Fetching primary parquet sources and computing exhibits.
research, intra-industry trade
Has intra-industry trade risen or fallen as GVCs deepened and China rose?
The Grubel-Lloyd index (Grubel & Lloyd, 1975, Intra-Industry Trade) measures how much of a country-product flow is two-way: similar goods crossing the border in both directions, the statistical signature of intra-industry specialisation. Computed at country × HS6 and trade-weighted across 238 economies and 5,022 HS6 lines, global intra-industry trade stands at 0.423 in 2024, against 0.438 in 1995. The net 30-year change is -15 GL × 1,000 points, with a peak of 0.457 in 1999.
methodGL × HS6 × country
window1995-2024
countries238 (BACI)
products5,022 HS6
global GL 20240.423
GL vs PCI corr0.438 (n=4,587)
Global intra-industry trade over thirty years
The weighted-average Grubel-Lloyd index pools country-product IIT shares across all 238 BACI reporters and all HS6 lines where the country has strictly positive exports and imports, weighting each observation by its two-way trade (X + M). This is the standard country-level operationalisation laid out in Grubel & Lloyd (1975), with the modern panel extension surveyed in Brülhart (1994, 'Marginal Intra-Industry Trade,' Weltwirtschaftliches Archiv 130: 600-613), and refined for vertical vs horizontal IIT by Fontagné& Freudenberg (1997, CEPII Working Paper 1997-06).
Figure 1
Global Grubel-Lloyd index, country-HS6 trade-weighted, 1995-2024
Global IIT fell by -15 GL×1,000 points between 1995 and 2024: from 0.438 to 0.423. The series peaks near 0.457 in 1999, broadly consistent with the GVC-expansion phase documented in Johnson & Noguera (2012, JIE) and the post-2011 flattening in Antràs (2020, JEP). A lower end-of-window value is the statistical footprint of asymmetric trade relationships (notably China-US and resource exporters) expanding faster than classic two-way manufacturing flows.
Method: GL_c,k = 1 - |X_c,k - M_c,k| / (X_c,k + M_c,k), computed at country x HS6, weighted by (X + M) across all country-product cells. Sources: Grubel & Lloyd (1975) Intra-Industry Trade; Brulhart (1994) Weltwirtschaftliches Archiv 130: 600-613; Fontagne & Freudenberg (1997) CEPII WP 1997-06. Data: CEPII BACI 202501 (retrieved 2026-04-28). Authors calcs.
Which HS Sections are most intra-industry intensive
Slicing the 2024 country-product GL pool by HS Section (21 sections of the WCO Harmonized System) exposes the structural composition of IIT. Chemicals, machinery, textiles and prepared foodstuffs tend to run higher, because production is fragmented across borders and varieties are differentiated, while primary-commodity sections sit lower because most producers are net exporters or net importers without offsetting reverse flows. This matches the horizontal-vs-vertical IIT taxonomy of Fontagné& Freudenberg (1997) and the differentiated-products mechanism of Krugman (1980, AER) on monopolistic competition in international trade.
Figure 2
Grubel-Lloyd index by HS Section, 2024
Top intra-industry country pairs over time
The pair-level Grubel-Lloyd applied to bilateral gross flows isolates the trade relationships that are most two-way. By construction this is an aggregate (not HS6) statistic computed on bilateral totals: GL_ij = 1 - |X_ij - X_ji| / (X_ij + X_ji) where X_ij is i's exports to j. Classic horizontal-IIT pairings (intra-EU DEU-FRA, USMCA CAN-USA) sit high and stable; pairs driven by unilateral comparative advantage (e.g., CHN selling to HKG or to large net-import regions) sit lower. The pattern speaks to the Helpman-Krugman (1985) monopolistic-competition trade benchmark relative to the Ricardian comparative-advantage benchmark.
Figure 3
Top-6 pairs by 2024 two-way trade: Grubel-Lloyd over 1995-2024
#
Pair
Two-way trade 2024
GL 2024
1
MEX-USA Mexico / USA
$728.2B
0.651
2
CAN-USA Canada / USA
$658.1B
0.784
3
CHN-USA China / USA
$603.8B
0.514
4
CHN-HKG China / China, Hong Kong SAR
$308.8B
0.206
5
CHN-JPN China / Japan
$301.1B
0.914
6
CHN-KOR China / Rep. of Korea
$294.2B
0.945
7
CHN-DEU China / Fed. Rep. of Germany (...1990)
Does intra-industry trade correlate with product complexity?
For each HS6 line in 2024 we compute global GL (weighted mean of country-level GL across the world) and plot it against the Product Complexity Index (PCI) from Hausmann, Hidalgo et al. (2014,The Atlas of Economic Complexity, Harvard CID). PCI is zero-centred and routinely negative for undifferentiated primary products, positive for complex differentiated goods. A positive correlation implies that more complex products are also more intra-industry.
Figure 4
Grubel-Lloyd index vs Product Complexity Index, HS6 in 2024
Horizontal vs vertical IIT: decomposing the global trend by complexity
Fontagné& Freudenberg (1997, CEPII WP 1997-06) distinguish horizontal IIT (differentiated varieties of similar quality, Krugman 1980 monopolistic-competition) from vertical IIT (quality-differentiated varieties along the value chain, Falvey 1981). Absent unit-value data at HS6 we use a complexity proxy: split HS6 lines at the median PCI in 2024 and compute the trade-weighted country-product GL index within each half each year. The high-PCI half (differentiated, complex) maps closer to horizontal IIT; the low-PCI half maps closer to homogeneous products with dominant Heckscher-Ohlin comparative-advantage patterns.
Figure 5
Trade-weighted GL index by HS6 complexity half, 1995-2024
The high-PCI half of HS6 lines runs at GL = 0.500 in 2024 against 0.322 for the low-PCI half, a gap of 178GL×1,000 points. Both halves trended down modestly since 1995 (0.512 high, 0.334 low), consistent with the asymmetric expansion of China-centred and resource-driven flows Antràs (2020, JEP) emphasises. The level difference between halves is the statistical signature of Fontagné- Freudenberg's horizontal/vertical taxonomy.
Horizontal vs vertical IIT: unit-value overlap test (Fontagné-Freudenberg)
Figure 5 uses PCI as a complexity proxy, but the canonical Fontagné & Freudenberg (1997, CEPII WP 1997-06) test compares export and import unit valuesat the country×HS6 level: if the ratio UVX/UVMis within ±15% (i.e. |ln(UVX/UVM)| ≤ 0.15), the two-way flow is horizontal(similar-quality varieties, Krugman 1980); otherwise it is vertical (quality- differentiated varieties along the value chain, Falvey 1981, Journal of International Economics 11(4): 495-511). Using TradeWeave's reporter-level unit-value table (2000-2019, CEPII BACI derivation), we decompose the weighted GL index into horizontal and vertical components.
Figure 6
Horizontal vs vertical IIT from unit-value overlap, 2000-2019
In 2019, horizontal IIT (unit-value overlap ±15%) runs at GL = 0.481, vertical IIT at GL = 0.444. Horizontal flows carry 24.0% of two-way trade on average over the sample window; the remaining 76.0% is vertically differentiated. The level gap between horizontal and vertical GL is the Fontagné-Freudenberg signature: quality-ladder flows (Khandelwal 2010, AER) register as two-way but are driven by North-South cost structures rather than the Krugman (1980) love-of-variety mechanism that anchors horizontal IIT.
Figure 7: Services IIT versus goods IIT, 2005-2024
The Figures 1-6 analyses operate on BACI goods trade. The IMF BPM6 balance-of- payments standard (IMF 2013, Balance of Payments and International Investment Position Manual, 6e) splits international trade into goods (credit and debit) and services (credit and debit), enabling a like-for-like computation of country-level Grubel-Lloyd on the aggregated services account. Using IMF BoP at country frequency (2005-2024), we compute the trade-weighted cross-country GL on services credit/debit flows and compare it with the analogous goods GL from the same source. Loý & Saucier (2020, WTO ERSD-2020-06) document that services IIT is systematically higher than goods IIT; the mechanism they emphasise is the task-level fragmentation of modern services (Baldwin 2016, The Great Convergence) that mirrors the goods-trade pattern of Grossman & Rossi-Hansberg (2008).
Figure 7
Services Grubel-Lloyd vs goods Grubel-Lloyd, country-aggregate, IMF BPM6
Figure 8. Cross-country distribution of country-aggregate Grubel-Lloyd, 2024
Figures 1-7 are trade-weighted aggregates that pool the world's 238 BACI reporters and mask cross-country dispersion. For each reporter c with at least 100 HS6 cells of two-way trade in 2024 we compute the country-aggregate GLc = trade-weighted mean of cell GL across products, and bin the resulting cross-country distribution. The histogram below shows how concentrated or dispersed intra-industry trade is across reporters in a single year. Krugman (1980, AER 70(5): 950-959) predicts horizontal IIT should rise with similarity in factor endowments and market size, so the upper-tail bins should cluster on advanced manufacturing economies (intra-EU and East-Asian core) and the lower-tail bins on resource exporters and small open economies with one-sided specialisation. The country-aggregate dispersion in 2024 is the cross-section that anchors the pair-aggregate evidence in Figure 3.
Figure 8
Distribution of country-aggregate Grubel-Lloyd index across BACI reporters, 2024
What the evidence says
IIT did not march steadily upward. After rising through the 1990s GVC expansion, the weighted global Grubel-Lloyd peaks near 1999 at 0.457 and then drifts lower, ending 2024 at 0.423. This tracks the deceleration of trade-to-GDP after the GFC and the asymmetric expansion of China-centred trade (Antràs, 2020, JEP).
IIT is a manufacturing-and-chemicals phenomenon. Machinery, chemicals, textiles, prepared foodstuffs run above 0.70; primary commodities and mineral products sit near or below 0.45. The Krugman (1980) differentiated-products mechanism operates where differentiation is possible; Heckscher-Ohlin dominates elsewhere.
Classic horizontal-IIT pairs remain the template. USMCA (CAN-USA, MEX-USA), intra-Asia manufacturing (CHN-JPN, CHN-KOR), and intra-EU pairs anchor the high end of the pair-level distribution; trade relationships driven by one-sided resource exports anchor the low end.
Complexity and IIT move together at the HS6 line level. Pearson r = 0.438 on n = 4,587 HS6 codes in2024: more complex goods are more two-way.
Open questions
Without unit-value data we use PCI as a complexity proxy (Figure 5); a proper horizontal-vs-vertical decomposition requires HS6 bilateral unit values, which BACI exposes at HS6×partner and could sharpen the Fontagné-Freudenberg (1997) typology.
How much of the post-2012 IIT drift down reflects China-US and China-RoW asymmetry versus resource-exporter basket effects? A counterfactual that fixes China's pattern at its 2012 level would answer.
Does marginal IIT (Brülhart 1994) track total IIT at the decadal level, or does adjustment pressure decouple from the stock of two-way trade during structural transitions (WTO accession, regional integration deepening)?
References
Antràs, P. (2020). 'De-Globalisation? Global Value Chains in the Post-COVID-19 Age.' Journal of Economic Perspectives, forthcoming. NBER WP 28115.
Brülhart, M. (1994). 'Marginal Intra-Industry Trade: Measurement and Relevance for the Pattern of Industrial Adjustment.' Weltwirtschaftliches Archiv 130(3): 600-613.
Falvey, R. E. (1981). 'Commercial Policy and Intra-Industry Trade.' Journal of International Economics 11(4): 495-511.
Khandelwal, A. (2010). 'The Long and Short (of) Quality Ladders.' Review of Economic Studies 77(4): 1450-1476.
Fontagné, L., & Freudenberg, M. (1997). 'Intra-Industry Trade: Methodological Issues Reconsidered.' CEPII Working Paper No. 1997-06.
Grossman, G. M., & Rossi-Hansberg, E. (2008). 'Trading Tasks: A Simple Theory of Offshoring.' American Economic Review 98(5): 1978-1997.
Grubel, H. G., & Lloyd, P. J. (1975). Intra-Industry Trade: The Theory and Measurement of International Trade in Differentiated Products. Macmillan, London.
Hausmann, R., Hidalgo, C. A., Bustos, S., Coscia, M., Simoes, A., & Yildirim, M. A. (2014). The Atlas of Economic Complexity: Mapping Paths to Prosperity. MIT Press / Harvard CID.
Helpman, E., & Krugman, P. R. (1985). Market Structure and Foreign Trade: Increasing Returns, Imperfect Competition, and the International Economy. MIT Press.
Johnson, R. C., & Noguera, G. (2012). 'Accounting for Intermediates: Production Sharing and Trade in Value Added.' Journal of International Economics 86(2): 224-236.
Krugman, P. R. (1980). 'Scale Economies, Product Differentiation, and the Pattern of Trade.' American Economic Review 70(5): 950-959.
Highest IIT-intensity section: Section 6 (Chemicals and allied industries) at 0.785. Lowest: Section 12 (Footwear, headgear, umbrellas) at 0.349. The ranking is dominated by how differentiated the section's goods are: machinery and chemicals run high, primary commodities and miscellaneous art/antiques run low.
Method: per-country per-section GL = 1 - |Sigma_k X_c,k - Sigma_k M_c,k| / (Sigma_k X_c,k + Sigma_k M_c,k) aggregated over HS6 within section, then trade-weighted across countries. Source: CEPII BACI 202501 (retrieved 2026-04-28) with HS Section map from WCO HS 2022. Authors calcs.
In 2024 the ten largest two-way pairs span $728.2B (MEX-USA, GL 0.651) down to $243.4B (CHN-RUS, GL 0.935). Pairs that anchor regional value chains run substantially more two-way than pairs that anchor a resource-for- manufactures bargain.
Method: pair GL on bilateral gross flows, GL_ij = 1 - |X_ij - X_ji| / (X_ij + X_ji). Top-6 pairs ranked by two-way trade in the latest year. Source: CEPII BACI 202501 (retrieved 2026-04-28) bilateral totals. Authors calcs.
$253.0B
0.731
8
CHN-VNM China / Viet Nam
$252.8B
0.744
9
DEU-USA Fed. Rep. of Germany (...1990) / USA
$246.9B
0.733
10
CHN-RUS China / Russian Federation
$243.4B
0.935
The Pearson correlation between HS6 GL and PCI across 4,587 product lines in 2024 is r = 0.438. The sign is positive: more complex products tend to be more intra-industry. This is consistent with the Grossman-Rossi-Hansberg (2008, AER) offshoring theory in which fragmentable, complex production chains generate task-level two-way flows, and with the Helpman-Krugman (1985) differentiated-products framework.
Method: HS6 GL = weighted average of country-level GL across countries for that HS6 line. PCI from Hausmann, Hidalgo et al. 2014 Atlas of Economic Complexity (Harvard CID). Points sized by 2024 two-way trade USD. Colour by HS Section. Top-600 by trade shown. Source: CEPII BACI 202501 (retrieved 2026-04-28); CID Atlas via pci_rankings.parquet. Authors calcs.
Method: HS6 lines split at median PCI (latest-year rank) following Fontagne & Freudenberg (1997) horizontal/vertical IIT taxonomy. Within each half, GL_c,k = 1 - |X_c,k - M_c,k| / (X_c,k + M_c,k) at country × HS6, trade-weighted by (X + M). Source: CEPII BACI 202501 (retrieved 2026-04-28); pci_rankings.parquet (Hausmann-Hidalgo). Authors calcs.
Method: for each country x HS6 x year with positive X and M, compute |ln(UV_X / UV_M)|; classify as horizontal (H) if ≤ 0.15, vertical (V) otherwise (Fontagne & Freudenberg 1997, CEPII WP 1997-06, Fig. 2). Aggregate trade-weighted GL within each group. Source: CEPII BACI 202501 (retrieved 2026-04-28) unit-value table (unit_values.parquet, reporter-level medians). Authors calcs.
In 2024, the cross-country services GL stands at 0.866 against 0.885 for goods, a gap of -19GL×1,000 points. Services credit and debit flows are more balanced within countries than goods flows, consistent with the Loý-Saucier (2020) observation that services trade runs on task-level two-way specialisation (business services, finance, travel) rather than the unilateral comparative advantage patterns that anchor much of goods trade. The gap has remained stable across the window, a pattern the WTO World Trade Report 2019 ('The Future of Services Trade') ties to the persistent-task-fragmentation hypothesis.
Method: country-year GL on aggregated credit vs debit flows; trade-weighted across reporting countries. Source: IMF BOP (BPM6), indicators 'Services, Credit/Revenue', 'Services, Debit/Expenditure', 'Goods, Credit/Revenue', 'Goods, Debit/Expenditure'. Reference: IMF (2013, BPM6); Loy & Saucier (2020) 'Trade in services: the key actors and their impact on world trade', WTO Staff Working Paper ERSD-2020-06.
Across 222 reporters with at least 100 HS6 two-way trade cells in 2024, the cross-country mean GL is 0.201, the median 0.144. The highest country-aggregate GLs sit on Switzerland (CHE) at 0.669n/aNetherlands (NLD) at 0.652, and Germany (DEU) at 0.648; the lowest sit on Dem. Rep. of the Congo (COD) at 0.005n/aSudan (SDN) at 0.007, and Guinea (GIN) at 0.007. The right-skew toward the upper bins matches Helpman & Krugman (1985)'s prediction that intra-industry specialisation is densest among similar-endowment partners; the left-tail mass corresponds to resource exporters and small open economies with one-sided specialisation.
Method: country-aggregate GL_c = trade-weighted mean of cell GL = SUM((1 - |X-M|/(X+M)) * (X+M)) / NULLIF(SUM(X+M), 0) over all HS6 cells with strictly positive X and M for reporter c in 2024; reporters with fewer than 100 such cells dropped to avoid sampling noise. Source: CEPII BACI 202501 (retrieved 2026-04-28). Reference: Krugman (1980) AER 70(5):950-959; Helpman & Krugman (1985) Market Structure and Foreign Trade.