Monopsony

Overview

Monopsony is market power held by a buyer rather than a seller — in labor economics, an employer's ability to pay workers less than their competitive marginal product because workers face frictions (search costs, relocation costs, few local alternatives) that make it costly to leave for a better-paying job.[1] Alan Manning's Monopsony in Motion: Imperfect Competition in Labor Markets (Princeton University Press, 2003) reframed this from the textbook single-buyer-town case to a general description of most real labor markets: because switching jobs is costly, the labor supply curve facing any individual employer slopes upward rather than being flat, giving even ordinary employers some wage-setting power.[1] The gap between a worker's marginal product and their wage under monopsony is a rent captured by the employer — the labor-market analogue of the rent a landowner captures from a fixed, undersupplied asset, though the underlying mechanism (search frictions, not fixed physical supply) is different from land.

Empirical work has since tried to measure this power directly. José Azar, Ioana Marinescu, and Marshall Steinbaum, "Labor Market Concentration" (NBER Working Paper 24147, 2017; published in the Journal of Human Resources 57(S), 2022), used job-posting data across thousands of local occupation-by-region labor markets in the US and found that a move from the 25th to the 75th percentile of employer concentration is associated with a wage decline of roughly 17%, consistent with concentrated employers exercising monopsony power.[2] This literature — along with related work on non-compete clauses and no-poach agreements — has become a central input into US antitrust discussions of labor markets since the mid-2010s.

The State of the Measurement Debate (2026)

Sydnee Caldwell, Arindrajit Dube, and Suresh Naidu — three of the field's leading empirical labor economists — published a comprehensive survey, "Monopsony Makes it Big," as NBER Working Paper 35608 in August 2026. The survey organizes monopsony power around a single central measure — the firm-specific labor supply elasticity — and identifies three distinct sources that can generate it: search frictions (Manning's original mechanism, above), preference heterogeneity (workers value non-wage job attributes differently, so firms don't compete on wage alone), and employer concentration (Azar, Marinescu & Steinbaum's mechanism, above). Distinguishing these sources matters because they imply different policy responses: a search-friction story points toward reducing job-search costs, while a concentration story points toward antitrust enforcement.

The survey's most consequential methodological contribution is a caution about relating estimated elasticities to actual wage markdowns: because firms differ in how they set wages in practice (posted wages, individual bargaining, or something in between), the same underlying elasticity estimate can imply very different amounts of rent actually captured by the employer — a warning against reading any single elasticity number as a ready-made "monopsony rent" figure. The paper also examines how monopsony interacts with efficiency wages (employers paying above the market-clearing wage to elicit effort or reduce turnover, a competing explanation for wage patterns sometimes mistaken for monopsony's opposite), and surveys the resulting policy implications across antitrust, labor regulation, immigration policy, and macroeconomics — closing with the authors' own list of open research priorities, i.e. this remains an active, unsettled measurement frontier rather than a solved one.

Relevance to the Rent Debate

Monopsony matters to this wiki because it is a buyer-side rent channel, distinct from the seller-side monopoly rents the wiki already covers under superstar firms and rising markups (De Loecker, Eeckhout & Unger). Where those pages ask whether firms extract rent by charging consumers more than a competitive price would allow, monopsony asks whether firms extract rent by paying workers less than a competitive wage would require — a mechanism that can coexist with, or substitute for, seller-side market power in explaining the falling labor share. Per the wiki's rent-gradient rule, monopsony sits well out on the contested frontier: unlike land, which is fixed in supply, labor markets are not literally scarce in the same sense, and the size of monopsony's aggregate effect on wages remains an active empirical dispute rather than a settled fact — a judgment the Caldwell-Dube-Naidu survey above reinforces directly: even the field's own leading empiricists treat the elasticity-to-markdown mapping as unresolved, not merely under-measured.

Radical Markets (Posner & Weyl, 2018) applies the concept in two chapters: Ch. 3 proposes a Visas Between Individuals Program partly to break employers' monopsony power over migrant workers, and Ch. 5 argues that dominant data platforms hold monopsony power over the data individual users generate, motivating the data as labor proposal to pay users for that input rather than let platforms capture it for free.

A Political Extension: Electoral Monopsony (2026)

An August 2026 NBER working paper by Carlos Fernando Avenancio-León, Adelina Barbalau, Cyndi Hou and Alessio Piccolo proposes that labour-market power converts into political power, coining the term electoral monopsonies: dominant local employers "using local labor market power to shape political preferences and electoral outcomes."[5] The paper opens with original survey evidence that workers at major local employers are more likely to experience employer political communication and to report that their employers influence their voting behaviour and career expectations, then models the mechanism as operating through workers' expectations about how wages and employment depend on who wins.[5]

The modelling distinction is the analytically useful part. A passive channel has workers simply internalising their employer's economic interests — a rational response to genuine exposure. An active channel has the dominant employer strategically shaping those expectations. Under the active channel, the authors argue, labour-market power "can generate political failures by inducing voters to oppose policies they would otherwise support," and can also contribute to polarisation, constrain the platforms parties are able to offer, and substitute for campaign spending.[5] Empirically, using US individual-level voting data and a shift-share design built on national industry-concentration shocks interacted with predetermined local employment shares, they find greater electoral monopsony power increases Republican voter turnout, with counterfactual estimates implying that reducing it "could have narrowly changed the outcomes of the 2016 and 2024 presidential elections."[5]

Two reasons this belongs on the wiki rather than only in a labour-economics reading list. First, it is a mechanism by which concentrated economic power reproduces the political conditions that sustain it — the structural analogue, on the labour side, of the homevoter dynamic on the land side, where those holding an asset vote to protect its value. Second, the lead author is the co-author of the assessment-gap literature on racial inequality in property taxation, and the two findings share a structure: a market institution that looks technical produces a distributional outcome that then entrenches itself politically.

The caveats are real and this page states them rather than the paper's headline. This is an unpublished working paper summarised here from its abstract, not a full read; the shift-share identification rests on the usual exclusion assumption that national industry-concentration shocks affect local voting only through local labour-market exposure; and a counterfactual that "narrowly" flips two presidential elections is, by construction, an extrapolation at the margin of the estimated effect rather than a robust finding about electoral outcomes. The turnout result is also directional in a specific partisan direction, which invites the usual caution about a single design carrying a politically loaded conclusion.

See Also

• Data as Labor — the Radical Markets Ch. 5 proposal built directly on platforms' monopsony power over user data
• Superstar Firms — the seller-side market-power account of the falling labor share, for contrast
• De Loecker, Eeckhout & Unger — markups — the parallel seller-side rent evidence
• Radical Markets (book page) — the discovery source for this page
• Rent-Seeking — the general concept of capturing rather than creating wealth
• Objection: homevoters will block LVT — the land-side analogue of the political-entrenchment mechanism electoral monopsony describes
• Narrative: Land and the Black-White Wealth Gap — the assessment-gap work by the same lead author

Sources

1. Alan Manning, Monopsony in Motion: Imperfect Competition in Labor Markets (Princeton University Press, 2003), Ch. 1 introduction. personal.lse.ac.uk/manning/work/mimintro.pdf (author's own posted copy) — used for the definition of monopsony as employer wage-setting power arising from labor-market frictions rather than a literal single buyer.
2. José Azar, Ioana Marinescu & Marshall Steinbaum, "Labor Market Concentration," NBER Working Paper No. 24147 (2017); published in Journal of Human Resources 57(S) (2022): S167–S199. nber.org/papers/w24147 — used for the concentration-wage elasticity estimate.
3. Eric A. Posner and E. Glen Weyl, Radical Markets: Uprooting Capitalism and Democracy for a Just Society (Princeton University Press, 2018), Ch. 3–5 — the discovery source; used for the VIP migrant-labor-monopsony proposal and the data-platform-monopsony argument. See also the wiki book summary.
4. Sydnee Caldwell, Arindrajit Dube & Suresh Naidu (2026), "Monopsony Makes it Big," NBER Working Paper 35608 (August 2026). nber.org/papers/w35608 — abstract fetched and read 2026-08-17; used for the firm-specific-labor-supply-elasticity framing, the three-source typology (search frictions, preference heterogeneity, employer concentration), the elasticity-to-wage-markdown measurement caution, the efficiency-wage interaction, and the antitrust/labor-regulation/immigration/macro policy scope. Scan depth: abstract only (B-claim); the full survey was not independently obtained this session.
5. Carlos Fernando Avenancio-León, Adelina Barbalau, Cyndi Hou & Alessio Piccolo (2026), "Firms as Electoral Monopsonies," NBER Working Paper 35676, August 2026, DOI 10.3386/w35676. nber.org/papers/w35676 — abstract page fetched and read 2026-09-17, with all quoted phrases verified verbatim against it — used for the "electoral monopsonies" coinage, the survey evidence on employer political communication, the passive/active channel distinction, the polarisation/platform-constraint/campaign-spending-substitution claims, the shift-share design and the Republican-turnout result, and the 2016/2024 counterfactual (B-claim; abstract-level only, the full working paper was not retrieved this pass, and no figures beyond those the abstract itself states are reported here).