Up unit vs down unit theory
More attention has been paid on the credit event recently by many sources and we have a lot more data available today. It is therefore even easier to analyze the state of the mortgage credit market than before. These two are no doubt the most significant pieces of information released in the last few months:
federalreserve.gov
Though just a guideline proposal, this document clearly outlined all the perils of toxic mortgage world.
firstamres.com
This is the best report for data.
I am going to try a different approach here. Basically, there are only so many moving parts. Some moving parts move up and some down. For each unit of part that moves down, one needs to move up just to stay in place. The equation is simple:
UU - DU = MDU
(UU = up units; DU = dn units; MDU = market direction units)
Here are the major but not all parts, in no particular order:
INCOME
INTEREST RATE
EQUITY APPRECIATION
CHANGE IN LOAN PRODUCTS AVAILABLE
CHANGE IN UNDERWRITING STANDARDS
Components such as income and underwriting standards can be controlled to some degree. Components such as interest rate and equity appreciation/depreciation are reactive to market conditions.
Looking back, we can see UU has been greater than DU by a huge margin, resulting in a very high positive MDU number. What is different this time vs previous cycles are the components that have time elements to them. As a result, up units appear to be much higher because some of the up units have embedded down units in them.
Churning of the subprime loans is one of the best example that I can think of. For each 2/28 loan that was originally originated, there are many up units. The artificial lowering of interest rate for many who otherwise should not be obtaining a loan created a chain reaction that all resulted in more up units.
Taking it a step further, actually quantifying these up and down units is not all that difficult. All you need is to establish a base. I will use 1% interest rate as the base and call that 1 unit. In other words, if rates (of any loan program) go up 1%, that is 1 DU (down unit) or go down 1%, that is 1 UU (un unit).
Now apply this 1% = 1 unit base to the 2/28 subprime loans. Without this program, a subprime borrower in the old days would be looking at paying 3%, 4% or more than a prime borrower. Let us just use 3%, then the 2/28 program has a 3 UU value directly. However, these 3 UUs are only good for 2 years and have to be returned or replaced at the end of the 2 years.
Again, you can quantify this quite easily.
For example, if a borrower's income rose during the 2 yrs in the amount needed to finance 3% increase in payments, then the net = zero.
For example, if market interest rate had come down 3% during the two years, then 3% increase in adjustment and 3% decrease in interest rate nets out to zero again.
If I am one of those big time wall street analysts with an army of support staff, I believe the data is available now to design a totally dynamic model. By plugging in each of the variables as they occur, the model should show how many loans may be impacted and the likely multiplier effects. Unfortunately, I have a staff of one (me) so I can only rely on approximation. So here is my model:
1% interest move = 1 UU (or DU) depends on up or down move.
Purchase = 10 DU (just to cover the transaction cost)
Refi = 1.5 DU (transaction cost either as pts or in rates)
1% appreciation (depreciation) = 1 UU (1 DU)
Income = UU or DU equivalent of 1% debt service
Income property cash flow = UU or DU equivalent of 1% debt service
For anyone who may still reading this, you can see it is possible to approximate the magnitude of the changes ahead. It is based on this model that I draw the following conclusions:
1. If real estate appreciation rate goes to zero, the MDU (market direction unit) will NOT be zero but rather a negative number.
2. The market under current conditions can only support a very small negative MDU number and for a very short period. That is because more DUs would be triggered automatically by 2/28s, then the 3/1 hybrids, then the 5/1 hybrids, then the 7/1 hybrids.
3. Negative MDU can reach a break point quite easily as borrowers in trouble exceed the critical mass level to trigger a catastrophic chain reaction. Of all the catalysts, financing is clearly the most likely fuse to set off the bomb. A LTCM type credit derivatives crisis today could easily shut down the mortgage finance market, without which the real estate market cannot function.
4. I see no policy, be it fiscal or monetary, that can serve as a safety net if this catastrophe occurs. Think about all the usual remedies such as lowering short term rates, tax incentives etc., all of them are very marginally available under current circumstances with minimal expected results. |
