The Treasury

Global Navigation

Personal tools


Past, Present and Future Developments in New Zealand's Terms of Trade - WP 06/09

4  Why have the trend and volatility changed?

4.1  Compositional change in New Zealand’s goods exports

4.1.1  Impact on the trend

Figure 5 above reveals that New Zealand export prices have risen relative to world commodity prices. There are a number of possible explanations for why this has occurred. Firstly, it may be that New Zealand goods experience a premium over other similar commodities overseas. This would suggest that the “law of one price” in international commodity markets does not hold and that commodity prices experienced by various countries do not converge in the long run. This could be because New Zealand is becoming more of a ‘price-maker’ in the international market. That is, its producers now have more price setting power now than they did in the past. Or it could be that other countries may have displayed higher productivity growth rates over time, allowing them to be more competitive at lower prices.

However, the more likely reason for New Zealand’s export prices outperforming world commodity prices is compositional change. Over history, the types of goods New Zealand exports have gone through significant transformation. Exporting has moved from a focus on a few primary commodities to the exporting of a much broader range of goods.[14] This compositional change is illustrated in Figure 7 below.

Figure 7 – New Zealand’s Good Export Composition (5 year moving average)
Sources: NZIER, Statistics New Zealand

As described by Briggs (2003) and shown in Figure 7, wool exports as a proportion of total goods exports have decreased significantly over time, from a peak of 75% in 1860, to just 3% in 2001.[15] With the exception of the 1860s “gold-rush” (when gold became the largest exported commodity) and the 1950’s Korean War (when demand for wool increased), this decrease in the proportion of wool exports has been relatively steady. The decline in the demand for wool (as a result of the development of cheaper man-made substitutes) and the development of other export products explain this decline in wool’s share of total goods exports.

The development of refrigeration in the late 19th Century saw dairy and meat products emerge as important export industries and by 1902, the value of wool exports was exceeded by the combination of dairy and meat exports.

Dairy, meat and wool products continued to be the major exported commodities up until the late 1960s. After this period there was an increasing contribution from the ‘other’ category of goods exports. ‘Other’ goods exports are made up of products such as fruit, aluminium and manufactured goods. The fact that the prices of manufactured goods (according to the Prebisch-Singer hypothesis) have not decreased by as much as primary commodities could partly account for the increases seen in New Zealand goods export prices relative to world commodity prices. The ‘other’ goods export category has increased considerably as a proportion of total goods exports in the last 40 years. This has been aided by trade treaties with New Zealand’s major trading partner Australia (the largest destination for manufactured exports). These treaties are the New Zealand Australia Free Trade Agreement in 1965 and Closer Economic Relations (CER) in 1982. This increase in the ‘other’ goods category has led to primary goods exports as a proportion of total goods exports decreasing from 87% in 1965 to just over 50% in 2005.[16],[17]

It is difficult to test empirically how the compositional change in New Zealand’s goods exports has affected the level of the terms of trade. There are feedbacks and endogenous movements that make separating out the specific effects complex. The real question of interest should be the ability of the economy to respond to external changes in both relative prices and conditions in the market place. An economy that is more dynamic will be able to benefit from these changes by shifting resources to areas where they are used more efficiently than an economy that is sluggish to respond to external changes. It is the expectation of future movements in relative prices that lead to shifts in resources. However, as it is impossible to know what people’s expectations of future movements are at any point in time, the impact of compositional change on the terms of trade must be tested in some other way. Grimes (2006) attempts to do this by holding export value shares constant at their 1972 levels and compares the implied terms of trade with the current terms of trade. He finds that if New Zealand was still exporting the same bundle of goods as it did in 1972, the terms of trade would be higher today. However, there are some issues which need to be considered with this form of analysis.

The use of export value shares as weights could bias the results as these shares are not independent of the price level in that year. It would be more appropriate to use export volume shares as suitable weights. However, accurate historical volume data is not available for all products and even if it were, the way in which volume or “real” measures are calculated is not totally independent from the price level as a base price series is needed as a deflator in order to calculate a “real” series. Related to this is the choice of the base year in which to hold export shares constant. If the base year chosen corresponded to a year that experienced some significant price shocks, then this would affect the individual export value shares. As an experiment, this paper tests different base years, namely 1960 and 1980 and compares them with Grimes’ base year of 1972. It is found that if the bundle of goods was held constant at their 1960 shares then the terms of trade would be considerably lower than they actually were in 2005, while if the 1980 export shares are used as a base, then the terms of trade would be slightly higher than they actually were in 2005. This is displayed in Figure 8. While the analysis of Grimes is more comprehensive than that used in this paper (as he uses much more disaggregated data on export shares and prices), the results using the 1960 and 1980 base years show that the choice of base year is important.

Figure 8 – Goods Terms of Trade Using Actual, 1960 and 1980 Export Shares
Source: NZIER, Statistics New Zealand, Author’s Calculation

Figure 8 illustrates that if New Zealand was still exporting the bundle of goods it was in 1960, the terms of trade would have been considerably lower. This shows that the compositional change in New Zealand’s goods exports has had a positive impact on the terms of trade over the past 40 years. The results using the 1980 export shares show that it is only in the last three years that the re-weighted terms of trade are higher. This is likely to be due to the significant increases in the prices of dairy and meat exports over these years and because dairy and meat exports made up a greater share of total goods exports in 1980 than currently, which results in the re-weighted terms of trade being at a higher level.

The destinations for New Zealand’s exports have also changed dramatically in the last century. New Zealand has moved from having a single country (i.e. the UK) as its main export market, to a scenario where New Zealand’s exports are now sent to a much wider range of countries. The UK now takes less than 5% of New Zealand’s goods exports compared with over 80% in the late 1930s. When the UK joined the EEC in 1973 New Zealand was “forced” to diversify the countries it exported to and Australia as well as many Asian countries increased their share. The “other” category of countries has increased the most and indicates the greater proportion of exports sent to Asian countries. This is shown in Figure 9.

Figure 9 – Destination of New Zealand’s Good Exports (5 year moving average)
Sources: NZIER, Statistics New Zealand

* The “other” category includes some EU countries prior to its formation.


  • [14]Although not the focus of this paper, New Zealand has also experienced a compositional shift towards services exports, as well as that which occurred within goods exports.
  • [15]These numbers are actual figures and therefore do not match accurately with the data from Figure 7 as that data has been smoothed.
  • [16]‘    Primary’ good exports are defined as the sum of wool, fishing, dairy, forestry, gold and meat exports.
  • [17]For more information on New Zealand’s primary sector from a historical perspective see Harrington (2005).
Page top