There has recently been some question raised about diversification and subordination levels. Recent subordination levels are somewhat higher as a percentage than in 07, but the real issue is subordination based on what. The recent issue of Chase was based on real underwriting and real debt cover ratios, and not the make believe values and projected cash flows of the earlier vintages. To just compare level of subordination misses the point, which is what is the underwriting protocol used. At least for the moment it is good and it is being done at what most of us believe is the bottom of the market. If you have good debt cover based on historic cash flows and current values, then most of us believe these should be good loans. Obviously if we have a major double dip recession or if there is a major black swan event in the world which disrupts the economy, then things will change. There is a real argument that true AAA should have 30% subordination to be true AAA so that it is really bullet proof. There is a lot of validity to that and it something the market will have to determine. In my view, if it is to be truly AAA and if it is to withstand the inevitable underwriting deterioration that will happen over time as the market ramps up, then something materially higher than 15% should be the test of AAA. While 15% may work when cash flows and values are at the bottom and headed up, they will not be sufficient when cash flows and values rise materially and we are back into the high volume and looser underwriting period in a few years. There needs to be an examination of relative underwriting to where is the market. That is the major key to avoiding another crisis.

It is an absolute fact that as markets rise and volume increases, the need for higher subordination should also increase, as the risk of a default rises as markets improve off the bottom. After 45 years in this business, it is very clear that as markets go up, risk goes up faster, and lenders get fooled into thinking they can lend more, and they can loosen underwriting. It is really the opposite. The more property values rise, the more lenders ramp up the leverage, the more values are driven higher and the greater the risk of default. At the bottom, as we seem to be at now, is when leverage ratios can be higher. When recovery is well established is when leverage ratios should be reduced and more attention should be paid to historic cash flows and not projections, and subordination levels need to be raised. Risk and recovery of values are in direct correlation. The secret is figuring out when are we far enough off the bottom to begin to ratchet down the leverage. Competitive forces inevitably force bad underwriting and eventually we are in a new crisis. The real way to avoid this is not lots of new regulations, but to have some controls over underwriting, and that is the job of the rating agencies.