The Anatomy of a Credit Bubble
Financial markets represent the collective result of individual actions. To fully understand how our current housing bubble was inflated, one needs to understand how the actions of the individual market participants impacted house prices. In my last analysis post, Your Buyer’s Loan Terms, I discussed future interest rates and debt-to-income ratios and their impact on future housing prices. In that post, I made a blanket assumption that interest-only and negative amortization loans will simply not be available in the future. It is a debatable assumption. In this post, I want to show more clearly how these two loan types created this bubble and why I believe they will not be available in the future. In short, I will describe the anatomy of a credit bubble.
To illustrate how this loosening and tightening of credit creates housing market bubbles, I will examine the last two bubbles similarities and differences. I will demonstrate how the bubble from 1987 to 1990 was very similar to our current bubble from 2001 to 2004. The last two years of our bubble, 2005 and 2006, were uncharted territory created by “innovation” in the lending industry.
How People Buy
When people decide they want to buy a house, they figure out how much they can afford and then go find something they want in their price range. For most people, what they can “afford” depends almost entirely upon how much a lender is willing to loan them. In the past, lenders would apply debt-to-income ratios and other affordability criteria to determine how much they were willing to loan. Buyers were generally limited in how much they could borrow because lenders were wise enough not to loan borrowers so much that they might default.
Buyers / borrowers behave much like drug addicts — they will borrow all the money a lender will loan them whether it is good for them or not. Most are not wise to the differences between the various loan types, and they have limited understanding on the risks they are taking on. This financially irresponsible borrower behavior is particularly bad here in California due to Southern California’s Cultural Pathology. If you need a primer on the various loan types, start with Financially Conservative Home Financing or Your Buyer’s Loan Terms.
Comparing the Bubbles
The circumstances during each bubble was different. Prices and wages were lower in the last bubble, interest rates were higher, the economies were different, etc. What is the same is the evaluation of personal circumstances each buyer goes through when contemplating a purchase. The cumulative impact of these decisions in represented in the debt-to-income ratios — how much each household pays to borrow versus how much they make. Comparing the trends in debt-to-income ratios provides a great tool for seeing how this bubble compared to the last one.
The chart above shows the historic debt-to-income ratios for California, Orange County and Irvine from 1986 to 2006. It is calculated based on historic interest rates, median home prices and median incomes. The last bubble is pretty obvious. In 1987, 1988 and 1989 people believed they would be “priced out forever,” so they bought in a fear frenzy. Mostly people stretched with conventional mortgages, but interest-only was used, and helped propel the bubble to a high level of unaffordability. Basically, prices couldn’t get pushed up any higher because lenders would not loan any more.
The Affordability Limit
When affordability limits are reached, prices must fall. This is caused by two related phenomenons:
- People will stretch to buy an asset that is appreciating; when appreciation stops, so does the stretching to buy. During the bubble rally, rising prices justifies paying too much because you obtain a return on your investment. Once prices quit rising due to the affordability limit (lenders won’t loan more money), there is no justification for the high prices, and people quit buying. The lack of buying causes volume to wither and inventories to spike; prices start coming down as sellers are forced to sell.
- Banks will not loan large percentages of the value for a depreciating asset, nor will they allow borrowers to utilize high debt-to-income ratios. Banks don’t like to lose money. Banks used to demand 20% down payments to give them a cushion if values dropped. Banks used to limit debt-to-income ratios to 28% to make sure the borrowers could afford to pay them back. The only assurance banks have for getting their money back is the value of the collateral (house). If house values start declining, banks want even more cushion to protect their investment. The era of 100% financing and 60% DTI is gone because houses stopped going up in value. As prices start to decline, down payment requirements will continue to rise and DTI will continue to fall which in turn reduces the number of available buyers which makes prices drop even further: a downward spiral.
What starts as buyers being unable and unwilling to buy turns into a downward spiral of tightening credit. This continues unabated until 20% down payments are the norm, and debt-to-income ratios fall back to their historically “safe” levels for banks of 28%. This is exactly what occurred from 1990-1997, and What is Past is Prologue.
Cheating on Affordability
Despite what the affordability charts show. People do not really make payments which are 62% of their gross pay. They cheat. They do this by utilizing risky financing options including interest-only and negative amortization.
Interest-only loans artificially “adds” affordability to the market because it allows for larger sums of money to be borrowed with lower payments. For example:
The median income in Irvine is $83,891. Applying a 28% DTI leaves a payment of $1,957. At current interest rates, a payment of $1,957 on a fixed-rate 30-year mortgage at 6.4% would finance $312,866. This same $1,957 payment on a 5-year ARM at 5.6% would finance $419,454. As you can see, the interest-only loan terms allows borrowers to increase their loans by 25% thus artificially increasing prices 25%.
2004’s False Top
On the DTI chart, notice the similarities between the periods 1987-1989 and 2001-2003. Both were rising DTI’s moving through the affordability zone pushing prices to the top of the range. If history had repeated itself, lenders would have become cautious in 2004 just as they did in 1990, and the market would have topped in 2004.
As you can see from the chart of available inventory above, 2004 would have indeed been the top. Inventory exploded, time-on-the-market went way up, and it looked like the party was over. It should have been; however, the lending industry “innovated” and came up with the negative amortization loan.
Negative Amortization Loans
When lenders “innovate” trouble is brewing. Banking has been around over 500 years. Everything has been tried at least once. Innovation in banking is a matter of trying something which has probably failed dozens of times before and hoping for a different outcome (the definition of insanity if you didn’t notice). It should not surprise anyone when the negative amortization experiment fails brilliantly.
From our example above, we can see how changing the terms from a conventional, fixed-rate mortgage with a 30 year term to an interest-only adjustable rate mortgage artificially increases home prices by 25%. Now look at what negative amortization does. In 2004, prices reached the limit imposed by interest-only. The only way to push prices higher would be to finance even larger sums with the same available payment. Option ARMs differ widely due to differences in their teaser rate, but for the sake of this calculation, I will assume a 3.75% teaser rate (I have seen them as low as 1%). The $1,957 payment finances $312,866 with a conventional mortgage, $419,454 with an interest-only mortgage, and a whopping $626,239 with negative amortization. In 2004, 2005 and 2006, people took out Option ARMs, bought the huge inventory spike from the summer of 2004, and sent prices into the stratosphere.
32% of loan originations in Orange County in 2006 were negative amortization (Option ARM).
Stop for a moment and ponder the math: the same payment now finances 100% more money. Is it any wonder our real estate market was 100% overvalued at the top? People purchasing with Option ARMs are buying at the rental equivalent value. From a financing perspective, the market is not overvalued. People are paying exactly what they should be paying. They are just doing it with loan terms which are going to destroy them — hence the term “suicide loan.”
These exotic financing terms are going away for one simple reason: foreclosures. People simply cannot make the payments when interest rates rise. If you look at the foreclosure rates in the early 90’s, you can see what happens when lenders get too loose with credit. Lenders overcooked the market then, and they got burned. You think they would have learned their lesson…
(One note on the foreclosures: defense industry layoffs are often blamed for the problems with the housing market. These layoffs came after the housing market was already in trouble. It slowed the recovery, but it was not the cause.)
Lenders faced high foreclosure rates in the early 90’s because they were too aggressive with their lending practices in the rally of the late 80’s: it was their own doing. As you can see from the above chart, the ultra-aggressive lending practices of the early 00’s are just now starting to show up in the foreclosures. Just as in the early 90’s, this is being caused by the past sins of the lenders: karma on grand scale. If does not take an expert to extrapolate from the chart above to see that foreclosures are going to shatter the old records set in the 90’s.
Price to Income Ratios
Just in case you still don’t believe there was a credit bubble. Examine the chart below of historic price-to-income ratios.
Very high price to income ratios signify borrowing large sums with small payments just as illustrated in the previous financing example. Ratio’s greater than 5 are considered very unaffordable and prone to high rates of default. The bubble of the early 90’s did not exceed 6 partly because interest rates were higher and partly because they did not use negative amortization.
Elimination of Exotic Financing
I have speculated that exotic financing is going to disappear. To be more accurate, exotic financing is going to become so expensive for borrowers as to render it practically useless. These loans will always be available, but the interest rate spreads will grow and the qualification standards will tighten to make them not usable. For example, in the heyday of negative amortization loans, lenders would qualify borrowers based only on the teaser rate payment without regard to whether or not they could afford the payment at reset. For more sophisticated borrowers, lenders allowed stated income or “liar loans.” Basically, a borrower would tell the lender how much they wanted to borrow, and the lender would fill out fraudulent paperwork showing the borrower was making enough money to afford the payment. This is amazingly irresponsible lending, but it was widespread. Now, lenders are requiring borrowers be able to actually afford the payments; of course, this makes many borrowers unable to obtain financing. That is credit tightening; that is how the downward spiral begins.
The next phase in the tightening of credit is an increase in interest rate spreads between prime loan terms and exotic loan terms. This is driven by the defaults and foreclosures. Mortgage rates for prime customers are very low because they rarely default. During the rally nobody was defaulting because prices were rising; people just sold if they got in trouble. This allowed banks to originates risky loans at very low interest rates because the loans didn’t look risky. Now that the market has stopped rising, the underlying risk is starting to show with dramatically increasing default rates. The true risks of these loans will become apparent over the next few years.
Banks have to make enough money on their good loans to pay for the losses on their bad loans and still make a profit. As exotic loans start showing very high default rates, banks have to start charging higher interest rates to cover the losses. Higher interest rates make for lower amounts of borrowing.
When banks realize the true risk associated with exotic financing terms, they may have to charge much more than they are charging for conventional loans. As you can see from the table above, if the interest rate is only 3% higher, the amount financed is 25% less. If the default rates are very high, no amount of interest rate spread can compensate the bank for the risk, and that loan program will be eliminated. This will be the fate of the negative amortization loan.
The 2006 vintage sub-prime negative amortization loans have already defaulted in record numbers. These loans are less than a year old. It is forecast that over 20% of these loans will default, and this is without a crashing housing market. If lenders have to make enough money on 4 loans to cover the loss on 1, interest rate spreads will be very high. If a negative amortization loan costs 13.75% rather than 3.75%, nobody will want it, and if lenders require borrowers to actually afford the 13.75% interest rate, nobody will qualify. Either way, negative amortization loans will die. The fate of stated income and interest-only loans may be no better.
When prices crash, defaults rates will increase for all borrower classes. Prime borrowers will not default at the high rates of sub-prime borrowers, but they will still default at rates higher than in the past; therefore, interest rates will rise for prime borrowers as well. The crash in house prices will cause all mortgage interest rates to rise.
What Happens Next?
Over the next several years, interest rates will rise to at least 8%, 20% down payments will become the norm, and debt-to-income ratios will fall back to their historically “safe” levels for banks of 28%. What will that do to prices?
At some point in the future, the market will bottom near the values above. How it gets there will depend on the number of foreclosures. If there are a great many foreclosures, it will happen quickly, if there are fewer, it could take longer. Either way, the median prices shown above will occur, it is just a matter of when. I have constructed two different scenarios as to how and when: Predictions for the Irvine Housing Market and How Bad Could Bad Get?
The conditions for the above disaster are already in place. Right now, we are in the lull before the storm, but the storm is coming; there isn’t much anybody can do about it.