What will financial institutions look like in the future? What will be the impact of the financial crisis on the shape of the financial industry? Government policy, market process and entrepreneurial innovation are all likely to have an impact on these questions.
Policy makers have in practice driven consolidation of the industry in the last 12 months, creating banks from the largest securities companies and consumer finance companies. Yet policy makers continue to discuss the separation of commercial banking and investment banking in some form, worry about ‘too big to fail problems’ and hypothesize regulatory changes that would drive industry structure. Markets are planned to be restructured from their original form to exchange traded variants. Offshore and off-balance sheet structures that have a long track record are now being wound up.
Little of the policy debate seems well informed by the actual realities of markets and businesses in the industry. Here we look at what market signals might be saying about the future shape of the financial sector and compare this to proposals for change in the industry.
A slippery slope – navigating financial risk
Several problems emerged within the financial sector, all with different outcomes in performance. Foremost amongst these are arguably failures in risk management. Some institutions within the financial sector, Credit Suisse and Goldman Sachs, started to pull back from the mortgage securities market as others, UBS and Merrill Lynch, increased their exposures.
There are also banks that largely avoided the use of off-balance sheet conduits. Some banks managed to minimise illiquid securities on their balance sheet, whilst others had huge inventories held for sale.
When liquidity became an issue for the financial sector, there were commercial banks that had strong liquidity from retail deposits, but also some that were underfunded. There were investment banks with much stronger committed liquidity support. Some banks, whilst they had exposures to problem areas, managed those risks better, either through hedging or with sufficient capital to absorb losses.
Some hedge funds performed well through the crisis. Those that did not, and many that failed, did not cause the systemic risks that had been feared in earlier incidents of financial upheaval.
How do we explain these wide differences in performance? Shouldn’t regulatory reformers be looking at the different performance of institutions to identify the direction of industry changes? If a regulatory framework had led to an extremely divergent performance, is it realistic to expect new rules made in the same fashion to deliver more uniformly stable performance?
Many of these differences in performance have origins in the industry structure. That structure has been strongly influenced by both present and historic regulation. The separation of commercial banking and securities companies in the United States was a legacy of post depression legislation.
Opportunities for finance and insurance companies to operate with more flexibility in some parts of their business are the result of blurring differences between financial products that invalidate old industry boundaries. Still, within every segment of the financial industry, there are banks, securities companies and insurers that managed risks better.
It follows that financial institution strategy, culture and management was also important. Responding to market, regulatory and institutional change, companies make choices. Some of these are wrong. Anyone working in a major financial institution is aware of the frailties of decision making in those firms.
As scale increases, information flows are weaker. As products become more complex, top management knowledge and expertise is typically weaker. As profits in some areas become larger, influence within institutions shifts. Rewarding risk managers for saying no to profit making, but excessively risky opportunities is difficult. Design of incentives in large companies is an inexact science, where teams and individuals make a contribution.
Within each of the specific markets where there were problems – mortgages, securitisation, credit default swaps, SIVs and conduits, commercial real estate, leveraged finance – there were micro-structures, regulatory issues and practices that need to be addressed. However, are there systemic issues that can be identified behind risk management failures in the recent crisis and these challenges?
In economics, the theory of the firm, pioneered by Ronald Coase, tells us that firms exist to reduce transaction costs from external markets. Given that large parts of financial markets are amongst the most ‘efficient’ markets that we know, it seems anomalous that in the financial sector there are so many institutions of such size and complexity that pursue greater scale despite the obvious costs. Given the pervasiveness of large firms with huge internal transactions and capital allocations that are to some extent outside the market, should we be surprised to see apparent ‘market failures’, which are in reality failures of managerial outcomes rather than market allocation of resources.
My hypothesis would be that many of the failures by firms that contributed to the crisis are the result of the massive structure of legislation, regulation, taxation, supervision, consumer protection and competing non-market provision of financial services that make markets less effective than they could be in the financial sector.
Asking what regulations we need to reduce the risk of financial crisis is the wrong question. So is asking how we should redesign institutions to reduce risk. The right issue is how to eliminate transaction costs in the financial sector that impair the market allocation of resources which make more pervasive ‘principal-agent’ problems and increase moral hazard.
Pouring oil on a slope – proposals for regulatory change. It follows that many of the proposals for change to the financial system are flawed. Rather than help improve management of risk, they pour oil on the slope.
Separation of commercial and investment banking
Separation of commercial and investment banking is one of the most common proposals from economists. Part of the idea is that regulators are very focused on depositor protection and that if depositors can be isolated from trading risks, securities activities can be left less regulated. ‘Too big to fail’ problems are reduced and the costs of depositor insurance is lowered.
A more radical version of the same proposal would require 100 per cent reserve backed depository institutions.
There are two problems with these proposals. First, it is practically at odds with recent experience that saw ‘universal’ banks as the survivors and largely eliminated big securities companies. Second, and more importantly, the distinction that might once have been meaningful is now moot. Deposit products can take many forms and depositors have shown a desire for yield that will always see them allocate funds away from a narrow depository institution.
It is not at all clear why depositors should be protected from securities trading risks, but not credit risks in traditional lending. Indeed in practice, interest rate mismatch-risks have often been as big an issue for bank profitability as credit. In modern financial markets, trading is also an important part of almost all commercial banks. Foreign exchange and interest rate risk management products, as well as simple debt securities like commercial paper, are an integral part of operations. It is not possible to draw a line between commercial and investment banking activities.
Compensation limits are proposed to reduce incentives to taking excessive risk and align shareholders with employee interests. Yet legislative proposals entirely miss the point that the ‘principal-agent’ problem at the core of bonuses paid for poor performance arises because of inadequate information, difficulty in framing suitable contracts and the high costs of moving to structures with superior incentives. Compensation limits will likely only ‘work’ to reduce incomes in circumstances where they are not needed and where performance attribution is easy and akin to a sales commission, e.g. equities broking or advisory businesses.
This would drive those activities to boutiques. Limiting compensation will face the same problem as managers of financial institutions where the time frame of profitability is long, capital allocation difficult to determine and individual contributions hard to assess. Promoting rather than limiting markets is a better solution to that problem.
Eliminating the ‘shadow financial system’
Eliminating the ‘shadow financial system’ of structures that have been largely ‘off balance sheet’ and domiciled in offshore jurisdictions is proposed to increase regulatory scrutiny and transparency. However, the jurisdictional competition provided by the ‘shadow’ system has been crucial to reducing transaction costs – it has allowed efficient dispersal of risks. These very factors led to the growth of the ‘shadow’ system. The answer to the shadow financial system problem is not to eliminate it, but to reduce transaction costs in all jurisdictions.
Change of regulatory responsibilities
Change of regulatory responsibilities seems inevitable. It also seems largely beside the point. Unified regulation worked poorly in the UK during this crisis, although arguably it was pre-empted by panicked political responses. Fragmented regulation worked poorly in the US. In fact, far more systemic risks emerged from heavily regulated companies (AIG) than lightly regulated companies (hedge funds).
The very process of changing regulatory institutions adds to costs and makes it more difficult for market participants to adjust to change. Notionally ‘functional’ regulation, e.g. treating providers of services with a similar economic effect equally, seems to make more sense than the institutional regulation of the US. However, identifying tomorrow’s systemic risks is more likely to come from competitive processes and competing jurisdictions than a single regulator with that focus.
The focus amongst politicians on bringing hedge funds under the umbrella of existing regulators is baffling. After the Long-Term Capital Management crisis, counterparties revisited their relationships with hedge funds, improved their control over collateral and improved management of leverage. These processes proved robust during the crisis, albeit causing problems for some funds. What government regulators can add to this process is not apparent.
Higher capital requirements
Higher capital requirements are proposed to build confidence, provide a buffer for losses to prevent failure and for the largest institutions to limit counterparty risks that could have systemic consequences. However, it can be argued that the very detailed Basel 2 capital requirements were contributors to the crisis as banks shifted systematically to lower capital allocation products and moved away from products where capital requirements were increasing.
Capital requirements are a large barrier to entry that makes it harder for new competitors to enter the market. Regulatory requirements have been manifestly too low in some areas, whilst giving counterparties false confidence about capital in others. In the hedge fund industry, prime brokers are effectively the regulators of leverage for their clients. Although they have made mistakes, they have done a much better job than government regulators of banks or ratings agencies. This might provide a hint of a better model for governing financial leverage and capital than the current approach.
Stress tests were designed to reassure counterparties and customers that institutions were financially sound. In practice they undermined confidence in institutions, created a long period of uncertainty and applied arbitrary rules that likely required some banks to raise capital that they did not need. This undermined private attempts to provide capital to the financial system. Financial institutions should continually test their financial stability under low probability, high cost scenarios and make plans to deal with those contingencies. Many did that reasonably effectively.
‘Sharing’ risks in securitised loans
‘Sharing’ risks in securitised loans is a more recent proposal to align the interests of originators of loans with investors. However, it will again require participants in the securitisation process to be larger and better capitalised institutions, reducing competitive pressures. Many investors do want originators to have ‘skin in the game’, but the proposed requirement to have a five per cent stake in any issue pre-empts more efficient and lower cost alternatives to the same problem.
Rating agencies arguably carry a significant burden of responsibility for the poor information they provided about risk for investors as the ‘AAA’ brand was broadened to cover different risks. The potential conflicts within the model where issuers pay for the service have been widely discussed. The power of incumbents was driven by lack of competition as much as the fee model. Alternative fee structures could not compete with the privileged position of incumbents that was mandated by regulators.
In short, none of the major proposals for industry reform address the underlying drivers of both company and specific market failures. Substantial regulatory changes are needed. There are good models of ‘deregulation’ that increase the scope of markets, adding to the stability and efficiency of the financial system.
Building a ladder on the slope
What form should regulatory and institutional reform take?
The imperative is to reduce the adaptation costs of the industry and allow markets to emerge as substitutes for managerial, legislative and regulatory decisions that all went astray in the crisis.
Institutional change and adaptation is very difficult in the financial sector. Elaborate licensing, high capital requirements and complex compliance issues mean very high barriers to entry.
This makes it difficult for new firms that are often smaller with better incentive structures and a more focused strategy to emerge. Both consolidation and break-up of financial institutions needs to be made much easier and at lower costs. This will allow specialists to set benchmarks as new markets emerge and allow more ‘boutique’ structures with less principle-agent problems to play a bigger role in the industry.
Taxation has been a pervasive driver of the structure of financial institutions, product development and leverage. Corporate and transaction taxes lead to innovative product structuring, but also reduce the efficiency of markets and can increase risk. Dealing with the tax issue is not a matter of closing loopholes, rather it requires less and lower taxes to reduce distortions. Only lower tax rates effectively reduce the incentive to amass excessive corporate debt caused by the non deductibility of the cost of equity funding. It should be no surprise that companies and financial institutions in the low tax jurisdiction Hong Kong have much more conservative gearing than many of their international peers.
Although currently not in favourable standing, the benefits of jurisdictional competition have been proved by the different performance of regulators globally through the crisis. The trend to convergence, especially focused on systemically important institutions, will reduce adaptation in the financial sector.
Although the world experienced a very global crisis, local problems cause specific responses and improvements from countries and companies. Many improvements in regulation globally came from the specific experiences of small countries like New Zealand in the 1980s or Asia in the 1990s.
Efforts to increase transparency in the financial sector are likely to be beneficial, but only to the extent that they are pursued at a reasonable cost. One of the ironies of the ‘Spitzer’ changes to equity research is that it reduced investment in research, the quality of research and the prospects of analysts identifying new information.
The reforms decreased access to company management and constrained opinionated research, which reduces the chance of identifying new areas of problems in the sector. Many accounting changes also make it more difficult to understand underlying financial trends. For example, mixing mark-to-market valuation changes with actual income and expenses in profit and loss statements obscures rather than clarifies. Competition to identify new information is likely to produce better outcomes than just requirements to disclose more information.
Incentive changes are also important. It is not as simple as reducing the size of payments or making them longer term. Looking within investment banks, there are many challenges in the most common compensation models. Profitability of trades might not coincide with the calendar year or reporting cycle. Charges for capital allocation to riskier activities are often not done well. Cross-subsidies between divisions are legion and allocations to management functions compared to ‘producers’ are difficult. However, these issues are so important for the industry that they cannot be left to central rules and plans. Management is constantly trying to improve incentives. Where they fail, good people leave to work for competitors or establish themselves as the new competition.
Note that many putative compensation ‘excesses’ have coincided with a move away from options-based incentives now that these are expensed. Simple models for expensing options often impose too high a cost on long dated and well out of the money options that can provide strong alignment between shareholder and staff interests.
Internal capital allocation to optimise returns is also made more difficult by proscriptive regulatory capital rules that open a gap between economic and regulatory capital. Regulatory capital processes should retreat from detailed and proscriptive approaches to simple, general rules that provide minimum guidelines.
Markets, left to their own devices before the financial crisis, were driving substantial changes, which at the margin improved the structure of the financial industry. These changes should not be hampered or ignored.
Parts of the industry were consolidating and seeing higher returns with scale. Processing parts of the financial sector were consolidating quickly, sometimes within large firms, sometimes with specialists consolidating across firms. Credit card and merchant processing, custody, trade finance and securities clearing were areas consolidating globally.
Alongside this was a trend towards less capital intensive parts of the industry moving into smaller or ‘boutique’ firms in areas like advisory and cash equities. This is probably largely a feature of better incentive structures in smaller firms, where human capital is more important than financial capital. Supporting this trend, new technology has made it easier to operate with ‘global’ coverage on a smaller scale.
Even in relatively capital intensive trading businesses, there was a shift of activity and talent from investment and universal banks to hedge funds.
Reasons for this would include the greater efficiency of external investors directly allocating capital between strategies, rather than internal management allocation in banks. In particular, external investors were good at removing low return capital from underperforming managers. Clearly, incentive structures were superior for many people within hedge funds. The ability to relatively easily establish and close hedge funds meant a dynamic and very competitive market.
Even within large banks, some trading activities moved into more hedge fund-like internal structures with mixed results.
Universal banks were also in the process of breaking up. Many very diversified institutions had exited asset management and insurance. Strategies in the largest financial firms were starting to diverge.
These trends were pointers to an industry that would emerge with some of the largest firms providing utility services to a much more fragmented, competitive and market based constellation of smaller firms. Specialisation within the industry was growing. The exchanges market was becoming much more competitive and many over-the-counter products were migrating to exchange traded alternatives. These trends were not mature, but arguably would have reduced ‘systemic’ risks.
In conclusion, attempts to restructure the financial industry are misguided and can in the long term compound inefficiencies and financial risks in the financial system. The alternative is to focus on making markets work better. This classically requires lowering transaction costs, reducing barriers to entry, increasing adaptability of financial institutions and very importantly, reducing tax burdens and distortions.
There are powerful models of this style of reform working well in financial markets around the world. Markets were ‘reforming’ themselves prior to the crisis. It would be a shame if the response to the crisis reduced innovation and adaptation in the financial sector and actually increased risk.