Housing as a Portfolio Asset: A Life-Cycle Analysis

By Bobby E. Waldrup, Seth C. Anderson, Sid Rosenberg,
and Vincent Shea

E-mail Story Print Story

Considerable debate surrounds the question of whether a home should be considered a component of one’s investment portfolio. Some argue that a home is frequently someone’s most important investment because it provides healthy returns via leverage, appreciation, tax deductibility of interest, and an absence of capital gains upon sale. University of Pennsylvania Professor Chris Mayer points out that one can expect the equivalent of 10%–12% pretax return over a typical seven- to 10-year timeframe. He attributes this to 3%–5% in capital appreciation and the fact that a homeowner does not have to pay rent.

Others, however, contend that homes are a consumption item and may not turn out to be a good investment. For example, New York City financial planner Gary Schatsky argues that, unlike other assets, selling the home to realize a profit is untenable, and that homes in general are illiquid assets. Some planners claim that home values can give people a false sense of security and cause them to neglect sound financial-planning strategies. Hence, many planners exclude one’s home when calculating net worth.

Stocks, Bonds, and Housing: Returns and Relationships

For the 34-year period ending in 2002, the authors computed the annual returns and return standard deviation for the Standard & Poor’s 500 Index benchmark and found annual returns to be 11%, with a standard deviation of 16% (Exhibit 1). For 10-year Treasury bonds, they found an annual return of 8.2%, with a standard deviation of 7.8%. For housing, they used existing single-family home prices over the 34-year period and found an annual return of 6.3%, with a standard deviation of 3.5%. (The authors acknowledge that the use of repeat sales measures might more accurately reflect an individual home’s appreciation rate, but chose a broader index of price changes that reflects the national upward trend as a more generic statistic.)

Because these asset classes have different mean returns and standard deviations, a coefficient of variation (standard deviation divided by mean return) was employed to compare riskiness per unit of return among the asset classes. Exhibit 2 indicates that stock returns display a higher variability per unit of return than do bonds or single-family home prices. Perhaps as important, single-family homes have displayed remarkable consistency in return variability over the last three decades.

The diversification benefits of combining these asset classes in a portfolio are a function of their return correlations. Year-to-year returns for single-family homes display little correlation with the returns of either stocks or bonds over the 34-year period (–0.18 and –0.16, respectively). Returns in single-family homes actually appear to track changes in the inflation rate more closely than they track the S&P 500 returns.

The following sections use these historical relationships to illustrate the impact of housing decisions at key points in the life cycle. The three junctures are:

• The decision to purchase a first home;
• The decision to trade up in price; and
• Decisions related to retirement.

Young Adults: Renting Versus Buying

Consider a young couple, the Smiths, that are 30 years from retirement, with an income resulting in a 33% combined marginal tax bracket. This couple currently has a portfolio comprising $15,700 in stocks and $15,700 in cash, with a current monthly rent expense of $1,469 [estimated as an average PITI (principal, interest, taxes, and insurance) and maintenance expenses of a median-priced home]. The Smiths are considering purchasing a median-priced home for $157,400 and financing it with a 6% mortgage. (All examples assume a 6% mortgage rate, 3% inflation rate, and 4% carrying costs for a house.)

The Smiths’ financial options are as follows: 1) to continue renting (assuming a 3% inflation rate) and invest the entire $31,400 in an S&P 500 index fund, or 2) to use the available cash of $15,700 to initiate a home purchase and leave the remaining $15,700 in stocks.

Exhibit 3 shows that the Smiths’ cumulative wealth impact will be higher ($615,323 versus $27,034) over a 30-year period under the home-purchase scenario. The wealth impact assumes a 4% annual cost for taxes, insurance, and maintenance, all of which grow at 3% annually. The wealth impact can be attributed to several factors:

• By purchasing a home, the Smiths have effectively frozen their “consumable” cost of housing, thereby stabilizing their payment stream to no longer increase with inflation, in contrast to rental payments.
• The Smiths’ investment in housing enjoys partial tax deductibility throughout the period and some exemption from capital gains at the end of the investment.
• Although not explicitly illustrated, by reducing the overall coefficient of variation of their wealth portfolio, the Smiths may now be more willing to take greater risks in their stock portfolio in order to achieve a greater long-term return.

Midcareer Family: Status Quo Versus Moving Up

There are three primary housing price subclasses: below-median homes, median homes, and above-median homes. Exhibit 4 shows that while the average home price rose 2.45% in 2002, median homes were projected to increase in price by 2.33%, versus an increase of only 1.30% for high-priced homes. The higher-priced homes were anticipated to increase at roughly one half the rate of median-priced homes. The Smiths’ choices and the ultimate importance of home price changes can be considered accordingly.

Now the Smiths are midcareer professionals, in the 33% combined marginal tax bracket. Their median-priced home, purchased 13 years ago, is currently valued at $157,400. They are considering the financial implications of moving up to a $315,000 house that would result in mortgage financing $238,083, after subtracting an 8% selling cost and a down payment of their accumulated equity, or continuing to live in their present house. Under either scenario, the couple wishes to pay off the house by age 62, their planned retirement date.

Initially, Exhibit 5 appears to show that their relocating would result in a cumulative net wealth impact of $212,384, versus an impact of $692,913 if they do not move. For the sake of conservatism, assume that they invest the differences between the two homes’ after-tax cash outflows in Treasury bills. If the Smiths had employed any inflation-beating alternate investment of this $480,304 difference, the gap in wealth effect would widen substantially.

This comparison assumes that the price appreciation rate of the high-priced home equals that of a median-priced one. It may be more appropriate to consider the impact of lower price increases for more expensive homes (Exhibit 6). At a 5% rather than a 6% return, the wealth impact of the high-price home falls to $91,130. When the rate of price increase declines to 3%, the high-priced house wealth impact is negative by $110,209.

The Smiths’ initial decision to purchase a median-priced home instead of renting produced a sound financial result. When the Smiths considered the option of moving up, they found that doubling the investment in their home does not necessarily translate into doubling their investment gain. Four reasons for this are:

• The investment utility (the avoided rental cost) remains relatively constant when moving up;
• The high-priced home is in a fundamentally different market. The percentage investment returns characterizing median-priced homes may exceed the returns from higher-priced homes;
• The cumulative after-tax outflows are substantially larger from moving up than from not doing so; and
• Purchasing a larger home extracts an opportunity cost in the form of forgone investments due to higher cash outflows.

Retirees: To Sell or To Stay?

Finally, consider now that the Smiths are approaching their planned retirement age of 62. The current value of their home is the median $157,400, and they have no remaining mortgage. The three options they are currently considering are:

• Selling the house, investing the proceeds in an annuity, and moving into a rental unit;
• Taking a reverse mortgage on the house and continuing to live there; or
• Obtaining an 80% mortgage on the home and using the proceeds to generate money for mortgage and living expenses.

Under the sell-and-rent scenario, the $157,400 price net of 8% selling fees would yield $144,808, to be used for a joint-survivor annuity. Given a joint life expectancy of 22 years, the purchased annuity would generate $865 in monthly income, which falls short of the Smiths’ $1,469 rental expense. Assuming a rental increase of 3% annually, some selected differences between rental expense and annuity revenue are seen in Exhibit 7. A sell-and-rent scenario skews the out-of-pocket cash flows to the later years of retirement, when such payments may be more onerous, depending upon the Smiths’ other financial assets. The growing discrepancy seen in years 1, 7, and 22 illustrates the impact of the geometrically increased rate in rent, given an offsetting level annuity. This scenario does not allow the Smiths to participate in future home price increases, although maintenance, tax, and insurance expenses are eliminated. Selling the home eliminates any related estate tax issues. The cumulative sum of negative cash flows is $333,339 for the sell-and-rent scenario.

Many people consider the Smiths’ second option, the reverse mortgage, to be a last resort. Such a mortgage would yield $474 income per month, which is short of the $525 monthly expenses for taxes, insurance, and maintenance. Exhibit 7 shows the annual shortfall resulting from the higher expenses for years 1, 7, and 22. Again, the out-of-pocket cash flows are skewed toward the later years. The negative cash flows total only $67,916, compared to $333,339 for the sell-and-rent option. Note that the home appreciation of 3% annually results in an approximate $23,000 net home value (market value less accumulated reverse mortgage) after year 22, resulting in a negative net worth impact of $44,940. If the Smiths’ remaining lives are relatively short, any residual home equity will be included in their estate; should inflation increase, their heirs would benefit from a higher home value.

The Smiths’ last option is to take an 80% mortgage on their house. Under this plan, it is assumed that they could obtain a $125,920 loan on a 22-year mortgage with a $10,475 annual payment. For simplicity’s sake, this option assumes equal mortgage and a reinvestment rate of 6%, which initially generates interest of $7,555, approximating the $7,800 standard deduction threshold because it reduces the net cash outflow, thereby making the option more appealing. In the later years of the mortgage, the declining interest expense is less appealing. It is assumed that the Smiths could invest the $125,920 at 6%, to yield $7,555 in the first year of the mortgage and a declining amount each year thereafter, owing to the excess of mortgage payments over the after-tax investment returns. Exhibit 7 shows total cash outlay over the 22 years for mortgage and home expenses of $386,803, with the largest outlays occurring later in the mortgage life. After year 22, the value of the home is $301,595, which is $85,208 less than the total cash outlays over the period.

Exhibit 7 shows that the Smiths would experience a more severe effect on the net wealth impact ($333,339) by selling and renting than by using a reverse mortgage or an 80% mortgage. The mortgage would generate taxable income for the retirees, but would also subject them to some degree of investment risk, depending upon the level of income sought. Any potential negative inflation impact would be greatest under the sell-and-rent scenario because of possible rent increases.

Potential estate consequences differ, depending on the Smiths’ lifespan (Exhibit 8). If they are short-lived, both the reverse mortgage and the 80% mortgage will result in the home being included in the estate. If they have long lives, the reverse mortgage will largely deplete the home’s value, whereas the 80% mortgage will result in the home’s full value being included in the estate. The reverse mortgage scenario will yield mixed results in different inflation environments, whereas the 80% mortgage choice will result in home appreciation, which may benefit the Smiths’ heirs, but may possibly result in higher estate taxes.