to consider the importance of trust in banking. Trust is interpreted as a rationally computed subjective belief that a counterparty will honour their obligations in a transaction, irrespective of whether they can be fully observed. In making their beliefs agents draw inferences from formal institutions, such as the legal framework, and also informal institutions, such as codes of conduct, customs and professionalism. Given the incomplete contracts and asymmetries of information inherent in much of modern banking, and how markets quickly respond to regulations, trust is shown to be a necessary requirement of efficient markets.

There are few sectors in the economy where trust is so highly valued as banking. Yet survey evidence since the crisis shows that trust has fallen sharply and continues to edge lower. This appears to be as a result of the seemingly endless banking scandals which have come to light. It is argued that the breakdown in the codes of conduct and professional norms was an outcome of merged businesses which trade in inherently different assets with different governance structures. While this led to rapid growth of capital markets, the complexity of the incomplete contracts created extraordinary opacity in large limited liability banks. Coupled with equity based compensation schemes, this created the means and incentive for a deterioration in conduct.

If bankers are to value appropriate codes of conduct which support trust, this will depend on the organisational environment of the banking system. This is consistent with ending the too-big-to-fail (TBTF) problem. The debate is often expressed by an analogy to a portfolio; either banks are homogeneous and each diversified, or banks are heterogeneous and the banking system is diversified. The key issue is whether a more diverse financial eco-system with speciality in specific tasks would improve transparency, simplify governance and encourage competition for trust.

E. Philip Davis (NIESR) gives a broad outline of comparative sectoral balance sheet developments in G-7 countries showing that there has been some convergence between countries traditionally seen as 'bank-dominated' and 'market-oriented'. For example, bank equity holdings in Germany and Japan are lower than in most other G-7 countries while there is a marked and striking convergence in corporate leverage. There remain contrasts, in particular in household debt which is much higher in market-oriented countries. Non-bank financial sectors, which contributed largely to the crisis, are much larger in the market-oriented countries, although they have also grown elsewhere. It could hence be argued that one lesson of the crisis is that these structural characteristics shown by the market-oriented countries are indicators of systemic vulnerability.

On the other hand, some bank-dominated systems such as in Germany and France were also afflicted even at the outset of the crisis, partly as a result of banks' focus on investing in complex structured products. Subsequently, there have been common difficulties with public debt, aggravated in some countries by a loose fiscal stance in the years leading up to the crisis. The balance sheets do not explain all aspects, for example Canada has remained robust despite similar portfolio patterns to the UK and US which were afflicted. Hence, while the balance sheets do suggest some lessons from the crisis they are also clearly only part of the story.

Martin Cihak and Asli Demirguc Kunt (World Bank) look further at the banking and market-oriented division in financial structure, notably during financial development and its implications for appropriate regulation. Whereas Davis' work shows the insights that can be gained from the flow of funds, the number of countries with such data is limited, restricting the statistical conclusions that can be drawn. Their analysis benefits from using the World Bank's wide-ranging database of financial development (freely available to researchers) which enables the authors to create proxies for bank and market development and to draw tentative conclusions based on a larger cross-section. For example, it is clear that there is a positive relationship between relative market development and economic development per se over 1970-2010. Whereas both bank credit and stock market value traded rise with GDP per capita, the relationship of the latter has a higher coefficient, implying a tendency for financial structure to shift to a market-based system at high levels of income. Markets are better able to finance the novel projects with intangible inputs that dominate in the advanced economies.

Looking at lessons from the crisis, a first point is that the relationship of banks, markets and growth did not break down over the crisis period, despite the UK and US being at the epicentre of the crisis. One additional reason could be the way market-based systems provide "multiple avenues of intermediation" (Davis, 2000). A second point is that the data suggest, again despite the role of the UK and US, that more bank-based systems have a higher incidence of crises. Financial crises are almost three times more likely in countries where the ratio of stock market value to private bank credit is less than one half. And of course many bank-based EU countries have become affected by the broader financial crisis, most recently Spain.

A regulatory implication is that it is inappropriate to limit evolution of the financial system towards a more diversified one. And, second, that complex market-based systems require a system of regulation addressing underlying incentives rather than the mechanistic-based system Basel agreements provide (although again the example of Spain suggests bank-based systems can also be vulnerable). The distortion of incentives prior to the crisis illustrate the point. For example, in the US official agencies and regulators may have responded to political and industry pressure to increase lower quality lending. More widely across countries, regulators did not enforce the powers they had. In a complex and opaque set of markets, market discipline failed to operate, while principal-agent problems promoting risk-taking were rife. The authors contend that increasingly complex regulation as proposed in Basel III will not prevent future crises, notably in market-based systems, as long as the incentive problems are not addressed. This mirrors findings in Armstrong (this issue).

Adrian Blundell Wignall, Paul Atkinson and Caroline Roulet (OECD) look at the need to change the business models of the major global interconnected banks (GSIFIs) in the light of their behaviour during the crisis. Their dominance of complex derivative markets implies a close link of such institutions to the market-based financial systems highlighted by Cihak and Demirguc Kunt (this issue) as needing regulation to focus closely on incentives. Meanwhile the poor coverage of derivative volumes in the flow of funds highlighted by Davis (this issue) shows how the authorities are often far behind the markets in terms of monitoring.

The OECD authors focus on three points, firstly TBTF problems raised by such institutions, secondly excessive leverage and third conflicts of interest in how such institutions behave and interact with their clients and regulators. First, TBTF is of particular relevance in over-the-counter derivatives markets where there is 'rehypothecation' of collateral implying multiple re-use and very high leverage. It is the implicit public subsidy of lender of last resort support that enabled the GSIFIs to underprice the related risk. Second, derivatives by their nature permit more leverage than underlying instruments and could be used by GSIFIs to minimise their overall capital requirements when using approved models under Basel II. The authors contend that derivatives are fundamental to the pre-crisis growth in bank leverage; and when volatility increased, the protection provided by netting and collateral proved inadequate, necessitating the massive provision of liquidity by central banks. Third, as regards conflicts of interest, one can highlight the incentives of rating agencies to overrate structured products--and regulatory arbitrage to reduce risk-weighted assets and hence capital needs relative to unweighted assets. Meanwhile principal-agent problems relating to dominance of managers over shareholders as well as the remuneration packages of the former ensure corporate governance is weak in restraining risk-taking.

Changing the business models of the GSIFIs is seen as essential to preventing a recurrence of the crisis. There needs to be separation at least of retail banking and securities business and removal of implicit guarantees on the latter as well as no access of securities subsidiaries to firm-wide capital. This would effectively reduce incentives for leverage in the derivatives business, forcing changes in business models to correctly price risk, as well as allowing firms to fail with less disruption, as hedge funds generally do (LTCM was an exception). The OECD favours a non-operating holding company model for financial reform as proposed by Vickers in the UK.

Kyuil Chung, Hail Park and Hyun Song Shin (Bank of Korea and Princeton University) also look at complex securities, in this case the impact of derivatives on systemic stability. The context is Korean demand for hedging risks, which in booms can lead banks to adopt vulnerable balance sheet positions, notably in terms of exposure to short dollar debt. This links to the broader topic of the vulnerability of economies, particularly emerging markets, to reversal of short-term foreign currency capital inflows built up during booms. These in turn give rise to a risk of mutually reinforcing banking and currency crises. Banking sector inflows in turn were a major feature of emerging markets during the build-up to the crisis, eclipsing in particular the amplitude of portfolio inflows.

In common with many aspects underlying the financial crisis, individually rational actions of agents such as hedging by exporters and asset managers (importers typically do not hedge) can contribute to systemic instability in the counterpart banking sector that sell the hedges. This is particularly the case since the financing of the hedging activity may itself generate domestic currency appreciation that in turn generates further hedging demand. Furthermore, banks were themselves in a period of intense competition for market share of such business, leading some exporters to over hedge and become vulnerable themselves to losses during the crisis. The importance of regulators focusing on incentives (Armstrong and Cihak and Demirguc Kunt this issue) is again underlined.

The authors' argument is that banks should not be left to finance such hedging, which entails accepting maturity transformation by, in effect, borrowing short-term in dollars against a long-term dollar claim. Instead institutional innovation in financial structure should take place, namely a public institution able to offer hedging services without taking on short-term foreign currency liabilities, and thus cutting the link between hedging and systemic vulnerability. The vehicle would be equity financed and structured so as to hedge the US dollar exposure. This is achieved by holding local currency assets valued in US dollars, thereby creating a natural short dollar position.

The authors remind us that the development of derivative markets, while offering key benefits to individual agents, can heighten systemic instability and this was shown powerfully during the financial crisis even in emerging market countries. Such structural innovations need to be viewed at a macroprudential level in order to see the nature of the challenges they pose to financial stability, including such aspects as the asymmetry between hedging needs of exporters and importers. And institutional innovation in the structure of the financial market may be needed to offset some of the externalities arising from financial change.

REFERENCES

Davis, E.P. (2001), 'Multiple avenues of intermediation, corporate finance and financial stability', Working Paper No. 01/115, International Monetary Fund, Washington DC.

Frydman, R. and Goldberg, M.D. (2011), Beyond Mechanical Markets, Princeton University Press.

La Porta, R., Lopez-de-Silanes, F., Shleifer, A. and Vishny, R. (1998), 'Law and finance', Journal of Political Economy, 106, pp. 1113-55.

Levine, R. (1997), 'Financial development and economic growth: views and agenda', Journal of Economic Literature, 35(4), pp. 688-726.

North, D.C. (1994), 'Economic performance through time', American Economic Review, 84(3).

Angus Armstrong and E. Philip Davis*

* National Institute of Economic and Social Research. E-mail: a.armstrong@niesr.ac.uk; e_philip_davis@msn.com.