Assessing the term structure of expected inflation using treasury inflation—protected securities: near real-time measures for those who need quick estimates

Business Economics, Jan, 2003 by Albert E. DePrince, Jr.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

(1.) Those plans called for the regular auction of five-, ten-, and thirty-year TIPS, after the initial (January 1997) ten-year TIPS auction. However, the broadness of the auction cycle diminished when the Treasury Borrowing Advisory Committee recommended elimination of the five-year TIPS in 1998 and later the thirty-year inflation-protected bond for reasons noted earlier in this study.

(2.) This is the compensation for the inability to accurately forecast inflation (Shen, 1995; Gong and Remolona, 1996).

(3.) These include the pure expectation, liquidity premium, preferred habitat, and segmented market hypotheses.

(4.) Other factors embodied in the TIPS may also affect the spread, including (1) "clientele effect," or demand for TIPS stemming from deferred income investment products; (2) difference in the timing of cash flow of conventional and inflation-protected securities; (3) tax treatment of the accreted principal in the TIPS; and (4) indexation lags (Cote et al., 1996; Circular, 1994).

(5.) This approach contrasts with the Federal Reserve methodology in which zero-coupon yields on the conventional Treasury securities are used rather than coupon bond yields to calculate expected inflation. They argue that the liquidity of the zero-coupon conventional Treasury securities and inflation-protected securities is in closer alignment than the liquidity of conventional coupon Treasury and inflation-protected securities (Greenspan, 2000). Also, because of the accretion of the TIPS principal, there is a closer match between the cash flows of zero-coupon bonds and the accreted TIPS principal than between the cash flows of TIPS and the conventional securities (Greenspan, 2000). However, other researchers are comfortable with conventional Treasury securities in computing expected inflation estimates (Shen and Corning, 2001). Additionally, the case for the conventional bond as the benchmark in the calculation was made in DePrince (2000).

(6.) A different technique was used to generate the long-term inflation forecasts. Monthly averages of the sets of spreads between the inflation-protected and conventional thirty-year bond were used to create a single time series of long-term inflation forecasts. Since April 2002, three spreads that were averaged. The resulting single series has an average maturity of about twenty-eight years. For simplicity, however, long-term inflation forecasts are reported as thirty-year forecasts in the text and in the graphs in which long-term expected inflation is portrayed, even though the average maturity is a bit less.

(7.) Measures of expected inflation were also generated for forecast horizons of six, seven, eight, and nine years. However, for reasons of simplicity, these estimates are not shown graphically.

REFERENCES

Bank of England (BOE). 1996. Gilts and the Gilts Market: Review 1995-6. Bank of England, United Kingdom. March.

Circular. 1994. Canada's Real Return Bonds. November 25.