What Asset Allocation Level Should You Use for Treasury Inflation-Protected Securities (TIPS)?

Over the past couple months, we’ve been discussing several interesting aspects of asset allocation and portfolio design. Thus far, we’ve explored how gold/precious metalsinternational equities, REITsshort-term bonds, and intermediate-term bonds perform as asset classes and if/how they should be weaved in to your asset allocation.

Continuing this investigation on portfolio construction, I wanted today to look in to the question of “what level, if any, of your portfolio should be allocated to Treasury Inflation-Protected Securities (more commonly referred to as TIPS)?”

Let’s get started!

 

WHY BOTHER CONSIDERING THE ADDITION OF TIPS AT ALL?

To begin, the first question that I suppose we should address is why it’s even worth considering adding TIPS to your portfolio in the first place.

As is the case with many elements of Modern Portfolio Theory and portfolio construction in general, TIPS theoretically should provide a favorable diversification benefit when incorporated with other components of your portfolio.

More specifically, this diversification benefit comes from the fact that the average return statistics of TIPS are not perfectly correlated with other commonly-included asset classes of a portfolio. In mathematical terminology, we can say that the diversification benefit is obtained because the correlation coefficients of TIPS with the other asset classes are not 1.

To demonstrate this in tabular form, I ran a correlation coefficient analysis of the annual returns of the Total US Stock Market, TIPS, and Short-Term Treasuries between the ~40 year period between 1972-2011 (using data from the Bogleheads.org Simba backtesting spreadsheet).

The correlation coefficient results can be seen in the table below. As you can see (green highlighted cells), the correlation coefficients between TIPS and the Total US Stock Market / Short-Term Treasuries are quite favorable at 0.20 and 0.03, respectively. What this means in English is that the returns of TIPS move in sync with the stock market in general only 1/5 of the time and with Short-Term Treasuries almost none of the time.

 

According to Modern Portfolio Theory principles, adding a poorly correlated asset class in to a portfolio can often decrease volatility while possibly, increasing returns. Thus, this is the motivation for looking at including TIPS in a portfolio/asset allocation.

In addition to a potential diversification benefit, TIPS provide a “pure hedge” against US inflation. In other words, unlike some other indirect hedges against inflation that may or may not be guaranteed, TIPS offer investors a guaranteed real rate of return, no matter what happens to inflation in the future. This also indicates that the real returns of TIPS have less variance than that of nominal bonds. Lastly, being Treasury securities, they have 0 default risk.

 

WHAT DO THE EXPERTS SAY ABOUT ADDING TIPS TO YOUR PORTFOLIO?

Before getting too far in to my own asset allocation analysis, I generally like to quickly review and summarize the thoughts that people much more qualified than I am have on a subject.

As such, listed below is a summary of what has been recommended in the books of several well-respected asset allocation authors regarding the incorporation of TIPS in to a portfolio:

  • Larry Swedroe (probably my favorite investing author I have found to date)
    • In Larry’s two books, The Only Guide to a Winning Investment Strategy You’ll Ever Need and The Only Guide to Alternative Investments You’ll Ever Need, Swedroe provides the most complete, definitive position that I could find in the literature of what allocation investors should consider giving to TIPS.
    • Essentially, he states that numerous academic studies have been conducted, and all come to the similar conclusion that TIPS should “dominate” the fixed income allocation for investors, especially in tax-sheltered accounts.
    • For example, he shared the results from 3 studies that concluded the optimal portfolio position based on risk-adjusted returns for the overall portfolio was between 50-100% of the fixed income allocation.
  • Burton Malkiel
    • In his famous and amazing book (2003 edition), A Random Walk Down Wall Street, Malkiel provides example asset allocations with 5% of the total portfolio allocated to TIPS (~25% of the fixed income allocation).
    • Unfortunately, it is never explained exactly how this 5% level is derived. The book only mentions that TIPS are a wise component to add to a portfolio.
  • William Bernstein (my 2nd favorite investing author I have found to date)
    • In his 2002 book, The Four Pillars of Investing, Bernstein displays sample portfolios/asset allocations with TIPS representing 4-20% of the total portfolio, depending on the investment horizon. 10% seemed to be a common allocation used within this range.
    • Generally, this equated to ~1/3 of the fixed income allocation. 
    • As was the case with Malkiel’s book, Bernstein doesn’t really get in to exactly how the allocation to TIPS was derived, but simply mentions that it is beneficial to have some exposure to the asset class.
  • Alexander Green
  • Although not a book (but simply a group of people with a lot of collective knowledge), on the Bogleheads Forum, the most popular fixed income allocation is a 50/50 split between nominal and TIPS bonds.

Conclusion from Literature – So, after looking through all of the books that have helped me build my portfolio over the years, there doesn’t seem to be a clear consensus among the various authors. On one hand, we have several that recommend committing about 1/3 of fixed income assets to TIPS, while on the other hand, Swedroe is recommending anywhere from 50-100% fixed income allocation to TIPS. Of course, this number will change as your life cycle allocation adjusts during different life stages.

 

HOW HAVE TIPS PERFORMED IN THE PAST COMPARED TO OTHER PORTFOLIO COMPONENTS?

In this case especially, the literature seemed to be rather inconclusive about what is exactly the correct amount an investor should allocate to TIPS.

The first thing that is interesting to examine when seeing how TIPS have performed compared to other common asset classes is to see how an investment made a long time ago (~40 years in this case) would have grown.

As such, shown below is the hypothetical growth of a $10k starting investment in Short-Term Treasures (red line), the Total US Stock Market (blue line), and TIPS (green line) between the years of 1972-2011.

As you can see, the investment in TIPS provided approximately (at least from the graph) the same return as that of Short-Term Treasuries. As would be expected, the equity asset class returned much higher than the fixed income investments.

The actual return data that produced the graph above is shown in the table below. First, as we would expect given that TIPS have a MUCH LONGER average maturity (~9 years) than the Short-Term Treasuries (~2 years), TIPS have higher volatility than the Short-Term Treasuries.

However, it is somewhat surprising to me that investors were not really compensated that much more at all for shouldering the increased risk of TIPS. For example, an investor only obtains an ~10% increase in average returns for TIPS in exchange for taking on an ~50% increase in volatility (standard deviation).

Essentially, what this data shows is that with TIPS, even though they have some attractive potential benefits, they don’t seem to have been very efficient compared to Short-Term Treasuries over the past 40 years.

 

TIPS ALLOCATIONS IN A 3-COMPONENT PORTFOLIO DESIGN

More important to us as portfolio design “engineers” is how an asset class will behave and/or benefit us when incorporated in a realistic portfolio/asset allocation.

To assess this for the TIPS asset class, I re-ran the portfolio analysis during the 1972-2011 period using a portfolio consisting 30% of fixed income (split between varying levels of Short-Term Treasuries and TIPS) and then a 70% allocation to the Total US Stock Market.

Shown below is the average annual return vs. risk graph that resulted from the analysis. Going from left to right, each plot point on the curve represents increasing allocations to TIPS, as described on the below table.

And, shown below is the exact data that was used to construct the return / risk curve above.

If we examine this data a little more closely, we see that numerically, the most efficient allocation to TIPS in terms of the highest return/risk ratio occurs at 1% TIPS, so very little at all (3% of fixed income allocation). However, in the grand scheme of things, the return numbers don’t change all that much as TIPS are incorporated, meaning there is not that significant of an effect.

Intriguingly, if a 60/40 equity/fixed income asset allocation split is used as opposed to the 70/30 employed above, the optimal level of TIPS becomes ~5% (12.5% of fixed income allocation). Furthermore, if a 50/50 split is used, the optimal TIPS level is found to be 8% (16% of fixed income allocation).

Overall, my 3-component TIPS analysis leaves me with the following question –

Why does my analysis predict much lower optimal levels of TIPS (3-16% of fixed income allocation) when the books I reviewed above recommended 25-100% of fixed income allocation?

I can think of several possible causes for the discrepancy. First, the data I’m using could be wrong (hopefully not). Second (and more likely), is the possibility that I am not comparing apples to apples, which in this case, means inflation-adjusted returns to inflation-adjusted returns.

To investigate the disparity in findings, I obtained the inflation data for the years being analyzed, subtracted it from the annual returns of the various asset classes, and generated the table below showing the summary of my findings.



When adjusting the returns for inflation (4.39% annual average inflation during the time period analyzed), the case for incorporating TIPS in to a portfolio becomes more significant, with the most efficient level being at a 7.4% TIPS allocation (of the total portfolio, or ~25% of the fixed income allocation) for the 70/30 equity/fixed income portfolio. If we use a 60/40 or 50/50 equity/fixed income overall asset allocation, the optimal allocation to TIPS becomes ~23% of the fixed income allocation.

You can view the complete set of numbers/calculations for my analysis by accessing the Google Docs Spreadsheet here.


CONCLUSIONS ABOUT TIPS ASSET ALLOCATION AND MY PERSONAL PATH FORWARD

So, after sifting through all this analysis, what’s the overall verdict on what asset allocation should be committed to the TIPS asset class?

Listed below are my key takeaways from this investigation:

  • TIPS provide the potential for a very significant diversification benefit due to their low correlation with both equities and nominal bonds.
  • In addition, since TIPS provide a solid hedge against inflation (something which is almost guaranteed to happen) by providing “real returns,” they are truly a hard asset class to ignore.
  • The literature recommends that between 25-100% of an individual’s fixed income allocation be committed to TIPS because of these benefits. It is often stated that TIPS should be the dominant holding in one’s fixed income allocation (meaning >50%).
  • In my own analysis of the ~40 year period between 1972-2011 using a 3-component portfolio of US equity, Short-Term Treasuries, and TIPS, I found that:
    • The return values HAVE to be adjusted for inflation in order for TIPS to have a significant case to be included in the portfolio.
    • After adjusting returns for inflation, the most efficient level for TIPS in a portfolio is ~25% of the fixed income position. 
    • I hypothesize that the TIPS allocation for maximal portfolio efficiency would have been higher if I used intermediate or long-term maturity bonds in my portfolio construction. However, since I like to use short-term bonds (which provide an indirect hedge against inflation since short-term rates are able to adjust fairly quickly as inflation changes), dominant levels of TIPS were not required to increase efficiency.

My Personal Path Forward – I want to lastly share how this analysis affects me personally. I currently use a 70/30 equity-fixed income asset allocation. 5% of my total portfolio (or ~17% of my fixed income position) is allocated to TIPS. Thus, I’m a little bit below both the optimal level found in my own analysis above and also what is suggested by most authors in the literature/Bogleheads.org. Because inflation is almost a certainty to occur in the future and the fact that TIPS provide a guaranteed real rate of return, I think it is prudent for me to increase my TIPS allocation.

Regarding the question of WHAT TIPS allocation I will use going forward, I believe I will stick to the lower end of the recommended spectrum, shooting for a ~25% fixed income allocation to TIPS, or 8% of my total portfolio. I don’t necessarily want to increase much higher than that due to the longer maturity that the Vanguard TIPS mutual fund I will use has (9 years). To keep my overall fixed income / equity allocation levels the same, I’ll exchange my short-term bond index fund allocation over to TIPS to make the required change.

A good question going forward perhaps is whether it is better to use the new Short-Term TIPS Fund or the regular TIPS Fund that Vanguard offers. I’ll put that on my list of things to analyze soon!

How about you all? Do you have any exposure to TIPS (or another inflation hedge) in your investing portfolio?

If so, what % of your portfolio does it constitute?

Share your experiences by commenting below!

Comments

  1. Great post. I think that TIPS are an excellent option to help round out a portfolio.
    My recent post What is the GI Bill and how much can you get?

    • Thanks so much for reading MMM! What level of TIPS allocation do you use?
      My recent post What Asset Allocation Level Should You Use for Treasury Inflation-Protected Securities (TIPS)?

  2. my2centopinion says:

    Thanks for the great analysis of TIPS. I've always known about the basic of a relative safe investment and the inflation protection. It seems like a good fit for people's allocation of fixed income.
    My recent post My Budget Blows

    • Thanks for your comment my2cent! Do you own any TIPS/TIPS funds?
      My recent post What Asset Allocation Level Should You Use for Treasury Inflation-Protected Securities (TIPS)?

  3. SiliconValleyGuy says:

    Thanks for the interesting post!
    To me it seems the analysis is a little off because much of the comparison is between mid-long term duration tips vs short term bonds. It is hard to tell how much impact you are getting simply from duration vs. the actual inflation adjustment aspect. I think the analysis would have been much more useful to isolate out the duration and do the comparison vs short term tips.

    • Very good point SVG! Do you know of where long term return/performance data for short-term tips can be found?

      I had thought about using that, but when I saw that it was a fairly recent inclusion to the Vanguard family of funds, I figured it would be hard to find at least 20 years of historical data.
      My recent post What Asset Allocation Level Should You Use for Treasury Inflation-Protected Securities (TIPS)?

      • SiliconValleyGuy says:

        Ah, good question. No not really, but Vanguard has a white paper that has some data re: short term tips, called “the long and short of tips”. I don't think it has all the data needed, though, but it is interesting reading in any case.

        • Thanks for sharing that SVG! I'll have to check that out. I'll also do a quick search around to see if any long term data on short tips is available elsewhere.
          My recent post The Size of Your Emergency Fund Should Be In Inverse Proportion to the Stability of Your Income

  4. David Smith says:

    Thank you for the excellent post. I'm always glad to see rigorous, quantitative analysis. Commodities and real estate can also provide inflation protection, so I'm curious how they perform with respect to TIPS. I'll attempt the analysis with MPT software.

    • Thanks for your comment David. Yeah I've definitely read about the use of those asset classes to protect against inflation. However, I've also heard the argument that they also don't provide adequate protection alone at times when it's needed most.

      I'm curious – what MPT software do you use?
      My recent post The Size of Your Emergency Fund Should Be In Inverse Proportion to the Stability of Your Income

      • I use Folio Explorer for MPT analysis. I’m not sure how to assess the “inflation protection” of TIPS vs real estate vs commodities in a quantitative manner, but here are a couple insights regarding volatility, diversification, and correlations. 1) In relation to the US total stock market, TIPS and short treasuries provide better diversification (lower correlation with US stocks) than real estate and commodities. 2) In low volatility portfolios with minimal equities, the allocation to short term treasuries exceeds the TIPS allocation. In higher volatility portfolios with moderate equities, the TIPS allocation exceeds short term treasuries. (This analysis uses historical prices from the following iShares ETFs: SHV, TIP, ITOT, IYR, and GSG. The analysis period is 2009-present due to the ETFs’ ages.)

  5. Tips weren’t issued until 1997. How do you have tips data going back to 1972?

  6. John Richards says:

    Hi, great site! I happened to be looking for some insight on TIPS when I ran across your website. I’ve struggled with the idea of including them in my portfolio, for similar reasons to the ones you stated above. I’m happy with my portfolio, and it’s strong performance, and what I think is fairly moderate risk. However, with interest rates so low, it makes sense to think about rising interest rates. I’m a big believer that deflationary forces will keep rates low for a couple decades, but I get a kick out of figuring this out anyways. I have used Portfolio Visualizer to model various portfolios over 40 years. I was gratified that the portfolio construction methodology I’d intuited after years of reading seemed to generate a superior return to risk profile (and so far has in real life) – at least in nominal terms. The portfolio includes a substantial allocation to Intermediate Treasuries. Substituting TIPS for some or all of the IT allocation only appears to hurt the portfolio, risk and return. I realize this is approaching the issue from a slightly different perspective. Anyways, if I back out inflation from those numbers, I don’t see how that would change the net portfolio results, but perhaps I’m oversimplifying by simply assuming I’d back inflation out of the total portfolio return – doing this asset class by asset class should still yield the same mathematical result(?), so I’m wondering what I’m missing. I’d appreciate any comments you could add to help clarify.

Speak Your Mind

*

CommentLuv badge