Hungarian Central Statistical Office home Experimental statistics home

Experimental statistics

The impact of changes in the terms of trade on the economy

Released: 28 January 2025

The change in terms of trade is basically a price shear, the relative development of export and import price changes.In case a country’s export price index surpasses that of the import, a higher volume of imports may be realised for a unit of export, and vice versa, this is usually called the improvement or deterioration of the terms of trade.

The terms of trade impact is evaluated in experimental statistics beside the trade in goods, by using the simple terms of trade index, from the total external trade perspective, too. The two terms of trade indices are moving together, most of the time, as the majority of the external trade is still the one in goods, meaning its price changes are decisive, differences, however, may occur. Based on analyses, the total external trade index is below the product range one when terms of trade basically improve (2009, 2013-2016, 2019-2020, 2023) and surpasses it when these basically worsen.

The terms of trade impact adjusted index value has been quantified, beside the GDP volume index, based on these it is apparent that the GDP index volume lessened by 0.9% in 2023, however considering the terms of trade index impact, a growth of 4.7% has been observed in terms of real income. In contrast, part of the significant GDP volume growth of 2021, the total of the 2022 one is annulled by the terms of trade deterioration.

Last but not least the impact of energy price changes in relation to terms of trade changes is presented. A negative rapport is observed between 2001 and 2023, meaning the higher the energy carriers’ price index the more significant the terms of trade deterioration and vice versa. The extremity of the last 4 years may be detected, too. The impact has been modelled as a regression relationship using mathematical statistical methods, its fine tuning may open new perspectives in understanding the relation between energy price and terms of trade changes.

Introduction, definition, history

Terms of trade impact experimental statistics deal with the quantification of the phenomenon, with revealing its possible causes and its impact assessment.

The change in terms of trade is basically a price shear, the relative development of export and import price changes. As for all price scissors, relative price changes may converge as quantitative changes in the aggregates when terms of trade change. In case a country’s export price index surpasses that of the import, a larger import may be realised for a unit of export, and vice versa, this is usually called the improvement or deterioration of the terms of trade.

„If we could sell something externally, other things being equal, for a higher price, the result is identical with producing and selling by the same input more externally of the given product. One could import more due to higher export income or repay debt (from this perspective is unimportant whether the volume or the price increase generated the result). The same is true for the terms of trade deterioration: producing more of a given product is of no use if less import could be paid for by its sales revenue; as the situation worsened, instead of improving. Naturally, it is important to know whether improvement or deterioration is due to changes in volume or prices (terms of trade), as we would like to have a better insight regarding the impact of our own work, as well as of the external factors. It has to be stressed that these two influences are the same nature in some regard, as such their joint review is well-founded” (Drechslrer—Szűcsné (1984) p. 118).

Above quote highlights why the idea of the presentation of the terms of trade impact on its own as well as within the framework of GDP volume change being worthwhile is widespread.

The simple terms of trade index (ratio of export and import goods price index) is part of the Hungarian Central Statistical Office (HCSO) publication activity, the HCSO, however, did not publish other indices up until now. The quantification of the phenomenon is set in a wider perspective in the experimental statistics on terms of trade impact, and (compared to previous publications) we use new analysis types, too. Analyses are based on yearly data; quarterly data analyses, regular data supply, an international outlook may be introduced on a later date. Analysed fields within the experimental statistics are:

First of all, it is worth taking a look at the history of the terms of trade impact quantification, and why did the interest for this topic gain popularity in the last years. Hungary laid stress on quantifying the terms of trade impact in the 1970s, due to the fact that following the 1973 oil crisis the export price index was even 20 percentage points below the import one. Several papers were published in the 1970s and 1980s about this topic, in regard to both the relative and absolute impact (Drechsler—Szűcsné [1979], Drechsler—Szűcsné [1984], Marton [1982]). In the years following the regime change basically terms of trade related analyses have not been published, in spite of the price hikes in import, although the term and its quantification became part of the standard curriculum (see theoretically Krugman—Obstfeld [2003], from empirical perspective e.g. Hunyadi—Vita [2002] or Hüttl—Vita [2005]).

Terms of trade related studies have a significant history. Empirical studies, used even today, (Nicholson [1960], Geary [1961]), have been written in the early sixties, followed by numerous methodological studies and calculations emerged, too. Although the methodology is nowadays part of the international regulation, quantifying the absolute impact is usually not compulsory (for instance in the ESA system) there are only related recommendations.

The relative disinterest of the 1990s, 2000s and the following interest-revival is not a methodological, but still an important aspect in regard to quantifying the changes in terms of trade. The process has been generated, starting from the 70s, by the gradual upturn of the external trade, its increasing role and mostly the crude material and oil price crisis. By the 90s alongside the classical exchanges between states the activity of multinational, international companies, their increasing outsourcing, overwrote previous rules of the game in the international economy. Due to knowledge-based economy, robotics and other trends the study of trade between states came once again into focus in the last years, and may further increase in the coming period (especially owing to certain degree of ratio relocation in production, for instance in the announced economic policy of Trump). Gábor Oblath from Hungary analysed in several papers the trend in the past years and described the methodology as well, e.g. Oblath [2019].

Methodology and calculations

1. Simple terms of trade index

The most commonly used index is the simple, net barter or commodity terms of trade one, the ratio of the relative price level of exported goods correlated with the price change of imported goods. It shows how much more or less import goods may be purchased by a country compared to one unit of export product, compared to the previous period. It is calculated by dividing the price index of the given country in a given period by the price index of imported goods. The formula is:

If the index value is above 1, meaning average export prices increase faster than import prices it can be said the country’s terms of trade improved. For a unit of export product more import goods may be purchased, or for one unit import less export has to be sold. For a more precise definition we should write:

Px and Pm stand for the export and import price index, Qx is the volume index of export, Vx is the export value index. The index value shows, in this case, the change of the export purchasing power from the base period to the reference one owing to the change in the import price level. Supposing Qx=1, meaning the export volume is unchanged, the result is the simple terms of trade index, with the interpretation that the index shows the change of the export purchasing power unit (unchanged quantity) owing to the different changes in export and import price levels. The simple terms of trade index is disclosed by the HCSO for many years now, under

Stadat → Prices→External trade prices→External trade in goods price indices and terms of trade indices.

Terms of trade impact should be evaluated not only from the trade in goods point of view, but from the external trade as a whole one, too. Figure 1 shows the simple terms of trade index value considering traditional trade in goods and total external trade, table 1 presents the value of the two terms of trade indices (in the case of the latter only from 2008).

Simple terms of trade index within external trade in goods and total external trade

(%)
Product range 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023
Products 98.3 101.8 99.9 98.4 98.7 100.6 101 100.8 101.5 99.6 99 100.6 102 96.4 93.2 106.4
Total external trade 99.1 100.8 100.0 98.6 99.0 100.5 100.6 100.7 100.7 99.6 99.1 100.3 102.1 96.9 94.8 106.4

The two kind of terms of trade indices are substantially moving together, as the majority of the external trade is still goods-linked, meaning their price changes are decisive, there are, however, certain exceptions, as the figure shows. It is obvious that the index calculated based on the total external trade is below the one calculated as per product range when terms of trade basically improve (2009, 2013-2016, 2019-2023) and surpasses it when these basically deteriorate.

2. The volume impact of the terms of trade change

The simple terms of trade index does not show the terms of trade change’s order of magnitude, as such this should be quantified; the methodology of it, however, is not uniform.

The System of National Accounts (SNA) is a uniform statistical framework, commissioned by the United Nations Statistical Commission, developed in cooperation by five international organisations – the UN, European Commission, OECD, IMF and the World Bank. The methodology highlights on several occasions that GDP volume growth is not identical with the increase in real income within the economy, the reason of the discrepancy being the changes in international terms of trade and the consequent trade gain or even loss1. The potential difference between real GDI and GDP volume may be significant if the value of export and import compared to GDP is high, furthermore if the content of goods and services in export and import is different. This may occur when the export of a country is predominantly made up of a few, primary goods (e.g. cocoa, sugar, oil) and imports mostly finished products, manufactured goods. It is advisable to consider trade gains and losses as integral part of the SNA, and it should always be calculated, even if there is no internationally uniform deflator (UNSC, 2009).

The internationally compatible regulatory framework of the European System of National Accounts (ESA 2010) contains the claim of quantifying trade gains and losses. Gross domestic product (GDP) may be considered both a production as well as an income index. As such the GDP volume index may be perceived as output volume change or change in the real value of income. From income point of view the unchanged price GDP shows the production volume and/or the real income of resident economic units. In terms of real income, a more precise meaning of this would be that it is the real income of residents for their participation in production, under the condition that it is used exclusively for the purchase of products and services produced in the given country. It should also be considered the volume of import purchased for a given export volume. Based on ESA 2010 real GDI may be obtained as the sum of so-called trading gain and the GDP volume. ESA 2010 recommends this formula for calculating the trading gain:

where X and M are export and import current prices, Px and Pm are export and import price indices.

P: price index, price gauge used for deflation.

The numerator of the first fraction is the actual current price based external trade balance, deflated with P price index. Selecting the corresponding P price index, according to ESA 2010, is the authority of the given country’s statistical office, by taking into consideration the individual circumstances of the country.

The measure of real GDI may be sensitive to the price index choice, hindering a consensus regarding the topic. Recommendations for choosing the price index may be categorised into three groups:

  1. Selecting Pm or Px depending if the external trade balance is positive or negative.
  2. Deflation with the mean of Pm and Px where the mean may be mathematical, harmonic or more complex, trade weighted mean.
  3. Deflation by other, general price index (e.g. consumer price index, GDP deflator price index).

In case of uncertainty the use of the mean of export and import price index is advisable (Geary method). The deflated value of the external trade balance, obtained in this way, is then reduced with the difference of the export and import values deflated by corresponding price indices, reaching the trade gain/loss originating from T terms of trade changes. The GDP volume may be modified with the value of T (Eurostat, 2013). The articles quoted in the first part of the publication analyse in detail the differences coming from the use of different deflator price indices, these are not reviewed here. Our calculations use, in line with the ESA recommendations, the Geary method as deflator2, the mean of the export and import price indices.

The quantitative effect of the terms of trade change has an impact on GDP as change calculated in income real value terms. This is why the calculation of the revised volume index is recommended. Let T be, as before, the value of the volume impact of the terms of trade change. The revised volume index in this case:

where GDP1 is the gross domestic product value of the reference period on unchanged, (base period) prices, the Hungarian routine calculates it by double deflation, GDP0 is the current price of the base period. It means that the revised index (not the unrevised one presented by the Hungarian statistical procedure) incorporates the quantitative impact of the terms of trade change on income.

At this point the income loss/gain coming from terms of trade changes should be defined. The presented methodologies give these values in million HUF, as the real value of income loss/gain resulting from the external terms of trade changes. As this value is generated within the external trade, it should be interpreted as equally less/ more income realised in export to be used in import – although this gain/loss is not necessarily used for import purposes. What proportion of this is realised in a given sector or at an economic operator cannot be determined through these means, so defining the measure of real GDP change from income perspective as a result of the terms of trade change is not possible with present apparatus. This is why the real change of the terms of trade is interpreted as decreasing/increasing factor of the real GDP value in income terms.

Calculation results, based on above, are presented henceforward. Figure 2 shows the volume impact (T value) magnitude of the terms of trade change compared to the conventionally presented GDP volume change (GDP1-GDP0). In most years the volume impact of the terms of trade change is not significant compared to the conventional change, however in recent years it became increasingly significant, in either positive (2020 and 2023) or negative (2021 and 2022) way.

Figure 3 shows, beside the GDP volume index (published by the HCSO, too) the value of the terms of trade change revised index . In this case, too, the difference in the case of the last four enlisted years is significant. The GDP volume lessened by 0.9% in 2023, but taking into account the effect of the changes in terms of trade, a 4.7% increase was experienced in terms of real income.. In contrast, part of the significant GDP volume growth of 2021 and the total of the 2022 one is annulled by the terms of trade deterioration.

3. The impact of energy price changes on the terms of trade change

There may be several reasons for terms of trade changes, perhaps the most significant factor for Hungary (due to its external trade structure) is the price change in the energy products’ import. This part is dealing with the connection between energy carriers’ import price index and the volume impact of the terms of trade change. Figures 4 and 5 show the link between the price index and the gain/loss, based on yearly data. Figure 4 presents the years between 2001 and 2023, showing a negative connection, the higher the energy carriers price index, the more significant the loss, and vice versa. It also displays the aberration of the last four years, in 2022, for instance, imported energy carriers’ prices grew on average by 161.5%, triggering the largest terms of trade deterioration of the last decades. Imported energy carriers’ prices almost halved on average in 2023, year-on-year, generating a huge terms of trade improvement. Figure 5 shows, by excluding the last four years, the link between the two phenomena, highlighting the fact that this negative, relatively strong comovement is true even without extremities.

The comovement is supported by a regression relation, too, using a linear regression, analysing how energy carriers’ price changes (the independent variable X being in this case) explain the volume impact of terms of trade change (dependent Y variable T)3 . The estimated model is Y=3172820-24119∙X, meaning a one percentage point higher price index decreases on average the real income by 24.1 billion HUF. Figure 6 displays the value of T and the value estimated through the energy carriers’ price changes by regression. Based on the trend depicted in the figure it is noticeable that the energy price index explains indeed quite well the measure of gain/loss, and offers the possibility of further analysis also, regarding when is there a greater discrepancy between the estimated and actual T, namely which are the years when, beside the energy carriers’ price changes, other important factors may be present, too. Such years are primarily 2007-2009 and 2021. Import energy prices more than doubled in 2021; based on the general trend a much larger deterioration may be estimated than the one that actually occurred.

Bibliography

Footnotes

  1. This fact is significant in national accounts calculations, too, as certain real income aggregates (GNI, GNDI, NNDI) originate from the real GDI, which is the sum of GDP volume and trade gain (or loss) (UNSC, 2009). ↩

  2. Hungarian economic statistics curricula usually deal with the Nicholson formula (deflator price index is the export price index), easily calculated and interpreted, according to our calculations, however, there is hardly any difference in case of the Nicholson and Geary formula when considering Hungarian data, as such we only present the latter. ↩

  3. As it is about time series, the possibility of residual autocorrelation may occur, distorting parameter estimations. This is why the Prais-Winsten algorithm has been used, as this filters the possible residual autocorrelations through an interactive error-correcting method. The final model is significant, with a 78% explanatory power. The value of the Durbin-Watson statistics is 1.8, also of an adequate value. ↩