How stitched is Bangladesh into cross-border production networks, and where in the chain does it sit? Three figures built from the OECD TiVA 2023 edition: a 26-year trajectory of backward and forward participation, a sectoral decomposition of the foreign-value-added embed, and an Antràs & Chor (2018)-style position index.
ISO3BGD
years covered1995-2020
backward, 202019.3%
forward, 202024.3%
rank by total GVC#1 of 86
Backward and forward participation
Backward participation measures foreign value added embedded in the country’s exports: how much of what BGD sells abroad was actually produced elsewhere. Forward participation measures domestic value added embedded in the exports of other countries: how much of what partners sell abroad was actually produced in BGD. The sum is total GVC participation; the ratio sorts upstream producers from downstream assemblers (Koopman, Wang & Wei 2014).
Figure 1
Bangladesh: backward and forward GVC participation, 1995-2020
In 2020, foreign value added accounted for 19.3% of BGD’s gross exports (backward), and BGD’s domestic value added made up 24.3% of partners’ exports (forward). Compared with 1995 (22.6% backward, 20.6% forward), the direction of change signals shallower downstream integration and deeper upstream supply to the rest of the world.
Aggregating across all exports hides the fact that different sectors use imported inputs in wildly different proportions. A country can look moderately integrated on average while one or two sectors (usually transport equipment, electronics, or apparel) are deeply dependent on imported intermediates. The chart below plots BGD’s foreign-VA share of gross exports for each ISIC Rev.4 sector in 2020, restricted to sectors that actually export.
Figure 2
Bangladesh: foreign-value-added share of exports by sector, 2020
The most import-intensive sector in BGD’s export mix is Textiles, apparel, leather (C13T15), where foreign VA is 25.8% of the sector’s gross exports. Sectors far below the country average are either primary-commodity exports (little imported input content by construction) or services (less tradable intermediate dependence). The dispersion across sectors is often wider than the cross-country dispersion of aggregates.
Source: OECD TiVA 2023 (DSD_TIVA_MAINSH@DF_MAINSH), EXGR_FVA share, partner=World, year=2020. Sectors: ISIC Rev.4 activity codes used in the OECD ICIO model.
Cite: Hossen, M. D. (2026). Bangladesh: foreign-value-added share of exports by sector, 2020. TradeWeave Workbench.copy permalinkcite
@misc{hossen_2026_fig-gvc-sector-BGD,
author = {Md Deluair Hossen},
title = {Bangladesh: foreign-value-added share of exports by sector, 2020},
year = {2026},
howpublished = {TradeWeave Workbench},
url = {https://tradeweave.org#fig-gvc-sector-BGD},
note = {Figure: Figure 2}
}
Upstream or downstream
The position index is (forward − backward) / (forward + backward). Positive values identify economies that supply intermediates to others (the inputs end up being re-exported by someone else); negative values identify economies that absorb intermediates and re-export finished or semi-finished goods. The measure is close to the Antràs & Chor (2018) upstreamness index, which builds the same intuition from the Leontief inverse of the Inter-Country Input-Output table.
Figure 3
Bangladesh: GVC position index, 1995-2020
In 2020, BGD’s position index is 0.11: a net upstream position in world value chains. Upstream economies (Taiwan, Korea, Chile, Saudi Arabia) supply intermediate inputs that partners re-export. Index values near zero indicate mid-stage hubs (Germany, Japan) equally engaged on both sides.
Source: derived from Figure 1. Antràs, P. & Chor, D. (2018) define a related upstreamness measure using Leontief inverse weights on final-demand destinations; the ratio here uses OECD TiVA backward and forward participation shares as a simpler approximation.
Cite: Hossen, M. D. (2026). Bangladesh: GVC position index, 1995-2020. TradeWeave Workbench.copy permalinkcite
@misc{hossen_2026_fig-gvc-position-BGD,
author = {Md Deluair Hossen},
title = {Bangladesh: GVC position index, 1995-2020},
year = {2026},
howpublished = {TradeWeave Workbench},
url = {https://tradeweave.org#fig-gvc-position-BGD},
note = {Figure: Figure 3}
}
Value-added exports: the Johnson-Noguera VAX ratio
Johnson & Noguera (2012, Journal of International Economics) defined the VAX ratio as domestic value added absorbed abroad divided by gross exports. A VAX below 1 means part of what BGD ships out is value added elsewhere (imported intermediates) or domestic value added that returns via round-trip intermediates. The OECD TiVA proxy used here is EXGR_DVA / EXGR at the total-economy level; it undercounts the full Johnson-Noguera measure (which subtracts reflected DVA absorbed at home) but preserves cross-country ranks.
Figure 4
Bangladesh: VAX ratio, 1995-2020
In 2020, domestic value added was 80.7% of BGD’s gross exports; in 1995 it was 77.4%. A falling VAX ratio indicates deeper backward participation (imported inputs feeding exports); a VAX close to 1 indicates an economy whose exports are almost entirely its own labour and capital (typical of commodity exporters and services-heavy economies).
Source: OECD TiVA 2023 (DSD_TIVA_MAINLV@DF_MAINLV). VAX proxy = EXGR_DVA / EXGR, activity=_T, partner=World. Full VAX definition: Johnson & Noguera (2012) J. Int. Econ. 86(2).
Cite: Hossen, M. D. (2026). Bangladesh: VAX ratio, 1995-2020. TradeWeave Workbench.copy permalinkcite
Total participation and Antràs-Chor upstreamness proxy
Antràs & Chor (2013, Journal of Political Economy) measure upstreamness as the average number of production stages between a sector and final use, computed from the Leontief inverse of the inter-country input-output table. Their index is bounded below by 1 (final-consumption goods) and increases with distance from final demand. The series below is a rank-preserving TiVA-share proxy, 1 + 3 × fwd / (fwd + bwd), mapped to the standard [1, 4] range. Total participation (backward + forward) tracks overall GVC intensity regardless of stage.
Figure 5
Bangladesh: total GVC participation and upstreamness proxy, 1995-2020
Total GVC participation in 2020 is 43.6% of gross exports (up from 43.2% in 1995). The upstreamness proxy is 2.67, on a scale where 1 is pure final assembler and 4 is pure raw-material supplier. Compare against Antràs, Chor, Fally & Hillberry (2012) point estimates: USA 2.81, China 2.91, Saudi Arabia 3.60, Vietnam 2.24.
Source: derived from OECD TiVA 2023 backward and forward shares. Upstreamness proxy is rank-preserving against the formal Antràs & Chor (2013) JPE index, which requires full Leontief-inverse computation on the ICIO table.
Cite: Hossen, M. D. (2026). Bangladesh: total GVC participation and upstreamness proxy, 1995-2020. TradeWeave Workbench.copy permalinkcite
@misc{hossen_2026_fig-gvc-ups-BGD,
author = {Md Deluair Hossen},
title = {Bangladesh: total GVC participation and upstreamness proxy, 1995-2020},
year = {2026},
howpublished = {TradeWeave Workbench},
url = {https://tradeweave.org#fig-gvc-ups-BGD},
note = {Figure: Figure 5}
}
Supplier concentration by HS chapter
GVC participation is volume; GVC vulnerability is concentration. If half of a country’s pharmaceutical imports come from a single partner, a choke-point event (sanction, export ban, factory failure) is a macroeconomic event. The chart below uses the CEPII GeoDep dataset (Arto, Rueda-Cantuche & Corsatea 2024, “Identifying strategic dependencies”) which computes, for every HS6 line, the share of imports sourced from the largest partner (share_odpt). We aggregate to the 2-digit HS chapter, weighting HS6 lines by Bangladesh’s import value in 2022, and report the top 20 chapters by import size.
Figure 6
Bangladesh: import-weighted top-supplier share by HS chapter, 2022
Across Bangladesh’s 20 largest import chapters in 2022, the most concentrated is HS 71 at 94.8% import-value-weighted single-supplier share; the least concentrated is HS 27 at 35.1%. Chapters above ~60% sit in Baldwin & Freeman’s (2022) choke-point range, where a supply shock from the dominant partner cannot be smoothly substituted.
Source: CEPII GeoDep (annual, HS6 × importer) joined to BACI 202501 HS6 imports as weights. share_odpt = import share of the largest origin partner per HS6. Aggregation: Σ (share_odpt × import_value) / Σ import_value within HS2 chapter, top 20 chapters by BACI imports for 2022.
Cite: Hossen, M. D. (2026). Bangladesh: import-weighted top-supplier share by HS chapter, 2022. TradeWeave Workbench.copy permalinkcite
SELECT SUBSTR(i.product_code, 1, 2) AS chapter,
SUM(g.share_odpt * i.import_value) / SUM(i.import_value) AS w_top1
FROM 'data/parquet/country_year_product/**/*.parquet' i
JOIN 'data/parquet/geodep.parquet' g
ON g.hs6 = i.product_code AND g.iso3_importer = 'BGD' AND g.year = i.year
WHERE i.year = 2022
AND i.country_code = 50
AND i.import_value > 0
GROUP BY SUBSTR(i.product_code, 1, 2)
ORDER BY SUM(i.import_value) DESC LIMIT 20;
Position change 2005-2020
The 2005-2020 window covers the last great wave of offshoring, the 2008 crisis, the Antràs & Chor (2018) “GVC slowdown,” the US-China tariff escalation of 2018-19, and the Covid-era supply-chain disruption. Each point below is one economy’s change in backward (Δ FVA/EXGR) against its change in forward (Δ IDC/EXGR) participation over this window. Economies in the upper-right deepened on both margins; lower-left economies retreated; the diagonal separates upstream (forward > backward gain) from downstream shifts. Bangladesh is highlighted in amber.
Figure 7
Change in backward vs forward GVC participation, 2005 → 2020
Between 2005 and 2020, BGD’s backward share moved by +0.6 pp and its forward share by -1.8 pp. Of 86 economies with data in both years, the cross-section is centred near zero on both axes, consistent with the post-2011 plateau documented by Baldwin & Freeman (2022).
Source: OECD TiVA 2023 (DSD_TIVA_MAINLV@DF_MAINLV). Backward = EXGR_FVA / EXGR; Forward = EXGR_IDC / EXGR. Each point is one economy; change computed as 2020 share − 2005 share in percentage points.
Cite: Hossen, M. D. (2026). Change in backward vs forward GVC participation, 2005 → 2020. TradeWeave Workbench.copy permalinkcite
@misc{hossen_2026_fig-gvc-change-BGD,
author = {Md Deluair Hossen},
title = {Change in backward vs forward GVC participation, 2005 → 2020},
year = {2026},
howpublished = {TradeWeave Workbench},
url = {https://tradeweave.org#fig-gvc-change-BGD},
note = {Figure: Figure 7}
}
The top 10 most GVC-integrated industries
Where inside Bangladesh’s export bundle does foreign value added actually sit? Ranking sectors by the level of foreign value added embedded in their gross exports (USD millions) instead of just the share corrects for scale: a 60% FVA share in a tiny sector is a smaller macroeconomic exposure than a 35% share in a sector that accounts for a quarter of gross exports. OECD (2021, Trade in Value Added: Key Insights) recommends this “share × size” reading for identifying the industries that carry a country’s real GVC footprint.
Figure 8
Bangladesh: top 10 industries by foreign value added in gross exports, 2020
No data available for this chart.
No matched FVA-level data for this country’s sectors in 2020.
Source: OECD TiVA 2023 (DSD_TIVA_MAINLV@DF_MAINLV and DSD_TIVA_MAINSH@DF_MAINSH). FVA USD in millions, EXGR_FVA at activity level, partner=World, 2020. Ranked by absolute FVA USD; labels show the sector-level FVA share of gross exports. Aggregation method: OECD (2021) "Trade in Value Added: Key Insights."
Cite: Hossen, M. D. (2026). Bangladesh: top 10 industries by foreign value added in gross exports, 2020. TradeWeave Workbench.copy permalinkcite
@misc{hossen_2026_fig-gvc-topsec-BGD,
author = {Md Deluair Hossen},
title = {Bangladesh: top 10 industries by foreign value added in gross exports, 2020},
year = {2026},
howpublished = {TradeWeave Workbench},
url = {https://tradeweave.org#fig-gvc-topsec-BGD},
note = {Figure: Figure 8}
}
Where the exports go: destination-market composition over time
TiVA tells us how much foreign value is embedded in BGD’s exports; BACI tells us where those exports are absorbed. The four lines below split Bangladesh’s gross goods exports across G7 (USA, CAN, GBR, DEU, FRA, ITA, JPN), EU27 ex-G7 (the other 24 EU member states), China, and Rest-of-World, in current USD. The post-2018 reorganisation of global production runs as much through the destination side (the China-decoupling and near-shoring narratives) as through the input-share side measured in TiVA, a point Antràs (2020, “De-Globalisation? Global Value Chains in the Post-COVID-19 Age,” NBER WP 28115) develops at length.
In 2024, BGD’s gross exports split as G7 51.8%, EU27 ex-G7 27.7%, China 2.0%, and Rest-of-World 18.5%. In 1995 the corresponding shares were 68.9% / 8.5% / 1.1% / 21.5%. The level series (USD) is plotted to keep cross-period comparisons grounded; the share figures above answer the relative-bloc question. A rising China line with a falling G7 line is the destination-side fingerprint of the decoupling narrative; a stable G7 line with a rising RoW line is South-South market diversification.
Source: CEPII BACI 202501 bilateral_year, exports from BGD by importer, 1995-2024, total_value scaled from thousands USD to USD. Bloc definitions: G7 = USA, CAN, GBR, DEU, FRA, ITA, JPN; EU27 ex-G7 = the other 24 current EU member states; China = CHN; Rest-of-World = residual. Framing: Antràs (2020) NBER WP 28115.
Cite: Hossen, M. D. (2026). Bangladesh: gross export destination split, 1995-2024. TradeWeave Workbench.copy permalinkcite
References. Antràs, P. & Chor, D. (2013). “Organizing the Global Value Chain.” Journal of Political Economy 121(5): 2127-2204. Antràs, P., Chor, D., Fally, T. & Hillberry, R. (2012). “Measuring the Upstreamness of Production and Trade Flows.” American Economic Review: Papers & Proceedings 102(3): 412-416. Antràs, P. & Chor, D. (2018). “On the Measurement of Upstreamness and Downstreamness in Global Value Chains.” In World Trade Evolution, Routledge. Johnson, R. C. & Noguera, G. (2012). “Accounting for Intermediates: Production Sharing and Trade in Value Added.” Journal of International Economics 86(2): 224-236. Koopman, R., Wang, Z. & Wei, S.-J. (2014). “Tracing Value-Added and Double Counting in Gross Exports.” American Economic Review 104(2): 459-494. OECD (2024). Trade in Value Added (TiVA): 2023 edition. Paris: OECD. Timmer, M. P., Erumban, A. A., Los, B., Stehrer, R. & de Vries, G. J. (2014). “Slicing Up Global Value Chains.” Economic Policy 29(80): 613-661.