To bridge the transatlantic productivity divide, Europe needs structural reforms—and AI
Comparisons between Europe’s sluggish economy and the dynamism of the United States have become a recurring theme in economic debate. Among the large and growing body of work on this topic are David Marsh’s book Can Europe Survive? and policy reports such as Mario Draghi’s “The future of European competitiveness,” which analyze the factors contributing to slower economic growth in the European Union (EU) since the turn of the millennium.
In particular, the analyses focus on the widening gap between strong productivity growth in the United States and near stagnation in the EU. Their policy recommendations emphasize structural reforms to make Europe’s economy more flexible and dynamic, foster a risk-taking culture similar to that in the United States, and encourage innovation and its commercial applications, thereby boosting productivity and economic growth.
While such recommendations are reasonable, it is important to recognize that structural reforms are difficult to implement and tend to face resistance from vested interests. Moreover, reforms take time to yield results and usually produce near-term pain, fueling popular backlash. These factors explain why the EU has made limited progress in implementing Draghi’s recommendations since September 2024. Nonetheless, if Europe wants to preserve its strategic autonomy or sovereignty, especially amid heightened geopolitical contention, it must seriously undertake such reforms.
In the foreseeable future, Europe should also aim to facilitate the diffusion of highly productive business practices already demonstrated by many high-tech firms, particularly in the United States, to domestic businesses. Such an effort could produce tangible results, helping improve economic performance. This challenge mirrors that facing the rest of the US economy, which lags behind the information technology (IT) and artificial intelligence (AI) sector in productivity performance.
Europe struggles to keep pace with US productivity
Since 2000, real GDP annual growth has averaged 1.3 percent in the EU, less than half that of the United States. Factors explaining this underperformance include rigid and burdensome regulations, fragmented markets for services and capital, high energy costs, slow labor-force growth amid population aging, and inadequate investment, including in research and development, particularly in IT and AI sectors. Against that backdrop, EU labor productivity has slowed—falling from 1.5 percent per year between 1999 and 2008 to 1.0 percent between 2010 and 2019 and 0.4 percent since then. By comparison, US labor productivity has averaged 1.5 percent per year over most of the post-2000 period, rising to 2.4 percent in the past two years.
Overall, the productivity gap favoring the United States over the Eurozone has grown to 41 percentage points—more than making up for the period from the end of World War II to the 1980s, when many European countries grew faster than the United States.
The tech sectors behind US success
The better productivity performance of the United States compared with the EU has been driven mainly by the IT sector and later reinforced by AI activities. Indeed, between 1988 and 2023, the cumulative total factor productivity growth in the IT sector exceeded 178 percent while remaining only 12 percent for non-IT sectors. As investment in the IT and AI sectors has outpaced overall capital expenditure, these activities have contributed increasingly to US economic growth.
In 2025, IT- and AI-related expenditures, adjusted for imported equipment, accounted for up to 25 percent of US GDP growth (estimated at 2.2 percent), despite comprising only 4.5 percent of the economy. Excluding the IT and AI sectors, the rest of the US economy would have grown by 1.65 percent in 2025—no different from the EU, estimated at 1.6 percent.
Why high-tech gains don’t reach the broader economy
It is not only in the United States that the IT and AI sectors—and, more broadly, the high-tech sector—enjoy stronger productivity growth than the rest of the economy. In fact, this phenomenon holds true across the member states of the Organization for Economic Co-operation and Development (OECD). According to OECD data, high-tech firms that have embraced technological advances in IT and AI are at the frontier of productivity, growing at an annual rate of 3.5 percent. By comparison, non-frontier firms have recorded a meagre 0.5 percent productivity growth, dragging down aggregate productivity figures.
Recent examples of non-frontier firms include those that have failed to embrace technological innovations, such as digital transformation and electric mobility trends. This includes many German automobile companies, which are now under intense pressure in their home market from Chinese competitors and have lost market share in China from 25 percent to 14 percent in recent years. Similarly, about 70 percent of US firms have failed in their digital transformation efforts due to a lack of a clear strategy, poor change management, and employee resistance—thus joining the ranks of laggards. More generally, non-frontier firms tend to be small and medium-sized enterprises, forming the backbone of the US and EU economies, employing the largest share of workers in sectors such as retail, construction, and traditional manufacturing.
The lack or slow pace of technology diffusion from high-tech frontier firms has left much of the US and EU economies populated by laggard firms in terms of adopting technologically enabled productivity improvements. This kind of slow diffusion of new technology isn’t new. For example, it took sixty years for the steam engine, thirty-two years for electricity, and fifteen years for personal computers and the internet to be widely adopted.
Several factors contribute to this phenomenon, including high fixed costs of investing in new technology, workforce training, and organizational restructuring to reap the benefits of innovation. Investment in intangible assets such as workforce training often produces a technology J-curve effect: productivity may dip initially as firms and employees adapt, before accelerating once the new methods are mastered. There is hope that AI technology diffusion might occur faster than in the case of the aforementioned examples, though adoption remains demanding, particularly in terms of the high costs of implementation hurting short-term profitability.
Learning from frontier firms
Instead of waiting for AI technology to naturally diffuse to the rest of the economy, policymakers, business leaders, and investors should actively facilitate, incentivize, and reward laggard firms to adopt productivity-enhancing practices demonstrated by frontier firms. Specifically, these measures should address barriers to technology diffusion. By doing so, laggard firms would improve their productivity, thereby boosting aggregate productivity and supporting stronger economic growth.
For the EU, encouraging laggards to catch up with frontier firms, both domestically and in the United States, could produce measurable results in the near term. But meanwhile, the EU and United States should continue to implement long-term structural reforms to develop the whole innovation ecosystem—including fostering a risk-taking culture to promote innovation and commercialization.
Importantly, for both the EU and the rest of the US economy, shortening the technology diffusion lag carries high stakes—failure would entrench the already dominant frontier firms, weaken market competition, and make it harder for other companies to catch up. This would keep large parts of the US and European economies in slow-growth lanes, exacerbate economic inequality, and fuel social divisiveness.
Hung Tran is a nonresident senior fellow at the Atlantic Council’s GeoEconomics Center, a senior fellow at the Policy Center for the New South, a former executive managing director at the Institute of International Finance, and a former deputy director at the International Monetary Fund.
At the intersection of economics, finance, and foreign policy, the GeoEconomics Center is a translation hub with the goal of helping shape a better global economic future.
Further reading
Mon, Jan 26, 2026
Inside the biggest Davos debates (other than Greenland)
Dispatches By Josh Lipsky, Alisha Chhangani
As the annual World Economic Forum in Switzerland ends, the issues discussed—from tariffs to AI—will continue to play out in all corners of the world.
Mon, Jul 21, 2025
Why European businesses are now stuck in the middle of an EU-China storm
Econographics By Alisha Chhangani
If relations continue to deteriorate, the world’s three major economies could find themselves in economic conflict and European businesses will be caught squarely in the middle.
Thu, Jan 30, 2025
US economic outlook 2025: It’s the productivity, stupid!
Econographics By Hung Tran
The range of forecasts for US economic growth in 2025 is unusually wide. Productivity is going to be a major reason for slow or strong growth prospects.