Is Gluskin's David Rosenberg Right about Utilities?

The list that emerged includes both index and actively managed funds.  Expense ratios vary considerably – compare XLU (0.18%) to  GABUX (1.4%), for instance.  The ETFs all track utility indices, which, I should note, are not necessarily as standardized as the Dow Jones Utilities Index.  FXU, for example, tracks a proprietary index called the StrataQuant Utilities Index.  Similarly, DBU tracks the Wisdom Tree Global ex-U.S. Utilities Index. 

Some of these funds hold non-U.S. positions that generate their earnings in currencies other than the U.S. dollar, making their returns subject to currency risk.  I focus on the aggregate risk levels associated with funds, and that currency risk is reflected as a component of the total risk in a fund. 

These funds have a wide range of yields.  To determine which are more attractive for income investors, I started by comparing the yields to the funds’ historical risk levels.  In general, riskier funds offer higher yields, but that is not uniformly the case. 

The chart below plots the 12-month yield of each fund against its trailing three-year volatility.  I have also used an optimizer to identify the “frontier” portfolios that deliver the maximum yield for each level of risk (the solid red line in the chart below).  For purposes of comparison, I have added yield versus risk for five portfolios that represent varying mixes of the S&P500 and an aggregate bond index – these are the green circles. (The green circle on the far left of the chart holds 60% bonds and 40% stocks.  The green circle on the far right holds 70% stocks and 30% bonds.)

For context, the Barclays Aggregate Bond Index (AGG) has a yield of 2.4% and the iShares 7-10 Year Treasury Fund (IEF) has a yield of 2.06%. 

Yield versus trailing three-year volatility for utility funds

The funds that deliver the highest yield at each level of risk (those that lie on the yield frontier) are all index ETFs: IPU, JXI, and XLU. The fixed income asset sub-class that compares most closely with the utility yield frontier is investment-grade corporate bonds (as represented by LQD), with a yield of 3.97%. 

There is no mandate, I should note, for the managers of these utility funds to make yield their primary goal.  In general, active fund managers seek to maximize total risk-adjusted return (alpha).  For this reason, I’m not surprised that the yields of the actively managed utility funds are below the risk frontier.  There’s a cluster of funds, with yields between 3.2% and 3.6%, that come quite close, though.

On the basis of the chart above, it is clear that many of these utility funds provide substantially more income than the aggregate-bond index, intermediate-term Treasury bonds, or generic portfolios of stocks and bonds. 

Utilities in the optimal yield portfolio

To more definitively determine whether utilities are an attractive asset class for portfolios designed to maxize income at a given level of risk, I ran optimizations to determine what asset allocation, among various combinations of utilities and fixed-income asset sub-clases, provides the most yield at each risk level. 

First, I selected a set of fixed income asset sub-class ETFs.  I also included a total market stock index (VTI) and both domestic and international utility ETFs.  Because the risk level of the S&P 500 over the past several years has been lower than the implied volatility of long-dated options on the S&P500, I set the forward-going volatility for the S&P 500 to be equal to its current implied volatility (21%).   I then ran a Monte Carlo optimization for the projected risk and return of various portfolios composed of these asset class ETFs. I used Quantext Portfolio Planner, a tool I developed, for this purpose.  Matching the volatility of the S&P 500 in the Monte Carlo simulation to the option implied volatility is part of the process of setting a forward-looking equity risk premium.  The volatility of the S&P 500 impacts the volatility of other risky assets.

The fixed income asset sub-classes included in the analysis are short-term Treasury bonds (SHY), intermediate-term Treasury bonds (IEF), long-term Treasury bonds (TLT), and investment-grade corporate bonds (LQD).  I also included an aggregate bond index (AGG). 

I ran the optimization to identify the portfolios with the highest yield for a series of target risk levels (volatilities) ranging from 6% (less than a third of the volatility of the S&P 500) to 16%.  For comparison, the projected volatility of LQD is 7.6%.  The results are shown below. 

To display all of these data in a single table, I have included the tickers, but not full names, of the funds; the full names are listed in the earlier table.  The first four funds (starting from the top of the table below) are the utility index funds that I previously found were on or near the yield frontier.  VTI is the Vanguard Total Market Index ETF, and the next five funds are fixed income ETFs.

Optimal Yield Portfolio Projections

For a portfolio with projected 6% volatility, the optimal portfolio yields 3.89%, and it is primarily allocated to bonds (in this case, 59.9% corporate bonds and 15.2% long-term Treasury bonds).  This portfolio has a 25% allocation to utility stocks.  The intermediate Treasury ETF, meanwhile, has projected volatility of 9% and a yield of 2.06%.  (The implied volatility of at-the-money put options on IEF expiring in January of 2014 is actually slightly higher than 9%, indicating I am conservative in my projections.) In other words, it is possible to get double the yield of intermediate Treasury bonds with less risk by investing almost entirely in a combination of corporate bonds and utilities. 

As we move to more volatile projections, the optimal portfolio contains fewer and fewer bonds and more and more international utility stocks (represented by IPU).  On the basis of yield versus risk, the international utility ETFs are more attractive than the domestic utility funds.  The iShares long-term Treasury ETF (TLT) has a yield of 2.8%, and this fund’s projected volatility is 20.9%, almost as volatile as the S&P 500.  This is a remarkably high estimate of volatility, but the longest-dated options on TLT confirm that this number is a good estimate.  You can build a portfolio almost entirely from utilities and corporate bonds with a projected volatility of 10% (less than half that of TLT) and a yield of 4.25% (which is 1.5 times the yield from TLT). 

As a further step, I performed optimizations using the funds with non-zero allocations from the table above, along with a series of individual utility stocks.  I did so because there is no particular reason to assume that the utility indices are the most efficient yield-generating combinations of utility stocks.  The resulting optimal portfolios are shown below.  (Again, to conserve space, I have identified stocks and funds only by their tickers.  The full names for all stocks and funds are given in an appendix.) In the chart below, individual stocks are shown in white, utility funds are shown in blue, and the two fixed-income funds are shown in salmon. 

Optimal-yield portfolio projections including individual stock allocations

For the lowest-risk portfolios, with projected volatility of 6%, including individual stocks increases the yield by 0.5%.  The added yield from including individual stocks increases with portfolio risk level, but not dramatically so.  The chart below summarizes these results.

Yield frontier with and without individual stocks

Allocations as high as 10% to individual utility firms expose a portfolio to heightened stock-specific risk.  Only advisors and investors who are comfortable with stock-specific risk should attempt to gain this extra 0.5% to 0.75% in additional yield.

Discussion and conclusions

While income investors have always favored utilities, the current environment makes them even more attractive.  The exceedingly low yields on Treasury bonds right now are accompanied by substantial risk, especially interest-rate risk. 

And the yields provided by utilities – particularly international utilities – are especially attractive.  While international utility stocks have currency exposure, their higher yield more than offsets that incremental risk. 

David Rosenberg, in his September 6 commentary, wrote that investors have historically bought bonds for yield and stocks for price appreciation.  Today, the opposite is true. 

Investors today buy Treasury bonds to benefit from ongoing quantitative easing and the resultant price appreciation; anyone seeking yield must now look to stocks, and utility stocks in particular.  But this makes sense only if one considers both yield and risk.  A portfolio consisting mainly of investment-grade corporate bonds and utilities, with only a modest allocation to Treasury bonds, provides the maximum yield at each risk level. 

My calculations show Rosenberg is right: Utilities and corporate bonds are considerably more attractive yield generators than Treasury bonds.

Geoff Considine is founder of Quantext and the developer of Quantext Portfolio Planner, a portfolio management tool.  More information is available at www.quantext.com.

Geoff’s firm, Quantext is a strategic adviser to FOLIOfn,Inc. (www.foliofn.com), an innovative brokerage firm specializing in offering and trading portfolios for advisors and individual investors.

Appendix: Stocks and funds analyzed

Read more articles by Geoff Considine