Engelen et al. discuss the banks and their ambition for unsustainable returns.
There's no reason to apologise for the actual profits (in British banks). A third of the money goes in tax, which means huge sums are going into schools and hospitals. . . . Another third goes to shareholders – and I'm not talking about fat cats driving Ferraris; I mean pension funds looking after the interests of all of us. (Angela Knight, incoming chief executive of the British Bankers' Association, in an interview with The Independent, Prosser 2007)
Banks have an ethical obligation to strive for sustainable value creation and stability, similar to the obligations we have for the environment. This entails the need, both in credit provision to customers and in internal affairs, to assess everything in terms of sustainability. That (implies) . . . for instance, that never again would a CEO of a German bank present a short term profit rate of 25 per cent as a realizable corporate goal, thereby raising expectations to insiders and outsiders which can never be fulfilled...(This is) an instance of idolatry. It reminds me of the dance around the golden calf. Luther said: 'to that which your heart goes out, is your God'. In our society money has become our God. (Bishop Wolfgang Huber, Berliner Zeitung, 24 December 2008, our translation)
These two quotations come from different moral worlds. Angela Knight represents the smooth processes of elite circulation and expedient public defences of the indefensible. Knight was a talented junior Conservative Treasury minister whose rise was blocked by electoral defeat in 1997 and afterwards built her second career as lobbyist and spokesperson for finance in Britain. In early 2007, as incoming chief executive of the British Bankers' Association, she was defending bank profits on the grounds that they boosted tax revenues and returns for pensioners. This trope about the poor and otherwise indigent rentier dates back to the mid-nineteenth century when (before the democratization of finance and increased life expectancy) the shareholders of railway companies were always widows and orphans. Bishop Huber, in late 2007, represents the awkward principle of individual Protestant conscience. In the Bishop's moral universe, banking represents not greed but idolatry or the worship of false gods, which is of course the first prohibition in all Protestant or reformed versions of the Ten Commandments. Just as in the case of Queen Elizabeth, the Bishop as an outsider has perhaps a much sharper perception of the issues around finance and crisis than most insiders.
If Queen Elizabeth's question was, why did nobody see it coming?, Bishop Huber's question is how did banking CEOs, like Josef Ackerman of Deutsche Bank, set unsustainable 25 per cent return on equity targets as 'a realizable corporate goal'. It is a great pity that, under current norms of elite circulation, a dissident like Bishop Huber will not be recruited into a non-executive board role where he could directly put his question to a bank CEO.
The aim of this chapter is to answer a more elaborate version of Bishop Huber's question. Specifically, we are concerned with two related questions about the pre-2007 period: why were high profit targets set in banking and how could returns of 15–25 per cent on equity be (unsustainably) delivered before the crisis? The high returns before the crisis are immediately puzzling because investment banking had always been a small-scale activity focused around new issues of securities and merger advice, which could not sustain giant companies; at the same time, retail banking had traditionally been a low return utility activity which, after 1980s deregulation, was increasingly exposed to increased competition. Sections 4.2 and 4.3 of this chapter answer our questions and explain the apparent paradox. Section 4.2 analyses the pressures on all public companies from the early 1990s under a stock market regime of shareholder value where sectoral and company excuses for low returns were not acceptable to fund managers or analysts. Section 4.3 analyses how banking business models changed with the rise of mass marketing in retail and proprietary trading in investment banking, so that, when leverage was added, these volume-driven business models could (unsustainably) produce high returns for shareholders.
But this chapter also tackles a series of equally interesting questions about the irrelevance of academics and regulators who either failed to register the transformational changes in banking business models or did not understand that the profits of change were unstable and unsustainable. Thus, the first section of the chapter begins by analysing the academic conceptualization of the role of banks in contemporary economies. In doing so, it considers the mainstream economics that dominates research-based departments and institutes, with commitment to markets and rationality assumptions, elaborated through algebraic formalization and published in a small number of highly ranked journals. We then turn to consider the voluminous, institutionalist varieties of capitalism literature. This filiates from Michel Albert and from Peter Hall and David Soskice, and represents an important part of the post-disciplinary world created by the splintering of previously unitary disciplinary formations like political science and sociology. The fourth and final section of this chapter considers the post-Keynesians and Minskians who can, with some justification, claim to have foreseen a crisis, because they had more generally understood the instability of finance-led capitalism. While we should be grateful for such prescience, it is also important to register the limits of Minsky's kind of technical analysis, which does not engage the political domain that is the theme of the second half of this book. Put simply, getting it (partly) right before the crisis is not the same as having a solution after the crisis.
Academic understanding: banks as intermediaries?
This section introduces and discusses academic understanding of banks in the mainstream economics literature, where banks are credited with functions like intermediation and risk transfer. This is a good example of how an overdeveloped interest in identifying the reasons for what exists has distracted mainstream economists from the more interesting task of evaluating stories about what exists. We then show that much heterodox social science, like the varieties of capitalism literature, does little better because the heterodox buy into the mainstream account of reasons so that intermediation between firms and households becomes taken for granted.
In mainstream economics and finance, banks have always posed a theoretical problem because, in informationally efficient markets, intermediaries like banks should not exist. The early literature on banking tried to explain this anomaly: from Gurley and Shaw (1955) onwards, the standard answer has been that banks exist as intermediaries because financial markets are informationally imperfect, with the consequent existence of transaction costs (Klein 1971; Benston and Smith 1976) and informational asymmetries (Leland and Pyle 1977). On this account, banks pool retail household deposits which are transformed into wholesale diversified claims on productive firms. Evaluation of credit risk, which individual households are not capable of performing themselves due to the existence of informational asymmetries, then becomes a very specific banking function in the economy that justifies the existence of banks. Banks achieve their profit objectives by intermediating between the economic units that have surplus funds (households) and those that require funds for investments (firms). This view of banks in mainstream economics continued for quite some time. As late as 1993, Bhattacharya and Thakor wrote that: 'to summarize, intermediation is a response to the inability of market-mediated mechanisms to efficiently resolve informational problems' (1993: 14).
From the early 1980s onwards, the changing framework conditions described in Chapter 2 have led an increasing number of mainstream economists to question the relevance of transaction costs and informational asymmetry and therefore of financial intermediation theory. This led to confused debate about whether disintermediation threatened credit-creating banks in a new historical period and how banks could still exist if their theoretical justification had more or less disappeared (Boyd and Gertler 1995; Miller 1998). In effect, as banks were not withering away, a new theory of banks was needed that would (like the old one) try to specify a new universal logic to explain the role of banks in the economy. In this context, Merton and Bodie (1995) shifted the debate to a functional approach to financial systems as opposed to an institutional approach, by emphasizing in their modelling what banks do rather than why they exist. Allen and Santomero (1997) then announced their new theory of financial intermediation as risk transfer within wholesale markets: instead of traditional theory, 'we offer in its place a view of intermediaries that centers on two different roles that these firms currently play. They are facilitators of risk transfer and deal with the increasingly complex maze of financial instruments and markets' (Allen and Santomero 1997: 1462).
The new emphasis on risk management displaced the old question in the financial intermediation literature about 'why banks exist?' but still retained the question about 'would banks survive?' in a world of disintermediation and reintermediation. Disintermediation has since become an important research topic in mainstream economics: for example, Schmidt et al. (1999) and Allen and Santomero (2001) investigated whether bank-based economies of continental Europe would converge to market-based economies. Schmidt et al. (1999) concluded that in the United Kingdom, Germany, and France, disintermediation had not been significant enough to change the role of banks but noted a relatively increasing role of markets in France. However, Allen and Santomero (2001) announced the decline of traditional banking business due to competition in intermediation and the associated rise of fee-producing activities by banks in the United States and the United Kingdom which, as we shall see, is a very significant development.
The world financial system has changed significantly in recent decades. In the US, banks and many other types of intermediaries have moved away from their traditional role of taking deposits and making loans. Although their share of intermediated funds has fallen they have not shrunk relative to GDP, and they remain an important part of the financial system. They have achieved this by moving away from simple balance sheet intermediation toward fee-producing activities. (Allen and Santomero 2001: 290)
The corporate form of banks also became an issue with a wave of banking mergers, in the 1990s, initially in the United States after deregulation, and then in the EU and Japan. Inter alia, these increased bank size and created large financial conglomerates which were actively encouraged by many national authorities. The EU Second Banking Directive of 1989 encouraged universal banking in the EU, as did the Treasury Report of 1991 and the Financial Services Modernization Act of 1999 in the United States. Meanwhile, in Japan, the Financial Services Agency's Financial Reform Programme had similar aims.
Explaining and analysing the existence of these new kinds of banking giants became another important research topic in mainstream literature before the crisis. Mainstream economics, of course, has struggled to find a rationale for corporate mergers when the empirics show that they do not generally create shareholder value for the acquiring firm. Mainstream finance has also struggled to find a rationale within the finance sector because studies of bank and financial services mergers report mixed results and do not consistently find scope or size economies. For example, econometric studies of the effects of bank mergers in different jurisdictions and different periods provide no convincing case for the existence of economies of scale and scope due to size and diversity (Rhoades 1993; Benston 1994; Milbourn et al. 1999; Vennet 2002; Casu and Girardone 2004; Campa and Hernando 2006; Stiroh and Rumble 2006).
So, after two decades of strenuous effort, mainstream economics updated its story on the rationale for banks by shifting towards risk transfer theories but, in doing so, it failed to explain why banks have become bigger and bigger multi-activity conglomerates, because size increases without any obvious economic rationale. Thus, as the crisis was breaking, mainstream economists were agnostic about many developments, arguing that more research was required. In 2007, after the crisis had started, Feldman and Lueck of the Federal Reserve Bank of Minneapolis returned to issues raised by Boyd and Gertler (1995) at the annual conference of the Federal Reserve Bank of Chicago in 1994. Nearly fifteen years later, Feldman and Lueck still could not decide the question of whether banks as intermediaries were in decline:
The new technology of banking continues to transform how banks operate. It also makes it possible for competitors, both new and old, to take on banks. . . . We simply do not have the data needed to adjust the balance sheets of all financial firms to properly determine their relative market share. (Feldman and Lueck 2007: 48)
This observation of mainstream indecision raises the question about whether heterodox approaches have done better in understanding developments in banking. There are many kinds of heterodox economics which derive their identity partly from being unlike mainstream economics. Below, we look at the varieties of capitalism literature which, in the decade after the mid-1990s, made strong inroads in the wider social sciences as a frame for looking at contemporary capitalism. This makes an interesting case because the varieties of capitalism preoccupation with cross-sectional national differences, understood in stereotyped bank versus market terms, blocked any recognition of the changes in banking which Allen and Santomero registered as risk transfer. The heterodox concept of banking was in many ways more conservative than the mainstream.
The master problem for the varieties of capitalism literature was not why do banks exist or survive but how does the financial system influence national economic performance. The varieties of capitalism literature is methodologically nationalist, and uses banking as an important variable in classifying economies according to the relative importance of banks and stock markets for productive firms. Finance is then a system embedded in an overall institutional structure of national complementarities. In Hall and Soskice's version (2001) of varieties of capitalism, the banks are 'patient' capital in a coordinated market economy (CME) and the stock market is footloose capital pursuing higher returns in a liberal market economy (LME).
We would argue that British firms must sustain their profitability because the structure of financial markets in a liberal market economy links the firm's access to capital and ability to resist takeover to its current profitability; and they can sustain the loss of market share because fluid labor markets allow them to lay off workers readily. By contrast, German firms can sustain a decline in returns because the financial system of a coordinated market economy provides firms with access to capital independent of current profitability; and they attempt to retain market share because the labor institutions in such an economy militate in favor of longterm employment strategies and render layoffs difficult. (Hall and Soskice 2001: 16)
Thus, the logic of financial intermediation is held to be variable by nationality in cross section and, on this basis, the literature turns to taxonomic issues. The problems of classification are then about identifying the metrics and criteria which measure the relative importance of stock market and bank, and also about resolving contradictory evidence in ambiguous national cases so that these can be related to ideal-type constructs such as CME.
Thus, Hall and Soskice and others have introduced empirics on the ratio of stock market capitalization to GDP, household portfolios, bank loans as source of corporate finance, and the role of banks in the governance system. Using these metrics to score different cases resulted in endless debates on how various measures give divergent results, while many measures are contestable even in seemingly clear-cut cases such as Germany. For example, Hackethal and Schmidt (2003) and Schmidt and Tyrell (2003) emphasize the importance of gross lending by banks, which accounted for 80 per cent of non-financial firms' finance between 1970 and 1996, but earlier studies by Mayer (1998), Edwards and Fisher (1994), and Corbett and Jenkinson (1996) used net lending as the measure and on this basis concluded that banks in Germany were not major providers of finance. There is then the further complication that change is measurable on such empirical criteria, but the nature and implications of any such change are often ambiguous. By the mid-2000s, everyone agrees 'the large for-profit banks in Germany have been systematically exiting their close relationships with industrial companies' (Vitols 2005: 364). However, researchers differ about what this change means: Vitols (2005: 12) anticipates an outcome which stops 'far short of convergence to the Anglo Saxon system', whereas Schmidt and Tyrell anticipate a long-run German 'transition to a capital market-based system' (2003: 50).
Finally, there is the problem of how to classify ambiguous and intermediate national cases. Thus, Schmidt and Tyrell credit Germany with a bank-based financial system and Britain with a capital market-based system before concluding the French system 'has not been stable' (2003: 47). Amable (2003) provides the most sophisticated attempt to classify empirical cases. Using cluster methodology and multiple criteria, he finds that Germany belongs to a group of countries where an ideal bank-based system exists. This group includes Japan, France, Austria, Italy, Portugal, and Spain, and it is characterized by a high credit to GDP ratio, an important share of insurance companies among institutional investors, little M&A activity, weak development of accounting standards, and a lagging venture capital sector. But we would also note that, as the empirics become more complicated, Amable finds five distinct variant types of capitalism where Hall and Soskice hypothesized just two.
Despite (or because of) this increasing sophistication, the varieties of capitalism literature has lost itself in typological discussions and failed to register the changing activities of banks in both bank and capital market-based economies. Because they were exploring their theory built around the bank versus market opposition, these authors failed to register the adjustment of reality whereby Landesbanks in Germany could divert from their national productive role of supporting SMEs to an international financial role of the greater fool, as they became depositories for toxic assets created by the US banks. In our view, this development was partly driven by an attempt to compete with shareholder value-driven banks and their highly profitable volume-driven business models. These more general developments are considered in the next two sections. Section 4.2 explains how the rise of shareholder value as the metric of corporate performance played out in banking, where it was possible to generate high profits before the crisis. Section 4.3 then turns to analyse the new business models in retail and wholesale banking which generated profits and remained undisclosed in the narratives about intermediation constructed in mainstream economics and varieties of capitalism literatures.
Shareholder value-driven banking
Shareholder value was the new corporate objective for quoted giant firms in the 1990s. The financial results of shareholder value by the 2000s were generally disappointing but not in the case of banks. This section examines the exceptional sector and explains how leverage was used to produce shareholder value in banks and other financial companies. It also observes that (unsustainable) profits sedated the critical faculties of non-executive directors, analysts, and fund managers who might otherwise have seen a crisis coming.
Froud et al. (2006) provides an overview of the impact of institutional investor demands for shareholder value on giant US and UK firms. The argument presented on outcomes has two elements: first, financial disappointment was inevitable because capital market pressures cannot shift product market limits in mature, competitive industries like car assembly where the delivery of value is almost impossible except through brief rallies which are cyclical and understood as such by everybody; and, second, the unintended consequence was a narrative turn as underperforming corporate CEOs rehearsed their excuses and claimed ever higher pay. That book did not focus on banks because of the difficulties of using the financial accounts of banks in making sense of their business models and hence the sectoral differences in value production were not highlighted.
While the bubble lasted, banking delivered a spectacular (albeit unsustainable) growth of profits in the 2000s which consolidated a long-run rise in the proportion of giant firm profits generated from finance in the United States, the United Kingdom, and other jurisdictions. Figure 4.1 presents data for the period since 1979 on the share of S&P 500 profits (income) generated by the finance, insurance, and real estate (FIRE) sector as a whole. The share of FIRE sector giant firms in total S&P 500 profits increased from 6.2 per cent in 1979 to 13.6 per cent in 1991 and then reached 39.1 per cent in 2002. Figure 4.2 shows the profits from deposit-taking (retail) banks over the decade from 1997 to 2007: retail banks (including commercial banks with investment banking activities) accounted for an average 11.3 per cent of total profit (income). At their peak in 2002 and 2003, retail banks accounted for more than 20 per cent of total S&P profits. Interestingly, the share of banks in S&P profits did not increase every year because oil and mining profits were high during the commodity boom that ran in parallel to the finance bubble.
As Figure 4.1 shows, FIRE's share in giant firm profits is substantially larger than the FIRE sector's share of US employment, which was more or less flat: in 2002, FIRE firms accounted for 39.1 per cent of S&P profit, and the FIRE sector employed 10.4 per cent of the workforce. This was hardly a miracle of efficiency because the profit was achieved by tying up a huge share of giant firm assets in finance. As Figure 4.1 suggests, the FIRE sector's share of giant firm assets was usually twice as high as its share of profits; moreover, its share of giant firm assets increased steadily in the 2000s from 58.3 per cent in 2000 to 69.4 per cent in 2007.
If the United States invented a new kind of unproductive (but hardly weightless) economy, so had the other high-income countries with the exception of Germany and Japan. The importance of finance plus commodities in generating giant firm profit in the bubble years emerges very clearly from Figure 4.3, which presents the trends in the FTSE 100 or the British giant firm sector. As Figure 4.3 shows, in the five years before 2007, the performance of the FTSE 100 was completely dominated by the commodities boom and the finance bubble. In those years, companies from oil and mining and from FIRE together accounted for more than 70 per cent of total FTSE 100 profits. The FIRE category is dominated by finance, which on average accounts for more than 30 per cent of all FTSE 100 profits over the bubble years when finance's share of UK employment was flat and its share of output was no more than 8 per cent. The miracle of the City of London and Wall Street was that such a small sector by traditional output and employment standards could generate such large profits, and in rigorous value terms this was more a trick than a miracle if two-thirds or more of giant firm assets were tied up in producing finance sector profits.
But stock market analysts were not very analytic about such issues. Instead, they praised profits at banks like Northern Rock (without inquiring how they were made) and endorsed moves like RBS' serial takeovers, including the overpriced acquisition of ABN Amro (without suspecting it would end in tears). Thus, according to ING in September 2006, 'Northern Rock has managed to transform itself into one of the UK's top residential secured lending powerhouses....We expect asset and profit growth to continue to outpace the market, and Northern Rock is the envy of its peers . . . ' (Sarangi 2006: 2). Analysts were apparently relaxed about RBS' valuation of ABN Amro in the bidding war with Barclays: analysts at Dresdner Kleinwort described the consortium approach by RBS, Santander of Spain, and Fortis of Belgium as 'close to perfect. . . . We believe they could offer up to €38 per share' (Treanor 2007). As Treanor (2007) also reports, Keefe Bruyette & Woods analysts estimated that the consortium could offer €39.5 per share, well above the €34 that had been suggested as the price being considered by Barclays. As long as shareholder value-driven banks were delivering profits, analysts sanctioned whatever they wanted to do.
Corporate governance was no more of a control via outside pressure from institutional investors or internal criticism by board directors. Proceduralized corporate governance, as invented by the Cadbury Report in the United Kingdom, has been greatly strengthened after successive crises (Ertürk et al. 2008). But this kind of corporate governance is more of an accelerator of initiatives in the absence of profit than a brake on irresponsibility when profit is being made. In general, a dearth of profits strengthens the scepticism of non-executive directors (NEDs) and the activism of outside shareholders, while an abundance of profits sedates the critical faculties of outsiders and becomes a matter of boardroom celebration as managers ride high on the tide of rising stock markets (Ertürk et al. 2008). Hence, there are few records of institutional shareholders in banks who asked awkward questions before 2007 about management rewards, risk management, and sustainability of growth in profits. Under the proceduralized corporate governance regime, these issues were directly the responsibility of NEDs. But, just as in the earlier case of Enron, the NEDs on the boards of US investment banks like Lehman or converted former UK building societies like Northern Rock did not question business models which appeared to be working. For example, the Lehman examiner Valukas' findings were:
Lehman's Board fully embraced Lehman's growth strategy. In a January 2007 Board meeting, the directors were informed of the large increase in the risk appetite limit for fiscal 2007, and of the firm's intention to expand its footprint in principal investments, and they agreed with Lehman's senior officers that Lehman needed to take more risk in order to compete. All of the directors told the Examiner that they agreed with Lehman's growth strategy at the time it was undertaken. (Valukas 2010: 76)
The simplest question was how so many banks of all kinds (in retail and wholesale across many jurisdictions) were able to turn in returns on equity of 15 per cent or more and thus become the sweethearts of the stock market. The short answer is through leverage or borrowing to expand capital employed in a favourable conjuncture: that is, the banks borrowed cheaply to expand their asset base from which they could make profits for shareholders whose equity base was much smaller. Hence, the banking sector's insatiable appetite for assets which we observed in the case of the US S&P 500. The arithmetic of leverage was such that a bank that had a meagre and declining return on assets (ROA) (or capital employed) could earn a handsome return on equity (ROE). Figures 4.4a and 4.4b present the relevant data for a group of British banks: these graphs show that the pattern was one of sustained, high ROE, combined with wafer-thin ROA ratios which declined in three of the four cases after 1998 and in all four cases after 2006. As Haldane (2009b) has noted, this in turn implies that banks' impressive performances for shareholders before the crisis were not the result of strong economies of scale or scope or high value-added innovation; instead, the driver was financial engineering in the form of higher leverage, new sources of funding, and an increasing velocity of transactions.
During the golden era, competition simultaneously drove down returns on assets and drove up target returns on equity. Caught in this cross-fire, higher leverage became banks' only means of keeping up with the Jones's. Management resorted to the roulette wheel . . . leverage increased across the financial system as a whole. Having bet the bank on black, many financial firms ended up in the red. (Haldane 2009b: 3)
A broader, sectoral view of banking in the United States, United Kingdom, and Mainland Europe after the mid-1990s directly shows the importance of leverage in all three areas and indirectly suggests that laggard national sectors and firms were under pressure to deliver at least 15 per cent ROE. Such stock market competition amongst the shareholder value-oriented banks in effect set performance targets for mutuals like building societies in the United Kingdom and Landesbanks in Germany. Haldane (2009b) analysed seventy global banks in 2007 and confirmed such ROE-driven behaviour amongst global banks:
First, the downward slope is consistent with global banks targeting a ROE, perhaps benchmarked by peers' performance. The Bank [of England]'s market intelligence in the run-up to crisis suggested that such 'keeping up with the Jones's' was an important cultural influence on banks' decision-making. Second, . . . banks kept up in this competitive race by gearing-up. Banks unable to deliver sufficiently high returns on assets to meet their ROE targets resorted instead to leveraging their balance sheets. (Haldane 2009b: 3)
Figures 4.5a and 4.5b present data on the profitability of large banks with assets of more than $50 billion in the United Kingdom and more than $100 billion in the United States and the Eurozone in 2010. The contrast between slim returns on assets and handsome returns on equity is very clear: the United States, the United Kingdom, and Eurozone banks all had ROA of less than 1.5 per cent but all three had converged on 15 per cent plus ROE by the mid-2000s. These graphs are also suggestive of pressure to conform to the norms of shareholder value. ROE is consistently higher in the United States, where the banking sector's rate of return never falls below 15 per cent, while UK banks have a different pattern of cyclicality but are never far behind. On the other hand, Eurozone banks start from lower levels of profitability, which in the mid-1990s is running at no more than 7–8 per cent ROE, but then rises and after 2003 converges on the 15 per cent achieved elsewhere.
The differences between ROA and ROE underline the attractiveness of leverage, allowing banks to enhance performance for shareholders with equity holdings (and for senior investment bankers whose bonuses were turnover related under the business model discussed in the next section). In an era of low interest rates and plentiful capital, every profitable investment made by a bank with a leverage ratio of twenty times equity capital, generates simply twenty times as much revenues, profits, and fees, as was the case in the overnight interbank money market before the crisis. But, the multiplicand is not guaranteed and it can all go wrong if asset prices start falling and markets close, and if the central bank does not then step in to provide cheap funds. Without such central bank support, leverage quickly turns into a crisis of insolvency and illiquidity because asset values have to be written down and losses recognized, as assets cannot easily be sold and cheap short-term funding is no longer available. Whatever happens in high street retail banking, the inevitable result in the wholesale markets is distressed parties, extreme risk aversion by sound counterparties, and a run on any vulnerable players within the leverage-dependent 'shadow banking system' (Gorton and Metrick 2010). Before 2007, such dysfunction was all in the future and the main symptom of the mid-2000s was a spectacular ballooning of bank balance sheets as the asset-hungry banks defended or increased their profits for shareholders.
As Figure 4.6 shows, bank balance sheets ballooned in all major jurisdictions, including those like France which did not include a major financial centre like London or New York. All this should have been a public cause for concern because the assets of the banking system were public liabilities if things went wrong. National governments had one way or another guaranteed retail deposits in most jurisdictions and, in any case, national governments would not easily allow large or interconnected banks to fail. A highly leveraged bank system could and did then have a frightening size in terms of total assets, which could be many times that of national GDP as illustrated in Figure 4.7: in the Dutch and the UK cases, the banking system was by this measure five times as large as GDP by 2008. And, as Turner has noted, this ballooning of bank assets in relation to GDP ' . . . was dominated not by the banks' relationship with UK households and companies, but by a complex mesh of intrafinancial system claims and obligations', fuelled by wholesale funds not by customer deposits (Turner 2010b: 17). Leverage was, therefore, the driving force behind the privatization of gains and the socialization of losses, multiplying profits and bonuses in good times and multiplying losses and bailouts in bad times.
Of course, hindsight is a wonderful thing and as soon as this crisis got under way, irresponsible individuals and risky corporate behaviour were identified and scapegoated. Northern Rock had relied on volatile short-term wholesale markets to finance the phenomenal growth of its mortgage book, rather than on reliable long-term retail deposits: in 2006, 76 per cent of Northern Rock's funding was from wholesale markets. In announcing its 2006 preliminary results, the bank stated that:
Northern Rock has four distinct funding arms enabling it to attract funds from a wide range of customers and counterparties on a global basis. In recognition of our broad and innovative access to a cost effective and diverse capital markets investor base, Northern Rock was awarded the prestigious International Financing Review's 2006 Financial Institution Group Borrower of the Year award. (Northern Rock Plc 2006: 9)
In another case, that of RBS, it was clear ex post that Fred Goodwin's strategic mistake was to overpay for ABN Amro, which was not a profit source but a burdensome set of liabilities given that ABN Amro had invested in toxic assets and contributed £20 billion to the £28 billion loss at RBS in 2008 – the biggest ever loss in British corporate history. British Prime Minister Gordon Brown was 'angry' at the 'irresponsible' actions of Fred Goodwin, who was knighted on Brown's advice in 2004 for services to banking (Bawden and Waller 2009). The representatives of proceduralized governance were more penitent and confessional, as they had been after previous crises. The International Corporate Governance Network (ICGN) represents institutional shareholders (private and public pension funds) responsible for managing global assets worth $15 trillion, accepted that poor governance was to blame:
It is true that shareholders sometimes encouraged companies, including investment banks, to ramp up short-term returns through leverage. . . . Corporate governance failings were not the only cause but they were significant, above all because boards failed to understand and manage risk and tolerated perverse incentives. (ICGN 2008: 1)
It is not surprising that bankers' bonuses quickly became an issue because they were an intelligible symbol of excess for a mass audience that did not understand leverage and such like. The G-20 demanded the implementation of new standards on bonuses acting on Financial Services Board's (an international coordinating body of central banks and financial authorities) advice. The Institute of International Finance, a global association of financial institutions, nodded in agreement (IIF 2009). Everything had changed and much was now being criticized and called into question. But, in all these debates, shareholder value-driven business models of banks were not questioned. Quite the opposite: governments that have been forced into bank nationalization, plan exit strategies of trade sale or flotation which will realize gains for the taxpayer because the banks have been turned around and are, once again, generating shareholder value. Generally, and for publicly quoted banks within the private sector, the aim is to ensure shareholder discipline through more effective governance of banks, which relates risk and reward but retains the shareholder value objective, while new entry strategies of increasing banking competition will inevitably attract shareholder value-driven firms, like supermarket chains, from other sectors. The continued dominance of shareholder value principles is above all demonstrated by the mandate for UK Financial Investments Limited (UKFI), the state holding company for nationalized banks. After the enforced nationalization of UK banks, the government could not think of what to do except to run them for shareholder value before selling them off, and nationalization was reinvented as a private equity style turn around (as discussed in Chapter 6). The chief executive of RBS will earn an incentive payment of £6.9 million if he doubles the share price (Jenkins 2009), that is acceptable to UKFI because the 'overarching objective [is] protecting and creating value for the taxpayer as shareholder' (UKFI 2009: 13). However, it is not clear whether the banks have been nationalized or the Treasury has been privatized as a new kind of investment fund.
Moreover, media reports and reconstructions of negotiations so far on the modification of the banking accord, Basel III, indicate that shareholder value considerations, and the protection of profits on equity, has been a major objective of regulators who are concerned not to undo the gains of leverage by raising capital reserve requirements so that profitability is spoilt. The capital reserve requirements that were published in August 2010 were much less stringent than the initial proposals that dated from December 2009. The size and definition of tier one capital in Basel III has been recalibrated in such a way that, according to Nomura, European banks will only need €200 billion in extra capital, a third less than under the December proposals, and US banks only $115 billion, half of what was in the December proposals. The commentariat has criticized this as caving in to industry lobbying, but insiders claim a higher rationale because 'regulators worldwide became alarmed that the combined effect of the (original Basel III) December proposals would hobble banks' ability to lend and shrink their profitability to potentially unsustainable levels' (Masters and Murphy 2010: 5). This excuse is interesting, both because regulators accept the banks' threat to withdraw lending without inquiring into what they are now lending on and because regulators do not question the appropriateness of shareholder value in banking, although this was what got us into this mess. The next section focuses on changes in the business models of retail and investment banks to explore how a shareholder value orientation encouraged the activities that undermined the banking sector.
Business model changes
If the crisis was about what happens when leverage is added to banks under pressure for shareholder value, the crisis is also tied up with changes in bank business models, or the ways in which they recover costs and generated profits. The crisis confusingly brought down banks with many different kinds of business models, including Wall Street investment banks like Bear Stearns and Lehman Brothers (which were leveraged wholesale traders), aggressive new-style UK retailers (like Northern Rock and Bradford and Bingley) which funded risky mortgage lending by wholesale borrowing, and the old-style reckless (like the Irish banks or HBOS) which lent unwisely on commercial property. The aim of this section is to set all this detail into perspective by presenting an overview of changes in wholesale and retail bank business models. In many advanced capitalist countries, the deregulation of finance was associated with a shift from old business models of intermediation to new business models of retail mass marketing and wholesale trading which made banks into transaction-generating machines.
Insofar as banks are intermediaries, their major source of income will be interest earned on the spread between interest paid to depositors and interest charged to borrowers, and a shift from intermediation can be crudely measured by a decline in interest income (or a rise in non-interest income) as a proportion of bank profits. Table 4.1 presents the relevant empirics for banks in six countries between 1984 and 2007. While banking systems still betray historically rooted institutional differences, in each of these countries, interest income as a source of overall banking revenue have declined vis-a-vis non-interest income. This last category combines retail fees and commissions earned by selling financial products like mortgages or pension plans, and from wholesale earnings either from various kinds of trading or from charges for services rendered at the nodes of the wholesale latticework.
As Table 4.1 illustrates, the simple average for six countries rises from 25.5 to 47.3 per cent in twenty-three years, with non-interest income significantly above this level in France as well as in the United Kingdom; the outlier is Germany where non-interest income hovers between 20 and 30 per cent, due to the federalist structure of the banking system in those two countries (Verdier 2000). Non-interest income after 2003 primarily came from trading income in wholesale markets, even in countries like France. The staggering ratio of 75.2 per cent in France in 2007 in Table 4.1 probably tells us what the likes of Jeróme Kerviel of Société Générale were engaged in before it blew up in 2008. However, the US banks had already started to suffer the impact of the dot-com crash on fee income after 2001, and the trading losses in CDO markets in 2007. The decline in interest income is driven by several distinct developments. These include the secular decline in nominal interest rates in the 1990s after the 'conquest of inflation', which had the unintended consequence of reducing spread between deposit and lending.
Figure 4.8 shows that the pressure continued through the bubble because in the United States, the United Kingdom, and in the Eurozone, net interest margin declined in the decade before 2007 from 3.0 to 2.0 per cent in the US case and from 1.5 to 1.0 per cent in the two other cases.
But this is only part of the story because, if traditional intermediation was inexorably less attractive, it encouraged retail and wholesale banks (and the new conglomerates combining both retail and wholesale) to get further into retail mass marketing and wholesale trading, which are of course connected through the markets because products like mortgages provide the feedstock for the wholesale markets. We first discuss the transformation in retail business models and then move to wholesale. In each case, we will emphasize the connection with leverage, complexity, and velocity which in wholesale is tied in to the 'comp ratio' business model which incentivizes senior bankers. Most critical analyses of the banking after the crisis focus on the hyperbolic growth of bank assets in relation to the productive sectors of the economy and the role complex securitization involving CDOs has played in making such bloated banks unstable. Turner (2010b), for example, insightfully analyses these secular trends in banking. But these studies tend to overlook the bank business models that drove these quantities and qualitative changes. Bank business models that are both shaped and justified by the shareholder value principle, but ultimately favoured the banking elites at the expense of shareholders, require a closer intellectual scrutiny.
The customer-facing changes in retail business model are less radical than those envisaged in the visionary new economy literature of the 1990s (see e.g. Evans and Wurster 1997; Llewellyn 1999), where retail bank branches and bookstores figured as examples of obsolescence because their activity would surely move onto the Web in ways that put incumbents with bricks and mortar investments at a serious disadvantage. Instead, incumbents in retail have moved the routine transaction, standing order, and account statement part of retail onto the Web but retained their branches because customers want them, and market share of mainstream retail business therefore requires a branch network. Consider the evidence from the United Kingdom: 80 per cent of consumers say that their preferred channel of arrangement for current accounts is a branch, and company shares of the current account market are closely correlated with the extent of a particular provider's branch network (Datamonitor 2008). Branches are also the material support of customer inertia which is such a striking feature of retail banking. Only 7 per cent of British current account customers switch in any twelve-month period and 65 per cent of consumers have held their current accounts for more than ten years (Ipsos MORI 2008). Before or after the crisis, retail consumers mistrust banking in general but have more trust in their primary bank. In a 2009 survey, the average consumer rating of trust in the banking sector was 2.1 out of 5 but the average trust rating in their (current account) primary bank was 3.4 out of 5 (Datamonitor 2009).
But this put retail banks under pressure. If market share requires retail branches which incur high rents and staff costs, how could banks recover these costs (especially when spreads on intermediation were declining)? One expedient was cross-subsidy from other retail activities like credit cards, with higher margins than account-based business where margins were also boosted by fee charges for overdrawn account holders and such like. But the major development was to use the bank branches for retail mass marketing of products like insurance policies, mortgages, and pension plans on which the retailer earned an upfront fee which was typically calculated either as a percentage of the value of the product or as a fixed amount. For example, most mortgages require typically a 2 per cent fee to cover arrangement and booking. This major shift is reflected in the visible changes of layout in retail bank branches. Intermediary retail banking operated out of branches that were dominated by the long counter across which money was paid in and out by bank clerks, while the manager in a side office made decisions about lending to households and businesses. By the 2000s, bank branches are dominated by cubicles and workstations where advisers sell financial service products like mortgages and pension plans, and decisions about personal loans are made over the phone in minutes by junior call-centre staff using FICO scores1 and other assessment techniques.
There were other, equally important but much less visible, changes because the new retail business models were increasingly volume-based and connected via mortgage securitization with the wholesale markets. In an old-style intermediation operation, the volume of lending depends on the availability of deposits, where limits are set by the balance sheet of the firm. The relation applies in the case of a national retail chain but is most obvious and constraining in the case of a community bank which operates a closed circuit between local savers and borrowers. Thus, James Stewart in Frank Capra's 1946 It's a Wonderful Life stopped a bank run by warning panicking savers about the effects their withdrawals would have on local borrowers. But, after deregulation and securitization, banks can obtain funds to lend and/or ramp up feeearning transactions by reaching beyond their balance sheets: banks can, for example, obtain new funds to lend by borrowing short-term on the overnight money market, and banks do not have to hold loans on their balance sheet because they can be securitized and sold on, so that the bank can lend and lend again. The new volume-based business model is the end result because, as in any retail operation, banks will try to cover high fixed costs by ramping volume.
The results are transformational. While some banks remained retail only (in that they do not have trading divisions), most banks were increasingly connected to the wholesale markets, especially through the securitization machine described in Chapter 2.
Figure 4.9 presents data on the size of the deposit base of UK commercial banks relative to total assets in 2007. Leading up to 2007, the sharp decline of the importance of saving accounts as a source of funding for banks is a direct reflection of the increasing amount of shortterm borrowing on money markets as well as the securitization of assets to provide alternative sources of funding. This was inevitably a source of new risks in retail banking, which has always had to manage a mismatch between short-term liabilities and long-term assets. Turner (2010b) underlines the systemic risk that arises out of this, what is usually termed a customer funding gap: 'this funding gap and reliance on wholesale funding created significant vulnerabilities for the UK banking system which crystallised in 2007 and 2008' (Turner 2010b: 17). The wholesale markets themselves then become the equivalent of James Stewart's panicky savers because retail banks are increasingly vulnerable to mood swings on the overnight money market and other debt markets, and to crisis-led closure of securitization processes which depend on willing counterparties and liquidity (Gorton and Metrick 2010).
The consequences of this retail revolution for the consumer are also very mixed. The volume requirement encourages aggressive selling of investment, insurance, pension, and mortgage products in the retail branches, while high fixed costs mean the stripping out of back-office functions and customer support which has no direct revenue stream. All this is covered by a large marketing spend on advertisements which feature friendly, helpful retail staff. Thus, in the United Kingdom, RBS has been promoting its NatWest brand under the slogan 'helpful finance'. A series of TV commercials by Yipp Films uses real staff and customers to illustrate how NatWest has 'money sense advisers' in more than one thousand branches. The voice-over sententiously claims 'they are not there to sell but to give you free impartial advice'. When the consumer body Which sent researchers to NatWest branches, it found that only four out of twenty sessions offered impartial advice while sixteen of twenty sessions ended in attempts to interest the consumer in NatWest products, which were the only products mentioned in six sessions (Bachelor 2009). The branch network is then the place where the implied promise of good advice is betrayed as an adviser on incentive pay 'sells to' the retail household.
The developments in the business model of wholesale finance have been rather different. Investment banks have always lived off fee income and the revolution here was about the changing sources of income with the relative decline of fee income earned from advice and issues, as 'proprietary trading' became the main source of profit for investment banks. In the late 1980s, investment banking earned fees by providing corporate advice to productive firms, especially on M&A, and by issuing bonds and shares for corporate customers. From the early 1990s, investment banks increasingly moved towards becoming principals in the financial markets and shifted into ownaccount trading by setting up trading desks for derivatives and such like – just as, more recently, investment banks have set up hedge funds and private equity funds in their own right (as well as advising others on such deals). The connection with volume and leverage is of course very strong because, as we have noted in Chapters 2 and 3, leverage increases the returns on profitable trades and deals by allowing traders to take larger positions.
The shift to proprietary trading should be straightforwardly measurable but it is difficult to track empirically, even for stand-alone investment banks that are public companies and obliged to report results by segment. The measurement problems are caused by limited accounting disclosure in the first half of the 1990s and the continuing absence of a standard set of segment categories for identifying and classifying investment bank activities. In 2005, when Goldman Sachs reported earnings 50 per cent higher than the analysts' forecast because of better than expected results in proprietary trading, Lex of the Financial Times commented that Goldman Sachs was a 'black box . . . albeit one that continues to deliver pleasant surprises' (Lex 2006). However, Bank of England economist Andrew Haldane, who has become famous for his unorthodox analysis of the crisis and banking as discussed in Chapter 1, is belatedly critical of this kind of commentary:
Many practitioners and policymakers were seduced by the excess returns to finance during that twenty-year golden era. Banks appeared to have discovered a money machine, albeit one whose workings were sometimes impossible to understand. One of the South Sea stocks was memorably 'a company for carrying out an undertaking of great advantage, but nobody to know what it is'. Banking became the 21st century equivalent. (Haldane 2009b: 2)
Figures 4.10a and 4.10b look inside the black box by presenting data on Goldman Sachs, a leading investment bank that has no retail activities. Goldman Sachs distinguishes between three categories: first, investment banking advice and services; second, proprietary trading and principal investments; and, third, investment management for clients. Already in 2003, revenues from proprietary trading were larger than those from traditional investment banking and their importance has since increased (albeit unsteadily because investment banking advice was and is a cyclical business). By 2007, proprietary trading accounted for almost two-thirds of Goldman Sachs' revenues and assets, and the bank was, in its own right, the largest hedge fund manager in the world. For European investment banks, Goldman Sachs and its American peers were the new pattern and exemplar. A McKinsey report in 2006 discussed how European investment and corporate banks played a lead role in developing credit and equity derivative markets, and makes the point that the late 1990s and early 2000s were a time of significant product innovation by these institutions as they sought to develop competitive positions against the US banks by emulating their prop trading successes (Roxburgh 2006).
The question of what has happened to proprietary trading since the crisis is illuminating. And the short answer is that it has become very much less visible because of the threat of re-regulation, but it has not gone away because it remains lucrative and its continuing importance is indicated by the effort going into new ways of finding funds for levered trades. There can be no doubt that, after the crisis, proprietary trading has to negotiate a new climate of disapproval which is encouraging all kinds of relabelling and circumvention. The British regulatory response has been more emollient as the regulator, the Financial Services Authority (FSA), accepts the banks' claim that proprietary trading played only a marginal role in the crisis (Hughes 2010), but the Obama administration has pressed ahead with curbs on proprietary trading by banks, and the legislature has passed the so-called Volcker rule that forces banks to unwind own account trading. The US banks have responded by reclassifying their activities and drawing a distinction between 'flow trading' that should not be counted as proprietary trading and 'pure prop trading'; the latter turns out to be of only marginal importance to the revenue and profit of banks, ranging from 10 per cent in the case of Goldman Sachs to 1–2 per cent in the case of European banks such as Barclays, UBS, and Credit Suisse. If this does not satisfy the Securities and Exchange Commission (SEC), the banks can no doubt in the next stage construct arms-length (but profitable) arrangements whereby their star traders set up independent prop trading firms and, following the analysis in Chapter 3, the nomadic war machine can take a different form, with a new group of irregulars armed and provisioned by the banks.
The maintenance of volume is critical because of the 'compensation (comp) ratio' practice embedded in the business model of wholesale banking in Wall Street, London, and other major centres. In investment banks like Goldman Sachs, or in the wholesale divisions of conglomerates banks like Barclays, the understanding is that the senior investment bankers will (through base pay and bonus) receive a definite percentage of net revenues, usually somewhere between 40 and 50 per cent. While there has been criticism of how bankers are paid in bonuses since the crisis, nobody has questioned the basic comp ratio principle that senior bankers should receive a semi-fixed proportion of the turnover. This makes the investment bank a kind of profit share arrangement between shareholders and senior investment bankers who, at firm and trading desk level, have a direct and simple incentive to increase turnover. Figure 4.11 shows the total compensation cost for all employees for the major US investment banks, Goldman Sachs, Merrill Lynch, and Lehman, up to the crisis, as well as the amount that went to shareholders in the form of dividend and/or cash (share buy-backs).
The comp ratio system combines the worst features of the soccer star system and of CEO pay, but under finance sector-specific conditions which encourage explosive results. In a major soccer club, the players will get a substantial chunk of around half of revenue and, as in investment banking, it is difficult to cut pay when revenue falls through relegation or non-qualification because of easy exit for star players; just as in a football club, the owners' returns get squeezed when times are bad. Looked at another way, an investment banker's pay is like that of UK and US senior executives, who are effectively paid according to company size (Froud et al. 2008). However, while CEOs in nonfinancial firms generally have limited opportunities to increase turnover organically if acquisition is blocked, senior investment bankers operate under different sectoral conditions where bricolage can be used to increase turnover by adding extra steps to transactions circuits, and leverage can increase the value of trades. From this point of view, the financial crisis was simply what happened when the comp ratio provided the incentive for bricolage, which was facilitated by the availability of new instruments and easy leverage. Turner (2010b) argues that proprietary trading that drives bonuses in investment banking enriches position-taking bankers rather than creating liquidity-related efficiency in financial markets:
. . . an emerging body of analysis which suggests that the multiple and complex principal/agent relationships which exist throughout the financial system, mean that active trading which both requires and creates liquid markets, can be used not to deliver additional value to end investors or users of markets, but to extract economic rent. (Turner 2010b: 40)
If we put together our analysis of retail and wholesale banking business models, this is not intermediation but banking for itself and banking has become a giant transaction-generating machine. As we have seen, the ongoing daily socio-economic cost of this in retail banking is mis-selling which undermines the bank's claim to help the customer; in wholesale banking the ongoing daily socio-economic cost is conflicts of interest which undermine the bank's claim to advise clients. As with retail mis-selling, this problem of wholesale conflict of interests is repeatedly discovered and punished, but not so fiercely as to prevent further misbehaviour. In investment banking, the pattern has been set by the ritualized interaction of US regulators with investment banks and conglomerates over conflicts of interest and mis-advice: a case is prepared and charges are filed before an out-of-court settlement is struck whereby the bank pays a fine without admitting guilt. The size of the occasional fine is such that this is an acceptable cost of doing business.
This pattern was very clear after the dot-com boom when the top ten investment banks in the United States (including Citigroup, Goldman Sachs, Merrill Lynch, Lehman Brothers, and JP Morgan) reached a general settlement with the authorities and paid a record total of $1,387.5 million in fines (SEC 2003). In the case of Citigroup, the fine settled issues arising from its multiple, conflicted relations with WorldCom: Citigroup provided analysis and advice on WorldCom shares for its institutional and retail clients, while it was also advising WorldCom management on strategy and lending to and underwriting securities for WorldCom. In addition, Citigroup served as the exclusive pension fund adviser to WorldCom and executed significant stock option trades for WorldCom executives, while at the same time conducting its own proprietary trading in WorldCom shares.
After the financial crisis, the same thing happened all over again. In Chapter 3, we have already noted Goldman Sach's involvement in the sale of a complex synthetic CDO, Abacus 2007-AC1, which caused a loss of $1 billion for investor customers of Goldman Sachs, but produced a profit for the hedge fund Paulson & Co., another client. Here, we can focus on the legal outcome: the 2009 SEC case against the bank ended when Goldman Sachs reached a settlement neither denying nor admitting guilt and paying a fine of $550 million, the largest such fine ever levied but no more than a week's trading revenue for Goldman (Treanor 2010). In 2010, the Director of the SEC's Division of Enforcement, Robert Khuzami, claimed 'This [Goldman] settlement is a stark lesson to Wall Street firms that no product is too complex, and no investor too sophisticated, to avoid a heavy price if a firm violates the fundamental principles of honest treatment and fair dealing' (SEC 2010). Our own analysis has rather differently suggested that violation of those principles is part of the business model.
Banking after a Minskian moment?
In this section, we pay our respects to those who got it right. The postKeynesians and Minskians did see a crisis coming, though we argue that they missed out on the reinvention of banking and banking business models so that their diagnosis fails to capture what is at stake in the financial sector. The chapter then ends with a brief overview of what is at stake if we follow this analysis, and the arguments of Chapters 2 and 3. This serves as a counterpoint to the discussion of the limited and hesitant politics of reform in the second part of this book before the issues are taken up again in Chapter 8, where we indicate the limits of technocratic reform.
Of course, not everyone fell for the story of beneficial financial innovation and the line about banks as credit and risk intermediaries. In an important article, Dirk Bezemer (2009) has reviewed the pre-2007 work of postKeynesians and Minskians, like Wynne Godley and Steve Keen, and demonstrated that this group did see the unsustainability of the credit bubble in housing and warned that this could produce some kind of crisis. On the law of large numbers in mainstream economics, we would of course expect some (outlier) individuals to anticipate this crisis; the obvious example is the maverick Nouriel Roubini, who identified the housing bubble in the first half of the 2000s and predicted that it would lead to crisis. But Bezemer's argument is that the heterodox as a group had a collective, theoretical basis for their prescience because 'accounting or (flow of fund) macro-economic models helped anticipate the credit crisis and economic recession' (2009: 1) when mainstream equilibrium models did not. Understandably, the heterodox critics were also unclear about the timing of crisis and about the importance of the shadow banking system which would make crisis much worse.
Even so, Bezemer's conclusion is uncomfortable for all who believe in academic league tables and credentialized knowledge. The heterodox economists who did see it coming were a marginal internal group within the profession who had lost out in thirty years of movement to and strengthening of the mainstream in prestige economics departments. Thus, one of the most relevant and accessible applied economists currently is Steve Keen, who is an associate professor at the University of Western Sydney (which is of course not the same as being a full professor at the University of Sydney). Mainstream economists have, since the crisis, continued to treat such living dissidents in the time-honoured academic way by not citing them at all. But the media and internal critics of the mainstream, like Paul Krugman and Joseph Stiglitz, did show a new interest in the founding fathers of the dissident schools who were all safely dead. As the Wall Street Journal rightly noted shortly after the outbreak of crisis: 'In time of tumult, obscure economists gain currency' (Lahart 2007).
The rediscovered founding fathers were John Maynard Keynes and Hyman P. Minsky but, as we shall see, the endorsement of Keynes was short-lived and the interest in Minsky was not particularly discriminating. Some, like Skidelsky (2010) in the United Kingdom, insisted that the giant state interventions to save banks and shore up aggregate demand proved Keynes had the answers, but that looks much less plausible three years after the crisis began in view of huge fiscal deficits and sovereign debt problems in high-income countries. Minsky gained more enduring recognition because his work provided a much more detailed analysis of the role of banks in the economy, and indeed he had predicted the recurrence of financial instability induced by bank lending and credit. The media took this up, so that the New Yorker and others enlisted Hyman Minsky as a prophet and labelled the crisis as a 'Minskian moment' (see e.g. O'Connell 2007; Thomas 2007; Wilson 2007; Cassidy 2008; Mason 2008), as the following quotation illustrates:
Twenty-five years ago, when most economists were extolling the virtues of financial deregulation and innovation, a maverick named Hyman P. Minsky maintained a more negative view of Wall Street; in fact, he noted that bankers, traders, and other financiers periodically played the role of arsonists, setting the entire economy ablaze. (Cassidy 2008)
So what exactly did Minsky argue? Like other post-Keynesians, he rejected the neoclassical notion of automatically equilibrating markets and insisted on the need for a theory of instability and an associated control practice: 'identifying a phenomenon is not enough; we need a theory that makes instability a normal result in our economy and gives us handles to control it' (Minsky 1986: 10). Minsky's distinctive contribution comes from his focus on financial markets, banks, and credit as causes of instability. Keynes had related instability to disequilibria in labour and product markets, initiated by speculation in liquid (stock) markets, but Minsky put the primary emphasis on instabilities arising in capital markets initiated by bank lending and debt. When the monetarists opened up a debate on free markets, full employment, and inflation with the Keynesians in the early 1980s, Minsky was one of the first to point out the growing importance of credit money issued by banks in influencing investment decisions.
One proposition that emerges from Keynes's theory is that, from time to time, a capitalist economy will be characterized by persistent unemployment. The neoclassical synthesis accepts this result, even though a deeper consequence of the theory, which is that a capitalist economy with sophisticated financial practices (i.e. the type of economy we live in) is inherently unstable, is ignored. (Minsky 1986: 100)
Interestingly, Minsky supposes that banks exist to create credit for productive investment activities, just as in mainstream economics. But, in contrast to neoclassical economists who see money as neutral, Minsky holds that the intermediary activity of banks can become speculative and disruptive.
In an economy in which the debt financing of positions in capital and financial assets is possible, there is an irreducible speculative element, for the extent of debtfinancing of positions and the instruments used in such financing reflect the willingness of businessmen and bankers to speculate on future cash flows and financial market conditions. Whenever full employment is achieved and sustained, businessmen and bankers, heartened by success, tend to accept larger doses of debt-financing. (Minsky 1986: 177)
To be able to distinguish conceptually between productive and speculative credit creation, and identify imminent bubbles, Minsky distinguishes between different types of 'financial structures'. When the present value of cash flows from capital or financial investments is sufficient to pay off debt, companies are what Minsky calls 'hedge financed'. When cash flows pay is sufficient to pay off interest only (and not the principal) then financing is 'speculative', according to Minsky; this happens when bankers tend to get innovative and help the economy grow. Under 'Ponzi financing', the third fragile 'financial structure' distinguished by Minsky, debt and financial innovation is used to pay off interest. This is the moment when the system becomes crisis-prone and is arguably what we witnessed in 2007.
The way in which a speculative boom emerges and how an unstable crisis-prone financial and economic system develops are of particular importance in any description of the economic process that is relevant for this economy. Instability emerges as a period of relative tranquil growth is transformed into a speculative boom. This occurs because the acceptable and the desired liability structures of business firms (corporation) and the organizations acting as middlemen in finance change in response to the success of the economy. (Minsky 1986: 173)
The eponymous 'Minsky-moment' refers to a bank-induced economic depression, when overvalued assets drop to their true value and most of the credit pumped into the economy suddenly leaves it again.
During periods of tranquil expansion, profit-seeking financial institutions invent and reinvent 'new' forms of money, substitutes for money in portfolios, and financing techniques for various types of activity: financial innovation is a characteristic of our economy in good times. . . . Therefore, in a capitalist economy that is hospitable to financial innovations, full employment with stable prices cannot be sustained, for within any full-employment situation there are endogenous disequilibrating forces at work that assure the disruption of tranquillity. (Minsky 1986: 178)
And at the end of the day it is the 'Minskian moment' that resets the expectations of future prices and values to a more 'healthy level'. This suggests that breakdown in financial markets, such as in the aftermath of the failure of Lehman Brothers, is nothing but unwillingness to trade resulting in lack of liquidity. The solution that follows from this is the injection of liquidity by central banks after the crisis, even if they earlier failed to 'take away the punch bowl'. Although banks are the endogenous source of bubbles and financial excess, their balance sheets after the crisis can be repaired by liquidity injection from the central bank so they regain their economic health to finance the next cycles of investments. This is Minsky's favoured technical fix which follows from his analysis, just like Keynes' preference for low interest rates flows from his argument about volatile expectations and the marginal efficiency of capital.
This Minskian apparatus of thought opens up a number of debates and issues, not least as to whether 2007 was truly a Minsky moment (Davidson 2008; Kregel 2008; Dymski 2009; Keen 2009). But, that debate should not distract from a more fundamental point about Minsky's a priori. Minsky was, like Keynes before him, what might be called a technical optimist about economic fixes for the problems generated by capitalism as a progressive system for generating welfare. Minsky believed in technical progress and explicitly endorsed the necessity of financial innovation: Minsky also believed in the possibility of steering a capitalist economy through new kinds of theoretically guided economic practices.
For Minsky, financial innovation was a special kind of creative destruction. It was intrinsic to banking because 'a banker is always trying to find new ways to lend, new customers, and new ways of acquiring funds, that is to borrow; in other words, he is under pressure to innovate' (Minsky 1986: 237). And thus financial innovation should not be suppressed: ' . . . capitalism without financial practices that lead to instability may be less innovative and expansionary; lessening the possibility of disaster might very well take part of the spark of creativity out of the capitalist system' (Minsky 1986: 328). This did not of course imply positive approval of any and all forms of financial innovation; in the 1990s after shareholder value, Minsky was pessimistic about the predatory nature of what he called 'money manager capitalism'.
The pension fund and mutual funds have made business management especially sensitive to the current stock market valuation of the firm. They are an essential ingredient in accentuation of the predatory nature of current American capitalism. . . . Money manager capitalism has led to a heightening of uncertainty at the firm and plant level; in particular, it has made the lot of middle management in firms unsure. (Minsky 1996: 363)
But more fundamentally, as we saw above, Minsky never renounced his optimism about technical fixes for banking and credit-led instabilities. Government and central banks can, as Greenspan would put it, 'mop up after the bubble bursts' and are together capable of sorting out the Ponzi-like balance sheets of banks and private companies before the economy launches itself to the next cycle of full employment and bust. Minsky thus provides us with a useful and unique vocabulary to discuss endogenous instability in present-day capitalism but simultaneously sees central banks as neutral, apolitical institutions with tools adequate for the job of stabilization. Framing capitalist instability in Minskian terms therefore directs us to a technocratic debate in both policy and academic circles about which policy interventions and which forms of bank regulation will restore order in credit and risk markets before they embark on another finance-led economic cycle.
Thus, Minsky shares the a priori of other heterodox economists who have never broken with the post-1945 pioneers who formulated 'hydraulic Keynesianism' and first conceived of the macro economy as a mechanism of stocks, flows, and ratios which generated employment, inflation, and all the rest. Before algebraic expression was standardized, this conception was materialized in the early plumbers' models of the macro economy as fluid moving in pipes between reservoirs. This basic concept has survived even though heterodox and mainstream economists have (after monetarism, Minsky, and all the rest) several times rethought which fluids and flows really matter; the corollary of the hydraulic concept is the idea of a technically informed policy practice (fiscal or monetary) which would steer the economy or damp fluctuations. But what if the problem is not knowing what to do as an expert policy adviser on the basis of a tight technical calculation of the best intervention in a world where economics contributes to perfectible rationality? What if the problem is understanding and changing other policymakers (including central bankers) whose loose, informal calculations of expedient intervention operate in a world of limited knowledge and ongoing mess? The post-Keynesians may have seen the crisis coming but are they prepared to provide advice? This is a serious question when the history of post-1945 economic management demonstrates the difficulty of constructing a credible technical management practice (Keynesian or monetarist) from limited knowledge, and under conditions of socio-political intrusion.
At this point it should be noted that our argument raises questions about Minsky's technical optimism which has been explicitly renounced by the current generation of Minskians and post-Keynesians who share our pessimism. Post-Keynesian economists tend to disagree with the mainstream media that this is a 'truly' Minskian moment (Davidson 2008; Kregel 2008; Dymski 2009; Keen 2009). For example, Davidson (2008) claims that the banking sector is not faced with a Minskian liquidity problem that can be resolved by a lender-of-last-resort central bank, but instead is suffering an insolvency problem that requires a different set of policy tools than Minsky proposed. Kregel (2008) argues that the current crisis is not an outcome of an endogenous Minsky process because the new banking of 'originate and distribute' led to the loss of banks' ability to evaluate credit as they no longer derive profits from credit intermediation. Dymski (2009) too emphasizes the transformation of banking from an interest income-driven machine of Minsky to a fee income-driven business, showing how this transformation invalidates a Minsky-inspired rescue of banking. Against this background, Keen argues that current problems of liquidity and solvency around wrecked balance sheets are beyond technical remedy and policy interventions so far have made things worse.
However, though I am proudly Minskian in my economics, I expect the bailouts to fail. Minsky, I fear, was an optimist. The basis for this opinion is the feeling that, even though Minsky gave Ponzi finance a key role in his 'Financial Instability Hypothesis', he did not foresee the extent to which misguided government action would rescue Ponzi financing from itself, and therefore renew it, in the name of systemic stabilisation. (Keen 2009: 9)
There is a large discrepancy between the pre-2007 romance of financial innovation and our own much darker story about finance and banking, as discussed in Chapters 2, 3, and 4 about conjunctural bricolage, the nomadic war machine, and the reinvention of banking through new volume-driven business models. Taken together, the message of these chapters is that finance and banking were not so much out of control as beyond control, so that some form of finance-led crisis was nearly inevitable and its consequences would be dire. And, though our analysis is more elaborate and multidimensional, we agree with the current generation of Minskians about the difficulty of expert management and mitigation by technical means. This is not an engineering problem and, on this basis, it is time to turn to a more explicitly political analysis in the second half of the book which examines whether what is beyond technical control can be brought under political control.
This is chapter 4 of the upcoming book After the Great Complacence by Engelen et al. These are the uncorrected author proofs, pre-published with permission of the author.