Category Archives: Asset Allocation

Goldman Sachs Predicts 4.5% 10-year Treasury Yields

Treasury BondGoldman Sachs just came out with a prediction that 10-year Treasury bond yields will rise to 4.5% by 2018 and the S&P 500 will provide 6% annualized returns over that same period.  The driver for this prediction is simply that the Fed is expected to raise the federal funds rate.

Because rising yields correspond to falling prices for bonds, Goldman’s forecast is that equities will substantially outperform bonds over the next several years.  If you are holding a bond yielding 2.5% (the current 10-year Treasury yield) and the Fed raises rates, investors will sell off their holdings of lower-yielding bonds in order to purchase newly-issued higher-yielding bonds.  If Goldman’s forecast plays out, bondholders will suffer over the next several years, while equity investors will enjoy modest gains.

Historical Perspective

This very long-term history of bond yield vs. the dividend yield on the S&P 500 is worth considering in parsing Goldman’s predictions.

Bond Yield vs. Dividend Yield

Source: The Big Picture blog

Prior to the mid 1950’s, the conventional wisdom (according to market guru Peter Bernstein) was that equities should have a dividend yield higher than the yield from bonds because equities were riskier.  From 1958 to 2008, however, the 10-year bond yield was higher than the S&P 500 dividend yield by an average of 3.7%.

Then in 2008, the 10-year Treasury bond yield fell below the S&P 500 dividend yield for the first time in 50 years.  Today, the yield from the S&P 500 is 1.8% and the 10-year Treasury bond yields 2.5%, so we have returned to the conditions that have prevailed for the last half a century. But the spread between bond yield and dividend yields remains very low by historical standards.  If the 10-year Treasury yield increases to 4.5% (as Goldman predicts), we will have a spread that is more consistent with recent decades.

Investors are likely to compare bond yields and dividend yields, with the understanding that bond prices are extremely negatively impacted by inflation (with the result that yields rise with inflation because yield increases as bond prices fall), while dividends can increase with inflation.  During the 1970’s, Treasury bond yields shot up in response to inflation. Companies can increase the prices that they charge for their products in response to inflation, which allows the dividends to increase in response to higher prices across the economy.  The huge spread between dividend yield and bond yield in the late 70’s and early 80’s reflects investors’ rational preference for dividends in a high-inflation environment.

What Has to Happen for Goldman’s Outlook to Play Out?

To end up with a 4.5% 10-year Treasury yield with something like a 2% S&P 500 dividend yield, the U.S. will need to see a sustained economic recovery and evidence of higher prices (inflation) driven by higher employment and wage growth.  In such an environment, investors will be willing to accept the lower dividend yield from equities because dividends grow over time and tend to rise with inflation.  This has been the prevailing state of the U.S. economy over the last fifty years.  Most recently, we had 10-year Treasury yields in the 4%-5% range in the mid 2000’s.  If, however, we continue to see low inflation and stagnant wages in the U.S. economy, bond yields are likely to remain low for longer.

Low Beta Market Sectors

With U.S. equity markets near their record highs and a bull market run that is starting its sixth year, the potential for a correction is a growing concern.  In addition, U.S. equity prices look fairly high when viewed in terms of the PE10 ratio.  Another factor that concerns some market watchers is that volatility (as measured by VIX) is at very low levels, reminiscent of 2007.  This type of complacency has historically been followed by increasing volatility, as levels return to their historical average, accompanied by a sell-off in higher-risk assets as investors adjust their portfolios to mitigate the effects of higher volatility.

Investors seeking to remain invested in equities at a target level but who want to reduce their exposure to market swings and to mitigate the impact of a rise in market volatility have historically been well-served by increasing their allocations to low-beta market sectors.  In this article, I will review the defensive value of low-beta allocations as well as examining the consistency of beta over time.

Beta measures the degree to which a security or a portfolio responds to a move in a benchmark index such as the S&P500.  A portfolio with beta equal to 80% (also written as 0.8) tends to go up 0.8% when the market rises 1.0% and vice versa.  Beta may be thought of as showing whether a security amplifies the moves in the benchmark (beta greater than 100%) or damps the moves in the benchmark (beta less than 100%).

How Beta Varies by Sector

The SPDR Select Sector ETFs provide a convenient way to break out the sectors of the U.S. equity markets by dividing the S&P500 into nine sectors.  These sectors illustrate how much beta varies.

Low Beta Market Sectors - 1

Betas and 10-year average annual returns for major sectors and indexes

The S&P500 has a beta of 100%, by definition.  Some readers may be surprised that emerging market stocks have beta of almost 140%, which means that emerging market equities tend to go up (down) 1.4% for every 1% gain (drop) in the S&P500.  Even before the market crash of 2008, emerging market stocks were high beta—this is not a new phenomenon.

There are three U.S. equity sectors with betas well below 100%: consumer staples (XLP), healthcare (XLV), and utilities (XLU).  It is often believed that low-beta equities have very low average returns.  In fact, a well-known but now widely-discounted model of equity returns (the Capital Asset Pricing Model, CAPM) assumes that beta of an equity or asset class corresponds directly to expected return.  High-beta asset classes have high expected return and vice versa.  Low-beta equities have historically substantially out-performed what would be expected on the basis of CAPM, however, and the past ten years is no exception.  These three sectors have all out-performed the S&P500 over the past ten years.  The return numbers shown here are the arithmetic averages, including reinvested dividends.

Low Beta Asset Classes in 2007-2008

The first question that is worth asking about beta is the degree to which beta corresponds to losses in really bad market conditions.  In the table below, I have tabulated beta calculated using three years of data through 2007 for each of the funds above, as well as the returns for each of these in 2008.

Low Beta Market Sectors - 2

Beta calculated through 2007 vs. 2008 returns

The three sectors with the lowest betas going into 2008 (consumer staples, healthcare, and utilities) had an average return of -22.3% in 2008, as compared to -36.8% for the S&P500.  An equity tilt towards these lower beta sectors could have reduced losses in that year.

Consistency of Beta through Time

The astute reader may notice that the betas calculated using ten years of data through May of 2014 (shown in the first table) are, in some cases, quite different from the betas calculated using three years of data through December of 2007 (shown in the second table).  Beta varies through time.  The betas calculated using three years of data through May 2014 provide an interesting contrast to the three-year betas through the end of 2007.

Low Beta Market Sectors - 3

Comparing betas for two 3-year periods

We are looking at two distinct 3-year periods, separated by almost six and a half years and, in general, low-beta sectors at the end of 2007 remain low-beta today and high-beta sectors back then are still high-beta.  The two most notable exceptions are international equities (EFA) and the technology sectors (XLK).  These changes notwithstanding, the three sectors with the lower betas in 2007 also have the lowest betas in 2014.

There are a number of factors that will determine whether any sector will weather a broad market decline better than others.  Beta is one important factor, but there are others.  In 2008, the financial sector suffered disproportionately large losses—well beyond what would have been expected on the basis of beta alone.  The underlying drivers of the 2008 market crash were most severe in the financial sector.  Small-cap stocks, by contrast, fell considerably less than the beta value of this sector would have suggested.

Low-Beta and Asset Allocation

Low-beta asset classes have historically provided some protection from market declines and increasing volatility.  There are a range of other considerations that potential investors should consider, however when creating a portfolio.  The selection of individual asset classes should be made with consideration of the characteristics of the total portfolio, including desired risk level, interest rate exposure, and income generation.  The target for total portfolio beta is primarily determined by an investor’s total risk tolerance.  A target beta level can be achieved both by choosing how to allocate the equity portion of a portfolio among sectors and by varying the balance between equity (stocks) and fixed income (bonds) investments.  Fixed income asset classes tend to have very low—even negative—values of beta.  In my next blog entry, I will explore these two approaches to managing beta at the portfolio level.

History suggests that low-beta sectors can provide some protection from market downturns.  The length of the current equity rally, and the substantial increases in equity valuations in recent years, are motivating some investors to consider their best defensive alternatives to protect against the inevitable reversal.  The question for investors to ask themselves is whether they are best-served by reducing portfolio beta by reducing their exposure to equities, by shifting some portion of assets from high-beta to low-beta sector, or both.

REITs and the Rally of 2014

There have been a number of surprises for investors in 2014.  Bonds have markedly out-performed stocks for the year-to-date.  The S&P500 is up 2.4% since the start of 2014, as compared to the Barclays Aggregate Bond Index, which is up by 3.5%.  Even more striking, the iShares 20+ Year Treasury Bond ETF (TLT) is up by 12.5% YTD and the Vanguard Long-Term Bond fund (VBLTX) is up by 11.4% YTD.  Interestingly, REITs have been also experienced a substantial run-up in 2014 so far.  The Vanguard REIT Index fund (VGSIX) is up 16.1% YTD and the iShares U.S. Real Estate ETF (IYR) is up by 14.4% YTD.

To understand the rally in REITs, it is useful to start with an overview of this asset class.  REITs are neither stock (equity) nor bond (fixed income).  A REIT uses investor money, combined with borrowed money, to acquire real estate.  The properties that they acquire can be office buildings (BXP, SLG), apartment buildings (EQR, UDR), healthcare facilities (HCP, HCN), or even digital data centers (DLR).  The returns vary between the specific types of property owned.  REITs make money from renting their facilities out.  Please note that I am discussing only equity REITs (those which own property).  There is a secondary form of REITs that buy baskets of mortgages.  These are called mortgage REITS (mREITs).

REITs are classified as ‘pass-through’ or ‘flow-through’ entities and pass at least 90% of their taxable income to their shareholders each year.  For this reason, REITs are often favored by income-oriented investors.

REITs have their own unique measures of value and drivers of performance.  Rather than price-to-earnings ratio, a measure of the valuation of stocks, the preferred measure of valuation for REITs is price-to-funds from operations (FFO).  A notable feature of REITs is their exposure to interest rates.  In general, stocks go up when interest rates rise, while bonds fall.  When interest rates decline, the reverse tend to be true: stocks decline and bonds rise in value.  REITs tend to have a fairly neutral response to changes in interest rates, although this varies.  To the extent to which REITs need to borrow in the future due to rolling credit or new financing, their costs rise when interest rates rise.  REITs often also have the ability to raise rents, however, so that their revenue can also rise when the costs of residential or commercial real estate rise with inflation.  In addition, because REITs own a physical asset (property), the value of the assets owned by the REIT tend to go up with inflation.

An examination of the performance statistics of REIT funds illustrates their properties quite clearly.

Chart 1

REIT funds vs. major asset classes (10 years through April 2014)—Source: Author’s calculations

The returns above are the arithmetic average of returns over the past ten years, including reinvested dividends.  Beta measures the degree to which an asset tends to amplify or mute swings in the S&P500.  The betas for VGSIX and IYR show that these funds tend to rise 1.3% for every 1% rise in the S&P500 (and vice versa).  These broad REIT funds tend to do well in rising stock markets and will also tend to perform worse than the S&P500 when the market crashes.  The S&P500 fell 36.8% in 2008, for example, and IYR fell 39.9%, for example.  The high beta is only one factor that determines the returns from these REIT index funds, however.  Despite a massive decline in real estate in 2008, the average annual return for these two REIT funds has been very high over the past ten years.  The volatility levels exhibited by these REIT funds is, however, very close to that of emerging market stocks, as shown above.

The best way to understand the unique features of REITs is to look at the correlations in the returns between asset classes, bond yields, and the returns from a traditional 60/40 portfolio (60% stocks, 40% bonds).

Chart 2

Correlations between monthly returns over the past ten years (through April 2014) for major asset classes, 10-year Treasury bond yield, and the returns from a 60/40 portfolio—Source: Author’s calculations

Equity indexes tend to have positive correlations to Treasury bond yields (because yield goes up when bond prices go down, and vice versa).  This effect is evident in the table above.  S&P500 stocks (represented by SPY) have a +30% correlation to the 10-year Treasury yield.  The same relationships hold for the other major equity classes in the table above (EFA, EEM, QQQ, and IWM), with correlations to the 10-year Treasury yield ranging from 20% (EFA) to 36% (IWM).  In 2013, Treasury bond yields rose and stock indexes gained substantially.  By contrast, the returns of bond funds (AGG, TLT) are negatively correlated to the 10-year Treasury bond yield.  The returns from these funds tend to be positive when Treasury yields are falling and vice versa.

Treasury bond yields follow interest rates and inflation.  When rates or inflation rise, bond yields rise because investors sell Treasury bonds to avoid getting caught with relatively low-yield bonds in a higher rate environment.  The selling continues until the yield on the bonds is in line with new rates.  This is why returns on the two bond funds are so negative correlated to Treasury yield.  When yields go up, bond prices fall and vice versa.

VGSIX and IYR have 6% and 7% correlations to the 10-year Treasury yield—which means that these REIT funds are not highly sensitive to movements in bond yields.  While investors were betting on rising yield in 2013, opinion seems to have shifted in 2014.  If you are looking for asset classes that don’t require a bet on whether interest rates will go up or down, REITs look pretty attractive.

The correlations between the returns from each asset class and the returns from a 60% stock / 40% bond portfolio show the degree to which adding these asset classes to a simple stock-bond mix is likely to provide diversification benefits.  The higher the correlation, the less diversification benefit you can expect.  The key idea in diversifying a portfolio is to combine assets with low correlations so that when one is losing money, others are likely to be doing something different (hopefully rising in value).  As compared to the major equity asset classes, the REIT funds have a lower correlation to the 60/40 portfolio.  This, in turn, suggests that REITs can provide more diversification benefit than equities to a portfolio that currently holds stocks and bonds.

In summary, the basic narrative to explain the rally in REITs goes something like this.  REITs provide substantial income compared to bonds and equities, as well as being essentially interest rate neutral.  While REITs are a volatile asset class, the relatively low correlation between REITs and equities provides diversification benefit to a portfolio of domestic stocks and bonds, as well as providing some modest protection from rising interest rates.  REITs appear have come back into favor after being sold off during and after the so-called taper tantrum of 2013, with yield-hungry investors leading the charge.  In a slow-growth environment, with the Fed indicating that they foresee an extended period of low rates, REITs look quite attractive.  The caveat to the positive view on maintaining an allocation in REITs is that there are substantial differences between the yield, risk, and interest rate exposure and that REITs have historically been a volatile asset class.  Unless you firmly believe that you can outsmart the broader market, REITs should be thought of as a long-term income producer and diversifier.  For those investors who have maintained a strategic allocation to REITs, the gains in 2014 YTD have helped to bolster portfolio returns as equities have provided very little to investors beyond their dividends.  The downside to this positive narrative, however, is that increased prices for REITs translate to lower yields for today’s buyers.

Related Links:

logo-folioinvesting

The brokerage with a better way. Securities products and services offered through FOLIOfn Investments, Inc. Member FINRA/SIPC.

 

Understanding How Fund Performance Comparisons Overstate Returns

May 13, 2014

Every investor has, I hope, read the standard disclaimer on mutual fund and ETF performance documents that ‘past performance does not predict future performance,’ or other text to that effect. Still, when you read a fund company’s statement that ‘x% of our funds have out-performed their category average’ or that ‘x% of our funds are rated 4- and 5-star by Morningstar,’ this seems meaningful. Similarly, if you read that the average small value fund has returns ‘y’% per year over the last 10 years, this also seems significant. There is a major factor that is missing in many of these types of comparisons of fund performance, however, that tend to make active funds look better than they really are (as compared to low-cost index funds) as well as making a mutual fund family’s managers look more skillful than they might actually be.

Continue reading

Different Perspectives on Home Ownership

The question of whether to buy or rent a home is of enormous economic significance for most families. Home equity represents the vast majority of American families’ net worth (see chart below). In a post in mid-2011, I discussed some of the major economic variables in the decision to buy a home. My conclusion was that buying a home made sense, but not because housing generates attractive long-term investment gains. The long-term data suggests houses have historically increased in value at a rate only slightly higher than inflation. The best arguments for owning a home, I argued, were historically-low interest rates, the mortgage interest tax deduction, and the inflation hedge provided by locking in your costs for shelter for the long-term. Since that time, housing prices and rents have risen dramatically. In the past week I read two thoughtful and interesting articles on the economics of owning a home that motivated me to revisit this topic for 2014.

Continue reading

A ‘New Era’ for Bonds?

The Wall Street Journal recently published an article titled How You Can Survive a New Era in the Bond Market.  The article suggests that investors adjust their bond allocations to tilt more towards high-yield (aka junk) bonds (both corporate and municipal) and global bonds, which tend to yield more than U.S. bonds.  This advice resonates with an Op Ed by Burton Malkiel, famed author of A Random Walk Down Wall Street, at the end of 2013.

The case against bonds is straightforward.  The best estimate for the expected future return from bonds is their current yield.  If you hold a bond until maturity, your total return will be very close to the current yield.  There are nuances to this rule.  With high-yield bonds, you should expect a total return that is a bit less than the current yield due to the fact that some of these bonds will probably end in default.  With bond funds, you don’t necessarily end up holding individual bonds until maturity, so the correspondence between current yield and expected return is a bit weaker.  Nonetheless, with current yields as low as they are, bond investors should not expect attractive returns from most bond classes.

Continue reading

60-Second Retirement Savings and Income Checkup

I think that the American public has largely tuned out the myriad studies showing that most households are woefully under-saving for retirement.  Even if we’d prefer not to think about this issue, however, it is crucial to regularly check on how we are doing.  There are two major questions.  First, during your working years, are you saving enough?  Second, during retirement, how much income can you sustainably plan to draw from your savings each year?  The good news is that there are some simple tools that you can use to do a fast estimate of how you are doing, how much you need to save to stay on track, or how to get on track. Continue reading