RDP 9312: A Re-examination of the Determinants of Australia's Imports 2. Trends in Import Penetration

The relationship between economic activity and imports is shown in Figure 1, where the annual change in real gross national expenditure (GNE) is compared with that of the volume of endogenous imports.[2] It is apparent that there is a positive correlation between real GNE and imports. In fact, typically, the change in the volume of imports is more than proportional to that in GNE. This is evident in each cycle of economic activity. The recession of 1982/83 was associated with a sharp fall in the growth of imports. Similarly, the recent recession has been associated with a fall in import growth. However, this fall has been less pronounced than that of 1982/83. Furthermore, in contrast to previous episodes, import growth has since recovered ahead of economic activity.

Figure 1: Endogenous Imports and Economic Activity

The relationship between relative prices and imports is shown in Figure 2. Here, the annual change in the volume of endogenous imports is compared with that of the relative price of imports (measured as the deflator for endogenous imports relative to the deflator for GNE).[3] A negative correlation between the series is evident for most of the period shown. The sharp increases in import prices between 1984 and 1986 corresponded to a fall in the volume of imports, whilst the fall in import prices in the late 1980s corresponded to a surge in import growth.[4] Recently, however, a positive correlation between the series has emerged: a rise in the relative price of imports has coincided with a recovery of import volumes.

Figure 2: Endogenous Imports and their Relative Price

A comparison of Figures 1 and 2 reveals that the recent history of import growth has some interesting properties. In 1990/91, whilst the economy was in recession, the relative price of imports was quite low. Such low relative prices in this period may have offset, to some extent, the downward pressure on import volumes associated with weak activity. More recently, the relative price of imports has increased, but this price increase is now coincident with the emergence of economic recovery. These offsetting influences make less clear the driving force for the recent growth in imports. What is clear, however, is that import penetration has increased.

As shown in Figure 3, the constant price value of imports as a share of GNE has exhibited a trend rise. Whilst there are cyclical movements around this trend, it appears that there has been a general ‘ratchetting up’ of import penetration. This may reflect an underlying structural change in the market for importables. Such change may be related to the increased openness of the Australian economy.

Figure 3: Import Penetration Ratio

The openness of the Australian economy is represented in this paper by changes in the effective rate of protection.[5] In 1973, there was a 25 per cent across-the board reduction in tariffs. The rate of protection then plateaued for a period, subsequently rising in the early 1980s. However, since its local peak in 1984, the effective rate of protection has fallen by almost half (see Figure 4). This represents an important structural change in the market for importables.

Figure 4: Effective Rate of Protection

Reductions in protection can be expected to affect the demand for imports, both directly and indirectly. A fall in protection increases the demand for imports directly by lowering their landed price. However, the effect of tariff changes on the landed price of imports is found, most often, to be relatively small – at least in the short run. Typically, it is swamped by the effect of exchange rate movements so that, in the short run, both the landed and free on board prices of imports move similarly.[6] In the long run, though, one might expect the effect of tariff reductions to generate a more discernible change in the landed price of imports. Reductions in protection may, as well, have important indirect effects on import demand. A credible commitment to the dismantling of protection will influence the behaviour of economic agents. As they come to recognise that the process is irreversible, they may adapt to this change so that, at a given level of output or relative price, they may demand more imports than would otherwise have been the case.

Reductions in protection will, however, also affect the domestic supply of import substitutes. A fall in the rate of protection increases the price of import substitutes relative to imports, reducing demand for the domestically produced good. Then, if producers recognise that the dismantling of protection will not be reversed, so that demand for import substitutes will be permanently lower, it can be expected that a corresponding reduction in the supply of such goods will occur. Using input-output data for selected years, Dwyer (1992) has shown that the share of gross domestic product (GDP) attributable to import-competing industries has fallen since the early 1980s. A proxy for the supply of domestically produced substitutes is considered here.[7] As shown in Figure 5, throughout the 1980s, such supply has fallen both absolutely and as a share of GDP.[8] The fall appears to have been accelerated during the two major recessions. However, if one abstracts from these recessions, it also appears that the trend decline in the supply of import substitutes has become significantly greater since the early 1980s, as the economy has become more open.

Figure 5: Supply of Import Substitutes

This result has an important implication. Comment on the growth in import penetration has typically been couched in terms of the impact of international integration on import demand.[9] And yet, it is evident in Figure 5 that greater openness has been accompanied by a fall in the supply of import replacements. Supply side influences may, therefore, be important determinants of the import decision.

We commence with the testable proposition that international integration has been accompanied by a structural change in the market for importables, impacting upon both the demand and the supply side. The framework for our analysis is outlined below.


Endogenous imports are those that exclude ‘lumpy items’ such as fuel, aircraft and defence equipment. These items do not typically respond to changes in domestic demand. For the purpose of this paper, imports of computers have also been excluded. [2]

Note that published price deflators for imports record import prices over the docks and, as such, are not the final price faced by the purchasers of imports. [3]

A detailed discussion of this episode can be found in Horton and Wilkinson (1989) and Wilkinson (1992). [4]

Measured by the Industry Commission's effective rate of assistance. An effective rate of assistance measures the percentage change by which all forms of assistance raise an industry's value added per unit of output. That is, it accounts for the fact that tariff and non-tariff protection is also applied to inputs to production. Plunkett, Wilson and Argy (1992) describe the construction of the series. [5]

This appears to have always been the case in Australia. See Gregory and Martin (1976) for a discussion of this with respect to the 25 per cent tariff reduction of 1973. They show that even with this massive tariff reduction, exchange rate effects dominated movements in the final price of imports. [6]

Industries are defined as import competing according to the ratio of the volume of imports to the volume of domestic sales reported in Kent and Scott (1991). The constant price GDP of these import-competing industries less exports is then expressed as a share of GDP for the economy as a whole. This approach yields results broadly consistent with those found by Dwyer (1992), using disaggregated data for selected years in which input-output data are available. For further discussion see Appendix 1. [7]

It might be argued that Australia would experience a secular decline in import substitutes as the economy becomes more developed and increases the share of non-tradeables in both production and consumption. However, the extent of the decline shown here exceeds that which can be explained by such secular change. [8]

In fact, as noted by Phillips (1989), this comment is often couched in terms of the ‘immorality’ of demand for imports. [9]