A sustained and increasingly technical argument about European economic performance has been running across economics blogs and Substacks since early 2026. The starting gun was fired by Seth Ackerman in a series of posts in February — “Eurpoors vs. Florida Man,” “Europe’s productivity keeps outpacing the US,” and “Productivity statistics, Paul Krugman, and Hicks’s shadows on the face of history” — arguing that the apparent divergence between US and European productivity is largely a statistical artefact of how GDP is measured. Dean Baker reached a similar conclusion in “The Grand Illusion: The US-Europe Growth Gap.” Paul Krugman took up the argument in May, developing it across four posts: “Is Europe in Economic Decline?” (May 10), “What Happens When Americans Realize How Miserable We Are?” (May 12), “Modeling the US-Europe Paradox” (May 12), and “Challenging the Narrative of European Decline, Continued” (May 17). The responses have been pointed. Noah Smith argued in “Yes, Europeans are poorer than Americans” (May 15) that the evidence for a real income gap is stronger than Krugman allows. Pieter and Luis Garicano, writing in Silicon Continent, mounted the most technically detailed case for the prosecution in “European stagnation is real” (May 12). The “Technology as Nature” blog responded directly to Baker in “Not So Illusory” (March 16), defending the constant-price methodology.
The core dispute can be stated concisely. Measured in constant prices, the standard method for tracking real economic growth, US GDP per capita grew 37 per cent between 2001 and 2024, while Germany grew 24 per cent and France only 18 per cent. This looks like a widening gap. Yet when the comparison is made at current purchasing power parity — the value of what each economy produces at prevailing prices — the gap has barely moved: France was 74 per cent of the US level in 2001 and 73 per cent in 2024. Krugman argues that the constant-price divergence is a statistical artefact of the US economy’s dominance of information technology production, where rapid output growth is paired with falling prices; the PPP measure, he claims, is the appropriate one for welfare comparison. His critics respond that constant-price divergence reflects a genuine and widening gap in productive capacity — one that matters for Europe’s ability to invest, innovate, and project geopolitical influence. I have surveyed these arguments, and the responses they have attracted, in my previous posts on this debate.
What I want to raise here is a question that neither side has asked: whether GDP, in any of its variants, is the right unit in which to conduct the argument at all.
The Problem with Prices
The fundamental problem with GDP as a measure of economic performance is not the familiar one — that it ignores inequality, or fails to count volunteer work, or treats a car accident as a positive because it generates repair bills. These are real limitations, but they are in principle correctable. The deeper problem is constitutive. GDP aggregates the value of goods and services by using market prices as weights. In a world in which prices reflect relative social costs — as competitive equilibrium theory holds they do — this is a defensible approach. In the world we actually inhabit, it is not.
Consider what market prices are measuring in the present economy. When Apple charges €1,400 for an iPhone in Paris, the price is not a signal from a competitive market reflecting marginal cost. It is the maximum that Apple can extract from French consumers given its patent monopoly, its vertical integration, and the switching costs it has carefully engineered over two decades. Roughly 40 per cent of that price is gross margin — rent, not production value. GDP counts it all. An economy that hosts Apple’s intellectual property registrations or its corporate headquarters receives credit in national accounts for productive activity that is substantially a matter of legal and regulatory geography rather than physical production.
The Garicanos deploy the combined equity market capitalisation of young US firms — $42.9 trillion, against roughly $5 trillion for all European listed companies — as evidence of European productive failure. It may indeed be evidence of that. But equity market capitalisation measures the present value of expected future profits, which is to say it measures the market’s estimate of future rent-extraction capacity as much as future productive capacity. A monopolist with a durable patent portfolio will have a higher market capitalisation than a competitive industry with identical physical output. The US technology sector’s capitalisation reflects both genuine innovative dynamism and a remarkable ability to erect and sustain barriers to competition. Treating the two as a single measure of productive superiority confuses the surgeon with the bill.
The Standard Commodity: A Different Measure
What would a genuine measure of productive capacity look like, if not one built from prices? Economic theory offers a suggestive answer, though not yet an operational one, in the concept of the standard commodity — an idea developed to address precisely the deficiency that afflicts GDP.
The problem GDP inherits from the price system is this: prices are not independent of the distribution of income between wages and profits, nor of the degree of market power firms exercise. When those things change, prices change — and so does measured GDP, even if the physical economy has not changed at all. A measure of output that is invariant to these distributional and competitive factors would need to be built not from prices but from the underlying structure of production itself: what each industry uses as inputs and produces as outputs, measured in physical quantities. The standard commodity is a composite good constructed from precisely these production relationships. Because it reflects what an economy can physically make — given its technology and resources — rather than what prices it can charge, it is unaffected by the rent-extraction capacity of dominant firms or by the legal geography of profit-booking. An economy that produces the same physical output as another but charges higher prices for it, because it has more market power, scores no better on this measure. The score improves only when the underlying productive capacity improves.
The immediate relevance to the Krugman debate is this. If a substantial fraction of the divergence in US and European GDP over the past quarter-century reflects growing US corporate market power — rising price-cost margins, increasing platform dominance, the agglomeration of IP rents in US-domiciled entities — rather than growing US physical productive capacity, then a standard-commodity measure would show a narrower divergence than either Krugman’s critics or the GDP data imply. The productive gap between the US and Europe may be real; it is very likely smaller, and differently distributed, than current measures suggest.
Neither Measure Escapes
It is worth noting that Krugman’s preferred measure GDP at current purchasing power parity does not fully escape the problem either, though it avoids some of its worst manifestations. PPP conversions estimate relative price levels using a basket of goods and services whose composition embeds contestable judgements about what counts as comparable output. US healthcare, which is very expensive and constitutes roughly 18 per cent of US GDP, enters the PPP basket as output whose high price is treated as reflecting high quality. Whether the additional cost of American healthcare represents additional value to American patients is, to put it gently, contestable. American healthcare is expensive partly because it is delivered through a highly concentrated, administratively complex system that extracts large rents from its captive market. On a production-structure measure, administrative complexity is not productive output; on the PPP measure, it registers as such.
Both the constant-price and the PPP measures, in other words, share the foundational assumption that prices — appropriately corrected — are a reliable guide to comparative value. In an economy of significant market power, administered pricing, and global supply chains designed partly around tax arbitrage, that assumption is under severe stress. They disagree about which prices to use; they agree, silently, that prices are what should be used.
What Follows
The distinction between productive capacity and rent-extraction capacity has direct implications for how Europe should read its own situation. European firms are more deeply embedded in global technology supply chains than national GDP figures suggest: German machine tools, Dutch semiconductor equipment, French aerospace components, and Italian precision engineering all contribute to final products whose value is attributed elsewhere in national accounts. The Trade in Value Added data compiled by the OECD reveal a more active European participation in the technology economy than headline GDP comparisons imply. The productive gap is real; it is smaller than constant-price GDP suggests, and it is concentrated in particular areas — above all, in the generation of new frontier technology firms — rather than spread uniformly across the economy.
The genuinely troubling divergence, and one that a production-structure measure would capture directly, is in the capacity to generate new high-value enterprises. Europe produces too few firms capable of operating at global technology frontiers, and those it does produce tend to be acquired before they can scale. This is partly a single-market problem, partly a capital-market problem, and partly a reflection of the self-reinforcing agglomeration dynamics that Krugman himself seminally described. The paradox is that Krugman’s own theoretical framework, which explains why Silicon Valley exists and why it persists, is the most compelling account of the European technology gap that the standard GDP debate has consistently failed to illuminate.
The debate between Krugman and his critics is, at root, a debate about what we are trying to measure and why. Krugman wants to measure welfare; the Garicanos want to measure economic power and future capacity. Both are legitimate questions. They are also compatible: European citizens can be living comparably to or better than Americans on most welfare measures, while the European economy loses ground on the measures that determine future investment capacity. The error is to treat these as a single question with a single answer. A production-structure measure, the kind of thing the standard commodity concept points towards, would help distinguish the two, and in doing so would clarify what European policy actually needs to address.
That is, I think, the most important gap in the current debate. It is not a small one.
Bibliography
Paul Krugman, Substack (paulkrugman.substack.com)
- “Is Europe in Economic Decline?” (May 10, 2026)
- “What Happens When Americans Realize How Miserable We Are?” (May 12, 2026)
- “Modeling the US-Europe Paradox (Very Wonkish)” (May 12, 2026)
- “Challenging the Narrative of European Decline, Continued” (May 17, 2026)
Noah Smith, Noahpinion (noahpinion.substack.com)
- “Yes, Europeans are poorer than Americans” (May 15, 2026)
Pieter Garicano and Luis Garicano, Silicon Continent (siliconcontinent.substack.com)
- “European stagnation is real” (May 12, 2026)
Seth Ackerman, The Informer (theinformer.substack.com)
- “Eurpoors vs. Florida Man” (February 2026)
- “Europe’s productivity keeps outpacing the US” (February 2026)
- “Productivity statistics, Paul Krugman, and Hicks’s shadows on the face of history” (February 2026)
Dean Baker, Beat the Press (CEPR)
- “The Grand Illusion: The US-Europe Growth Gap” (February 2026)
Technology as Nature
- “Not So Illusory” (March 16, 2026)
Background
- Piero Sraffa, Production of Commodities by Means of Commodities (Cambridge University Press, 1960)
- OECD Trade in Value Added (TiVA) Database
- World Input-Output Database (WIOD)