In Part I of this article, I explained why I have issues with the traditional idea that individuals should provide for their required level of retirement income (beyond what is provided by Social Security and any pensions) entirely with assets with zero risk of loss of principal (e.g. Treasury bonds). In Part II, I discuss the alternative approaches.
There are two investments that have zero loss of principal: traditional Treasury bonds and Treasury InflationProtected Securities (TIPS), which are Treasury bonds with embedded protection against inflation.
I agree with the notion that people need to save and invest so as to be able to provide a very reliable and consistent income stream in retirement. Zvi Bodie has presented a compelling argument that investments in stocks do not become less risky as you hold them for longer periods, so that investors cannot rely on stocks as part of their required income stream. I have performed detailed analysis of Bodie’s argument and I agree with his argument: the magnitude of loss that you can face with an equityheavy portfolio increases the longer you hold the portfolio. As I noted in Part I, William Bernstein has recently advocated for a portfolio in which all of your required income is provided by Treasuries and annuities, largely consistent with Bodie.
I have two reasons why I disagree with the notion of an allTreasury portfolio, whatever the mix of TIPS and traditional Treasury bonds:
1) The required level of wealth for people to retire on a 100% Treasury bond portfolio is entirely out of reach for almost everyone.
2) The real return (return after adjusting for inflation) for long dated TIPS is below zero.
To illustrate and explain my positions, I am going to analyze a portfolio designed for conservative investors who are near or in retirement, the Folio Investing Target 2010 Conservative Folio. This is a portfolio allocation designed for a conservative investor who retired in year 2010 or thereabout. The definition of how much risk such an investor can take is, admittedly, somewhat arbitrary.
As of this writing, the yield on 20year TIPS is right at zero. For longer maturities, the yield turns very modestly positive and for shorter terms, the yield becomes more negative. I believe that it is reasonable to estimate that a TIPS portfolio built for a 30year retirement will have an average of zero real yield (no yield beyond keeping up with inflation), given the current yields in the market. In other words, for simplicity, I am simply going to estimate that a TIPS ladder will have zero real yield and zero risk. This means that a 100% TIPS portfolio will keep pace with inflation, but deliver no return beyond inflation. An investor planning on a 30year retirement can provide a riskfree constant inflation adjusted income of $33,333 on a $1 Million portfolio invested entirely in TIPS. This portfolio will last exactly 30 years. If an investor retires at age 65 and invests 100% in TIPS, his portfolio will reach zero value at age 95.
The Target 2010 Conservative Folio
Folio Investing maintains a series of portfolios designed on the basis of expected retirement date (Target Date Folios) and the Target 2010 Conservative Folio is part of this series. The Target 2010 Conservative Folio holds 79.5% of its assets in bonds of various types and 20.5% in equities and commodities (see table below).
Symbol

Company

Weight

Vanguard Bd Index Fd Inc Intermed Term

3.50%


Vanguard Bd Index Fd Inc Total Bn

16.75%


Vanguard Bd Index Fd Inc Short Trm Bond

18.00%


Powershares Db Cmdty Idx Track Unit Ben

2.75%


Ishares Trust S&P Natl Mun B

16.00%


Powershares Global Etf Trust Insur Natl

18.00%


Ishares Tr Us Tips Bd Fd

7.25%


Vanguard World Fds Energy Etf

14.75%


Vanguard World Fds Inf Tech Etf

1.50%


Vanguard World Fds Utilities Etf

1.50%

Source: Folio Investing
[Note: The holdings are reevaluated and updated (as needed) on a quarterly basis.]
The portion of the portfolio not allocated to bonds is primarily in energy stocks (VDE). An additional 2.75% of the portfolio is allocated to commodities (DBC). Both of these asset classes are intended to provide a hedge against inflation since the prices of energy and other commodities are a major component of inflation. The remaining 3% of the portfolio is allocated to technology stocks (VGT) and utility stocks (VPU).
The bond allocation in this portfolio includes a range of classes, including traditional Treasury bonds, TIPS, municipal bonds, and some corporate bonds (through the inclusion of corporates in an aggregate bond fund).
The Target 2010 Conservative Folio has a trailing 12month yield of 2.61% as compared to a trailing 12month yield of 2.07% for IEF, the iShares Barclays 710 Year Treasury Bond Fund. TIP, the iShares Barclays TIPS Fund, has a trailing 12month yield of 2.7%. The 2010 Conservative Folio’s mix of assets has a trailing fouryear risk (volatility) level that is below that of IEF. The equity portion of this portfolio reduces the sensitivity to interest rates—a key concern. If rates rise, the value of a portfolio of Treasury bonds declines. The asset allocation of the Target 2010 Conservative Folio (2010C), by contrast, is estimated to be neutral with respect to interest rates.
To determine how much income this type of portfolio could provide, I have run a Monte Carlo simulation that generates thousands of possible future outcomes for all of the various assets in the portfolio, accounting for the relationships between each asset class. Monte Carlo simulations are not perfect, of course, and rely on the quality of all of the assumptions built into the model. Under the Monte Carlo projections, the 2010C portfolio is projected to generate an average annual return of 4.5% per year, with annualized volatility (risk) of 4%. To provide some context for this result, the Monte Carlo projections generate annual average return of 3.1% per year for TIP, the iShares Barclays TIPS ETF and 2.2% per year for BND, the Vanguard Total Bond Market ETF. The trailing 12month yield for TIP is 2.7% and the trailing 12month yield for BND is 2.93%, according to Morningstar. Without getting too detailed here, simply note that the Monte Carlo projected return for BND and TIP are fairly close to their trailing 12month yields. In other words, the Monte Carlo projection suggests that the future expected total return is similar to the trailing yield for these bond funds, consistent with a common estimate that expected return for bonds should be roughly equal to their yields.
The model portfolio could provide a range of possible returns in any given future year. For this reason, the ability of the portfolio to sustain a specific stream of income is expressed in terms of statistics. For our model investor, retiring at age 65 and drawing $33,000 per year in income and increasing that amount by 3% per year to keep up with inflation, there is a 95% probability that he will be able to sustain this income level until age 96.
Naturally, anyone would prefer a 100% probability of funding their retirement for 30 years (the 100% TIPS solution) vs. a 95% probability of funding retirement for 30 years. The problem, of course, is that with the 100% TIPS portfolio you are also certain to have zero remaining in your account after thirty years. With the Target 2010 Conservative Folio, the median value of your portfolio is projected to be $700,000 at age 95.
Clearly, there are uncertainties and risks in all solutions. In the case of a 100% investment in TIPS, you are totally certain to be able to fund your $33,000 per year, and increase that amount with inflation, but you are also guaranteed to have nothing left at age 95. With the Target 2010 Conservative Folio, you have the risk that your investments will deliver returns below expectations, as well as, the risk that the Monte Carlo model is creating a statistical projection of the future that is too rosy.
The Target 2010 Moderate Folio, a portfolio designed for investors retiring in 2010 but who are able to bear a somewhat higher risk level, can sustain more income and with a higher probability of funding a 30year retirement than the Target 2010 Conservative Folio. I chose the Target 2010 Conservative Folio for this discussion because it has the lowest level of investment risk of any of the Target Date Folios. There are many different portfolio choices that an investor might make, depending upon his or her risk tolerance.
Conclusions
The question for an investor at the start of retirement is the degree to which he or she is willing to bear some level of uncertainty in future outcomes. TIPS provide complete certainty of preserving the current purchasing power of your money (keeping up with inflation) until some point in the future. The problem is that TIPS currently provide zero or negative return beyond keeping up with inflation. A portfolio which invests in a mix of bond classes, as well as having a modest allocation (20% in this case) to risky assets such as stocks and commodities, appears to provide higher expected total return, albeit with accompanying risks.
For some investors, complete reliance on TIPS may be the best solution. TIPS are the best asset class available for preserving purchasing power against the effects of inflation. In effect, TIPS act like insurance against inflation. Any type of certainty or insurance comes with a cost. It is, in fact, only rational that investors will have to give up some expected return in order to ‘insure’ against inflation.
The debates over the issue of whether retirees should hold TIPS and other safe assets exclusively or maintain some exposure to risky assets such as stocks will never be unequivocally resolved. I hope with this article that I have at least helped to illuminate the implications of the different choices. To further explore this choice, I recommend Zvi Bodie and Rachel Taqqu’s book, Risk Less and Prosper, which I reviewed earlier this year and Moshe Milevsky’s Are You A Stock or a Bond?. I have also written an extended article that focuses on the quantitative aspects of this issue.
Related Links:
 401(k)s for the SelfEmployed
 The Challenge of LongTerm Income: Part 1
 From the Portfolioist Book Shelf: Risk Less and Prosper by Zvi Bodie and Rachelle Taqqu
Folio Investing The brokerage with a better way. Securities products and services offered through FOLIOfn Investments, Inc. Member FINRA/SIPC.
How do you determine interest rate sensitivity of a portfolio? Noticed your comment:
“The asset allocation of the Target 2010 Conservative Folio (2010C), by contrast, is estimated to be neutral with respect to interest rates.”
Ah–good question. I measure sensitivity of the portfolio to interest rates by calculating the correlation between portfolio return and tenyear Treasury yield. The correlation is very close to zero in this case–hence, there is little interest rate exposure.
Geoff, in order to compute the Monte Carlo estimation, what parameter values (interest rates, returns of the different asset classes, etc) did you use as input? To what time span do they correspond (one year, 2 years, etc)?
Thanks in advance
Hi Sam:
The Monte Carlo simulation is Quantext Portfolio Planner, which I designed. I have written about 200 articles on this model. The rates of return from fixed income include both interest rates and price appreciation components–this is a ‘total return’ model. The MC output is intended to be longterm expected returns and are not designed to be for a specific outlook period such as 1 or 2 years.
Pingback: Sector Watch: Spotlight on Defensive Strategy « Portfolio Investing Blog: Portfolioist
Geoff, Just finished translating Gregory Hight’s calculations for “Diversification Effect” (DE) into Excel. His paper is here: http://www.fpanet.org/journal/CurrentIssue/TableofContents/DiversificationEffect/ Very simply, DE is a measure of how effectively imperfect correlations reduce portfolio risk. (0 – 100%)
Knowing (and long admiring) your emphasis on diversification, I thought I’d see how well this retirement portfolio measured up. Result: your allocation of these assets reduces (unsystematic) portfolio risk 77%. Aplus Geoff.
Hight’s work on diversification metrics complements what I’ve read of your work …and I’m not surprised that you’ve provided us – again – with a thoughtful portfolio and handsome metrics. Grateful.
Pingback: Saving and Investing for Retirement: Part One « Portfolio Investing Blog: Portfolioist
Pingback: Saving and Investing for Retirement: Part Two « Portfolio Investing Blog: Portfolioist
Pingback: Saving and Investing for Retirement: Part Three « Portfolio Investing Blog: Portfolioist