Interview with Larry Summers (Harvard University, Secretary of the Treasury under President Clinton and Director of the National Economic Council under President Obama)
Lawrence H Summers, Charles W. Eliot University Professor at Harvard University. He previously served as the Secretary of the Treasury for President Clinton, Director of the National Economic Council for President Obama and the Chief Economist of the World Bank.
Interview conducted 29th January 2025
—
Q: How would you characterise Bidenomics? What do you think its objectives were, and how do you think it performed?
I think the Bidenomic strategy was: much greater faith in strong aggregate demand to produce the benefits of a high-pressure economy; a belief in fairly activist industrial policy to resurrect American manufacturing and to address key challenges around resilience and security; and a substantial effort to signal a greater commitment to policies directed at ‘predistribution’, as distinct from redistribution, by supporting – in various ways – labour in the economy, including through aggressive antitrust policies. Net, it was a move to a substantially more involved government in economic policy – with the adoption of almost all advanced justifications for greater government involvement.
Q: How would you describe the reality of what the Administration actually achieved compared to what was originally intended?
I think they did a fair amount of what they planned to do. They had a set of ideas around the care economy, which they were less able to implement than they had hoped. They were able to implement their infrastructure and green economy initiatives, but on a smaller scale than they had hoped. But I would say they were able to get a fair amount of the program adopted.
Q: Do the various elements you describe add up to a coherent set of ideas and a new way of policymaking, as advocates claim?
There was some scope to pick and choose among the ideas. It’s not coherent in the sense that a sensible person would be for all of it or for none of it, but it has a set of common themes that run through it. I think it has some important and valid aspects, and some others that I think are less compelling.
Ex ante, I thought the macro stimulus was being substantially overdone – at the time. And I think that was because the program was assembled from the bottom up, by bringing together what was seen as politically compelling rather than from any macroeconomic analysis. No macroeconomic analysis of the gap in the economy versus the magnitude of the fiscal stimulus would have supported a program of the size that was adopted. All one has to do is see that nominal GDP – essentially aggregate demand, or in monetarist language ‘money times velocity’ – increased by 35% during the Biden term. The GDP gap might have been 6–8% when Biden came in. If you set policy in a way that generates 35% nominal GDP growth, it is hard to believe you won’t have inflation substantially above 2% unless you’re starting with an immense real GDP gap. That was not hard to see when the program was launched.
I think that was a substantial error—economically and politically. This is the third time in 58 years that progressives have come a cropper around inflation. If inflation had appeared more under control, Richard Nixon would not have been elected in 1968, because that was a very close election. If inflation had not symbolised a country out of control, Ronald Reagan would not have been elected in 1980. And if inflation had not been the premier economic issue, Donald Trump would not have been elected in 2024. That was an enormous mistake.
There was a political error of judgment, which I wouldn’t necessarily have known was an error when it was made. There was a deep conviction that it was important to provide child tax credits because if you put child tax credits in, there would then be a clamour for them to continue, and that would represent a significant step forward for children in America. In the event, there was no such demand. Republicans did not allow them to continue, and no one in the congressional election mentioned the failure of Republicans to allow them to continue because they were unmourned. While I personally supported the child tax credits, fiscal irresponsibility just to get the camel’s nose under the tent was not a plausible political strategy.
I’m not sure how history will judge the infrastructure and green programs. The recognition of the need to jump-start renewable energy technologies and resolve chicken-and-egg problems was an important and valuable initiative. We’ve seen a profound change in progressive thinking about climate change, away from pricing carbon toward finding technologies that are renewable and have energy costs no higher than fossil fuel costs. I think this program was catalytic in that regard and that was an important step.
I think it was important to address infrastructure. Unfortunately, the Achilles’ heel of progressives is a focus on conceptualisation rather than implementation. If one looks at the path of real infrastructure spending, as opposed to nominal spending, real spending increased much less than expected due to various bottlenecks. The stories of billions of dollars being allocated with very few charging stations built are troubling. I don’t think we really quite know what the incremental efficacy of the green investment programme has been. For example, I don’t fully understand the administration’s fixation with producing solar units in the United States. It seems to me that they’re not a security-sensitive technology, and there are a lot more jobs involved in installing them than in producing them. So I didn’t see the rationale for the degree of protectionism that was applied.
Semiconductors were handled with more objectives trying to be pursued, such as unionised labour and childcare, than I would have recommended, but those initiatives were valuable and contributed towards security, given the extent of concentration in Taiwan and China (a very dangerous part of the world) for semiconductor production.
I think the general attitude of the administration of hostility towards business was politically costly and probably carried with it a set of economic costs as well. The Federal Trade Commission was extremely active, and by regularly losing court cases with a somewhat incoherent populist doctrine, it set back the cause of active competition policy while, at the same time, complicating American leadership by attacking potential national champions.
Q: Were you surprised by the dynamism of the labour market response to the ARPA stimulus?
I don’t know for sure. I remarked early on about what I called the Truro, Massachusetts, model. Every winter on Cape Cod, GDP in Truro goes down by 60–70%, and every May, when the tourists return, the restaurants and shops re-open, and GDP effectively doubles or triples within a month. I thought there were important respects in which the broader economy was in a similar state of suspended animation during the pandemic as Truro during the winter; and that was a more self-correcting dynamic, less in need of dramatic action by government, than was sometimes suggested.
I haven’t studied the European experiences closely enough. Europe faced a very different type of energy price spike than the U.S., for instance. But I remain sceptical that the level of inflation we experienced was necessary for a reasonable recovery. And I think that inflation was also associated with significant reductions in real wages.
Q: In 2021, many were worried about the opposite: a weak demand recovery leading to an even worse economic downturn. Is it right to say that policymakers should have foreseen inflation?
I thought those analyses were missing the point, in the same way that the analyses done in 1943 forecasting a depression after World War II were missing the point. They were failing to understand that when people are unable to spend, they hoard savings, and then the spending took place later. They were failing to appreciate how much was being done to support disposable income. I thought there was a need to maintain a steady path of household disposable income. There was not a need to achieve record levels of growth in disposable income, which is what was done as a result of the stimulus program.
Q: Was the “hot economy” necessary for the strong labour market recovery the US saw from 2021?
Another analytic error that people made was this: suppose eight Indian restaurants existed in Brookline, Massachusetts, and three closed because of the strain of COVID. When COVID ended, there was still probably the same number of people who wanted to eat Indian food. They went to the five restaurants rather than the eight restaurants, and the five each had to hire more waiters and were maybe more efficient (in the sense that there were more economies of scale), albeit with less variety.
I think the fact that America permits more creative destruction, and as a consequence enjoys more creation – and that it benefits from the flexibility of American institutions – is probably more “the lesson” from the recovery. I think there is evidence that those parts of America which moved beyond excessive unemployment insurance faster enjoyed more rapid recoveries. I think it’s more a tribute to general differences in resilience and flexibility in America’s institutions, than a commentary on the specific policies the Biden administration adopted. I wouldn’t say that the Biden administration were pushing toward less flexibility rather than more; I just don’t think they were focused on flexibility as much as I would have liked them to be.
Q: Against those pre-existing strengths (labour market flexibility, avoiding an energy shock, the strengths in the tech industry) how much did Biden micro-economic policies matter for US growth?
The fact that the world’s great tech companies are disproportionately American has to be seen as having a great deal to do with the strength of the American economy in this period. We could have failed to preserve financial stability or could have allowed a cascading decrease in aggregate demand. But I think it’s the system, more than specific policies pursued in those years, that deserves dominant credit.
Q: Earlier you differentiated between solar panels and semiconductors. Are you saying that national security concerns justified a more protectionist approach for chips but not for solar panels?
In general, if you argue for industrial policy and you mention jobs, you’re probably making a mistake. There are rationales for industrial policies, but they don’t include the creation of jobs. That suggests an excess in industrial policies.
Q: Some argue that the location of jobs matters as much as the industry itself.
There is a case, perhaps even a reasonably strong case, for a range of place-based policies. I’m not convinced that is well enough understood, or well enough done. The tendency – from the analysis I’ve seen – is that the tendency of the Biden administration to target programmes on high-unemployment areas was not particularly pronounced.
Q: If you were advising governments in the UK and Europe, what lessons—positive or negative—would you highlight from the Bidenomics experience?
Be very careful of inflation. You have to be mindful of some degree of fiscal sustainability.
You have to be aware that people don’t check into hotels that they won’t be allowed out of. Employers who can’t fire people won’t hire them. Labour market flexibility is at a premium.
You have to be mindful of the difficult of implementation and executive in the public sector, and projects need to not have a hundred different objectives to satisfy a hundred different constituencies. And veto power should not be distributed promiscuously..
And you need to be supportive of entrepreneurship, because ultimately that’s where lots of energy comes from.
Those are the lessons I would emphasise.
Q: You mentioned the intellectual shift in climate policy from pricing to technology – what is the translatable lessons there for Europe?
I’d be looking to get the best technologies from anywhere, implementing them, and supporting entrepreneurship and innovation around key technologies.
If I were Britain, I’d recognise the immense asset I have in being home to several of the world’s great universities. I would keep them great. I’d use them as magnets for talent, some of which would settle in Britain. I would try to build as much activity out of them as possible, in the way Route 128 was built out of MIT or Silicon Valley out of Stanford. And I would think of that as a major source of comparative advantage.
Also, I wouldn’t mind buying solar panels from China.
Q: If you had to summarise Bidenomics in one surprising positive and one disappointing negative, what would they be?
The biggest negative was the miscalculation on inflation. The thing that was positive was the CHIPS Act – showing you can meaningfully get stuff going with strategy.
ENDS