I love the very idea of arbitrage — ‘risk-free money’. Anywhere else, this would be called a scam and the would be overtones of theft. But in the world of finance, where it is understood that markets are (mostly) efficient and that there can be no such thing as a free lunch (mostly), nobody seems to mind if you happen to find a lunch that’s free because they are safe in the knowledge that it won’t be free for long. Interestingly, some of the thought leaders in capital markets theory do describe a number of free lunches. But because they know that free lunches are not possible, they describe it as reward for extra risk taken.
I spent a year or two working with a treasury derivatives trader building treasury derivative arbitrage models. These were humungous Excel sheets loaded to the gills with add-in functions, some custom C code, and pulling in realtime data, detecting small differences in pricing between synthetic and real securities. An example: if you take a euro-swiss franc swap and a swiss franc-dollar swap you can create synthetic euro-dollar swap. This has (just about) all the characteristics of a real euro-dollar swap but it might have slightly different cost. Where the difference exists, it’ll be tiny but if you have a billion euros on your books overnight, it may be something you can take advantage of ‘for free’. Most of this simple type of arbitrage really does not exist in the markets any more, but when we used more complex treasury derivatives, and worked out all the permutations across a basket of ten currencies, there were more opportunities.
I still remain interested in financial arbitrage, and on the lookout for simpler ways of consistently making money — but that’s for another post.