PRODUCT VARIETIES, EXTENSIVE AND INTENSIVE MARGINS OF TRADE: EVIDENCE FROM CANADA ISLAM, Sadequl *

PRODUCT VARIETIES, EXTENSIVE AND INTENSIVE MARGINS OF TRADE: EVIDENCE FROM CANADA ISLAM, Sadequl * Abstract Using disaggregated data at the HS ten-digit level, this paper computes Canada s intensive and extensive margins in the U.S. market with reference to the Rest of the World, the European Union, China, and Japan. The empirical findings for the 1992-2012 period show that Canada s intensive margin declined with reference to the Rest of the World, the European Union, and China. The evidence also demonstrates that percentage changes in the intensive margin dominate percentage changes in Canada s relative market share in the United States, especially during the Great Recession. The paper also examines export varieties from Canada and selected countries. JEL Code: F1 Keywords: Intensive margin, extensive margin, export varieties 1. Introduction Canadian trade with the United States is one of the largest bilateral trade relationships in the world, induced by a high degree of economic integration under the NAFTA ( among others, Castillo Ponce and Ramirez Acosta, 2008). Two-way merchandise trade between the two countries increased from $ 189.2 billion in 1992 to over $616.4 billion in 2012. Trade expansions can be associated with a) increased varieties of tradable goods, b) increased volumes of traditional goods,c) increased quality of traded goods, and d) increased efficiency of exporting firms. Traditional trade theories ( the Ricardian model and the Hecksher-Ohlin model) emphasized the role of productivity and factor endowments in determining a country s comparative advantage and highlighted the gains from specialization in exports. In the 1980s, several economists ( among others, Krugman, 1979, Helpman and Krugman, 1985) developed a New Trade Theory which incorporated consumers preference for product varieties and increasing returns to scale. This new theory explains how increasing returns to scale and product differentiation lead to intra-industry trade. More recently, research done by some economists, notably, Melitz (2003) and Bernard, Eaton, Jensen, and Kortum (2003) has led to a New New Trade Theory which highlights the role of heterogeneous firms in international trade. According to this theory globalization and trade liberalization lead to shifts of production from inefficient firms to efficient firms within an industry. There is now a large empirical literature on trade flows that directly or indirectly explores the relevance of the above theories. One strand of the empirical literature has investigated the effects of trade liberalization on firm productivity and product diversification ( among others, Baldwin and Gu, 2004, 2003). Another strand of the empirical literature has concentrated on assessing the extensive and intensive margins of trade ( Bernard, Jensen, Redding, and Schott, 2009; Hillberry and McDaniel, 2003; Hummels and Klenow, 2005). Growth of trade along the intensive margin can be defined as an increase in the value of * Sadequl Islam, Department of Economics, Laurentian University, Sudbury, Ontario, Canada P3E 2C6. Email: sislam@laurentian.ca, sadequlislam@sympatico.ca

trade involving existing trade relationships. The intensive margin can be further decomposed into two components: a) changes in volumes and b) changes in unit prices. The extensive margin can be defined as growth of trade attributable to an increase in the number of product varieties. The relative contributions of the extensive margin and the two sub-components of the intensive margin have implications for trade theories and policies. Analysis of extensive and intensive margins of trade enhances our understanding of trade patterns and the relative efficiency with which economies allocate resources. The main objectives of this paper are the following: 1) To examine the relative contribution of extensive and intensive margins of Canada s trade. 2) To evaluate the role of new product varieties in the extensive margins of Canada s exports. 3) To explore the impact of the Great Recession of 2008-09 on extensive and intensive margins of Canada s exports. The paper contributes to the empirical literature in three ways: first, it measures the extensive and intensive margin of Canada s trade with the United States using the latest trade data at highly disaggregated level (the ten-digit level of the Harmonized System ). Second, it explores the role of new product varieties in Canada s exports to the United States. Third, it investigates the effects of the Great Recession on Canada s margins of exports. 2. Methodology and Data For availability of data, the paper follows the methodology of Hummels and Klenow ( 2005) in computing the extensive margin and the intensive margin. Following Hummels and Klenow (2005), the intensive margin of Canada with respect to, for example, the Rest of the World, in the U.S. market is defined as: Imrw = " i! Icu " i! Icu p p cui x x cui (1) where imrw = intensive margin of Canada with respect to the Rest of the World in the U.S. market. p = the price of the relevant product, x= quantity of export, c refers to Canada while r refers to the Rest of the World, and u is the basket of goods exported by Canada to the U.S.A. The intensive margin of Canada, accordingly, is Canada s value of exports relative to the value of exports from the Rest of the World in those categories in which Canada exports to the United States (I cu ). The extensive margin, Emrw, is defined as: Emrw= " i! Icu " i! I p p x x (2) I cu is the set of observable categories in which Canada has positive exports to the United States, that is, X cui > 0. 98

Islam, S. Product Varieties, Extensive And Intensive Margins Of Trade: Evidence From Canada The Rest of the World has positive exports to the United States in I categories. Accordingly, Emrw equals Canada s exports to the United States in I cu relative to the exports from the Rest of the World. The extensive margin ranges from 0 to 1. The value of the extensive margin can be interpreted as how much of the exports by a country, for example, Canada to the United States can be explained by export varieties. It can also be interpreted as a weighted count of Canada s export basket relative to the export basket of the Rest of the World. The ratio of Canada s exports to that of the Rest of the World to the United States equals the product of the intensive margin and extensive margin, that is: Csrw = Imrw. Emrw = " i! Icu " i! I p p cui x x cui The paper concentrates on Canada s trade with the United States for the 1992-2012 period. It uses the trade database of the United States Department of Commerce. This database provides annual trade data involving Canada, other countries, and the United States. The paper also uses the database of profiles of Canada s exporters provided by Statistics Canada. The paper compares export varieties from Canada and selected countries including China. 3. Basic Statistics Before analyzing the intensive and extensive margins of Canada with respect to the Rest of the World and selected countries, it is useful to present some basic statistics. Table 1,Panel A, reports top ten export categories (at the two-digit HS code level) from Canada to the United States in 1992. Several points can be highlighted. First, the dominant product group is vehicles (HS 87) which accounted for 28.6% of Canada s exports to the United States and 36.8% of total imports of vehicles into the United States. Second, the next dominant product group is mineral fuel (HS 27) which accounts for 11 % of Canada s export basket and 19.5% of US import of mineral fuel. The third dominant product group in Canada s export basket to the United States is nuclear reactors and parts. Third, some product groups had very high shares in the US import market: wood pulp (84.2%), wood and wood articles ( 70.3%), and aluminum and aluminum articles (61.8%). However, in 1992, these product groups had low weights in Canada s export basket. Finally, electric machinery (HS 85) which can be considered a dynamic product group, accounted for only 5.1% of Canada s exports and 7.4% of US imports of the same product group. Panel B in Table 1 shows top ten export products ( at the two-digit level) from Canada to the United States in 2012. Several observations can be made. First, mineral fuel (HS 27) became the dominant product group in 2012 with higher shares in Canada s export basket and the US import basket, compared to 1992. Second, several product groups, vehicles, nuclear reactors, electrical machinery, aluminum, paper, and wood have lost market shares in the United States. Finally, plastics and plastic articles (HS 39) accounted for higher shares in Canada s export basket and US import basket in 2012, compared to 1992. 99 (3)

Table 1. Top Ten Export Products ( two-digit level) from Canada to the U.S.A Panel A,1992 All Commodities 98629968 SCX MSU 87 Vehicles, Except Railway Or Tramway, And Parts Etc 28245787 28.6 36.8 27 Mineral Fuel, Oil Etc.; Bitumin Subst; Mineral Wax 10802651 11.0 19.5 84 Nuclear Reactors, Boilers, Machinery Etc.; Parts 7689620 7.8 10.1 48 Paper & Paperboard & Articles (inc Papr Pulp Artl) 5983102 6.1 74.3 85 Electric Machinery Etc; Sound Equip; Tv Equip; Pts 5070919 5.1 7.4 44 Wood And Articles Of Wood; Wood Charcoal 4478409 4.5 70.3 98 Special Classification Provisions, Nesoi 4063653 4.1 30.8 76 Aluminum And Articles Thereof 2121873 2.2 61.8 39 Plastics And Articles Thereof 1863985 1.9 23.2 47 Wood Pulp Etc; Recovd (waste & Scrap) ppr & pprbd 1793309 1.8 84.2 Panel B, 2012 All Commodities 323,936,455 SCX MSU 27 Mineral Fuel, Oil Etc.; Bitumin Subst; Mineral Wax 103,371,998 31.9 24.4 87 Vehicles, Except Railway Or Tramway, And Parts Etc 57,581,891 17.8 24.0 84 Nuclear Reactors, Boilers, Machinery Etc.; Parts 20,648,675 6.4 6.7 39 Plastics And Articles Thereof 10,215,080 3.2 24.3 98 Special Classification Provisions, Nesoi 9,780,728 3.0 18.4 85 Electric Machinery Etc; Sound Equip; Tv Equip; Pts 8,255,090 2.5 2.8 76 Aluminum And Articles Thereof 6,831,513 2.1 43.9 48 Paper & Paperboard & Articles (inc Papr Pulp Artl) 6,719,039 2.1 44.5 71 Nat Etc Pearls, Prec Etc Stones, Pr Met Etc; Coin 6,522,698 2.0 10.1 44 Wood And Articles Of Wood; Wood Charcoal 6,318,393 2.0 49.0 Notes: 1. Figures are in thousands of US dollars. 2. SCX = Share (%) of the relevant commodity in Canada s total exports to the USA. 3. MSU= Market share (%) of Canada in total US imports for the relevant commodity. 4. Source of data: US Census Bureau, Department of Commerce. Table 2 presents data on Canada s exporting establishments and value of exports. It is evident from Table 2 that the number of exporting establishments in the manufacturing sector (NXM) increased from 17,689 in 1996 to 21,999 in 2006; however since 2006, the figure declined. The exporting establishments in the manufacturing sector as a percentage of total establishments shows a downward trend from 45.2% in 1996 to 41% in 2009. It is also evident from Table 2 that Canada s exports of manufactured products (VXM) fluctuated with a rising trend during 1996-2006; however, since 2006, the figure shows absolute declines. Finally, it can be observed from Table 2 that the value of exports of manufactured products as a percentage of total exports shows a downward trend from 62.9% in 1996 to 52.3% in 2009. Table 3 shows the number of establishments by export destination. The number of establishments exporting to only the United States (NXUS) fluctuated with a declining trend during 2006-2009. The number of these establishments as a percentage of the total number of establishments (PNXUS) also shows a downward trend. The number of establishments exporting to the United States as well as other countries (NXUSO) appears to show a rising trend. Table 3 also shows that the number of establishments exporting to only other countries (NXO) increased from 6,020 in 1996 to 9,845 in 2009. 100

Islam, S. Product Varieties, Extensive And Intensive Margins Of Trade: Evidence From Canada The number of these firms as a % of total establishments also increased from 15.4% in 1996 to 20.7% in 2009. Table 2. Number of Canada s Exporting Establishments and Value of Exports Year NXM VXM TNX TVX PNXM PVXM 1996 17,689 161,631 39,154 257,159 45.2 62.9 1997 18,585 176,948 41,197 278,960 45.1 63.4 1998 19,128 196,870 42,553 293,885 45.0 67.0 1999 19,761 223,647 44,543 325,766 44.4 68.7 2000 20,408 248,641 46,391 378,357 44.0 65.7 2001 20,700 233,097 47,668 367,239 43.4 63.5 2002 21,309 241,029 49,483 359,148 43.1 67.1 2003 21,593 216,552 50,643 348,129 42.6 62.2 2004 21,739 234,290 52,910 378,497 41.1 61.9 2005 21,981 234,231 53,559 401,508 41.0 58.3 2006 21,999 236,892 53,489 404,344 41.1 58.6 2007 21,514 235,566 52,540 413,683 40.9 56.9 2008 20,970 225,027 51,648 448,815 40.6 50.1 2009 19,529 171,997 47,637 329,120 41.0 52.3 Notes: 1. NXM = Number of exporting establishments in the manufacturing sector. 2. VXM = Value of exports in millions of C $ from the manufacturing sector. 3. 3. TNX = Total number of exporting establishments. 4. TVX = Total value of exports. 5. PNXM = NXM as a % of TNX. 6. PVXM = VXM as a % of TVX. 7. Source of Data: Statistics Canada, A profile of Canadian Exporters, 1996-2009. Table 3. Number of Canada s Exporting Establishments by Export Destination. Year NXUS PNXUS NXUSO PNXUSO NXO PNXO TNX 1996 24,366 62.2 8,768 22.4 6,020 15.4 39,154 1997 27,487 66.7 7,921 19.2 5,789 14.1 41,197 1998 29,181 68.6 7,683 18.1 5,689 13.4 42,553 1999 31,374 70.4 7,680 17.2 5,489 12.3 44,543 2000 33,498 72.2 7,550 16.3 5,343 11.5 46,391 2001 34,081 71.5 7,980 16.7 5,607 11.8 47,668 2002 34,472 69.7 8,682 17.5 6,329 12.8 49,483 2003 32,452 64.1 10,361 20.5 7,830 15.5 50,643 2004 32,047 60.6 11,298 21.4 9,565 18.1 52,910 2005 31,851 59.5 11,775 22.0 9,933 18.5 53,559 2006 32,046 59.9 11,679 21.8 9,764 18.3 53,489 2007 30,806 58.6 11,676 22.2 10,058 19.1 52,540 2008 29,507 57.1 11,714 22.7 10,427 20.2 51,648 2009 26,950 56.6 10,842 22.8 9,845 20.7 47,637 Notes: 1. NXUS = Number of exporting establishments exporting to only United States. 2. TNX = total number of exporting establishments. 3. NXUSO= number of establishments exporting to the United States and other countries. 4. NXO= number of establishments exporting to other countries. 5. PNXUS = NXUS as a % of TNX. 6. PNXUSO = NXUSO as a % of TNX. 7. PNXO = NXO as a % of TNX. 8. Source: Statistics Canada, A profile of Canadian Exporters, 1996-2009. 101

Table 4. Value of Canada s Exports by Export Destination Year VXUS PVXUS VXUSO PVXUSO VXO PVXO TVX 1996 111,870 43.5 129,339 50.3 15950.0 6.2 257,159 1997 133,260 47.8 124,667 44.7 21034.0 7.5 278,960 1998 131,452 44.7 144,670 49.2 17763.0 6.0 293,885 1999 149,323 45.8 161,276 49.5 15167.0 4.7 325,766 2000 172,816 45.7 190,916 50.5 14625.0 3.9 378,357 2001 187,418 51.0 165,290 45.0 14531.0 4.0 367,239 2002 163,186 45.4 181,681 50.6 14282.0 4.0 359,148 2003 162,028 46.5 172,617 49.6 13483.0 3.9 348,129 2004 176,886 46.7 183,886 48.6 17726.0 4.7 378,497 2005 164,065 40.9 216,302 53.9 21142.0 5.3 401,508 2006 187,306 46.3 188,869 46.7 28169.0 7.0 404,344 2007 188,555 45.6 195,494 47.3 29633.0 7.2 413,683 2008 220,437 49.1 184,045 41.0 44333.0 9.9 448,815 2009 156,444 47.5 134,586 40.9 38090.0 11.6 329,120 Notes: 1. VXUS = Value of exports (Millions of C$) by establishments which export only to the U.S.A. 2. VXUSO = Value of exports by establishments which export both to the U.S.A. and other countries. 3. VXO = Value of exports by establishments which export only to other countries.4. PVXUS = VXUS as a % of total value of exports (TVX). 5. PVXUSO = VXUSO as a % of TVX. 6. PVXO = VXO as a % of TVX. 7. Source: Statistics Canada, A profile of Canadian Exporters, 1996-2009. Table 5. Number of Export Products (HS Ten-digit level) from Canada and Selected Countries to the U.S. A. Year Canada UK France Germany Italy 1992 10,174 9,121 8,168 9,382 8,158 1993 10,389 9,374 8,418 9,470 8,428 1994 10,954 9,816 8,937 9,959 8,991 1995 11,370 9,995 9,119 10,087 9,217 1996 11,600 10,352 9,378 10,312 9,585 1997 12,142 10,867 9,852 10,807 10,123 1998 11,854 10,665 9,685 10,557 9,911 1999 11,921 10,444 9,699 10,545 9,890 2000 12,007 10,543 9,861 10,725 10,218 2001 11,827 10,318 9,820 10,619 10,051 2002 12,101 10,379 9,780 10,835 10,375 2003 11,995 10,285 9,801 10,773 10,306 2004 11,914 10,353 9,775 10,832 10,369 2005 11,919 10,169 9,701 10,737 10,387 2006 11,874 10,139 9,735 10,809 10,546 2007 11,761 9,862 9,555 10,593 10,524 2008 11,505 9,666 9,273 10,424 10,261 2009 11,093 9,246 8,943 10,067 9,820 2010 11,105 9,475 9,005 10,217 10,101 2011 11,160 9,682 9,192 10,454 10,380 2012 11,206 9,787 9,154 10,449 10,549 Source: Census Bureau, US Department of Commerce. 102

Islam, S. Product Varieties, Extensive And Intensive Margins Of Trade: Evidence From Canada Table 6. Number of Export Products (HS Ten-digit level) from Selected Countries to the U.S. A. Year Japan Brazil India China Russia Mexico 1992 9,089 3,773 3,446 6,602 794 5,848 1993 9,091 3,860 3,872 7,002 1,542 6,098 1994 9,438 4,145 4,296 7,591 1,805 6,752 1995 9,433 3,954 4,603 7,918 1,853 7,672 1996 9,617 3,967 4,873 8,283 1,763 8,339 1997 10,009 3,989 5,425 9,043 1,924 8,816 1998 9,765 3,923 5,610 9,254 1,956 8,562 1999 9,765 4,199 5,740 9,671 2,087 8,646 2000 9,752 4,700 6,163 10,206 2,282 8,700 2001 9,535 4,752 6,346 10,310 2,242 8,563 2002 9,640 5,221 6,734 11,068 2,297 8,673 2003 9,671 5,602 6,997 11,472 2,355 8,616 2004 9,695 5,856 7,306 11,995 2,239 8,721 2005 9,839 5,990 7,827 12,702 2,125 8,802 2006 9,938 6,055 8,177 13,114 2,085 8,817 2007 9,791 5,840 8,286 13,161 1,897 8,733 2008 9,868 5,577 8,302 13,090 1,764 8,649 2009 9,400 5,110 8,017 13,053 1,658 8,558 2010 9,518 5,129 8,301 13,236 1,673 8,645 2011 9,667 5,119 8,577 13,467 1,738 8,864 2012 9,727 5,180 8,770 13,614 1,745 8,929 Source: Census Bureau, US Department of Commerce. Table 4 presents data on value of exports by export destination. It is evident that value of exports by establishments which export only to the United States (VXUS) shows a rising trend until 2008.The value of exports by these establishments as a percentage of total exports shows a fluctuating pattern with 43.5% in 1996 and 47.55 in 2009. The value of exports by establishments serving the United States and other countries as a % of total exports (PVXUSO) appears to show a declining trend. Finally, the value of exports by establishments serving other countries as a % of total exports (PVXO) seems to show an increasing trend, with 6.2% in 1996 and 11.6% in 2009. Table 5 and Table 6 reports the number of export products at the HS ten-digit level from Canada and selected countries to the United States. Several points from Table 5 and Table 6 can be highlighted. First, the number of export products from Canada fluctuated during the 1992-2012 period; the figure stood at 11,206, lower than the peak year of 1997, when it was 12,142.The fall in export varieties during 2007-2009 can be attributed to the recession in the United States. It appears that the number of export varieties has increased in recent years (2010-2012) following the US recovery. Second, given Canada s geographical proximity, it is not surprising that Canada exports more varieties of goods compared to U.K., France, Germany, and Italy. The number of export varieties from these four countries also decreased during the US recession ( 2008-2009) and increased during 2010-2012. Third, it is evident from Table 6 that among the selected countries, China 103

exports the highest number of export products to the United States. The number of export varieties more than doubled from 6,602 in 1992 to 13,614 in 2012. Fourth, the number of export varieties also shows a significant upward trend for Brazil, India, Russia, and Mexico; however, Russia still export the fewest varieties of products to the United States Finally, the number of export products from Japan fluctuated with no clear trend. The figures on export varieties from the selected countries are consistent with the hypothesis that there is an inverted-u relationship between stage of industrial development and diversification of exports ( Imbs and Wacziarg, 2003): a country at a lower stage of development is likely to export fewer varieties compared to a country at an intermediate stage ( for example, China) while a country at an advanced stage of development ( for example, Germany and Japan) also export fewer varieties of products. 4. Intensive and Extensive Margins This section examines the intensive and extensive margins of Canada s exports to the United States with reference to the Rest of the World, the European Union, China and Japan. Table 7. Canada s Intensive and Extensive Margins in the US with Respect to the Rest of the World. Year Imrw Emrw csrw pccsrw pcimrw pcemrw 1992 0.247.920 0.227 1993 0.257.920 0.236 0.040 0.040 0.000 1994 0.258.929 0.240 0.014 0.004 0.010 1995 0.258.936 0.136-0.565-0.590 0.026 1996 0.272.940 0.256 0.629 0.643-0.014 1997 0.252.943 0.238-0.073-0.076 0.003 1998 0.251.933 0.234-0.015-0.004-0.011 1999 0.255.944 0.241 0.028 0.016 0.012 2000 0.246.949 0.233-0.031-0.036 0.005 2001 0.249.939 0.234 0.002 0.012-0.011 2002 0.231.951 0.220-0.062-0.075 0.013 2003 0.227.944 0.214-0.025-0.017-0.007 2004 0.222.952 0.211-0.014-0.022 0.008 2005 0.221.947 0.209-0.010-0.005-0.005 2006 0.207.940 0.195-0.073-0.065-0.007 2007 0.201.963 0.194-0.005-0.029 0.024 2008 0.203.945 0.192-0.009 0.010-0.019 2009 0.179.949 0.170-0.122-0.126 0.004 2010 0.18.944 0.170 0.000 0.006-0.005 2011 0.176.944 0.166-0.022-0.022 0.000 2012 0.176.945 0.166 0.001 0.000 0.001 Notes: Imrw= Intensive margin with respect the rest of the world, Emrw= extensive margin with respect to the rest of the world, csrw= Canada s exports relative to the exports from the rest of the world to the U.S.A., pccsrw= Percentage change in csrw, pcimrw = percentage change in Imrw, and pcemrw= percentage change in Emrw. Source: Computed from data available from the Census Bureau, US Department of Commerce. It is evident from Table 7 that Canada s intensive margin of exports appears to show a downward trend from 0.247 in 1992 to 0.176 in 2012. It is worthwhile to note that the intensive margin measures the ratio of Canada s exports to exports from the Rest of the 104

Islam, S. Product Varieties, Extensive And Intensive Margins Of Trade: Evidence From Canada World to the United States involving the same set of goods. The extensive margin (Emrw) was 0.945 in 2012 with no clear trend during 1992-2012. The figures on the extensive margin are higher than those found by the Office of Chief Economist, Department of Foreign Affairs and Trade, Canada (2013) involving Canada and Chile. It is evident from Table 7 that the Canada s exports relative to exports from the Rest of the World (csrw) reveals a downward trend from 0.227 in 1992 to 0.166 in 2012. It should be noted that csrw is the product of intensive and extensive margins. Table 7 also reports the values of percentage changes in intensive margin (pcimrw), percentage changes in extensive margins (pcemrw), and percentage changes in csrw (pccsrw); the sum of pcimrw and pcemrw equals pccsrw. It can be observed that during most of the years the percentage change in the intensive margin (pcimrw) dominated the percentage change in csrw ( pccsrw). For example, in 2009, a recession year, Canada s csrw declined by 12.2% This was associated with a value of -0.126 for pcimrw; that is Canada s intensive margin declined by 12.6%. The decline in Canada s intensive margin could be interpreted as a reflection of Canada s loss of competitiveness with respect to the Rest of the World. Table 8. Canada s Intensive and Extensive Margins in the US with Respect to the European Union. Year Imeu Emeu cseu Pcimeu Pcemeu Pcseu 1992 1.082 0.891 0.964 1993 1.144 0.905 1.035 0.056 0.016 0.072 1994 1.151 0.915 1.052 0.006 0.010 0.016 1995 1.169 0.917 1.072 0.016 0.003 0.018 1996 1.160 0.920 1.068-0.007 0.003-0.004 1997 1.110 0.933 1.035-0.044 0.013-0.031 1998 1.034 0.925 0.956-0.071-0.009-0.080 1999 1.042 0.949 0.990 0.008 0.027 0.035 2000 1.084 0.935 1.014 0.039-0.015 0.024 2001 1.030 0.923 0.950-0.051-0.014-0.064 2002 0.945 0.948 0.895-0.086 0.027-0.059 2003 0.922 0.945 0.871-0.024-0.003-0.028 2004 0.943 0.959 0.904 0.022 0.015 0.037 2005 0.999 0.935 0.934 0.058-0.025 0.033 2006 0.986 0.923 0.910-0.013-0.013-0.027 2007 0.942 0.948 0.894-0.045 0.027-0.018 2008 0.979 0.942 0.923 0.038-0.006 0.032 2009 0.847 0.947 0.802-0.145 0.005-0.140 2010 0.929 0.935 0.869 0.092-0.013 0.080 2011 0.916 0.933 0.855-0.014-0.002-0.016 2012 0.889 0.955 0.849-0.030 0.023-0.007 Notes: Imeu= Intensive margin with respect the European Union, Emeu= extensive margin with respect to the European Union, cseu= Canada s exports relative to the exports from the European Union to the U.S.A., pcimeu= Percentage change in Imeu, pcemeu = percentage change in emeu, and pcseu= percentage change in cseu. Source: Computed from data available from the Census Bureau, US Department of Commerce. Table 8 presents Canada s intensive, extensive margins, and related variables with respect to the European Union. It is evident that Canada s intensive margin (Imeu) increased 105

during 1992-1995 but declined since then. The extensive margin (Emeu) fluctuated with a value of 0.891 in 1992 to 0.955 in 2012. Canada s exports relative to exports from the European Union (cseu) also appears to show a downward trend. Table 8 also reports percentage changes in the three variables. As observed previously in Table 7, Table 8 also reveals that percentage changes in the intensive margin dominates the percentage change in cseu. For example, in 2009, pcseu was -0.14. This was associated with a value of - 0.145 for pcimeu; this suggests that the intensive margin declined by 14.5 % outweighing a slight positive change ( 0.005) in pcemeu. Table 9 reports the intensive margin (Imch), extensive margin (Emch), the share of Canada s exports relative to China s exports to the United States (csch), and percentage changes in these three variables. It is striking to note that Canada s intensive margin with reference to China has steadily declined from 4.34 in 1992 to 0.8 in 2012. Table 9. Canada s Intensive and Extensive Margins in the US with Respect to China Year Imch Emch csch Pcimch Pcemch Pccsch 1992 4.340 0.883 3.834 1993 3.980 0.886 3.526-0.087 0.003-0.084 1994 3.684 0.899 3.311-0.077 0.014-0.063 1995 3.477 0.912 3.170-0.058 0.014-0.043 1996 3.257 0.929 3.026-0.066 0.019-0.046 1997 2.832 0.944 2.673-0.140 0.016-0.124 1998 2.638 0.923 2.434-0.071-0.023-0.094 1999 2.651 0.916 2.430 0.005-0.007-0.002 2000 2.447 0.943 2.308-0.080 0.029-0.051 2001 2.272 0.931 2.115-0.074-0.014-0.088 2002 1.772 0.943 1.670-0.249 0.013-0.236 2003 1.541 0.943 1.454-0.139 0.000-0.139 2004 1.368 0.953 1.303-0.120 0.011-0.109 2005 1.249 0.955 1.193-0.091 0.002-0.089 2006 1.099 0.956 1.051-0.127 0.001-0.127 2007 1.020 0.967 0.986-0.075 0.012-0.063 2008 1.053 0.955 1.005 0.032-0.013 0.019 2009 0.797 0.957 0.763-0.278 0.003-0.275 2010 0.796 0.955 0.761-0.001-0.002-0.003 2011 0.827 0.955 0.790 0.037 0.000 0.037 2012 0.800 0.951 0.761-0.032-0.004-0.037 Notes: Imch= Intensive margin with respect China, Emch= extensive margin with respect to China csch= Canada s exports relative to the exports from China to the U.S.A., pcimch= Percentage change in imch, pcemch= percentage change in Emch, and pccsch= percentage change in csch. Source: Computed from data available from the Census Bureau, US Department of Commerce. This suggests that while Canada exported to the United States more than four times compared to China, in 2012, Canada exported only four-fifths of China s exports involving the same set of goods. The evidence on Canada s intensive margin with respect to China suggests that competition between Canada and China in the U.S. market has increased; this is contrary to the findings of an earlier study by Boileau (2007). It appears from Table 9 that the extensive margin of Canada shows an upward trend, increasing from 0.883 in 1992 to 0.951 in 2012. Canada s exports relative to China s 106

Islam, S. Product Varieties, Extensive And Intensive Margins Of Trade: Evidence From Canada exports ( csch) also demonstrates a downward trend. It is also evident from Table 9 that the percentage change in the intensive margin (pcimch) dominates the percentage change in csch (pccsch). In 2009, for instance, the value of pccsch was -0.275. This was associated with a value of -0.278 for pcimch and a value of 0.003 for pcemch. A similar scenario occurred in 2002. Finally, Table10 presents Canada s intensive, extensive margins, and related variables with reference to Japan. It is clearly evident that Canada s intensive margin (Imj) steadily increased from 1.067 in 1992 to 2.366 in 2012. This suggests that with respect to Japan Canada s competitive position improved during 1992-2012. The value of the extensive margin (Emj) was 0.935 in 2012 with no clear trend during 1992-2012. The value of csj clearly shows an upward trend, implying a rising trend in Canada s exports relative to Japan s exports. It can be observed from Table 10 that the percentage change in the intensive margin (pcimj) dominates the percentage in Canada s exports relative to Japan s exports (pccsj). Table 10. Canada s Intensive and Extensive Margins in the US with Respect to Japan Year Imj Emj csj Pcimj Pcemj Pccsj 1992 1.067 0.949 1.012 1993 1.107 0.937 1.037 0.037-0.013 0.024 1994 1.135 0.949 1.078 0.025 0.014 0.038 1995 1.235 0.947 1.169 0.084-0.003 0.082 1996 1.428 0.948 1.353 0.145 0.001 0.146 1997 1.475 0.932 1.375 0.032-0.017 0.016 1998 1.505 0.945 1.422 0.020 0.014 0.034 1999 1.600 0.949 1.518 0.062 0.004 0.066 2000 1.650 0.955 1.576 0.031 0.007 0.037 2001 1.779 0.961 1.710 0.075 0.007 0.082 2002 1.793 0.961 1.722 0.008-0.001 0.007 2003 1.983 0.947 1.877 0.101-0.015 0.086 2004 2.076 0.951 1.975 0.046 0.005 0.051 2005 2.257 0.932 2.104 0.083-0.020 0.063 2006 2.248 0.908 2.041-0.004-0.026-0.030 2007 2.285 0.954 2.180 0.017 0.049 0.066 2008 2.634 0.925 2.438 0.142-0.030 0.112 2009 2.459 0.960 2.362-0.069 0.037-0.032 2010 2.476 0.930 2.303 0.007-0.032-0.025 2011 2.614 0.936 2.446 0.054 0.006 0.060 2012 2.366 0.935 2.213-0.100 0.000-0.100 Notes: Imj= Intensive margin with respect Japan, Emj= extensive margin with respect to Japan, csj= Canada s exports relative to the exports from Japan to the U.S.A., pcimj= Percentage change in Imj, pcemj= percentage change in Emj, and pccsj= percentage change in csj. Source: Computed from data available from the Census Bureau, US Department of Commerce. 5. Conclusion Induced by multilateral and regional trade liberalizations, bilateral trade between Canada and the United States has expanded in absolute terms. Canada s exports to the United States is still dominated by natural resource-based products and vehicles. 107

The empirical findings show that the number of Canadian manufacturing establishments exporting to only the United States as a percentage of total number of establishments has declined. Furthermore, the value of exports from establishments serving only the US market as a percentage of total exports has also declined. The data on export varieties show that Canada export more products than the selected countries; however, China export more varieties than Canada. The empirical findings reveal that Canada s intensive margin declined with reference to the Rest of the World, the European Union, and China. The decline was quite pronounced with reference to China. Canada s intensive margin however, increased with reference to Japan. Furthermore, the percentage change in Canada s exports relative to the selected regions/countries were dominated by percentage changes in the intensive margin rather than the extensive margin. The findings also show that during the Great Recession in the United States, Canada s share relative the shares of the selected regions/countries in the United States declined significantly. This was associated with a significant decline in Canada s intensive margin. Accordingly, the evidence suggests that Canada s competitive position in the U.S. market has weakened. References Baldwin,J.R.and Gu,W. (2004). Trade Liberalization: Export-market Participation,Productivity Growth, and Innovation. Oxford Review of Economic Policy, 20 (3), 372-92. Baldwin,J.R. and Gu,W.(2003). Participation in Export Markets and Productivity Performance in Canadian Manufacturing. Canadian Journal of Economics, 36(3), 634-54. Bernard,A., Jenson,J., Redding,S., and Schott,P. (2009). The Margins of US Trade. CEP Discussion Paper No. 906, Bernard,A.B.,Eaton,J., Jensen,J.B., and Kortum,S. (2003). Plants and Productivity in International Trade. American Economic Review, 93(4), 1268-1290. Boileau,D. (2007). Competition from China is Not Largely responsible for Canada s Decline in Share of U.S. Imports, APS-004. Castillo Ponce,R.A. and Ramirez Acosta,R. (2008). Economic Integration in North America. Applied Econometrics and International Development, 8(2), July-December. Helpman,E. and Krugman,P. (1985). Market Structure and Foreign Trade. Cambridge, MIT Press. Hillberry,R. and McDaniel,C. (2003). A Decomposition of North American Trade Growth Since nafta. USITC Working Paper 2002-12-A. Hummels,D. and Klenow,P.J. ( 2005). The variety and Quality of a nation s Exports. American Economic Review, 95(3), 704-723. Imbs,J. and R. Wacziarg (2003), Stages of Diversification. American Economic Review, Vol. 93(1), 63-86. Krugman, P. (1979). Increasing Returns, Monopolistic Competition, and International Trade. Journal of International Economics, 9(4)469-479. Melitz,M. (2003). The Impact of Trade on Intra-Industry Reallocations and Aggregate Industry Productivity. Econometrica, 71(6), 1695-1725. Office of the Chief Economist, Foreign Affairs and International Trade Canada. The Economic Impact of the Canada-Chile Free Trade Agreement, May 2013. Statistics Canada (2011). A Profile of Canadian Exporters, 1996-2009. Journal published by the EAAEDS: http://www.usc.es/economet/eaat.htm 108