The globalisation that goes unseen

We have been telling the same story for years. Global supply chains are shortening, reshoring is on the rise, geopolitics is taking precedence over economics, and globalisation is entering its twilight years. Then comes an OECD study suggesting a more intriguing possibility: perhaps part of globalisation is not disappearing at all; it is simply that we are no longer able to see it. The study is entitled ‘Participation in GVCs through capital goods and services’ and is particularly interesting for the way it changes the picture. Traditional statistics on trade in value added (TiVA) track the international flow of intermediate inputs effectively, but treat capital as final demand. If an Italian factory buys German machinery, American software or uses technology developed elsewhere, a significant part of its dependence on foreign sources thus tends to disappear from the reconstruction of the production chain.

The OECD is attempting to reintegrate that capital into the chain. And the result is remarkable. Taking into account machinery, equipment, infrastructure and intangible assets such as software, research and development, and databases, the average participation of OECD countries in global value chains increases by around 11 percentage points. The share of services incorporated into manufacturing exports rises by around 9 percentage points. This does not mean that the old statistics were wrong. It means that they were better at capturing an economy in which trade coincided with the physical movement of goods, and less effective at capturing an economy in which value shifts without being recorded at customs. And this is where the story becomes political. For years, we have interpreted every change in production chains as evidence of globalisation’s retreat: reshoring, friendshoring, tariffs, subsidies, investment controls, economic security, trade wars. All true. But it does not automatically follow from these phenomena that the world is becoming more nationalised. Supply chains can change their geographical location, replace some suppliers with others, incorporate more technology and more services, and yet continue to be international. Another OECD report this summer showed that, in 2024, the use of imported goods and services in global production was close to historic highs and that trade linked to global supply chains accounted for around 17 per cent of global GDP. The most interesting finding, therefore, is not the end of interdependence, but its transformation. Globalisation in 2005 was evident in Chinese ports, in containers and in relocated factories. Today’s globalisation is also evident in an Italian company that uses American software, a German machine, a Dutch patent and research developed in France. This helps to explain why ‘de-risking’ and ‘deglobalisation’ are not synonymous. Reducing a critical dependence on China may mean increasing the number of suppliers in Vietnam, India, Mexico or Eastern Europe. Producing more semiconductors in the West may mean building new supply chains that rely on Dutch machinery, Japanese materials, American design and European energy. Economic security can make supply chains less concentrated without making them any less global. Diversification does not mean autarky. Resilience does not mean self-sufficiency.

There is, however, an even more important point. The greater the weight of intangibles becomes, the more the distinction between industry and services becomes outdated. A machine tool is no longer just steel: it contains software, support, sensors, data, design, maintenance and patents. A car is increasingly an industrial product built on a growing array of services. If one measures only what physically crosses the border, one underestimates precisely the part of the economy that is growing the most.

For Italy, the lesson is twofold. A country integrated into European supply chains, with strengths in engineering, machinery, components, design and industrial services, risks being misrepresented if one looks solely at the trade balance or the number of factories that have been brought back home. For an economy like ours, the greatest danger is not being too interdependent, but becoming peripheral in relation to the new interdependencies built around technology, capital, research and advanced services.

This does not mean denying the problems of globalisation. Strategic dependencies exist, geopolitics matters, some suppliers are risky, and certain production sectors need to be protected or diversified. But it is one thing to correct globalisation and quite another to declare it over. The OECD’s findings serve to remind us that before celebrating the demise of an economic system, we must be sure we know where to look. Because it may well be that globalisation is not disappearing. It may simply have become more difficult to measure.