The position and length of Canadian supply chains

Colin Scarffe
July 2022

Key points

1. Canada specializes in the early (upstream) production stages of the supply chain. In general, Canada’s exports will be earlier in the value chain and closer to raw materials, while Canada’s imports will come from later in the value chain and be closer to final goods. This means that, in general, the Canadian economy is more susceptible to disruptions that happen in the later stages of supply chains, and is less susceptible to disruptions from the beginning stages of supply chains.
2. A few notable industries—such as natural resource industries—exemplify Canada’s role as early producers in supply chains. Despite exporting different baskets of goods and services to each trading partner, Canadian exports to all trade partners are, on average, located in the early stages of the supply chain.
3. Since 1997, Canadian exports have moved to slightly earlier production stages (closer to raw materials) in the value chain, while Canada’s import positioning has stayed relatively constant. The shift in exports to earlier stages occurred in the 2000’s due to a growing share of oil and gas extraction in exports and a diminishing share of auto vehicle manufacturing.
4. Since 2010, there have been no major trends for Canada’s trade with the U.S., Mexico, or the rest of the world. The one exception has been China. While Canada’s exports to China still skew closer to the early stages of the value chain and closer raw materials, over the past 10 years these exports have moved further along the value chain closer to the final stages of production. Likewise, while Canada’s imports from China still skew closer to the final stages of the value chain and closer to final goods, over the last 10 years these imports have shifted towards the early stages of the value chain and have become closer to raw materials.

1. Introduction

Several events have recently brought supply chains, or global value chains (GVCs), into the public spotlight.^{Footnote 1} In the early days of the COVID-19 pandemic, many countries shut production facilities in an effort to contain the pandemic’s spread. Later, a shift in consumption from services to goods increased the pressure on supply chains, and transportation networks to keep up with demand. Prior to the pandemic, rising protectionism and trade tensions among countries that have historically been pro-free trade, were already beginning to influence the positioning and structure of global supply chains. In the post-COVID period, protectionism along with increased disruptions from climate change events will likely keep supply chains in the spotlight for many years to come. In the Canadian context, a more specific question emerges: how might supply chain disruptions affect Canadians and Canadian businesses?

In order to analyze the vulnerability and risks of a supply chain disruption to Canada, it is important to understand where Canada and Canadian firms fit into GVCs. On the one hand, if Canada’s production mainly happens at the beginning—or upstream—of a GVC, the primary concern would be a negative demand shock to Canadian industries. On the other, if Canadian production happens at the end—or downstream—of a GVC, the primary concern would be a supply shock to key inputs. If Canada’s production is in the middle of a GVC, both supply and demand shocks are concerns. This paper will attempt to answer where Canada fits into the global production process by assessing the upstreamness and downstreamness of Canadian production and trade.

This paper is organized as follows: section 2 presents the data and provides a detailed discussion on the methodology used to measure the position of production and trade for Canada. Section 2.1 provides simple measures of upstreamness and downstreamness, which are essential for understanding the results. Section 2.2 presents more complex measures, which are the basis for most of the paper. The measures in section 2.2 are mathematically technical; however, they follow the same logic as the simple measures in section 2.1. Section 3 presents the results for the upstreamness and downstreamness of Canadian industries. Section 4 provides the results for Canadian production and trade positioning in 2019 as well as Canada’s main trading partners. Section 5 examines how Canada’s upstreamness and downstreamness has changed over time. Lastly, section 6 concludes.

2. Data and methodology

All of the data in this paper comes from Statistics Canada’s symmetric Input-Output (IO) tables. An IO table models the structure of an economy, detailing how industries acquire their inputs (either intermediate inputs or value added) and what they do with their output (producing either intermediate inputs for other industries or final goods). The most recent Canadian IO tables have 234 goods and services industries, which will be the basis for measuring Canada’s GVC positioning in this paper. At the time of writing, the latest IO table was 2019 and is the basis for sections 3 and 4. In section 5, the IO tables from 1997-2019 are used.

This paper is an attempt to classify where Canada fits into the production process. Two complementary measures are used to quantify the production process:

• Output upstreamness from final goods
• Input downstreamness from value added

Output upstreamness refers to how far are an industry’s outputs from being a final good (i.e. consumed directly, a capital good, or exported). The output upstreamness measures the industry positioning by examining the use of the outputs. On the other hand, input downstreamness refers to how far are an industry’s inputs from being a value added input (i.e. labour, capital, or natural resources). The input downstreamness measures the industry positioning by examining the source of the inputs.

The auto industry can be used as an example for the two measures, and are used in figure 1 as the two end points. When the metals used to make a car are mined, the outputs are far from being a final good, and the inputs are close to being natural resources. Resultantly, mining is both an upstream (based on the outputs) and not a downstream (based on the inputs) industry. The metals are then processed, which is then used to construct the frame of the car. At this point, the outputs become less upstream (closer to a final good) and the inputs are more downstream (further away from value added). Finally, the car frame (along with many other components) are assembled into a car and sold as a consumer good. The car is a final good (not upstream from anything) while the inputs are have all been through several stages of production prior to final assembly (far away from value added).

Importantly, not all supply chains will be the same length; two industries that have similar upstream positions might have very different downstream positions and vice-versa. For example, automobile and light-duty motor vehicle manufacturing and education both produce final goods almost exclusively. However, autos are a downstream industry (most of the inputs have been through multiple stages of production) whereas education is not a downstream industry because most of the input is value added in the form of labour and human capital. This is a case where one industry (autos) has a longer supply chain than another (education), and hopefully clarifies why it is important to examine both the inputs and outputs. For the rest of the paper moving up the value chain and further away from final goods will be referred to as becoming more upstream, and moving down the value chain and further away from raw materials will be referred to as becoming more downstream.

2.1 Two simple measures

This sub-section will attempt to quantify where an industry falls in the production process using the simplest relations possible. This sub-section will use the simple measures found in Antràs and Chor (2018). The total use (or output) of a sector can be divided into two classifications: intermediate outputs—which are used as inputs in other industries—and final goods—which include outputs that are directly consumed, capital goods, and exports. The basic relation is:

Text version

Y_i is equal to the sum from j to n of Z_ij plus F_i

The total use of sector i (Y_i) is equal to the sum of all intermediate output (Z_ij) from sector i to each sector j (including sector i itself), plus the output that are final goods (Fi). The simple measure of output upstreamness is the share of an industry’s output that is used as an intermediate inputs for other industries. Alternatively, the simple measure is one minus the share of the industry’s output that is final goods. Formally:

Text version

U_i is equal to the sum from j to n of Z_ij divided by Y_i which is equivalent to 1 minus F_i divided by Y_i

A sector where the output is largely final goods will have an upstreamness score, or , close to zero. A sector where the output is primarily intermediate inputs for other industries will have a score close to one. For example, sectors that process raw materials are upstream; they will have large intermediate outputs (Z­_ji) and small final outputs (F­i). These sectors will have upstream scores close to one. On the other hand, construction sectors are not upstream as construction is generally not used as an intermediate input for other industries. They will have limited intermediate outputs and be mostly comprised of final outputs; resultantly, construction industries have an upstream score close to zero.

An alternative method to using the output of an industry is to focus on the inputs into an industry to measure the position in the production process. The basic relation for inputs, and the counterpart of equation (1) is:

Text version

Y_i is equal to the sum from i to n of Z_ij plus VA_j

Where the total inputs (Yj) of industry j (which is the exact same as the output Y_i in equation (1) when i equals j) is equal to the sum of intermediate inputs (Z_ij) used by sector j from each sector i, plus the value added (VAj) in sector j. In general, there are three sources of value added: labour, capital, and natural resources. The simple measure of input downstreamness is the share of the industry’s input that are intermediate outputs from other industries—or one minus the share of the industry’s input that is value added. Formally:

Text version

D_j is equal to the sum from i to n of Z_ij divided by Y_j which is equivalent to 1 minus VA_j divided by Y_j

If an industry uses many intermediate inputs (such as construction) then it is a downstream industry and will have a higher downstream score (i.e. a score close to one)—the inputs have passed through many stages of production before entering this industry. On the other hand, if it has a high value added (such as natural resource extraction) then it is not a downstream industry and will have a score close to zero. Using the network diagram found in the appendix, the two simple measures of upstreamness and downstreamness measure the lines “a.” and “b.” for all industries.

Figure 2 presents a summary of the four types of industry characterizations. Re-enforcing a point made earlier, not all industries have the same length of value chain. Many goods industries will go through many stages of production and will therefore have a longer value chain. Many service industries do not have multiple stages of production (that are measurable by an IO-table) and will therefore have a short value chain. When an industry is in the middle of a long and complex value chain, it is possible that an industry is both more upstream and more downstream than another.

2.2 Using the IO structure for a single economy

More complex measures of upstreamness and downstreamness are developed in this section taking advantage of the IO table’s structure and well developed algebra. The measures in this section are based on the work of Antràs et al. (2012), Miller and Temurshoev (2015), and Antràs and Chor (2018). Figure 3 depicts the basic structure of an IO table.

Instead of only calculating the share of intermediate goods in a sector’s output, the relation between sectors can be used to count the number of steps in production it takes for the output of a sector to become a final good. Starting with a closed economy (setting inventories, exports and imports to zero) and letting , equation (1) can be re-written:^{Footnote 2}

Text version

Y_i is equal to the sum from j to n of a_ji multiplied by Y_j plus F_i

Y_i is equal to F_i plus the sum from j to n of a_ji multiplied by F_j plus the sum from j and k to n of a_ji multiplied by a_jk multiplied by Y_k

Y_i is equal to F_i plus the sum from j to n of a_ji multiplied by F_j plus the sum from j and k to n of a_ji multiplied by a_jk multiplied by F_k plus the sum from j, k, and z to n of a_ji multiplied by a_jk multiplied by a_kz multiplied by F_z plus etc.

The key part of the final line of equation (5) is that each term specifies the number of steps needed for outputs to become final goods. The first term requires only one step to become a final good (that is the output itself), the second term requires the refinement of one additional industry before becoming a final good (two steps), the third term requires the refinement of two additional industries before becoming a final good (three steps) etc. In order to measure of upstreamness of production, the numbers of steps needed to become a final good are counted.

Text version

U_i multiplied by Y_i is equal to one times F_i plus two times the sum from j to n of a_ji multiplied by F_j plus three times the sum from j and k to n of a_ji multiplied by a_jk multiplied by F_k plus four times the sum from j, k, and z to n of a_ji multiplied by a_jk multiplied by a_kz multiplied by F_z plus etc.

Although equation (6) seems like it is a summation of an infinite number of terms, as long as the “a’s” are less than one, it is a geometric sum which simplifies to [I-A]-2F, where A is a matrix of the a_ji’s and F is the vector of final demand by industry. Using the relation for F in the first line of equation (5), it simplifies further to:

Text version

U is equal to the Identity minus the A matrix to the exponent minus 2, multiplied by the F vector, all divided by the Y vector. Which is equal to the Identity minus the A matrix inverse, multiplied by the Y vector and all divided by the Y vector

The third term in equation (7) is the Leontief inverse multiplied by industry share of output and is the basis for IO multipliers. Using a result found in Miller and Temurshoev, the output upstreamness in (7) can be interpreted as the total number of forward linkages in a given industry. An industry such as electricity generation has high output upstreamness because it has many forward linkages. Many countries, such as China, choose to subsidize industries with high forward linkages because many industries indirectly benefit from the subsidy.

A similar process can be done for the input downstreamness. Letting , equation (3) can be re-written as:^{Footnote 3}

Text version

Y_i is equal to the sum from j to n of Y_j multiplied by b_ji plus VA_i

Y_i is equal to VA_i plus the sum from j to n of VA_j multiplied by b_ji plus the sum from j and k to n of VA_k multiplied by b_jk multiplied by b_ki plus the sum from j, k, and z to n of VA_z multiplied by b_jk multiplied by b_kz multiplied by b_zi plus etc.

Similar to equation (5), the key part of equation (8) is that each term represents how far away the inputs are from being primary factors (i.e. the value added). The first term includes primary factors for industry i. In the second term, the primary factors have been transformed by one stage of production first. In the third term, the primary factors have been transformed by two stages of production, etc. Counting all stages of production results in:

Text version

D_i is equal to one times VA_i divided by Y_i plus two times the sum from j to n of VA_j divided by Y_i multiplied by b_ji plus three times the sum from j and k to n of VA_k divided by Y_i multiplied by b_jk multiplied by bki plus four times the sum from j, k, and z to n of VA_z divided by Y_i multiplied by b_jk multiplied by b_kz multiplied by b_zi plus etc.

D is equal to VA transpose multiplied by the identity minus the B matrix to the exponent minus two, all divided by the Y vector

D is equal to Y transpose multiplied by the identity minus the B matrix inverse, all divided by the Y vector

 (Where VA’ is the transpose of the value added vector, and B is the matrix of b_ji’s)

This time, the total input shares are multiplied by the Ghosh inverse, or the input counterpart to the Leontief inverse. In the same way that the output upstreamness in (7) can be interpreted as the number of forward linkages, the input downstreamness in (9) can be interpreted as the number of backward linkages. Industries that have a high number of backward linkages are also often targets for subsidies because they can increase demand in a large number of industries.

The upstreamness and downstreamness measures are different—although mathematically and conceptually quite similar—ways of measuring the same thing. Miller and Temurshoev (2015) highlight the mathematical links between the two methods. One of the main takeaways from Miller and Temurshoev is that when the upstreamness and downstreamness of each industry is weighted by its share of output, then these two measures will give the exact same result. However, when the industry upstreamness and downstreamness measures are weighted by exports and imports, they will tend to give different results. These more complex measures of upstreamness and downstreamness attempt to account for all the lines in the network diagram found in the appendix.

Lastly, the terms set to zero—net inventories, imports, and exports—need to be accounted for on the output side of the economy. Following Antràs et al. (2012), the following adjustment is made to the elements within the A matrix:^{Footnote 4}

Text version

a_ji hat is equal to a_ji multiplied by Y_i divided by Y_i minus N_i plus M_i minus X_i

As many of the results in this paper will be about Canada’s role in international supply chains, the results are based on the Canadian IO tables. This means that exports and imports are placed as if they were going through a Canadian production process. In practice, after leaving Canada or arriving in Canada, the traded products could have different values of upstreamness and downstreamness depending on the partner economy’s IO table; however, it is expected that industries would have similar upstream and downstream placements across economies, and thus the values would be similar.^{Footnote 5}

3. Industry Results

To begin the analysis, the complex measures of upstreamness and downstreamness are plotted against the simple measures for each industry in figure 4.

Figure 4 - Text version

Industry	Description	Upstream	simple upstreamness	Domestic downstream	simple downstreamness
BS111A00	Crop production (except cannabis, greenhouse, nursery and floriculture production)	2.87167	0.735631765	2.150010079	0.525807653
BS1114A0	Greenhouse, nursery and floriculture production (except cannabis)	1.7092	0.561149877	2.125945159	0.537895993
BS111CL0	Cannabis production (licensed)	1	0	2.120325405	0.601437024
BS111CU0	Cannabis production (unlicensed)	1	0	1.517747591	0.320952155
BS112A00	Animal production (except aquaculture)	2.84956	0.955159005	2.940425936	0.811927441
BS112500	Aquaculture	2.2801	0.663758006	2.395199977	0.576290043
BS113000	Forestry and logging	3.94806	0.963858438	2.335320097	0.600445657
BS114000	Fishing, hunting and trapping	2.2646	0.661100345	1.897825183	0.397365135
BS115A00	Support activities for crop and animal production	3.77445	0.98515593	1.464160797	0.219021784
BS115300	Support activities for forestry	4.51861	0.990993547	2.065102171	0.481418805
BS211110	Oil and gas extraction (except oil sands)	3.54886	0.895507564	1.886295564	0.457472313
BS211140	Oil sands extraction	3.72287	0.964119778	1.849122489	0.441233508
BS212100	Coal mining	4.12172	0.996743196	1.651248943	0.316209241
BS212210	Iron ore mining	4.80345	0.991239631	1.776463324	0.367518713
BS212220	Gold and silver ore mining	6.08769	0.995847192	1.911010169	0.443032174
BS212230	Copper, nickel, lead and zinc ore mining	6.72409	0.997689722	1.733903333	0.358113742
BS212290	Other metal ore mining	6.40919	0.996865148	1.883497087	0.431377968
BS212310	Stone mining and quarrying	2.93943	0.991268351	1.866465849	0.431887772
BS212320	Sand, gravel, clay, and ceramic and refractory minerals mining and quarrying	3.3527	0.9869198	1.857648281	0.422271754
BS212392	Diamond mining	2.82647	0.935078751	2.025182828	0.48181972
BS21239A	Other non-metallic mineral mining and quarrying (except diamond and potash)	3.81175	0.973989445	1.744938287	0.366369881
BS212396	Potash mining	3.6722	0.93711891	1.54632371	0.28593562
BS21311A	Support activities for oil and gas extraction	3.49268	0.996652135	1.732959304	0.340458394
BS21311B	Support activities for mining	2.84554	0.622293346	1.856345719	0.429928959
BS221100	Electric power generation, transmission and distribution	2.54896	0.644419654	1.576984444	0.302512256
BS221200	Natural gas distribution	2.32221	0.543578723	1.356094227	0.194960871
BS221300	Water, sewage and other systems	2.44468	0.617210163	1.941724375	0.45396451
BS23A000	Residential building construction	1	0	2.198079061	0.550591291
BS23B000	Non-residential building construction	1	0	2.120154094	0.539058376
BS23C100	Transportation engineering construction	1	0	2.25845308	0.611563521
BS23C200	Oil and gas engineering construction	1	0	2.145927349	0.591905821
BS23C300	Electric power engineering construction	1	0	1.995496551	0.465345428
BS23C400	Communication engineering construction	1	0	2.570571686	0.724996702
BS23C500	Other engineering construction	1	0	2.21185182	0.587085255
BS23D000	Repair construction	2.80497	0.990339795	1.901220533	0.411102503
BS23E000	Other activities of the construction industry	2.96031	0.959565375	1.637890012	0.303316895
BS311100	Animal food manufacturing	2.91006	0.688690129	2.933252409	0.811915581
BS311200	Grain and oilseed milling	2.54539	0.745793253	2.796831427	0.839854389
BS311300	Sugar and confectionery product manufacturing	1.60802	0.383084358	2.60080699	0.682018473
BS311400	Fruit and vegetable preserving and specialty food manufacturing	1.41508	0.261436321	2.53486352	0.690763967
BS311500	Dairy product manufacturing	1.59317	0.377864156	3.082729574	0.769595627
BS311600	Meat product manufacturing	1.8234	0.485498749	3.113859365	0.765215591
BS311700	Seafood product preparation and packaging	1.96312	0.561076991	2.697514953	0.785497355
BS311800	Bakeries and tortilla manufacturing	1.45773	0.287047465	2.440895961	0.612462175
BS311900	Other food manufacturing	1.59143	0.375528952	2.502667711	0.662213412
BS312110	Soft drink and ice manufacturing	1.71074	0.379196533	2.499030813	0.669080482
BS312120	Breweries	1.81378	0.439171636	2.066971983	0.492273594
BS3121A0	Wineries and distilleries	1.42166	0.255745048	2.215379337	0.556271119
BS312200	Tobacco manufacturing	1.1779	0.085638356	1.893090626	0.437600665
BS31A000	Textile and textile product mills	2.0376	0.613816969	2.496560857	0.65664368
BS31B000	Clothing and leather and allied product manufacturing	1.24707	0.15101278	2.359811541	0.622388048
BS321100	Sawmills and wood preservation	3.10252	0.969955772	2.566857543	0.691123473
BS321200	Veneer, plywood and engineered wood product manufacturing	2.46069	0.976097783	2.451314013	0.638263049
BS321900	Other wood product manufacturing	2.46288	0.972872525	2.486967992	0.658782415
BS322100	Pulp, paper and paperboard mills	3.4561	0.889047044	2.558739891	0.70189512
BS322200	Converted paper product manufacturing	2.57063	0.749183108	2.596356817	0.67346355
BS323000	Printing and related support activities	2.8825	0.959381073	2.241681632	0.542935192
BS324110	Petroleum refineries	2.80772	0.705523639	2.566917036	0.824725391
BS3241A0	Petroleum and coal product manufacturing (except petroleum refineries)	3.28331	0.969596886	2.786718736	0.773859352
BS325100	Basic chemical manufacturing	4.20469	0.95822152	2.488762295	0.677516785
BS325200	Resin, synthetic rubber, and artificial and synthetic fibres and filaments manufacturing	3.81166	0.987750355	2.562114436	0.660144831
BS325300	Pesticide, fertilizer and other agricultural chemical manufacturing	3.74347	0.945048651	2.485909975	0.696837367
BS325400	Pharmaceutical and medicine manufacturing	2.05504	0.740128943	2.447699201	0.646991891
BS325500	Paint, coating and adhesive manufacturing	2.99439	0.927938981	2.584384321	0.679130278
BS325600	Soap, cleaning compound and toilet preparation manufacturing	1.66751	0.333848023	2.433349647	0.645720456
BS325900	Other chemical product manufacturing	3.48806	0.952159465	2.602346746	0.679265336
BS326100	Plastic product manufacturing	2.84054	0.924144599	2.519050813	0.645253122
BS326200	Rubber product manufacturing	2.99197	0.85076222	2.627688216	0.677662672
BS327A00	Non-metallic mineral product manufacturing (except cement and concrete products)	2.63798	0.875719313	2.273767442	0.59725372
BS327300	Cement and concrete product manufacturing	2.47102	0.983008919	2.30832111	0.626034676
BS331100	Iron and steel mills and ferro-alloy manufacturing	3.77453	0.998893629	2.702303094	0.769255962
BS331200	Steel product manufacturing from purchased steel	3.79807	0.997739598	2.857345957	0.744575814
BS331300	Alumina and aluminum production and processing	4.19864	0.943801343	2.763510789	0.709116456
BS331400	Non-ferrous metal (except aluminum) production and processing	5.92302	0.998427851	3.11642242	0.921606431
BS331500	Foundries	3.56682	0.995607938	2.461208673	0.604881137
BS332100	Forging and stamping	3.19631	0.978932277	2.630405704	0.686055094
BS332A00	Cutlery, hand tools and other fabricated metal product manufacturing	2.66562	0.881686264	2.228201014	0.540240057
BS332300	Architectural and structural metals manufacturing	2.68622	0.981679782	2.441413667	0.614673292
BS332400	Boiler, tank and shipping container manufacturing	2.62978	0.848304633	2.449282504	0.606106283
BS332500	Hardware manufacturing	2.82275	0.975824656	2.470517839	0.632699382
BS332600	Spring and wire product manufacturing	2.5951	0.932308417	2.558657443	0.663550008
BS332700	Machine shops, turned product, and screw, nut and bolt manufacturing	2.9348	0.951143772	2.161600909	0.50390712
BS332800	Coating, engraving, cold and heat treating and allied activities	3.63683	0.987004268	2.104835866	0.498247959
BS333100	Agricultural, construction and mining machinery manufacturing	2.42104	0.481681412	2.387369859	0.612250924
BS333200	Industrial machinery manufacturing	1.73415	0.30880507	2.287559593	0.575982465
BS333300	Commercial and service industry machinery manufacturing	1.92653	0.395635938	2.247375686	0.568407319
BS333400	Ventilation, heating, air-conditioning and commercial refrigeration equipment manufacturing	2.25888	0.784567897	2.316387952	0.574334478
BS333500	Metalworking machinery manufacturing	2.29544	0.569925061	2.266625337	0.56402597
BS333600	Engine, turbine and power transmission equipment manufacturing	2.91369	0.854598274	2.280605334	0.569830326
BS333900	Other general-purpose machinery manufacturing	2.31306	0.562259308	2.427463106	0.622817327
BS334100	Computer and peripheral equipment manufacturing	1.29483	0.133923217	2.308542238	0.593402867
BS334200	Communications equipment manufacturing	1.92206	0.520583683	2.226227859	0.567977675
BS334A00	Other electronic product manufacturing	2.08094	0.637082093	2.178415939	0.544224288
BS334400	Semiconductor and other electronic component manufacturing	3.22036	0.964480512	2.366357248	0.611821174
BS335100	Electric lighting equipment manufacturing	2.18396	0.720278844	2.301171752	0.590037294
BS335200	Household appliance manufacturing	1.31759	0.183013877	2.549413434	0.685393758
BS335300	Electrical equipment manufacturing	2.74345	0.798425794	2.503882283	0.6307656
BS335900	Other electrical equipment and component manufacturing	2.64091	0.900664066	2.697279255	0.664299641
BS336110	Automobile and light-duty motor vehicle manufacturing	1.02227	0.017451335	3.284526338	0.867929312
BS336120	Heavy-duty truck manufacturing	1.06225	0.033032446	2.983044993	0.789445974
BS336200	Motor vehicle body and trailer manufacturing	1.21227	0.151628777	2.732421639	0.686559104
BS336310	Motor vehicle gasoline engine and engine parts manufacturing	2.40071	0.906630568	2.861175184	0.729702582
BS336320	Motor vehicle electrical and electronic equipment manufacturing	2.52787	0.908437664	2.774152404	0.737247737
BS336330	Motor vehicle steering and suspension components (except spring) manufacturing	2.48638	0.939116685	3.076573432	0.786517643
BS336340	Motor vehicle brake system manufacturing	2.26303	0.919908435	2.829692166	0.725955221
BS336350	Motor vehicle transmission and power train parts manufacturing	2.48932	0.934847973	2.636885647	0.659609225
BS336360	Motor vehicle seating and interior trim manufacturing	2.53551	0.947137853	3.129844292	0.809704827
BS336370	Motor vehicle metal stamping	2.59763	0.991256676	2.769744699	0.712649046
BS336390	Other motor vehicle parts manufacturing	2.37732	0.925488975	2.751755099	0.716421567
BS336400	Aerospace product and parts manufacturing	2.58316	0.646780993	2.39078192	0.610241913
BS336500	Railroad rolling stock manufacturing	1.52757	0.250117615	2.801114614	0.751035187
BS336600	Ship and boat building	1.26187	0.092799363	2.176120496	0.53256977
BS336900	Other transportation equipment manufacturing	2.03099	0.504487509	2.789129145	0.728836282
BS337100	Household and institutional furniture and kitchen cabinet manufacturing	1.46335	0.400750543	2.262457509	0.56838406
BS337200	Office furniture (including fixtures) manufacturing	1.32858	0.229829793	2.356319349	0.594968711
BS337900	Other furniture-related product manufacturing	1.30522	0.225524062	2.519080957	0.663237818
BS339100	Medical equipment and supplies manufacturing	1.88342	0.656696939	2.133692385	0.531620025
BS339900	Other miscellaneous manufacturing	1.76762	0.379777319	2.525858064	0.694439949
BS411000	Farm product merchant wholesalers	2.89497	0.803116684	1.753524288	0.415014905
BS412000	Petroleum and petroleum products merchant wholesalers	2.51291	0.616454317	1.74994155	0.400706552
BS413000	Food, beverage and tobacco merchant wholesalers	1.64728	0.355281633	1.821202165	0.419591088
BS414000	Personal and household goods merchant wholesalers	1.7029	0.4076122	1.840703953	0.433728431
BS415000	Motor vehicle and motor vehicle parts and accessories merchant wholesalers	1.9806	0.507873756	1.865938853	0.440424516
BS416000	Building material and supplies merchant wholesalers	2.46651	0.828991806	1.575480114	0.29971945
BS417000	Machinery, equipment and supplies merchant wholesalers	2.31708	0.566740569	1.562379074	0.290621052
BS418000	Miscellaneous merchant wholesalers	2.88775	0.733366238	1.826596962	0.424802046
BS419000	Business-to-business electronic markets, and agents and brokers	2.47143	0.739031648	1.634032624	0.340042067
BS441000	Motor vehicle and parts dealers	1.39989	0.17406543	1.639248842	0.337851227
BS442000	Furniture and home furnishings stores	1.43283	0.255674906	1.918279415	0.486682532
BS443000	Electronics and appliance stores	1.55112	0.297288159	1.709057223	0.378098548
BS444000	Building material and garden equipment and supplies dealers	2.04877	0.56380019	1.682407146	0.358689092
BS445000	Food and beverage stores	1.23062	0.133114605	1.674768231	0.362325605
BS446000	Health and personal care stores	1.39705	0.232890629	1.720615386	0.387475663
BS447000	Gasoline stations	1.55218	0.2497332	1.588268999	0.317843373
BS448000	Clothing and clothing accessories stores	1.15881	0.066049104	1.85815063	0.465271706
BS451000	Sporting goods, hobby, book and music stores	1.26982	0.130308346	1.746205053	0.406196284
BS452000	General merchandise stores	1.42567	0.222506395	1.675254546	0.367262788
BS453A00	Miscellaneous store retailers (except cannabis)	1.36882	0.192945546	1.787500717	0.42982697
BS453BL0	Cannabis stores (licensed)			1.820989077
BS453BU0	Cannabis stores (unlicensed)	1	0	1.352668471	0.206741067
BS454000	Non-store retailers	1.46057	0.235617947	1.890797742	0.478389428
BS481000	Air transportation	1.95024	0.487813057	2.30948197	0.58848278
BS482000	Rail transportation	3.01809	0.54654162	1.752637656	0.360777974
BS483000	Water transportation	3.46182	0.648451212	2.704830246	0.756164111
BS484000	Truck transportation	3.19648	0.69689718	2.328688361	0.598417887
BS485100	Urban transit systems	1.08014	0.052172155	2.903289846	0.91967832
BS48A000	Other transit and ground passenger transportation and scenic and sightseeing transportation	1.76155	0.661515346	1.809103495	0.38002427
BS485300	Taxi and limousine service	2.20508	0.650500758	2.146842796	0.556342303
BS486A00	Crude oil and other pipeline transportation	3.5854	0.499869961	1.689886804	0.36558455
BS486200	Pipeline transportation of natural gas	2.87248	0.377655607	1.419194066	0.215644752
BS488000	Support activities for transportation	3.19821	0.78354627	2.154839161	0.553109719
BS491000	Postal service	2.51219	0.705066776	1.417474204	0.194556121
BS492000	Couriers and messengers	2.98069	0.960679492	2.129607634	0.541253028
BS493000	Warehousing and storage	3.23033	0.80892483	1.5333092	0.273378034
BS511110	Newspaper publishers	2.54536	0.755329178	2.061483076	0.527453406
BS5111A0	Periodical, book and directory publishers	2.31608	1.155199807	2.071828794	0.543855868
BS511200	Software publishers	2.34452	0.519148547	1.706947223	0.380192502
BS5121A0	Motion picture and video industries (except exhibition)	3.61875	0.775933191	2.696701808	0.745445916
BS512130	Motion picture and video exhibition	1.25856	0.126889321	2.341236859	0.576274083
BS512200	Sound recording industries	2.72307	0.984549323	2.392154459	0.640161648
BS515100	Radio and television broadcasting	2.73187	1.006577513	2.234310871	0.55655593
BS515200	Pay and specialty television	2.86044	1.060936238	2.583441926	0.656984661
BS517000	Telecommunications	1.84645	0.409374371	1.861802768	0.427685516
BS518000	Data processing, hosting, and related services	3.09578	0.893971268	1.789436257	0.418163231
BS519000	Other information services	2.99971	1.192138429	1.971091502	0.518313249
BS521000	Monetary authorities - central bank	2.15585	0.993047101	1.604028931	0.322721298
BS5221A0	Banking and other depository credit intermediation	2.20166	0.598456279	1.436930524	0.240494607
BS522130	Local credit unions	1.85705	0.445764937	1.76063294	0.420856283
BS522200	Non-depository credit intermediation	2.27577	0.613912758	2.08619845	0.606746645
BS522300	Activities related to credit intermediation	2.60561	0.81085757	1.908661463	0.492972294
BS52A000	Financial investment services, funds and other financial vehicles	1.98227	0.49002267	2.22700826	0.653004136
BS524100	Insurance carriers	2.03895	0.547296513	1.967532972	0.574397401
BS524200	Agencies, brokerages and other insurance related activities	3.03303	0.974842098	1.602554508	0.33132397
BS531100	Lessors of real estate	1.76179	0.40841503	1.678017963	0.369206791
BS531A00	Offices of real estate agents and brokers and activities related to real estate	1.37703	0.231783141	1.964468427	0.502148784
BS5311A0	Owner-occupied dwellings	1	0	1.215401435	0.122080862
BS532100	Automotive equipment rental and leasing	2.88867	0.977662466	1.764119001	0.401866434
BS532A00	Rental and leasing services (except automotive equipment)	2.69761	0.807365708	1.835006813	0.419450897
BS533000	Lessors of non-financial intangible assets (except copyrighted works)	3.09108	3.049211852	1.941166946	0.521938752
BS541100	Legal services	2.45426	0.72171341	1.541439867	0.289273411
BS541200	Accounting, tax preparation, bookkeeping and payroll services	2.81896	0.897981519	1.604330053	0.318392534
BS541300	Architectural, engineering and related services	2.67258	0.967744554	1.669929512	0.356896726
BS541400	Specialized design services	2.80082	0.949299106	1.851879236	0.451220537
BS541500	Computer systems design and related services	2.48313	0.637284265	1.72579918	0.399505567
BS541600	Management, scientific and technical consulting services	3.08791	0.892846537	1.660759332	0.350022542
BS541700	Scientific research and development services	1.54558	0.20082742	1.634196153	0.336771417
BS541800	Advertising, public relations, and related services	2.90584	0.921044936	1.851970518	0.448446391
BS541900	Other professional, scientific and technical services	2.7155	0.797845583	1.875023397	0.465750418
BS551113	Holding companies	3.38224	0.964763656	1.792994435	0.443252248
BS561100	Office administrative services	3.11086	1.639374276	1.601466842	0.322621925
BS561A00	Facilities and other support services	3.20338	0.995328841	1.684076754	0.36490546
BS561300	Employment services	3.32399	0.990467108	1.698223067	0.373533893
BS561400	Business support services	3.03609	1.084908944	1.692571305	0.367776489
BS561500	Travel arrangement and reservation services	2.32563	0.736381703	1.778639749	0.414473485
BS561600	Investigation and security services	2.65099	0.921963714	1.55436338	0.287642523
BS561700	Services to buildings and dwellings	2.801	0.899334499	1.926784769	0.448442605
BS562000	Waste management and remediation services	2.76827	0.902947307	1.857747883	0.451007329
BS610000	Educational services	1.92258	0.524105768	1.841697227	0.440449893
BS621100	Offices of physicians	2.09608	0.9835059	1.494762218	0.257185631
BS621200	Offices of dentists	1.07534	0.066061147	1.691738543	0.353260567
BS621A00	Miscellaneous ambulatory health care services	1.69141	0.623246319	1.345311676	0.17983782
BS623000	Nursing and residential care facilities	1.28013	0.251870807	1.441182723	0.230530307
BS624000	Social assistance	1.32435	0.277415675	1.931823438	0.466691114
BS71A000	Performing arts, spectator sports and related industries, and heritage institutions	2.12225	0.538986952	1.980222175	0.489210606
BS713A00	Amusement and recreation industries	1.50509	0.278792079	1.935106969	0.48010444
BS713200	Gambling industries	1.25842	0.128331573	2.141099927	0.597420039
BS721100	Traveller accommodation	1.86283	0.518676216	1.868365846	0.40929875
BS721A00	Recreational vehicle (RV) parks, recreational camps, and rooming and boarding houses	1.1811	0.116289441	1.782818916	0.393241659
BS722000	Food services and drinking places	1.45553	0.229250199	2.262937238	0.540029084
BS811100	Automotive repair and maintenance	1.91656	0.389684381	1.820699746	0.406501945
BS811A00	Repair and maintenance (except automotive)	2.52193	0.726240764	1.768311236	0.383878141
BS812A00	Personal care services and other personal services	1.35812	0.161119265	1.868668497	0.43899018
BS812200	Funeral services	1.06179	0.047004081	1.980366254	0.454976237
BS812300	Dry cleaning and laundry services	1.78082	0.509293855	1.958142369	0.463057495
BS813000	Business, professional and other membership organizations	2.95258	0.977221803	1.77965562	0.38368481
BS814000	Private households	1	0	1	0
NP610000	Educational services	1.18703	0.126483638	1.655733356	0.327057167
NP621000	Ambulatory health care services	1.32512	0.283321672	1.807066957	0.4048132
NP624000	Social assistance	1.20588	0.175774291	1.565681522	0.287851078
NP710000	Arts, entertainment and recreation	1.24048	0.125819368	1.999244994	0.50534606
NP813100	Religious organizations	1.15096	0.076551751	1.555470617	0.284632016
NP813A00	Grant-making, civic, and professional and similar organizations	1.09657	0.059021834	1.998799495	0.531233744
NP999999	Other non-profit institutions serving households	1.12725	0.081604548	2.188129558	0.634448518
GS611100	Elementary and secondary schools	1.01998	0.014069625	1.252170529	0.13299835
GS611200	Community colleges and C.E.G.E.P.s	1.16423	0.103047809	1.480707868	0.251106038
GS611300	Universities	1.10948	0.052225301	1.405546929	0.19993684
GS611A00	Other educational services	1	0	1.550903949	0.28885979
GS622000	Hospitals	1.10419	0.075402464	1.639294664	0.351159949
GS623000	Nursing and residential care facilities	1.04679	0.042069273	1.415687669	0.216036435
GS911100	Defence services	1.01432	0.007478455	1.642387843	0.331350558
GS911A00	Other federal government services (except defence)	1.16395	0.090116787	1.64653794	0.347463887
GS912000	Other provincial and territorial government services	1.11687	0.065211948	2.080145417	0.621436637
GS913000	Other municipal government services	1.26945	0.142839285	1.608579275	0.320798035
GS914000	Other aboriginal government services	1	0	1.967790869	0.510970785
PRM100000	Taxes on products	#N/A	#N/A	#N/A	#N/A
PRM200000	Subsidies on products	#N/A	#N/A	#N/A	#N/A
PRM300000	Subsidies on production	#N/A	#N/A	#N/A	#N/A
PRM400000	Taxes on production	#N/A	#N/A	#N/A	#N/A
PRM500000	Wages and salaries	#N/A	#N/A	#N/A	#N/A
PRM600000	Employers' social contributions	#N/A	#N/A	#N/A	#N/A
PRM700000	Gross mixed income	#N/A	#N/A	#N/A	#N/A
PRM800000	Gross operating surplus	#N/A	#N/A	#N/A	#N/A
TOTAL	Total	#N/A	#N/A	#N/A	#N/A

  The simple measures have a tight correlation with the complex measures of upstreamness and downstreamness—particularly on the downstream side. On the upstream side, there is more variability, especially for the industries that are the most upstream (those that produce almost exclusively intermediate goods for other industries). This is because a greater share of the outputs will be influenced by the upstreamness of other industries—thus allowing for greater variation in their own upstreamness.

The high degree of correlation between the simple measures and the complex measures is expected and a beneficial feature which serves as a check on the theory and calculations; if there was a major divergence, it could be debated as to which measure is better and thus lead to arbitrary conclusions. However, since they lead to similar findings, the choice of measure is less important. Given that the complex measures attempt to treat the intermediate goods heterogeneously, which accounts for variable supply chain length, these will be the measure of choice for the rest of the paper. Next, figure 5 has the complex measures plotted against one another.

Figure 5 - Text version

Industry	Description	Upstream	Domestic Downstream	Overall upstreamness	Overall downstreamness
BS111A00	Crop production (except cannabis, greenhouse, nursery and floriculture production)	2.871669586	2.150010079	2.06556106	1.962935314
BS1114A0	Greenhouse, nursery and floriculture production (except cannabis)	1.709198407	2.125945159	2.06556106	1.962935314
BS111CL0	Cannabis production (licensed)	1	2.120325405	2.06556106	1.962935314
BS111CU0	Cannabis production (unlicensed)	1	1.517747591	2.06556106	1.962935314
BS112A00	Animal production (except aquaculture)	2.849559474	2.940425936	2.06556106	1.962935314
BS112500	Aquaculture	2.280099642	2.395199977	2.06556106	1.962935314
BS113000	Forestry and logging	3.948059902	2.335320097	2.06556106	1.962935314
BS114000	Fishing, hunting and trapping	2.264604035	1.897825183	2.06556106	1.962935314
BS115A00	Support activities for crop and animal production	3.774453536	1.464160797	2.06556106	1.962935314
BS115300	Support activities for forestry	4.518608956	2.065102171	2.06556106	1.962935314
BS211110	Oil and gas extraction (except oil sands)	3.548858026	1.886295564	2.06556106	1.962935314
BS211140	Oil sands extraction	3.722869939	1.849122489	2.06556106	1.962935314
BS212100	Coal mining	4.121720706	1.651248943	2.06556106	1.962935314
BS212210	Iron ore mining	4.803448779	1.776463324	2.06556106	1.962935314
BS212220	Gold and silver ore mining	6.087688845	1.911010169	2.06556106	1.962935314
BS212230	Copper, nickel, lead and zinc ore mining	6.724092658	1.733903333	2.06556106	1.962935314
BS212290	Other metal ore mining	6.409186794	1.883497087	2.06556106	1.962935314
BS212310	Stone mining and quarrying	2.939429669	1.866465849	2.06556106	1.962935314
BS212320	Sand, gravel, clay, and ceramic and refractory minerals mining and quarrying	3.352696932	1.857648281	2.06556106	1.962935314
BS212392	Diamond mining	2.826472063	2.025182828	2.06556106	1.962935314
BS21239A	Other non-metallic mineral mining and quarrying (except diamond and potash)	3.811747552	1.744938287	2.06556106	1.962935314
BS212396	Potash mining	3.672204916	1.54632371	2.06556106	1.962935314
BS21311A	Support activities for oil and gas extraction	3.492675473	1.732959304	2.06556106	1.962935314
BS21311B	Support activities for mining	2.845539791	1.856345719	2.06556106	1.962935314
BS221100	Electric power generation, transmission and distribution	2.548957339	1.576984444	2.06556106	1.962935314
BS221200	Natural gas distribution	2.322205724	1.356094227	2.06556106	1.962935314
BS221300	Water, sewage and other systems	2.444678935	1.941724375	2.06556106	1.962935314
BS23A000	Residential building construction	1	2.198079061	2.06556106	1.962935314
BS23B000	Non-residential building construction	1	2.120154094	2.06556106	1.962935314
BS23C100	Transportation engineering construction	1	2.25845308	2.06556106	1.962935314
BS23C200	Oil and gas engineering construction	1	2.145927349	2.06556106	1.962935314
BS23C300	Electric power engineering construction	1	1.995496551	2.06556106	1.962935314
BS23C400	Communication engineering construction	1	2.570571686	2.06556106	1.962935314
BS23C500	Other engineering construction	1	2.21185182	2.06556106	1.962935314
BS23D000	Repair construction	2.804972676	1.901220533	2.06556106	1.962935314
BS23E000	Other activities of the construction industry	2.960309147	1.637890012	2.06556106	1.962935314
BS311100	Animal food manufacturing	2.910059612	2.933252409	2.06556106	1.962935314
BS311200	Grain and oilseed milling	2.545386598	2.796831427	2.06556106	1.962935314
BS311300	Sugar and confectionery product manufacturing	1.608016606	2.60080699	2.06556106	1.962935314
BS311400	Fruit and vegetable preserving and specialty food manufacturing	1.415081648	2.53486352	2.06556106	1.962935314
BS311500	Dairy product manufacturing	1.593172297	3.082729574	2.06556106	1.962935314
BS311600	Meat product manufacturing	1.823403086	3.113859365	2.06556106	1.962935314
BS311700	Seafood product preparation and packaging	1.963121493	2.697514953	2.06556106	1.962935314
BS311800	Bakeries and tortilla manufacturing	1.457729039	2.440895961	2.06556106	1.962935314
BS311900	Other food manufacturing	1.591426888	2.502667711	2.06556106	1.962935314
BS312110	Soft drink and ice manufacturing	1.710741141	2.499030813	2.06556106	1.962935314
BS312120	Breweries	1.813775075	2.066971983	2.06556106	1.962935314
BS3121A0	Wineries and distilleries	1.421661381	2.215379337	2.06556106	1.962935314
BS312200	Tobacco manufacturing	1.177903497	1.893090626	2.06556106	1.962935314
BS31A000	Textile and textile product mills	2.037599248	2.496560857	2.06556106	1.962935314
BS31B000	Clothing and leather and allied product manufacturing	1.247073258	2.359811541	2.06556106	1.962935314
BS321100	Sawmills and wood preservation	3.102520976	2.566857543	2.06556106	1.962935314
BS321200	Veneer, plywood and engineered wood product manufacturing	2.46068603	2.451314013	2.06556106	1.962935314
BS321900	Other wood product manufacturing	2.462875103	2.486967992	2.06556106	1.962935314
BS322100	Pulp, paper and paperboard mills	3.456095659	2.558739891	2.06556106	1.962935314
BS322200	Converted paper product manufacturing	2.570629871	2.596356817	2.06556106	1.962935314
BS323000	Printing and related support activities	2.882497376	2.241681632	2.06556106	1.962935314
BS324110	Petroleum refineries	2.80772147	2.566917036	2.06556106	1.962935314
BS3241A0	Petroleum and coal product manufacturing (except petroleum refineries)	3.283314285	2.786718736	2.06556106	1.962935314
BS325100	Basic chemical manufacturing	4.204689324	2.488762295	2.06556106	1.962935314
BS325200	Resin, synthetic rubber, and artificial and synthetic fibres and filaments manufacturing	3.811661053	2.562114436	2.06556106	1.962935314
BS325300	Pesticide, fertilizer and other agricultural chemical manufacturing	3.743469305	2.485909975	2.06556106	1.962935314
BS325400	Pharmaceutical and medicine manufacturing	2.05504125	2.447699201	2.06556106	1.962935314
BS325500	Paint, coating and adhesive manufacturing	2.994394964	2.584384321	2.06556106	1.962935314
BS325600	Soap, cleaning compound and toilet preparation manufacturing	1.667514809	2.433349647	2.06556106	1.962935314
BS325900	Other chemical product manufacturing	3.488062815	2.602346746	2.06556106	1.962935314
BS326100	Plastic product manufacturing	2.840542542	2.519050813	2.06556106	1.962935314
BS326200	Rubber product manufacturing	2.991974182	2.627688216	2.06556106	1.962935314
BS327A00	Non-metallic mineral product manufacturing (except cement and concrete products)	2.637976331	2.273767442	2.06556106	1.962935314
BS327300	Cement and concrete product manufacturing	2.47102323	2.30832111	2.06556106	1.962935314
BS331100	Iron and steel mills and ferro-alloy manufacturing	3.774528249	2.702303094	2.06556106	1.962935314
BS331200	Steel product manufacturing from purchased steel	3.798073896	2.857345957	2.06556106	1.962935314
BS331300	Alumina and aluminum production and processing	4.198638824	2.763510789	2.06556106	1.962935314
BS331400	Non-ferrous metal (except aluminum) production and processing	5.923020948	3.11642242	2.06556106	1.962935314
BS331500	Foundries	3.566816041	2.461208673	2.06556106	1.962935314
BS332100	Forging and stamping	3.196309725	2.630405704	2.06556106	1.962935314
BS332A00	Cutlery, hand tools and other fabricated metal product manufacturing	2.665619086	2.228201014	2.06556106	1.962935314
BS332300	Architectural and structural metals manufacturing	2.68622234	2.441413667	2.06556106	1.962935314
BS332400	Boiler, tank and shipping container manufacturing	2.62977502	2.449282504	2.06556106	1.962935314
BS332500	Hardware manufacturing	2.822754643	2.470517839	2.06556106	1.962935314
BS332600	Spring and wire product manufacturing	2.595104375	2.558657443	2.06556106	1.962935314
BS332700	Machine shops, turned product, and screw, nut and bolt manufacturing	2.934801693	2.161600909	2.06556106	1.962935314
BS332800	Coating, engraving, cold and heat treating and allied activities	3.636829526	2.104835866	2.06556106	1.962935314
BS333100	Agricultural, construction and mining machinery manufacturing	2.421038798	2.387369859	2.06556106	1.962935314
BS333200	Industrial machinery manufacturing	1.73415123	2.287559593	2.06556106	1.962935314
BS333300	Commercial and service industry machinery manufacturing	1.926527383	2.247375686	2.06556106	1.962935314
BS333400	Ventilation, heating, air-conditioning and commercial refrigeration equipment manufacturing	2.258881382	2.316387952	2.06556106	1.962935314
BS333500	Metalworking machinery manufacturing	2.295439809	2.266625337	2.06556106	1.962935314
BS333600	Engine, turbine and power transmission equipment manufacturing	2.913694266	2.280605334	2.06556106	1.962935314
BS333900	Other general-purpose machinery manufacturing	2.313057741	2.427463106	2.06556106	1.962935314
BS334100	Computer and peripheral equipment manufacturing	1.294833594	2.308542238	2.06556106	1.962935314
BS334200	Communications equipment manufacturing	1.922056612	2.226227859	2.06556106	1.962935314
BS334A00	Other electronic product manufacturing	2.080935294	2.178415939	2.06556106	1.962935314
BS334400	Semiconductor and other electronic component manufacturing	3.220355508	2.366357248	2.06556106	1.962935314
BS335100	Electric lighting equipment manufacturing	2.183962244	2.301171752	2.06556106	1.962935314
BS335200	Household appliance manufacturing	1.317590725	2.549413434	2.06556106	1.962935314
BS335300	Electrical equipment manufacturing	2.743446768	2.503882283	2.06556106	1.962935314
BS335900	Other electrical equipment and component manufacturing	2.640906636	2.697279255	2.06556106	1.962935314
BS336110	Automobile and light-duty motor vehicle manufacturing	1.022271973	3.284526338	2.06556106	1.962935314
BS336120	Heavy-duty truck manufacturing	1.062250899	2.983044993	2.06556106	1.962935314
BS336200	Motor vehicle body and trailer manufacturing	1.212265364	2.732421639	2.06556106	1.962935314
BS336310	Motor vehicle gasoline engine and engine parts manufacturing	2.400707743	2.861175184	2.06556106	1.962935314
BS336320	Motor vehicle electrical and electronic equipment manufacturing	2.527866394	2.774152404	2.06556106	1.962935314
BS336330	Motor vehicle steering and suspension components (except spring) manufacturing	2.486383143	3.076573432	2.06556106	1.962935314
BS336340	Motor vehicle brake system manufacturing	2.263026827	2.829692166	2.06556106	1.962935314
BS336350	Motor vehicle transmission and power train parts manufacturing	2.489324265	2.636885647	2.06556106	1.962935314
BS336360	Motor vehicle seating and interior trim manufacturing	2.535510496	3.129844292	2.06556106	1.962935314
BS336370	Motor vehicle metal stamping	2.597634519	2.769744699	2.06556106	1.962935314
BS336390	Other motor vehicle parts manufacturing	2.377320417	2.751755099	2.06556106	1.962935314
BS336400	Aerospace product and parts manufacturing	2.583158814	2.39078192	2.06556106	1.962935314
BS336500	Railroad rolling stock manufacturing	1.527567411	2.801114614	2.06556106	1.962935314
BS336600	Ship and boat building	1.261869507	2.176120496	2.06556106	1.962935314
BS336900	Other transportation equipment manufacturing	2.030994153	2.789129145	2.06556106	1.962935314
BS337100	Household and institutional furniture and kitchen cabinet manufacturing	1.463347508	2.262457509	2.06556106	1.962935314
BS337200	Office furniture (including fixtures) manufacturing	1.328584824	2.356319349	2.06556106	1.962935314
BS337900	Other furniture-related product manufacturing	1.305223216	2.519080957	2.06556106	1.962935314
BS339100	Medical equipment and supplies manufacturing	1.883415404	2.133692385	2.06556106	1.962935314
BS339900	Other miscellaneous manufacturing	1.767623186	2.525858064	2.06556106	1.962935314
BS411000	Farm product merchant wholesalers	2.894967796	1.753524288	2.06556106	1.962935314
BS412000	Petroleum and petroleum products merchant wholesalers	2.51290741	1.74994155	2.06556106	1.962935314
BS413000	Food, beverage and tobacco merchant wholesalers	1.647277094	1.821202165	2.06556106	1.962935314
BS414000	Personal and household goods merchant wholesalers	1.702904199	1.840703953	2.06556106	1.962935314
BS415000	Motor vehicle and motor vehicle parts and accessories merchant wholesalers	1.98059621	1.865938853	2.06556106	1.962935314
BS416000	Building material and supplies merchant wholesalers	2.466513015	1.575480114	2.06556106	1.962935314
BS417000	Machinery, equipment and supplies merchant wholesalers	2.317079945	1.562379074	2.06556106	1.962935314
BS418000	Miscellaneous merchant wholesalers	2.887750004	1.826596962	2.06556106	1.962935314
BS419000	Business-to-business electronic markets, and agents and brokers	2.471433216	1.634032624	2.06556106	1.962935314
BS441000	Motor vehicle and parts dealers	1.399894983	1.639248842	2.06556106	1.962935314
BS442000	Furniture and home furnishings stores	1.432829437	1.918279415	2.06556106	1.962935314
BS443000	Electronics and appliance stores	1.551116871	1.709057223	2.06556106	1.962935314
BS444000	Building material and garden equipment and supplies dealers	2.048772405	1.682407146	2.06556106	1.962935314
BS445000	Food and beverage stores	1.230620485	1.674768231	2.06556106	1.962935314
BS446000	Health and personal care stores	1.397047685	1.720615386	2.06556106	1.962935314
BS447000	Gasoline stations	1.552177554	1.588268999	2.06556106	1.962935314
BS448000	Clothing and clothing accessories stores	1.158808069	1.85815063	2.06556106	1.962935314
BS451000	Sporting goods, hobby, book and music stores	1.269824633	1.746205053	2.06556106	1.962935314
BS452000	General merchandise stores	1.425668876	1.675254546	2.06556106	1.962935314
BS453A00	Miscellaneous store retailers (except cannabis)	1.368815936	1.787500717	2.06556106	1.962935314
BS453BL0	Cannabis stores (licensed)	1	1.820989077	2.06556106	1.962935314
BS453BU0	Cannabis stores (unlicensed)	1	1.352668471	2.06556106	1.962935314
BS454000	Non-store retailers	1.460567251	1.890797742	2.06556106	1.962935314
BS481000	Air transportation	1.950236154	2.30948197	2.06556106	1.962935314
BS482000	Rail transportation	3.018089118	1.752637656	2.06556106	1.962935314
BS483000	Water transportation	3.461818749	2.704830246	2.06556106	1.962935314
BS484000	Truck transportation	3.196480747	2.328688361	2.06556106	1.962935314
BS485100	Urban transit systems	1.08013538	2.903289846	2.06556106	1.962935314
BS48A000	Other transit and ground passenger transportation and scenic and sightseeing transportation	1.761551649	1.809103495	2.06556106	1.962935314
BS485300	Taxi and limousine service	2.205081868	2.146842796	2.06556106	1.962935314
BS486A00	Crude oil and other pipeline transportation	3.585395249	1.689886804	2.06556106	1.962935314
BS486200	Pipeline transportation of natural gas	2.872476919	1.419194066	2.06556106	1.962935314
BS488000	Support activities for transportation	3.198210098	2.154839161	2.06556106	1.962935314
BS491000	Postal service	2.512193713	1.417474204	2.06556106	1.962935314
BS492000	Couriers and messengers	2.980693729	2.129607634	2.06556106	1.962935314
BS493000	Warehousing and storage	3.23032505	1.5333092	2.06556106	1.962935314
BS511110	Newspaper publishers	2.545360764	2.061483076	2.06556106	1.962935314
BS5111A0	Periodical, book and directory publishers	2.316081105	2.071828794	2.06556106	1.962935314
BS511200	Software publishers	2.344516098	1.706947223	2.06556106	1.962935314
BS5121A0	Motion picture and video industries (except exhibition)	3.618746545	2.696701808	2.06556106	1.962935314
BS512130	Motion picture and video exhibition	1.258560694	2.341236859	2.06556106	1.962935314
BS512200	Sound recording industries	2.72306772	2.392154459	2.06556106	1.962935314
BS515100	Radio and television broadcasting	2.731872333	2.234310871	2.06556106	1.962935314
BS515200	Pay and specialty television	2.860444453	2.583441926	2.06556106	1.962935314
BS517000	Telecommunications	1.846446095	1.861802768	2.06556106	1.962935314
BS518000	Data processing, hosting, and related services	3.095780466	1.789436257	2.06556106	1.962935314
BS519000	Other information services	2.999711712	1.971091502	2.06556106	1.962935314
BS521000	Monetary authorities - central bank	2.155852638	1.604028931	2.06556106	1.962935314
BS5221A0	Banking and other depository credit intermediation	2.201661669	1.436930524	2.06556106	1.962935314
BS522130	Local credit unions	1.857049179	1.76063294	2.06556106	1.962935314
BS522200	Non-depository credit intermediation	2.275768202	2.08619845	2.06556106	1.962935314
BS522300	Activities related to credit intermediation	2.605612524	1.908661463	2.06556106	1.962935314
BS52A000	Financial investment services, funds and other financial vehicles	1.982273415	2.22700826	2.06556106	1.962935314
BS524100	Insurance carriers	2.038949873	1.967532972	2.06556106	1.962935314
BS524200	Agencies, brokerages and other insurance related activities	3.033028752	1.602554508	2.06556106	1.962935314
BS531100	Lessors of real estate	1.761790219	1.678017963	2.06556106	1.962935314
BS531A00	Offices of real estate agents and brokers and activities related to real estate	1.377028292	1.964468427	2.06556106	1.962935314
BS5311A0	Owner-occupied dwellings	1	1.215401435	2.06556106	1.962935314
BS532100	Automotive equipment rental and leasing	2.88867017	1.764119001	2.06556106	1.962935314
BS532A00	Rental and leasing services (except automotive equipment)	2.697612379	1.835006813	2.06556106	1.962935314
BS533000	Lessors of non-financial intangible assets (except copyrighted works)	3.091084577	1.941166946	2.06556106	1.962935314
BS541100	Legal services	2.454257133	1.541439867	2.06556106	1.962935314
BS541200	Accounting, tax preparation, bookkeeping and payroll services	2.818959098	1.604330053	2.06556106	1.962935314
BS541300	Architectural, engineering and related services	2.672584742	1.669929512	2.06556106	1.962935314
BS541400	Specialized design services	2.800818399	1.851879236	2.06556106	1.962935314
BS541500	Computer systems design and related services	2.483125772	1.72579918	2.06556106	1.962935314
BS541600	Management, scientific and technical consulting services	3.087912365	1.660759332	2.06556106	1.962935314
BS541700	Scientific research and development services	1.545582645	1.634196153	2.06556106	1.962935314
BS541800	Advertising, public relations, and related services	2.905839895	1.851970518	2.06556106	1.962935314
BS541900	Other professional, scientific and technical services	2.715500181	1.875023397	2.06556106	1.962935314
BS551113	Holding companies	3.382238162	1.792994435	2.06556106	1.962935314
BS561100	Office administrative services	3.110863264	1.601466842	2.06556106	1.962935314
BS561A00	Facilities and other support services	3.203380793	1.684076754	2.06556106	1.962935314
BS561300	Employment services	3.323992394	1.698223067	2.06556106	1.962935314
BS561400	Business support services	3.036092668	1.692571305	2.06556106	1.962935314
BS561500	Travel arrangement and reservation services	2.325628698	1.778639749	2.06556106	1.962935314
BS561600	Investigation and security services	2.650985996	1.55436338	2.06556106	1.962935314
BS561700	Services to buildings and dwellings	2.800997596	1.926784769	2.06556106	1.962935314
BS562000	Waste management and remediation services	2.768269943	1.857747883	2.06556106	1.962935314
BS610000	Educational services	1.922583551	1.841697227	2.06556106	1.962935314
BS621100	Offices of physicians	2.096082738	1.494762218	2.06556106	1.962935314
BS621200	Offices of dentists	1.075342466	1.691738543	2.06556106	1.962935314
BS621A00	Miscellaneous ambulatory health care services	1.691407954	1.345311676	2.06556106	1.962935314
BS623000	Nursing and residential care facilities	1.280134014	1.441182723	2.06556106	1.962935314
BS624000	Social assistance	1.324351087	1.931823438	2.06556106	1.962935314
BS71A000	Performing arts, spectator sports and related industries, and heritage institutions	2.122253723	1.980222175	2.06556106	1.962935314
BS713A00	Amusement and recreation industries	1.505086366	1.935106969	2.06556106	1.962935314
BS713200	Gambling industries	1.258424792	2.141099927	2.06556106	1.962935314
BS721100	Traveller accommodation	1.862827496	1.868365846	2.06556106	1.962935314
BS721A00	Recreational vehicle (RV) parks, recreational camps, and rooming and boarding houses	1.18109991	1.782818916	2.06556106	1.962935314
BS722000	Food services and drinking places	1.455530141	2.262937238	2.06556106	1.962935314
BS811100	Automotive repair and maintenance	1.916562472	1.820699746	2.06556106	1.962935314
BS811A00	Repair and maintenance (except automotive)	2.521933581	1.768311236	2.06556106	1.962935314
BS812A00	Personal care services and other personal services	1.358121266	1.868668497	2.06556106	1.962935314
BS812200	Funeral services	1.061785917	1.980366254	2.06556106	1.962935314
BS812300	Dry cleaning and laundry services	1.780817209	1.958142369	2.06556106	1.962935314
BS813000	Business, professional and other membership organizations	2.95257563	1.77965562	2.06556106	1.962935314
BS814000	Private households			2.06556106	1.962935314
NP610000	Educational services	1.18702669	1.655733356	2.06556106	1.962935314
NP621000	Ambulatory health care services	1.325115344	1.807066957	2.06556106	1.962935314
NP624000	Social assistance	1.205883381	1.565681522	2.06556106	1.962935314
NP710000	Arts, entertainment and recreation	1.240480382	1.999244994	2.06556106	1.962935314
NP813100	Religious organizations	1.150958985	1.555470617	2.06556106	1.962935314
NP813A00	Grant-making, civic, and professional and similar organizations	1.096574213	1.998799495	2.06556106	1.962935314
NP999999	Other non-profit institutions serving households	1.127254852	2.188129558	2.06556106	1.962935314
GS611100	Elementary and secondary schools	1.019980799	1.252170529	2.06556106	1.962935314
GS611200	Community colleges and C.E.G.E.P.s	1.164229511	1.480707868	2.06556106	1.962935314
GS611300	Universities	1.109475349	1.405546929	2.06556106	1.962935314
GS611A00	Other educational services	1	1.550903949	2.06556106	1.962935314
GS622000	Hospitals	1.104193127	1.639294664	2.06556106	1.962935314
GS623000	Nursing and residential care facilities	1.046789994	1.415687669	2.06556106	1.962935314
GS911100	Defence services	1.01432046	1.642387843	2.06556106	1.962935314
GS911A00	Other federal government services (except defence)	1.163945433	1.64653794	2.06556106	1.962935314
GS912000	Other provincial and territorial government services	1.116873546	2.080145417	2.06556106	1.962935314
GS913000	Other municipal government services	1.269447966	1.608579275	2.06556106	1.962935314
GS914000	Other aboriginal government services	1	1.967790869	2.06556106	1.962935314
PRM100000	Taxes on products	#N/A	#N/A	2.06556106	1.962935314
PRM200000	Subsidies on products	#N/A	#N/A	2.06556106	1.962935314
PRM300000	Subsidies on production	#N/A	#N/A	2.06556106	1.962935314
PRM400000	Taxes on production	#N/A	#N/A	2.06556106	1.962935314
PRM500000	Wages and salaries	#N/A	#N/A	2.06556106	1.962935314
PRM600000	Employers' social contributions	#N/A	#N/A	2.06556106	1.962935314
PRM700000	Gross mixed income	#N/A	#N/A	2.06556106	1.962935314
PRM800000	Gross operating surplus	#N/A	#N/A	2.06556106	1.962935314
TOTAL	Total	#N/A	#N/A	2.06556106	1.962935314

The average GVC positioning for all Canadian industries has an upstream measure of 2.07, and a downstream measure of 1.96.^{Footnote 6} This means that on average, it takes four production stages to convert value-added activities into final goods. While the averages are quite similar for upstreamness and downstreamness—as stated earlier, without the open economy adjustment they would be the exact same—the upstream measure has a wider spread than the downstream measure. The weighted standard deviation for upstreamness is just over 1, while the standard deviation for downstreamness is just over 0.2.^{Footnote 7}

There is a wider spread on the outputs produced by industries (the upstreamness) than the inputs consumed by industries (the downstreamness). Some industries produce outputs that will, on average, take 6 to 7 more steps to become final goods; however, other industries will produce outputs that are consumed right away. The inputs to industries have a shorter span. No industry has only value added activities, thus no industry has a downstream score of exactly one; concurrently, no industry has inputs with a downstream score higher than 3.3.^{Footnote 8} The reason is that industries are classified by what is produced, rather than what is consumed. Some industries will produce a single good or service, which leads to some industries having more extreme upstream scores. However, each industry will use several intermediate inputs, thus creating more centred downstream scores.

The industries that are below and to the left of the average (the large black dot and dashed lines) have the shortest value chains and will have the least spillovers to other industries if they experience a shock or disruption. These are largely services that operate independently of other industries. While the spillovers may be minimal, many of the industries are important for the economy and society, such as hospitals, where most of the inputs are value-added activities and most of the outputs are consumed as final goods. Consequently, a disruption to hospitals is unlikely to have many spillovers into supply chains.^{Footnote 9} Of the 234 industries that are in the IO tables, 50 of them, representing 31.1% of the value in the Canadian economy, are in the bottom left quadrant.  These industries have the shortest GVCs and rank low on both upstreamness and downstreamness.

The industries that are in the bottom-right (below average upstreamness, above average downstreamness) of figure 5 are downstream industries at the end of a GVC. These are mainly manufacturing and construction industries where the inputs have already been through at least two stages of production, and the outputs have fewer than two stages left to produce a final good. The most downstream industries are auto and food manufacturing industries. Both autos and food use inputs that go through multiple stages of production first, and both will have a high proportion of final goods as their output. These industries are vulnerable to supply shocks, and shocks to these industries will cause demand shocks up the value chain. The exact interpretation of downstreamness is the total number of backwards linkages for an industry. The more backwards linkages, the higher the chance to be hit with a supply disruption, and at the same time, create a demand disruption for multiple industries further upstream. For example, if there is a negative supply shock (for example, a semi-conductor shortage) to the automobile and light-duty motor vehicle manufacturing industry, a negative demand shock for various motor-vehicle parts manufacturing industries would ensue. That would have a ripple effect to other manufacturing industries, and eventually a negative demand shock for various upstream industries that process raw materials and mining. Some 53 industries, representing 25.1% of the value in the Canadian economy, are in the bottom right quadrant that are downstream industries.

The industries that are in the upper-left (above average upstreamness, below average downstreamness) of the diagram are upstream industries at the start of a GVC. These include a large variety of industries including natural resources, utilities, and some services industries that have not been through many stages of production, but produce intermediate inputs for other industries. The most upstream industries are in the mining sector. These industries are most vulnerable to demand shocks and if these industries experience a disruption, it causes a supply shock down the value chain. The exact interpretation of the upstreamness measure is the total number of forward linkages. The more forward linkages an industry has, the greater the susceptibility to demand shocks while creating supply shocks for other industries further downstream. For example, a disturbance to copper, nickel, lead, and zinc ore mining will cause the intermediate inputs for other industries to become more expensive and/or scarce; this will, in turn, raise prices throughout the value chain and eventually cause final goods to be more expensive. Some 60 industries, representing 24.9% of the value in the Canadian economy, are upstream industries.

The final section are the industries that are in the upper-right of the diagram; these high-upstream, high-downstream industries sit in the middle of complex GVCs. Disruptions to these industries cause supply effects for some industries and demand effects for others. The industry in the top and to the right of figure 5, non-ferrous metal production and processing, is one of the most downstream industries while at the same time being one of the most upstream. The average input into this industry has already been through 3 stages of production, and the outputs of this industry will go through another 6 stages of production before becoming a final good. An interference in this industry would cause a demand disruption for many of the upstream industries such as mining, while simultaneously causing a supply disruption for manufacturing industries. Some 71 industries, representing 18.9% of the value in the Canadian economy, are both high upstream and high downstream industries.

Interestingly, industries that have both high upstream and high downstream measures have the highest total amount of industries (71), whereas industries that have both low upstream and low downstream measures have the lowest number of industries (50). However, the opposite is true when it comes to value; the high upstream and downstream industries represent 18.9% of the value in the Canadian economy, while the low upstream and downstream industries represent 31.1% of the value in the Canadian economy. The reason is due to the fact that services, which tend to fall into the neither upstream nor downstream category, are some of the largest industries with a wide range of activities. Products and industries tend to be classified by what is produced; this makes for easy classification when it comes to goods industries—it’s easy to differentiate between the process that makes auto and truck suspension components and auto and truck brake components. Thus, many goods industries produce a homogenous good (particularly those at the beginning or end of value chains) and are resultantly smaller. However, for services, what is produced is harder to track since there is no physical good. Consequently, this leads to larger and more heterogeneous industries.^{Footnote 10} Of the 10 largest industries by value, 9 of them are in service industries that would not typically belong to a value chain, or classified as inputs in an IO table (e.g. owner-occupied dwellings, other provincial government services, etc.). This is seen more easily in figure 6, which aggregates from figure 5 using industry share of value.

Figure 6 - Text version

	Upstream	Downstream	Size
Agriculture and forestry	2.82	2.36	$   100,104,894.00
Mining, oil and gas	3.94	1.83	$   207,992,641.00
Utilities	2.52	1.56	$     61,916,376.00
Construction	1.28	2.13	$   354,156,554.00
Manufacturing	2.60	2.64	$   761,246,021.00
Government Services	1.13	1.67	$   591,807,389.00
Services	2.03	1.77	$1,988,161,840.00

Of note, figure 5 and figure 6 have similarities to previous work on supply chain vulnerability done by the Office of the Chief Economist. (Boileau and Sydor, 2020) In their paper, Boileau and Sydor attempt to measure the vulnerability of Canadian industries to potential supply and demand disruptions. While the analyses share some commonalities, the methodologies are distinct. Ultimately the goal of the previous paper was to classify the vulnerability of industries, whereas this paper attempts to classify where industries fit into supply chains.

4. Results for Canadian trade

One of the main tasks of this paper is to characterize how Canada fits into GVCs—does Canada specialize in activities that are closer to the beginning or end of the value chain. In order to make this characterization, Canada’s international trade is examined. To begin, figure 7 displays an aggregate upstream and downstream position for production, exports, and imports, using a weighted average.

Figure 7 - Text version

	Overall upstream	Overall downstream	Import Upstream	Import downstream	Export upstream	Export downstream
Goods share
Goods & Services	2.07	1.95	2.41	2.40	2.77	2.29
Goods	2.88	2.43	2.42	2.51	2.91	2.45
Services	1.77	1.78	2.35	1.89	2.37	1.83
	Production		Imports		Exports

The first noteworthy feature of the above graphs is that Canadian production of goods and services has a much shorter supply chain than Canadian trade. The reason is that services have shorter value chains (see figure 6) and make up the majority of Canadian production. Goods industries^{Footnote 11} make up only 26% of Canadian production even though they make up 83% of imports and 74% of exports. Thus, the increased share of goods makes it natural for Canadian international trade to have longer value chains than Canadian production. Excluding services, Canadian production has a similar supply chain positioning and length to exports.

The next important finding is that Canadian exports are, on average, in the early parts of the supply chain, while Canadian goods imports come from the later stages of the supply chain. Canadian goods exports have an upstream score of 2.91, while having a downstream score of 2.45. This means that, on average, Canadian exports will go through more production stages once they leave the country than they had been through prior to export. Put differently, most of the production stages will happen once the goods leave Canada. Goods imports are more balanced—the downstream score is only 0.1 higher than the upstream score. Although this range is much narrower for imports than for exports, it still means that the majority of the production occurs prior to the imports arriving in Canada

Another way of examining Canadian trade is to compare the scores for exports and imports. While the downstream scores are similar (only marginally higher for imports), exports are more upstream than imports. There are two implications from this; the first implication is derived from the  definitions of upstreamness and downstreamness: Canadian imports are closer to final goods than Canadian exports. Again, this implies that Canada specializes in the early stages of the value chains. The second is that given the similarity of the downstream score, the larger upstream score implies that Canadian exports tend to be involved in longer value chains and conversely, Canadian imports come from shorter value chains.

Building on the evidence presented in figure 7, figure 8 shows the positioning of Canadian trade with its major trade partners: the U.S., China, the European Union (excluding the United Kingdom, the EU-27), Mexico, and the rest of the world (ROW). This suggests that Canada’s specialization in the early parts of the supply chain are not due to any particular trade relationship, but rather a feature of Canadian trade in general.

Canadian exports have a similar upstream and downstream scores that does not vary much by destination. The downstreamness index has a narrow range of only 2.1 (EU) to 2.4 (Mexico). The upstreamness index also has only a slightly broader range of 2.7 (U.S.) to a high of 3.1 (rest of the world—the grouping of countries that are not the U.S., EU-27, China, or Mexico partners). This means that exports will require more production steps after being exported than they went through prior to being exported. In other words, Canadian exports to all destinations are generally at the beginning of the supply chain.

On the import side, the upstreamness and downstreamness measures tell different stories. The downstream measure is largely the same across origins, with the range being less than 0.2. However, the upstream measure for imports varies more than the three other measures (import and export downstreamness, and export upstreamness). This may indicate that not all Canadian imports are similar. The imports from Mexico and China are much less upstream (i.e. closer to a final good) than the imports from the U.S. and the rest of the world. China historically has been known as the place of final assembly across Asia, and thus it is expected that Canadian imports from China would contain more final goods (Fang et al. 2010). For Mexico, the key driver for the low upstream measure is the autos trade. While Mexico and Canada are both apart of the integrated North American auto production process, Mexico does much of the final assembly, and thus exports many final goods (Klier & Rubenstein 2017). On the other hand, imports from the U.S. and the rest of the world are more upstream (in fact, more upstream than downstream), indicating that Canada uses many of these imports for further production.

Figure 8 - Text version

	Upstreamness	Downstream	Downstreamness
Overall	2.065561	1.951852	-1.95185
Imports	2.410709	2.396357	-2.39636
Imports from US	2.438315	2.406339	-2.40634
Imports from China	2.15041	2.38247	-2.38247
Imports from Mexico	1.912343	2.463016	-2.46302
Imports from EU	2.246564	2.35629	-2.35629
Imports from ROW	2.649954	2.336203	-2.3362
Exports	2.770931	2.293868	-2.29387
Exports to US	2.693764	2.33795	-2.33795
Exports to China	2.875192	2.193513	-2.19351
Exports To Mexico	2.720074	2.391734	-2.39173
Exports to EU	2.809331	2.108257	-2.10826
Exports to ROW	3.118857	2.230808	-2.23081

Based on the narrow range of value chain positioning, it would seemingly be plausible that similar products are traded.  However, this would be an erroneous conclusion as the exports and imports are different for each market, despite having proximate upstream and downstream values. To explore the idea of similarity, the Finger-Kreinin (1979) similarity index is used. The details of the index can be found in the appendix, and the results can be found in figure 9. In the table, the upper triangular blue matrix are the results for the import similarity, and the lower triangular pink matrix are the results for the export similarity. As a guideline, anything below 0.67 is not similar, and anything above 0.67 is a similar.

Contrary to the upstream and downstream analysis in figure 7, the industries that Canada imports from are slightly more similar than the types of industries from which Canada exports. Only two pairs (the U.S. and total, and EU-27 and ROW) have similar exports, whereas there are five pairs of similar imports. The main takeaway is not that Canadian imports across different origins are similar, but rather Canadian exports to different destinations are dissimilar.

While the export industries vary by destination, the upstream and downstream export scores are comparable across destinations. In other words, the products Canada exports to its major trading partners are different, but the role Canada plays in the supply chain is similar. This suggests that it is not the case that Canada specializes in specific products that are located at the beginning of the supply chain, but rather, Canada specializes in  early supply chain activities.

Canada’s specialization in the early value chain activities implies that Canadian exports are more exposed to demand shocks than supply shocks. This is not to say that supply shocks cannot cause problems for Canadian industries—returning to the industry analysis in figure 5 and 6, all industries use some intermediate goods—but in general, there’s more potential for there to be downstream problems rather than upstream problems for Canadian exports. The COVID-19 pandemic is a good illustration to highlight Canada’s vulnerabilities. There was (or still is) a supply issue of securing enough semi-conductors which has led to slowdowns in many industries—particularly the auto industry. But in general, most of the supply issues have been for consumer goods while it has been domestic factors (notably plant closures to protect workers’ health) that has restricted Canadian exports more than any upstream import factors.

5. Changes in the value chain positioning over time

The next objective of this paper is to determine how Canada’s supply chain positioning has changed over time. In order to determine the dynamics of Canada’s value chain positioning, the Canadian IO tables from 1997-2019 are used. IO tables have changed over the years, but the figures in this section use both the most current and the 1961 link that is comparable across all the years.^{Footnote 13} Figure 10 has the results for production.^{Footnote 14}

Figure 10 - Text version

Year	Upstream production	Downstream production	Upstream Production L61	Downstream production L61
1997			2.052368	1.96496
1998			2.040677	1.973754
1999			2.063947	1.982313
2000			2.181081	2.031535
2001			2.148402	2.033965
2002			2.114286	2.017359
2003			2.109613	1.998667
2004			2.133451	1.997454
2005			2.124599	1.986915
2006			2.099149	1.973743
2007			2.072368	1.953419
2008			2.086246	1.955894
2009			1.970784	1.937937
2010	2.01502	1.948295	1.98851	1.951664
2011	2.052181	1.948191	2.017935	1.952592
2012	2.042345	1.962459	2.006087	1.967972
2013	2.052839	1.952023	2.021133	1.956189
2014	2.066442	1.948776	2.019107	1.955458
2015	2.04214	1.968422	2.001193	1.97501
2016	2.024537	1.964143	1.98868	1.968009
2017	2.049934	1.958005	2.011476	1.960806
2018	2.080502	1.97842	2.042379	1.983365
2019	2.065561	1.962935	2.027306	1.968216

If the late 1990s are excluded from the sample, then it would be viable to say that the measures have trended downwards (indicating shorter value chains), if only slightly. This would likely be attributable to the increasing amount of services that are present in the economy. However, with those 3 years in the late 1990s included, figure 9 shows that Canada’s production in 2019 is fairly similar to what it was in the late 1990s, despite increasing slightly in the early 2000s and decreasing towards the end of that decade.^{Footnote 15} Next, figure 11 has the results for Canadian imports and exports over time.

Figure 11- Text version

Year	Upstream Exports	Downstream Exports	Upstream Exports L61	Downstream Exports L61	Upstream imports	Downstream imports	Upstream imports L61	Downstream imports L61
1997			2.465844	2.351965			2.328859	2.372683
1998			2.422833	2.348355			2.316361	2.363843
1999			2.412602	2.328825			2.333863	2.340835
2000			2.598223	2.370903			2.474356	2.433985
2001			2.572156	2.393827			2.433631	2.477576
2002			2.527997	2.38319			2.38989	2.443717
2003			2.553337	2.334409			2.376117	2.410174
2004			2.604059	2.342673			2.413269	2.412442
2005			2.621439	2.309948			2.393718	2.394996
2006			2.614195	2.29942			2.364844	2.388552
2007			2.613612	2.262457			2.326699	2.361469
2008			2.708647	2.222661			2.35677	2.364803
2009			2.556023	2.247745			2.267516	2.352607
2010	2.705172	2.248785	2.591745	2.259898	2.351914	2.336354	2.297897	2.357903
2011	2.802975	2.234569	2.66362	2.244884	2.409397	2.336092	2.342796	2.358842
2012	2.761184	2.284401	2.617315	2.297647	2.402211	2.361436	2.333625	2.391315
2013	2.760201	2.267538	2.637667	2.27889	2.37692	2.360312	2.32936	2.383776
2014	2.814699	2.250902	2.643952	2.264014	2.424325	2.379293	2.347889	2.408602
2015	2.70333	2.344692	2.56337	2.364356	2.396656	2.410122	2.333523	2.44363
2016	2.650417	2.351743	2.524124	2.369031	2.383474	2.409631	2.325556	2.435202
2017	2.721139	2.311921	2.589275	2.328611	2.384982	2.403591	2.329252	2.427787
2018	2.78589	2.325511	2.654446	2.344905	2.423394	2.430048	2.370339	2.458726
2019	2.770931	2.316028	2.630826	2.332947	2.410709	2.416582	2.354543	2.445508

The positioning and dynamics of Canadian exports and Canadian imports tell two very different stories. For Canadian imports, the upstream and downstream position is more or less the same, and there has been very little movement in the measures since 1997. Canadian exports, on the other hand, have a sizeable gap between their upstream and downstream measures, which has widened over the last 20 years. This means Canadian exports can be expected to go through more stages of production outside of Canada than the number of production stages that they have been through preparing to be exported.

The dynamics for exports have a clean story—and for people familiar with Canadian trade, an expected narrative. Most of the dynamics for exports have to do with a growing share of oil and gas extraction, coupled with a shrinking share in manufacturing. Between 2001 and 2008, the share of oil and gas extraction in exports increased 8.9 percentage points (p.p.), from 8.2% to 17.2%. Oil and gas extraction also became slightly more downstream (going from 1.6 to 1.4) while maintaining the same upstreamness score of 3.6. Conversely, motor vehicle manufacturing lost export share, decreasing from 13.2% of exports in 2001 to 6.9% of exports in 2008. It became slightly more downstream (going from 3.1 to 3.3) while also becoming slightly less upstream (1.2 to 1.1). Thus, this is the case of an upstream industry gaining share while a downstream industry lost share. The narrative also fits other dynamics of the graph—the price of oil plunged at the end of 2014 and beginning of 2015 causing a short recession in Canada. On the graph, this corresponds to the sudden decrease in upstreamness and increase in downstreamness. Since then, oil prices have somewhat recovered, but are still not back to the level observed in the early 2010’s.

Lastly, the IO tables do not have detail on Canada’s major trading partners prior to 2010, but the dynamics can still be assessed for the 2010s.^{Footnote 16} Figure 12 has the dynamics for the U.S., China, Mexico, and the R.O.W.

Figure 12 - Text version

Year	Upstream exports to U.S.	Downstream exports to U.S.	Upstream imports from U.S.	Downstream imports from U.S.	Upstream exports to China	Downstream exports to China	Upstream imports from China	Downstream imports from China	Upstream exports to Mexico	Downstream exports to Mexico	Upstream imports from Mexico	Downstream imports from Mexico	Upstream exports to R.O.W.	Downstream exports to R.O.W.	Upstream imports from R.O.W.	Downstream imports from R.O.W.
2010	2.656252	2.252924	2.320752	2.375038	3.131811	2.257471	1.940894	2.343068	2.564782	2.374005	1.907106	2.451987	2.885365	2.208689	2.539857	2.243232
2011	2.74291	2.23425	2.360115	2.375151	3.25308	2.213647	1.973582	2.349651	2.624766	2.33961	1.986739	2.471679	3.014894	2.206684	2.629174	2.242712
2012	2.686767	2.287139	2.361166	2.400392	3.215888	2.232067	2.010874	2.36251	2.620182	2.341809	2.056524	2.500228	3.001887	2.258434	2.597252	2.267244
2013	2.708645	2.272028	2.356569	2.393868	3.1924	2.220273	2.027427	2.378891	2.597019	2.352358	2.033196	2.496093	2.952482	2.230521	2.529922	2.266789
2014	2.767383	2.245822	2.423342	2.395362	3.049549	2.217263	2.052949	2.413266	2.597685	2.403496	1.95896	2.58214	3.062742	2.232587	2.563422	2.308444
2015	2.634517	2.372703	2.413616	2.422968	2.93942	2.250718	2.073069	2.40724	2.566274	2.413555	1.901721	2.628013	3.01761	2.243195	2.488862	2.353828
2016	2.596502	2.384847	2.409054	2.425851	2.832652	2.219448	2.084535	2.395853	2.524205	2.419995	1.893764	2.628526	2.952132	2.226964	2.446765	2.348578
2017	2.675366	2.337432	2.403252	2.415619	2.863305	2.193514	2.099481	2.388191	2.705608	2.391734	1.852889	2.70213	3.005663	2.200302	2.474473	2.342967
2018	2.746335	2.351373	2.456631	2.439542	2.892404	2.235522	2.15321	2.422427	2.728913	2.391607	1.917966	2.695759	3.037466	2.213135	2.47099	2.379719
2019	2.72613	2.334945	2.438315	2.423539	2.840369	2.211275	2.15041	2.399004	2.649505	2.375213	1.912343	2.655973	3.050652	2.22499	2.465816	2.378654

The dynamics for Canada’s trade with the U.S. and Mexico have been stable for the last 10 years and are the same as described in figure 7: Canada’s exports to both the U.S. and Mexico are more upstream than Canada’s imports, while Canadian imports from Mexico have a low upstream value and a relatively high downstream value. Once again, this is consistent with Canada specializing in the early stages of the value chain. The dynamics of Canada’s trade with the rest of the world are also consistent with Canada moving towards earlier parts of the supply chain: exports have become more upstream, while imports have become less upstream and slightly more downstream.

At any static point in time, Canada’s trade with China fits the narrative of Canada exporting early stages of value chains and importing final goods; however, the dynamics for Canada’s trade with China have changed substantially over the past 10 years. Canadian exports to China have become less upstream than they were at the beginning of the decade, while imports from China have become more upstream. This result is unexpected because it contradicts the recent narrative attached to Chinese trade. For much of the 2000s, China was the final assembly point for many supply chains across Asia and they held a high percentage of foreign content in their exports (see, for example, Scarffe 2020, or Kee and Tang (2016) for firm level evidence). However, the narrative has been changing into one where China is doing more of the value added activities along the supply chain. Chor, Manova, and Yu (2021) calculate Chinese production and trade upstreamness (the same measure presented in this paper) using Chinese IO tables between 1992-2014. They find that while Chinese exports have had a relatively stable upstreamness, Chinese imports have had a sharp rise in their upstreamness. They conclude that China’s role in the supply chain is expanding. This narrative contradicts figure 11 which suggests China’s supply chains are contracting rather than expanding.

Using a shift-share to decompose the change in positioning into within industry effects and between industry effects reveals that the decrease in upstreamness for exports has been driven by changes in export shares between industries, while the increase in upstreamness for imports has been balanced between the two effects. For Canadian exports, there are three industry types driving these results. First, mining and similar industries—particularly non-ferrous metal production and processing—have become a smaller share of Canada’s exports to China. These are upstream industries and the decreasing shares mean that Canadian exports become less upstream. Second, automobile and light-duty motor vehicle manufacturing went from 0.04% of exports in 2010 to 2.5% of Canadian exports to China in 2019. Autos are not upstream and increasing these exports lowers the upstream score of Canadian exports to China. Lastly, university services and other service industries (such as food and drinking places)—which are not upstream nor downstream—are becoming a larger share of Canadian exports to China.

It may be possible to reconcile the decrease in the mining sector exports with Chor et al.; perhaps China is sourcing those imports from other countries or producing more domestically. However, the increasing share of the latter two industries are more difficult to harmonize with Chor et al. The larger share of autos and larger share of service exports are the result of rapidly growing affluence in China. It would be expected that the increase in buying power would affect China’s trade with all countries, instead of being a particular aspect of the Canada-China relationship. The data for Chor et al. stop in 2014 so it is possible that more recent analysis is congruent; however, the trend of increased enrollment by Chinese students at Canadian post-secondary institutions started at least as early as the late 90s and thus thinking that the last 5 year would reveal a different trend is unconvincing.  A definitive explanation would require examining China’s multilateral trade patterns and is beyond the scope of this work.

6. Conclusion

The purpose of this paper is to categorize where Canada fits into global value chains. Key findings indicate that Canadian exports are more upstream (outputs are further away from final goods) and less downstream (inputs are closer to value added) than Canadian imports. This is consistent with the notion that Canada specializes in producing goods in the earlier portion of the value chain while importing goods from the end of the value chain. Further supporting this statement is the finding that Canadian exports to all  major trading partners (and the rest of the world in aggregate) are more upstream and less downstream than Canadian imports,. Despite the fact that for each trade partner, Canada exports and imports a different set of products. This means that Canadian consumption may be more vulnerable to supply shocks as Canada imports many final goods, while Canadian exports are more vulnerable to demand shocks as Canada’s production happens mainly at the start of the value chain. There are some major industries that produce final goods and are more susceptible to supply shocks (such as the auto industry and manufacturing in general); but in aggregate, most of Canada’s exports are in the earlier portions of supply chains and are more susceptible to demand shocks.

The gap between the upstreamness and downstreamness of Canadian exports has grown since the late 1990s. Most of this happened in the 2000’s due to a growing share of oil and gas extraction—an upstream industry with outputs that are far from a final good—and a diminishing share of auto vehicle manufacturing—a downstream industry where the outputs are largely final goods. Analysis of trade with the U.S., Mexico, and the rest of the world suggest that Canada’s place in the value chain has not changed in the last 10 years. However, contrary to the existing literature, Canada’s exports to China have become less upstream while imports have become more upstream. Whether this is specific to Canada’s trade with China, or an entirely new pattern in Chinese trade is left to future work.

References

Antràs, P., Chor, D., Fally, T., & Hillberry, R. (2012). Measuring the upstreamness of production and trade flows. American Economic Review, 102(3), 412-16.

Antràs, P., & Chor, D. (2018). On the measurement of upstreamness and downstreamness in global value chains (pp. 126-194). Routledge.

Boileau, D., & Sydor, A. (2020). Vulnerability of Canadian Industries to Disruptions in Global Supply Chains. Global Affairs Canada. Ottawa.

Chor, D., Manova, K., & Yu, Z. (2021). Growing like China: Firm performance and global production line position. Journal of International Economics, 130, 103445.

Fang, Tony, Caroline Gunterberg, and Emma Larsson. "Sourcing in an increasingly expensive China: Four Swedish cases." Journal of Business Ethics 97, no. 1 (2010): 119-138.

Finger, J.M. and Kreinin, M.E. (1979) Measure of Export Similarity and Its Possible Uses. Economic Journal, 89, 905-912. http://dx.doi.org/10.2307/2231506

Kee, H.L., Tang, H. (2016). Domestic value added in exports: theory and firm evidence from China. American Economic Review, 106 (6), 1402–1436.

Klier, T., & Rubenstein, J. (2017). Mexico’s growing role in the auto industry under NAFTA: Who makes what and what goes where. Economic perspectives, 41(6),1-29.

Miller, R. E., & Temurshoev, U. (2017). Output upstreamness and input downstreamness of industries/countries in world production. International Regional Science Review, 40(5), 443-475.

Scarffe, C. (2020). The Canada-China Global Commerce Picture and Supply Chain Links. Global Affairs Canada. Ottawa.

Statistics Canada. (2020). Symmetric input-output tables, 2019.

Statistics Canada. (nd). Symmetric input-output tables, 1997-2018.

Appendix

The below network diagram attempts to show how value added turns into a final good, and how the measures of supply chain positioning works. The simple measures of supply chain positioning account only for the share of inputs that are value added, and the share of outputs that are final goods—lines “a.” and “b.”, respectively. The more complex measures attempt to account for all the different industry linkages, specifically, how many productions steps have the intermediate inputs already been through, and how many more production stages the intermediate outputs will go through.

Note: Industry i (the gold box) is downstream from the blue circles and is upstream to the purple circles.

Technical Appendix A: open economy adjustment

This is a rephrasing of the arguments put forth in Antràs et al. (2012). Re-writing equation (1) to include imports, exports, and net inventories:^{Footnote 17}

Text version

Y_i is equal to the sum from j of Z_ij plus F_i plus X_i minus M_i plus N_i

The problem is that in order to construct inter-industry flows, the imports, exports, and net inventories need to be decomposed into the share that becomes final goods, and the share that become intermediate goods. This means that the share of gross output that is used as intermediate inputs in other industries (including cross-border) is:

Text version

Delta_ji is equal to a_jiYj plus X_ji minus M_ji plus N_ji divided by Y_i

X_ji is equal to delta_ji multiplied by X_i

M_ji is equal to delta_ji multiplied by M_i

N_ji is equal to delta_ji multiplied by N_i

However, in practice we lack data on inter-industry cross-border flows; additionally we lack data on inter-industry flows of net inventories.^{Footnote 18} In order to proceed an assumption is made that for each industry i,  the share of exports, imports, and net inventories all have the same share of output that are used as intermediate outputs and final outputs. In other words:

Putting these three assumptions into equation (11), the delta matrix becomes:

Text version

Delta_ji is equal to a_ji times Y_j divided by Y_i minus X_i plus M_i minus N_i

Which is equivalent to changing the A matrix to:

Text version

a_ji hat is equal to a_ji multiplied by y_i divided by Y_i minus N_i plus M_i minus X_i

Technical Appendix B: Finger-Kreinin similarity measure

The Finger-Kreinin (1979) similarity measure uses a simple formula where the product or industry shares are compared to the shares of a different basket. The similarity measure takes the lower of the two shares for each industry, and sums across industries. Letting a and b be vectors with components denoted by subscript i, the similarity between the two vectors can be calculated as:

Text version

S as a function of a and b is equal to the sum from I to N of the minimum of the two elements a_i and b_i, such that the sum from I to N of a_i is equal to one, and the sum from I to N of b_i is equal to one.

Thus, if the vectors a and b were identical, the similarity would be one. If there was no overlap between the two vectors, the similarity would be zero.

As a benchmark, a simulation of two vectors of length 234 (the number of Canadian industries) were randomly generated from a uniform distribution, and then a similarity measure was constructed. This simulation was performed 10,000 times using Stata 16. The average similarity measure was 0.67 with a standard deviation of 0.02. Using this as a benchmark, anything below 0.67 is not similar, and anything above 0.67 is a similar.

It is worth noting that the length of the vector is important. As the number of components of the vector increases, the expected share of each component decreases. As the expected share of each component decreases, the expected value of the similarity index also decreases. Thus the use of 0.67 as a benchmark similarity cut-off stated in the paper is only valid when the vectors are of length 234.