This analysis of the Longer-Term Exchange Rate Anchors case study will attempt to forecast the exchange rates of two currencies, the Australian dollar and the Indonesian Rupiah. A better understanding of current currency exchange rates is needed to be able to predict future exchange rates. This case study allows students to look forward to the trend of forecasting future exchange rates.

Kieran J. Walsh

Harvard Business Review (UV7652-PDF-ENG)

December 12, 2018

### Case questions answered:

- When considering the idea of PPP, please have a conversation within your group as to which index the “iPad” or “Big Mac” is more to the spirit of the theory of PPP. You should include a couple of paragraphs summarizing your discussion in the report.
- Please select two countries one Advanced Economy (AE) and one Emerging Market Economy (EME).
- Consider the expected inflation file given for the case, and the current exchange rate from the Big Mac Index, compute the expected future exchange rates for your selected countries assuming PPP holds with certainty for 2020.
- In the data files, you have monthly average currency exchange rates of the home currency for the country to one U.S. dollar. Since the data is annual, it is repeated twelve times to cover the monthly frequency of the exchange rate data from 1994 to 2018. I have lagged the economic data by six months already to keep the timing of the information close. As mentioned in the case IMF uses both PPP as well as REER regression methodologies to compute “long-run” anchors for exchange rates. The IMF model uses more than 20 variables in their regression. If we assume that U.S. inflation is constant (it is of course not constant, but does typically over the long-run averages about 2% annually) we can estimate the future rate by using a simple regression between inflation and the exchange rate. Please estimate this regression for the data on your two countries selected. Use the regression output, and both the current inflation and expected inflation of the selected countries to estimate the future exchange rate for 2020 for both countries. Now, run the same regression, but include the prior year’s exchange rate. Comment on why this might be a good idea and re-estimate the expected exchange rate.

〖Rate〗_(t,j)= ∝ + β_1*〖Inflation〗_(t,j)+ ϵ_(t,j)

〖Rate〗_(t,j)= ∝ + β_1*〖Inflation〗_(t,j)+ β_2*〖Rate〗_(t-1,j)+ ϵ_(t,j) - Given in the data is the implied foreign exchange rate using PPP, run a regression between this figure and the actual rate, what do these results imply for both of the selected countries?

〖Rate〗_(t,j)= ∝ + β_1*〖PPPRate〗_(t,j)+ ϵ_(t,j) - Assume now that the basic model estimated from four is not complete. You now assume, however, that a model containing four variables: Current Account Balance (as a percentage of GDP), the Unemployment rate, and net trade (the difference of imports and exports as a percentage of GDP) does a good job of predicting exchange rates and the lag of the prior year’s rate. Estimate this model and again use the forecast data to estimate the 2020 exchange rate.

〖Rate〗_(t,j)= ∝ + β_1*〖Current Accounts〗_(t,j)+β_2*〖Unemployment〗_(t,j)+β_3*〖Trade〗_(t,j)+ β_4*〖Rate〗_(t-1,j)+ ϵ_(t,j) - Using your knowledge of how exchange rates should work and the data provided as a team consider and estimate one additional model. You should include in the report the logic behind the proposed model. Estimate the model and the forecast rate for 2020.
- Is there a common theme for your selected countries between the information provided in the “iPad” and “Big Mac” Indexes, as well as the four different exchange rates forecasted in 4,6, and 7? Of all the models you explored, which seems like it would do the best job of forecasting the exchange rate? Lastly, consider what you would do if you were a currency market speculator for the two countries you selected. What would you do if you were the CEO of a multinational with large import exposures from these countries?

Not the questions you were looking for? Submit your questions & get answers.

## Case answers for Longer-Term Exchange Rate Anchors

This case solution includes an Excel file with calculations.

1) This analysis of the Longer-Term Exchange Rate Anchors case study will attempt to forecast the exchange rates of two currencies, the Australian dollar and the Indonesian Rupiah. A better understanding of current currency exchange rates is needed to be able to predict future exchange rates.

There are two indexes, the Big Mac Index and the iPad Index, that attempt to compare the purchasing price parity, or currency using a basket of goods, of countries worldwide. The indexes are calculated by dividing the price of their product (Big Mac or iPad) in one country by the price in another country.

The resulting exchange rate is then compared to the benchmarked currency to understand if the currency is on par, undervalued, or overvalued.

The Big Mac Index will be used in this analysis for two main reasons. First, the price of a Big Mac is proportionately the same across countries due to the fact its ingredients are locally sourced, making it reflective of the local economic conditions in each of its respective countries. Assuming that McDonald’s prices its Big Mac’s in a relatively uniform manner, with price-to-cost ratios similar across markets, then the price of a Big Mac can be a useful tool to measure its currency.

In contrast, the iPad Index compares the price of imported iPads, which are manufactured in China and priced for each country by Apple. Second, these indexes attempt to compare the price of a basket of goods to determine if a currency is on par, undervalued, or overvalued, and ultimately expose any opportunities for arbitrage.

Since the ingredients for Big Macs are sourced locally, McDonald’s does not control arbitrage. In contrast, because Apple prices the iPad for each specific country, they can influence arbitrage opportunities.

2) The two countries we are selecting for our analysis are Indonesia for the Emerging Market Economy (EME) and Australia for the Advanced Economy (AE).

3) 10,725.12 IR/AUD$ expected spot rate in one year of Indonesian Rupiah to Australian dollar.

4)In the data files, you have monthly average currency exchange rates of the home currency for the country to one US dollar. Since the data is annual, it is repeated twelve times to cover the monthly frequency of the exchange rate data from 1994 to 2018.

I have lagged the economic data by six months already to keep the timing of the information close. As mentioned in the case, the IMF uses both PPP as well as REER regression methodologies to compute “long-run” anchors for exchange rates. The IMF model uses more than 20 variables in their regression. If we assume that US inflation is constant (it is, of course, not constant, but does typically over the long-run averages about 2% annually), we can estimate the future rate by using a simple regression between inflation and the exchange rate. Please determine this regression for the data on your two countries selected.

Use the regression output and both the current inflation and expected inflation of the selected countries to estimate the future exchange rate for 2020 for both countries. Now, run the same regression, but include the prior year’s exchange rate. Comment on why this might be a good idea and re-estimate the expected exchange rate.

#### Unlock Case Solution Now!

Get instant access to this case solution with a simple, one-time payment ($24.90).