The issue these days is that there are few market areas where you can really buck the system. By ‘buck the system’, I mean find alternative trading strategies to really gain competitive advantage. The last real move in this direction was the creation of the credit derivatives market ten years ago by JPMorgan. Credit derivatives are now a $12 trillion market – about the size of America’s annual GDP – but even that is becoming a commodity.

There appear to be three profitable directions for trading strategies in the future. The first is to create much more complex trading strategies using technology tools to win. The second is to create much more complex products using technology tools to win. The third is to move towards simplicity and outsource the trading operation to those who can afford the complex technological capabilities to manage this on your behalf. All three rely on technology tools to win.

These strategies are not mutually exclusive but can be mutually inclusive, depending on your overall market objectives. Let’s look at what each of these strategies mean in practice.

The first focuses upon creating much more complex trading systems using algorithmic trading. Algorithmic trading has historically been the domain of engineers and scientists. By way of example, of the key players involved in creating the credit derivatives markets, most were either science or mathematics graduates. In the words of the leader of the JPMorgan credit derivatives team at the time: "I wanted to be a rocket scientist and ended up working on Wall Street". Actually, the real leader of the team - Peter Hancock now of Integrated Finance Ltd – was a science student at Oxford University who wanted to be an inventor...you get the drift.

These rocket scientists started the market towards more automated trading back in the late 1990’s using quantitative trading strategies and highly automated systems. This market began simplistically with buy-side hedge funds and sell-side brokers looking to find arbitrage and other complex plays in the equities and FX markets. For example, the original tools would have allowed an investor to find micro-opportunities in real-time and make trading wins of a reasonable size. For example, if the value of Microsoft (MFT) shares increase by more than 0.5% and the value of IBM (IBM) shares decrease by more than 0.5% at the same time, then buy MFT and sell IBM.

Today, the tools are being used more and more to create cross-asset class plays in all markets and are getting really clever. For example, if the value of MFT increase by more than 0.5% within a 30-minute window during which IBM shares decrease by more than 0.5%, then buy MFT and sell IBM whilst taking a futures option on IBM purchase and MFT sale for one month. In addition, if Oracle shares rise whilst SAP fall during the same 30-minute window, follow with Oracle buy and SAP sell with a hedged futures option. Finally, if these movements occur after a 3 point rise in the Nasdaq, then shift FX strategy from euro to US Dollars, with a sales option on the dollar.

In fact, you could add as many parameters as you wanted until you were happy you had played all the different criteria that might be required during that period. But it doesn’t end there.

I was at a recent conference where one key algorithmic firm discussed two new concepts of algo trading using the concepts of “TiVo playback” and “graphic equalisers”. What they were getting at is the idea of placing all your complicated ideas and strategies for trading into the system, and then seeing how they would have played out over the past 30-day trading cycle. To do this, the system stores all the trade and market data for the last month and then allows you to fast forward, reverse, pause and stop the market activities during that cycle in a style similar to using a TiVo or Sky+ player.

Now the incredibly clever bit. Whilst you are playing your TiVo market simulations, you can see what works and does not work as a trading strategy through a sort of graphic equaliser. The system, in other words, shows you green for good stuff and red for bad.

The more you find trading strategies that work, the more you store those in the good stuff and they become inbuilt into the software algorithms. Equally, if your strategies didn’t work, you highlight those and that also becomes inbuilt into the software algorithms as the bad stuff. The result is that the software learns what works and doesn’t work and builds that into future trading patterns.

The impact of this is that it provides asset managers the ability to test strategies and moves their trading advice away from their brokers and onto their systems. So, you may well ask, what’s the point of active trading when the software tools are getting so good that they will ultimately always know the best trades to make with perfect timing?

In other words, if we create a ‘perfect’ market where all trading is so sophisticated using algo tools that you can only keep up if you use algorithmic trading, then those without algo tools are dead meat.

Well, not necessarily. And this leads to the second phase of market development which, as mentioned, is complementary to algorithmic trading.

If everyone has algo tools that learn trading strategies and always strike perfect trades with perfect timing, then those in the human world have to go and find something else to do, such as more complex credit derivatives.

When the JPMorgan team created credit derivatives in the late 1990’s, they were looking for a way to create something that would be hard to copy. A bit like swaps, arbitrage and other futures and options, it was really a way to find another market.

Sounds easy, but if it was then every bank’s investment team would be winners. What the JPMorgan team succeeded in doing is creating something that was hard to copy because, by the time others knew what it was, they had taken the lion’s share of the market.

By way of a brief explanation. The way credit derivatives work is like an insurance policy for a bank’s credit risk. For example, the bank takes a basket of loans that mix $1 billion to Panama, $5 billion to Mexico and $2.5 billion to Columbia, and then offers the option on these loans defaulting as a credit derivative. The investor only has to pay on the derivative if the borrowers default. Meanwhile, if the loans are paid off, then the investor has made a packet of cash from the premiums the banks pays to offset their credit risk.

This has contributed to an acceleration in lending, and rising exposures. For example, of General Motors’ stated losses of over $10 billion last year, over $2 billion were hidden in the finance arm GMAC as a result of an exposure to over $200 billion in credit derivatives. Similarly, in March 2006, Austria's Bawag Bank announced that it had used offshore companies to mask almost €1 billion of losses for similar investments that almost turned the bank insolvent in 2000.

Even with such risks, banks love credit derivatives because it takes possibilities of substantial loan losses off their balance sheet and onto someone else’s, such as General Motors. That is why it is a market that has ballooned over the past decade from nothing to $12 trillion – bigger than the USA’s GDP – but is also now a market that has commoditised.

Like any great trading initiative – swaps, structured finance, arbitrage – the more popular the initiative, the more money it makes, the more others want to get in on the action, the more competitors start to buy the people who make the money, the more commoditised the market becomes. After ten years, credit derivatives are reaching market maturity and so the next big play is now being looked for. That big play is likely to be around complex cross-asset class trading strategies that combine equities, bonds, options, derivatives and FX.

There are huge dangers in this of course, and successful execution demands deep technological know-how. However, the only way in which markets can operate is to seek out and trade off risks against rewards, and the greater the risks, the greater the rewards.

Perhaps the approach is best summed up from a line out of Michael Lewis’s book Liar’s Poker, which described life in Salomon Brothers in the late 1980’s. In the book, Salomon’s head of law, Donald Feuerstein, took great delight in finding “chinks in the regulator’s armour” which he could use in ways to buck the markets and make greater returns. In other words, the more markets are commoditised and regulated, the more complex and sophisticated the products and strategies which need to be used to generate returns. In today’s world of turbo-charged algorithmic cross-asset class trading strategies, that can only be delivered through massive investments in systems.

Which brings us to our third strategy.

The technological turbo-charging of the investment markets has meant that you have to have a huge cheque book and pen to be a player these days. Those firms that cannot aford to enter the game as a player in their own right will end up giving their trading to a third party to manage. After all, if you can outsource your customer service centre to Mumbai, why not do the same with your investment banking division if you can’t afford to be a top player?

Therefore, there will be a range of firms who find algo trading and derivatives beyond their reach. These firms will ditch their trading desks and get someone else to do it for them. By way of example, Acme Trading Desk Services Ltd offers outsourced trading desk facilities and soon finds that Acme supports Hedge Fund A on a Monday, Fund Manager B on a Tuesday, Investment Manager C on a Wednesday, Insurance Group D on a Thursday and maybe takes Friday off because the systems need a bit of an overhaul.

This vision is not too far away, and is in fact becoming even more likely thanks to forthcoming legislation such as MiFID in Europe and CP176 from the UK’s Financial Services Authority that relates to soft commissions and bundled brokerage.

The combination of regulatory drivers, technology chargers and market makers’ margins, will deliver a future world in financial trading that will only be played by those who have the depth of pocket and technological know-how to be a player.

Chris Skinner is a director of TowerGroup and founder of Balatro.
Web links: www.towergroup.com and www.balatroltd.com
Author's email: Chris Skinner