The latest mainstream thinking in economics
Michael Roberts is an Economist in the City of London and a prolific blogger.
Cross-posted from Michael Roberts’ blog
This graphic is licensed under the Creative Commons Attribution-Share Alike 4.0 International license.
ASSA is a massive conference of economists from around the world, organised by the American Economics Association. (AEA). Every year thousands attend and there are hundreds of sessions and papers. Most of these are mainstream, but there are sessions organised by heterodox organisations like the Union for Radical Economics (URPE). This year’s conference took place in Philadelphia.
Each year, I divide my report into two posts: the mainstream and the radical heterodox. Let’s start with the mainstream. As in 2025, the dominating theme was artificial intelligence (AI) and its impact on economies. There was a livestreamed session entitled AI and Productivity: is this time different? The assembled speakers pretty much agreed that AI would be a game changer, at least for US productivity growth. Yes, AI’s impact on productivity growth could take the form of a J-curve, namely that its adoption in industry could briefly lower productivity if skilled workers were sacked and new workers took time to adapt to AI usage. But eventually, AI adoption would become a general purpose technology (GPT) that would deliver a step-change in labour productivity. That was the theme of the first paper by Erik Brynjolfsson of Stanford University, the AI optimist and J-curve theorist.
Economists from the US Fed and the Brookings Institution took this optimism further in their paper. Some labour-saving innovations, such as the light bulb, temporarily raise productivity growth as adoption spreads, but the effect fades when the market is saturated; that is, the level of output per hour is one-off higher, but the growth rate is not. In contrast, two types of technologies stand out as having longer-lived effects on productivity growth. First, there are technologies known as general-purpose technologies (GPTs), as above. Second, there are inventions of methods of invention (IMIs). IMIs increase the efficiency of the research and development process, generating new ideas more quickly and cheaply; the compound microscope is an example. These economists argue that AI has the characteristics of both a GPT and an IMI. “Because both GPTs and IMIs promote productivity growth for extended periods, it is reasonable to expect genAI will have a noteworthy impact on productivity.”
Economists from the OECD took this up in their paper by attempting to quantify the impact of AI on productivity growth, arguing that AI could contribute between 0.3-0.9 percentage points to annual TFP growth over the next decade. Total factor productivity is a neoclassical economics measure of ‘innovation’ as a ‘residual’ in the aggregate account of labour productivity. What the OECD guys basically claimed was that if labour productivity growth were, say 1% a year, it would eventually be doubled by AI adoption and applications.
Economists in another session calculated the likely gain in labour productivity. “First, we argue that AI can already be seen in productivity statistics for the United States. The production and use effects of software and software R&D (alone) contributed (a) 50 percent of the 2 percent average rate of growth in US nonfarm business labor productivity from 2017 to 2024 and (a) 50 percent of its 1.2 percentage point acceleration relative to the pace from 2012 to 2017.“
Second, taking additional intangibles and data assets into account, “we calculate a long-run contribution of AI to labor productivity growth based on assumptions that follow from the recent trajectories of investments in software, software R&D, other intangibles, and productivity growth in both US and Europe. Our central estimates are that AI will boost annual labor productivity growth by as much as 1 percentage point in the United States and about 0.3 percentage point in Europe.”
So AI optimism ruled at ASSA.
US tariffs and the future of the dollar was another theme of the main sessions. Some New Keynesian economists from Brown University in their paper reckoned that, with tariffs, there should be a 1.6 pp decline in US output, a 0.8 pp rise in inflation, and a 4.8% appreciation of the dollar. Even just the threat of tariff increases would lead to a 4.1% appreciation of the dollar, 0.6% deflation, and a 0.7 pp decline in output. So why has the dollar weakened since Trump’s tariff moves last April? The authors claim that this is due to ‘tariff policy uncertainty’. “Uncertainty reduces demand and generates a risk-premium wedge that weakens the currency of the tariff-imposing country.” They concluded that Trump’s tariff policy thus suggests “the dollar’s “exorbitant privilege” may not be permanent.” Major policy errors, including trade policies that heighten uncertainty, may accelerate an erosion of confidence.
This argument complemented the view of the long-standing mainstream expert on debt, Kenneth Rogoff. His presentation was based on his latest book, Dollar decline; the failures of mainstream economics and epochal crisis – reviews – Michael Roberts Blog. Rogoff argues that, such is the rising level of public debt in the US that it threatens the privileged position of the US dollar in world markets. This risk is made worse by Trump’s establishment of cryptocurrency stablecoins as part of the government’s borrowing tools.
But the dollar’s role as a reserve currency and as a safe haven still has a long way to run. The dollar’s effective exchange rate against other currencies is still well above where it was in 2007. The Long Depression since 2008 has actually been a period of dollar strength as other major G7 economies weakened relative to the US and the BRICS currencies made little headway.
Did the US Federal Reserve make mistakes in its monetary policies both during and after the COVID pandemic that led to the significant spike in inflation rates after the end of the slump? Should the Fed change its mandates that maintain a balance between employment and inflation and instead concentrate on controlling inflation as other central banks do? What will happen to the Fed when Jay Powell leaves this summer and is replaced by an appointee of Trump’s?
These questions were considered by the great and good in mainstream economics, including former Fed chair, Janet Yellen. Echoing the hardline Fed chair Paul Volcker of the 1980s, some speakers reckoned that the dual mandate was no good and helped inflation get out of control because the Fed did not move quick enough to raise its policy rate and reverse quantitative easing (QE) adopted during the COVID slump.
Others said that the problem was that monetary policy was helpless when fiscal profligacy from government spending, as during COVID, was so huge. ‘Fiscal dominance’ meant that the Fed could not stop inflation rising. Naturally, Janet Yellen, Biden’s former Treasury Secretary, denied that there had been ‘fiscal dominance’ on her watch.
What was interesting was that all agreed that the Fed has little or no traction over inflation, whether because of ‘fiscal dominance’ or because price rises were supply, not demand, driven. This is hardly an endorsement for the purpose of the Fed, whether it remains ‘independent’ of political influence (apparently good) or comes under the thumb of Trump.
Indeed, this view that inflation is supply-driven was supported in another session where two economists argued that the key condition for inflation bursts is supply deficiency—caused either by exogenous shocks to aggregate supply (supply disruptions), or by excessive aggregate demand (supply-constrained demand booms). “Aggregate demand booms, even big ones, that do not trigger supply deficiencies, are not inflationary.” Viewing supply issues as the root cause for inflationary episodes provides an accurate account of when and where inflation occurs. Supply deficiencies typically lead to high, but ultimately moderate, inflation rates. This was a confirmation of my own view.
In the session on the Fed, AI appeared as the future saviour of the fight against inflation, not Fed policy. It was argued that a sharp boost in productivity growth from AI adoption and diffusion of around 0.5% bps a year would help the US public debt to GDP ratio. But that would not be enough. So the other thing that all agreed on was for the need to end ‘fiscal profligacy’. Yellen called for the cutting of Federal entitlement programs, like welfare and medicare (but not defence), and increased taxes to close the budget gap. So yet another round of ‘austerity’.
There is an obsession with public debt levels among the mainstream and little discussion of the impact of high private debt (including corporate debt) that weighs down on the profitability of capital. Nobody points out that the biggest jumps in public debt ratios have been caused by the collapse of the financial sector (2008) leading to the bailouts through public sector borrowing; or in the shutdown of the supply sectors during COVID leading to huge government spending. The public debt crisis is the result of the private sector’s collapses – and it is working people that must ‘normalise’ this debt by accepting cuts in welfare and other services and/or absorb significant tax increases, while the rich and the corporations avoid all.
AI may be the saviour of capitalism’s productivity growth, but China is the opposite, at least according to Robert Gordon, the mainstream expert on US productivity trends. Faster productivity growth in manufacturing than in the private business sector was once a reliable feature of the US economy. During 1972-2009 manufacturing productivity growth was one percent faster than in the private sector (3.1 vs. 2.0 percent). But during 2009-24 its growth was a full 1.5 percent slower than in the private sector (-0.1 vs. 1.4 percent). The slowing of manufacturing innovation, as measured by TFP accounts for 40% of this slowdown but the other 60% is due to a reduction in investment growth. And this slowdown was accelerated by the productivity success in higher value-added sectors in China after it entered the WTO in 2001. The China ‘shock’ broke the back of US manufacturing prowess. Tariff policy is unlikely to reverse the decline in US manufacturing productivity.
Be the first to comment