The Credit Bubble
The Great Housing Bubble was not really about housing; it was about credit. Most financial bubbles are the result of an expansion of credit, and the Great Housing Bubble was no exception. Housing just happened to be the asset class into which this capital flowed. It could have been stocks or commodities just as easily, and if the government gets too aggressive in its actions to prevent a collapse in housing prices, the liquidity intended to prop up real estate prices will likely flow into some other asset class creating yet another asset price bubble.
The root causes of the Great Housing Bubble can be traced back to four interrelated factors:
- Separation of origination, servicing, and portfolio holding in the lending industry.
- Innovation in structured finance and the expansion of the secondary mortgage market.
- The lowering of lending standards and the growth of subprime lending.
- Lower FED funds rates as an indirect and minor force. 
The Federal Home Loan Mortgage Corporation, also known as Freddie Mac, was created by Congress in 1970 to make possible a secondary mortgage market to provide greater liquidity to banks and other lending institutions to facilitate home mortgage lending. The Federal National Mortgage Corporation, also known as Fannie Mae, was originally created by the Federal Housing Authority (FHA) in 1938. In the beginning, Fannie Mae would securitize FHA loans, and it was the first to create a secondary mortgage market. In 1968, the company was privatized to remove its debt from the balance sheet of the Federal Government. Fannie Mae’s role in purchasing FHA loans was replaced by the Government National Mortgage Association, also known as Ginnie Mae. Both Freddie Mac and Fannie Mae are private corporations that have the implied backing of the Federal Government even though their activities are explicitly not guaranteed (until they were taken into conservatorship in September 2008). Collectively Freddie Mac, Fannie Mae and Ginnie Mae are known as Government Sponsored Entities or GSEs, and they are responsible for maintaining a secondary market for mortgage backed securities.
Fannie Mae and Freddie Mac buy and sell mortgage loans to create a secondary market. [ii] Mortgage originators bring groups of loans to the two companies which will either buy the loans to hold in their own portfolios, or they will bundle these loans together into securities in a process known as a “swap.” In a swap program, the originator provides the group of loans, and Fannie Mae and Freddie Mac promise the originator they will receive payments from the pool–whether Fannie Mae and Freddie Mac receive said payments or not. This guarantee is tantamount to insurance as the two companies are taking on all risk of default for a small annual “guarantee fee,” usually equal to 20 basis points (0.2% of the guarantee amount). Fannie Mae and Freddie Mac have strict loan origination guidelines because of the insurance they are providing. In the terms of the mortgage industry, “conforming” loans are those loans that meet the underwriting standards of Fannie Mae and Freddie Mac. In the later stages of the rally in the Great Housing Bubble, more and more mortgage loans were being originated that did not conform to Fannie Mae’s and Freddie Mac’s standards. The asset-backed securities (ABS) market packages these non-conforming loans into collateralized debt obligations and garnered significant market share. Despite their more conservative lending standards, Fannie Mae and Freddie Mac guaranteed many loans that performed poorly in the fallout of the Great Housing Bubble. They guaranteed many exotic loan types with inflated appraisals and committed many of the same errors as asset-backed securities (ABS) issuers during the bubble.
Figure 12: Percentage Held of Household Mortgage Debt, 1971-2006
As the secondary mortgage market continued to grow, lending institutions began to sell the loans they originated rather than keeping them in their own portfolios. The banks began to make money by originating and servicing loans rather than by keeping them and earning interest. This was a radical change in lending practices and incentives; lending institutions stopped being concerned with the quality of the loans because they did not keep them, and instead they became very concerned with the volume of loans originated and the fees these generated. The originators were only concerned with meeting the parameters set forth by buyers of mortgage backed securities in the secondary market. When the parties purchasing these loans reduced standards to the point where everyone qualified, loan originators gave everyone loans. Lower lending standards opened the door for lenders to provide loans to those with low FICO scores in great volume: subprime borrowers. When combined with the widespread belief that home prices would never go down, the combination inflated the Great Housing Bubble.
Figure 13: Subprime Originations, 1994-2006
Subprime lending as an industry barely existed prior to 1994. There were few lenders willing to loan to people with poor credit, and there was no secondary market to purchase these loans if they were originated. The growth of subprime was the direct result of the lowering of lending standards created by the change of incentives brought about by the creation of the secondary market. These factors alone were not enough to create the Great Housing Bubble, but they provided the basic infrastructure to allow the delivery of capital that caused house prices to take flight. The catalyst or precipitating factor for the price rally was the Federal Reserve’s lowering of interest rates in 2001-2004.
Many mistakenly believe the lower interest rates themselves were responsible by directly lowering mortgage interest rates. This is not accurate. Mortgage interest rates declined during this period, and this did allow borrowers to finance somewhat larger sums with the same monthly loan payment, but this was not sufficient to inflate the housing bubble. The lower Federal Funds rate caused an expansion of the money supply, and it lowered bank savings rates to such low levels that investors sought other investments with higher yields. It was this increased liquidity and quest for yield that drove huge sums of money into mortgage loans.
Structured finance is an innovation of the finance industry on Wall Street. It is a method of redistributing risk based on complex legal and corporate entities such as corporations, limited liability companies or some other kind of legal entity capable of entering into contracts. The shares or other interests in structured financial entities are derivatives that obtain their value from an underlying asset. Any asset that has a regular cashflow can be pooled through structured finance to create an asset-backed security. This cashflow can be split among various parties and valued based on the risk of repayment. For instance, the most common form of structured finance utilized to inflate the Great Housing Bubble was the collateralized debt obligation or CDO. A CDO derives its value from the underlying, asset-backed securities which in the Great Housing Bubble were generally bundles of mortgage loans. Mortgage loans generate a steady cashflow stream as individual homeowners pay their mortgage obligations, and these loans are collateralized by residential real estate. In the event of default on the mortgage held by a CDO, a house can be put through foreclosure to satisfy the mortgage debt and thereby return capital to the CDO.
In any asset-backed security, assets are bundled together to reduce risk and make the asset more attractive to investors. In contrast, if an individual buys a mortgage loan from a lender in order to receive the interest payments, this investor assumes all the risk of default. The default loss risk might be low, but if one party must bear this risk, the investor significantly discounts the security to compensate. However, if this individual investor buys a small share of a large pool of mortgage loans, the investor reduces risk exposure significantly and thereby the discount for purchasing it. The value of the security is increased by pooling and thereby lowering the risk. Also, for an individual investor to purchase a mortgage loan requires a significant equity investment as mortgage loans are often in the hundreds of thousands of dollars. If a number of mortgage loans are pooled and sold off to many investors as shares or interests in a financial intermediary like a CDO, the equity requirement can be lowered considerably thus opening this type of investment to a broader investment community. It is the spreading of risk and the lowering of equity thresholds that makes structured finance such an appealing investment tool.
Figure 14: Structure of a Collateralized Debt Obligation
Collateralized debt obligations, like other asset-backed securities, are divided in segments known as tranches (rhymes with launches). These tranches are typically titled: senior, mezzanine and equity based on their risk exposure. There is no single structure or formula for a CDO, and many contain numerous subdivisions resulting in more segments than the three described. Similar to the lien order of mortgage obligations, these tranches are paid in order of priority. The senior tranche is paid first, the mezzanine tranche is paid next, and finally the equity tranche is paid any remainder. Since these obligations are paid in order, the senior tranche has the least risk exposure and lowest returns, and the equity tranche has the highest risk and greatest potential for return. To further lessen risk (and make the transaction even more complicated) insurance policies are often issued to insure the buyer of a senior tranche against loss. These policies known as credit default swaps were a very lucrative business during the Great Housing Bubble. It was such good business that many insurers took excessive risks and lost a great deal of money when house prices declined. [iii]
The real magic of structured finance is its ability to take assets of low investment quality and turn it into something viable. George Soros aptly titled his book, “The Alchemy of Finance.” [iv] Like the alchemists of medieval Europe, modern investment bankers try to turn lead into gold. The syndicators who create and manage collateralized debt obligations assess the risk of loss on the underlying asset and break it down into three categories corresponding to the three tranches. The equity tranche in a CDO assumes the expected risk of loss. For example, if subprime loans expect an 8% loss from defaults, then the equity tranche will be 8% of the CDO. The syndicator typically keeps this equity tranche as part of their incentive fee, but practically speaking, the discount would be so steep it is hardly worth selling. If defaults losses are less than 8%, they see tremendous profits, and if it is over 8%, they see nothing. The Mezzanine tranche assumes the risk beyond the expected risk. If the average default loss is around 8%, and the highest default loss ever recorded is 24%, the mezzanine tranche exists to take on this risk. There is a very good chance they will see most or all of their money because the average default loss is being absorbed by the equity tranche. The senior tranche is supposed to have no risk from default loss. The line between mezzanine and senior is at or beyond the highest default loss rate ever recorded. This is not to say there is no risk, but it would take an unprecedented event to see any losses in this tranche–something like the collapse of the Great Housing Bubble.
Syndicators of collateralized debt obligations go to the open market to raise sufficient capital to buy the necessary securities and cover their fees. Since there is very little risk to the senior tranche holders, they require a lesser return on their investment. Although they own 76% of the CDO and receive 76% of the cashflow, they will pay more than 76% of the capital costs of the syndication (close to 85%) and still receive their required rate of return because the underlying subprime loan pool is paying in excess of the return required by senior tranche holders. The mezzanine tranche has more risk, and they will require a higher rate of return more closely approximating the interest rate on the underlying subprime mortgage. The remaining cost of the syndication is raised by the mezzanine tranche. The equity tranche raises no additional capital, and it is generally kept on the books of the syndicator as a bonus.
One can argue that structured finance creates greater efficiency in our financial system because capital is freed to pursue other objectives. Although, it can also be argued, as Warren Buffet has, that derivatives, the product of structured finance, are “financial weapons of mass destruction.” (Buffet, 2002) Both arguments stem from the same characteristic of these securities: excessive debt. When the loan that became part of the collateralized debt obligation was originated, this money was created out of nothing by the originating lender. This is how all money is created in a fractional reserve banking system (Heffernan, 2005). As long as there is sufficient cashflow, debt creation is normal; however, when excessive debt is created and available cashflow cannot service this debt, the system experiences the very serious problem of insolvency which can lead to monetary deflation–the disappearance of lender-created money into the ether from which it was created.
If an individual investor wanted to buy a mortgage loan, the purchase would proceed with equity rather than lender-created money. However, once packaged into a CDO, the senior tranche is often purchased by an investment banker or another lender which also created this money from nothing. Since the equity tranche raises no capital, the mezzanine tranche may be the only money in the structure not created by a lender out of the ether. With so little “real” money in the deal, there is very little buffer between what would be a loss of invested capital and a banking loss of created capital. There is a tipping point where the debt service exceeds the cashflow, and when this tipping point is reached, the entire debt structure may collapse in a deflationary spiral. [v] The structured finance products such as collateralized debt obligations and their derivatives are highly leveraged instruments with a very sensitive tipping point. These instruments are also highly sensitive to short term credit availability and lending rates. The long-term CDOs were often financed by continually rolling over short term debt. Rising cost of short-term debt would take a while to cause problems, but a sudden withdrawal of credit availability, as was witnessed during the credit crunch, meant desperate sales for those who owned these instruments. Monetary deflation was a major concern to the Federal Reserve as the Great Housing Bubble began to deflate.
The use of structured finance techniques in the syndication of collateral debt obligations was not by itself a problem causing the Great Housing Bubble. This was part of the infrastructure for delivering capital to the mortgage market which began with the creation of the secondary mortgage market. In the aftermath of the crash of house prices, collateralized debt obligations received a bad reputation as dangerous securities unworthy of the safe, “AAA” ratings they received from the companies that evaluate the creditworthiness of financial instruments. The advantages of structured finance did not disappear because of problems with the market or the ill-advised ratings these securities received. Collateralized Debt Obligations as syndicators of mortgage-backed securities nearly disappeared in 2008. However, they did not go away, and they will continue to be an integral part of the capital delivery system providing money for buyers to purchase residential real estate.
Systemic Risk in the Housing Market
Credit rating and analysis of collateralized debt obligations and all structured finance products are integral to the smooth function of the secondary market for mortgage loans. A credit rating agency is a company that analyzes issuers of debt and debt-like securities and gives them an overall credit rating which measures the issuer’s ability to satisfy its debt obligations. There are more than 100 major rating agencies around the world, and three of the largest and most important ones in the United States are Fitch Ratings, Moody’s and Standard & Poor’s. A debt issuer’s credit rating is very similar to the FICO score of an individual rated by the Fair Isaac Corporation widely used in the United States by institutional lenders. Of greater importance to the housing market, the credit rating agencies also analyze and rate the creditworthiness of the various tranches of collateralized debt obligations traded in the secondary mortgage market.
Credit ratings are widely used by investors because they provide a convenient tool for comparing the credit risk among various investment alternatives. The analysis of risk is crucial in determining the interest rate a syndicator will need to offer to attract sufficient investment capital. From the other side of the transaction, it is important to the investor who is comparing the interest rates being offered by various investments. The ratings agencies provide this critical, third-party analysis both sides of the transaction can rely upon for unbiased, accurate information. When the ratings agencies are doing their job well, there is greater efficiency in capital markets as syndicators of securities are obtaining maximum market values, and investors are minimizing their risks. This efficiency in the capital markets leads to better resource utilization and stronger economic growth.
Unfortunately for many investors in collateralized debt obligations during the Great Housing Bubble, the ratings agencies did not provide an accurate or credible rating of many CDO tranches. When the housing market pricing declined, many CDO tranches were subsequently downgraded. In defense of the agencies, they were providing an analysis of risk based on existing market conditions. Their reports contained caveats concerning downside risks in the event market conditions changed, but this list of risks is standard in any analysis and widely ignored by investors who are counting on the rating to be a market forecasting tool rather than the market reporting tool it really is. Credit rating agencies are not in the business of market forecasting or evaluating systemic risks.
There is a deeper problem with the ratings agencies that began to surface in the Great Housing Bubble. Ratings agencies used to charge investors for their risk analysis, but there was a transition to charging the issuers instead. As one might imagine, there are reports that ratings agencies were concerned if they gave CDOs poor ratings, their primary source of income would go elsewhere. This put pressure on the agencies to overlook certain problems or merely list them as footnotes to their reports rather than lower a rating due to a foreseeable contingency such as a decline in house prices.
Mortgage Default Losses
There is risk of loss in any investment, and losses in collateralized debt obligations arise from the difference in the book value of the underlying mortgage note and the actual resale value of the collateral on the open market, if this collateral is subject to foreclosure. There is an important distinction that must be made between the default rate on a mortgage loan and the resultant loss incurred when a default occurs. High mortgage default rates do not necessarily translate into high mortgage default losses and vice-versa.
Subprime loans have had comparatively high default rates since their introduction. When subprime mortgages began to capture broader market share starting in 1994, the rate of home ownership in the United States began to rise. The increasing use of subprime loans and the subsequent increase in home ownership rates put upward pressures on house prices. As house prices began their upward march, the default losses from subprime defaults began to fall because the collateral was obtaining more resale value, or was being sold by the subprime borrower before foreclosure. This made subprime lending, and its associated high default rates, look less risky to investors because these default rates were not translating into default losses. As time went on and prices continued to rise, subprime lending established a track record of investor safety which drew more capital into the industry. However, since the relative safety of subprime lending was entirely predicated upon rising prices, it was an industry doomed to fail once prices stopped rising.
Take this phenomenon to its extreme and its instability becomes readily apparent. Imagine a time when prices are rising, perhaps even due purchases by subprime borrowers, and imagine what would happen if 100% of the subprime borrowers defaulted without making a single payment. It would take approximately one year for the foreclosure and relisting process to move forward, and during that year, the prices of resale houses would have increased. When the lender would go to the open market to sell the property, it would obtain enough money to pay back the loan and the lost interest so there would be no default loss. What just happened? Lenders became de facto real estate speculators profiting from the buying and selling of homes in the secondary market rather than lenders profiting from making loans and collecting interest payments. This profiting from speculation is the core mechanism that disguised the riskiness of subprime lending. When these speculative profits evaporated when prices began declining, the subprime industry imploded and its implosion exacerbated the decline of home prices.
One of the major failings of the credit markets in the Great Housing Bubble was the failure to take a holistic view and evaluate the systemic risks involved. A standard credit analysis reviews various risk parameters and attempts to rate the impact of each. The implicit assumption is that the total risk is equal to the sum of the parts; however this is not necessarily the case. Synergy is when the whole is greater than the sum of its parts, and there is a strong synergy in default loss risk in collateralized debt obligations that became apparent during the Great Housing Bubble. The credit rating agencies failed to identify this risk synergy until after the fact.
The risk of default loss in a tranche of a collateralized debt obligation is directly related to the default loss risk in the underlying mortgage notes. There are six general areas of credit default loss risk that may be evaluated independently, but their interactions are often synergistic in nature: creditworthiness risk, high combined-loan-to-value default risk, high debt-to-income ratio risk, fraud and misrepresentation risk, investment perception risk, and resale value risk. Of these general areas of risk, market valuation is most responsible for creating synergistic effects and amplifying default losses. Since many of the more “innovative” loan programs entered the market during a time of rising prices, there was no history of performance of these securities in other market conditions making it very difficult to assess the impact a down market would have on default rates. As it turns out, exotic loan programs do not perform well in any conditions other than a raging bull market.
Every mortgage loan that is originated contains an evaluation of the creditworthiness of the borrower who is responsible for making timely mortgage note payments. The most common evaluation tool is the FICO score. Prime borrowers have the highest FICO scores, they are considered the lowest default risk, and they receive the lowest interest rates as a result. Subprime borrowers have the lowest FICO scores, they are considered the highest default risk, and they receive the highest interest rates. This is the best documented and most carefully evaluated risk parameter. Before many of the loan programs were introduced during the Great Housing Bubble, FICO scores strongly correlated with default rates. This correspondence broke down in the price decline when the bubble popped because the other risk factors proved to have a greater influence than expected.
High CLTV Defaults
The combined-loan-to-value (CLTV) is the total debt of all mortgage obligations as a percentage of the appraised value of a particular property. A high CLTV generally corresponds to a low downpayment, but as resale values fell in the market crash, the CLTV rose for many borrowers as a consequence of falling prices. Although all borrowers with high CLTV loan balances show high default rates, it is important to distinguish between those borrowers who had a high CLTV because of a low downpayment and those who had a high CLTV because of falling values. Even though downpayments are a sunk cost and irrelevant to the market value of a house, they do have a strong psychological impact on the behavior of homeowners. [vi] People who put little or no money of their own money into the purchase of real estate exhibit greater default rates because they are not losing much of their money. Most people really do not care if the lender loses money, particularly if they will not have to repay the lender for the loss or incur tax penalties on the forgiven debt. When borrowers have less of their money in a transaction they are less likely to sacrifice to stay current on their mortgage note obligations, and they are more likely to default if resale values decline, particularly if their payments are greater than the cost of a comparable rental.
Fraud and Misrepresentation Risk
Most purchasers of collateralized debt obligations did not realize there was a huge amount of fraud and misrepresentation in the underlying loans they were purchasing. High CLTV financing, particularly the widely offered 100% financing, became the ideal tool for fraud. Fraudulent transactions require “straw buyers” willing to sacrifice their credit for a fee (or identity theft,) appraisers willing to inflate the houses value, and realtors and mortgage brokers either willing to go along with the transaction for cash or too ignorant to see the truth. In a transaction, the straw buyer purchased a house for greater than its true market value, and the excess payment was used to pay off the corrupted parties. Fraud was much easier to commit with 100% financing because the bank loaned the full amount of an inflated appraisal. It is much harder to commit fraud when the bank only loans 80% of a property’s value. Most often the seller was in on the scam and was using the transaction to get out of a bad deal, but sometimes sellers were also innocent victims. The straw buyer had no intention of repaying the loan from the start, and the property quickly went into foreclosure.
A more common problem was misrepresentation of income. Stated-income loans, also known as “liar loans,” were very common during the bubble rally. People would simply make up a number that qualified them for a loan and state it on their mortgage application. One of the assumptions purchasers of CDOs made was that the originators of the underlying loans made sure the borrowers in reality made enough money to pay back the loan. Often times the extent of the loan originators’ due diligence was examining the borrower’s signature on the loan application and trusting in the veracity of the signatory. This was a very serious problem for valuing an interest in a CDO because there was no way to accurately determine the viability of the income stream when the income of those responsible for paying the underlying mortgage notes was in doubt.
High DTI Defaults
The debt-to-income ratio is the total amount of payments compared to gross income expressed as a percentage. A lender evaluates the DTI of the mortgage loan as well as the total DTI of all borrower indebtedness when making a determination of creditworthiness. Historically, a borrower could not have a mortgage DTI in excess of 28% and a total DTI greater than 36% to qualify for a loan because debt burdens in excess of these figures proved to have high default rates. Despite this historical knowledge, lenders widely ignored these standards in the Great Housing Bubble in the quest for more customers. During the rally, few of these people defaulted because they were offered even more debt through home equity lines of credit from which they could make mortgage payments, and the few who did get into financial problems simply sold their house to pay off the mortgage. During the rally, people were keen to take on mortgage debt because interest rates were low, and it was a necessary tool for obtaining real estate and its commensurate appreciation benefits. It did not matter to buyers if 50% of more of their gross income was going toward debt service if the property itself was providing the additional income necessary to sustain their lifestyle. Of course, this only works when prices are increasing rapidly. Once prices stopped rising, the property could no longer provide additional income, and the borrowers had to make the crushing payments out of their true income. Without the benefits of appreciation, borrowers quickly found the burden of a high debt-to-income ratio overwhelming, and many borrowers defaulted because the payments were too much to handle–just as the lessons from history said they would be.
Investment Perception Risk
One of the biggest fallacies pushed on the general public is the notion that residential real estate is a great investment. This idea caused people to view houses as an investment and treat them accordingly. [vii] When the participants in a housing market perceive houses as an investment, they will more easily default on the loan than if they viewed the house solely as a home for their family. People develop emotional attachments to their family homes, and they will sacrifice much in order to keep it. People behave in a more businesslike manner when they view a house as an investment, and they are willing to give up the house if the investment does not perform as planned. When faced with the reality that house prices were not going to continue to go up and payments were in fact going to continue to cause losses, many people decided to stop making payments and allow their investment go into foreclosure. Financially, it was the logical decision given the alternative of continuing to make payments on a losing investment. When the “Great American Dream” of home ownership was tainted by investment motives, it became a nightmare for speculators and CDO investors alike.
Resale Value Risk
The biggest risk faced by buyers of collateralized debt obligations is the default loss risk of the underlying mortgage when the collateral for the mortgage (the house) is overvalued in markets characterized by low affordability. The greatest risk of default is based on changes in the resale value of homes. All other default loss risk factors are masked when prices are increasing, and they are amplified when prices decline. Valuation risk is the ultimate synergistic factor.
There are three methods of appraising the resale value of residential real estate: the comparative-sales approach, the cost approach, and the income approach. The comparative-sales approach uses recent sales of similar properties in the market because comparable sales reflect the behavior of typical buyers in the marketplace. The cost approach determines market value by calculating the replacement cost of an identical structure plus the cost of the land or lot upon which the house would sit. The income approach determines market value by analyzing market rents of comparable properties and applies the gross rent multiplier of expected rents. Most lenders give the greatest weight to the comparable sales approach when establishing market value before applying any loan-to-value limitations to the loan amount. The income approach is generally only considered for non-owner occupied homes. [viii] The three-test approach to appraising market value as used during the Great Housing Bubble is fraught with risk and is seriously flawed.
The comparative-sales approach reinforces the delusive behavior and irrational exuberance of a financial mania. If everyone is overpaying for real estate, the comparative-sales approach simply enables greater fools to continue overpaying for real estate. Since market prices for houses which serve as loan collateral fall to fundamental valuations based on income after the financial mania runs its course, mortgages originated based on the comparative-sales approach have a great deal of market risk not reflected in the pricing of collateralized debt obligations based on the underlying mortgage loans.
The cost approach has an even greater level of market risk. The cost of a structure may represent a relatively small percentage of the market value of real estate in high-value markets. In some of the most overvalued markets during the bubble, the replacement cost of the structure may have been $250,000 while the value of the underling land was $450,000; however, since the market value of land is a residual calculation based on the market value of the property, the value of the land cannot be determined independently of the house situated on it. Either the comparative-sales approach or the income approach must first be applied to establish the market value of the property before any calculation of the market value of the land can be determined. In short, since the cost approach is dependent upon another valuation method, it is not useful as an independent method of property valuation. Also, since the valuation of land is extremely sensitive to small changes in the valuation of the property, the cost approach is misleading with respect to the valuation of residential real estate.
The only reliable method for the valuation of residential real estate is the income approach, and it is the only approach that is widely ignored by the lending community. It has been demonstrated in previous residential market bubbles in California and in major metropolitan areas in other states that once a price decline begins, prices fall to fundamental valuations based on income and rent. [ix] The reason for this is that once the speculative investment incentive is removed from the market, buyers do not support prices until there is a new reason for them to buy: they can save money versus renting. Comparative rents are the fundamental valuation of residential real estate. Mortgage default loss risk is low only when market prices are in line with comparative rents or when market prices are increasing. Default loss risk is low when prices are in line with rents because a property can be converted from owner-occupied to a rental unit and the payment can still be covered. Default loss risk is low when prices are rising because a borrower experiencing financial difficulty can always sell the property to repay the loan. Unfortunately, once market prices increase above the level of comparative rents, they endure a period of decline back to comparative rent levels; therefore, if lenders continue to use the comparative-sales approach, they will enjoy a temporary period of low market risk while prices increase and another painful period of losses when prices decrease. As was demonstrated in the aftermath of the Great Housing Bubble, these periods of lender losses can imperil the entire banking and financial system. The only way to prevent the pain of loss is to recognize the end-game risks when prices are increasing and choose not to participate in that lending environment. Many lenders did not participate in the crazy lending of the Great Housing Bubble, and they were not significantly damaged in the aftermath; however, the hunger for mortgage loans from the CDO market compelled many lenders to participate or get buried by their competitors. The only real market-based solution to the problem of originating bad loans must come from the CDO market.
The CDO Market Solution
The solution to preventing future bubbles in the residential real estate market lies in the market for collateralized debt obligations and conforming loans insured by the government sponsored entities (GSEs). The GSEs created the secondary mortgage market in the 1970s, and the CDO market is the extension of this market bringing large amounts of investment capital to residential real estate. During the Great Housing Bubble the CDO market did not properly evaluate the risk of default on the underlying mortgage notes they pooled.
If the CDO market were to evaluate mortgage default loss risk based on the income approach rather than the comparative-sales approach, the performance of CDOs would be greatly improved, and investor confidence would return to the market. It is only after the risks are properly evaluated that capital would return to this market. If the CDO market evaluates risk based on the income approach, the lenders that originate loans hoping to sell them to CDOs would be forced to do the same. If lenders originate loans based on the income approach, the irrational exuberance that creates financial bubbles would not be enabled. People would still be free to overpay for houses with their own money, but the scope and scale of financial bubbles would be limited to the funds of buyers, and the banking system would not be imperiled by the foolishness of the market masses when prices fall to fundamental valuations based on rent and income.
 Most participants in the housing market believe changes in interest rates are responsible for changes in housing prices (Case & Shiller, The Behavior of Home Buyers in Boom and Post-Boom Markets, 1988). There is actually very little correlation between interest rates and house prices. Interest rates are very important for determining the amount a borrower can obtain in a loan, but other factors are more critical to determining the actual price of real estate.
[ii] Most of the technical data on the secondary market found in this section comes from the paper Fussing and Fuming over Fannie and Freddie: How Much Smoke, How Much Fire? (Frame & White, 2005). Another paper with excellent historical background on the evolution of the secondary mortgage market is The Housing Finance Revolution (Green & Wachter, 2007) by Richard Green and Susan Wachter. Perhaps the finest overview of the functioning of the secondary mortgage market with respect to subprime is Understanding the Securitization of Subprime Mortgage Credit by Adam B. Ashcraft and Til Schuermann (Ashcraft & Schuermann, 2008).
[iii] In the paper Who Holds the Toxic Waste? An Investigation of CMO Holdings (Haubrich & Lucas, 2006), the authors provide a good background on the CMO (CDO by another name) market. Their analysis of who holds the bad paper is suspect due to lack of data. Also, many “insurers” were not insurance companies. These were private, unregulated firms who often had little financial ability to make good on their obligations.
[iv] (Soros, 1994)
[v] (Burdekin & Siklos, 2004)
[vi] Even if a homeowner’s house is worth less than the mortgage, there is still option value in the property. If the homeowners is not far underwater, it may not take much time for values to return and provide them with equity in the property.
[vii] Robert Shiller’s surveys of market attitudes in 1989 showed 95% of respondents in the bubble markets of San Francisco and Orange County said they thought of their purchase as an investment at least in part. Also, the tendency to view housing as an investment is a defining characteristic of a housing bubble (Case & Shiller, Is There a Bubble in the Housing Market, 2004). Since housing bubbles portend of disastrous declines, the investment motive as a risk to CDOs is very real.
[viii] FHA Guidelines for appraisals only require the use of the income approach for income producing properties.
[ix] Real Estate prices in California have bubbled 3 times since the 1970s. After prices peaked in the late 70s and then again in the late 80s, prices declined until they came back into alignment with historic fundamental valuations. This is strong confirmation of the theory of buyers waiting for rental equivalence before purchasing when prices drop.