RDP 9005: Real Exchange Rates and Australian Export Competitiveness 4. Results
August 1990
- Download the Paper 1.1MB
Pre-Float Period
Five real export exchange rate indices were calculated in this paper – one for each of the alternative weighting schemes considered. In all of the following figures, an appreciation of the exchange rate implies a rise in the real exchange rate, and hence a fall in competitiveness. In Figure 1, the moving average and the third country weighted real exchange rate series are compared with the ABARE real exchange rate index.[14] Overall, these three series exhibit broadly similar movements. For all series, the appreciation of the early seventies was almost entirely eroded by about 1977, and from then until around 1984 there was little movement in the series.
The most notable feature of Figure 1 is that the third country index is at a higher level than the other indices for almost all of the period.
This indicates that our level of export competitiveness has been lower over the seventies and eighties than the other measures suggest. The gap between the three measures first opened during the real appreciation of 1973/74, and widened during 1977/78 when the third country index depreciated by less than the other two indices. The divergence in the measures is dominated by the higher weight placed by the third country index on the real exchange rate between Australia and the USA or Canada, and the lower weight on the exchange rate between Australia and Japan.
In Figures 2 and 3, the real exchange rates between Australia and each of its six major trading partners are shown to further highlight this point. The sharp appreciation of the real exchange rate with Canada and the US in 1973/74 is a major factor in the rise in the third country index. The sharp fall in the real exchange rate against Japan in the second half of the seventies also helps explain the larger falls in the bilateral indices compared with the third country index.
It is also interesting to note that for most of the seventies and eighties Australia has been less competitive with the USA and Canada than it was at the beginning of the seventies. This is particularly important given that we produce very similar products to these countries, and thus compete with them widely in Australian and overseas markets.
Our real exchange rates with Germany, New Zealand and the United Kingdom have not shown as dramatic trends as the other three exchange rates over the pre-float period.
Post-Float Period
Table 2 summarizes the changes in the three main indices since the float in December 1983. This table shows that the appreciation of the Australian dollar to December 1989 has largely eroded the real depreciation since the float.
| Period | Annual | MA | Spliced | 80/89 | 3rd Country |
|---|---|---|---|---|---|
| Mar Q 84 to Sep Q 86 | −33.2 | −32.7 | −32.6 | −32.9 | −29.5 |
| Sept Q 86 to Dec Q 89 | +31.2 | +30.6 | +31.5 | +31.6 | +28.4 |
| Mar Q 84 to Dec Q 89 | −12.4 | −12.1 | −11.3 | −11.7 | −9.4 |
Referring again to Figures 2 and 3, we can see that of the 48% real depreciation against the Japanese Yen between March quarter 1984 and September quarter 1986, less than half was eroded – i.e. in December quarter 1989 the index was still 26% below its level of March quarter 1984. In contrast, although the Australian dollar depreciated by over 25% against the US dollar between March quarter 1984 and September quarter 1986, this fall was entirely eroded – by December 1989 the index was only five percent above its level of March quarter 1984. Examining the real exchange rate with Germany, New Zealand and the United Kingdom shows that Australia was in a substantially more competitive position in December quarter 1989 than it was in March 1984. Overall, however, the increase in competitiveness from the depreciation of the Australian dollar in 1985 and 1986 was substantially diminished, as Figure 1 clearly shows.
Composition of Changes
Table 3 shows the percentage changes in the different components of the real exchange rate index based on 1980/89 export shares. A fixed weight index is used to abstract from the effects of changing weights. The depreciation in the index to September 1986 was largely driven by nominal exchange rate depreciation, despite an eleven percent increase in relative price levels. The subsequent appreciation of the real exchange rate occurred as a result of both exchange rate and relative price increases. Overall, however, this table clearly shows that the main influence on changes in the real exchange rate over the period considered was nominal exchange rate change, rather than relative inflation.
| Period | RER Index | Relative Prices | Exchange Rates |
|---|---|---|---|
| Mar Q 84 to Sep Q 86 | −33 | +11 | −39 |
| Sept Q 86 to Dec Q 89 | +32 | +13 | +16 |
| Mar Q 84 to Dec Q 89 | −12 | +25 | −29 |
Comparison with Other Measures
The behaviour since the float of several different measures of Australia's real exchange rate are shown in Table 4. This table compares the real exchange rate index calculated by ABARE and the IMF with the measures developed in this paper.
| Period | MA | 3rd Country | IMF | ABARE |
|---|---|---|---|---|
| Mar Q 84 to Sep Q 86 | −32.7 | −29.5 | −34.3 | −34.2 |
| Sept Q 86 to Dec Q 89 | +30.6 | +28.4 | +29.9 | +28.3 |
| Mar Q 84 to Dec Q 89 | −12.1 | −9.4 | −14.7 | −15.6 |
The real effective exchange rate calculated by the IMF is a geometric fixed weighted average of CPIs, expressed in a common currency.[16] The IMF index takes account of both bilateral and third country competition. The weights are calculated by aggregating four sets of weights based on exports and imports of manufactured goods and primary products for the period 1980–1982. Weights for manufactured goods are simple averages of bilateral and third country weights. Primary products are weighted according to shares in world imports and exports on a commodity-by-commodity basis, aggregated across a large number of commodities. The index is based in 1985.
The IMF index provides an estimate of 14.7% for the fall in Australia's real exchange rate since the float. This estimate is larger than that suggested by both the bilateral and third country based indices. This may be due to the combination of third country effects upon imports and internal trade not measured by the indices calculated in this paper. The differences may also reflect the fact that our indices are weighted using more recent trade data.
The ABARE real exchange rate index is an arithmetic average of CPIs, adjusted for exchange rates. The index is based in 1980, with weights calculated according to the trade share of twelve of Australia's trading partners, updated annually. The ABARE index shows a net fall of 15.6% since the float. The differences between our index and the ABARE index relate mainly to the methodology and coverage of the respective indices.[17]
Footnotes
The results of the annual, spliced, and 1980/89 weighted series were found to be very similar to the MA series and so are not presented here. Our preferred index is the MA index, which represents a compromise between using fixed weights and changing weights every year. [14]
The percentage changes are calculated as geometric weighted averages of the respective components, using 1980/89 export shares as weights. The sum of percentage changes of the components will not equal the change in the index, as the sum will only equal the product when percentage changes are small. [15]
See IMF (1990) for a discussion of the methodology behind this index. [16]
Using the same countries and methodology as ABARE, we estimate a fall of 14.8% since the float. This difference may reflect different data definitions and our use of smoothed quarterly weights. By extending the coverage of the ABARE index to 22 countries and retaining their methodology, this fall is reduced to 10.1%. Changing the base year from 1980 to 1985 reduces the change since the float to 13.8% for the 12 countries, then 10.3% when extended to 22 countries. Using a geometric averaging technique, the change in the index equals 11.1% for the 22 countries. [17]