Robust and Resilient Finance
Modern banks – and most other financial institutions – trade in securities, and the growth of such trade is the main explanation of the growth of the finance sector. Lending to firms and individuals engaged in the production of goods and services – which most people would imagine was the principal business of a bank – is literally an insignificant part of their balance sheets. For British banks today, it amounts to less than 5% of their total assets.[1] These assets consist mostly of the liabilities of other financial institutions, and conversely the liabilities of banks are predominantly owed to other financial institutions.
The finance sector establishes claims against assets – the operating assets and future profits of a company, or the physical property and prospective earnings of an individual – and almost any such claim can be turned into a tradable security. ‘High-frequency trading’ is undertaken by computers which are constantly offering to buy and sell securities. The interval for which these securities are held by their owner may – literally – be shorter than the blink of an eye. Competition to speed up the critical links in the global financial system – above all between Chicago and New York, but also across the Atlantic and over the pole to Tokyo – has reduced the time such communication links take to close to the physical limits established by the speed of light.[2]
World trade has grown rapidly, but trading in foreign exchange has grown much faster. The value of daily foreign exchange transactions is almost a hundred times the value of daily international trade in goods and services.[3] The annual volume of payments processed in Britain is £75 trillion, about forty times Britain’s national income.[4] Trade in securities has grown rapidly, but the explosion in the volume of financial activity is largely attributable to the development of markets in derivatives, so called because their value is derived from the value of other securities. If securities are claims on assets, derivative securities are claims on other securities, and their value depends on the price, and ultimately on the value, of these underlying securities. Once you have created derivative securities, you can create further layers of derivative securities whose values are dependent on the values of other derivative securities – and so on. The value of such derivative contracts outstanding is three times the value of all the physical assets in the world.[5]
What is it all for? What is the purpose of this activity? And why is it so profitable? Common sense suggests that if a closed circle of people continuously exchange bits of paper with each other, the total value of these bits of paper will not change much, if at all. If some members of that closed circle make extraordinary profits, these profits can only be made at the expense of other members of the same circle. Common sense suggests that this activity leaves the value of the traded assets little changed, and cannot, taken as a whole, make money. What, exactly, is wrong with this common sense perspective?
Why we trade
Perhaps the central insight of modern economics – the core of Adam Smith’s Wealth of Nations – is that trade can benefit both parties to an exchange. Specialisation allows parties with different skills, different resources, or simply experience acquired through narrow focus, to deal advantageously with each other. I keep a dairy herd, you have a coffee plantation. And together we can both enjoy milky coffee. The wealth of nations resulted from the rejection of the mercantilist view – recently revived by Donald Trump – that trade was a process in which one party took advantage of the other.
Modern financial economics treats risk as a commodity like milk or coffee. People have different preferences and capabilities in their approach to risk and their ability to manage it, just as they have different tastes in food, or farm different kinds of land, or hold different agricultural skills. Trade between them benefits both parties. In this way, markets in risk enable the inescapable risks of modern life to be handled more efficiently.
If this analogy between risk and other commodities were valid, the standard tools of economics could be applied to the trading of risk. This approach has been the basis of financial economics for half a century. The metaphor has many attractions for those who work in financial markets, implying that the claims of market efficiency that are made for trade in milk and coffee are equally applicable to trade in foreign currency and credit default swaps. The larger the volume of trade, the wider the scope of markets, the greater the benefits from free exchange in securities markets.
But you can have too much of a good thing. That trade can benefit both parties does not imply that all trade does: and if not all trade is a process in which one party tricks the other in to giving something away, some is indeed like that… But three centuries after the great French economist Colbert provided a definitive exposition of the mercantilist doctrine, a different strand of thought in modern economics revives the notion that trade is tricky by stressing ‘information asymmetry’ – people trade because they have different knowledge, or different perceptions of the same knowledge.[6] And this is often true in risk markets.
These two approaches to thinking about trade in risk have a long history. Michel Albert, a French economist turned insurance company boss, offered an entertaining account of the development of the global insurance market in the eighteenth century.[7] He explained how, in Edward Lloyd’s coffee house in London, leisured English gentlemen gathered to gamble on the fate of ships at sea. The value of these positions ebbed and flowed with the tides, their fortunes were buffeted by the weather. A thousand miles away, Swiss villagers came together to agree that if a cow died, they would take collective responsibility for replacing it. The English traded risks, the Swiss mutualised them. The Swiss practised gemeinschaft; the English, not knowing the meaning of gesellschaft, equated it with wagering. The elision would have profound consequences centuries later.
The rise of derivative trading
In 1997, a London barrister, Robin Potts QC was asked by ISDA, the International Swaps and Derivatives Association, to review the new market in credit default swaps. Were participants in this market the modern counterparts of the gentlemen of Lloyds – engaged in a wager? Or were the buyers and sellers of credit default swaps more akin to the Swiss villagers, sharing the risks of disease and disasters?
Mr Potts expressed Albert’s distinction in legal terms. He drew attention to the famous case of Carlill v. Carbolic Smoke Ball Company. In 1892 a race-going judge, Sir Henry Hawkins, defined the meaning of a wager in English law. ‘A contract by which two persons, professing to hold opposite views touching the issue of a future uncertain event, mutually agree that dependent on the determination of that event one shall win from the other’. Insurance is different. The essence of insurance is (in the less felicitous words Mr Potts drew from an even earlier judgement) ‘a contract to indemnify the insured in respect of some interest which he has against the perils which he contemplates he will be liable to’.[8]
These two strands in the historic development of risk markets – laying bets on the interpretation of incomplete information, and the socialisation of individual risks – are still at the centre of how risk and insurance markets operate today. And, astonishingly, London and Switzerland remain key centres of the global insurance market. The villagers have descended from alpine meadows to the impressively prosperous urban centres of Zürich and Münich. The iconic Lloyd’s building – designed by Richard Rogers and perhaps the most striking of all City of London office buildings – is barely a hundred yards from the place where Edward Lloyd’s customers first smelt the coffee. And Lloyd’s is still the principal global location for marine insurance.
In the twentieth century both Lloyds and the Swiss/German industry were primarily reinsurance markets. A policyholder’s insurer will normally handle the routine administration of premiums and claims, but large losses are ceded, at an appropriate price, to reinsurers with specialist risk evaluation skills. Even in the late twentieth century, the organisational form reflected the historic origins. Lloyds was supported by its ‘names’, (mostly) English individuals of means and social standing who hoped to derive a regular income from profitable underwriting, but put their personal wealth on the line to meet any losses. Munich Re and Swiss Re were financial behemoths, pooling risks globally and maintaining large capital reserves to meet future losses.
At Lloyd’s the coffee house tradition continued with business conducted in ‘the Room’. The Lutine Bell at its centre reflected the maritime history. (The bell had been salvaged from a bullion ship which had sunk in 1799 and was rung once to signify a wreck and twice to record a safe return). A broker placing a risk would attempt to find a lead underwriter willing to accept a substantial portions of it: if the lead underwriter was well regarded, others would follow or behalf of their own ‘names’. This mixture of cooperation and competition encouraged individual underwriters to develop specialist skills and to share the responsibility for detailed diligence of individual risks. This practice of syndication was equally common in the sharing of large loans among major banks. The exchange of risks benefited all parties to the trade, in a manner of which Adam Smith would have approved.
By the 1980s, however, ‘the Room’ was becoming a trading floor like those of other exchanges. If you can earn profit by selling reinsurance contracts, you can also earn profit by selling contracts for the reinsurance of reinsurance. You could even create contracts for the reinsurance of reinsurance of reinsurance. And so on. The outcome was a nexus of contracts known as the LMX spiral, so elaborate and complex that it was simply impossible to ascertain the underlying risks to which the holder was exposed. When the Piper Alpha oil rig went on fire in the North Sea in 1987, killing 200 people and triggering what was then the world’s largest marine insurance claim, underwriting names who had never heard of Piper Alpha discovered they had reinsured it over and over again. The total value of claims at Lloyds turned out to be ten times the value of the underlying loss.[9]
The trade in risk that had occurred was not the spreading of risk on more and broader shoulders. The operator of the rig, Occidental Petroleum, was better placed to assess, monitor and accept the risk than those people who ultimately had to pay for the costs of the accident. Some ‘names’ suffered losses on a scale which forced the sale of their homes and blighted their lives. There were reports of suicides. The trade in risk that had occurred was between people who had some understanding of what they were doing – and of the nature of the risk – and people who did not. It was the world of Colbert and Ruskin, not that of Friedman and Greenspan. Far from spreading risk and placing it in the hands of people well placed to manage it, the LMX spiral concentrated it in the hands of people who had no capacity to manage it at all.
A Lloyd’s underwriter denounced to me in extravagant language the ignorance and incompetence of the agents who had promoted these structures, and brought Lloyd’s close to collapse. Why, I asked, had he not blown the whistle on the individuals concerned? He looked at me with a pitying gaze and simply said: because they were willing to buy risks at a price at which I was delighted to sell them. And this would be a forerunner for the development of modern financial markets more generally.
Trouble in the Teton Mountains
In 2005 the Federal Reserve Bank of Kansas held a symposium at the agreeable Wyoming resort of Jackson Hole. The purpose was to honour Alan Greenspan, who would soon retire from his position as chairman of the Federal Reserve Board. Raghuram Rajan, then chief economist at the International Monetary Fund, queried the value of recent innovation in financial markets and warned of troubles ahead.[10]
Rajan’s paper was not well received. The principal discussant was the vice chairman of the Federal Reserve Board, Don Kohn. Kohn treated the speech as an attack on what he called ‘the Greenspan doctrine’, which proclaimed the virtues of the financial innovations which Rajan had queried. Kohn made a robust defence of these innovations. ‘By allowing institutions to diversify risk, to choose their risk profiles more precisely, and to improve the management of the risks they take on, they have made institutions more robust’. He went on to explain that ‘these developments have also made the financial system more resilient and flexible—better able to absorb shocks without increasing the effects of such shocks on the real economy’.[11]
This was indeed Chairman Greenspan’s view. Had he not explained that ‘these instruments enhance the ability to differentiate risk and allocate it to those investors most able and willing to take it’?[12] If Kohn was critical, he was at least polite: Larry Summers described Rajan’s views as ‘Luddite’, and likened his thinking to those who would substitute runners and horses for cars and aeroplanes. Complexity, Summers argued, was inseparable from progress.[13]
The keynote address at Jackson Hole was delivered by Robert Rubin, who had from 1995 to 1999 served as President Clinton’s Treasury Secretary. Rubin had previously been a Goldman Sachs executive and, after leaving government, would receive more than $100m for his work in non-executive roles at Citigroup between 2000 and 2009.[14] Rubin was a key figure in Summers’ transition from the denigration of ‘ketchup economics’ to enthusiastic supporter of financialisation, and had groomed the Harvard academic to be his successor. When Brooksley Born, chair of the Commodity Futures Trading Commission, had sought to extend the regulation of derivative markets, Rubin, Summers and Greenspan had led the opposition and supported legislation which excluded financial contracts from the remit of the agency.
The following year, another Federal Reserve Board Governor, Ben Bernanke, reiterated Kohn’s claim: ‘Banking organisations of all sizes have made substantial strides over the past two decades in their ability to measure and manage risks’. Such advances resulted, he said, in ‘greater resilience of the banking system’.[15] Bernanke was the Princeton professor and student of the Great Depression, who had earlier proclaimed ‘the Great Moderation’.
The chairman of the Federal Reserve Bank of New York similarly applauded the work of risk managers when he addressed their annual conference. Timothy Geithner told his audience: ‘Financial institutions are able to measure and manage risk much more effectively. Risks are spread more widely, across a more diverse group of financial intermediaries, within and across countries. These changes have contributed to a substantial improvement in the financial strength of the core financial intermediaries and in the overall flexibility and resilience of the financial system in the United States’.[16]
The breathtaking scale of these misapprehensions proved no obstacle to the subsequent advancement of those who held them. When the global financial crisis broke George W. Bush had already appointed Bernanke to succeed Greenspan as Chairman of the Federal Reserve Board. Kohn became his vice-chairman. President Obama formed his economic team by appointing Geithner as Treasury Secretary and Summers as Chairman of the National Economic Council. (Rajan left the International Monetary Fund at the end of 2006 and returned to India where in 2013 he became Governor of the Reserve Bank). But talking truth to power proved no more welcome in India than in Wyoming and Rajan’s contract was not renewed.
In the 1970s, David Henderson, an economist with extensive experience in government, had delivered for the BBC a series of lectures on ‘the unimportance of being right’: ‘one thing that you might think would count, but which in fact is given no attention whatever, (in securing promotion to positions of influence) is whether your advice has been any good’.[17] Henderson went on to observe that ‘if this is possibly true of the British public service, it is I think more obviously true of the economics profession, both in this country and elsewhere’ and quoted another commentator, Samuel Brittan, ‘it is much more important for a paper to be competent than for it to be right or enlightening’.[18] If there has been any change since Henderson wrote, in either government or the academy, it has been for the worse.
The problems of complexity
The complexity of modern finance has been designed, and has operated, principally to benefit financial intermediaries rather than users of financial services. The claims of Greenspan, Geithner and others that the innovative use of new instruments made the financial system more robust were false. Interdependencies between financial institutions have increased to a point at which the system as a whole displays fragility born of complexity.
The phrase ‘too big to fail’ came into wide use in the global financial crisis to describe the dilemma policymakers faced in resolving the affairs of systemically important financial institutions. The phrase provoked the justified rejoinder that ‘too big to fail is too big’. But ‘too big to fail’ misses the key point. Financialisation has led to increases in the size of financial institutions, but the central problem is not size but complexity. Size in banking can enhance stability, at least up to a point. Britain avoided significant bank failures in the twentieth century precisely because its banks were big, in contrast to the collapse of the fragmented US banking industry in 1933.[19] The failure of the UK banking sector in 2008 occurred, and was traumatic, not because the sector had become more concentrated, but because it had become more complex.
Lehman, the business whose failure turned the instability of 2008 into a crisis which jeopardised the functioning of the global financial system, was not, in any ordinary sense of the phrase, a business of economic importance. If it was a systemically important financial institution, it was not an important financial institution. The business provided no services to the real economy which were not available elsewhere, and few services to the real economy at all. The company was badly run, and operated primarily for the benefit of its own staff, especially its most senior executives. But Lehman was massively interconnected. At the time of its bankruptcy, the company had over 200 subsidiaries around the world, and approaching one million outstanding transactions, almost entirely with other financial institutions. The narrow consequence of this interdependence was that the winding down of the company’s convoluted structure will employ lawyers and accountants for a decade. The broader consequence was that financial institutions with exposure to Lehman were uncertain of the value of their claims.
Because so many institutions had dealings with Lehman, uncertainty was contagious. Even businesses with little direct engagement with the failed bank were uncertain about the value of their claims against institutions which did. The collapse of confidence spread throughout the financial system, with adverse effects on the non-financial economy which persist to the present day. Lehman was not too big to fail, but it was too complex to fail.
The historian Joseph Tainter has studied a wide variety of civilisation collapses – the many sophisticated societies which once thrived, but exist no longer.[20] The floodplains of the Tigris and Euphrates in Mesopotamia were the site of the first urban civilisation and the development of modern agriculture. Ancient Rome finally succumbed to the barbarians at the gate. The Mayan and Chacoan civilisations of North America were highly sophisticated societies which we know only as archaeological sites. The growth of complexity in social and economic interaction is the distinctive mark of civilisation, but complexity, and the inequality and specialisation that go with it, entail diminishing returns. Eventually the social and political costs of managing that complexity became overwhelming and promote internal decomposition.
Parallels between the progress of financialisation and the decline and fall of the Roman empire may seem at first sight far fetched. Yet there are obvious analogies. The Roman Empire disintegrated because of the ultimately counterproductive consequences of the growth of complexity, its inability to manage the scale of organisational problems raised, its increasing attention to ritual disconnected from an external world, and its incapacity to effect substantive self-criticism or self-repair. We can see all these issues in the modern financial system.
The lesson is – on the one hand – to eschew unnecessary complexity, and – on the other – to give close attention to the management of such complexity as is unavoidable. The unplanned evolution of the financial services network contrasts with the conscious design of other utility networks, such as electricity. The overriding need for system stability is embedded in the thinking of everyone engaged in electricity supply. And anyone who thinks electricity supply less complicated than the financial system knows little about the complexities of maintaining the stability of an electricity grid. But it has not been usual to think about the financial system in the systemic way which is natural to operators of other networks. And despite recent experience of the consequences of system failure, it is still not usual to think in this way.
The organisational sociologist Charles Perrow has studied the robustness and resilience of engineering systems in different contexts, such as nuclear power stations and marine accidents.[21] Robustness and resilience require that individual components of the system are designed to high standards. Demands for higher levels of capital and liquidity are intended to strengthen the component units of the financial system. But, the levels of capital and liquidity envisaged are inadequate – laughably inadequate – relative to the scale of resources required to protect financial institutions against panics such as the global financial crisis. More significantly, resilience of individual components is not always necessary, and never sufficient, to achieve system stability. Failures in complex systems are inevitable, and no one can ever be confident of anticipating the full variety of interactions which will be involved.
Engineers responsible for interactively complex systems have learnt that stability and resilience requires conscious and systematic simplification, modularity which enables failures to be contained, and redundancy which allows failed elements to be by-passed. None of these features – simplification, modularity, redundancy – were characteristic of the financial system as it had developed in 2008. On the contrary. Financialisation had greatly increased complexity, interaction and interdependence. Redundancy – as, for example, in holding capital above the regulatory minimum – was everywhere regarded as an indicator of inefficiency, not of strength.
In Perrow’s analysis, systems lack robustness if they are interactively complex – everything depends on everything else – and tightly coupled – the tolerance for error is low. The interactive complex and tight coupling of a nuclear power station is an inescapable consequence of prevailing technology. Paradoxically, attempts to increase resilience by incorporating many layers of safety provision may make the system less robust by increasing its complexity. An assembly line is complex but not interactively complex – it depends on a linear sequence of events in which each step logically follows the preceding one. Such a process may be tightly or loosely coupled. The moving belt of the traditional car plant’s assembly line demonstrates tight coupling, while the normally leisurely processing of a book from manuscript to publication is loosely coupled – no one is surprised at the author’s late delivery, nor is the production process upset.
Robust systems are typically linear. The delivery system that connects a toy manufactured in China to the mailman at my front door contains many links, but whenever a single link fails, and they often do, there is ample scope for recovery. Moreover, neither the failure nor the recovery have material consequences for other elements of other delivery chains.
The collapse of the new economy bubble in 2000 proved much less severe and enduring in its consequences than the global financial crisis of 2008. The scale of the imaginary wealth first created, then destroyed was not necessarily greater in that phase of equity market mispricing than in the later episode of credit market mispricing. But the rise and fall of technology stocks did not involve the complex interdependencies between financial institutions of the kind which marked the credit boom – there was little interactive complexity and coupling was fairly loose.
On the other hand, the Japanese stock market and property bubble – in which Japanese banks were heavily implicated – did lasting damage to the Japanese economy, because the rise and subsequent fall of asset prices had multiple consequences for other parts of the financial sector and for the balance sheets of industrial companies. As a result of complex interactions created by the growth of high frequency trading and the wide use of exchange traded funds (packages of existing securities which are themselves bought and sold like other securities), it is far from certain that an equity market meltdown today could be accommodated with the same equanimity as in 2000.
Increases in interactive complexity and tighter coupling were the very innovations which the participants at Jackson Hole celebrated. Firms themselves became interactively complex. Most of the conglomerate businesses which failed in 2008 were brought down by activities peripheral to their principal business. The failure of Lehman was not the cause of the global financial crisis – that was far more deep seated. But Lehman made no contribution to the real economy commensurate with the damage done by its failure. The public costs of its interactive complexity far exceeded any public benefits. Lehman’s reliance on overnight financing, whose disappearance provoked its abrupt collapse, epitomised tight coupling.
Many aspects of the modern financial system are designed to give an impression of overwhelming urgency – the endless ‘news’ feeds, the constantly changing screens of traders, the office lights blazing late into the night, the young analysts who find themselves required to work 30 hours at a stretch. But very little that happens in the finance sector has genuine need for this constant appearance of excitement and activity. Only its most boring part – the payments system – is an essential utility on whose continuous functioning the modern economy depends. No terrible consequence would follow if the stock market closed for a week (as it did in the wake of 9/11) – or longer; if a merger were delayed, or large investment project postponed for a few weeks; if an initial public offering happened next month rather than this. A millisecond improvement in data transmission between New York and Chicago has no significance whatever outside the absurd world of computers trading with each other.
The tight coupling is simply unnecessary, the perpetual flow of ‘information’ part of a game traders play that has no wider relevance, the excessive hours worked by many employees a tournament in which individuals compete to display their alpha qualities in return for large prizes. The traditional bank manager’s culture of long lunches and afternoons on the golf course may have yielded more useful information about business than the Bloomberg terminal.
Lehman – an ill-managed purveyor of unneeded products – represented exactly the kind of business which should fail in a well functioning market economy. The view that it was a mistake for the US government to permit Lehman to collapse is expressed, not by people who miss the services that Lehman provided, but by people who regret the consequences of its failure. The lesson is not that policymakers should try to prevent such failures but that public processes should ensure that similar failures are more easily contained.
[1] Bank of England (Bankstats tables B2.1 and B2.1.1 (Jan. 2018))
[2] C. Steiner. “Wall Street’s Speed War”, Forbes (9 Sept 2010), accessed 2 Mar 2018
[3] “Triennial Central Bank Survey: Foreign Exchange Turnover in April 2016”, Bank for International Settlements, (Sept 2016) p. 3; “World Trade Statistical Review 2017”, World Trade Organisation (2017) pp. 14-5
[4] “Payment Statistics January 2018”, PaymentsUK (19 Feb 2018) accessed 5 Mar 2018
[5] “Statistical Release: OTC Derivatives Statistics at End-June 2017”, Bank for International Settlements (Nov 2017), p. 2; Credit Suisse Research Institute, “Global Wealth Report 2017: Where Are We Ten Years After the Financial Crisis?”, Credit Suisse, (14 Nov 2017)
[6] See for example G. Akerlof, “The Market For ‘Lemons’: Quality Uncertainty and the Market Mechanism”, Quarterly Journal of Economics, Vol. 84, No. 3 (1970); M. Spence, “Job Market Signalling”, Quarterly Journal of Economics, Vol. 87, No. 3 (1973)
[7] M. Albert, Capitalism Against Capitalism (1993)
[8] R. Potts, “Credit Derivatives: Opinion”, ISDA (1997) p. 4
[9] Deeny v. Gooda Walker, [1994] EWHC 300 (Comm)
[10] R. Rajan, “The Greenspan Era: Lessons for the Future”, Speech to Jackson Hole Conference (2005)
[11] D. L. Kohn, “Commentary: Has Financial Development Made the World Riskier?”, Proceedings, Federal Reserve Bank of Kansas City (2005) pp. 371-9
[12] A. Greenspan, “Financial Derivatives”, speech to the Futures Industry Association, Boca Raton, FL, (19 Mar 1999)
[13] L. H. Summers, “General Discussion: Has Financial Development Made the World Riskier?”, Proceedings, Federal Reserve Bank of Kansas City (2005) pp. 387-97
[14] E. Dash and L. Story, “Rubin Leaving Citigroup; Smith Barney for Sale”, The New York Times (9 Jan 2009)
[15] B. S. Bernanke, “Modern Risk Management and Banking Supervision”, remarks at the Stonier Graduate School of Banking, Washington, DC (12 Jun 2006)
[16] T. F. Geithner, “Risk Management Challenges in the US Financial System”, remarks at the Global Association of Risk Professionals 7th Annual Risk Management Convention & Exhibition, New York City (28 Feb 2006)
[17] P. D. Henderson, “Two British Errors: Their Probable Size and Some Possible Lessons”, Oxford Economic Papers, Vol. 29, No. 2 (1977) pp. 159-205
[18] S. Brittan, Is There an Economic Consensus? (1973)
[19] M. Carlson and K. J. Mitchener, “Branch Banking as a Device for Discipline: Competition and Bank Survivorship During the Great Depression”, Journal of Political Economy, Vol. 117, No. 2 (2009)
[20] J. Tainter, The Collapse of Complex Societies (1990)
[21] C. Perrow, Complex Organisations: A Critical Essay (1972); Normal Accidents: Living With High Risk Technologies (1984)