Jason Furman on Productivity, Competition, and Growth

Jason Furman — professor of economic policy at Harvard and former chair of the Council of Economic Advisers under Obama — joins Tyler Cowen for Ep. 103 to diagnose America’s productivity slowdown, argue that rising market concentration depresses investment, and set out a policy agenda built around immigration, R&D, and a more aggressive approach to tech platforms.

Key ideas

1. Rising market concentration is suppressing business investment. Furman argues that monopoly power in concentrated industries reduces both the pressure and the incentive to invest. Research by Thomas Philippon and Jan Eberly shows a correlation between concentration increases and investment shortfalls across sectors — a relationship that, in Furman’s view, policy can and should address.
2. Immigration is the single highest-leverage lever for productivity growth. Asked to name two or three priority changes as a hypothetical ‘productivity czar’, Furman answers ‘more immigration’ three times before listing anything else. Micro evidence — patents filed, and spillovers to neighbouring American researchers — supports the claim in a way aggregate historical comparisons cannot.
3. The China shock literature teaches us about labour-market rigidity, not trade. The finding that displaced manufacturing workers suffered larger and more persistent losses than expected reveals weaknesses in how the American labour market connects people to jobs, enables mobility, and cushions wage drops — not a reason to rethink trade openness per se.
4. ‘Merger to monopoly’ deserves more scrutiny than organic growth. Furman distinguishes Walmart (organic expansion, each store out-competing its local rivals) from Facebook and Google, which assembled their positions through acquisition. Where growth was purchased rather than earned, the presumption of efficiency weakens — and the case for intervention strengthens.
5. The US fiscal position is manageable but not a free lunch. Low real interest rates below the growth rate allow a sustained primary deficit, but the current trajectory implies a steady-state debt approaching 400 per cent of GDP. Furman counsels exploring how far that can stretch gradually, not committing to it outright — and notes that borrowing to finance consumption, not investment, still has real future costs.

Content

The investment shortfall and market concentration

Furman opens by distinguishing two readings of sluggish US investment. Some of the shortfall is benign: an intangible economy — one built on software, brands, and data rather than factories — simply requires less physical capital. But a portion reflects genuine policy failure. Concentrated industries generate monopoly rents (the excess profit a firm earns by facing limited competition, much as a dominant supplier can charge above cost) that reduce both the competitive pressure to invest and the need to defend market share. Furman draws on Thomas Philippon’s work and Jan Eberly’s sector-level research to show that the geographic and sectoral pattern of investment shortfalls tracks the pattern of rising concentration.

Cowen pushes back with the hospital sector — concentrated, yet heavily invested — and with tech, where high market dominance has not obviously produced high prices. Furman concedes the cases but holds the general point: in healthcare, market power demonstrably raises prices, and the hospital mergers that followed the 2000s consolidation wave produced few efficiency gains other than, perhaps, lower wages for nurses. The aggregate concentration indices are crude, he acknowledges — retail concentration reflects local-level competition and efficiency gains, not monopoly — but on balance they have predictive power for profitability, markups, and the divergence between returns to capital and safe Treasury yields.

Productivity: diagnosis and remedies

Asked the cause of productivity slowdowns after 1973 and again after 2005, Furman offers a partial verdict. Reduced public infrastructure investment and falling federal R&D spending (down as a share of GDP since the 1960s) account for only a few tenths of a percentage point each. Concentration may add another tenth or two to the recent slowdown. But part of the question is mis-specified: the growth of the 1950s and 1960s was unusually fast, shaped by post-war catch-up and one-off projects (the interstate highway system, the lunar programme). The internet’s productivity gains were largely captured in the level of output a decade or more ago and do not add to growth in any given recent year.

Furman’s forward agenda as a hypothetical ‘productivity czar’ consists of: more immigration of talented workers who patent and generate spillovers for domestic researchers; sharply increased federal R&D spending; business tax reform that allows immediate expensing of investment and extends R&D credits to internalise positive externalities (the benefit an innovating firm confers on others, for which it currently captures too little); YIMBY housing reform to allow workers to move to high-productivity cities (San Francisco, Boston, Washington) that artificial planning constraints have priced them out of; and investment in higher education.

On place-based policies for lagging regions, Furman is sceptical. Algorithmic formula-based transfers (more Medicaid matching when state unemployment rises, longer unemployment insurance in downturns) make sense. Discretionary regional development — picking a cluster to build — has a poor track record, and Congress has no appetite for expert guidance on which places deserve resources.

Labour-market fluidity and the China shock

Furman reframes the China shock literature as evidence about labour markets rather than trade. The finding — that workers displaced by Chinese import competition fared worse and for longer than earlier models predicted — reveals three pathologies: an unusual incidence of long-term unemployment, involuntary part-time work, and labour-force exit. These pathologies, Furman argues, track a broader secular decline in labour-market fluidity — the rate at which workers move between jobs and firms create and destroy positions — that predates the China shock and shapes how any large economic disruption propagates.

The drivers of lower fluidity include: land-use constraints that make it expensive to relocate to better labour markets; occupational licensing requirements that impede switching sectors; and healthcare tied to employers, which locks workers into jobs they would otherwise leave. Whether the Affordable Care Act improved the last of these is an open empirical question Furman identifies as worth studying.

On the pace of recovery from the 2008 recession, Furman cites Alan Krueger’s finding that no OECD country, over the previous fifty years, had reduced an elevated unemployment rate faster than 0.7 percentage points per year. The US recovery was, by that metric, roughly on pace — or slightly faster than the 1980s recovery. GDP growth underperformed forecasts; unemployment recovery did not.

Big tech and the limits of ‘price equals zero’

Furman draws a taxonomy of tech incumbents by how they grew. Amazon he treats as largely organic — each expansion into a new product category competed on its merits — and accordingly places lowest on his concern list, though not off it. Facebook and Google he classifies as ‘merger to monopoly’: Facebook owns three of the main social networking platforms because it purchased two of them; Google assembled most of its product suite through acquisition. Where that is the route to dominance, the presumption of efficiency weakens and the case for scrutiny strengthens.

Cowen challenges the consumer-harm framing — the user-facing price is zero, there are many ways to connect socially, and online advertising is cheaper than the media it displaced. Furman responds on two levels. First, higher advertising prices caused by platform concentration are embedded in the cost of every product sold through those channels, so the consumer pays indirectly. Second, the correct counterfactual is not the world before Facebook and Google but the world with five competing platforms rather than two — lower ad prices, more innovation, and less exposure to privacy costs that Furman treats as a distinct and non-zero harm.

On privacy regulation, Furman has mixed views on GDPR (the EU’s General Data Protection Regulation): privacy matters, but cumbersome compliance requirements advantage large platforms over smaller rivals. He favours asymmetric regulation — a code of conduct for dominant players that does not apply to small entrants — as a way of combining meaningful constraint with preserved entry incentives. His UK report recommended exactly this for what became the Digital Markets Unit.

Fiscal space, interest rates, and the debt trajectory

Furman applies a basic fiscal arithmetic (the relationship between borrowing costs and economic growth that determines whether a country’s debt grows or shrinks over time) to the US position: when the real interest rate on government debt is below the economy’s real growth rate, a government can run a primary deficit (spending more than it collects in taxes, before interest payments) indefinitely without the debt ratio rising. That condition currently holds.

But the current primary deficit implies a steady-state debt of around 400 per cent of GDP — a level Furman regards as feasible but imprudent to commit to deliberately. His preferred approach: explore how far the envelope can stretch slowly, not lock in a trajectory; optionally lock in low rates through long-term borrowing; and be alert to the asymmetric tail risk that rates can rise by far more than they can fall from present levels. On Brexit, he reads the cost as a medium-term microeconomic one — modestly worse supply chains, lower import-export efficiency — in the range of a couple of percentage points of GDP, rather than a large upfront shock.

Related

• What Makes Economies Grow — theme; Furman supplies the frontier-economy variant of structural transformation — keeping the growth engine dynamic through competition, immigration, and labour-market fluidity
• Jason Furman — speaker; Harvard economist and former chair of the Council of Economic Advisers
• Tyler Cowen — host
• Joe Studwell on Africa, Asia, and What Development Actually Requires — adjacent episode on the same show; Studwell’s defence of industrial policy and Furman’s scepticism of place-based regional policy make an instructive contrast
• What Makes a Great Investor — theme; Furman’s fiscal arithmetic (r vs g, primary deficits, debt trajectories) informs how to think about sovereign borrowing costs and capital allocation