What they are saying about The Great Housing Bubble
“The author, Larry Roberts, is best known for his daily posts as IrvineRenter on the Irvine Housing Blog. Long before Lehman crashed, Fannie Mae was taken over, and even before home prices were dropping nationally, he was one of the few voices presenting real information on the housing bubble.
The author’s background is in new housing development in Southern California. It was a good start to understanding how things worked. Supplemented by knowledge from countless posters at the housing blog, he has been able to show why home prices couldn’t stay elevated. Price to income ratios, price to rent ratios, and other factors detailed in the book showed how far out of line prices had become by 2006. A full year before house prices started to crash, he was predicting it, and many of the crash’s details. While some people are permanently bullish or bearish on housing, the best are able to understand and explain the mechanisms, tell you what will happen in what sequence.
The Great Housing Bubble is an excellent read, and an important one.”
Brian Whitworth – Principal, FinancialPatents.com
Fundamental Valuation of Houses
The fundamental value of all housing prices is equivalent rents. Rents define the fundamental value of real estate because rental is a direct proxy for ownership; both rental and ownership provide for possession of property. Equivalent rents are a major component of the United States Government’s Consumer Price Index (CPI).  According to the US Department of Labor, “This approach measures the change in the price of the shelter services provided by owner-occupied housing. Rental equivalence measures the change in the implicit rent, which is the amount a homeowner would pay to rent, or would earn from renting, his or her home in a competitive market. Clearly, the rental value of owned homes is not an easily determined dollar amount, and Housing survey analysts must spend considerable time and effort in estimating this value.” Prior to the first California housing bubble in the late 1970s, the housing cost component of the CPI was measured using actual price changes in the asset. When this bubble created an enormous distortion in this index, the rental equivalence model was constructed. It has been used to smooth out the psychologically-induced housing price bubbles ever since.
An argument can be made for the real cost of construction as the fundamental valuation of houses. If house prices in a market fall below the cost of new construction, no new houses will be built because a builder cannot make a profit. If there is continuing demand for housing, the lack of supply will create an imbalance which will cause prices to increase. When new construction becomes profitable again, new product will be brought to market bringing supply and demand back into balance. If demand continues to be strong, builders will increase production to meet this demand keeping prices near the real cost of construction.
Based on a theory of rational market participants, one would expect that when prices go up and the cost of ownership exceeds the cost of rental, people choose to rent rather than own, and the resulting drop in demand would depress home prices: The inverse would also be true. Therefore, the proxy relationship between rental and ownership would keep home prices tethered to rental rates. However, this is not the case. [ii] If there were only a consumptive value to real estate, the cost of ownership and the cost of rental probably would stay closely aligned; however, since there is an opportunity to profit from speculative excesses in the market, rising prices can lead to irrational exuberance as buyers chase speculative gains.
Rental rates tend to keep pace with wages because people normally pay rent out of current income. As people make more money, they compete for the available rentals and drive prices up at a rate about 1% greater than the overall rate of inflation. [iii] There are times when supply and demand issues in local markets create fluctuations in this relationship, but as a rule, rents track wages pretty closely. Since house prices are tied to rents, and rents are tied to wages, house prices are indirectly tied to wages. When house prices increase faster than wage growth, the price levels become unsustainable, and if the differential is too great, a bubble is inflated. [iv]
Figure 10: National Rent-to-Income Ratio, 1988-2006
Ownership Cost Math
A useful way to look at the cost of housing is to evaluate the total monthly cost of ownership. There are 7 costs to owning a house. Although some of these costs are not paid on a monthly basis, they can be evaluated on a monthly basis with simple math. These costs are:
- 1. Mortgage Payment
- 2. Property Taxes
- 3. Homeowners Insurance
- 4. Private Mortgage Insurance
- 5. Special Taxes and Levies
- 6. Homeowners Association Dues or Fees
- 7. Maintenance and Replacement Reserves
The mortgage payment is the first and most obvious payment because it is the largest. It is also an area where people take risks to reduce the cost of housing. It was the manipulation of mortgage payments that was the focus of the lending industry “innovation” that inflated the housing bubble. The relationship between payment and loan amount is the most important determinant of housing prices. This relationship changes with loan terms such as the interest rate, but it is also strongly influenced by the type of amortization, if any. Amortizing loans, loans that require principal repayment in each monthly payment, finance the smallest amount. Interest-only loan terms finance a larger amount than amortizing loans because none of the payment is going toward principal. Negatively amortizing loans finance the largest amount because the monthly payment does not cover the actual interest expense.
Property taxes have long been a source of local government tax revenues. Real property cannot be moved out of a government’s jurisdiction, and values can be estimated by an appraisal, so it is a convenient item to tax. In most states, local governments add up the cost of running the government and divide by the total property value in the jurisdiction to establish a millage tax rate. California is forced to do things differently by Proposition 13 which effectively limits the appraised value and total tax revenue from real property. [v] Local governments are forced to find revenue from other sources. Proposition 13 limits the tax rate to 1% of purchase price with a small inflation multiplier allowing yearly increases. [vi] The assessed value can only increase 2% a year regardless of actual market appreciation. The assessed value is set to market value when the property is sold. Often the lender will compel the borrower to include extra money in the monthly payment to cover property taxes, homeowners insurance, and private mortgage insurance, and these bills will be paid by the lender when they come due. If these payments are not escrowed by the lender, then the borrower will need to make these payments. The total yearly property tax bill can be divided by 12 to obtain the monthly cost.
Homeowners insurance is almost always required by a lender to insure the collateral for the loan. Even if there is no lender involved, it is always a good idea to carry homeowners insurance. The risk of loss from damage to the house can be a financial catastrophe without the proper insurance. A standard policy insures the home itself and its contents. Homeowners insurance is a package policy which covers both damage to property and liability or legal responsibility for any injuries and property damage by the policy holder. Damage caused by most disasters is covered with some exceptions. The most significant exceptions are damage caused by floods, earthquakes and poor maintenance.
Private Mortgage Insurance
Mortgages against real property take priority on a first recorded, first paid basis. This is known as their lien position. This becomes very important in instances of foreclosure. The first mortgage holders get paid in full before the second mortgage holder get paid and so on through the chain of mortgages on a property. In a foreclosure situation, subordinate loans are often completely wiped out, and if the loss is great enough, the first mortgage may be imperiled. Because of this fact, if the purchase money mortgage (first lien position) exceeds 80% of the value of the home, the lender will require the borrower to purchase an insurance policy to protect the lender in event of loss. This policy is of no use or benefit to the borrower as it insures the lender against loss. It is simply an added cost of ownership. Many of the purchase transactions during the bubble rally had an 80% purchase money mortgage and a “piggy back” loan of up to 20% to cover the remaining cost. These loan pairs are often referred to as 80/20 loans, and they were used primarily to avoid private mortgage insurance. There were very common during the bubble.
Special Taxes and Levies
Several areas have special taxing districts that increase the tax burden beyond the normal property tax bill. Many states have provisions which allow supplemental property tax situations. The State of California has Mello Roos fees. A Community Facilities District is an area where a special tax is imposed on those real property owners within the district. This district is established to obtain public financing through the sale of bonds for the purpose of financing certain public improvements and services. These services may include streets, water, sewage and drainage, electricity, infrastructure, schools, parks and police protection to newly developing areas. The taxes paid are used to make the payments of principal and interest on the bonds.
Homeowner Association Dues and Fees
Many modern planned communities have homeowners associations formed to maintain privately owned facilities held for the exclusive use of community residents. These HOAs bill the owners monthly to provide these services. They have foreclosure powers if the bills are not paid. It is given the authority to enforce the covenants, conditions, and restrictions (CC&Rs) and to manage the common amenities of the development. It allows the developer to legally exit responsibility of the community typically by transferring ownership of the association to the homeowners after selling off a predetermined number of lots. Most homeowners’ associations are non-profit corporations, and are subject to state statutes that govern non-profit corporations and homeowners’ associations. In cases where a large number of houses are unsold, in foreclosure, or are owned by lenders, remaining homeowners may encounter large increases in assessments. In some cases, the additional cost can become unaffordable to remaining homeowners pushing more of them to sell or be foreclosed on by their own homeowners association.
Maintenance and Replacement Reserves
An often overlooked cost of ownership is the cost of routine maintenance and the funding of reserves for major repairs. For example, a composite shingle roof must be replaced every 20-25 years. It may take $100 a month set aside for 20 years to fund this replacement cost. Also, condominium associations often levy special assessments to undertake required work for which the reserves are insufficient. In the real world, most people do not set aside money for these items. Most will attempt to obtain a Home Equity Line of Credit (HELOC) to fund the repairs when they are necessary. Of course, this assumes a property has appreciated and that such financing will be made available.
There are two other variables people often consider when evaluating the cost of ownership that is not included in the prior list: income tax savings and lost downpayment interest. When a borrower takes out a home loan, the interest is tax deductible up to a certain amount. For borrowers in the highest marginal tax bracket, the savings can be significant, and this can make a dramatic difference in the true cost of ownership. However, this benefit diminishes over time as the loan is paid off and the interest decreases. Plus, contrary to popular belief, it is never good financial planning to spend $100 to save $25 in taxes. Also, these benefits are almost universally overestimated by people considering a home purchase. Renters considering home ownership will need to remember that they will be giving up the standard deduction when they itemize to obtain the Home Mortgage Interest Deduction (HMID). [vii] A “married filing jointly” taxpayer will forgo a $10,700 deduction in 2007. This reduces the net impact of the HMID. Anecdotally, even those in the highest tax brackets usually do not get more than a 25% tax savings.
This is the forgotten benefit of a conventionally amortizing loan: forced savings. Most people are not good at saving. The government recognized this years ago when they started taking money out of people’s salaries to pay income taxes because they knew people would not do it on their own. People who become homeowners during their lifetimes often have the equity in their home as their only source of retirement savings other than social security. To accurately calculate the cost of ownership, this hidden savings amount needs to be deducted from the total cost of ownership because this money will generally come back to the borrower at the time of sale. Since taxpayers in the United States get a capital gains exemption up to $250,000 per person or $500,000 per couple, this savings amount does not need to be adjusted for capital gains taxes in most circumstances.
Lost Downpayment Interest
Unless 100% financing is utilized, a cash downpayment will generally be withdrawn from an interest bearing account to purchase a house. The monthly interest that would have accrued if the downpayment money was still in the bank is a cost of ownership. This is perhaps the most overlooked ownership cost. For instance, if you are putting 20% down on a $244,900 property, you will be taking $48,980 from a bank account where it would have earned 5% in 2007. This $2,449 in interest comes to $204 in lost interest the moment this money gets tied up in real property. If someone chooses to rent rather than buy, this interest income would be earned. Of course, this earned income is also taxed, so 75% of this number is the net opportunity cost of a downpayment.
To establish the cost of ownership, each of these costs, if applicable, must be quantified. When the total monthly cost of ownership is equal to the rental rate, the market is considered to be at fair value for owner-occupants. In fact, this is the equilibrium in most real estate markets across the nation. In a strange way, the bubble did not upset this equilibrium. The use of negative amortization loans with artificially low teaser rates allowed borrowers to obtain double the loan amount with the same monthly payment: double the loan; double the purchase price. This is how prices were bid up so high so fast without a commensurate increase in wages. The elimination of these loans is also the reason prices collapse.
Running the Numbers
Below is a typical cost of ownership for a $244,900 Median property in the US (2006):
Equation 1: Cost of Ownership for 2006 Median Property in United States
|$ 244,900||Purchase Price|
|$ 48,980||Downpayment @20%|
|$ 195,920||Mortgage @ 80%|
|$ 1,238.35||Mortgage Payment @ 6.5%|
|$ 204.08||Property Taxes @ 1%|
|$ 51.02||Homeowners Insurance @ 0.25%|
|$ 51.02||Special Taxes and Levies @ 0.25%|
|$ 100.00||Homeowners Associate Dues or Fees @ $100|
|$ 306.13||Maintenance and Replacement Reserves @1.5%|
|$1,950.60||Monthly Cash Cost|
|$ (278.06)||Tax Savings @ 25% of mortgage interest and property taxes|
|$ (177.11)||Equity hidden in payment|
|$ 153.06||Lost Downpayment Income @ 5% of Downpayment|
|$ 1,648||Total Cost of Ownership|
- The mortgage payment assumes a 30-year fixed-rate conventionally amortized mortgage at 6.5% interest.
- The property taxes are set at the 1% limit imposed by Proposition 13.
- The homeowners insurance is estimated at one-quarter of one percent per year.
- Private Mortgage Insurance is estimated at one-half of one percent per year. It is not included in the calculation above because this example utilized 80% financing. If the financing amount required PMI, the costs would have been over $100 a month higher.
- Special Taxes or Levies (Mello Roos) is estimated at one-quarter of one percent per year. Some neighborhoods do not have Mello Roos as the bonds have been paid off. Some Mello Roos fees are as high at 1%.
- HOA dues are estimated at $100: some are lower, and some are much higher.
- Maintenance and replacement reserves are estimated at 1.5%. This may be the most contentious estimate of the group because most people assume they will simply borrow their way around these costs when they are incurred. This certainly has been the pattern during the bubble years when credit was free flowing. This method of home improvement and maintenance may be significantly more difficult as the credit crunch and declining values make financing much more difficult to obtain. In any case, these costs are real, and failing to acknowledge them denies the realities of home ownership.
- The sum of the above costs is the monthly cash cost of ownership. A homeowner may not write a check for each of these costs every month, but the costs are still incurred, and renters do not pay them.
- The tax savings are based on the maximum interest payment at the beginning of a loan amortization schedule. This tax savings will decline each month as the mortgage is paid off. Contrary to popular belief, this is not a bad thing. Also, the property taxes are also deductable, but Mello Roos are not fully deductible (even though most people mistakenly deduct it).
- The opportunity cost of lost interest assumes a 5% interest rate on the downpayment reduced by 25% for taxes on this earned income.
The actual cost of ownership on a typical $244,900 property would be approximately $1,648 per month. Some will be higher and some will be lower, but the calculation above, when adjusted for the specific property details being examined, yields the cost of property ownership.
So what general relationships can be inferred from the ownership cost breakdown provided above? First, notice the relationship between monthly cost and price. This property is worth 154 times the monthly cost when you fully examine the cost of ownership. Also, notice the relationship between monthly payment and price. This property is worth 198 times the monthly payment. Common mistake homebuyers make when considering a home purchase is to look at only the payment and ignore the other costs of ownership. Most assume, or have been told by realtors and mortgage brokers trying to make a commission that the tax benefits offset the other costs of ownership. Clearly, this is not the case. The true cost of ownership is about 30% higher than the monthly payment.
The price-to-cost and price-to-payment relationships become important when one wants to evaluate the relative value of the property compared to market rents. Since housing is a consumer good that can be obtained through either renting or owning, it is rational to compare the costs of each method of possessing property to see which provides a better value to the consumer. Just as stocks have price-to-earnings ratios (PE Ratios) used to establish relative value, houses have a price-to-rent ratio to establish relative value. [viii] When a property can be rented for an amount equaling its monthly cost of ownership, it is at rental parity. This is the breakeven point where a consumer would be indifferent in financial terms to own or to rent. Of course there are reasons to own or to rent which are not financial, but from a strictly financial standpoint, this is where the fundamental value lies.
The price-to-rent ratio is very sensitive to changes in interest rates. When interest rates are low, the cost of money is low, so larger sums can be borrowed and vice-versa. Nationally, the price-to-rent ratio increased steadily from 1988 through 2004 in a range from 157 to 199 while mortgage interest rates declined from 10.34% in 1988 to 5.84% in 2004. This increase in price was mostly the result of lowered interest rates as the out-of-pocket expense remained relatively constant. The dramatic increase in prices after 2004 was not supported by incomes or rents, and it is part of the evidence of a real estate bubble. [ix]
The price-to-rent ratio is also the basis for a commonly used valuation measure used in the property management business, the Gross Rent Multiplier (GRM). The GRM is a convenient way to evaluate whether or not a rental rate will cover the monthly cost of a particular property. It was developed by landlords seeking a method to quickly evaluate the purchase price of a property to see if it would be a profitable investment. When performing such an evaluation, a cashflow investor will typically look for a GRM near 100 to find a property with positive cashflow. This method can also be easily adapted to calculate the breakeven point where an owner/occupant would break even compared to renting. Considering the full cost of ownership–including those costs often ignored–the price-to-rent ratio and Gross Rent Multiplier is lower than most think. The GRM is a convenient measure of value because it spares you the toil of performing the above, detailed calculation to evaluate a large number of properties.
Figure 11: National Price-to-Rent Ratio, 1988-2007
 There are a number of research papers discussing the pros and cons of various methodologies for calculating equivalent rent. Hedonic Estimates of the Cost of Housing Services: Rental and Owner-Occupied Units (Crone & Nakamura, 2004) Treatment of Owner-Occupied Housing in the CPI (Poole, Ptacek, & Verbrugge, 2005).
[ii] Robert Shiller noted “that real owners’ equivalent rent and real building costs track each other fairly well, as one might expect. But neither of them tracks real home prices at all, suggesting that some other factor – I will argue market psychology – plays an important role in determining home prices.”
[iii] Depending on the market, rental rates grow at a rate around 1% over the rate of inflation. Rental rates are closely aligned with income growth, and in markets where income growth is strong, rental rates increase at approximately the same rate.
[iv] John Krainer, chief economist for the Federal Reserve Bank of San Francisco, pointed out in 2004 “The price-rent ratio for the U.S. and many regional markets is now much higher than its historical average value.” (Krainer & Wei, House Prices and Fundamental Value, 2004) This is one of the first papers (other than those by Robert Shiller) to recognize the data was pointing to a housing bubble.
[v] The full text of the Proposition 13 law can be found at http://www.leginfo.ca.gov/.const/.article_13A
[vi] In California, the first half of regular secured property tax bills are due November 1st, and delinquent after December 10th; the second half are due February 1st, and delinquent after April 10th each year. If the delinquent date falls on a Saturday, Sunday, or government holiday, then the due date is the following business day.
[vii] All information regarding tax rates comes from the Internal Revenue Service. http://www.irs.gov/
[viii] There are many studies that have mentioned the use of price-to-rent ratios as being similar to price-to-earnings ratios of stocks. Some of the studies are good, and some are not. Bubble, bubble, toil and trouble is of the latter variety (Haines & Rosen, 2006). Typical of these studies is that they will look at the data, see the obvious signs of a bubble, and proceed to dismiss the obvious as something else. Even though the national data for price-to-rent clearly shows a bubble, even in their own graphs, the authors dismiss the idea because “all real estate is local.” The paper was written for the Federal Reserve, but it could have been written for the National Association of Realtors. Another silly statement they make is “Thus, what appears to be a bubble in some markets might just be a reflection of normally high volatility in those markets.” This is like saying “what appears to be a bubble isn’t a bubble because bubbles are normal in these markets.” When the authors can look right at the data and not understand what they are seeing, there is little hope the paper will draw the right conclusions.
[ix] The study A Trend and Variance Decomposition of the Rent-Price Ratio in Housing Markets by Sean D. Campbell, Morris A. Davis, Joshua Gallin, and Robert F. Martin (Campbell, Davis, Gallin, & Martin, 2005) uses method of estimating the investment premium people pay for homes in bubble markets based on the expectation of future rental growth. This entire approach is flawed as it assumes people are investing based on cash flows. This would be a rational approach, but most people who buy in financial manias know little or nothing about cashflow or how to value it. The real reason they are “investing” is to capture speculative price changes. Trying to determine a fundamental valuation based on cashflow is an interesting exercise in math and statistics, but it completely fails to capture the real motivation behind buyers in the marketplace.