Despite coming off of the worst housing recession in recent years, Americans’ enthusiasm for buying a home remains remarkably strong. A recent survey conducted by Zillow and Pulsenomics LLC found that almost two-thirds of Americans think that buying a home is the best long-term investment a person can make.
But is all this enthusiasm warranted? Is buying a house really a good investment? There are plenty of non-monetary benefits to owning a home. Owning a home and putting down roots has always been intimately tied to the American Dream. But what about the pure dollars and cents—is buying a home a good financial call? As always, it depends on your individual situation, but generally the data suggest that compared to the stock market, buying a home has produced similar or better returns with less risk—especially over longer horizons.
If you compare the increases in the prices of homes and stocks by looking at the S&P 500, stocks seem much more attractive. Since 1975, the S&P 500 has increased more than twentyfold while over the same period, the Zillow Home Value Index, which tracks the value of the median house in the United States, has only increased in price fivefold.
Given this fact, how can we say that housing is a good investment? Because price increases aren’t the whole story. Stocks pay dividends—if you buy a stock and later sell it, your total return depends on how much you sold that stock for, but also on any dividends you received while you owned it. Similarly, as a homeowner you don’t just get the increase in the price of your home, you also get the benefit of being able to rent it out to someone or live in it yourself for free (later we will talk about how mortgages affect this analysis).
Ignoring these dividends and rents would make our analysis incomplete, and when we incorporate them using the Zillow Rent Index to measure rents, the results change dramatically. Once dividends and rents are included, and after accounting for taxes over the period from 1975 to present, the annual return of the S&P 500 is 10.4 percent. The annual return on housing is 11.6 percent. That might sound like a small difference, but over a long period of time it can compound to quite a big difference. An annual return of 11.6 percent means that over 40 years, a dollar would grow to almost $80. With an annual return of 10.4 percent, that same dollar would grow to just over $52 during the 40-year period. A difference of only 1.2 percent each year compounds into a difference of more than 50 percent over 40 years.
All things being equal, a higher return is better. But it is important to also consider the risk involved. Stocks, for example, tend to have a higher return than Treasury bonds. But Treasury bonds are much safer, so it is perfectly sensible for someone to prefer to buy a bond over a stock.
How does the risk of the stock market compare to the risk of the housing market? Generally, it will depend on your investment horizon—the longer the horizon, the smaller the risk. The figure below shows a common measure of the risk of an investment: the standard deviation of returns. The higher this measure is, the more volatile the investment. This standard deviation of return for both stocks and housing is calculated at various horizons using data from 1975 to present.
As the figure shows, the longer you hold either investment, the less risky they become. Over short periods of time there can be large swings, but generally those swings tend to cancel each other out in the long-run. However, at every horizon, homeownership appears to be the less volatile, safer investment.
One factor we haven’t yet considered in this analysis is the impact of a mortgage on return and risk. When comparing buying stocks with buying a house, in both cases we assumed that the buyer would put down cash to make the purchase. In reality, of course, most people who buy a home do so with the help of a mortgage.
Buying a home with a mortgage is an example of leverage—borrowing money to buy an asset. Leverage will generally increase the return, but also the risk. If you borrow $80,000 to buy a $100,000 home, you only have to put down $20,000 up front. If the home’s value increases to $120,000, you have just turned $20,000 into $40,000 (your share of the $120,000 after accounting for the $80,000 loan). But if the value of the home falls to $80,000, then you have effectively lost your entire $20,000.
Stocks can also be bought using leverage (known as buying on margin), but buying on margin is generally more expensive than getting a mortgage to buy a home. Moreover, because the interest paid on a mortgage is tax deductible, there is a tax benefit to buying a home with leverage that investors in the stock market do not get. This makes housing an even more attractive asset.
Based on the numbers, it appears that buying a home can be a very good financial decision compared to investing in the stock market. However, as the recent housing recession highlights, there are definite risks. In both the housing market and the stock market, past performance is never a guarantee of future returns—just because housing has historically been a good investment does not promise that it will continue to be so in the future. Also, unlike stocks, it is very costly and time-consuming to sell or buy a home. A potential homebuyer should consider that if something unexpected happens like a job loss, they might find themselves trapped with an expensive mortgage. Nevertheless, based on the data, it appears that Americans’ love for housing is not unfounded, and for the right buyer in the right situation, housing can be a great investment.
Favilukis, Jack and Ludvigson, Sydney C. and Van Nieuwerburgh, Stijn, The Macroeconomic Effects of Housing Wealth, Housing Finance, and Limited Risk-Sharing in General Equilibrium (May 2010). NBER Working Paper No. w15988.
Giglio, Stefano and Maggiori, Matteo and Stroebel, Johannes, Very Long-Run Discount Rates (April 2014).
 To extend the sample, we chain the ZHVI backwards using the Federal Housing Finance Agency’s House Price Index. Similarly, later in the brief the Zillow Rent Index is chained backwards using data from the U.S. Census.
 We compute returns on the S&P 500 using the following formula: Returnt+1=(Pt+1+Dt+1)/Pt where Pt is the price of the index in quarter t, and Dt are the dividend payments in quarter t. Similarly the return to housing is calculated as: Returnt+1=(Pt+1+Rt+1)/Pt where Pt is the median value of homes, Rt and is the median rent of homes. The returns on housing are then adjusted to account for depreciation and maintenance (1 percent a year) and property taxes (.67 percent a year after accounting for the deductibility of property tax from income taxes), while the returns to S&P 500 are adjusted for capital gains tax rate of 15 percent. These numbers are calculated quarterly and then expressed as an annual return.
 This is broadly consistent with the academic literature on returns to housing computed using a variety of methodologies. See Favilukis, Ludvigson and Van Nieuwerburgh (2010) and Giglio, Maggiori and Stroebel (2014) for example.