The Term Structure of Expected Inflation Rates

Similar documents
The Information Content of the Yield Curve

An Orthogonal Polynomial Approach to Estimate the Term Structure of Interest Rates

Estimating Maximum Smoothness and Maximum. Flatness Forward Rate Curve

Resolution of a Financial Puzzle

Dynamic Replication of Non-Maturing Assets and Liabilities

Equilibrium Asset Returns

Instantaneous Error Term and Yield Curve Estimation

Problems and Solutions

We consider three zero-coupon bonds (strips) with the following features: Bond Maturity (years) Price Bond Bond Bond

Advanced Financial Economics Homework 2 Due on April 14th before class

A Note on Long Real Interest Rates and the Real Term Structure

Dynamic Hedging and PDE Valuation

Classifying Corporate Bonds: A Simple Approach

Estimating Term Structure of U.S. Treasury Securities: An Interpolation Approach

Practical example of an Economic Scenario Generator

Jaime Frade Dr. Niu Interest rate modeling

Crashcourse Interest Rate Models

Modelling the Term Structure of Hong Kong Inter-Bank Offered Rates (HIBOR)

Mixing Di usion and Jump Processes

The Fisher Equation and Output Growth

Claudia Dourado Cescato 1* and Eduardo Facó Lemgruber 2

Svensson (1994) model and the Nelson & Siegel (1987) model

ECON FINANCIAL ECONOMICS

ECON FINANCIAL ECONOMICS

The Duration Derby: A Comparison of Duration Based Strategies in Asset Liability Management

25. Interest rates models. MA6622, Ernesto Mordecki, CityU, HK, References for this Lecture:

Mind the Trap: Yield Curve Estimation and Svensson Model

Brooks, Introductory Econometrics for Finance, 3rd Edition

Estimating term structure of interest rates: neural network vs one factor parametric models

Lecture 3: Factor models in modern portfolio choice

Derivatives Options on Bonds and Interest Rates. Professor André Farber Solvay Business School Université Libre de Bruxelles

Equilibrium Term Structure Models. c 2008 Prof. Yuh-Dauh Lyuu, National Taiwan University Page 854

2 Control variates. λe λti λe e λt i where R(t) = t Y 1 Y N(t) is the time from the last event to t. L t = e λr(t) e e λt(t) Exercises

ECO 317 Economics of Uncertainty Fall Term 2009 Tuesday October 6 Portfolio Allocation Mean-Variance Approach

Smooth estimation of yield curves by Laguerre functions

A VALUATION MODEL FOR INDETERMINATE CONVERTIBLES by Jayanth Rama Varma

Improving Nelson-Siegel term structure model under zero / super-low interest rate policy

A THREE-FACTOR CONVERGENCE MODEL OF INTEREST RATES

Statistical and Computational Inverse Problems with Applications Part 5B: Electrical impedance tomography

Index Models and APT

Consumption and Portfolio Choice under Uncertainty

Indian Sovereign Yield Curve using Nelson-Siegel-Svensson Model

Course MFE/3F Practice Exam 2 Solutions

Introduction to Bond Markets

Credit Risk Modelling: A Primer. By: A V Vedpuriswar

Forward Contracts. Bjørn Eraker. January 12, Wisconsin School of Business

Fixed-Income Options

Optimal Portfolio Selection

( ) since this is the benefit of buying the asset at the strike price rather

EVALUATION OF CALLABLE BONDS: FINITE DIFFERENCE METHODS, STABILITY AND ACCURACY

Exercises on the New-Keynesian Model

The Fixed Income Valuation Course. Sanjay K. Nawalkha Gloria M. Soto Natalia A. Beliaeva

Mean-Variance Analysis

Pricing Dynamic Solvency Insurance and Investment Fund Protection

1.1 Basic Financial Derivatives: Forward Contracts and Options

Statistical Methods in Financial Risk Management

CB Asset Swaps and CB Options: Structure and Pricing

Interest rate models and Solvency II

The duration derby : a comparison of duration based strategies in asset liability management

Mathematics of Finance Final Preparation December 19. To be thoroughly prepared for the final exam, you should

Black-Scholes Option Pricing

Investment hysteresis under stochastic interest rates

INSTITUTE OF ACTUARIES OF INDIA

The Nelson-Siegel-Svensson Model for U.S. Treasury Securities and Its Interpretation

Is asset-pricing pure data-mining? If so, what happened to theory?

CSCI 1951-G Optimization Methods in Finance Part 00: Course Logistics Introduction to Finance Optimization Problems

Random Variables and Applications OPRE 6301

INSTITUTE OF ACTUARIES OF INDIA

Sharpe Ratio over investment Horizon

Characterization of the Optimum

Lecture 5: Review of interest rate models

Homework 3: Asset Pricing

Problem set 5. Asset pricing. Markus Roth. Chair for Macroeconomics Johannes Gutenberg Universität Mainz. Juli 5, 2010

Pricing Convertible Bonds under the First-Passage Credit Risk Model

Intertemporally Dependent Preferences and the Volatility of Consumption and Wealth

Applied Macro Finance

Advanced Modern Macroeconomics

dt + ρσ 2 1 ρ2 σ 2 B i (τ) = 1 e κ iτ κ i

The term structure model of corporate bond yields

Chinese Bond Market: A Need for Sound Estimation of Term Structure Interest Rates

MORNING SESSION. Date: Wednesday, April 30, 2014 Time: 8:30 a.m. 11:45 a.m. INSTRUCTIONS TO CANDIDATES

Recent Advances in Fixed Income Securities Modeling Techniques

MS&E 348 Winter 2011 BOND PORTFOLIO MANAGEMENT: INCORPORATING CORPORATE BOND DEFAULT

Libor Market Model Version 1.0

Consumption- Savings, Portfolio Choice, and Asset Pricing

Chapter 8. Markowitz Portfolio Theory. 8.1 Expected Returns and Covariance

GDP, Share Prices, and Share Returns: Australian and New Zealand Evidence

The Fixed Income Valuation Course. Sanjay K. Nawalkha Gloria M. Soto Natalia A. Beliaeva

Interest Rate Curves Calibration with Monte-Carlo Simulatio

Int. Statistical Inst.: Proc. 58th World Statistical Congress, 2011, Dublin (Session CPS001) p approach

An Equilibrium Model of the Term Structure of Interest Rates

Counterparty Credit Risk Simulation

INSTITUTE AND FACULTY OF ACTUARIES. Curriculum 2019 SPECIMEN SOLUTIONS

CHAPTER 14 BOND PORTFOLIOS

To apply SP models we need to generate scenarios which represent the uncertainty IN A SENSIBLE WAY, taking into account

A new approach for scenario generation in risk management

LECTURE NOTES 3 ARIEL M. VIALE

B. Combinations. 1. Synthetic Call (Put-Call Parity). 2. Writing a Covered Call. 3. Straddle, Strangle. 4. Spreads (Bull, Bear, Butterfly).

An Intertemporal Capital Asset Pricing Model

Two Equivalent Conditions

Transcription:

The Term Structure of Expected Inflation Rates by HANS-JüRG BüTTLER Swiss National Bank and University of Zurich Switzerland 0 Introduction 1 Preliminaries 2 Term Structure of Nominal Interest Rates 3 Term Structure of Real Interest Rates 4 Term Structure of Expected Inflation Rates 5 Conclusions

2 0 INTRODUCTION Goal: estimating the term structure of expected inflation rates of investors of riskless non-indexed bonds. Earlier attempts: FRANKEL [1982]: macroeconomic framework, nominal interest rate = real interest rate + expected inflation rate (Irving FISHER s hypothesis). Stanley FISCHER [1975]: theoretical work on indexed bonds: nominal interest rate = real interest rate + expected inflation rate interest premium. The interest premium may have either sign. Financial literature confirmed this result within quite different frameworks. To name a few: FAMA AND FARBER [1979], COX, INGERSOLL AND ROSS [1981, 1985, henceforth

3 CIR], LUCAS [1982], BENNINGA AND PROTOPAPADAKIS [1985], BREEDEN [1986]. Empirical literature neglected the interest premium so far. Exceptions: EVANS [1998] estimates the time-varying interest premium, but not the expected inflation rate (exogeneous). REMOLONA, WICKENS AND GONG [1998] estimate both the interest premium and the expected inflation rate (time series, selected nominal and index-linked bonds). We estimate the term structures of expected inflation rates and interest premia by means of the extended CIR model at a given moment in time. We proceed in three steps: 1. We use a non-linear optimization to determine the nominal instantaneous forward interest rates from observed prices of coupon-bearing, non-indexed government

bonds (DELABAEN AND LORIMIER, 1992). This methodology is superior to the known methods because it can explain any term structure of interest rates. the nominal spot interest rates are determined by numerical integration rather than differentiation (higher numerical accuracy). 2. We fit the extended CIR model to the nominal term structure obtained in the first step by means of a non-linear regression subject to certain constraints. From the estimated model parameters, the term structure of real spot interest rates is calculated. 3. The term structures of expected inflation rates and interest premia are calculated from the estimated model parameters obtain in the second step. 4

5 Result of long-term inflation forecast by means of yield curve. Expected Inflation Rates in Percent p. a. 9 8 7 6 5 4 3 2 1 0 Expected Spot Inflation Rates Observed Spot Inflation Rates Forecast of One-Year Inflation Rates = Expected One-Year Forward Inflation Rates Observed One-Year Inflation Rates February 1999-1 Years 0 1 2 3 4 5 6 7 8 9 10 Date 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Fig. 6: Inflation Forecast on 30 Dec. 1991 Correct trend.

6 1 PRELIMINARIES Spot interest rate P t, T = exp R, c t, T T t (1-1) P t T R R, c t, T = ln P t, T T t Price of a pure discount bond Settlement date Maturity date Spot interest rate = n (nominal), r (real) c Continuous compounding Instantaneous spot interest rate r, c t R, c t, t = lim t T ln P t, T T t (1-2) (1-3) r Instantaneous spot interest rate ( T)-year forward interest rate t, T, = exp F, c t, T, T (1-4)

7 F, c t, T, = ln t, T, T (1-5) F The forward price of a pure discount bond, fixed at date t and paid at a later date T when the bond will be delivered. The bond matures at date ( T t). The ( T)-year forward interest rate F, c t, T, = R, c t, t R, c t, T T t T (1-6) Instantaneous forward interest rate ƒ, c (t, T)=F, c (t, T, T) = lim T F, c (t, T, ) = R, c (t, T)+ R, c(t, T) T T t (1-7) ƒ The instantaneous forward interest rate = T = t ƒ, c (t, ) d = R, c (t, T) T t (1-8)

8 P t, T = exp R, c t, T T t = exp = T ƒ, c (t, ) d = t (1-9) ƒ, c (t, T)= ln P t, T T = P t, T T P t, T (1-10)

2 TERM STRUCTURE OF NOMINAL INTEREST RATES 9 Task: calculate the nominal spot interest rates from a sample of observed prices of couponbearing bonds. Known methods: Bootstrap method (most popular). Regression of bond prices on discount factors (CARLETON AND COOPER [1976]). Spline methods (MCCULLOCH [1971, 1975], VASICEK AND FONG [1982]). Assumed functional form (NELSON AND SIEGEL [1987], SVENSSON [1995]). All these methods have many drawbacks as mentioned in SHEA [1984, 1985]. Goal: in order to obtain a forward rate curve as smooth as possible, minimize the sum of squared differences in instantaneous forward interest rates subject to the constraint that the theoretical bond prices do not deviate from the observed

bond prices by more than a given tolerance error (DELABAEN AND LORIMIER [1992], LORIMIER [1995]). Outline of the Delbaen-Lorimier method Suppose we observe L prices of coupon-bearing bonds B obs (t, T N, c), = 1, 2,, L, at date t. N Face value or redemption value of bond c Vector of coupons Ascending order of maturity dates T 1 T 2 T L. Let time be divided into equal time steps of length t. Determine the number of time steps of length t, denoted as H, according to H T L t t 10 (2-1) We wish to find the instantaneous forward rates at the end points of these time steps. Since ƒ(t, t) = r(t), there are H forward rates

to be optimized. For example, with (T L t) = 30 years and t = 90 days, H = 120. Theoretical cash price of coupon-bearing bond: 11 B n t, T N, c, q k = = Σ c, k P n t, t + + q k k =0 + N P n t, T k = = Σ c, k exp R n, c t, t + + q k + k k =0 q + N exp R n, c t, T T t where = q T t, = T t q, =1,2,, L. (2-2) B n Theoretical price of a nominal couponbearing bond q Coupon periodicity Number of coupon dates Fraction of the first coupon period

Replace the spot interest rates by the instantaneous forward interest rates: 12 R n, c t, t + + k q + k q = t + + q k = ƒ n, c (t, ) d, = t k =0,, ; =1,, L. (2-3) Approximate the integral by the mean value of the upper and lower step function:

13 t t + + k q = ƒ n, c (t, ) d j =, k 1 1 2 t j + j +1 Σj =0 +, k +, k +1, k 2 t, k where =1,2,, L; k =0,1,, ; + q k, k =, 0 t, k H;, k = + q k, k t j =ƒ n, c (t, t + j t), (j =0,1,,, k,, H); H +1 H. (2-4) k k Number of instantaneous forward rates to be considered for the kth coupon date of the th coupon-bearing bond Fraction of the last time step for the kth coupon date of the th coupon-bearing bond

Minimize the squared differences in instantaneous forward interest rates subject to the condition that the relative pricing errors fall into a given tolerance range: 14 j = H ( 1,, H ) = min j Σj =1 1, 2,, H where j j j 1, 0 given, subject to 2 B n t, T N, c, q B obs t, T N, c, q =1,2,, L. 1 100 (2-5) Pricing error tolerances, ( = 1, 2,, L). Results of the DL methodology Many expirements applied to the theoretical term structure proposed by VASICEK [1977]. The Delbaen-Lorimier methodology performs much better than the bootstrap method when the term structure is sufficiently bent.

The Delbaen-Lorimier methodology is able to extract any term structure from a sample of bond prices. Example: wave-like term structure. Instantaneous forward interest rate: ƒ n, c (t, T)=a + b (T - t)+ 100 1 where a =0.01, b = 0.002667, c = 2 1000, d = 4 15. Spot interest rate: sin c + d (T - t) 15 (2-6) R n, c (t, T)=a + 1 b (T - t) 2 2 d (T t) 100 d (T - t) sin c + sin 2 d (T - t) 2 (2-7)

16 Annually Compounded Spot and Forward Rates in Percent p. a. 5,5 5,0 4,5 4,0 3,5 3,0 2,5 2,0 Spot Wave Spot DL Forward Wave Forward DL Spot Bootstrap Yield to Maturity Spot NS Forward NS 1,5 1,0 0 1 2 3 4 5 6 7 8 9 Remaining Years Figure 1: Wave Term Structure 10 11 12 13 14 15

17 Interest Rates in Percent p. a. 8,5 8 7,5 7 6,5 6 5,5 5 4,5 4 3,5 3 2,5 2 1,5 1 0,5 Spot Rates on 30 Dec. 1991 Forward Rates on 11 Feb. 1999 Forward R. on 15 March 1999 Forward Rates on 30 Dec. 1991 Forward on 4 Jan. 1999 Spot Rates on 11 Feb. 1999 Spot Rates on 15 March 1999 Spot Rates on 4 Jan. 1999 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 Remaining Years Figure 2: Term Structure of Nominal Interest Rates

Possible Explanations of the steep yield curve Domestic Inflation? Unlikely. Imported Inflation? May be. Adjustment to higher interest rates in the EU? 18

19 3 TERM STRUCTURE OF REAL INTEREST RATES If there are no indexed bonds traded in financial markets, then we must rely on a model which is able to explain nominal and real interest rates simultaneously. Two candidate models: CIR [1985] and BAKSHI AND CHEN [1996]. Outline of the CIR model 2 Real instantaneous spot interest rate: dr r, c (t)= r r, c (t) dt + r r, c (t) dz 1 (t), (3-1) 0,, r r c z 1 Continuously compounded real instantaneous spot interest rate Speed of adjustment Long-run equilibrium value Constant volatility parameter Gauss-Wiener process

20 Price of a real pure discount bond: P r (t, T)=A(t, T) exp B(t, T) r r, c (t), where A(t, T)= + + T t 2 exp 2 + + exp T t 1 +2 B(t, T)= P r 2 exp T t 1 + + exp T t 1 +2 = + 2 +2 2, = 2 2 Price of a real pure discount bond (3-2) Factor risk premium Processes for the consumer price level and the inflation rate: dr y, c (t)= 2 2 r y, c (t) dt + 2 r y, c (t) dz 3 (t), 0 2, 2, 2 dp(t)=r y, c (t) p(t)dt + p p(t) r y, c (t) dz 2 (t), dr y, c,dp = 2 p r y, c p, 0 p 1 r y c Instantaneous spot inflation rate (3-3)

21 2 2 2 z 3 p p z 2 Speed of adjustment Long-run equilibrium value Constant volatility parameter Gauss-Wiener process Consumer price level Constant volatility parameter Gauss-Wiener process Correlation coefficient between z 2 and z 3 Covariance operator

22 Price of a nominal pure discount bond: P n (t, T)=P r (t, T) C(t, T) 2 C(t, T)= exp Dt, T, r y, c (t) 2 exp 2 + 2 p + T t 2 2 + 2 p + exp T t 1 +2 Dt, T, r y, c (t) = 2 exp T t 1 1 2 p r y, c (t) 2 + 2 p + exp T t 1 +2 = 2 + 2 p 2 +2 2 2 1 p 2 2 = 2 2 2 2 2 (3-4) P n Price of a nominal pure discount bond Define 11-by-1 parameter vector = [,,,, 2, 2, 2, p,, r r c(t), r y c(t)]

23 Spot interest rates in the CIR model R r, c t, T = B(t, T) r r, c(t) ln A(t, T) T t R n, c t, T = R r, c t, T + 2 ln C(t, T) + Dt, T, r y, c (t) T t (3-5) R r c R n c Real spot interest rate Nominal spot interest rate Take the limit as T t R r, c t, t = r r, c t R n, c t, t = r n, c t = r r, c t +1 p 2 r y, c (t) (3-6) r r c r n c Real instantaneous spot interest rate Nominal instantaneous spot interest rate Process for the nominal instantaneous spot interest rate dr n, c (t)= r r, c (t) + 2 2 r y, c (t) dt + r r, c (t) dz 1 (t) +1 p 2 2 r y, c (t) dz 3 (t) (3-7)

24 Find parameter vector by constrained regression H min R n, c t, j R n, c t, j Σj =1 R n, c t, j R n, c t, t + j t R n, c t, j R n, c t, t + j t s. t. R n, c t, t + (k) t = R n, c t, t + (k) t, k =1,, m; m 11. 2 (3-8) Indicator vector which locates the interest rates to be constrained

25 0 Interest Rates in Percent p. a. 8,5 8 7,5 7 6,5 6 5,5 5 4,5 4 3,5 3 2,5 2 1,5 1 0,5 Nominal Rates on 15 March 1999 11 Feb. 1999 4 Jan. 1999 Nominal Spot Rates on 30 Dec. 1991 Real Rates on 30 Dec. 1991 Real Spot Interest Rates on 15 March 1999 11 Feb. 1999 4 Jan. 1999 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 Remaining Years Figure 3: Real Spot Interest Rates

26 4 EXPECTED INFLATION RATES Define random inflation rates correspondingly p R y c ƒ y c p(t) = exp R y, c t, T T t p(t) = exp = T ƒ y, c (t, ) d = t t T. Consumer price level Spot inflation rate p(t), Instantaneous forward inflation rate Expected values of inflation rates (4-1) R y, c t, T T t = = T ƒ y, c (t, ) d = t = ln p(t) p(t) (4-2) Expectation operator Since ln(p(t) / p(t)) ln( {p(t) / p(t)}) by Jensen s inequality, relationship (4-1) reads for the expected values as follows:

27 p(t) = exp R y, c t, T exp ln p(t) T t = exp = T ƒ y, c (t, ) d = t exp ln p(t) (4-3) The expected instantaneous spot inflation rate is equal to the expected instantaneous forward inflation rate: ln p(t) + = T = t ƒ y, c (t, ) d =ln p(t) (4-4) Differentiate with respect to maturity date ƒ y, c (t, T) dt = dln p(t) = dp(t) p(t) Equ. (4-5) is stochastic equivalence to equ. (1-10). Take expectations: (4-5) ƒ y, c (t, T)= r y, c (T), t T. (4-6) Obtain the expected instantaneous spot inflation rate by recurrence for small time steps:

28 r y, c (t + j t)= 1 2 t r y, c t +[j 1] t + 2 2 t, j =2,3, r y, c (t + t)= 1 2 t r y, c t + 2 2 t, j =1. (4-7) The expected instantaneous forward inflation rates define the whole term structure of expected inflation rates. Finally, the interest premium becomes R n, c t, T = R r, c t, T + R y, c t, T c t, T (4-8) c Continuously compounded interest premium In the CIR framework, the interest premium consists of two terms, namely, of the variance of the consumer price level and of the so-called wealth premium. The wealth premium depends on the investor s attitude towards risk as well as on the covari-

ance between real wealth and the future inflation, which have either sign. Hence, the interest premium may be positive or negative. 29

30 Expected Inflation Rates in Percent p. a. 6 5,5 5 4,5 4 3,5 3 2,5 2 1,5 1 0,5 0 Expected Spot Inflation Rates Start and End of Expected One-year Forward Inflation Rates, resp. Interest Premia Expected Instantaneous Forward Inflation Rates Nominal minus Real Spot Interest Rates 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 Remaining Years Figure 4: Expected Inflation Rates on 15 March 1999

31 0 0,5 1,5 1 Expected Inflation Rates in Percent p. a. 2,5 2 3,5 3 8,5 8 7,5 7 6,5 6 5,5 5 4,5 4 Expected Instantaneous Forward Inflation Rates Start and End of Expected One-year Forward Inflation Rates, resp. Nominal minus Real Spot Interest Rates Interest Premia Expected Spot Inflation Rates 0 1 2 3 4 5 6 7 8 9 10 Remaining Years Figure 5: Expected Inflation Rates on 30 December 1999

Why are the present inflationary expectations so high? We offer three possible explanations for this phenomenon. 1. Investors of government bonds might expect a high future domestic inflation. In view of the present non-existent inflation, and in view of the present slackness of the Swiss economy, this seems rather an unplausible explanation. 2. The investors might expect a high foreign, that is, imported inflation. Again, this seems an unplausible explanation because the inflation is presently very low in those European countries which are the main trading partners of Switzerland. 3. The investors might expect the Swiss nominal spot rates to adhere to the higher interest rate level of the European Union (EU), in particular, 32

because they might expect that Switzerland will join the EU in the near future. We feel that the third explanation is the most conceivable one for three reasons. a. The CIR model to evaluate the term structure is a model of a closed economy as well as a model that is not able to explain other effects than the expected inflation rate and the interest premium. b. A possible Swiss memberhip in the EU was not a political issue at the end of 1991. Hence, the term structure considered for this date should not contain the possible effect that the Swiss interest rates could adhere to the higher interest rate level of the EU. The term structure considered at the end of 1991 forecasts very low future inflation rates of about 0.3% 33

34 per annum for the years 1999 to 2001 as shown in figure 6. c. An inspection of the variance of the consumer price level in the course of future time, shows that this variance calculated for a recent term structure is twice as big as the variance for the term structure considered at the end of 1991. Hence, the investor s uncertainty about future variations of the consumer price level has risen substantially, although the inflation rate fell substantially during the past decade.

35 5 CONCLUSIONS In this paper, we infer the term structure of expected inflation rates from a sample of observed prices of coupon-bearing bonds by means of a three-step procedure. In the first step, the nominal instantaneous forward interest rates are optimized. In the second step, the extended CIR model is fitted to the optimized nominal spot interest rates. By this procedure, the real spot interested rates are determined. Finally, the estimated CIR model parameters allow the calculation of the expected instantaneous forward inflation rates by means of a simple recurrence relationship. The expected inflation rates coincide quite well with the observed inflation rate for a sample observed at the end of 1991.

We conjecture that the present high expected inflation rate obtained from the CIR model is a spurious inflationary expectation. Rather, the result relates to the difference in interest rates between Switzerland and the EU countries. Therefore, we plan to investigate the European term structure of expected inflation rates. 36