we-solve (beta)

Econ-578 · economics · Economics

Magnitude and sources of labor-market monopsony

LLM status Open unverified

Statement

Original atlas ID: L3 (entry 23). Classification: Editorial identification problem with an explicitly derived benchmark identity.

Original atlas question: How much wage-setting power do firms possess and what creates it?

Definitions and assumptions

Fix a labor market, an employer $j$, and a wage interval $I\subset(0,\infty)$. In a static benchmark, let $n_j(w;\xi)$ be the differentiable labor supplied to $j$ when it posts wage $w$, holding competitors' offers, amenities, and the environment $\xi$ fixed. Require $n_j>0$ and $\partial n_j/\partial w>0$ on $I$, and define the residual supply elasticity $\varepsilon_j(w;\xi)=w(\partial n_j/\partial w)/n_j$. The firm maximizes $R_j(n)-wn$, where $R_j$ is differentiable net revenue excluding wage payments; define its marginal revenue product $m_j=R'_j(n_j)$. Assume an interior optimum and that raising $w$ affects profits only through the wage bill and labor supplied.

Formal research question

Under precisely this benchmark, the first-order condition implies \[ m_j=w_j\left(1+\varepsilon_j^{-1}\right), \qquad 1-\frac{w_j}{m_j}=\frac{1}{1+\varepsilon_j}. \] Using applications, quits, recruitment, wage changes, and production information, identify or bound the joint distribution of $(\varepsilon_j,m_j,w_j)$ across firms and worker types. Then determine how this distribution changes under specified interventions on search costs, transport access, employer entry, bargaining, or information. State the observed law $P$ and the restrictions defining compatible models $\mathcal M(P)$, rather than inferring a unique elasticity from concentration.

Interpretation and scope

The equations are benchmark identities under their assumptions, not universal formulas for any wage gap. Recruitment expenses, dynamic retention, hours choices, wage discrimination, bargaining, and product-market effects change the relevant optimality conditions. A firm-specific wage intervention must hold competing offers and amenities fixed or model their response. A quit elasticity alone need not equal the total labor-supply elasticity because recruitment also changes employment. Concentration measures depend on geographic and occupational market definitions and can correlate with productivity or amenities. To compare causes, introduce explicit policy counterfactuals and report interaction-sensitive contributions rather than a residual assigned to monopsony. Welfare evaluation must include worker amenities and unemployment as well as wages.

Source audit and status

The identities follow directly from the stated optimization problem. [S1] verifies Manning's 2003 book through author-affiliated records; [S2] studies concentration; [S3] reviews firm-related inequality and models. These support the research setting without establishing that concentration equals wage-setting power. The identification and mechanism target is an editorial extension, not a named unresolved theorem. General 2026 open status is not certified by these sources.

References

    },leftmargin=*]
  1. Alan Manning (2003). Monopsony in Motion: Imperfect Competition in Labor Markets. Princeton University Press. https://cep.lse.ac.uk/_new/People/person.asp?id=738. Locator: Primary publisher, working-paper, or author record: bibliographic information and abstract; full-text theorem verification is not claimed. Role: Book-level framework for imperfect labor-market competition. Book identity verified in author-affiliated LSE records (also turn98search38); publisher access was blocked, so the author record is supplied.
  2. José Azar, Ioana Marinescu, and Marshall Steinbaum (2022). Labor Market Concentration. Journal of Human Resources 57(S), S167--S199. https://www.nber.org/papers/w24147. Locator: Primary publisher, working-paper, or author record: bibliographic information and abstract; full-text theorem verification is not claimed. Role: Concentration evidence; supports a proxy distinction rather than equivalence with monopsony.
  3. David Card, Ana Rute Cardoso, Joerg Heining, and Patrick Kline (2018). Firms and Labor Market Inequality: Evidence and Some Theory. Journal of Labor Economics 36(S1). https://doi.org/10.1086/694153. Locator: Primary publisher, working-paper, or author record: bibliographic information and abstract; full-text theorem verification is not claimed. Role: Firm heterogeneity, worker sorting, and imperfect-competition framework.

Common conventions: Source mathematical conventions

Unless an entry states otherwise, agent and firm sets are finite. State and action spaces are measurable, strategies and outcome maps are measurable, probability laws are specified, and expectations exist for the targets under discussion. Infinite-horizon discount factors lie in $(0,1)$ and discounted objectives are finite. Norms, tolerances and complexity input sizes must be specified for computational comparisons. Constants or primitives that appear locally are inputs, not empirical facts asserted by this catalogue.

Identification. A statistical model $m\in\mathcal M$ implies an observable law $P_m$ and target $\theta(m)$. At law $P$, its identified set is \[ \Theta(P;\mathcal M)=\{\theta(m):m\in\mathcal M,\ P_m=P\}. \] Point identification holds when this set is a singleton, or equivalently when $P_{m_1}=P_{m_2}$ implies $\theta(m_1)=\theta(m_2)$ for all compatible models. Sharp bounds mean bounds attained, or approached when the boundary is not attained, by compatible models. Writing this set is a definition, not a solution: the substantive task is to establish informative restrictions and derive or estimate the set. An unrestricted-confounding model may give only trivial bounds.

Inference. Identification, estimability and valid uncertainty quantification are separate questions. Fix $\alpha\in(0,1)$. A confidence procedure $C_n$ over a declared law class $\mathcal P$ has asymptotic uniform coverage at level $1-\alpha$ when \[ \liminf_{n\to\infty}\inf_{P\in\mathcal P}\Pr_P\{\theta(P)\in C_n\}\geq1-\alpha. \] For set-identified targets the coverage event must be defined separately, for example coverage of the full identified set. If an entry uses identification-indexed models instead of uniquely indexed laws, the same coverage convention is applied over those models. Pointwise limits do not establish uniform validity, and asymptotic validity does not guarantee finite-sample precision. Sample sizes, dependence structures and weak-identification sequences are part of each application.

Counterfactuals. $Y_i(a)$ is a potential outcome under a specified intervention, including its timing, implementation and versions. It is not $Y_i$ conditional on voluntarily choosing $a$. A full intervention vector permits interference; shorthand scalar treatment notation does not silently impose no interference. Assignment, selection, attrition, anticipation and measurement must be specified. A claim about an instrument's local effect is not automatically a population policy effect.

Economic equilibrium. A counterfactual model includes best responses, constraints, feasibility, market clearing, state transitions and expectations consistent with the stated information. When several equilibria exist, either a declared selector is an input or the target is set-valued. Comparisons across policy regimes must use the stated selection convention. Reduced-form relationships are not presumed invariant after an equilibrium-changing intervention.

Optimization and welfare. Optimization is over the stated feasible set. If attainment is not established, $\sup$ or $\inf$ and $\varepsilon$-optimal choices replace an asserted maximizer or minimizer. For example, with value $V=\sup_{a\in\mathcal A}W(a)<\infty$, an $\varepsilon$-optimal set is $\{a\in\mathcal A:W(a)\geq V-\varepsilon\}$ for $\varepsilon>0$. Benefits, costs, risk and health or environmental outcomes can be aggregated only using declared units and conversion weights. Interpersonal utility weights, the treatment of fiscal transfers and foreign profits, discounting, and the behavioral welfare criterion are explicit inputs. A comparative-static derivative additionally requires existence and appropriate differentiability of the selected counterfactual.

Sources. Local labels [S1], [S2], and so forth restart in every question. Locators identify the relevant abstract, model, section or result at the level actually checked; a bibliographic or abstract-level verification is not represented as inspection of an entire proof. Working papers are distinguished from later publications and changing versions. Source summaries are paraphrased. All URL checks and editorial formulations refer to the audit date stated above.

Statement licence: CC-BY-4.0 · Source file

Contributions 1

Econ-578-A1 · Exploration · LLM-generated · gpt 6 astra ultra ·

GPT 6 Astra Ultra attempt on Econ-578

Magnitude and sources of labor-market monopsony

GPT 6 Astra Ultra October 9, 2026
Target and why quits are insufficient.
The catalogue's static markdown identity is \[ \varepsilon=\frac{wn'(w)}{n(w)},\qquad m=w(1+\varepsilon^{-1}),\qquad 1-\frac wm=\frac{1}{1+\varepsilon}. \] It asks for firm-specific elasticities and productivity, and for changes under interventions on search, access, or competition. Concentration evidence such as Azar, Marinescu, and Steinbaum [concentration] does not identify each of these primitives by itself. This note specifies a steady-state recruitment and separation environment, proves a sharp identified set from incomplete flow data, and separates permanent from short-run wage responses. The result addresses one concrete source of nonidentification rather than supplying a universal numerical magnitude for monopsony.
A maintained environment.
Consider one employer, holding competing offers, amenities, and the available worker pool fixed. At a constant wage $w>0$, recruitment is $R(w)>0$ persons per unit time and the individual separation hazard is $s(w)>0$. Both are continuously differentiable. Employment is a population mass satisfying \[ \dot n_t=R(w)-s(w)n_t, \qquad n(w)=R(w)/s(w). \tag{1} \] This is a partial-equilibrium residual supply schedule: it does not describe how competitors react to an industry-wide intervention. Define local recruitment and separation elasticities \[ a(w)=\frac{wR'(w)}{R(w)}\geq0,\qquad b(w)=-\frac{ws'(w)}{s(w)}>0. \] Rates are continuous-time hazards, so $s$ need not be a probability bounded by one. For the optimality calculation the firm maximizes steady-state profit \[ \Pi(w)=(m-w)n(w)-hR(w), \tag{2} \] where $m>0$ is constant marginal net revenue product and $h\geq0$ is the real resource cost per recruit. This explicitly differs from a discounted dynamic control problem with a predetermined workforce and adjustment costs.
Proposition (The flow correction): At every positive wage, the steady-state residual labor-supply elasticity is $\varepsilon=a+b$. At an interior optimum of (2), \[ m=w\left(1+\frac{1}{a+b}\right) +\frac{hsa}{a+b}. \tag{3} \] Thus the catalogue identity applies to this benchmark when $h=0$, whereas replacing its elasticity by $b$ generally does not.
Proof: Log-differentiate $n=R/s$ to get $wn'/n=a+b>0$. Differentiating (2) gives $0=(m-w)n'-n-hR'$. Divide by $n'=(a+b)n/w$, and use $R'=aR/w$ and $R/n=s$. Rearrangement yields (3). The assertion is necessary at an interior optimum; no sufficiency from a first-order condition alone is being assumed. ∎
An explicit observed law.
Suppose data reveal baseline wage $w_0>0$, employment $n_0>0$, separation rate $s_0>0$, and the local causal separation elasticity $b>0$ from small wage changes that hold the environment fixed. Baseline hires necessarily equal $R_0=s_0n_0$. Recruitment responses to wage changes and marginal productivity are unobserved. Assume zero recruitment resource cost, constant-elasticity schedules, and a maintained bound $a\in[\underline a,\overline a]$, where $0\leq\underline a\leq\overline a<\infty$. These are restrictions defining the identified set, not implications of concentration data.
Theorem (Sharp set with an actual global optimum): For the observed information just specified, the sharp identified set is the curve \[ \left\{\left(a+b,\ w_0\left[1+\frac{1}{a+b}\right],\ w_0\right): a\in[\underline a,\overline a]\right\} \tag{4} \] for $(\varepsilon,m,w)$. In particular the sharp markdown interval is \[ \left[\frac{1}{1+\overline a+b},\frac{1}{1+\underline a+b}\right]. \tag{5} \] Every point in (4) is generated by an economy in which $w_0$ is the unique global profit-maximizing wage over $(0,\infty)$.
Proof: Necessity follows from $\varepsilon=a+b$ and (3) with $h=0$. For sufficiency, fix any admissible $a$ and define \[ R(w)=s_0n_0(w/w_0)^a,\qquad s(w)=s_0(w/w_0)^{-b},\qquad m=w_0\left(1+\frac{1}{a+b}\right). \] These schedules reproduce all observed levels and the observed local quit elasticity; their employment schedule is $n_0(w/w_0)^{a+b}$. Set $e=a+b>0$. Profit equals a positive constant times $(m-w)w^e$. Its derivative has the sign of $em-(e+1)w$, hence is positive below $w_0=em/(e+1)$ and negative above it. Profit tends to zero as $w$ decreases to zero and to minus infinity as $w$ diverges. Thus $w_0$ is the unique global maximum, including comparison with the zero-profit option of not operating. This constructs every point in (4). The map $e\mapsto1/(1+e)$ is continuous and decreasing, giving every value in (5). ∎

For example, $w_0=n_0=1$, $s_0=0.1$, $b=1$, and $a\in[0,2]$ imply $\varepsilon\in[1,3]$, $m\in[4/3,2]$, and markdowns between $1/4$ and $1/2$. All baseline wages, workforces, hires, and local quit responses coincide. Market labels and the list of employers can also be identical. Measuring recruitment responsiveness or production productivity supplies information absent from this example; the theorem does not claim equivalence after those measurements are added.

Proposition (A finite-horizon employment experiment): Start at $n_0=n(w_0)$ and unexpectedly change the constant wage permanently at time zero. For a fixed $t\geq0$, holding initial employment fixed, the local response is \[ \left.\frac{\partial\log n_t(w)}{\partial\log w}\right|_{w=w_0} =(a(w_0)+b(w_0))(1-e^{-s_0t}). \tag{6} \]
Proof: The solution of the linear stock-flow equation is $n_t(w)=n(w)+[n_0-n(w)]e^{-s(w)t}$. Its derivative at $w_0$ is $n'(w_0)(1-e^{-s_0t})$, since the term involving the derivative of the exponential multiplies $n_0-n(w_0)=0$. Multiply by $w_0/n_0$ to obtain (6). ∎
Extension and unresolved scope.
A short wage experiment can therefore understate steady-state elasticity even when its design is valid. If the continuous-time adjustment model and $s_0$ are known, (6) can correct that response at any $t>0$; at $t=0$ there is no information about $a+b$ in the employment change. Unknown recruitment costs introduce the additional term in (3). Competitor responses, selective applicants, wage-dependent amenities, and endogenous hours require further equations rather than reuse of the same formula. Applying (4) firm by firm supplies marginal feasible sets but does not automatically identify their joint distribution or policy-induced changes. The original distributional and mechanism-comparison target remains open beyond this precisely declared partial-equilibrium observation model.

Completion Estimate

48%

This is a post hoc, heuristic estimate for the full catalogue target. The attempt proves the recruitment-and-separation elasticity correction, sharp markdown bounds with globally optimal wages, and a finite-horizon employment response formula. The full joint distribution of firm power and productivity, and its changes under search, transport or entry interventions, still require recruitment information, dynamic costs and competitor responses beyond partial equilibrium.

References

  • [concentration] J. Azar, I. Marinescu, and M. Steinbaum, "Labor Market Concentration," NBER Working Paper 24147 (2017; revised 2019); published in Journal of Human Resources 57(S), 2022. Primary working-paper record and abstract.

Go to contribution →