The New Manufacturing Arc

Between 1995 and 2023, China's share of world goods exports rose from 4.52% to 15.13% in this BACI extract - a gain of 10.61 percentage points that dwarfs every other country. No single economy comes close, and any honest map of the last three decades of trade has to put that fact first.

But the second tier of the same leaderboard tells a quieter, more distributed story. Rank the world’s exporters by how many points of global share they added over those 28 years, and after China the next six names are not the established powers of 1995. They are Viet Nam (+1.77pp, to 1.88%), India (+1.2pp, to 1.95%), the United Arab Emirates (+1.2pp), Mexico (+1.03pp, to 2.67%), Poland (+1.01pp) and Türkiye (+0.73pp). They sit, in order, at ranks 2 through 7 on the all-country share-gain list, immediately behind China.

Spread across a map, these six trace an arc: Southeast and South Asia, the Gulf, North America, and Central and Eastern Europe. None of them individually rivals China. Together they describe where the share that did not go to China - and the share that left the older manufacturing powers - actually landed. This is a story about the reallocation of export share, not a causal account of industrial policy, and the rest of the piece keeps that distinction in view.

The slope chart makes the scale difference unmissable. China’s line climbs 10.61 points; the steepest of the others, Viet Nam, climbs 1.77. In raw share terms China gained roughly six times what Vietnam did. Treating the arc as a 'rival’ to China would be a category error, and we do not.

What the leaderboard does show is breadth. Viet Nam effectively started from nothing - 0.11% of world exports in 1995 - and reached 1.88%, a more than seventeen-fold rise in share. India and the United Arab Emirates each added 1.2 points, but from very different starting points and for very different reasons, as later sections show. Mexico and Poland both added almost exactly a full point (1.03 and 1.01), extending the arc into North America and the European Union. Türkiye rounds it out at +0.73.

For context, the eighth-placed gainer overall is the Russian Federation at +0.60 points - below every country in the arc and driven by a different (energy-price) dynamic. The arc, in other words, is not an arbitrary selection: it is essentially the contiguous block of the world’s largest manufacturing-share gainers once China is set aside. A reader should still keep two caveats close: these are nominal USD shares, so price moves and exchange rates are baked in, and a country’s share reflects its size as much as its dynamism.

Plotting world export share year by year from 1995 to 2024 shows that the arc is not one phenomenon but several, with different timing and shape.

Viet Nam's curve is the most dramatic: a near-flat line through the late 1990s that bends sharply upward through the 2010s and is still rising, reaching a peak of 2.21% in 2024 - above its 2023 reading and the highest in the series. India climbs more steadily and also peaks late, at 2.01% in 2024. Mexico, which already held 1.63% in 1995, shows a slower percentage climb but the highest absolute level of the group, topping out at 2.83% in 2024.

The United Arab Emirates is the outlier in shape: its share peaks earlier, at 1.73% in 2022, and is visibly sensitive to the commodity cycle rather than to a factory build-out - a hint, confirmed below, that the UAE belongs to this list for reasons unlike the others. Poland and Türkiye both reach their series highs in 2023 (1.5% and 1.21% respectively), the European end of the arc still trending up at the end of the window. Because these are annual shares of a growing world total, a flat line means 'growing with the world’ and only a rising line means ‘gaining ground’.

Decompose each country’s exports into the 21 Harmonized System sections and a common pattern appears for the manufacturers: the share of Machinery and electrical equipment (HS Section XVI) rises, and lower-value categories recede.

Viet Nam is the textbook case. In 1995 its top export sections were Vegetable products (23.64%), Mineral products and Textiles; machinery and electrical equipment was a rounding error at 1.19% of exports. By 2023 that single section was 54.37% of everything Vietnam exported - a swing of +53.18 points - while the old mineral-products line collapsed from 21.52% to 1.42%. Poland and Türkiye show milder versions of the same move: in both, machinery is now the largest section (27.31% for Poland), while textiles - once Türkiye’s top earner at 36.39% - fell to 14.15%. Mexico was already a machinery exporter in 1995 (34.45%); its visible shift is the rise of Transport equipment (HS XVII) to 26.62% - the automotive build-out.

The United Arab Emirates breaks the pattern, and that is the point. Its top 2023 export section is still Mineral products at 47.29%, down from 78.08% but still dominant, and its fastest-growing section is Precious stones and metals (+17.37 points, to 19.31%) - the signature of a gold-trading and re-export hub, not a factory floor. India sits in between: machinery rose to 16.65%, but its mix stayed broad, with Chemicals and Mineral products close behind and textiles falling from 26.25% to 8.49%. Note that these are HS goods sections only; India’s large services exports are outside BACI entirely.

Where each country’s exports go - using the reporter’s own exports-to-partner records - shifted just as sharply between 1995 and 2023, and the direction is revealing.

Viet Nam's single largest market in 1995 was Japan, taking 31.67% of its exports. By 2023 the top destination was the USA at 27.76%, with China second at 20.21% - a country that is simultaneously Vietnam’s largest supplier of inputs. That pairing is the clearest data signature in this story of a 'China + 1’ assembly model: components in from China, finished goods out to the United States.

Mexico shows the opposite of diversification. The United States already took 77.4% of Mexican exports in 1995 and still took 75.76% in 2023 - a deep, durable bilateral relationship rather than a widening web. Poland is the European mirror image: Germany remains its top market at 24.77%, and Poland reads less like a standalone exporter than like an integrated tier of the German industrial supply chain. India is the most genuinely rebalanced: the United States leads at 19.39%, but the United Arab Emirates has risen to second at 6.53%, reflecting Gulf-facing energy and goods trade. These are export destinations by value, and they say nothing about shipping routes - only about where the buyers booked the trade.

Lining up the four core arc economies for 2023 shows that similar share gains can sit on very different structures.

Viet Nam is the most export-specialised and the most transformed: $423.9B of exports, a goods-trade surplus of $146.2B, world export rank 14, and a more than 78.3-fold increase in nominal exports since 1995 - the steepest in the group. Its exports are also the most concentrated by product, with machinery alone at 54.37%. Mexico is the largest exporter of the four ($602.5B, rank 8) and the most single-market dependent: 75.76% of its exports go to the United States.

India stands apart on balance. It runs by far the largest goods deficit of the four - about $226.3B (Poland also runs a smaller deficit, about $25.1B, while Vietnam and Mexico post surpluses) - because its imports ($667.4B, import rank 7) far exceed its goods exports, and its export mix is the most diversified, with its top section only 16.65%. Poland is the smallest exporter here ($339.0B, rank 22) and the most Europe-embedded, sending 24.77% of its exports to Germany. Four countries, one shared headline of rising share - and four genuinely different economic engines underneath.

This is a measured account of one thing: the reallocation of world export share among countries, computed from CEPII BACI bilateral goods-trade values, 1995 to 2023, with 2024 shown where the series extends. Every figure here is reproducible from the committed generator script against the same data.

It is not a causal explanation. The numbers show that share moved toward Viet Nam, India, the United Arab Emirates, Mexico, Poland and Türkiye; they do not, by themselves, attribute that movement to tariffs, wage gaps, trade agreements, 'friend-shoring’ or any specific policy. Several caveats bind the interpretation. First, all values are nominal US dollars, so commodity-price swings and currency moves are included - which is exactly why the United Arab Emirates, a minerals-and-gold hub rather than a manufacturer, appears on a 'manufacturing arc’ leaderboard, and why its share tracks the energy cycle. Second, share reflects size: a percentage point means something different for a country starting at 0.11% than for one starting at 1.63%. Third, BACI is goods only - a critical omission for India, whose services exports are a large part of its real external sector and are entirely absent here. Finally, partner figures are exporter-reported destinations by value, not evidence of physical shipping routes. Read as a map of where goods-export share went, the arc is solid. Read as a verdict on who ‘won’, it would overreach the data.