Many clients view real estate as an investment with stable returns, yet moderate risk. In this post, we take a closer look at real estate as a portfolio component.
Real estate is a hard asset. As such, its utilitarian value is independent of many risk factors, including inflation, or recession. When it comes to real estate as an investment, we expect it to appreciate, just like we would for any other investment. We can capture the appreciation of an investment in its total return, which we then decompose into two components: the capital gains, and the income stream from dividends or interest.
To capture the capital gains, we need a measure for the market value of our real estate investment. The Case-Shiller Home Price indices are designed to measure the price level of existing single-family homes in the U.S. and thus seem well suited for the task.
The Case-Shiller Indices use the “repeat sales method” to calculate their index. This method uses sales data on properties that have sold at least twice and captures their appreciation. This method is widely recognized as the premier methodology for indexing housing prices. The following chart shows some variants of the Case-Shiller Home Price Indices:
From the graph, we learn that real estate is overall developing reasonably similar across the nation. The metropolitan areas seem to appreciate a little better than the national average. Seattleites will, of course, insist that their city is quite different: the chart shows indeed that Seattle has been doing better than other metropolitan areas, especially since 2011, but was also hit a bit harder during the recent slowdown in Q4/2018. Moving forward, we will be using the U.S. National Index.
The Case-Shiller Indices only consider arms-length transactions, at which both the buyer and the seller act in their best economic interest when agreeing upon a price. This method excludes transactions between family members or repossessions of properties by mortgage lenders at the beginning of foreclosure proceedings. Also, because of using the repeat sales method, newly constructed homes are excluded from the index calculation process until they have been sold at least twice. We will get back to these methodology details later.
An essential part of an investment property’s total return is the income stream through the rent received. This income stream props up the total return and helps to offset price fluctuations.
We have not been able to find a reliable data-source for historical price-to-rent ratios. Instead, we have taken a different approach to model the rent income. As we found typical price-to-rent ratios to range between 15 and 30, we pegged the ratio to 17 at the beginning of our observation period. Moving forward, we are using the Home Price Indices to scale the property values, and the Consumer Price Index to scale the rent. While the absolute values might suffer from a constant offset, the development over time should reflect market reality well:
The chart shows nicely how the property values fluctuate against rental income. As a precursor to the 2008 recession, property values have hugely outpaced rents. The recession reset the relationship, and since then, the ratio is slowly creeping up again. Also, the ratio seems to show some softening since late 2018.
Now that we have captured both the appreciation in value and the income generated, we can calculate the total return of an investment property. Here is the chart:
This looks like a fantastic investment, combining excellent performance with low volatility. Often when things look that great, they are too good to be true. Let’s revisit the method of calculating the Home Price Index, which is the main driver for the returns. We notice a few shortcomings:
- The index uses 3-months smoothing, which suppresses a lot of the noise, but also some of the real price volatility.
- The index is based on actual sales, which are weighted to reflect their share of the market. This method can be tuned to work very well in times of equilibrium. However, when a market turns stale, and the total volume of properties sold drops, this will add to the smoothing effect.
- The index only captures sales at arms-length. Due to this, transactions under distress, as they occurred in 2008, are not included, further adding to the smoothing effect.
To summarize these shortcomings, it is very likely that the total return we modeled is significantly smoother than reality.
REITs to the Rescue
A real estate investment trust (“REIT”) is a company that owns and operates income-producing real estate. These REITs are often trade publicly on major exchanges, which makes them very liquid. This liquidity in turn makes sure that REIT prices immediately reflect all available market information. Let’s compare our simple model to a market index for REITs:
We can see how over the course of 30 years our model and REITs agree quite nicely, suggesting that we have identified the main drivers of investment return. Also, we could say that REITs are a good proxy for owning real estate.
However, we can also see the dramatic smoothing effect caused by using the Case-Shiller Index. This smoothing not only removes a lot of the volatility, but it also adds lag. As owners of real estate, we typically don’t experience volatility, simply because we don’t receive new quotes regularly enough.
REITs as a Portfolio Component
REITs create an equity curve as ragged as the stock market, so it is probably fair to compare these two. The following chart compares the equity curves on a logarithmic scale, and also adds their drawdowns:
We make the following observations:
- Over the past 30 years, the total return of real estate was about on par with the stock market.
- The 2002 recession has only marginally affected real estate, while during the 2008 recession real estate has done much worse than the stock market.
- Starting with the 2008 recession, the real estate market seems strongly correlated to the stock market.
Beta is a measure of co-movement between two assets. A beta of 1 suggests that for each 1% change in the benchmark, the asset will also change by 1%. The following chart shows the 1-year rolling beta of REITs versus the S&P 500:
The chart shows how, over the past 30 years, real estate typically had a positive correlation to the stock market. This correlation was especially augmented during the 2008 recession, which, of course, originated in real estate. However, the beta has declined since and ranges around 0.5 in mid-2019.
This positive beta suggests that a portfolio combining stocks with REITs is not as diversified as we might wish. Still, because the beta is most often smaller than 1, combining both asset classes will improve the overall risk profile of a portfolio. The following chart compares a portfolio created from 50% stocks and 50% REITs to the S&P 500:
It is not obvious if this is truly an improvement. Here are our observations:
- In the years from 1988 to 2000, a portfolio combining Stocks with REITs was trailing a pure stock portfolio.
- During the 2001 recession, the portfolio greatly benefitted from the real estate component.
- During the 2008 recession, real estate even increased the portfolio’s drawdown.
- In the years after 2009, the portfolio performs only marginally different from pure stocks.
While real estate is certainly no cure-all, we tend to believe that real estate is an excellent addition to most portfolios, as it helps with diversification. If we assume that the next recession, unlike the previous one, is not tightly linked to real estate, this diversification will improve our portfolio’s risk profile.
There are multiple approaches to investing in real estate. In this post, we have shown how REITs behave quite similarly to owning a rental property. Selecting a single investment property is as tricky as picking a single stock to buy and hold. REITs address this issue through their broad diversification. Further, investments in individual properties don’t scale easily. Investors will face a reinvestment risk, which REITs address through their affordable shares. Mostly, the decision is a matter of personal preference.
To make these findings on real estate investing actionable, we need to decide upon the optimal allocation towards real estate. We believe that investors should consider all of their assets when making portfolio decisions. However, when doing so, we find that most investors already have more than enough real estate in their portfolio: their primary residence.
We very much hope that we’ve been able to explain the role that real estate can play in your portfolio. At Bertram Solutions, we pride ourselves in doing independent research, and this post is a good example of our approach to analyzing investments.