Summary of Economic Exposure Impact of Krona Devaluation on Spectrum Manufacturing ab
*Includes a gain of $150,000 on loan repayment.
This section presents a workable approach for determining a firm’s true economic exposure and susceptibility to exchange risk. The approach avoids the problem of using seat-of-the-pants estimates in performing the necessary calculations.5 The technique is straightforward to apply, and it requires only historical data from the firm’s actual operations or, in the case of a de novo venture, data from a comparable business.
This approach is based on the following operational definition of the exchange risk faced by a parent or one of its foreign affiliates: A company faces exchange risk to the extent that variations in the dollar value of the units cash flows are correlated with variations in the nominal exchange rate. This correlation is precisely what a regression analysis seeks to establish. A simple and straightforward way to implement this definition, therefore, is to regress the changes in actual cash flows from past periods, converted into their dollar values, on changes in the average exchange rate during the corresponding period. Specifically, this involves running the following regression:6
where
ΔCFt =CFt − CFt-1, and CFt equals the dollar value of total affiliate (parent) cash flows in period t
ΔEXCHt = EXCHt − EXCHt-1, and EXCHt equals the average nominal exchange rate (dollar value of one unit of the foreign currency) during period t
u = a random error term with mean 0
The output from a regression such as Equation 11.3 includes three key parameters: (1) the foreign exchange beta (ß) coefficient, which measures the sensitivity of dollar cash flows to exchange rate changes; (2) the t-statistic, which measures the statistical significance of the beta coefficient; and (3) the R2, which measures the fraction of cash flow variability explained by variation in the exchange rate. The higher the beta coefficient, the greater the impact of a given exchange rate change on the dollar value of cash flows. Conversely, the lower the beta coefficient, the less exposed the firm is to exchange rate changes. A larger t-statistic means a higher level of confidence in the value of the beta coefficient.
However, even if a firm has a large and statistically significant beta coefficient and thus faces real exchange risk, it does not necessarily mean that currency fluctuations are an important determinant of overall firm risk. What really matters is the percentage of total corporate cash-flow variability that is due to these currency fluctuations. Thus, the most important parameter, in terms of its impact on the firm’s exposure management policy, is the regressions R2. For example, if exchange rate changes explain only 1% of total cash-flow variability, the firm should not devote much in the way of resources to foreign exchange risk management, even if the beta coefficient is large and statistically significant.
Limitations
The validity of this method is clearly dependent on the sensitivity of future cash flows to exchange rate changes being similar to their historical sensitivity. In the absence of additional information, this assumption seems to be reasonable. However, the firm may have reason to modify the implementation of this method. For example, the nominal foreign currency tax shield provided by a foreign affiliate’s depreciation is fully exposed to the effects of currency fluctuations. If the amount of depreciation in the future is expected to differ significantly from its historical values, then the depreciation tax shield should be removed from the cash flows used in the regression analysis and treated separately. Similarly, if the firm has recently entered into a large purchase or sales contract fixed in terms of the foreign currency, it might decide to consider the resulting transaction exposure apart from its operating exposure.
5 This section is based on C. Kent Garner and Alan C. Shapiro, “A Practical Method of Assessing Foreign Exchange Risk,” Midland Corporate Finance Journal, Fall 1984, pp. 6-17.
6 The application of the regression approach to measuring exposure to currency risk is illustrated in Garner and Shapiro, “A Practical Method of Assessing Foreign Exchange Risk,” and in Michael Adler and Bernard Dumas, “Exposure to Currency Risk: Definition and Measurement,” Financial Management, Summer 1984, pp. 41-50. We use changes, rather than levels, of the variables in the regression because the variables are nonstationary. In addition, such a regression may include lagged values of EXCHt given that sales and costs often respond with a lag to exchange rate changes.