3 Terms of trade, real gross domestic income and real gross domestic product

View the most recent version.

3.1 Data and presentation

The data for gross domestic product (GDP), final domestic demand, exports, and imports come from Statistics Canada's Provincial Economic Accounts. Commodity price data, provincial industry and export data, and exchange rate information come from Statistics Canada's CANSIM database. To illustrate the impact of relative price changes on real income, two sets of analysis are presented. The first part of the analysis focuses on Canada as a whole from 1981 to 2007. However, it is primarily concerned with examining the response to the relative price changes after 2002. The historical record is employed, where necessary, as a stylized example for comparison. Because examining different points across business cycles can influence results, an effort has been made to examine similar phases of Canada's business cycles. Where necessary, the 1981-to-1982 and the 1991 recessions are omitted when average growth rates are calculated.

The second part examines the response to the post-2002 relative price shifts. However, rather than examine macroeconomic variables affecting all of Canada, it focuses on the varying regional impacts of the relative price change, and how they have changed over time. Ultimately, as the analysis shows, understanding how the provinces respond is key to understanding how Canada responds.

3.2 Canadian real gross domestic product compared with real gross domestic income

Canada is a resource-rich nation that engages extensively in trade, predominantly with the United States. Although there is a substantial manufacturing industry, located principally in the provinces of Quebec and Ontario, resources remain a mainstay of many areas of the economy. Consequently, Canada's trading gain is intimately linked to changes in commodity prices.

Figure 1 illustrates this relationship by plotting the Bank of Canada's commodity price index against an index of Canada's trading gain. In each case, the indices are set to equal 100 in 2002. Over the entire 1981-to-2007 period, changes in commodity prices have been mirrored by changes in Canada's trading gain. Prior to 2003, the trading gain oscillated within a stable band that followed commodity price cycles. After 2002, commodity prices have risen to record levels, driving up Canada's trading gain.

Figure 1
Changes in commodity prices and trading gain, 1981 to 2007

Additional factors have also affected the trading gain. First, the Canadian dollar has appreciated by nearly 50% relative the U.S. dollar, lowering import costs, especially for manufactured goods. ³ Second, the emergence of Asia, primarily China, as a manufacturing super-power has driven down the prices of many consumer products on a global scale (Francis 2007).

As a result, unlike previous periods when trading gains rose, the post-2002 trading gain is the result not just of resource-related export price increases, but also of lower import prices. In fact, import prices declined on average by 2.6% per year from 2003 to 2007 while export prices rose by about 1% per year.

Consequently, the difference between real gross domestic product (GDP) growth and real gross domestic income (GDI) growth is larger after 2002 than in almost every previous year. Figure 2 shows the contribution to real GDI growth that comes from real GDP growth and from the trading gain for the 1982-to-2007 period. In most years, the contribution to real GDI growth that came from relative price changes was modest, and it followed commodity price cycles. After 2002, however, the combination of falling import prices and rising export prices led to large, ongoing trading gains.

Because Canada has a modest trade surplus—it typically makes up less than 5% of GDP—the Salter ratio effect is muted in Canada (Table 1). The terms-of-trade effect, however, is significant. The average share of exports and imports in nominal GDP ranged between a low of 24% in the 1980s and a high of 41% in the late 1990s. Because Canada engages heavily in trade but only runs a modest surplus, it is more susceptible to terms-of-trade shifts than Salter-ratio changes.

Figure 2
Real gross domestic income growth, by source

Table 1
Real gross domestic income growth decomposed

A country can capitalize on its terms-of-trade improvement in different ways: by transforming its stream of exports into a larger stream ofimports, lowering its exports or raising its savings. If the first option occurs, import growth should outpace export growth, which is in fact what has occurred in Canada (Figure 3). From 2003 to 2007, real imports have expanded at an average annual rate of 6.1% while exports have risen by an average of 1.3%. Higher consumption and investment levels have resulted as the domestic economy has absorbed the increased stream of imports. From 2003 to 2007, consumption growth has risen at an annual average rate of 3.8%, while investment has risen at an average rate of 6.7%. Both consumption and investment outpaced real GDP's average annual growth of 2.7%, and are more in line with the 3.9% annual average growth in real GDI. ⁴

Figure 3
Real imports and real exports, 1997 to 2007

Analysis of Canada as a whole points to two main features. First, the relative price changes affecting the Canadian economy after 2002, while drawn from similar sources as previous shifts, have a more pronounced effect than they had previously. Second, price changes have a different composition; in particular, while resource export prices have risen, import prices have also fallen.

3.3 Provincial real gross domestic product compared with real gross domestic income

While the contribution to Canada's real GDI growth from its trading gain is historically large after 2002, the impact of relative price changes have also been significant for many of the provinces. Because the Canadian provinces have broadly differing economic structures, relative prices changes have led to significant differences in economic performance. To a large extent, the differences across regions stem from the type of commodity whose price changes.

During the 1980s and 1990s, it was unusual for a broad range of commodity prices to increase or decrease in tandem. In fact, energy and industrial material commodity price movements had a negative correlation of -0.62 from 1981 to 1989 (Table 2). Food and energy commodities exhibited a moderate positive correlation, while food and industrial materials were weakly correlated. During the 1990s, the correlations across commodities tended to remain weak, except for food and industrial commodity prices. After 2002, however, there has been a strong positive correlation between commodities price movements.

Because commodity prices tend not to move in tandem, changes in commodity prices have historically led to a redistribution of purchasing power across provinces. The post-2002 commodity-price movements have similarly redistributed purchasing power, but in an importantly different manner. Prior to 2002, a rise (fall) in energy-producing provinces' purchasing power was accompanied by a decrease (rise) in the purchasing power of energy- importing provinces; after 2002, a redistribution has occurred because certain provinces have seen larger gains than others. Only in one instance do the relative price changes detract from real income growth after 2002. The difference comes from how provincial economies have dealt with increased commodity costs and falling world manufactured goods prices.

Table 2
Correlations in commodity price movements

Prior to examining real GDP, the trading gain and real GDI, a discussion of the structure of the provincial economies will serve to illustrate why the differing composition of price changes in the 1980s, 1990s and 2000s is important. A breakdown of each province's economic structure is provided in Table 3. For each province, the average share of nominal business sector GDP for each industry is shown.

GDP in the most eastern province, Newfoundland and Labrador, is dominated by mining, oil and gas extraction, which is a relatively new phenomenon. Traditionally, the Newfoundland and Labrador economy had been dominated by fisheries and the service sector. However, following the collapse of the cod stocks in the late 1990s, and the initiation of offshore oil and gas extraction, the make up of the province's economy has changed dramatically.

The economies of Prince Edward Island, Nova Scotia and New Brunswick are broadly similar. Agriculture and fisheries activities play an important role in GDP, particularly in Prince Edward Island which produces a significant number of potatoes and mussels. Mining, oil and gas extraction is relatively unimportant, except in Nova Scotia, where there are offshore energy deposits. In each of these provinces, manufacturing and finance, insurance, real estate and leasing (FIREL) activities are important sources of value added.

The central economies of Quebec and Ontario form the manufacturing heartland of the Canadian economy. They are also the largest provincial economies. While manufacturing has the largest GDP share, FIREL, retail, wholesale and professional, scientific and technical services industries also make significant contributions to these economies. FIREL is particularly large in Ontario, which is the centre of the Canadian financial industry.

Table 3
Nominal value added shares, 2003-to-2004 average

Although their physical geography is similar, the Prairie provinces of Manitoba, Saskatchewan and Alberta have notably different economic structures. Although Manitoba's agricultural value added share is larger than Quebec's or Ontario's, in most other respects it is quite similar to the central Canadian economies. Saskatchewan, while more dependent on agriculture, as well as Alberta, generates a significant portion of value added from mining, oil and gas extraction. These provinces, especially Alberta, are the major oil producers in Canada. British Columbia's major source of value added comes from manufacturing, while retail trade, transportation and FIREL also have important shares.

Although the value added shares illustrate the different regional economic structures, they do not sufficiently illustrate the degree of difference across the regions on their own. The mix of commodities produced, particularly in the goods sector, has important implications for how the provinces respond to terms-of-trade and Salter-ratio shocks. Table 4 shows the export share for broadly grouped product types from each province to destinations outside of Canada.

An examination of the export shares indicates that the provinces are less diversified than the value added shares suggest. In almost every province, the proportion of exports accounted for by the top two product groupings exceeds 60%. In some cases, such as Newfoundland and Labrador and Alberta, more than half of all exports by value are related to a single type of product. In British Columbia and Ontario, forestry and automotive exports, respectively, make up over 40% of exports. Although each of these two provinces has a manufacturing industry, the export shares reinforce the notion that manufacturing in Ontario is dominated by automotive products, while manufacturing in British Columbia is dominated by wood products.

Table 4
Nominal export shares by product type, 2003-to-2007 average

The concentration of exports in particular product areas means that relative price changes, either through the terms of trade or the Salter ratio, will have differential impacts across provinces. As a result, real GDP and real GDI changes take on very different magnitudes between 2003 and 2007. Figure 4 plots the average annual growth rate of real GDP and real GDI for Canada and for each of the provinces from 2003 to 2007. While including the trading gain does lead to some minor changes in the rankings of which provinces have grown the fastest (mostly due to changes in the ranking of oil-producing provinces), the trading-gain-adjusted measure illustrates how much stronger the growth in the purchasing power ofincome has been relative to the volume of income produced across most provinces.

Figure 4
Real gross domestic product compared with real gross domestic income average annual growth rates, Canada and provinces

During the 2003-to-2007 period, one of the more striking changes is the decline in import prices. The decline in import prices, induced by the appreciating Canadian-U.S. exchange rate and the integration of Asian manufacturing into the world economy, have contributed to Canada's trading gain. However, each of the provinces has benefited to a different degree.

Although Canada has seen an increase in its terms of trade due to falling import prices and rising export prices, many provinces have, in fact, seen import and export prices move in the same direction (Figure 5). In Newfoundland and Labrador, Nova Scotia, New Brunswick and Saskatchewan, both import and export price indices have risen. However, the rise in export prices is larger than the rise in import prices, leading to terms-of-trade improvements in each case. In Ontario, import and export prices declined as the currency appreciated; however, the decline in import prices is more pronounced leading to small terms-of-trade improvements. In Quebec, Manitoba, Alberta and British Columbia, import prices fell while export prices rose, a situation that potentially generates large terms-of-trade improvements. Prince Edward Island is the only province to have experienced a terms-of-trade deterioration.

While the terms of trade tend to contribute to real income growth, the impact of the Salter ratio is less consistent across provinces. Inter-provincial transfers, which take many forms including governmental transfers, redistribute income from richer to poorer regions. These transfers of funds affect the level of aggregate expenditures in both the sending and receiving provinces. Since the impact of the Salter ratio is weighted by net exports as a share of nominal GDP, inter- provincial transfers can make the Salter ratio effect more pronounced at the provincial level. ⁵

Figure 5
Import and export price index average annual growth, 2003-to-2007 average

The contribution to the trading gain from the terms of trade and the Salter ratio for each of the provinces, and for Canada, is shown in Table 5. The first two columns show the average annual contributions during the 2003-to-2007 period. The last four columns show the average annual contributions for each relative price ratio during the 1993-to-2002 and the 1983-to-1990 periods, respectively. The contributions during the years 2003 to 2007 help to explain the size of the trading gain that has been realized, while the contributions during the earlier periods illustrate why the Canadian trading gain was modest during the 1980s and 1990s and pronounced after 2002.

Table 5
Contribution to real gross domestic income growth from Salter-ratio and terms-of-trade changes across periods

During the 2003-to-2007 period, terms-of-trade changes have been the dominant factor affecting provincial trading gains, especially for the energy-producing provinces of Newfoundland and Labrador, Saskatchewan and Alberta. Energy-importing provinces have also seen terms-of-trade improvements contribute positively to their trading gains. The Salter-ratio contribution, on the other hand, has been negligible in many cases. Nevertheless, there are particular provinces where the contribution from the Salter ratio is important. In Ontario from 2003 to 2007, the Salter ratio and terms of trade have had, on average, similar, offsetting effects. Similar offsetting effects have occurred in Prince Edward Island and New Brunswick, although the magnitude of the terms-of-trade effect is slightly larger than the Salter ratio effect in these cases.

Strikingly, during the 2003-to-2007 years, provincial trading gains have been mostly positive. In every province, except for Prince Edward Island, the trading gain has contributed to real income growth. In previous periods, the trading gain was not as widespread. During the 1983-to-1990 period, when energy prices were falling, trading gains tended to contribute to real income growth in the energy-importing provinces such as Quebec, Ontario, Nova Scotia and New Brunswick and they detracted from real income growth in the energy-exporting provinces of Alberta and Saskatchewan (Newfoundland and Labrador did not develop its offshore oil and gas industry until the 1990s). A similar pattern emerged in the 1990s, when trading gains detracted from real income growth in energy-importing provinces and contributed to real income growth in energy- exporting provinces.

The dichotomy between the 1980s, 1990s and post-2002 years is well illustrated by Alberta, Quebec and Newfoundland and Labrador (Figures 6 to 8). Alberta's real GDI rose and fell much more rapidly than its real GDP, as changes in energy prices affected the purchasing power ofits income. The effect was consistent across decades—fluctuations in energy prices always drive trading gains.

Figure 6
Real gross domestic product compared with real gross domestic income in Alberta

Figure 7
Real gross domestic product compared with real gross domestic income in Quebec

In Quebec, trading gains were positive in the early 1980s and after 2002. In the earlier period, the positive effect had come from falling energy prices, while in the later years the positive effect came in spite of rising energy prices. Movements in other commodities, which were not well correlated in the 1980s, and falling prices for other imports overcame the negative effects of rising energy prices on Quebec's purchasing power after 2002.

In Newfoundland and Labrador, there was little effect from relative price changes during the 1980s. During the 1990s, the province expanded its resource base when offshore oil and gas began being marketed. When prices for energy began rising after 2002, real GDI in Newfoundland and Labrador accelerated sharply.

Figure 8
Real gross domestic product compared with real gross domestic income in Newfoundland and Labrador

The two major differences between the post-2002 years and the 1980s and 1990s—the widespread commodity price rises and the integration of Asian manufacturers into the global economy—have allowed almost all provinces to benefit to some degree. Because the source and composition of the relative price changes were substantially different from previous periods, the response of the provinces has been more consistent across the country and, in many cases, stronger. The offsetting nature of energy price shocks from the past has been overcome in the post-2002 years.

³ . See Amano and van Norden (1993) or Bailliu and King (2005) for a discussion of the determinants of the Canadian-U.S. exchange rate.

⁴ . Canada's investment in machinery and equipment is heavily dependent on imported goods. As a result, investment is more elastic than consumption to changes in import prices. From 2003 to 2007, import prices declined at an average annual rate of 2.6%, contributing to investment growth.

5 . The System of National Accounts does not produce data on these inter-provincial income flows; accordingly, current account balances at the provincial level are not available. As a result, the analysis cannot move from producing estimates of real gross domestic income to real gross national income.