OPM = Other People's Money.
It's the general idea although the details are everything. Without going into yield differentials the idea is actually to be time independent and risk neutral. You make money by instantaneous fluctuations where market inefficiency causes sufficiently similar securities to diverge in price. As you know this is called arbitrage. It's the details of "sufficiently" that causes the problems and you have to have dissimilar vehicles to have a spread in the first place. When I used the term, "translation", I meant the intended zero expected return on the hedge becomes a stochastic variable of time. If you're short side is rising faster than your long side, you'll eventually have a margin call. That means your risk zero spread components develop price change as a function of time when you want no change as a function of time. If the short security changes each day more than the long security, you must make adjustments that supersede the pure arbitrage function. That is you add or subtract to longs or shorts based on translation rather than on fluctuation.
The width of the spread is caused by normal factors of market risk. A GNMA is perceived to be more risky than a T-Bond. T-Bonds change price faster than GNMAs, so it is possible to have negative transient spreads between those two. Arbitrage either assumes you will be in and out of the market quickly to avoid uncompensatable trend effects or that the exogenous factors effecting the components are either small or manageable. The latter is called risk arbitrage. You believe you can manage what happens over time. This is an ok assumption 99% of the time.
Th entire hedging industry could go to cover all their shorts or take off one leg only and it would have no lasting effect on prices of these components. It wouldn't have much of an instantaneous effect either. You might have an intraday spike up and spike down, but that would be it. Prices of anything reflect fundamental expectations, not market clearing deviations. This is an extremely important to understand for anyone involved in investment of any form. It is called elasticity of price change relative to marginal demand or supply. When the elasticity persists in time, you get a translating state, a trend.
Thus mortgage rates are independent of fools gambling. Mortgage rates will rise if the entire interest rate structure rises due to a rise in inflation. The spread will continue to widen if the Japanese continue to factor and the FED continues to pump. I believe that mortgage rates are bottoming and the T-bond is going to fall off a cliff. As soon as the exogenous factor slows, bang, you'll see the spread narrow very quickly as everyone including the other Japanese, start to unload their fat T-bonds in an inflating world. Sell the breakout because there won't be any looking back.
Merriweather was gambling that FED would do the right thing, the rational self-interest economy preserving thing and lean against the US propensity to inflate. It was a Leeson kind of gamble. If they could only stay the position. The fact that he went bust means big money for us by patiently waiting to use the tapas he generated. He was right about the circumstances of US inflation and it's going to cost we the people plenty. |
