60 Years' Worth of Financial Engineering: Is Any of It Positive Innovation?

by: Rortybomb

Felix responds to my previous post about 401(k)s and still believes his “thesis that financial innovation over the past couple of decades has been, on net, a bad thing.” Felix isn’t necessarily looking for a specific instrument that has been good but is making a more general indictment of the financialization of the economy, but I’m going to throw one out. I suppose this depends on where you draw the line, but I’ll suggest the Interest Rate Swap market as a piece of positive financial innovation from the 1980s. Many noise traders have blown up otherwise good portfolios with them no doubt, but on average I think they have been a win, for specific reasons.

As opposed to monstrosities like reverse convertibles, it is very clear what one is getting, and isn’t designed to trick people from their money. It is one item that adjusts one series of cash flows. We have good reason to believe the interest rate yield curve is rationally priced, and with a quick excel check you can figure out a fair value for an interest rate swap. Retail investors should never be allowed near one, of course, but for informed investors they aren’t trying to figure out the likelihood of events deep in the tail like a CDS, or something that is opaque, hidden, and not necessarily subject to arbitrage forces downward, like how housing will do in the next 5 years. They are easy to get out of, and they can actually hedge, or speculate on, a real risk easily. I’d like to see fewer school boards take large positions in this market, but on average I think it has been a win.

Is there anything to salvage in the last 60 years of financial innovation post-crisis? There are three waves of financial innovation:

1950s-1970, Portfolio Selection: This is your CAPM, mean-variance analysis, diversification arguments and index funds. This sets up future stock price movements to be the results of correlations, and where you get your heavy asset allocation statisticians. It is also why when you look at stock information, say at Wolfram Alpha, there’s a mess of correlation charts and data. Beta is more or less dead now, having been performed so hard by market participates that it has lost its ability to predict what it was intended, and is now replaced with factor models and momentum effects.

Asset allocation has taken it on the chin lately but I don’t think it will go away. I don’t have a lot of opinion on equity allocation, not having to deal with it directly, but I’d like to say that we are exposed to a lot of systematic risk in the market lately, and that the theory was exactly that you couldn’t diversify away from this aggregate market risk, just the idiosyncratic risk of individual firms.

1970s: Black-Scholes. Call and put option pricing. Binomial models to supplement Black-Scholes to non-constant volatility and correlation models. I think the bevy of call and put options available on exchanges have been a net win. It gives us more information on the evolution of prices in stock in a medium term period, and allows people to hedge, or speculate, on the evolution of a stock.

1980s-on: Here I’m mostly in agreement with Felix, though I think it is important to view the changes in the context of the larger change towards a capital markets banking system, and the instruments that were needed to start up this system and keep it running after the wave of late 1990s deregulation kicked in. I’m not sure if we can go back easily, though repealing a fair amount of the banking deregulation would be a start.