International Financial Markets Prices and Policies Second Edition 2001 Richard M. Levich 16C Measuring and Managing the Risk in International Financial Positions Chap 16C, p. 1 Overview ❿ Measuring Economic Exposure to FX Risk Regression Approach: Sensitivity of the firm s MV to FX changes Scenario Approach: What if? analysis Value-at-Risk (VAR): With 95% probability, how much might our portfolio change in value in a day, a week, or a month? ❿ Limitations in estimating economic exposure What assumptions are we making? Can the approach be implemented? By whom? Chap 16C, p. 2 1
The Concept of Economic Exposure (1 of 2) ❿ MV/ S reflects economic exposure of the firm. Because MV=NPV of all future CF, if an FX change impacts any CF, this represents exposure ❿ Economic exposure is a very broad concept involving The firm: Including production and marketing operations Its products: Whether the products are highly differentiated with low price elasticity of demand, or generic products with a high price elasticity The industry: Whether it is highly competitive vs. oligopolistic, worldwide vs. local, with high or low barriers to entry. Chap 16C, p. 3 The Concept of Economic Exposure (2 of 2) ❿ The channels of economic exposure include The firms customers, suppliers and competitors There can be direct as well as indirect sources of exposure ❿ Economic exposure may be different in the short-run vs. the long-run In the short-run, the firm has less flexibility to change its prices, production process, and products In the long-run, the firm has more flexibility to change its prices, production process, and products. Plus, exchange rate changes are more likely to be offset by local price changes (Purchasing Power Parity) Chap 16C, p. 4 2
The Regression Approach ❿ The regression approach directly measures the exposure of a firm to exchange rate changes by estimating the relationship between the firm s market value at time t (MV t ) and the spot rate (S t ) using the equation: MV t = a + bs t + e t Units: MV ($), a($), b( ), S($/ ), e($) ❿ The coefficient b measures the sensitivity of the market value of the firm to the exchange rate. Chap 16C, p. 5 Examples of the Regression Approach DM Asset Exposure Market Value of the Firm and $/DM Rate DM Liability Exposure Market Value of the Firm and $/DM Rate 23,500 19,500 23,000 19,000 Market Value 22,500 22,000 21,500 Market Value 18,500 18,000 17,500 21,000 17,000 20,500 16,500 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 $/DM Rate $/DM Rate No DM Exposure (Asset or Liability) Market Value of the Firm and $/DM Rate 21,000 20,800 20,600 Market Value 20,400 20,200 20,000 19,800 19,600 19,400 19,200 19,000 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 $/DM Rate Chap 16C, p. 6 3
Interpreting Regression Results ❿ To interpret the regression analysis, three results need to be examined: The magnitude of b. b > 0 an asset exposure in the foreign currency b < 0 a liability exposure b = 0 no exposure to the exchange rate The t-statistic of b. Statistical significance is necessary for confidence in the results. The R 2 of the regression. R 2 measures the percentage of variation in the market value explained by the exchange rate. Chap 16C, p. 7 A Numerical Regression Example MV t = a + b S t + e t Market Value of Firm XYZ and $/Euro Rate Monthly Data: January 1999 - July 2000 25,500 1.2000 25,000 1.1500 Market Value (in US$) 24,500 24,000 23,500 Partial Results b = 5,093 (1,464) t=3.5 R 2 = 0.416 1.1000 1.0500 1.0000 0.9500 US$/Euro Spot Rate 23,000 0.9000 Dec-98 Feb-99 Apr-99 Jun-99 Aug-99 Oct-99 Jan-00 Mar-00 May-00 Jul-00 Market Value US$/Euro Chap 16C, p. 8 4
The Regression Approach - Extensions ❿ To measure the firm s exposure to multiple exchange rates, a multiple regression can be estimated: MV t = a + b 1 S $/,t + b 2 S $/,t + b 3 S $/,t + e t ❿ If the firm has data on cash flows at the level of a subsidiary or project, the exposure of these smaller units can also be measured: CF t = a + bs t + e t Chap 16C, p. 9 The Scenario Approach ❿ Given a scenario, we can estimate the firm s cash flows (and its market value) conditional on an exchange rate path. ❿ The scenario approach is well suited to a spreadsheet analysis where one is encouraged to ask a variety of what-if questions. Best case, worst case, most likely case, etc. ❿ What if regarding the time path of Exchange rates Prices paid for inputs Sales volumes Prices received for outputs Chap 16C, p. 10 5
Examples of the Scenario Approach Consider the impact of a permanent 5% appreciation of the US$, holding all other factors constant. Present Value of Cash Flows (Millions) $39.577 $35.222 A The slope measures the exposure of the firm at the initial exchange rate. - 15% - 10% - 5% 5% 10% 15% $/A$ $0.5435 $0.5682 $0.5952 $0.6250 $0.6563 $0.6875 $0.7188 A$/$ A$1.84 A$1.76 A$1.68 A$1.60 A$1.52 A$1.45 A$1.39 O A* Chap 16C, p. 11 Examples of the Scenario Approach Suppose the firm can pass along part of the exchange rate change to its Australian customers. Present Value of Cash Flows (Millions) $39.577 $35.222 B A The slope of BOB* is flatter than AOA* since the firm has less exposure now. - 15% - 10% - 5% 5% 10% 15% $/A$ $0.5435 $0.5682 $0.5952 $0.6250 $0.6563 $0.6875 $0.7188 A$/$ A$1.84 A$1.76 A$1.68 A$1.60 A$1.52 A$1.45 A$1.39 O A* B* Chap 16C, p. 12 6
The Value at Risk (VAR) Approach ❿ The Wrong Question : How much could I lose on a single day? (EVERYTHING!!) ❿ The Right Question : What is the most I could lose on a single day with 95% (or 90% or 99% or...) confidence that I will not lose more than that amount? ❿ Assuming normal price distributions, calculate the loss in value of the portfolio if an unlikely (say, 5% chance) adverse price movement occurs. ❿ The result of this calculation is the value at risk (VAR) Chap 16C, p. 13 What value could the future spot rate take with 5% probability? 9 8 7 6 µ=1.50; σ=0.05 $/ spot rate 5 4 µ=1.50; σ=0.08 3 2 1 0 1.25 1.30 1.35 1.40 1.45 1.50 1.55 1.60 1.65 1.70 1.75 mu 1.50, sigma 0.05 mu 1.50, sigma 0.08 Chap 16C, p. 14 7
Assumptions and Features of VAR ❿ VAR is a probability statement about the potential change in value of a portfolio resulting from changes in market factors over a specified period of time. ❿ VAR depends on the underlying probability distribution of market factors. ❿ VAR requires estimates of variance and correlation of underlying market price changes ❿ VAR increases with the length of the time period (increases with the time) ❿ Check VAR example in Box 16.5 + Discussion Board Chap 16C, p. 15 Limitations in the Regression Approach ❿ Using market data presumes that financial markets are efficient, and that share prices respond quickly and appropriately to exchange rate changes. ❿ The regression approach produces estimates that are subject to estimation error. ❿ For the exposure coefficient to be useful, the relationship between spot rate changes and MV must remain stable in the future. Many things could change. ❿ The approach is unsuitable for newly organized or reorganized firms for which there is not a large sample of consistent observations. Chap 16C, p. 16 8
Limitations in the Scenario Approach ❿ The number of scenarios is infinite Analyze likely, plausible scenarios Analyze unlikely, implausible scenarios (Stress testing) Analyze rudimentary or complex scenarios? How to aggregate information across scenarios? ❿ Draws on firm specific information Price elasticity of demand for outputs Price over cost mark-up ❿ Draws on industry specific information What will competing firms do? Exploit price advantages for market share, profit, R&D, etc. Chap 16C, p. 17 Limitations in the VAR Approach ❿ Statistical parameters must be estimated (volatility of asset returns σ, correlation among asset returns ρ) Can use historical or implied methods to estimate these ❿ Assumption about normality of underlying distribution of price changes Market prices may experience abnormal jumps ❿ Market policies may be subject to abnormal shifts (new President, Congress, Fed chairman) Chap 16C, p. 18 9
Empirical Evidence on Firm Profits, Share Prices, & Exchange Rates ❿ During the Bretton Woods pegged-rate period, the general stock market index tended to move up (down) immediately after a devaluation (revaluation) of the local currency. ❿ Studies also indicated that exposure coefficients vary from firm to firm within the same industry and over time, and that exchange rate changes can have a substantial impact on the overall economy. Chap 16C, p. 19 Summary on Exposure Measurement ❿ Exposure seeks to estimate MV/ S ❿ Task is complex and data intensive Regression: Past equity market values and spot FX rates Scenario: Prices and quantities determine cash flows VAR: Mean, volatility and correlation of asset returns ❿ Each technique produces an estimate that depends on various assumptions ❿ Exposures are likely to be smaller in the long-run than in the short-run Because of PPP, International Fisher Effect, and competitive forces across firms Chap 16C, p. 20 10