Autonomisation, financialisation, neoliberalism

Arbitrage is defined as any technique to profit from differences in price between identical assets in different markets. At its simplest, if an asset is priced at £1000 in one country and at £1200 in another, I can buy in the cheaper market and sell in the more expensive and profit from the difference. In real life the transactions required are usually more complex and often involve the use of hedging and swaps to cover the difference in risk between similar or related but not identical assets.

It is a fundamental tenant of modern economics and financial modelling that we can assume that that markets are free from arbitrage opportunities. In real world, of course, this is not the case, arbitrages do exist, and in practice traders will look for arbitrage and attempt to profit from it. Much of the automatic algorithmic trading that happens today is designed to take advantage of just this effect. This can itself be seen as a self-correcting mechanism, to some degree - as arbitrages are discovered and exploited they should naturally cease to exist. In the simple example above as more people buy the £1000 asset and sell the £1200 one the excess demand should raise the price of the first asset and the excess supply reduce the price of the second until they match and the trade ceases.

The existence or non-existence of arbitrage may seem of minor consequence, however, I want to argue that in fact unravelling this concept reveals much about the workings of the economy today and the political economy of neo-liberalism.

In order for arbitrage to be sufficiently small that it can be ignored from a broad economic perspective, a number of features are required to hold in a given market. I am going to examine three conditions and the political consequences that arise from them:

• markets need to be efficient
• there need to be no transaction costs
• transactions need to be instantaneous

Efficiency

The efficient market hypothesis (EMH) states, at its strongest, that all information about an asset is already in the price. This means that the past behaviour of an asset cannot affect the future price of that asset, be that a share, a security, a piece of property or anything else for sale in a public market. If shares in a company have been increasing for the last month, that should have no bearing on what the price does tomorrow, because the events that led to those rises have already been taken into account and future changes will be in response only to future events. The market is "efficient" at processing information, such that publicly known knowledge is instantly encoded into asset prices.

Mathematically this means that prices follow what is known as a Markov process, and it is a core assumption for almost all financial mathematics which allows the construction of tractable models.

Of course the EMH isn’t true; markets are not efficient. A cursory examination of the historical data reveals that there is short term positive and long term negative serial correlation in prices. If a share has been increasing in price for a week, it probably will go up tomorrow, though only on average and to a small extent, while over a longer period it will revert back towards the mean behaviour of the market as a whole. Knowing when it is likely to switch is, unfortunately, almost impossible.

Indeed, at its strongest the EMH is self-contradictory. If all public knowledge is already encoded in prices there is no need for any market participant to pay attention to real world events or to inform themselves of that information - it cannot give them an opportunity for advantage. In that case, by what mechanism does the market encode the information? Price changes must occur a finite time after the event to which they react. Nevertheless, the hypothesis is generally considered illuminating, if not exact, and most mainstream economics assumes we are "close" in some sense to efficiency, such that for all intents and purposes it can be assumed to be approximately true. 1

The most obvious political implication of market efficiency is that markets are therefore "better" than non-market organisation, e.g. than state/public/common ownership or control in terms of allocation and distribution of resources. This is, however, not quite what the term means. There is a deliberate (or simply ignorant) conflation of two distinct meanings of the word efficient, a quite narrow technical definition and a broader colloquial one.

For markets to be efficient they need information, and this information needs to be publicly available to all market participants - and potential participants - private information creates distortions and biases - hence the prohibition on insider trading. 2

This desire for information to be public is important, and a facet of neo-liberal thought which rarely receives the attention it deserves. There is often an assumption that open data - where data are free to access and use without restriction - is an inherently progressive and left-wing idea. Because it seems to fly against the desire of capital to enclose and protect private property, freely available data is seen as unquestionably a good thing. And, indeed, it offers huge potentials and advantages over closed, proprietary data. As Cathy O'Neil recently wrote, however, freely available to all doesn't mean the same thing for everyone and a radical opening of data against the competition stifling effects of intellectual property finds support amongst the neo-liberal right wing as well.3

As by-strategy wrote of Julian Assange's neoliberal utopianism: "Assange’s politics are neoliberalism’s ideal image of itself, entirely consistent with its politics to the extent that it radicalises them." 4 Assange considers himself a strong proponent of markets and part of wikileaks job to free them from the entrenched control of state and corporate bureaucracies intent on maintaining their personal control. It unleashes a more radical, more free, more creative capitalism from the stagnation of monopoly.

"Assange’s most lengthy articulation of his own politics comes in a lengthy interview with Forbes. Asked “Would you call yourself a free market proponent?”, Assange replies “Absolutely. I have mixed attitudes towards capitalism, but I love markets”. The stance that is ambiguous to capitalism, but in favour of markets represents the more extreme variants of neoliberalism, whereby capitalism (while it actually exists) plays second fiddle to an idealised vision of how markets function available on a minor scale within currently existing capitalism. Assange continues: “To put it simply, in order for there to be a market, there has to be information. A perfect market requires perfect information…For a market to be free, people have to know who they’re dealing with”. How does Wikileaks fit into this scenario? For Assange, through the act of leaking information, Wikileaks is providing better information in order for the market of international politics to work better. The question of informational asymmetry is a complex one in neoliberal circles, with a long history. Whereas neoliberalism in the variant of the Chicago School of Economics tends towards a model of equilibrium where actors have perfect information about the market, the Austrian school of Economics, favoured by the more radical anarcho-capitalist believe that information is unevenly distributed throughout a market system, and that to increase overall information enables better price setting thus improving the efficiency of the market.

Assange’s philosophy here blends Austrian and Chicago School approaches. Accepting the Austrian approach of informational asymmetry as the current situation, but believing that increased distribution of knowledge as a result of leaking would tend towards the Chicago assumption of perfect information. In the situation of perfect information, so runs the theory demonstrated mathematically by Keith Arrow and Gérard Debreu, then market transactions will tend towards a Pareto optimal state, where no actor can be made better off without making another worse off - a state that is a mathematical formalisation of Adam Smith’s notion of the “invisible hand”. Hence “WikiLeaks is designed to make capitalism more free and ethical”."

Transaction costs

Transactions within almost all financial models are assumed to have no costs. If I buy some asset and immediately sell it again in the same market there should be no associated cost. The price to buy and asset should be equal to the price to sell it and there should be no taxes, commissions or tariffs. These costs are again seen to distort markets - as well as being difficult to model formally.

It is interesting to consider at this point the meaning of distortion in this context. The model isn't distorted from reality, but inversely, reality is distorted and needs to be corrected. Actually existing social relations mean the world doesn't conform to theory, world therefore needs to be restructured in line with the "correct", predictable framework. No specific evidence is required to prove that a market distortion is intrinsically problematic, its very nature as a deviation from the model is itself enough to prove the existence of a problem in need of intervention.5

The theoretical framework requires costs to be at least negligible, if not non-existent. Tariffs, taxes and commissions need to be eliminated not simply, as is often implicitly or explicitly thought, because they take a cut of your profits, but because their existence is inherently disruptive to the economic models.

Of course in reality very little attempt is made to get rid of commissions, which act a direct gain for the industry, extracting rent from elsewhere in the economy, but we do see attempts to reduce taxes, tariffs, capital controls and so on. To see opposition to these as simply motivated by greed, by a desire to protect immediate profits is mistaken and misunderstands the underlying foundations of radical neo-liberal political economy. People genuinely believe their ideology.

Temporality and Dynamics

If I know the exchange rate between two currencies A and B, and between currency B and a third currency C, the no arbitrage condition fixed the value of the third exchange rate between currencies A and C. If I change a quantity of pounds into dollars, then from dollars into euros the exchange rates must be such that changing currency a final time back into pounds from euros returns me the same quantity with which I began.

This equality only holds if all three transactions are completed instantaneously (and if there are no costs to the exchanges). In reality transactions take some finite time to execute, within which fluctuations can and will occur. Prices of assets, interest rates, exchange rates and so on fluctuate constantly. As some shock in the real world occurs or as supply and demand of particular commodities alter, information takes time to propagate and prices to adjust. The faster this happens the closer we stay to equilibrium and the closer we expect to stay to models built to maintain these relationships at all points.

Ironically, then, in order to have a system which can be described by fixed, static relationships, which we can model with as little concern for dynamics as possible, we require the real system to be as interconnected and fast-moving as possible. Markets need to operate fluidly, without frictions or viscosity. They must be liquid and complete. That is, there must be someone always-already there to take the other end of any trade, with assets they can immediately and simply exchange. We have to financialise and securitise every market, marketise every commodity, commodify every material good and social relationship.

Yet these systems that flow so quickly and so freely are in danger of overshooting, completely connected markets can generate feedback and loops which in turn create unexpected emergent behaviour not visible in the simple aspatial and atemporal models. The speed of transaction and connectedness do not instantaneously equilibrate us back into a neo-classical framework but generate new distortions, cycles or even chaos. And so in increasing the autonomous flow of capital, directed by high frequency trading algorithms designed to expect static relationships, the markets create flash crashes, sudden shocks that shouldn't exist and whose effects feed back into both reality and model6 . The world is forced to conform to the assumptions of economic models, and in so doing reveals further flaws, themselves yet more impetus for further restructuring.

Speculation

Within this paradigm of complete liquidity we find framework with which we can analyse and understand speculative trading.

Consider a business which requires an ongoing supply of some material commodity as part of its operation: a car manufacturer who needs aluminium, a cafe chain that needs coffee, trains that need oil, say. The company needs to buy a certain quantity each month. Prices at the consumer end are sticky, it is difficult to vary them continuously higher and lower each month, but prices of raw materials at the wholesale level do fluctuate, in some cases quite substantially. A number of options exist which allow the company to hedge this risk, paying, on average, a small amount of money to reduce their future uncertainty.

The company could buy futures in the commodity, that is an agreement to buy some quantity at some set point in the future as a price fixed today. Their risk is then totally eliminated, future costs are exactly known and can more easily be planned around. Alternatively, one can buy an option. This is a contract which says you may, but do not have to, buy in the future at a set price. If the real prices is higher than the option price you exercise the option and save money, if it is lower you simply buy in the market as usual and allow the option to lapse. Essentially the option fixes a maximum price. Uncertainty is not totally eliminated but is reduced compared to the absolute vagaries of the market.7

The ability to buy derivatives does not lie only with those for whom a natural uncertainty in material production exists, however. Anyone can buy options as a purely speculative action, watching for opportunities to buy at an option price below the market and immediately sell, profiting from the difference. In practice the material assets underlying the trades are never affected, the owner of the options never has to physically take possession. The option itself assumes the value of the difference between the price it allows its owner to buy at and the real price open to everyone else. The option becomes an asset and can be directly sold. Capital flows completely autonomised from production. This purely speculative action, which has no direct bearing on physical distribution and allocation of resources is often seen as particularly parasitical on the real economy.

Within the logic of neo-liberalism however, these trades are seen as increasing liquidity of the derivative market. This increased liquidity should allow the market to discover the 'correct' price. Market efficiency tells us that the price people are willing to pay corresponds identically to the fundamental value of the asset. The price of derivatives then tells us what the market thinks about the future performance of the real economy. Though finance capital can now flow autonomously, where it flows is seen as indicative of real action. Reflexively capital then alters the real economy, changes in which then return to the movements of the simulacra of the market. Cause and effect are reversed, capital moves by itself and the real world adjusts in response.

To the criticism that traders are manipulating markets rather than feeding into a process of agglomeration of diverse opinions from which a democratic common view emerges, the solution offered is to increase financialisation. Markets with more participants, more liquidity, more capital, will be harder to manipulate, respective of a broader swathe of opinion. Yet this only deepens the problem, making finance and those best able to control it even more dominant.

The alternative, to leave speculative trading and focus on fundamentals is, however, also difficult to sustain. As profits generally fall, growth stutters and yield proves hard to find, capital has little choice but to concentrate wherever returns can be found. If hedge funds begin to cut into a company's profits, manipulating prices and extracting rents, competition forces them to engage directly with financialisation and attempt to claw back interest and rent from elsewhere. The process of financialisation is one which moves hand-in-hand with globalised production and not one which can be simply disposed of.

As Cristian Marazzi writes, "financialization is the other side of the post-Fordist capitalism coin; it is its “adequate and perverse” form... finance permeates from the beginning to the end the circulation of capital. Every productive act and every act of consumption is directly or indirectly tied to finance."8

In order to escape, the entire edifice of global capitalism needs to be deconstructed, reform within the system is not simply possible.

• 1For a good mainstream analysis of the virtues and flaws of the EMH see, for example, John Kay's the long and the short of it.
• 2Note however that Milton Friedman, for one, thought that insider trading should be allowed on the basis that it will increase the speed with which information reaches the market. Moreover, the trades themselves should not even need to be made public, the pressure put on prices being sufficient to convey the meaning. Information and prices seen as so synonymous that even the existence of new information about the real economy is not required so long prices are public.
• 3http://mathbabe.org/2012/12/29/open-data-is-not-a-panacea/
• 4http://by-strategy.tumblr.com/post/30283200728/julian-assange-also-neoliberal-utopian
• 5 I deliberately say intervention, as neo-liberalism is in fact not a doctrine of non-intervention, but in fact requires the intervention of the state - see here by-strategy again for readings on the history of neo-liberalism and the formula of "strong-state, free market". http://by-strategy.tumblr.com/post/33794260004/on-early-neoliberalism
• 6http://withoutpity.tumblr.com/post/35231162719/on-automation-and-algorithms
• 7The language of eliminating uncertainty is in fact problematic; risk is not really removed in any case, only moved. Risk can be diversified and spread, but the movements of oil prices, the discovery of new resources, the weather, consumer sentiment and so on are fundamentally uncertain, and no amount of financial complexity can alter that fact. Pooling risk can make it more manageable - that is the basis for insurance - or it can change a larger number of smaller isolated crises into a singular systematic one. In any event the risk must re-materialise somewhere to someone's detriment.
• 8The Violence of Financial Capitalism, Cristian Marazzi