Are TIPS the Real Deal?: A Conditional Assessment of their Role in a Nominal Portfolio

This paper documents predictable time-variation in the real return beta of U.S. Treasury inflation protected securities (TIPS) and in the Sharpe ratios of both indexed and conventional bonds. The conditional mean and volatility of both bonds and their conditional correlation are first estimated from predetermined variables. These estimates are then used to compute conditional real return betas and Sharpe ratios. The time-variation in real return betas and the correlation between TIPS and nominal bonds coincides with major developments in the fixed income market. One implication of this predictability is that portfolio managers can assess more efficiently the risk of investing in TIPS versus conventional bonds. Conditional Sharpe ratios indicate that over the sample period, TIPS had superior volatility-adjusted returns relative to nominal bonds. This finding is striking in view of the absence of a major inflation scare during the sample period from February 1997 through August 2001, but is loosely consistent with the possibility that TIPS elevated rather than reduced Treasury borrowing costs. On the other hand, mean-variance spanning tests indicate that TIPS did not enhance the mean-variance efficiency of diversified portfolios. Are TIPS the “Real” Deal?: A Conditional Assessment of their Role in a Nominal Portfolio Since the first auction of U.S. Treasury Inflation Protected Securities (TIPS) in January 1997, TIPS have become a significant component of debt issuance, representing roughly 1/3 of 10and 30-year Treasury gross auction amounts from 1997 through 2001. One motivation for offering TIPS was the Treasury’s belief that lower borrowing costs would result from meeting an unsatisfied demand for debt securities that offer a fixed real interest rate and, hence, are immune from inflation increases. An additional benefit of TIPS is that policymakers and market participants would be able to use the yield differential between nominal bonds and TIPS to determine market participants’ inflation expectations, subject to assumptions concerning risk premiums. As TIPS issuance has increased in recent years, fixed income investors have considered the strategic and tactical roles that these bonds should play in portfolios. A major concern is that TIPS will underperform their nominal counterparts, especially when performance is assessed relative to a nominal benchmark. When a portfolio manager allocates funds to TIPS, tracking error risk or the risk of underperforming a nominal benchmark index increases. This risk rises as the correlation between the nominal bond index and TIPS returns falls. Hence, a portfolio manager who is able to forecast the correlation between TIPS and nominal returns will be better able to manage tracking error risk. 1 See Sack (2000) for a discussion of using TIPS and nominal bonds to uncover implied inflation rates and related issues. See Price (1997) for a detailed discussion of indexed bonds.