The Marginal Revolution Podcast

Episode: Favorite Models: Spence on Monopolies, Harberger on Incidence, Solow on Growth

Hosts: Alex Tabarrok [A], Tyler Cowen [B]
Date: October 7, 2025

Overview

In this episode, Tyler Cowen and Alex Tabarrok delve into their favorite foundational economic models: Michael Spence's analysis of monopoly and quality choice, Arnold Harberger's two-sector tax incidence model, and Robert Solow's model of economic growth. Throughout, they discuss what these models explain, their real-world applications, their limitations, and how they continue to shape economic thinking.

1. Spence Model: Monopoly and Quality Choice

(Main Discussion: 00:10–20:42)

Model Overview

• The Spence (1975) model considers how monopolists decide not just how much to produce, but also the quality of that output.
• Key Question: Do monopolists always lower product quality, or can they choose higher quality too?

Key Insights

• A monopolist is focused on maximizing revenue, principally by targeting the marginal consumer—those just on the edge of purchasing.
• Changes in quality can affect all consumers but are chosen to sway the marginal buyer.
• This decision might raise or lower quality for everyone, sometimes decreasing total social value versus a planner's choice.

Illustrative Example (02:17–06:08)

• Alex: Tom values a good at 9, Dick at 5, Harry at 3.
  • Monopolist’s options: sell 1, 2, or 3 units at descending prices.
  • If improving quality bumps Harry’s willingness to 4, selling 3 units at 4 becomes optimal—even if Tom’s value drops from 9 to 6.
  • Quote: “Even though...the social value of the good has actually gone down...the monopolist only cares about the effect on Harry.” (03:13)

Applications & Intuition

• Restaurant smoking policy: Allowing smoking might appeal to a marginal smoker, even if it hurts existing non-smoking customers.
• Monopolists aren't planners—they don't consider aggregate welfare, only revenue maximization at the margin.

Cowen’s Critique (06:08–15:29)

• The simple assumption that monopolies create low-quality goods is challenged: quality moves both ways depending on numbers and margins.
  • Quote: “The Spence model shows you that a monopolist could choose higher quality or lower quality, and neither follows a priori.” (06:08)
• Applications: News media shift—ex: The New York Times moving from moderation (for ad revenue/eyeballs) to catering to "infra-marginal" loyal subscribers (with the rise of subscription models).
  • Analogy to HBO and pay TV: serving “hardcore” fans versus “marginals.”
• Tyler argues Spence underweights the importance of infra-marginal, enthusiastic consumers as viral marketers and core revenue drivers in the digital era.
  • Quote: “It’s the person who barely is willing to buy your thing who's not going to talk about it. And you need people to talk you up.” (11:39)

Relating to Educational Publishing (16:09–20:42)

• Tabarrok: When writing their economics textbook, their aim was to delight infra-marginals, not maximum popularity.
• Cowen: Publishers often pressure for content that appeases the average marginal adopter ("can this get through committee?"), but products with passionate infra-marginal fans endure and thrive.
  • Quote: “They don’t care that people love our book. They only want it to be popular.” (17:08)
• Demonstrates the tension between optimizing for passionate loyalty versus broad, marginal adoption.

2. Harberger Model: Incidence of the Corporate Income Tax

(Main Discussion: 20:42–33:54)

Model Summary

• Arnold Harberger’s 1962 paper was pioneering in showing how taxes on one sector (corporate) affect all sectors (including non-corporate).
• Moves analysis from partial equilibrium (single sector) to general equilibrium (multiple sectors, interdependencies).

Key Insights

• Taxing corporations lowers returns on capital in all sectors—not just in the taxed sector.
  • Quote: “If you tax the corporate sector...capital can suffer in both sectors. Again, a revelation...but it shocked people.” (22:30)
• Capital flees taxed sectors for untaxed ones (or abroad), leading to broad-based impacts.
• Labor also bears some burden—estimates suggest labor may bear up to half the corporate tax burden (directly and via pension funds).
  • Quote: “I think it’s plausible to think labor might bear half the burden of a corporate tax...” (24:10)
• If land is included, inelastic supply means it can end up absorbing much of the tax.

Applications

• Real-world international tax policy: small, open economies can’t tax capital heavily (it flees abroad), so even social democracies like the Nordics often treat capital income gently.
• For city congestion taxes, much of the burden may end up on landlords rather than commuters, due to relative elasticities.

Model Evolution and Critique

• The baseline model assumes fixed levels of savings/capital—subsequent contributors like Larry Summers expanded to consider more elastic responses.
• New empirical work suggests more of the tax might be passed on to consumers via price increases than the model predicts.
  • Quote: “Some of the core results...can be like up to 50% of the total, gets translated into higher prices for consumers... Not resolved in my opinion, but a puzzle.” (33:15)
• Business intuition (“We’ll just pass the tax on!”) has more truth than previously credited.

Model Impact

• The Harberger model is foundational for thinking about tax incidence in economics, trade theory (connected to Heckscher-Ohlin), and the evolution of more complex modeling (e.g. Krugman, Dixit–Norman).

3. Solow Model: Economic Growth

(Main Discussion: 33:54–54:54)

Model Summary

• The Solow (1956) model is a “workhorse” macro framework analyzing long-run growth through capital accumulation, diminishing returns, and technological progress.
• Fundamental insight: Separates sources of growth—capital deepening, labor increases, and "ideas" (total factor productivity, TFP).

Key Insights

• Poor countries can grow rapidly at first (“catch-up growth”): marginal returns to capital are high, depreciation low.
• As capital accumulates, more investment is needed to offset depreciation; growth slows unless TFP improves.
  • Quote: “China...grew at 10%...even though its institutions are much worse than the institutions of the United States...Solow model explains this.” (38:05)

Model Implications

• The distinction between catch-up and cutting-edge growth is critical.
• The model’s parameters are easily estimated; one can predict convergence rates—but real-world deviations are substantial.
• Growth accounting: attributes output growth to changes in capital, labor, TFP.

Critiques & Limitations

• Cowen: Separating ideas from capital may be unrealistic—especially in a technology-driven economy where “ideas are embodied in investment.”
  • Quote: “I'm not sure we should separate capital and ideas so cleanly.” (43:16)
• At today’s tech frontier, the blurring of ideas and capital undermines classic Solow accounting.
• The model doesn’t explain persistent divergence—not all countries catch up, even with plausible institutions (e.g., Uruguay, Argentina).
  • Quote: “My inclination is...maybe they're not going to [catch up].” (46:55)
• Scale effects (important in a tech world) and culture/ambition are not captured by the model.

Applications & Modern Twist

• By introducing noise (shocks to capital, labor, TFP), the Solow model becomes the foundation for real business cycle theory.
• Tyler: The distinction between “once-and-for-all” versus ongoing growth is challenged—perhaps innovation and luck, not smooth progress, drive history.
  • Quote: “So much of economic growth is just that every period you get a new once and for all change.” (53:16)

Concluding Thoughts on Modeling

• Simple models are valuable even when wrong—they highlight where to look for new insights and force clarification of assumptions.
  • Alex: “Models matter not just when they're right, but also when they're wrong, because they point you in a direction you might not have previously thought about.” (54:16)

Memorable Quotes & Moments

• On Monopolists and Quality:

  • Alex: “What this tells you is that the monopolist isn't thinking the same way as the social planner.” (04:51)
  • Tyler: “The Spence model shows you that, that a monopolist could choose higher quality or lower quality, and neither follows a priori.” (06:08)

• On Publishers and Textbooks:

  • Alex: “They don't care that people love our book. They only want it to be popular.” (17:08)

• On Harberger Model:

  • Tyler: “If you tax the corporate sector under a lot of reasonably general assumptions, the rate of return on capital goes down equally in both sectors...capital can suffer in both sectors.” (22:30)
  • Tyler: “Labor bears a chunk...a typical estimate might be a third.” (24:10)

• On Solow Model and Growth:

  • Alex: "China had very little capital. So the capital that it first applied was very productive..." (38:05)
  • Tyler: "My inclination is just to think the countries that haven't developed by now, for whatever reasons, maybe they're not going to..." (46:55)
  • Alex: “Models matter not just when they're right, but also when they're wrong, because they point you in a direction you might not have previously thought about.” (54:16)

Important Timestamps

• Spence Monopoly Quality Model: 00:10–20:42
• Harberger Tax Incidence Model: 20:42–33:54
• Solow Growth Model: 33:54–54:54

Summary Tone

Energetic, nerdily enthusiastic, debate-driven. Both hosts embrace deep dives, friendly disagreement, and marry real-world stories to economic abstract reasoning, lending a relatable and lively atmosphere throughout.

Recommended Further Learning

• Modern Principles of Economics by Cowen & Tabarrok (textbook with these models).
• Alex’s YouTube video on the Solow Model (as praised by Tyler).

This episode is a masterclass in why core economic models are enduringly useful, what we should demand of our models, and how nuanced their real-world application must be. Both hosts take listeners inside the economist’s mindset—skeptical, flexible, empirical, and always seeking relevance for today’s world.