Interview with Brian Deese (Director of the National Economic Council under President Biden)
Brian Deese was the Director of the White House National Economic Council, where he advised President Biden on domestic and international economic policy. During the Obama Administration, he served as acting director of the Office of Management and Budget and deputy director of the National Economic Council. Previously, he also served as the global head of sustainable investing at BlackRock
Interviews conducted 3rd March 2025 and 4th March 2025
Q: How do you characterise “Bidenomics”? What do you think its objectives were and how do you think it performed?
The term has been thrown around and misappropriated in lots of different ways and I don’t think it’s constructive to try to distil “Bidenomics” into an individual sentence. But if you made me identify the key components, I think there are two things that are notable and new.
One is a commitment and an effort to drive a strong, labour market-centric recovery from a countercyclical perspective.
Two, a more ambitious willingness to experiment with and put in place industrial strategies—meaning using public capital to try to drive private investment in areas of the economy of strategic importance.
If you lift up and ask, “what’s the overall [economic] record?” It’s a recovery that, in terms of how fast it was and how fair it was, was remarkable—miraculous in some respects, unpredicted and unexpected. And then the beginnings of a new architecture of an industrial strategy for which I believe— the initial results are quite promising, but the ultimate impact will depend on implementation over many years, because it was by design a long-term effort.
Q: Was the economic response to the pandemic shaped by the conclusions you’d drawn from the financial crisis?
I think that has been broadly overstated by both proponents and opponents who have sought to suggest the Biden administration’s approach was to take a 180-degree different view.
President Biden himself and his team were acutely aware of both the economic and political lessons learned.
Economically, it was important to recognise the risks associated with emergency fiscal policy and to appreciate the value of insurance. Politically, given our congressional system, there was the risk of not being able to come back with additional legislation. In many ways, the lesson from the Obama administration was less about economics alone and more about political economy—highlighting the political risks and constraints involved in legislating additional policy down the line.
At the core, there was a view that President Biden held strongly, that if he was going to make an error, he wanted to make the error in the direction of driving a stronger, worker-centric, labour market-centric recovery. But similarly, an effort to understand the circumstances, weigh costs and benefits, and get as close to the right answer as possible.
Q: How do you respond to the criticism that the 2024 election outcome was a result of inflation, and that ARPA (the American Rescue Plan Act) contributed to that by increasing the level of demand-driven inflation?
First, that’s a stylised reaction to the headlines. If Kamala Harris had won the election, that wouldn’t have been the diagnosis in the headlines.
There was an ex post political rationalisation that said, “inflation was the principal issue; that created a global anti-incumbency bias; and therefore, to what degree did Biden policies contribute to that inflation?”. Much of the political analysis in the US is that the actual trajectory of inflation in 2023 and 2024, leading into the election, was improbably positive, in the sense that inflation came down faster, with less attendant economic pain, than most forecasters thought was possible.
The political question was, why was Biden, and then ultimately Harris, incapable of benefiting politically from what was an improbably rapid decline in inflation, given the ongoing strength of the labour market? So that’s a political and a public sentiment question.
Obviously, then opponents of Biden and detractors of Biden and people who want to attack Biden say, “Well, that’s a condemnation of the fact that the ARP was too large.” But I think it’s important for your analysis, to disaggregate: there’s a lot of political questions embedded here around the politics of inflation that get conflated with the economics.
On the economics, every serious analysis that has been done has come to the conclusion that the ARP, and in particular, the marginal size of the ARP that people debate, was not the principal or primary driver of the US inflation dynamic. Post analysis suggest that it was marginal. You can have a reasonable debate about whether the ARP was responsible for half a percentage point of marginal inflation, or a percentage point, or maybe even a percentage point and a half.
But on the politics, the idea that having peak inflation at 7.5% or 8% rather than 9% would have made a meaningful difference in public sentiment or the ultimate political disposition is one that those critics have to contend with and don’t usually contend with because they paint with a broad brush.
So on the economics, you can have a reasonable debate about whether the ARP at the margin was responsible for that marginal impact, and whether on the economics of it, that was a cost that outweighed the benefits that it provided. On the politics, it’s just too easy to conflate, because the election went the way that it went. It’s too easy to conflate and say, “Well, that was obviously a bad decision, ex post.” Ex ante, that’s not really the way that you would think about it.
Q: One of the things which is very striking to us in the US recovery is the dynamism in the labour market around small business creation, job creation, and wage rises at the low end of the wage distribution. Was that something you were expecting or planning for?
Our objective in designing the countercyclical fiscal response early on in the administration was to do two things: one, to try to drive a strong, labour market-centric recovery, and two, to take out insurance against a set of known unknowns and risks on the horizon that we knew existed but were difficult to put parameters around, principally associated with the shape of the pandemic and the pandemic response, but also other known unknowns.
On that second piece, we couldn’t have predicted the precise nature of those unknowns, but we did end up living through them in terms of Delta and Omicron on the pandemic side, and then Russia’s invasion of Ukraine and the geopolitical challenges that exacerbated the supply side of the global crisis.
On the labour market side, our hope and objective was to provide a bridge and support at both the household balance sheet level and the small-business balance sheet level that would put people in a position to continue taking risk, continue seeking job opportunities. So, in a sense, it was our hope to drive the kind of dynamism we’ve seen in the labour market.
Were we expecting precisely the outcome in terms of the degree of wage growth and the degree of small business creation? No, although I will say that we did look at how economies have responded from prior massive pandemic shocks—obviously, there are very few modern analogues, so you have to go back to a century ago to find analogues that are most relevant.
One of the markers is bursts of innovation and entrepreneurship on the back end, which can be connected to the nature of the shock. We had that on our radar screen. It would give us too much credit to suggest we were trying to engineer the results on the small business and entrepreneurship side that we’ve seen as an outgrowth. But I do think it is one of the less recognised aspects of the US economic performance over the last 24, 36 months.
Q: Looking back, do you think the nature of the US pandemic response was a better way to [respond] than the more furlough-based response you saw in the UK, France, Germany, and other European countries?
It’s a fascinating question, and one that I’ve thought a lot about. Of course, it’s a hard question because you have to say, “Is it a better response consistent with the underlying system?”
Our labour market— the US labour market— and the US social safety net was such that if we wanted to try to achieve the alternative [European] response, we would have had to have created, in real time, social safety-net structures that we don’t have a lot of muscle memory with.
We saw, for example, the PPP— the Paycheck Protection Program— our efforts to stand up more of a furlough-based response was challenged by the deficiencies in our own system. I do think this ended up being a more constructive response consistent with the system that we had in the US.
Q: Some people would say, “The Administration overdid the fiscal response, and that caused inflation.” Others might argue, “They overdid the fiscal response, but the labour market outcomes were brilliant.” And then there’s a third view: “They might have slightly overdone fiscal policy, and it may have been somewhat inflationary, but without that approach, the strong labour-market recovery would not have occurred—since it depended on a heated labour market—and that was ultimately a price worth paying.” What do you think?
I know very few serious people who would take your first frame as, “They overdid fiscal policy, and therefore caused inflation”—implying fiscal expansion alone was the primary driver of inflation. Even the most vociferous proponents of that first perspective essentially argue, “They overdid fiscal policy, and therefore inflation ended up perhaps 30% higher than it otherwise would have been.”
Those people aren’t really saying the administration fundamentally overdid fiscal policy or that the ARP itself was excessive. They’re saying the ARP was perhaps $500 billion to a trillion larger than it needed to be, meaning you’re not looking at something like 30% more inflation, but rather closer to 10 or 15%.
I’m more in category three. We had a perspective on what would be the ideal amount of fiscal to do in that environment. We then worked through the political process to pass a bill and ended up with an even larger response than where we had started. We thought that on the balance of risk, that was a better risk to take, both because of the insurance against unknowns and because of the potential to drive a strong, labour market-driven recovery.
I don’t think you get the strength and shape of the recovery otherwise. Economically, when balancing those risks, I think that was a reasonable trade-off at the time, and it’s worked out quite well for the U.S. The politics layered on top are complicated—but that’s my take on the economics.
Q: Could you reflect briefly on how the administration’s initial plans translated into actual policies once they encountered the realities of U.S. politics?
The administration came in with a relatively clear view of what it was aspiring to achieve, both from the perspective of near-term countercyclical policy and longer term investment policy and reforms. It laid that strategy out in a straightforward way over the course of the first quarter of 2021.
What the Biden administration did was embark upon a challenging 18-month effort to try to legislate that agenda through the narrowest congressional margins that any modern president has operated with. If you go back and look at what we released in February and March of 2021, it’s a pretty comprehensive and straightforward assessment saying, “Here is our policy vision, here is what we want to get accomplished.”
What happened over 18 months was a political process—with a 50–50 Senate and the narrow margin in the House—of figuring out, realistically, how much of that vision we could get done. That process defined those 18 months.
Regarding the initial vision—where there was a greater focus on the human-capital side of supply-side progressive economics, such as increasing labour supply through investments in care or the Child Tax Credit—did those become less important or less central to the Biden administration’s policies? The reality is, that broader vision was what the administration sought to achieve, and we accomplished what we realistically could, given the political economy at the time.
Both supporters and detractors might argue, “The Biden administration shifted sharply toward industrial policy and away from a human-capital focus.” But whether you’re a supporter or a detractor, that argument gives too much credit to our policy vision and too little attention to the politics of a closely divided Congress, where most legislation needed bipartisan majorities to pass.
Q: Could you lay out your thinking and approach to industrial policy, and how much of it you view as genuinely realising your ideas versus how much was shaped by political compromise?
To answer the last part, this is an area where, despite a political process that was painful, complicated, and slower than we hoped, we largely achieved the vision we set out. I began laying out this vision publicly in speeches as early as June of 2021, and continued doing so over the following years.
At core, there’s nothing earth-shattering about our industrial strategy approach, because it was grounded on the idea that we would identify broad strategic sectors where we identified an economic and national security imperative. We assessed that, left to its own devices, private actors in the market would underinvest compared to what would be economically optimal, and then to the degree we could, use long-term, technology-neutral public incentives to crowd in private capital to build more capacity at scale.
We identified relatively early on three broad sectors that we were motivated by: infrastructure, clean energy, and innovation technology— with microelectronics at the centre of that. To a large extent, we were able to put in place the architecture of a more long-term and more stable public incentives to invest in those categories than we have had as a country in some significant period of time.
Wherever we could, we tried to structure those in a way where they were neutral in applications and didn’t require specific governmental decisions around specific firms, precisely because of the concerns around risks of industrial policy. In some places, there was no alternative than to do so— for example, chips and semiconductors— and in those cases, we tried to build an approach that would allow us to reach our goal while mitigating the risks
The core theory behind the Biden industrial strategy wasn’t unprecedented from an economics point of view. It was just more ambitious and more unapologetic than prior efforts.
We were quite intentional in the design. Wherever possible, we wanted to create stable long-term, technology-neutral incentives that didn’t require a narrow and competitive government-run process of picking individual grantees or individual winners. That was in front of our own minds. That shouldn’t be the dominant policy tool, that should be the exception, where we have to.
Where you see – for example in hydrogen or semiconductors – a strategy that looks much more like a government competitive policy of picking individual grantees and winners, it was based on our assessment that there was no credible alternative, either because of consolidation and concentration in the market—semiconductors—or the lack of a market [for hydrogen] and [the industrial policy] was innovation forcing and market shaping on the front end.
By far the dominant delivery mechanism of dollars—whether that’s for broadband or it’s for battery technology— was largely in tax incentives that are designed to be efficient and also help to reduce those challenges [around picking winners].
With respect to place, we thought about this in two respects. One: how can we build incentives into the architecture of the industrial strategy to encourage investments in areas where there will be a high return to that investment, where there’s lower utilisation, lower labour costs, and therefore higher potential return to those investments. I would say that was implicit in our policy goal and in the policy design and outcomes I would say there are some notable high points and then some things to learn from.
As an example the concept of giving an adder for energy communities in the core of the Inflation Reduction Act tax credits was, in my view, right in design. From a policy perspective, you would want to keep it narrowly targeted and have a high bar in terms of defining energy communities, so that that incremental incentive really had a lot of juice.
That concept, as titrated through the Congressional process, means we ended up with a broader definition that allowed more members of Congress to see into them. In fairness, the Biden administration then interpreted that guidance in a way that was broad, not narrow, and decided to placate the people who just voted for the bill. As a result, you end up with a definition of energy communities that is quite broad.
But I think the energy communities adder is still doing good at the margin. One of the lessons—both in design and execution—is that prioritising specific places is sensible, but the political economy of doing so is challenging.
If you lift up and look at the distribution of private investment in clean energy and semiconductors, it’s striking that 80% of it is going into municipalities with below-median education levels and below-median high school graduation rates. That’s not the typical pattern of traditional private investment. We’ve conducted considerable research on this point, and I think it’s both relevant and important.
It is incumbent on those who want to be more aggressive on industrial strategy to be clear and analytically precise about the goal—what they’re trying to accomplish—and to be prepared to defend the economic and national strategic interest in that goal. There is a risk, when you get into industrial strategy, of simply declaring that “X is strategic, therefore it requires an industrial strategy, and thus the strategy is justified”, without putting enough pressure on whether X truly is strategic.
Now, there is a—what I would call—a kind of comfortable but not entirely intellectually rigorous view that says: “Industrial strategy makes sense for something strategic like semiconductors, but when you get into areas like batteries or solar panels, those seem less strategic.” Let’s start with semiconductors. On what model do you defend the idea that having 30% of leading-edge semiconductor production in the United States meaningfully enhances our national security? I start from this—why is it that we’re comfortable accepting these national security-adjacent strategies?
Some of it is that when national security implications are raised, economists and economic models sometimes say, “that’s not our purview. Someone else defines the national security part of the equation, and we simply insert it into our model.”
We used to do this exercise internally inside the Biden administration. We had a model that said: what does it cost to get what share of leading-edge semiconductor production in the United States? Why is 30% the right amount? Our model equilibrated at X amount of cost for 30%,? But what’s the incremental cost to get to 40% and how much more valuable is that from us from a geostrategic perspective? What if it costs half as much to get to 20%? How valuable is that?
Our tools in economic policy making are not particularly good at that. Actually we should be as exacting on that question, because in a sense, given the way that our national security industrial complex works, I’m more worried about the use of a national security justification to justify an industrial strategy around critical mineral X or an industrial strategy around shipbuilding. That’s what’s happening in the US right now—an industrial strategy because it’s a national security priority. Well, on what model and at what cost and at what relative value [is that justified]?
I’m a big believer in the need to be more analytically robust about this. The framework that treats semiconductors and dual-use technology as the easy cases, while viewing batteries and solar panels as edge cases, strikes me as somewhat lazy. Why, exactly, are batteries—or storage and the upstream components that go into storage—in our national or geostrategic interest? That’s something we need to justify clearly, but I’m prepared to defend it. I’m prepared to argue as equally as strategically important as semiconductors.
Then there’s solar panels. It’s easy to dismiss them as simply green technology, but consider polysilicon: we’re now producing high-grade silicon in the U.S. that also has dual-use applications in chips. So where, and in what context, do we define that as strategic or as delivering value for money? Hydrogen and nuclear technology—these are hard questions. I actually think that this is a place where, in general, to do industrial policy more robustly, we all need a more rigorous framework.
On the “Everything Bagel” critique —the critique is a straightforward one that says if you try to do, in your industrial strategy, more jobs, more powerful labour, more powerful communities, more ancillary services in a community, more national security, more economic security, more place-based vitality, and you prioritise all of those equally, you end up with an inefficient outcome.
I have a reasonably high amount of sympathy for the idea that if you’re going to do industrial policy, you should identify a clear goal. Going back to semiconductors, the North Star is: get to 30% leading-edge semiconductor capacity in the US by 2030. The harder question then is, what are the tools and tactics that are most likely to get you there? If you believe you need a durable industrial commons with suppliers and supply chains, not just a fab in Phoenix, then you might believe you need to subsidise small and medium-sized employers in the US. You can still have a North Star goal and a broader policy that supports the goal.
Q: Another critique would argue there’s been an overemphasis on manufacturing jobs as the primary path to creating good jobs. A better industrial policy aimed at generating good jobs would instead target services—improving productivity in sectors like retail or healthcare. How would you respond to that critique?
I think this is a place where I’m probably aligned with those who say the principal goal of industrial policy shouldn’t be job creation of a particular type. For example, if the North Star is 30% leading-edge fab production in the United States, then it’s reasonable to expect we can do that in a way that improves job quality in some categories and adds to productivity, and therefore has a broader positive macro impact. But it’s not necessarily a massive job creator on its own. If you look at these initiatives purely from a job-creation perspective, they’re not going to be the most efficient way to make jobs.
So a critique of critique you put forward might be: if you have industrial policy goals, you need to ask what capabilities are necessary. There’s an under-recognition about the value of muscle memory in manufacturing processes and supply chain capabilities to build industrial capability —what Dan Wang and others talk about regarding China.
If you look at why we’re terrible at building nuclear power plants in the US – it costs twice as much, takes twice as long as in South Korea or the UAE – partly that’s our regulatory system and structure that discourages risk taking. But partly it’s because we’ve lost the muscle memory from building only one every 15 or 20 years. You need to think downstream about the industrial commons necessary to get these things done – but the wrong way to think about that would be to think about the job outputs you’re going to get.
The flip side of that is that I’ve never seen a politically successful initiative where politicians don’t claim that the thing they are going to do is going to create better jobs or more jobs for people. The criticism that people who are promoting industrial policy should never talk about job creation doesn’t align with the reality of how politics work. It doesn’t keep me up at night that politicians talk a lot about creating jobs If you’re a senator from Arizona or Ohio, 80% of what you’re saying is: “This is exciting because we’re creating more good jobs.”
On the one hand from a policy perspective, you shouldn’t design semiconductor industrial policy purely by asking, “How many jobs does this create?”. On the other hand, in the production and distribution of semiconductor industrial policy, it’s basically fine— even good—that people identify that one of the results is more good jobs.
Not to belabour the point, but part of what the United States can achieve by successfully executing industrial strategies is proving out a model that can help make similar efforts in the UK and elsewhere more successful.
It drives me crazy to talk about nuclear power in the US because there’s this collective assumption—almost treated as a law of physics—that it takes 10 to 15 years to build a plant here. Unlike the laws of physics that are actually in play in nuclear, it’s not a law. In fact, if we were to set a goal to build 10 gigawatts of nuclear by 2030 and if we walked back to determine exactly what we’d need to do to make that happen, and then actually did it, the spillover benefits to the world would be quite significant.
One, we would have driven down the cost of deployable technology elsewhere – some of those 10 gigawatts would come from AP1000s which are a proven technology, but we would prove out innovation in process that would have spillover benefits. There would be a number of EPCs that have done it in the US that could do it in the UK. Two, we would drive innovation in small modular nuclear and other technologies. Three, you would show a roadmap of what would actually take to up-end a regulatory architecture, have the model between EPCs and government milestone contracting payments. There are these collective problems in post-industrial OECD economies that are correlated. I believe that when some of us break these down, it’s obviously good for our national economies but there will be spillover benefits too.
Q: What are the lessons from the Bidenomics experience for countries in Western Europe, especially the UK?
One is—even before Trump—we’ve shifted into a new geopolitical paradigm. Rather than fighting that reality or trying to recreate some yesteryear, the right approach isn’t about whether we should pursue a more active and energetic industrial strategy, but how to do it effectively.
Second, clarity of the goal and simplicity and efficiency of the delivery mechanism matter greatly. The clearer the goal—and the clearer the public debate around the national and economic security priorities—the better. It’s neither advisable nor feasible to build UK or European industrial capabilities simultaneously in wind, hydrogen, nuclear, shipbuilding, and semiconductors. There should be a national debate to prioritise clearly. The more efficiently you can use subsidies to encourage scale at the margin, while avoiding the known risks of industrial strategy like capture, the better.
I’ve heard repeatedly that our UK or European friends found the IRA somewhere between annoying and distracting—largely because it was so effective. Companies found it easy to interface with, and it provided certainty for investment. That’s a positive lesson we should learn from, rather than fight against.
Obviously, harmonisation across borders is important. Saying this today—on March 4th, after the tariff announcement—I recognise that cross-border harmonisation will only get more complicated, but it should clearly remain the objective.
Q: Do you think there are also lessons on the macro-side, on policymaking and politics, that might translate outside the US?
The shape and strength of the US recovery reflect the broader fiscal policy response—not specifically Biden or “Bidenomics,” but the collective response of the U.S. government. Both the design and magnitude of that fiscal policy created a stronger and fairer recovery. Not the strongest or fairest possible, but clearly stronger and fairer than it otherwise would have been.
In hindsight, there will be important lessons from this. In fact, if we spend too much time debating the incremental margin—like the last $500 billion of the ARP—we risk losing sight of the broader question: what can we learn from the overall U.S. response?
I think we’ll ultimately look back at this as an example of a robust response, emphasising household balance sheets and labour-market strength, and draw many positive lessons from it. Not uniformly positive—there will certainly be areas where we can learn and improve—but overall, that’s what I hope and expect we’ll take away.
ENDS