A service of the

Download article as PDF

­A recurring question over the last few decades, not only in Europe but in much of the developed world, is: Why has economic growth slowed? This recent slowdown seems very different from the explicable and, in many ways, anticipated fall in growth rates following the rapid catch-up growth after World War II. The experience of the 21st century – growth rates of GDP per capita falling to 1% or even to 0% – seems to imply that something has gone wrong within the European economy.

Looking at the situation from afar, I am less certain that this represents some kind of failure of policy or direction. The decline in economic growth rates is consistent across the developed world. The U.S., Canada, South Korea and Japan, among others, have also experienced notable drops in their growth rates in the last two decades. Compared to Europe, some of these – the U.S. and Canada – still have higher growth rates, while others – Japan, in particular – seem to be stuck with the same anemic growth that European policymakers and economists worry about.

One can be tempted to look for global reasons for this, as it affects so many nations with such different economic histories; China’s entry to world trade is an obvious point of emphasis here. But I am less certain that Europe and the rest of the developed world are living through some kind of persistent global shock related to innovation or poor policy. Rather, I think what we see in these areas is the consequence of being rich and having already experienced rapid growth in the past. We are the grandchildren that Keynes (1932) speculated over in his essay, living in an economy that can fulfill material needs for nearly everyone, and leaving us open to make choices that may not translate to rapid economic growth.

The standard ways that we examine economic growth are not equipped to deal with that. Textbook models of economic growth (including one that I wrote) presume that the productivity growth driving growth rates in the long run is persistent. Robert Solow took that productivity growth rate as a given, but even when Paul Romer and others developed theories to explain that productivity growth rate, a defining feature of those models was the delivery of a constant growth rate of productivity in the long run. Nearly every paper on economic growth written in the last 40 years is built to deliver a “balanced growth path” that implies that growth rates of the mid- to late 20th century will go on forever. Hence, the recent slowdown in growth appears to be an anomaly.

What would economic growth look like if we no longer presume that growth must be constant? It is useful to take seriously the process of structural change and what it implies about demand for the output of different industries – food, clothing, housing, transportation, health care, etc. – and how that interacts with productivity growth in those industries to determine aggregate economic growth.

The demand for different industries is subject to both income and substitution effects. In Europe’s recent economic past, living standards were much lower and the influence of income effects on demand across industries was likely more salient. The most obvious case of this is the transition out of agriculture. Europe (like essentially every other economy) gained productivity in agriculture that lowered the relative price of food but notably shifted workers, capital and spending into non-agricultural industries like manufacturing as people demanded appliances, cars and better housing with the money they saved. As manufacturing tended to have rapid productivity growth rates, this accelerated economic growth during the middle of the 20th century.

But there were limits to how much people wanted to spend on manufacturing as well, much like with food, and as productivity growth continued in that industry, it led to increased spending in service industries. Unlike that earlier transition, services did not necessarily have high productivity growth, and hence demand was pushing spending into low growth products. The essential story here, identified by William Baumol in the 1960s, is consistent with his explanation for the “cost disease of services”.

I think that developed areas like Europe, the U.S. and Japan must acknowledge that this process need not stabilize around a persistent growth rate of aggregate GDP. As we have become very rich, the power of income effects to drive structural change is likely waning and our economies are governed more by the substitution effects working through our demand for output of different industries.

Notably, data suggest that we treat the output of most industries as complements, and hence are unwilling to adjust our spending in response to relative price changes. This means that our expenditure on industries tends to move the same direction as its relative price, and therefore in the opposite direction to its relative productivity growth. We spend more on industries whose productivity growth is low, and less on industries whose productivity growth is high. This has immediate and direct effects on aggregate economic growth, which – roughly – uses that spending to weight the productivity growth of each industry. It need not be the case that productivity growth within industries is slowing down, but rather that the nature of our demand is now acting against higher growth.

If we view industries as complements, then over time economic growth will continue to drop to match the growth rate of the slowest-growing industries, and they will occupy the majority of our spending. Unlike the presumption of persistent balanced growth, the nature of economic growth in developed economies is likely one of steadily declining growth rates.

Is this a problem? To the extent that certain fiscal policies were designed under the presumption of persi­­stent growth, this creates hard political discussions about the right way to allocate spending and set tax rates. But those are not economic problems per se. It is hard to say our preferences are wrong, despite the effect on economic growth, just as it would be hard to say that our preferences are wrong because we decided to stop eating at a certain restaurant or buy a certain brand of clothes.

Perhaps Keynes would not be that surprised by the situation that his grandchildren find themselves in, and he would not see it as dire. He imagined that the structure of the economy would be quite different once we were rich enough to accommodate almost any material need. This does not imply there are not difficult decisions to make in response, but appreciating that our own wealth has led us into a world of slow growth is necessary to avoid chasing a solution – persistently high growth rates – that is out of reach for Europe and the rest of the developed world.


Keynes, J. M. (1932), Economic Possibilities for our Grandchildren (1930), Essays in Persuasion, Harcourt Brace, 358-373.

Download as PDF

© The Author(s) 2023

Open Access: This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/).

Open Access funding provided by ZBW – Leibniz Information Centre for Economics.

DOI: 10.2478/ie-2023-0070