The Treasury

Global Navigation

Personal tools


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

2  Why are the terms of trade important?

2.1  Terms of trade trends and economic growth

As movements in the terms of trade reflect changes in relative prices, it is often unclear how these movements affect the real economy. Although this has been debated extensively in the literature to date, there is still no consensus view about how trends in the terms of trade impact on economic growth.

The most common view is that the terms of trade has a positive impact on economic growth. An increase in export prices relative to import prices allows a larger volume of imports to be purchased with a given volume of exports. The implied increase in the real purchasing power of domestic production is equivalent to a transfer of income from the rest of the world and can have large impacts on consumption, savings and investment. The terms of trade can also be thought of as a rate of return on investment and therefore a secular improvement in the terms of trade leads to an increase in investment and hence economic growth. A graphical illustration of the income effect of a movement in the terms of trade is shown in Figure 2. Real gross domestic income (RGDI) measures the purchasing power of the total income generated by domestic production. The difference between real GDP and RGDI is defined as the terms of trade effect. The appreciation of the terms of trade over 2004 led to a boost in real incomes and this is shown by RGDI exceeding real GDP over 2004 and 2005.

Figure 2 – Terms of Trade Effect on Purchasing Power
Source: Statistics New Zealand

Although the changes to real incomes from terms of trade movements can be seen in Figure 2, the total economy-wide impacts of terms of trade movements are hard to quantify. Changes in the terms of trade can have different macroeconomic impacts depending on the composition of the relative price movements. If a fall (rise) in the terms of trade is due to a decrease (increase) in export prices, then this will initially impact on exporters before indirectly affecting households. However, if a fall (rise) in the terms of trade is a result of an increase (decrease) in import prices (for example, oil prices), this is likely to affect households and businesses more directly and the macroeconomic shock will be different.

Shifts in production and resources are often associated with relative price movements. An external shock such as an increase in prices which benefits one sector of the economy often leads to increased investment in this area or resources shifting to this sector from another. An example of this was the significant dairy conversions of the 1990s. Land used for dairy farming increased by over 35% between 1990 and 2000 while land used for sheep and beef farming fell by close to 13%.[3] This was due mainly to changes in relative prices as New Zealand dollar dairy prices increased by 31% during this period as opposed to only a 1% increase in combined meat and wool prices. The impact on economic growth as resources shift from one sector to another is difficult to interpret, as they are likely to be coupled with productivity changes and therefore the sole impact of a relative price change is hard to quantify.

Harberger (1950) and Laursen and Metzler (1950) were some of the first to look at the impact of a terms of trade shock on an economy. They suggested that a deterioration in the terms of trade will reduce a country’s real income (or increase real expenditure for a given income level) consequently decreasing savings, through consumption smoothing behaviour. This later became known as the Harberger-Laursen-Metzler effect. Obstfeld (1982) and later Kent and Cashin (2003) extended this idea and showed that the duration or persistence of terms of trade shocks are important when determining the effect on an economy. A longer or more persistent shock may result in lower investment and potentially higher saving in anticipation of lower future output.

Much of the current literature looking at the relationship between a secular trend in the terms of trade and economic growth has concentrated on explaining cross-country differences between developing and industrialised nations. Some of the first to approach this idea were Prebisch (1950) and Singer (1950) who proposed that developing countries had experienced a downward trend in their terms of trade relative to developed countries. This theory (later known as the Prebisch-Singer hypothesis) has been one of the most extensively researched ideas in development economics. It states that over time, the price of primary commodities relative to the price of manufactured goods should decrease. Cashin and McDermott (2002) state that this is a result of a lower income elasticity of demand for commodities as well as smaller productivity increases for manufactured goods, while Gillitzer and Kearns (2005) suggest that manufactured goods are much less homogeneous than commodities and therefore producers have more price setting power. Related to this is the observation that primary commodities are generally products with low barriers to entry. These products are therefore more likely to have experienced increased competition with price pressures and tight margins than manufactured goods for which there is less competition as the products are more difficult to produce.

Grilli and Yang (1988), in a commonly cited study - and whose data is the basis for much of this field of research - found evidence supporting the idea of decreasing real commodity prices (the ratio of primary commodity prices to manufactures prices). More recently studies by Lutz (1999) and Cashin and McDermott (2002), using different methodologies, also find evidence supporting the Prebisch-Singer hypothesis.

However, there are studies that disagree with the Prebisch-Singer hypothesis of a secular decline in real commodity prices. The debate surrounds the appropriate use of deterministic or stochastic trends and the calculation of structural breaks. Powell (1991) found that after allowing for three breaks in the series, non-oil commodity prices and manufactured good prices are cointegrated, implying that the commodity terms of trade is stationary and therefore not declining over time. Kellard and Wohar (2006) allowed for two structural breaks and found little evidence in support of the Prebisch-Singer hypothesis. They find that for the majority of commodities, a single downward trend is not the best representation but rather a “shifting trend” which often changes sign over the sample period is more appropriate. They argue that previous literature that finds evidence of a single downward sloping trend as support for the Prebisch-Singer hypothesis is too simplistic. This downward trend may in fact encompass several trends and therefore simplifying the result to a single downward trend may be misleading to policy makers.

Rather than looking at evidence for a secular decline in developing countries’ terms of trade, other studies have focussed on the growth effects of terms of trade trends. Barro and Sala-i-Martin (2004) found that the growth rate of the terms of trade had a positive impact on growth (among other economic and demographic variables) as part of their cross-country regression analysis. Other studies which also find evidence of the positive economic growth effects of terms of trade movements include, inter alia, Mendoza (1997) and Easterby, Kremer, Pritchett and Summers (1993).

Although less common, there are studies that find empirical evidence to suggest a negative relationship between the terms of trade and economic growth. Hadass and Williamson (2001) found that the growth performance of developing nations was reduced by global terms of trade shocks between 1870 and World War I relative to developed countries. However, they found that in fact the terms of trade in these developing countries increased more than it did for the developed nations. Although they do not decide on an explanation for this, they suggest a possible reason. They suggest it could be due to what has come to be known as “resource curse”. Sachs and Warner (1995, 2001) tested this idea empirically and suggest that resource-rich countries generally grow more slowly than resource-poor countries and any relative price shock that increases the value of these resources (particularly natural resources) will hamper development. This may happen for a number of different reasons, including a decline in the competitiveness of other economic sectors (Dutch Disease), the crowding out of human capital through the underinvestment in institutions and education, or as a result of corruption from the mismanagement of revenues from the natural resource sector. Note that the source of the curse is not natural resources, but government mismanagement when resources are present.

2.2  Terms of trade volatility and economic growth

A more recent development in the literature is to study the impact of terms of trade volatility on economic growth. Unlike the debate surrounding the growth effects of long-term trends in the terms of trade, it is generally agreed in the literature that terms of trade volatility has an adverse effect on economic growth. This is usually tested through the channels of uncertainty on investment decisions where increased volatility or uncertainty is associated with increased risk. As the terms of trade can be thought of as a return on investment, increased risk generally leads to a reduction in investment.

As with the trend analysis in the terms of trade, much of the literature looking at the growth effects of terms of trade volatility has focused on cross-country differences, particularly between developing and developed countries. Blattman, Hwang and Williamson (2003) tested the impact of terms of trade trends and volatility on economic growth. They showed that terms of trade movements are a very important determinant of economic growth, with volatility in the terms of trade much more significant than the long-term trends. They also found that this was particularly the case for developing nations whose exports are dominated by primary commodities. Industrialised nations, which are more likely to have a broader export structure, appeared to be significantly less affected by both the trend and volatility of the terms of trade. This therefore makes it hard to interpret for New Zealand. The fact that New Zealand’s goods exports have been dominated by primary commodities leads one to believe that the terms of trade is more important to New Zealand than other industrialised countries.

Mendoza (1997) also looked at the impact of terms of trade volatility on economic growth through the development of a stochastic growth model. In his model, growth in the terms of trade positively impacts consumption growth through increases in income, while volatility in the terms of trade has a negative impact on consumption growth through risk aversion. To test his model’s key findings he ran cross-country panel regressions of developed and developing nations and concluded that terms of trade shocks account for close to half of the differences seen in cross-country growth rates. Like Mendoza, Turnovsky and Chattopadhyay (1998) found a negative relationship between economic growth and terms of trade volatility. However, rather than look at the differences between developed and developing nations, they focused only on small developing nations. One could argue that New Zealand, although not typically classed as a developing nation, has some similarities in the fact that its exports are dominated by primary commodities. Like the findings of Blattman et al (2003), one can draw important implications from this research for New Zealand. It shows that for primary commodity exporters, like New Zealand, it is important for economic growth if volatility in the prices of these commodities and hence the terms of trade is reduced.

It is also commonly agreed upon in the literature that commodity prices have a significant impact on the variation in a country’s terms of trade. Cashin and McDermott (2002) report that on average real commodity prices have declined by approximately 1% per year. However, some prices have changed by up to 50% in a single year. Because of this volatility, interpreting results concerning trends in commodity prices (and the terms of trade) is difficult. Kellard and Wohar (2006) suggest that researchers should move away from analysis of the Prebisch-Singer hypothesis and more towards theoretical and empirical work examining local trends. This may allow a better understanding of the drivers of commodity prices and better policy analysis. Both types of analysis are undertaken below.


  • [3]These statistics are all courtesy of the Ministry of Agriculture and Forestry (MAF).
Page top