Interrogating the Algorithm: Debt,Derivatives and the Social Reconstruction of Stock Market Trading

Abstract

This article examines the rise of algorithmic trading and the invention of new financial instruments, their implications and effects. It politicizes and denaturalizes the technological ‘innovations’ that have created networks of control over groups that states and corporations have designated as threatening, then examines how such ‘innovations’ have transformed stock market trading from a geographically-fixed, paper-based institution to a high-speed electronic casino. Finally, it discusses how and why financial actors have amassed the social, economic and political power that allows them to engage in practices that cause immense social harm while remaining virtually untouched by state regulatory agencies, civil and criminal law.

Keywords

corporations capitalism surveillance technology crime and control

Introduction

This article is about the financial instruments that generated the sub-prime mortgage crisis of 2008 – Collateralized Debt Obligations (CDOs), Mortgage-Backed Securities (MBSs), ‘secured credit card receivables’ – and the algorithms and surveillance technologies that powered them. The value of these instruments is derived through sophisticated computerized mathematical models where algorithms assess value, financial products such as derivatives link formerly disparate forms of capital and blend them into one tradable instrument, then rating agencies assess the riskiness of the newly created instruments. The paper argues that these new modes of ‘asset securitization’, defined as the process of transforming ‘illiquid assets into tradable securities’ (Ellul, 2005: 16), were invented by the world’s most powerful financial actors as an instrument (very successful, as it turned out) to maximize their profit levels. It politicizes both the hardware at the core of finance-led capitalism and the neoliberal regulatory framing that enables the expansion, intensification and reproduction of the dominance of finance over production, that is, the discourses and policies at the heart of ‘financialization’. It argues that financialization – ‘a pattern of accumulation in which profit making occurs increasingly through financial channels rather than through trade and commodity production’ (Krippner, 2005: 174) – represents a fundamental shift in the nature of capitalism.

The goal of the article is to show how the sophisticated manipulation of breakthroughs in technology and mathematical science by financial actors, plus this sector’s monumental hubris, have made their financial manipulations possible. Legitimated by the discourse of ‘progress’ and ‘innovation’, their right to gamble with the life chances, pensions and jobs of millions of global citizens has never been seriously challenged. As we shall see, regulatory oversight has been minimal: regulators typically lack the budgets, specialized knowledge and, in most countries, the economic, political and cultural permission to purchase the latest, fastest, most powerful computers, maintain the vast server farms and retain the highly remunerated algorithmic gurus required to keep up with deep-pocketed business organizations. As this paper will show, the algorithmic surveillance technologies employed in stock market trading today constitute a highly effective form of technologically-enabled resistance that allows dominant financial institutions to evade civil, regulatory and criminal law. Although globalized trade has multiplied the amount of social harm financial traders can cause, regulatory permission was neither required nor sought for any of the ‘innovations’ discussed in this article. The power of financial capital and of the actors and institutions that act in its name has become hegemonic.

This paper will, first, trace the two factors that have enabled financialization today: the transformation of exchange markets and the explosion of technological innovations.¹ This historical perspective allows us to denaturalize and politicize the ‘common sense’ perspectives that equate progress and prosperity with the practices of capitalism. Second, it will track the exponential growth of the financial sector and the invention of new instruments of debt. Third, it will assess attempts to regulate and sanction exchange markets, banks and the financial sector in general. As we shall see, civil, regulatory and criminal law have been either woefully ineffective or entirely absent from their obligation to deter and sanction financial harm. Laws, particularly criminal laws, stutter into existence, if at all, only after each global crisis and financial meltdown.

I. Stock Markets and Technological Surveillance

Historicizing Stock Market Trading

In the 18th century John Law, a Scot from Edinburgh, came up with the idea of selling shares – ‘transferable, tradable instruments’ – to pay off French debt incurred through the Napoleonic wars (Ferguson, 2008). The very first stock market ‘bubble’ – where share prices rise to unsustainable heights before suddenly crashing – happened shortly thereafter, when Mr Law sold hundreds of shares to land companies seeking to move into the massive Louisiana territory (Gleeson, 1999; Schama, 1990). By the 19th century selling government securities had become one of the primary functions of a nation’s central bank, as important as controlling the money supply and setting prime interest rates (Braithwaite and Drahos, 2000: 91). These strategic powers allow central banks to regulate both the credit and the monetary system, the two essential, interconnected processes at the core of finance capitalism. Credit – ‘the power to allow or deny other people the possibility of spending today and paying back tomorrow’ (Strange, 1994: 88) – is based on trust, the expectation that the lender will get the money or goods back at some future time. Subjective judgments on the trustworthiness of creditors and credit instruments, then, shape the entire monetary system, impacting markets of production, currency and rates of exchange.

The first ‘modern’ stock markets were meeting places where those with capital could ‘nation-build’ by finding worthy entrepreneurs looking for capital to make ‘things’ – battleships, steam engines, automobiles, refrigerators. Stock markets were geographically fixed institutions, located in and centered on a particular nation-state, where human traders shouted buy-and-sell bids. Traditional stock exchanges were mutual, not-for-profit entities owned by their members, who typically enjoyed exclusive trading rights on the exchange and in that region or state. Record-keeping was pen and paper based.

Over the last two decades all this changed. All the major exchanges have been converted into for-profit commercial companies, open to outside investors and frequently selling their own shares on the stock market they operate (a process called ‘demutualization’). They now compete with new players and venues for new listings and the fees they generate (Snider, 2013). Trading itself has become an almost entirely electronic process, deterritorialized, abstract and virtual, dominated by Electronic Communication Networks (ECNs), algorithmic trading and spin-offs such as High Frequency Trading (HFT), Special Investment Vehicles and hedge funds. Today trading is more about the making and selling of specialized financial products than the making of ‘things’: profits from the financial sector were only 16 percent of total profits in the US economy in 1975, but by 2007 this had risen to 41 percent (McNally, 2011: 85–6). In this new ‘virtual economy’ profits, it was said, would be made through the trading of knowledge, information and symbolic assets such as brands. However, in fact they were made from the invention of thousands of new ways to package and sell debt. Thus consumer debt in the US, fed by relentless marketing and easy credit access, doubled from 1980 to 2007. But debt in the financial sector quintupled, from 25 percent of the American GDP in 1982 to 121 percent in 2008 (McNally, 2011: 86).

While the main goal of those who developed, adapted and adopted these instruments was (and is) to make it cheaper to buy and sell stocks, thereby increasing the number of transactions (the volume of sales) and maximize profits (in the language of financiers this is known as optimizing ‘liquidity’ and ‘price discovery’), the consequences of this uncontrolled social experiment are immense. The risk is exacerbated by the fact that governments under the sway of neo-liberal economic doctrines have cut or privatized programs that were formerly public and universal, such as old age pensions (Harvey, 2007), leaving people at the mercy of private market forces. These two factors – the creation of new debt instruments and the removal of social security support – have made citizens and institutions around the globe more dependent on world financial markets, more vulnerable to theft, misrepresentation and market collapse than ever before. And those who pay the price for this ‘casino capitalism’ are not the banks and large-scale investors who caused the problems in the first place – after the 2008 financial crisis they received government bailouts and have been enjoying record profits, bonuses and share prices ever since. Those who lost their houses, jobs and pensions were not the privileged few but the disadvantaged majority.

Historicizing Technological Development

The history of the technological ‘innovations’ that make this casino capitalism possible must also be de-naturalized and historicized. The nature and uses of technological innovations (discursively labeled ‘progress’), the complex systems of trading and regulation we see today, were not inevitable. This system of governance and control has been socially and politically created, nurtured, protected and sponsored by economic and political elites. The algorithmic surveillance technologies employed in stock market trading today, technologies that have transformed the buying and selling of stocks and enabled the creation of ever more complicated instruments of debt, are typically depicted as the serendipitous result of random inventions by ‘geeks’. Not so. As this section will show, technologies are human inventions, all are shaped by the political and economic forces dominant in a particular historic period. This is no less true of algorithmic technologies. The computer, the internet and the algorithms that program surveillance were ‘a consequence of specific processes of research and development guided by focused questions … [and] financed by agencies with … expectations of a return’ (Smart, 1992: 44).

Surveillance, ‘the collection, processing and analysis of personal information about individuals or populations … to regulate, control, govern, manage or enable their activities’ (Coleman and McCahill, 2011: 192), sprang from deep-seated human proclivities to watch over and protect their bodies, territory and culture (Smart, 1992). This often involved seizing the material and nonmaterial resources of those seen as rivals or enemies. Thus surveillance technologies originally developed as weapons of war: from radar and the science of cryptography to the unmanned drone. Computers evolved from British efforts during the Second World War to decipher German communications by breaking the main German coding system, and from the logistics of warfare – the need to supply and control large numbers of geographically scattered forces. Thus surveillance technologies were ‘invented’ to further domination and control by certain groups over others, not to maximize creativity or facilitate democratic decision-making. It is not surprising, then, that computer design has never gone far down that road.²

In addition to aggressing or defending against external enemies, various elites developed what became surveillance technologies to control the internal enemy. The perceived need to control ‘the lower orders’ motivated Bentham’s invention of the Panopticon, an architectural design and technological device for a new kind of institution, the penitentiary, that allowed the few (guards) to scrutinize the many (prisoners). The object was to reform the undesirables – the lazy and dissolute, the drunkards, debtors and thieves – by confining them where their every act was visible to their jailers, but inmates would never know whether they were being observed or not (Foucault, 1977). Surveillance here was a form of disciplinary power: the tyranny of constant visibility was meant to force those confined to internalize the dictates (work habits, values and morality in that order!) of the dominant elites of that period. The 19th and early–mid 20th century saw the construction of many institutions – schools, asylums and workhouses – modelled on the Panopticon.

Business organizations had their own ‘othered’ populations to control, namely their employees. Since the earliest days of capitalism businesses have been obsessed with finding ever more sophisticated (and intrusive) mechanisms to manage, discipline and ultimately eliminate human employees from the production process. Business has thus been a major player in the development of surveillance technologies, constantly commissioning studies to tell those at the top better ways to control their workforce. ‘Innovations’ to ‘increase productivity’ range from the 19th-century Taylorist workplace to the computer-monitored electronic workplace of today (Ball, 2010). The dream of controlling every element of production has been realized. The productivity of every call-centre employee – number of calls, number of sales, call duration minus the time taken for any bathroom or lunch breaks – is instantly available to the employer, employee and co-workers (Ball, 2010). Similar conditions prevail in factories, warehouses, delivery trucks, etc.

The digitization of records in separate corporate and state bureaucracies has given these bodies access to the activities, values and history of citizens and consumers. As files became mobile, transferable across and between private and public institutions, data could migrate. Individual profiles could be traced through data collected in bits of information from a myriad of sources, then re-assembled to suit the priorities and interests of the institution involved. The ‘computerized dataveillance’ made possible through this integration of surveillance capabilities (Haggerty and Ericson, 2006: 4) has become a key component of both state and commercial governance. Our ‘data doubles’ in cyberspace, however partial or erroneous they may be, determine our chances of boarding a plane, getting a credit card or being admitted to university.

On the consumer side of the equation the burgeoning surveillance technologies created through ‘the mutual constitution of digitization and commodification’ (Mosco, 2004: 156) have been appropriated as an endless 24 hour global marketing opportunity. Ever more knowledge about customers, their likes and dislikes, their personalities and areas of vulnerability and weakness, are now exploited. With the customary hubris of capital, permission to gather such information has usually been assumed (after all, who could quarrel with a new profit-generating opportunity!). But if consent should waver, it is encouraged through systems of rewards (platinum level Visas, frequent flyer points, affinity cards) and punishments (refusing access, denying credit, flagging ‘malingerers’).

On the state side, surveillance technologies experienced exponential growth following the attacks of 9/11 in the United States and 7/7 in Britain. The ‘military industrial complex’ no longer captures the immensity of the new surveillance society; it is now variously labelled a ‘global-security industrial complex’ (Boyle and Haggerty, 2009) or the ‘bordersecurity industrial complex’ (Barry, 2009). These terms indicate the degree to which military discourses, policies and rationales for surveillance and security have penetrated everyday modes of governance (Coaffee and Wood, 2006). The dangerously blurred line between military and civilian operations (Bigo, 2005) grants de-facto normative permission for the security industry to penetrate public and private life – as underlined by the Edward Snowden disclosures.

As this historical picture makes clear, the state-subsidized surveillant assemblage was designed to extend control over internal and external ‘enemies’ (workers, ‘criminals’ and ‘terrorists’), and to deliver greater profits to business. This legacy has serious implications: although ‘we’, the employee, consumer and citizen, can and do use these technologies in a vast number of ways and for a variety of purposes (new applications emerge almost daily), the basic ‘choices about deployment and anticipated use’ were made long ago (Mosco, 2004: 16). They are built into the design of hardware and much of the software. This does not mean it is impossible to resist or change them to serve different purposes, but it does set the bar for non-elite, non-specialist groups extremely high. However, as the next section demonstrates, those with massive economic and cultural power, the financial elites in the business of stock market trading, have successfully adapted key electronic ‘innovations’ to serve their own needs and interests with minimal difficulty.

II: Algorithmic Trading: Inventing New Instruments of Debt

This section outlines the exponential growth of the financial sector and the invention of new instruments of debt. It argues that dominant financial actors and institutions have created these instruments to maximize their profits, evade transparency and resist regulation. Digitized trading, it is claimed here, rewards those with the financial and political capital to develop and employ the fastest, most powerful computers and pay the (very high) salaries of the experts required to constantly tweak the algorithms and keep ahead of competitors. Regulatory agencies (as well as small competitors, mortgagees and retail investors) are put at a huge disadvantage because they cannot compete. As the primary federal regulatory agency in the United States, the Securities and Exchange Commission (SEC), pointed out when requesting a budget of $1.04 billion for the 2012 financial year: ‘a number of financial firms spend many times more each year on their technology budgets alone’ (http://www.sec.gov/about/secfy12congbudgjust.pdf, 14 February 2011, accessed 29 December 2012).

The invention of new financial instruments cannot be disentangled from the logics of neoliberalism, financialization and the mammoth increase in hegemonic corporate power they produced. The election of Ronald Reagan in the United States and Margaret Thatcher in the United Kingdom officially launched the religion of neoliberalism, which ‘swept across the world like a vast tidal wave of institutional reform and discursive adjustment’ (Harvey, 2007: 23).

Neoliberal theory posits that ‘prosperity’ and ‘efficiency’ are maximized when markets are ‘set free’ from government ‘interference’. The state should concentrate on reinforcing institutions that uphold market values such as private property, and ensure that the infrastructures that make capitalism possible – good roads, schools and prisons – are maintained. Government is to accomplish all this, despite cutting taxes for business and entrepreneurs, by increasing its ‘efficiency’. ‘Efficient’ governments cut or eliminate welfare, unemployment insurance and other public benefits, sell off state enterprises and deregulate or privatize state services.

The Reagan government began this process with the 1980 Depository Institutions Deregulation and Monetary Control Act, which removed ceilings on the interest rates commercial banks could offer. In the deregulatory climate of the 1980s and 90s, banks and insurance companies lobbied heavily and ultimately successfully to rid themselves of rules restricting where and how each could invest, the products they could sell, and the capital cushion each had to maintain. As trade volumes doubled and tripled, the distinction between commercial and investment banks continuously eroded. In 1999 with the passage of the Gramm-Leach-Biley Act (also known as the Financial Services Modernization Act), the last barriers came down, leaving market actors ‘free’ to pursue profits as they saw fit.

The other pillar of this process is financialization. As noted earlier, financialization is the practice of applying a monetary (exchange) value to everything that can be touched (tangible goods, services, natural resources) or imagined (futures, intangible concepts and processes). The goal, highly profitable for certain elites, is to turn ‘it’, whatever ‘it’ is, into a financial instrument, or a derivative of a financial instrument, that can be traded on some kind of exchange. Financialization turns all world assets – literally everything that can be imagined – into a commodity. This is the rationale behind the aforementioned demutualization of stock exchanges and the transformation of markets from trading shares in products to trading risks. A flood of speculative, unregulated financial instruments emerged to take advantage of these opportunities. Foremost among these (and not coincidentally key factors in the credit collapse of 2008) were mathematically-based ways of hiding risk by conceptualizing, bundling assets of varying worth, and selling them through financial products such as credit default swaps (CDS), Collateralized Debt Obligations (CDOs) and Mortgage-Backed Securities (MBSs).

For example, MBSs originate when a bank lends X dollars to an individual to buy a house. The bank then sells this mortgage to an outside investor and takes a profit. The bank will typically have an agreement to sell all its loans to a single issuer such as Fannie Mae or Freddie Mac. The mortgage then becomes part of a mortgage pool. (Government sponsored enterprises like Freddie and Fannie were favored because they were seen as lower risk: in 2003, three quarters of the $7 trillion in securitized mortgages were in home loans) (Ellul, 2005: 16).

The assets in this pool of mortgages are then subdivided into several classes of securities, a process called ‘tranching’. The GSE places these mortgages in a Special Purpose Vehicle (SPV) which it then insures against default (another device meant to minimize risk) and sells to investors around the world. The buyer is promised regular interest plus a return of the principle at maturity (Ellul, 2005). The bank that issued the original mortgage and all those who combined it with other mortgages, sliced and diced, bought and sold this new financial instrument – all extracting their fees along the way – are now off the hook if the borrower is unable to keep up the mortgage payments or if house prices fall – the scenario that set off the crisis of 2008.

The basis of these products is the algorithm. Trading floors are now ‘computerized visualizations of price fluctuations and yield curves’ (De Goede, 2011). Algorithmic trading is defined as ‘electronic stock market trading that uses a complex computer algorithm to control the buying and selling of stocks, shares and other investment products’ (Heywood, 2011). In theory, algorithmic trading can be used by any trader in any electronic financial market; in fact, it is typically used by large market players – hedge, pension and mutual funds – to execute large trades. A 2009 report estimated that the 300 security firms and hedge funds specializing in algorithmic trading made $21.8 billion in profits in 2008 (Heywood, 2011). The algorithm itself, devised by experts in mathematics, physics and computer science, instructs computers to buy or sell according to a formula that monitors and monetizes factors that could conceivably affect the price of stocks – wheat harvests in Russia, the buzz on Twitter, the health of the Euro, a change of government in Cambodia – thousands of datasets are continually accessed. Which datasets are chosen and how they are weighted is a closely guarded secret, but algorithms that deliver profits are subject to constant attacks through ‘predatory trading’, a term that refers to algorithms designed to spy on or crack the codes of other algorithms. To keep ahead of competitors, therefore, new formulae must be constantly devised – the average ‘shelf life’ of an algorithm before it is ‘gamed’ is down to 14 days (Malakian, 2011). The result is a kind of technological ‘arms race’ among traders, and between traders and regulators.

Algorithmic trading has been made possible by the development of Electronic Communication Networks (ECNs), known as Multilateral Trading Facilities (MTFs) in Europe, digitized computer systems that ‘bring together multiple parties … interested in buying and selling financial instruments. … Systems can be crossing networks or matching engines operated by an investment firm. … Instruments may include shares, bonds and derivatives’ (Banks, 2010). ECNs were originally devised to cut costs and increase the speed of trading by replacing brokers operating through ‘open-outcry’ on trading floors or via telephone (and receiving a commission for each transaction) with computers. Trades mediated by humans, it is said, were more prone to ‘clearing errors’ and were limited geographically. ECNs/MTFs allow financial products to be traded outside the traditional stock exchange – the physical location of buyers and sellers is irrelevant (Mizrach and Neely, 2006: 529).

High Frequency Trading (HFT) is a special class of algorithmic trading where computers initiate orders based on information received electronically, enabling orders to be completed before any human trader has had time to process the information. HFT allows those with access to it to ‘zip in and out of markets’, capitalizing on misprices or split second differences in prices across exchanges (Bowley, 2011: B5). While the profits on each trade may be miniscule (0.0625 or.01 per share) (Heywood, 2011), traders can execute multiple thousands of trades per second. Indeed Nasdaq claims that the average order is completed in 98 microseconds – 98 millionths of a second (Bowley, 2011: B5). To be successful, HFT requires a computer infrastructure that is reliable, fast and networked into many other networks.

‘Dark Pools’ (DPs) are another trading mechanism enabled by surveillance technologies. A DP is a digital trading platform ‘containing anonymous, nondisplayed trading liquidity that is available for execution’ (Banks, 2010: 3). Translated, this means market players outside the DP are kept ‘in the dark’, they have no way of knowing that these particular buy and sell orders exist and therefore cannot take advantage of them. This ‘nondisplayed liquidity’, through which selected traders can complete transactions before other market players even know the opportunity is available, is legitimated on the grounds that it allows firms to protect ‘sensitive information’ and avoids the market disruption (for example, panic buying or selling) that can occur when large buy or sell orders are made public. To obtain access to a Dark Pool subscribers, typically ‘institutional investors, broker-dealers and market makers’ (Banks, 2010: 14), must have an account with a particular ECN (also called a Multilateral Trading Facility, an MTF) which in turn must be registered with a national regulator. Only members can view orders. Profit is derived when the orders of those in a DP are executed at a better price than the ‘base reference price’ available on the open market. This advantage is made possible by speed: their trades are completed split seconds before ‘normal’ market players can act (Banks, 2010: 6).

These algorithmically induced ‘sneak previews’ have created a two-tiered financial market – one for powerful insiders with access to the required technology, a second one for the retail investor whose orders are merely completed as an ‘afterthought’ (Salkever, 2009). It is, in effect, a disguised form of insider trading that flies under today’s regulatory radar, benefitting financial insiders with privileged institutional, economic and cultural resources. Moreover, when machines rather than humans are programming automatic trades, their interaction can create disastrous, uncontrollable feedback loops. Sudden drops in market values – for example the ‘flash crash’ of 6 May 2010 when the Dow Jones Industrial Average dropped 573 points in 5 minutes – are now ‘routine’ (Stokes, 2011). They are created by the interaction of ‘legions of powerful, superfast trading algorithms’ that interact to create ‘a market that is incomprehensible to the human mind and impossible to predict’ (Stokes, 2011). The introduction of code takes markets ever further from human control, and the levels of technical expertise required for market success fortify exclusionary boundaries and lessen the chances of democratic participation and control. What are the implications when ‘a volatile and unpredictable trading process with effects that are impossible for humans to predict’ is mixed with a ‘specialized computer science framework’ (Molnar, 2011, personal communication)? And when this is done by institutions more concerned with maximizing their profit levels than examining the societal damage? As the next section makes clear, relying on regulators for protection against the predations of financial elites is distinctly unwise.

III: Algorithmic Trading and Governance

This section deals with regulatory inadequacy in the face of the technological superiority of the financial sector. The point is not, however, to advocate a simple technological ‘fix’ to address regulatory inadequacy (were it to be possible). The technological advantages of financial actors and the complexities of algorithmic trades are symptoms, empirical manifestations of the political, economic and cultural power and unchallenged hegemonic dominance of corporate/financial elites. This structurally rooted imbalance is what makes it exceedingly difficult for any institution to take meaningful action against the forces driving financialization.

This is complicated by the fact that the institutions responsible for governing stock markets and trading are numerous and complex. A wide array of public and private regulatory bodies with powers, jurisdiction and responsibilities varying by region and by nation-state are charged with overseeing a vast patchwork quilt of regulations. Because of this the following discussion will focus on public (state) regulatory regimes, particularly the world’s largest and richest regulatory agency, America’s Security and Exchange Commission (the SEC).

As noted above, all financial regulators suffer from an imbalance of power vis-a-vis those they are charged with governing. Demutualization, the expansion of trading platforms, algorithmic high-speed trading and its digitization have worsened their position. However, governments responding to the 2008 financial crisis are speedily progressing through the typical reform cycle, from initial outraged rhetoric to ‘how could this possibly happen’ to legislative proposals and the search for low lying ‘bad apples’ to take the blame. With the passage of the Dodd-Frank Wall Street Reform and Consumer Protection Act in the United States on 21 July 2010 and the well publicized 150 year prison sentence handed to Bernie Madoff on 30 June 2009 for running a Ponzi scheme),³ the market banking meltdown of 2008 is well into the legislative phase. In crisis periods, regulatory provisions that were deregulated or ignored during boom times are revived, maximum penalties increased, regulatory budgets re-visited and sometime restored, and the neoliberal anti-regulatory rhetoric of state legislatures is muted (Snider, 2000, 2009). But once the media spotlight shifts – to government deficits in the aftermath of the 2008 crisis – regulatory agency budgets are attacked and accusations that governments ‘went too far’ reappear. Throughout each pro-regulatory reform cycle, business’s omnipresent army of enablers – tax lawyers, accountants, investment advisors, and most recently PhDs in mathematics and computing – are working to devise new ways to evade, avoid, or nullify each new set of regulations (Braithwaite, 2005; McBarnet, 2004). Their inventions proliferate unchecked until the next wave of financial meltdowns occurs and the cycle begins again.

However, because every crisis brings a new crop of regulations, saying regulatory surveillance has been inadequate does not mean it is absent. Systems of automated, real-time monitoring of digitized trades and trading patterns began in the 1990s following the ‘Black Monday’ flash crash of 19 October 1987 when the Dow Jones Industrial Average dropped 508 points (Brady, 1988), the first two being the Advanced Detection System (ADS) and SOMA (Surveillance of Market Activity) (Brown and Goldschmidt, 1996; Kirkland et al., 1999). The real-time system used in the Canadian province of Ontario (Canada being the only developed country without a federal regulator) is typical of those employed around the world. Surveillance of daily trading activity here has been contracted out to a non-profit firm, Market Regulation Services (Williams, 2008). MRS uses data on ‘normal’ trading volumes and ‘average’ share price fluctuation to ferret out ‘abnormal’ trading activities. When an anomaly is discovered the system is programmed to issue an ‘Alert’. Since there are now thousands of trades every hour, the number of Alerts issued depends on the sensitivity of the computer program. The number of Alerts followed up, on the other hand, varies with the number and expertise of the employees hired by MRS, their jurisdictional mandate, legal powers and interpersonal skills. Their mandate will typically have been negotiated with the targets of the regulation themselves – financial actors with extensive social, economic and political resources – and will probably, therefore, be minimal (Coleman and McCahill, 2011; Snider, 2009). Stock market regulators quickly learn that measures to protect individual debtors and investors must not cut trading volumes, profits, or interfere with ‘liquidity’ and ‘competitiveness’.

In addition, regulatory agencies in the developed world have passed regulations specifically targeting the Brave New World of algorithmic trading. Back in 1997, the SEC passed Regulation OHR to ‘increase transparency’ and ensure that public orders ‘compete directly in the establishment of quotes’ (Banks, 2010: 164). This was followed by Regulation ATS in 1998, aimed at defining and overseeing electronic trading platforms ‘to ensure their integration into the national market system’ (Banks, 2010: 8). It required alternate trading venues to register as broker/dealer and disclose to the SEC the prices and sizes of its best-priced orders. This was followed by Regulation FD requiring companies to ‘ensure fair and equal distribution of information … to the marketplace’ (Banks, 2010: 8–9) and in 2007 by Regulation NMS whose goal was to ‘promote … competition from electronic platforms’ (Banks, 2010: 10). Its counterpart, MiFID in the European Union, has similar provisions and goals.

After (not before) the May 2010 ‘flash crash’, the SEC required the installation of circuit breakers that automatically stop trading if a stock’s price has fluctuated more than 10 percent in 5 minutes. The SEC has also proposed (not instituted) an audit trail, a database containing information on every trade aimed at ‘help[ing] regulators keep pace with new technology and trading patterns’ (Stoke, 2010). Mary Shapiro, then Chair of the SEC, has mused about requiring trading algorithms to include a governor which would limit the size and speed at which trades could be executed; others have suggested ‘speculative position limits’ (LaRocco, 2011; Morgan, 2011). Such measures represent attempts by regulators to ‘create a level playing field between different … trading venues’ and, as an aptly titled article put it, ‘catch up with technological advances’ (Gomber and Gsell, 2006: Abstract).

One of the most ambitious real-time surveillance (Alert) schemes is the SMARTS system, devised by a group of academics at University of Sydney, the Capital Markets Cooperative Research Centre (CMCR)⁴ (Brown and Goldschmidt, 1996). They claim to have developed the regulatory holy grail, an ‘outcomes-based framework’ that measures both the efficiency and the integrity of a market (www.smartsgroup.com). An efficient market, defined as one which ‘minimize[s] transaction costs while maximizing price discovery’ (Aitken and Harris, 2011: 23), is measured through a series of algorithms that assess price impact, price discovery and market resiliency. A fair market (integrity), defined as ‘one that minimizes the extent to which market participants engage in prohibited trading behaviours’ (Aitken and Harris, 2011: 24), is measured through proxies for insider trading, market manipulation and broker-agent conflict. The SMARTS group claims these ‘state of the art solutions’ will ‘empower exchanges, regulators and broking houses’ and ‘set[s] the benchmark for surveillance systems worldwide’ (www.smartsgroup.com). The scheme requires massive computing power to access – and in some cases create – the multitude of datasets required, and large salaries to attract and keep the specialists required to implement, program, update and repair it. And like all Alert systems, its efficacy will ultimately depend on the willingness and ability (the economic, political and cultural capital) of regulatory staff to intervene in the highly profitable games of powerful economic actors.

Meanwhile, Wall Street institutions specializing in HFT, such as Goldman Sachs, have been lobbying heavily against provisions of the Dodd-Frank Wall Street Reform and Consumer Protection Act, to ensure that their lucrative trading practices remain unregulated (Doering and Rampton, 2010). Since the Act was signed in 2010, scores of lobbyists have descended on Washington, all seeking to shape the new rules – presently very vague⁵ – in ways that favor the dominant financial players they represent. According to the Centre for Responsive Politics, $3.49 billion was spent on lobbying in 2009 alone. One organization, the US Chamber of Commerce, spent $144,496,000 in 2009 and a total of $651,035,680 over the preceding 12 years (Center for Responsive Politics, 2010). The Securities and Exchange Commission had annual operating budgets less than this – the SEC’s annual budget was $970 million in 2009, a 7 percent increase over 2008 (US Securities Exchange Commission, 2010: 6–7). It has requested $1.407 billion for the 2012 financial year. The enormous clout of Wall Street is aptly summarized as: ‘[L]aws are written their way. Treasury secretaries are drawn from their corner offices. Regulatory agencies are run with their well-being in mind’ (Frank, 2011).

Normative chatter continues in the dozens of national and international organizations specializing in financial regulation – bodies such as the International Accounting Standards Board (www.ifrs.org), the Bank for International Settlements, the EU Council of Economic and Finance Ministers, the International Organization of Securities Commissions (IOSCO) and FINRA (the Financial Industry Regulatory Authority (iosco.org). Regulators ‘dialogue constantly’ seeking a ‘general consensus’ on what should be done about HFT and its ‘Cheetah Traders’, Dark Pools and related practices (Chilton in LaRocco, 2011). None of these international agencies, however, can lay charges, assess sanctions or fine offenders – their sole purpose is to persuade (they have no coercive power) nation-level regulators to adopt their recommendations.

While regulators prevaricate, business continues to invent new ways to profit maximize with the potential to be even more destructive. High Frequency Trading firms have recently begun seeking ‘new asset classes’ to find ‘pricing inefficiencies’ that will yield ever more profits (for them).⁶ Algorithms are now being designed to ‘detect complex correlations between different asset classes’ (Morgan, 2011: online). HFT already makes up ‘about 50% of European trading and about a third in US markets’, according to the US Commodity Futures Trading Commission (CFTC) (LaRocco, 2011). And despite the Great Recession ushered in by the manipulations of financial elites, a recession that has caused so much pain for non-elites, ‘the country’s top 25 hedge-fund managers were by 2009 taking in $25 billion’ (Peck, 2011: 62, emphasis added), which was more than they had taken in the pre-crash days of 2007.

In the battle to control those who prime the pump of financial capitalism, criminal law has been virtually absent. At one level of analysis, this is no surprise. To study corporate crime is to realize that equally harmful acts are not equally punished. Despite the enormous social harm this technological arms race has wreaked on the peoples of the world – the economic equivalent of a nuclear explosion, with world GDP declining 6 percent and $25 trillion obliterated overnight in 2008 – none of the activities described here have been legally constituted as crimes. The enormous ideological, symbolic and life-threatening power of the sovereign state to criminalize and sanction, so overused against traditional law-breakers, has been kind to those whose greed and fraud cause financial disasters. Despite studies showing that the illegal acts of corporations are responsible for far more injuries, deaths and financial loss than the traditional thieves, muggers and murderers who fill You-Tube videos and newspapers, the legal framework, discourse and sanctions visited on privileged offenders is very different (Slapper and Tombs, 1999). Many researchers – virtually the whole corpus of corporate and white-collar crime literature from Sutherland (1945) on – have tried to excuse or explain the relative absence of criminal sanctions for corporate offences, citing the complex nature of corporate crime, the lack of expertise and lenience of judges and juries, difficulties residing in the mens rea requirements of Western law, difficulties in identifying and proving criminal liability in a multi-layered corporate structure, the expense of criminal proceedings, and much else (Braithwaite, 2005; Haines and Gurney, 2003; Parker, 2006; Shapiro, 1984; Shearing, 1993).

The privileged position of entrepreneurial, profit-maximizing activities, individual or corporate, is further enabled by the disagreements that rage among scholars. Economists, business professors, political scientists and sociologists, the ‘authorized knowers’ in this domain, have for decades debated whether (plus where, when and why) criminal law ‘should’ be used.⁷ Since criminalizing the corporation – holding it and its executives accountable for fraud, theft, negligence, injury, and manslaughter and/or for the economic and human damage it causes – is hugely expensive for governments, they have seized academic legitimations with alacrity. For any major case the costs of prosecution can easily consume the annual budget of a regulatory agency, because corporations have entire departments poised to exploit every legal loophole and file every possible injunction and constitutional challenge.⁸ Political costs may also be astronomical: drops in campaign donations or electoral support often result when corporations enlist friendly media and ‘experts’ to support their perspective, and/or threaten local jobs by talking of moving to less regulated areas. Thus it is not surprising that, as Friedrichs (2013: 21) points out: ‘at this writing not a single high-level private or public sector executive or official has been convicted of criminal charges in relation to [the] financial meltdown’.

This section has discussed the normalization and neutralization of the algorithmic patterns driving accumulation in financial capitalism and the passive responses of regulatory agencies. Given the structural and historical roots of regulatory weakness, the chances that regulatory experiments with direct surveillance will prevent and/or sanction the market frauds, bankruptcies and ‘flash crashes’ of the future are not good. As one expert has observed (Austin, 2010: 452), regulatory agencies are ‘directly in the “firing line” over problems in the markets’ and under heavy pressure to produce immediate results. And governments under such pressure ‘typically pursue the “soft” targets’ (the small independent brokers and firms) while ignoring investigations ‘where the trail … is too difficult, complex, expensive or just takes too long’. Over five years after the 2008 debt and market crisis, national and international bodies are still ‘consulting’ with their ‘stakeholders’ (that is, the primary perpetrators!) and debating ‘appropriate’ measures which might (or might not) be turned into binding regulatory statutes.

Conclusion

As we have seen, cultural and policy responses to the ‘innovations’ of financial capitalism have generally followed scenarios laid out by dominant corporate actors. While this is no conspiracy, it is certainly not accidental. To be sure, the state-sponsored regulatory apparatus has its own specific knowledge claims, interests and agendas, its distinctive terrain, and actors which exert their own ‘distinct determinative forces’ (Garland, 2006: 421). However, its rationalities, mentalities and strategies are shaped within, and to a considerable degree reflect, the ‘common sense’ of financialization and neo-liberalism. Similarly, while competitions between different constituent groups with different claims to knowledge all try ‘to (re)shape relations of governance and have their ideas and interests taken up and represented in the political realm’ (Froats, 2011: 35), the knowledge claims that become dominant are the ‘wisdom’ of financial capitalism – markets and market forces are always ‘right’, and the key task of regulators is to facilitate market forces and ‘investor confidence’. Thus, especially during boom periods, it is literally unthinkable for political authorities or regulators to take measures that would seriously threaten finance capital.

This is why modern legal institutions have remained so inadequate in dealing with financial crime despite the immense social harm they have caused. This is why Big Banks, Wall Street traders and their equivalents all over the world have assumed the right to experiment with people’s life chances and prospects, albeit under the thin legitimation of discourses of ‘innovation’. This is why capital’s obsession with time and technology as profit-maximizing devices has not been seriously challenged by any governance regime – even though it is openly admitted that nothing of value is produced by the creation and incessant high-speed trading of ‘Credit Default Swaps’, Dark Pools and similar vehicles (Soederberg, 2010). To challenge the technological arms race described here, the ‘normalization’ and ‘neutralization’ of algorithmic patterns driving accumulation and the ‘normative chatter’ around these developments, weakening the hegemonic dominance of financialization will be required.

Footnotes

Funding

The author wishes to thank the Social Sciences and Humanities Council of Canada for research grant no. 410-2009-0472.

Notes

References

Aitken

Harris

(2011) Evidence-based policy making for financial markets: A fairness and efficiency framework for assessing market quality. Journal of Trading 6(3): 22–31.

Austin

(2010) Government to the rescue: ASIC takes the reins of the stock market. Company and Securities Law Journal 28: 444–459.

Ball

(2010) Workplace surveillance: An overview. Labor History 51(1): 87–106.

Banks

(2010) Dark Pools: The Structure and Future of Off-Exchange Trading and Liquidity. New York, NY: Palgrave-Macmillan.

Barry

(2009) The national security complex – integrating military, intelligence, and homeland security. Available (accessed 19 May 2010) at: http://borderlinesblog.blogspot.com/2009/09/national-security-complex-integrating.html

Bigo

(2005) The globalization of (in)security? Reflections on the field of the professionals of unease management and analysis of the transnationalization of the (in)securitization process. Cultures and Conflicts 58: 53–100.

Bowley

(2011) The new speed of money, reshaping markets. New York Times, 2 January, B5.

Boyle

Haggerty

(2009) Spectacular security: Mega-events and the security complex. International Political Sociology 3(3): 257–274.

Brady

(1988) Brady Report: Presidential Task Force on Market Mechanisms (1988): Report of the Presidential Task Force on Market Mechanisms. Washington, DC: US Government Printing Office.

10.

Braithwaite

(2005) Markets in Vice, Markets in Virtue. Annandale: Federation Press.

11.

Braithwaite

Drahos

(2000) Global Business Regulation. Cambridge: Cambridge University Press.

12.

Brown

Goldschmidt

(1996) ALCOD IDSS: Assisting the Australian stock market surveillance team’s review process. Applied Artificial Intelligence 10(6): 625–642.

13.

Coaffee

Wood

(2006) Security is coming home: Rethinking scale and constructing resilience in the global urban response to terrorist risk. International Relations 20(4): 503.

14.

Coleman

McCahill

(2011) Surveillance & Crime. London: SAGE.

15.

Committee on Financial Services (2010) Brief Summary of the Dodd-Frank Wall Street Reform and Consumer Protection Act. Washington, DC: Committee on Financial Services.

16.

Doering

Rampton

(2010) CFTC weights crack-down on high-frequency trades. Available (accessed 13 July 2012) at: http://www.reuters.com/article/idUSTRE66C6A220100713

17.

De Goede

(2011) How to Fight a Derivative: Society and Space–Environment and Planning. Available (accessed 21 November 2010) at: http://www.societyandspace.com

18.

Ellul

(2005) The economics of asset securitization. Business Review Q3: 16–25.

19.

Ferguson

(2008) The Ascent of Money: A Financial History of the World. New York, NY: Penguin.

20.

Foucault

(1977) Discipline and Punish: The Birth of the Prison. Toronto: Random House of Canada.

21.

Frank

(2011) Easy chair: The age of Enron. Harper’s, 2 August: 7–11.

22.

Friedrichs

(2010) Trusted Criminals: White Collar Crime in Contemporary Society, 4th edition. Belmont, CA: Thompson Wadsworth.

23.

Friedrichs

(2013) Wall Street: Crime never sleeps. In: Will

Handelman

Brotherton

(eds) How They Got Away with It: White Collar criminals and the Financial Meltdown. New York, NY: Columbia University Press, 3–26.

24.

Froats

(2011) The discursive power of risk: Rewriting the Goudge Report on paediatric death investigation in Ontario. Unpublished master’s thesis, Queen’s University, Kingston.

25.

Garland

(2006) Concepts of culture in the sociology of punishment. Theoretical Criminology 10(4): 419–447.

26.

Glasbeek

(2002) Wealth by Stealth. Toronto, ON: Between the Lines Press.

27.

Gleeson

(1999) Millionaire: The Philanderer, Gambler and Duelist Who Invented Modern Finance. New York, NY: Simon & Schuster.

28.

Gomber

Gsell

(2006) Catching up with technology: The impact of regulatory changes on ECNs/MTFs and the trading venue landscape in Europe. Journal of Competition & Regulation in Network Industries (Special Issue on the Future of Alternative Trading systems and ECNs in Global Financial Markets) 1(4): 535–557.

29.

Haggerty

Ericson

(2006) The New Politics of Surveillance and Visibility. Toronto, ON: University of Toronto Press.

30.

Haines

Gurney

(2003) The shadows of the law: Contemporary approaches to regulation and the problem of regulatory conflict. Law & Policy 25(4): 353–380.

31.

Harvey

(2007) Neoliberalism as creative destruction. The Annals of the American Academy of Political and Social Science 610: 21–44.

32.

Heywood

(2011) Algorithmic Trading. Available (accessed 30 May 2011) at: http://australiatrading.com

33.

Kirkland

Senator

Hayden

Dybala

Goldberg

Shyr

(1999) The NASD regulation advanced-detection system (ADS). AI Magazine 20(1): 55–68.

34.

Krippner

(2005) The financialization of the American economy. Socio-Economic Review 3(2): 173–208.

35.

LaRocco

(2011) CFTC’S Commissioner Chilton: ‘Cheetah Traders’ need to be regulated. Available (accessed 8 June 2012) at: www.cnbc.com/id/43328839

36.

Lyon

(2003) Technology vs ‘terrorism’: Circuits of city surveillance since September 11. International Journal of Urban and Regional Research 27(3): 666–678.

37.

Malakian

(2011) What I Learned at Sifma 2011. Available (consulted 30 August 2011) at: www.waterstechnology.com/waters/opinion/2086349/anthony-malakian-learned-sifma-2011

38.

McBarnet

(2004) After Enron: Governing the corporation, mapping the loci of power in corporate governance design. In: O’Brien

(ed.) Governing the Corporation: Regulation and Corporate Governance in an Age of Scandal and Global Markets. London: J. Wiley, 89–111.

39.

McNally

(2011) Global Slump: The Economics and Politics of Crisis and Resistance. Oakland, CA: PM Press.

40.

Mizrach

Neely

(2006) Advantages of ECNs (electronic communications networks) in the secondary treasury market. Federal Reserve Bank of St. Louis Review 88(6): 527–41.

41.

Morgan

(2011) HTF firms sniff out new asset classes. Financial Times, 30 May.

42.

Mosco

(2004) The Digital Sublime: Myth, Power, and Cyberspace. Cambridge, MA: MIT Press.

43.

Mügge

(2009) Tales of tails and dogs: Derivatives and financialization in contemporary capitalism. Review of International Political Economy 16(3): 514–526.

44.

Parker

(2006) The compliance trap: the moral message in responsive regulatory enforcement. Law & Society Review 40(3): 591–622.

45.

Peck

(2011) Can the middle class be saved? The Atlantic, September: 60–78.

46.

Rose

Miller

(1992) Political power beyond the state. British Journal of Sociology 43(2): 173–205.

47.

Salkever

(2009) Wall Street Ripoff 2.0: High Speed Trading and Deep, Dark Pools. Available (accessed 1 August 2011)at: http://www.dailyfinance.com/story/investing/wall-street-ripoff-2-0-high-speedtrading-anddeep-dark-pools/19116311

48.

Schama

(1990) Citizens: A Chronicle of the French Revolution. Toronto, ON: Vintage Books.

49.

Shapiro

(1984) Wayward Capitalists: Target of the Securities and Exchange Commission. New Haven, CT: Yale University Press.

50.

Shearing

(1993) Towards a constitutive conception of regulation. In: Braithwaite

Grabosky

(eds) Business Regulation and Australia’s Future. Sydney: Australian Institute of Criminology, 67–79.

51.

Slapper

Tombs

(1999) Corporate Crime. London: Longman.

52.

Smart

(1992) Modern Conditions, Postmodern Controversies. London: Routledge.

53.

Snider

(2000) The sociology of corporate crime: An obituary (or: whose knowledge claims have legs?). Theoretical Criminology 4(2): 196–206

54.

Snider

(2006) Making change in neo-liberal times. In: Balfour

Comack

(eds) Criminalizing Women. Black Point: Fernwood Publishing, 323–342

55.

Snider

(2009) Accommodating power: The common sense of regulators. Social & Legal Studies 18(2): 179–197.

56.

Snider

(2013) The technological advantages of stock market traders. In: Handleman

Wills

(eds) How They Got Away With It: White-Collar Crime and the Financial Meltdown. New York, NY: Columbia University Press, 151–70.

57.

Soederberg

(2008) A critique of the diagnosis and cure for ‘Enronitis’: The Sarbanes-Oxley Act and neoliberal governance of corporate America. Critical Sociology 34(5): 657–680.

58.

Soederberg

(2010) Corporate Power and Ownership in Contemporary Capitalism. London: Routledge.

59.

Stokes

(2011) Algorithms take control of Wall Street. Wired Magazine, 27 December.

60.

Strange

(1994) States and Markets: An Introduction to International Political Economy. London: Pinter.

61.

Sutherland

(1945) Is white-collar crime crime? American Sociological Review 10(2): 260–71.

62.

US Congress Office of Technology Assessment (1987) The Electronic Supervisor: New Technologies, New Tensions. Washington, DC: US Congress Office of Technology Assessment.

63.

US Securities Exchange Commission (2010) Strategic Plan, Fiscal Years 2010–2015. Available (accessed 10 December 2013) at www.sec.gov/about/secstratplan1015f.pdf

64.

Williams

(2008) The lessons of ‘Enron’: Media accounts, corporate crimes and financial markets. Theoretical Criminology 12(4): 471–499.