In the previous post, I discussed whether or not buying a house today is a smart financial decision. I recommend reading it before reading this post, but if you’re a millennial who hates to read a lot, then the short answer would be no. Or probably not.
Not everyone wants to buy today. Some are considering various alternatives for investment. Some do not have enough free cash flow, while others are waiting for a bubble to explode. Among millennials, some don’t even want to buy a house at all. So, you need to determine not only if but when you should buy a house. To help you decide, we will take a brief tour and see if buying a house has ever been a smart financial decision.
When we try to assess the profitability of investment, we need to make the assessment based on what we would do in each case. Checking historic returns, however, can sometimes lead to a cognitive bias, as we are tempted to analyze results in retrospect. For example, we now know that investing in the San Francisco housing market five years ago would have been a good investment, but would it have been good if we had thought about it in 2012?
Real estate vs. stocks—Empirical comparison
Thanks to the Case-Schiller index, we have data from 1987 to 2017 for the top metropolitan areas in the U.S. Moreover, we also have national data from 1890. According to the national data, the real (inflation adjusted) home price index rose by 63% in this 127-year period. The nominal home price index was multiplied by 51 during this same time, but I remind you that Coca Cola cost a nickel until 1959.
I don’t even need to refer to historic stock market returns to know they are well over 63% for the same period. If you had invested $100 in 1928, that money would be worth $328,000 today—a nominal multiple of 3,280—and the average return of stocks is in the range of 7%-12% (depends on how it is measured).
Many methods are used to measure the return of stocks, and such measurement needs to be adjusted for dividends. If you want to get the real value, it also has to be adjusted for inflation. Measuring real estate returns is even more complicated. You can’t live in the stock market, so if you want to put a roof over your head, you can’t invest all your money in stocks, and the return of your house should reflect the rent that you’re not paying (the cost of housing services). This is even before considering the mortgage, which costs you in interest and hurts the value of your money when you are young and compounding should be like a neon light flashing on a dark road. A mortgage does allow you to leverage your money, however, and thus leverage the value of your assets—something that a regular investor does not do with stocks.
Still, according to Robert Schiller, Nobel Prize winner and the inventor of the Case-Schiller index, if you remove all forms of dividends and compare the Standard and Poor’s U.S. composite stock price index since 1871 with Shiller’s own real U.S. home price index since 1890, stock market capital gains have outperformed the housing market’s capital gains. Here’s a graph showing the differences since 1975:
Real estate vs. stocks—Theoretical comparison
Stock markets carry a bigger risk than real estate. The stock market standard deviation is bigger than that of the real estate market, and, consequently, a bigger reward is anticipated. But standard deviation matters mostly when we discuss short periods of time, as over long periods we assume that the return would be close to the expected value. What is the expected value? Past performance is no indication of future results, but as the sample becomes bigger, we have more clarity regarding the projections of the behavior of any financial vehicle.
What we know about stocks is that their returns, sometimes double digit, can make your money worth much more over time and usually more than returns from real estate. How much bigger? A stock’s value should reflect the company’s earnings. Let’s imagine that company X is selling product Y. If the economy is growing, the company will probably grow with it. If inflation sets in, the company would probably raise its prices. Also, if the company is doing well, it will also pay dividends to its shareholders.
From an economic perspective, then, we would anticipate that stocks would rise in real GDP growth and the growth of inflation together with some extra dividends. Accordingly, the components for stock growth are:
- Inflation—worldwide, central banks declare that their goal is an inflation rate of 2%. If inflation is indeed 2%, then we expect 2% derived from this component.
- Real GDP growth—if the economy is growing, we expect that our stocks will rise at least at the same rate. Generally speaking, public companies should rise more than the GDP, since the GDP is also comprised from governmental expenses, which usually moderate the growth.
- Others—dividends also contribute to the yield, while fees deduct a bit from the yield.
All three components are usually worth about 5%–8%, but the inflation rate in the U.S. today is 2.2%, and the real GDP growth is approximately 2%, yet stocks are beating the accumulated 4%–5% by a wide margin (the S&P 500 had a 10% return in 2016 and is poised for a bigger return in 2017). Why? First, the big indices are influenced by top stocks, which are growing fast and strong nowadays, such as the FANG. Also, the low interest rate puts more money on the market. Second, when times are bad, stock prices will probably go down more than the GDP and the inflation rate, which takes us back to the expected value over time.
Real estate vs. stocks—Normative comparison
Let’s elaborate on the theoretical side of the comparison and see what would be the normative comparison, meaning what we would expect the results to be. I mentioned before that it would be reasonable to expect a 5%–8% annual return for stocks. So what would we expect for real estate?
Let’s consider the components of real estate price growth:
- Inflation—as in the stock market, we would expect that the price would rise at the rate of inflation.
- Real GDP? Why would the GDP affect the real estate market? It can allow people to acquire more housing services than they needed before, but I expect to see that the correlation here is weak—out of every new dollar earned, some money will go to consumption, and some will be saved, and out of every new dollar used for consumption only a fraction would go to the real estate market (to consume more expensive housing services).
- Other factors may change the demand and supply sides of the curve. For instance, a high net migration to the city not followed by the same increase in the number of houses would push the numbers up, but is it a sustainable gain? Probably not, as the increase in prices would make the return of the market bigger for the project initiators and thus increase the number of housing starts. Also, the government might be forced to release more projects to the market. In the long run, an increase in the demand side should not increase prices, but they would be affected in the short term.
From a rational point of view, the real estate market as a whole should increase no more than the rate of inflation. Most of the increases that are more than the inflation rate are primarily in specific places. For example, the prices in Manhattan rose during the last few centuries more than the rate of inflation because in some places demand will always be stronger than the supply, as there is a constant shortage on the supply side. In general, though, over long periods of time, the real estate premium over inflation is slim. While looking for long-term statistics, I found this graph:
This graph shows the prices in a neighborhood in Amsterdam, a city that was well established even at the beginning of the seventeenth century. The graph shows that even though prices fluctuated from 1628 to 2008, real prices increased by only 100%, which may seem to be a lot, but for a period of almost four hundred years it’s only a 0.35% increase per year. As you can see in the graph, a few years before 2008 the real return was flat or even negative.
As with the Manhattan example before, I believe that real estate prices can overachieve only if the available residential area is limited, for example, as a result of topographic boundaries or other limitations on a city’s growth. For instance, a place where real estate prices have increased and exceeded the increase in stock prices is Singapore, which is a small island that has experienced an economic boom over the last few decades, as you can see in the table below.
Even in Hong Kong, which has geographic similarities to Singapore, the stock market has beaten the real estate market.
Empirically, buying stocks has been better than buying real estate over the last century. Theoretically and normatively, I think this trend will stay this way, as stocks are destined to gain more over long periods of time than real estate. I think that the primary problem with real estate is not even those mentioned above but rather that people tend to ignore the most important rule of financing—Diversify your portfolio.
By buying a house, you put all your cards in one city, in one neighborhood, in one house. Imagine that the city is in economic turmoil. Not only is your job at risk, but now your biggest asset has also depreciated dramatically. Just think about people living in Detroit a few years ago, working with Ford/GM and owning a house in the city. Stocks are a different ball game. You can buy ETF of stocks, you can put your money in indices of overseas economies, and you can even buy commodities or bonds and get a very wide exposure to the world economy. So, before investing in real estate, just keep in mind that diversification will be much harder when all your eggs are in one basket and the risks can be very specific. Watch out.