RDP 9304: Exchange Rate Pass-Through: The Different Responses of Importers and Exporters 6. Analysis of Results

Estimation of the pass-through relationship for Australia has shown that first stage pass-through of changes in the exchange rate is complete in the long run for the prices of both imports and manufactured exports. This result satisfies priors about Australia as a small open economy. Nonetheless, a qualification needs to be made. In the pass-through literature, it is implicitly assumed that movements in the exchange rate are exogenous. Changes in the exchange rate may well be associated with changes in the domestic prices of traded goods relative to those abroad and thus be endogenous. In such a case, establishment of a cointegrating relationship may merely indicate that purchasing power parity (PPP) holds in the long run. If so, a result of complete pass-through would be found by default.^[27]

However, it is well established that Australia has experienced exogenous shocks to the exchange rate. This is most apparent during the episode of exchange rate overshooting during the mid 1980s (O'Mara, Wallace and Meshios 1987). It has been shown that during such episodes of exchange rate movement, as well as generally, there has been a reversion of the domestic price of traded goods to their long-run equilibrium. Thus the issue of potential endogeneity of the exchange rate does not appear to detract from the finding of complete exchange rate pass-through.

Whilst it is concluded that exchange rate pass-through is complete in the long run, this is not so in the short run. The short-run dynamics of adjustment to long-run equilibrium contain important information about changes in relative prices. In particular, they illustrate the different responses of domestic import and export prices to exchange rate shocks.

6.1 Import Price Pass-Through

It was shown that following depreciation of the currency, adjustment of import prices is fairly rapid with most of the price effect passed on within two quarters and full pass-through occurring in about one year. Furthermore, this pattern of price adjustment appears to have been maintained throughout the estimation period. This is evidenced by the results of recursive regressions.

A stable pattern of pass-through indicates that there is symmetry in the response of import prices to exchange rate movements.^[28] Furthermore, it indicates that the pattern of pass-through has not altered even during the recent recession: despite subdued domestic demand, foreign price-setting agents appear not to have altered their pricing policies.^[29] The rate of pass-through being experienced during the latest episode of depreciation is, in fact, typical. Consequently, the hypothesis of a change in the response of foreign price-setting agents to Australia's latest episode of currency depreciation is rejected. Again, this may be expected to occur for a small open economy which is a price taker in world markets.

This finding has several implications. First, if the present rate of pass-through accords with historical experience, it follows that import prices over the docks will increase further as they move towards their long-run equilibrium value.

Second, if the current pattern of pass-through is typical, the exchange rate has not become less effective in inducing changes in the relative producer prices of imports and non-traded goods. Consequently, the lack of substitution in domestic production between importables and non-traded goods, as implied by high rates of import penetration, must be attributable to factors other than the exchange rate.

Third, if patterns of pass-through are normal with respect to the price of imports over the docks, the direct inflationary consequences of depreciation are being kept in check by other factors. These include developments in world prices and the second stage of price adjustment.

World prices of traded goods have been subdued since 1989/90. Thus low world prices have, in a sense, insulated the economy from the short-term inflationary stimulus provided by currency depreciation.

Furthermore, there appears to be significant price adjustment at the second stage: local distributors appear not to be passing on the full cost of an imported item to consumers. This is suggested in Figure 11 where it is shown that the final retail price of imports has moved by proportionately less than that over the docks.^[30] The relative rates of change in the free on board and retail prices of imports contain information about the second stage of price adjustment. If it is assumed that cost structures remain stable, an increase in the free on board price of imports relative to their retail price represents a narrowing of the distributors' margin. On this basis, it can be inferred that there was considerable pressure on margins during the record currency depreciation of the mid 1980s. Subsequently, pressure on margins eased, especially as the currency appreciated.^[31] During recent quarters, margins appear to have been squeezed again so that direct inflationary effects of depreciation are being absorbed, to some extent, at the second stage.

One might speculate that an attempt will be made to ‘catch up’ on lost margins which will generate inflationary pressure. However, the extent of pass-through at the second stage is governed by the scope for distributors to alter existing relative prices of final goods. Essentially, this is a function of the degree of substitutability between imports and domestically produced goods. Much of the record currency depreciation in the mid 1980s was absorbed at the second stage because market demand curves for final goods were sufficiently elastic for full pass-through to be sub-optimal with respect to profits. During the recent albeit less dramatic fall in the exchange rate, it appears that currency depreciation is again being absorbed at the second stage. This time, an environment of low inflation is imposing an additional constraint on pass-through: if the price of a domestic substitute has not changed, the scope for altering the retail price of an import is further limited. This increases the pressure for cost reduction in order to preserve viable margins. Thus second stage pass-through is essentially a microeconomic issue. The way in which it impacts on the macroeconomy is recommended for future research.

6.2 Export Price Pass-Through

It was shown that following depreciation of the currency, adjustment of manufactured export prices is significantly slower than that of imports. In fact, it takes almost two years for pass-through to be near complete. The slow adjustment of manufactured export prices relative to that for imports has a balance of payments implication. If import prices respond fairly rapidly to exchange rate movement, but manufactured export prices persist at levels below their long-run equilibrium, there results an effect on the current account balance analogous to that of the J-curve. This appeared to occur in the mid 1980s. Conversely, in the late 1980s, actual manufactured export prices remained above their long-run equilibrium, making a small positive contribution to the current account balance.^[32] The differential rate of response of import and export prices to the exchange rate implies, other things being constant, that there is some degree of endogeneity in the nation's terms of trade.^[33]

The response of manufactured export prices to exchange rate changes differs to that of import prices not only with respect to speed. Whereas a typical pattern of import price pass-through can be established, this is not so for manufactured exports. The results of recursive regressions indicate that since 1986/87 there has been a significant increase in the rate of pass-through. This suggests a change in the pricing behaviour of exporters.

This possible change in pricing policy is reinforced in Figure 12 in which it is shown that the export price of manufactures and the corresponding domestic price track each other closely until 1986/87. After this time, manufactured export prices deviate significantly from the domestic price. This break accords with the increase in the rate of exchange rate pass-through. It also accords with the increase in export orientation of the Australian manufacturing industry (Menzies and Heenan 1993; Reserve Bank of Australia 1992). Furthermore, the share of contracts invoiced in Australian dollars has fallen considerably.^[34] In combination, these factors suggest that there has been a shift in focus from pricing to the domestic market to parity with the world price. Certainly, such a result is consistent with the increase in international integration of the Australian economy.

If such a shift in pricing policy has occurred, it suggests that with continued international integration, the endogeneity in the nation's terms of trade that emerged during the mid 1980s will be dissipated, along with attendant effects on the balance of payments.

Footnotes

In fact, proponents of the PPP doctrine would argue that, ultimately, complete pass-through will occur for any economy and so not be peculiar to the case of a small open economy. [27]

That is, the extent to which domestic import prices rise when the currency depreciates equals the extent of their fall when the currency appreciates. This finding is contrary to reports that the pattern of pass-through was different during the episode of currency appreciation in the late 1980s (Prices Surveillance Authority 1989). [28]

Confirmation of this was provided by testing the speed of pass-through with respect to the effect of domestic demand pressures. This was done by including a measure of the output gap in the short-run dynamics of the P-H ECM and the UECM. However, this variable was found to be insignificant in both models. Whilst this does not imply that macroeconomic conditions are irrelevant to the speed of pass-through, there is no statistical indication that foreign suppliers consistently take account of Australian domestic conditions when setting the prices of their exports to Australia. [29]

The imported items component of the CPI and the implicit price deflator for imports are not strictly comparable due to compositional differences and alternative weighting systems. Thus the import price index for consumption goods is compared to the imported items component of the CPI. An attempt has been made to reconcile the basket of imported goods contained in the CPI and the index of free on board prices of imported consumption goods. Details are available on request. [30]

In fact, during the late 1980s margins appeared to recover to such an extent that the Prices Surveillance Authority (1989) conducted an inquiry into the issue of exchange rate pass-through. [31]

However, to the extent that manufactured exports are still a relatively small component of total export earnings, this effect is not likely to be very large. [32]

This finding is consistent with that of Treasury Department (1986). [33]

Unpublished ABS data indicate that the share of contracts for exports of manufactures invoiced in Australian dollars has fallen from around 73 per cent in 1984/85 to around 67 per cent in 1991/92. Conversely, in most years, around 37 per cent of contracts for total merchandise exports are invoiced in Australian dollars. [34]