Thinking about a career in stock market trading? Start by learning how to swim, because the global financial centres of the future may be located in the middle of the ocean. According to this fascinating report by BBC science and technology reporter Jason Palmer, mid-ocean island chains may become the optimal location for financial hubs because high-frequency traders are seeking to exploit the speed of light.

High-frequency trading carried out by computers often depends on the varying prices of a financial instrument in two separate geographic stock markets. So exactly how fast the data travels – and how far it has to travel – can make the difference between a profit and a loss. As Palmer outlines in his report, Harvard physicist Alexander Wissner-Gross told a meeting of the American Physical Society this week that financial institutions are increasingly looking at ways to exploit this trick of the light. The only limit is the speed of light, and trades can now travel at nearly 90 per cent of this ultimate limit.

Indeed, one of the many people to have recognised the profit potential of high-frequency trading is Irish businessman Brian Conlon, last year’s winner of the Ernst & Young entrepreneur of the year award. His company, the Newry-based First Derivatives, started developing and selling its own software for “low-latency, high-volume” trading three years ago.

“If you think about highly liquid stock like IBM or Microsoft, the price is going to change maybe five, 10, 20 times a second,” he explained in an interview with Barry O’Halloran last September. “So if you’re going to make a decision like ‘I want to buy IBM when the price hits 100’, then you have to make that decision within microseconds or milliseconds. If you don’t get your order into the exchange, then the price will have changed and the price will be invalid. Similarly, if you look at any trading screen, they are ticking many times a second. If you’re doing a program trade you have to make that decision fast or else your price is out of date.”

First Derivatives’ software is programmed to make these decisions – buy IBM at 100 and sell it at 110 – without human intervention and the risk of all those nasty “fat finger” losses. Algorithmic trading such as this is driving the mushrooming volume in daily global financial trades – in New York alone, the number of transactions carried out every day has gone from five million a decade ago to over a billion today. As a result, Conlon’s company, which he started in a spare bedroom 15 years ago, is doing very nicely, thank you.

But why might a chunk of Wall Street eventually relocate to, say, the Azores? It’s because the latencies in global fibre-optic links – the time delay for a signal to shift it from one global financial centre to another – are lower in some locations for certain trades. Last year, Dr Wissner-Gross mapped the optimal points where financial transactions should originate in order to maximise the chances of buying low and selling high. The answer was midway between the two major financial hubs involved in the trade. This often meant the ideal location was at nippy high latitudes or mid-ocean.

Unsurprisingly, companies weren’t too keen on deploying floating data centres in places where there’s nothing but blue sea on the horizon – not even a Costa Coffee – so now Dr Wissner-Gross is helping firms work out which geographic stock markets they are best positioned to trade from their current location. But there’s still a clear physical advantage, and therefore competitive advantage, to moving, he maintains. Desert island trades, here we come. How does the Atlantic Stock Exchange or the WaveStoxx 1000 sound?

Just remember: if you can make money at 90 per cent the speed of light, you can lose it that fast too.