On ‘Spoofing’

On 21 April​ , the financial trader Navinder Singh Sarao was arrested in West London. The US authorities are seeking to extradite him to stand trial in Illinois after charges were issued against him by the US Department of Justice. The DoJ alleges he was in the habit of ‘spoofing’ futures markets, by entering orders without genuinely intending to buy or sell, and that this contributed to his trading profits of about $40 million between 2010 and 2014. He is said to have done all this from his home, a semi-detached house on a residential street in Hounslow. Spoofing isn’t new – in fact it is quite common – and although it was explicitly outlawed in the financial reform laws introduced in the US in 2010, legislation was already in place that could have been used to prosecute spoofers. Sarao is, however, so far as I can tell, only the second person to face criminal charges in any jurisdiction for spoofing. The first was a New Jersey futures trader, Michael Coscia, who was indicted in October last year.

The trading of shares and futures is now anonymous and electronic. You no longer stand in a crowded trading pit, shouting and gesticulating. You sit quietly at your computer – often it’s doing the trading for you – and enter bids to buy or offers to sell. Those bids and offers are transmitted electronically to the exchange’s computer system, which maintains what traders call the ‘order book’. This is essentially a list of all the bids to buy or offers to sell a particular stock or other financial asset that have been neither executed nor cancelled. Bids and offers get executed when the exchange’s system finds a bid to buy and an offer to sell at the same price.

Experienced human traders and sophisticated computer-trading algorithms pay close attention to the order book, because it gives a sense of supply and demand. If, for example, the order book contains a lot more offers to sell than bids to buy, then supply seems to exceed demand, and it looks likely that prices are about to fall. Anyone trying to sell will often then reduce their asking prices to attract buyers, so prices do indeed fall. The salience of the order book gives spoofers, human or algorithmic, their opportunity. Start by selling, let’s say, five S&P 500 index futures at the current market price. (These are contracts tied to the level of the S&P 500 share index. Five of them – a small trade by futures-market standards – are the equivalent of shares worth around $500,000.) Then enter one or more very large offers to sell, but at prices slightly higher than the market price, so they won’t be executed. That will cause supply to look as if it exceeds demand, so prices will probably fall a little. Now cancel your big spoofing offers, which have done their job, and simultaneously buy five futures for less than you sold them for a few seconds ago. You’ve made a profit: most likely not huge, but repeatable.

Of course spoofing isn’t as easy as I’ve made it sound. You need steady nerves, intense concentration and quick reflexes to do it yourself with keyboard and mouse. Doing it by computer means either writing a program or modifying a commercially available one, though you can pay someone to do that for you. There’s also always the risk that the big bids or offers you intend to cancel are executed before you’re able to remove them from the order book, in which case you could be left nursing a large loss.

What you didn’t have to fear, until very recently, was going to jail. You might well have received the kind of message that the Chicago Mercantile Exchange sent Sarao, reminding him that orders ‘are expected to be entered in good faith for the purpose of executing bona fide transactions’. If you repeatedly received these warnings and ignored them, an exchange or a regulator might have taken administrative action against you. That could mean a financial penalty, and perhaps even loss of one’s licence to trade, but it wasn’t a criminal matter. Why has it become so?

There have been suggestions of a connection between Sarao’s trading and the wild market convulsions of 6 May 2010 that have become known as the ‘flash crash’. The DoJ alleges that between 11.17 a.m. and 1.40 p.m. (Chicago time) a fifth or more of all the orders to sell S&P 500 index futures were Sarao’s. It would be remarkable if a single trader, operating on his own, could constitute so large a part of one of the world’s most important markets. However, the DoJ also says that Sarao cancelled his main sell orders at 1.40.12.553 p.m., although he stayed active on a smaller scale over the next five minutes. The millisecond time stamp matters, because that was around a minute before the worst of the big plunge in prices. Wholesale cancellation of sell orders would have lessened, not increased, the chances of a plunge, and so at the very least other forces must have been at work.

Rather, the explanation of the criminalisation of spoofing is that attitudes to it have changed. Spoofing was harder to pull off in the public arena of a Chicago trading pit, but if you did it successfully you weren’t punished, even informally. As a former pit trader told me, ‘It was just thought of as “boys will be boys” and [as in a] poker game, bluffing was not something that was thought of as wrong, immoral or illegal. In fact it was in some ways admired.’

That attitude survived the early years of electronic trading. Jakob Arnoldi of Aarhus University interviewed an exchange’s head of surveillance for an article in the journal Theory, Culture & Society. ‘Five years ago,’ Arnoldi was told, ‘everybody would say: where’s the problem? Why are you programming these stupid algos [algorithms that can easily be spoofed]? It’s your fault. But now it is market manipulation.’ Arnoldi concludes that regulators are trying to make the world safe for algorithms. That’s part of the story, but it’s also important to bear in mind that few markets have ever been places of untrammelled competition; they nearly always contain at least fragments of moral order. That was certainly true of Chicago’s trading pits. Spoofing might have been acceptable, but if you reneged even once on a deal with another pit trader (deals were either verbal or agreed by eye contact and hand signal, and thus not realistically enforceable in law), you risked being informally shut out from trading by everyone else, perhaps for good.

Instead of face-to-face markets based on direct human interaction, we now have something more like the ‘perfect markets’ of some economists’ imaginations: individualistic, atomistic, anonymous. Moral questions, however, have been displaced, not eliminated. There is a growing sense that the electronic order book, precisely because it is now the heart of so many financial markets, has to be pure. All orders must be ‘entered in good faith’; the poker player has no place. But formal regulation and the criminal law may be insufficient to police matters of this kind. Good faith is a matter of intent, and intent is hard to prove – that must be one reason for the dearth of prosecutions. Behaviour in markets is subtle, nuanced and complicated. It’s not possible exhaustively to define the boundary beyond which clever, flexible strategy becomes demonstrable bad faith, and beyond which bad faith becomes crime. It’s a deep and difficult issue, and calls to mind an old Marxist word: contradiction.