MultiFamily Market Outlook - February 24, 2005 (Print All Articles)

Multifamily: Still a Great Investment

Share prices of the publicly traded multifamily companies tracked by the NAHB for its Multifamily Stock Index (MFSI) have performed incredibly well over the past six years, declining in only one year, and then by only 4%. By contrast, over the same period of time the S&P 500 has had a cumulative return of 5%, punctuated by annual returns ranging from a high of 34% to a low of minus 23%. While it is unlikely that the MFSI can continue to consistently outperform the S&P 500, it is highly probable that the MFSI will remain much less volatile than the market as a whole.

In January 2002, the NAHB introduced the MFSI to help the multifamily industry and investors better track the performance of public firms principally involved in multifamily ownership and management, and to allow for comparisons between the MFSI and other major stock indices. In order for meaningful historical comparisons to be made, we set the starting point for tracking the performance of all the firms that qualified for inclusion in the MFSI at December 31, 1998. 

Since then, the MFSI has increased by slightly over 100%, for a compound rate of return of better than 13.5% per year. During the same six years, the S&P 500 with dividends reinvested increased by 5.1% for a compound rate of return of 0.8% per year. That is, since December 1998 the compound rate of return for the MFSI has been 12.7 percentage points higher than the S&P 500 with dividends reinvested. As a result, $1,000 invested in the MFSI is now worth $2,141 but only $1,051 if invested in the S&P 500. 

The Year in Review

Over the past 12 months the MFSI and the S&P 500 have enjoyed very divergent rates of return, with the MFSI consistently outperforming the S&P 500 month after month. The year began with the MFSI 78% higher than the market, and that margin grew eight times in the past 12 months, setting four new records along the way. It closed last month at 104%, the highest January reading ever.

The very different historical rates of return exhibited by these two indices can be seen by looking at Figure 1. It shows that since the middle of 2000, the performance gap between the two indices has continually widened over time.

In particular, from late 1999 through December 2002, the S&P 500 declined by about 40% while the MFSI increased by 25%. Since then both indices have increased, with the MFSI rising by 50% and the S&P 500 by about 60%.

Figure 2 looks at the same data, but reports only the level of the indices on January 31 for each of the past seven years. Here the profoundly divergent growth rates between the MFSI and the S&P 500 are even more pronounced.

While the indices moved largely in tandem through early 2000, during the next several years the S&P 500 continually fell while the MFSI held steady at slightly below 1,500. As a result, the performance gap between the two indices grew very large by early 2003 and has remained very large ever since.  

This result is further explained by looking at Figure 3, which shows the 12 month rate of return ending January 31 for the past six years.

For the years ending January 31, 2001, 2002 and 2003, the MFSI dramatically outperformed the S&P 500. But in only the first of these three years did the MFSI have a significantly positive return. In the latter two years, the MFSI was virtually stagnant while the S&P 500 performed remarkably poorly. More recently, the S&P has regained its footing and has registered a positive return for two straight years for the first time since the end of the tech bubble of the late 1900s.

The Past Sheds Light on Future Trends

With all the earlier information in hand we now can closely examine Figure 4 (which shows the cumulative performance differential, in percent, between the MFSI and the S&P 500 since January 1999) for hints about future movements of the two indices. Recall that when the line goes up, the MFSI is outperforming the S&P 500, and when it falls the opposite is true.

Figure 4 shows that the past six years can be separated into three distinct periods. The first ends in early 2000 and is characterized by movement of the line around zero, suggesting that both indices exhibited similar behavior. The second period runs from March 2000 through September 2001, a period of only 18 months. During that time, the line rose 90 percentage points, from a starting point of –10% to a peak of almost 80%, recording six new highs along the way. To increase that much, that fast, is simply remarkable.  For purposes of comparison, during the 40 months since September 2001 (the third period), the line rose by only 25 percentage points. This illustrates that roughly 80% of the superior performance of the MFSI over the past six years is attributable to a brief 18-month span ending in September 2001.

A key difference between these three periods is interest rates. During the first period, the interest rate on 10-year Treasury bonds rose by 200 basis points, and both indices moved sideways. During the second period, the interest rate on 10-year Treasuries tumbled by about 200 basis points in an effort by the Federal Reserve to get the U.S. economy out of recession. And, in retrospect, it is now clear that it was the unique combination of a recession not brought on by a prolonged series of large interest rate hikes that set the stage for the dramatic rise in the MFSI when rates began their descent to extremely low levels starting in March 2000.   

Since September 2001, the decline in interest rates has been modest, only 70 basis points.  And, with long-term rates expected to rise over the next year, interest rates will not propel the MFSI like they did during most of 2000 and 2001. Moreover, with the economy no longer in recession, corporate earnings are likely to remain reasonably healthy. As a result, the rest of the market is likely to rise even as interest rates edge up. In short, we are entering a period of rate hikes and relatively good corporate earnings, precisely the opposite of what occurred during the heady days between March 2000 and September 2001. This suggests that the MFSI will be hard-pressed to maintain its historical edge over the rest of the market.

Multifamily Stocks are Less Volatile

Despite lower rates of return in the future, the MFSI should remain less volatile that the rest of the market. This is because of the significant role dividend payments play in most of the MFSI firms. To understand why dividends are so important in reducing volatility, consider how interest rate movements effect the price of two 10-year bonds: a zero- coupon bond, which behaves similarly to a non-dividend-paying stock, and a bond that pays interest on an annual basis, and which approximates the behavior of a REIT. 

Because the zero-coupon bond pays all interest and principle at the end of the 10 years, every interest rate change dramatically alters the present value of the zero-coupon bond, because all the money is discounted 10 years each time the interest rate changes. By contrast, because the regular bond pays interest annually, it has only the principal amount and the tenth interest payment due at the end of the tenth year, and thus is less affected by changes in interest rates. In other words, because the first nine interest payments are all due before the tenth year, rate changes affect those nine payments less than if they were due at the end of the tenth year.

Multifamily Stocks to Bring Lower, but Steady, Returns

Over the past six years, the MFSI has dramatically outperformed the market as a whole, as measured by the S&P 500. Between March 2000 and September 2001, the relative performance of the MFSI compared to the S&P 500 was stunning. with the return during that period accounting for 80% of the superior performance of the MFSI. Unfortunately, we are now entering a period of rising rates and good corporate earnings, conditions opposite to what prevailed during the critical 18-month period ending September 2001. Despite a high probability of lower returns in the future, the MFSI should remain considerably less volatile than the market as a whole because of the importance of dividend payments.

Starts — Steady On Into 2005

The rate of housing starts in buildings with five or more units basically held steady as 2004 drew to a close. Five-plus starts in December came in at a (seasonally adjusted) annual rate of 291,000 — up just 1.7% from November. The preliminary estimate for total five-plus units started in 2004 is 303,700 (a number that will be subject to revision over the next couple of months). If that holds, it would be down 3.7% from 2003. But 2003 was the strongest year for five-plus production since 1989, and anything over 300,000 has to be considered a success for the multifamily sector. (NAHB's forecast for 2005 calls for slightly under 300,000).

Although the rate at which  five-plus permits were issued slipped 5.7% to 367,000 in December, that nevertheless tied for the fourth-highest month in 2004. Moreover, the backlog of unused permits surged by 10.7% — so there's a lot in the pipeline, and that's a positive sign for multifamily starts at the beginning of 2005. Data for the 4th quarter are not yet available, but the condo share of multifamily starts rose dramatically through the first three quarters — from 24%, to 33%, and finally 40%. Demographics and other factors have favored condos for some time, and many industry analysts have been looking for a shift like this in the starts figures.

Real Rents Go Up

In December, real residential rents regained some of the ground they had lost since August, according to the most recent Consumer Price Index (CPI) numbers released by the Bureau of Labor Statistics. In nominal terms, the residential rent component of the CPI increased at a seasonally-adjusted annual rate of 2.3% (which is, in fact, slightly below its average rate of increase over the past year). But the overall CPI meanwhile declined at a 0.6% rate. While prices on most items crept up during the month, food prices remained flat, and energy prices dropped substantially. The movement in energy prices was driven, as usual, by a large change in prices for petroleum-based energy products.

As a result, the real rent index (which adjusts residential rents for inflation) moved back up from 107.5 to 107.8.  Although that's a substantial increase, the index had been at 108.0 earlier in the year, and as high as 108.4 in late 2003.

Forecast: Good Forward Momentum

The advance estimate of economic growth for the fourth quarter of 2004 was surprisingly weak, although an upward revision is virtually inevitable, and good forward momentum apparently carried into the early part of this year.

The employment report for January showed decent growth in payroll employment and a surprising drop in the unemployment rate. That conveyed mixed messages, as the glow of lower unemployment was clouded by the specter of a tighter labor market that could limit low-inflation economic growth down the line. The actual inflation situation remains remarkably benign, although core inflation has been firming up and the Federal Reserve has been concerned about a number of specific inflationary threats. As widely expected, the Fed hiked short-term interest rates by another quarter point at the February 2 FOMC meeting, raising the federal funds rate target to 2.5%. The FOMC decision was unanimous, and was coupled with approval of requests by all 12 Federal Reserve Banks for quarter-point increases in the discount rate.

Minutes released from the December 14 FOMC meeting revealed a lot of discussion about stubbornly low long-term rates. Some members argued that expectations of longer-term economic growth had been marked down, while others believed that the extended period of accommodative monetary policy created so much liquidity in financial markets that risk-taking still was being encouraged. It’s also likely that the shift in the Fed's policy since mid-2004 has held down inflation expectations and that the Fed’s ongoing commitment to a “measured” pace of adjustments has controlled expectations of future short-term rates -- keeping downward pressure on long rates. Whatever the proper explanation, it’s clear that persistent forces are holding long-term rates down.

As a result, we’ve just trimmed our forecast for long-term rates across the 2005-2006 period by about 30 basis points at year-end. We still expect moderate increases, partly because we believe that market participants have overly optimistic views about future inflation and the amount of policy tightening that lies ahead.