Hyperbolic but not altogether inaccurate December 20, 2012 at 3:27 pm

Felix Salmon on the ICE purchase of NYSE Euronext:

Exchange mergers… aren’t actually boring at all: they’re an indication that the financial-services industry is desperately trying to protect the rents it can collect by means of consolidation. There are lots of stock exchanges, and none of them make much money. By contrast, there are relatively few derivatives exchanges, they tend not to compete directly with each other, they tend not to compete on price, and they’re wildly profitable.

One Response to “Hyperbolic but not altogether inaccurate”

  1. I’d say it’s definitely true. If I buy 100 shares of GOOG on exchange X, I can trivially go out and sell them at exchange Y, dark pool Z or even just bilaterally trade them with a buddy of mine. If I buy 10 contracts of the S&P index futures at the CME, the only way I can sell is to go back through the CME.

    If the CME charges higher fees I can go out and trade some other index future that closely mimics the E-mini S&P, but then there’s a whole host of strategic and operational issues. What’s the margin, how much liquidity is there, what’s the clearing counter-party risk, how well does the other index future actually hedge on my strategy horizon, etc.

    Stock exchanges have to compete on every single order, usually enforced mechanically by a smart order router. Future exchanges only have to compete at a much higher, fuzzier, longer-term, more locked-in level. At the end of the day whether you pay 1 bps or 0.1 bps in exchange fees matters to the smart order router, but not so much to the CFO.

    Of course the solution is simple, simply mandate that future contracts be fungible across major exchanges. But policy is moving in the opposite direction. Both in terms of centralizing derivatives clearing and discouraging the fragmented liquidity commonly found in equity markets.