Sunday, August 26, 2007

Sub-prime mortgages, risk, and causes of financial crises

The sub-prime mortgage market is the current financial "crisis" in the United States. Comparing sub-prime mortgages today with the S&L crisis of 30 years ago provides interesting parallels and shows how the different mortgage market today has resulted in a different problem than 30 years ago.

Those under 45 or so won't remember Savings and Loans. Back in the 1960's and early 1970's there were several types of financial institutions. One which has not survived is the "Savings and Loan." A Savings and Loan could only accept money in savings accounts and only loaned money for house mortgages (perhaps also home improvement loans, I forget the exact rules). Savings and Loans could not provide checking accounts, auto loans, etc. which banks offered. On the other hand, S&Ls could have many branch locations, while banks were limited to a single office.

During this time most home mortgages were provided by S&Ls. The S&L would charge 6% and later 8% for mortgages, while offereing 3% (later 5%) for savings. Maximum interest rates were regulated by the government, limiting the rate charged for mortgages or paid for savings.

The S&L crisis started in the mid 1970's but hit its climax in the early 1980s with scandals about shady deals and bankrupt S&Ls. While many people remember the scandals of the 1980s, fewer remember the original causes in the 1970s.

To explain the S&L crisis, one needs to look at way the system worked and how it reacted to the high inflation and interest rates of the 1970s. Prior to about 1975 interest rates in the U.S. were fairly stable and regulated. As mentioned above, S&Ls loaned money at 6%-8% for home mortgages. Adjustable rate mortgages were illegal, and most mortgages were for 30 years. They were also "assumable", meaning that a buyer of the home could take over the payments (and balance) of the existing mortgage.

Another factor was that the secondary mortgage market did not exist. S&Ls and banks held their mortgages the full term of the loan, getting their monthly payments. Today's practice of bundling mortgages and selling them developed later in the 1980s and 1990s.

In the mid 1970s inflation started rising, peaking at over 13% in 1980. Prior to 1974 the inflation rate rarely exceeded 3-4%. With inflation rising, many people were not satisfied with the 5% paid on savings accounts (limited by law). They started moving money into higher yield investments and out of the traditional banking system. As money moved out of banks and S&Ls, they lobbied for higher interest rates, and the "certificate of deposit" was created. CDs could pay higher interest but had minimum balance and term restrictions. CDs stemmed the loss of money but increased the costs to banks and S&Ls. Remember that banks cannot create money out of thin air. If they loan money they need deposits to back up that money (yes, they sometimes only need deposits for a percentage of their loan balance, but they still need the deposits). The increased costs of higher interest paid on savings could only be made up by increasing loan rates. Banks were in fairly good shape -- most bank loans were for a few years (car loans were typically 3 year) so the bank could get income from increased rates fairly soon. Savings and Loans, however, were in a bind.

Savings and Loans had their money tied up in 30 year mortgages. While the typical American would move every 5-10 years, assumable mortgages meant that the low 6% of many mortgages had years and years to run. Savings and Loans could write new mortgages, but the new income could not make up for the higher interest paid for savings. Thus, by the late 1970s S&Ls were paying higher interest on savings than they received for mortgages. Since S&Ls held their loans until paid off, they were in a financial bind.

At this time S&Ls were given more freedom to do other things. They could offer checking accounts and begin to loan money and invest in other areas. However, this added flexibility couldn't solve the underlying problem and S&Ls proceeded to merge, fail, and in some cases commit fraud. By 1990, very few S&Ls remained, and those which did were called "Savings Banks" and were much closer to standard commercial banks.

Today we won't get an S&L crisis of this type. Since the S&L crisis the secondary mortgage market has evolved, and most mortgages are packaged and sold to investors. This avoids the S&L crisis because the buyers of mortgage backed investments pay up front and receive interest over the life of the mortgages. If mortgages are paid off early (due to falling rates) the mortgage backed investment is paid off early. If interest rates go up, the investor still gets the same amount from a mortgage backed investment (if backed by fixed rate loans). Assumable mortgages are also a thing of the past, so with the mobility of the US populations most mortgages probably exist less than 10 years, with the home owner moving or refinancing within that time.

With the changes in the market, the problem of long term loans financed by short term deposits no longer exists.

However, the new mortgage market has created its own distortions which reveal a new problem in the banking system. Today mortgages are originated by banks or mortgage brokers. These are then sold into the investment market after which the originating bank or broker has no part in rest of the life of the mortgage loan. Thus, mortgage originators are most concerned with getting a loan approved so they can receive their commission, a lump sum when the loan is sold.
This shows the source of the current sub-prime mortgage problem. Lenders have been becoming more and more lax over time. In the case of mortgages, there is intense competition to get a loan. I personally would receive several letters each WEEK offering to refinance my mortgage (I still average about 1 a day for a new credit card). Those who did not qualify for a fixed rate loan could get an adjustable loan at a lower rate. In addition, loan standards are lower. Traditional mortgage loans were for 80% of the home's value. FHA would allow up to 95% or 97% (with mortgage insurance). Today lenders offer up to 125% of a home's value, especially in shorter term home equity loans.

These new markets worked very well so long as home prices were rising and interest rates remained low. However, interest rates are rising. Because rates are rising, home prices are stalling (since part of the rise in home prices the last 15 years has been because a given mortgage payment could pay for more home as rates fell). With stalling prices and rising rates on adjustable mortgages, people who could barely afford their home are in trouble.

We now have a situation with rising default rates. This has severalcauses. One is people not realizing that their mortgage payment would rise and not being able to afford the rising rate.
Another cause is liberal lending practices by lenders who a more concerned with the immediate return of originating a loan than the long term prospects of repayment. A final cause is investors in mortgage backed securities not realizing how risky their investments are.

Tying this rather rambling discussion together, the S&L crisis of the 1980s was caused because S&Ls (by government regulation) had to make and hold onto long term loans based on short term deposits. This could not stand up to a rise in interest rates. Today's secondary mortgage market avoids the S&L type crisis but creates its own crisis -- liberal lending practices resulting in excessive foreclosures. In each case the "rules of the game" have created the environment. In the past, regulated rates that could not respond to high inflation. Today a system which allows excessively liberal lending along with a belief by many homeowners that housing prices would continue to rise indefinitely.

Unfortunately, the banking industry has adopted excessively liberal lending practices as a norm. Banks come close to forcing credit cards and credit card debt on consumers. The constant offers and credit line increases has been a minor problem for years and will become worse as time goes on, especially if interest rates continue to rise.

Wednesday, May 9, 2007

Obama and fuel economy

Barack Obama is criticizing U.S. auto makers for poor fuel economy, and claiming that Japan and China average much higher (about 45mpg rather than the US 29mpg). While this speech makes good politics, and Senator Obama will get praise for giving it in Detroit, it ignores a few unpleasant facts.

Yes, fuel economy is higher in some other countries. That's because the cars are all much smaller than in the U.S. When we look at cars sold in the U.S. by domestic and foreign auto makers, all manufacturers average about 27-33mpg according to the NHTSA web site (http://www.nhtsa.dot.gov/cars/rules/cafe/FuelEconUpdates/2003/index.htm
). The overall average is 29mpg, with the domestic makers around the average. The fact is that there isn't very much variation in the fleet economy (not counting companies like Ferrari at 14mpg).

Japanese cars may average 45mpg in Japan, but the Japanese don't sell those cars in the U.S. Instead, they sell the large cars that US consumers will buy. I was talking to somebody a couple days ago who was driving his "economy" car, a Lexus (Japanese make) which got 30mpg highway (less city) as opposed to his Lexus SUV that averages about 15mpg.

There is nothing stopping Japanese makers from selling small high mileage cars in the U.S. But a simple look at the Honda Civic shows the trend. From 1973 until today, the Civic grew from 140 inches long to 177 inches long. Curb weight went from 1500lbs to about 2900lbs. Fuel economy has dropped, the original getting 40mpg highway, the current one 30mpg. I can't find 1973 overall fuel economy for the Civic, but today's model is in the mid 20's.

Certainly the hybrid Civic, like the Prius, provides much higher mileage. And Detroit has perhaps made a mistake in emphasizing ethanol fuel instead of hybrids, but we're still left with the question of fuel economy in the United Stetes. Small cars don't sell.

Airport security

Looking at airport security, it appears that if there is another major incident, we'll be back to the same complaints as after 9/11. After 9/11 there were complaints about poorly paid, poorly educated contract workers doing the security checks. Passing through an airport checkpoint yesterday, the ID check was not being done by a TSA officer, instead an employee of a private firm was checking security. And looking at the TSA employees, I'm not sure how much better the workers are.

The irony is that according to the 9/11 commission report, several 9/11 hijackers were flagged by airport security. However, the only response at the time was to search their luggage. The system wasn't setup to stop suicide attacks. If it were, the much maligned security company employees would have at minimum reduced the number of hijackers on the planes.

However, given the higher awareness of commercial aircraft security, the next attack will most likely come from some other direction.

Friday, February 9, 2007

Starbucks at the airport

I noticed something interesting waiting for a flight out of San Diego the other day. There was the usual food kiosk selling sandwiches, pizza, chips, drinks, etc. It had the usual inflated airport prices ($2.25 for a $.75 bag of chips, etc).

Around the corner was a Starbucks. The prices were about the same as the prices at the local Starbucks. No noticable markup for the airport.

Is Starbucks being nice or the other food seller grabbing monopoly profits?

Tuesday, February 6, 2007

rent, lies, and statistics

A news story yesterday said "Renters will dig deeper in 2007" (USA Today headline). While the story speaks of higher rents, and they appear to be rising, the story also shows how statistics can be used to mislead and possibly create a story where there really is none.

The story states that rents will rise 5% in 2007. It also says that they will have risen 14% over the 3 years from the end of 2004 until the end of 2007. Note that a 14% rise is around 4-5% per year.

Apparently to show how this will hurt people, the story goes on to say that wages will rise "4%, adjusted for inflation" over a similar period.

Notice the first impression a reader will get -- 14% rent increase, 4% wage increase. Yet that's not what it says. The wage increase is adjusted for inflation, the rent increase apparently is not. Looking at inflation numbers -- the consumer price index, inflation in 2005 and 2006 was 3.4% and 2.5%. Assuming 3% inflation in 2007 (about where it's been the last 10 years), the numbers total to about 9% inflation for 3 years.

Now compare the numbers. The wage increase is 9% (inflation) plus 4% (real increase), or about 13%. Rents will rise 14%. A 1% difference, not very significant.

So the news story is actually a non-story. Rather than renters digging deeper, it appears that for the last 3 years rents have roughly matched in inflation.

This provides a perfect example of the maxim, "there are lies, damn lies, and statistics".