The Curious Behavior of TIPS

Jared Kizer takes a look at the clear empirical relationship between these fixed income securities and the credit risk premium.

I’ve followed the TIPS market fairly closely since I started in the advisory business in 2003, and we’ve used them for a portion of many clients’ fixed income portfolios (and continue to do so today). The year I started in the business was only a few years after the inception of the market, and we now have about 20 years of returns history to examine. In the early years, there was a very logical argument that TIPS should dominate the fixed income portfolios of individual investors because they were essentially the perfect fixed income security, issued by the U.S. government and linked to inflation. Since credit risk is redundant for most investors and individuals have spending that grows with some measure of inflation, TIPS seemed to have everything an investor could want. Further, the historical correlation between unexpected inflation and stock returns has been negative while it was expected to be positive for TIPS. This meant TIPS had the potential to provide more powerful diversification of stocks when compared with nominal Treasuries.

The reality of TIPS, however, has been more muddled. TIPS have exhibited bizarre behavior for stretches of history (e.g., in the years following the inception of the market and during the financial crisis). In the early years, real yields on TIPS were extremely high — which one might reasonably attribute to the market learning about the securities. During the financial crisis, however, TIPS substantially underperformed nominal Treasuries and I’m still not sure exactly why.

Here, I focus on another aspect of TIPS I’ve not seen documented. For some reason, TIPS returns have been highly correlated with the credit risk premium. Is this due to the relative illiquidity of TIPS compared with Treasuries? It’s possible, but why has this persisted with a market value of TIPS in excess of $1 trillion and for a security issued by the U.S. government? Or is it related to institutional ownership of TIPS who may tend to sell them during periods of stress to generate liquidity? I have no idea what the answers are, but as we will see below, the empirical relationship between TIPS and the credit risk premium is clear.

For this piece, I use monthly returns for the Barclays U.S. TIPS Index, U.S. Treasury Index and U.S. Corporate Bond Index for most of the findings below. Barclays began reporting returns on the TIPS index starting in March 1997 with returns now updated through June 2018. To start, the correlation of the monthly returns of TIPS to Treasuries over the full history has been 0.67 while the correlation to investment-grade corporate bonds has been 0.70. This is not what I would have expected and is our first clue that there’s something potentially interesting here. If we examine the correlation starting in 2010 and forward, it’s basically the same result with the TIPS-to-Treasury correlation at 0.67 and the TIPS-to-corporates correlation at 0.76. So, the results below can’t be attributed to either the early history of TIPS or only to performance during the financial crisis. Let’s dig a bit deeper by looking at a scatter plot of the monthly excess returns of TIPS relative to the investment-grade credit premium (IGCP). The excess returns of TIPS are the monthly returns after subtracting out nominal Treasury returns.

Figure 1: TIPS Excess Returns vs. IGCP (3/1997–6/2018)

We see here a clear positive correlation between the two. The excess returns of TIPS have tended to be lower when the IGCP is below its mean and higher when the IGCP is above its mean. Now, let’s look at the same graphic using only data from 2010 and forward.

Figure 2: TIPS Excess Returns vs. IGCP (1/2010–6/2018)

Here, the data points lose some of the extremes compared with Figure 1 but the basic result remains: The excess returns of TIPS have been positively correlated to the IGCP. Further, the correlation between the two across both periods is virtually identical at roughly +0.50. I think it’s fair to say no one would have projected this type of relationship. The base cases would have been either zero correlation to the IGCP or potentially even negative correlation.

Who’s to say what is driving this relationship? My best guess is that some of the largest institutional holders of TIPS are using them as either collateral or as part of multi-asset class strategies that necessitate selling TIPS during periods of relative market stress, inducing the relationship we see above. The implications are that investors should understand that TIPS may not diversify equity market risk as strongly as one would expect and may warrant a lower target allocation than some of the original thinking noted above.

This commentary originally appeared August 3 on MultifactorWorld.com

By clicking on any of the links above, you acknowledge that they are solely for your convenience, and do not necessarily imply any affiliations, sponsorships, endorsements or representations whatsoever by us regarding third-party Web sites. We are not responsible for the content, availability or privacy policies of these sites, and shall not be responsible or liable for any information, opinions, advice, products or services available on or through them.

The opinions expressed by featured authors are their own and may not accurately reflect those of the BAM ALLIANCE. This article is for general information only and is not intended to serve as specific financial, accounting or tax advice.

© 2018, The BAM ALLIANCE^®