we-solve (beta)

Econ-620 · economics · Economics

Effectiveness of antitrust remedies

LLM status Open unverified

Statement

Original identifier: I10 (atlas item 80). Classification: Remedy effectiveness causal evaluation research program. Original tractability label: Medium (editorial, not a complexity result).

Definitions and assumptions

Let a case $j$ have predecision covariates $X_j$ and eligible remedies $r\in\mathcal R_j$, including divestiture bundles, conduct restrictions, and interoperability obligations. The counterfactual $Y_{jt}(r)$ records quality-adjusted prices, output, entry, and service quality over a declared horizon $t=1,\ldots,T$. Remedy assignment is generally selected by the case, negotiations, and anticipated harm. Define enforcement resources, compliance monitoring, and buyer suitability as components of the regime. A target population of cases is specified independently of which cases were studied. Noncompliance is a post-remedy response, not a baseline confounder to control away automatically.

Formal research question

Identify or bound policy contrasts $\tau_t(r,r')=\mathbb E_{j\sim F_*}[Y_{jt}(r)-Y_{jt}(r')]$ and a restoration criterion relative to an identified no-merger competitive baseline $Y_{jt}(0)$. For a prespecified discrepancy function $d$, assess \[ \Pr_{j\sim F_*}\!\left(\max_{1\leq t\leq T}d(Y_{jt}(r),Y_{jt}(0))\leq\varepsilon\right), \] with a declared tolerance $\varepsilon$. Determine conditions for causal identification using randomized implementation where feasible, credible matched or synthetic comparison markets, or structural restrictions. Compare remedies by consumer welfare net of implementation and monitoring resource costs, with explicit selection sensitivity.

Interpretation and scope

Durable restoration means success over a stated finite horizon; finite data cannot prove that firms will never circumvent a remedy. A remedy can improve outcomes compared with an unremedied merger while still failing to restore the no-merger baseline. These are different contrasts and should be reported separately. Divestitures require a viable buyer, retained capabilities, and access to complementary assets; counting assets transferred does not establish competitive effectiveness. Conduct remedies depend on monitoring, enforceability, and firms' adaptation, making implementation part of the treatment rather than incidental detail. Case studies selected after unusual transactions do not automatically generalize to ordinary or blocked cases. Synthetic controls require credible counterfactual fit and no contaminating spillovers, and selection on unobserved anticipated harm can defeat matched comparisons. A useful extension would register the remedy, target outcomes, comparison strategy, and data access before implementation, then follow entrants and consumers beyond the immediate transaction. Cross-case comparison must allow heterogeneous industries and remedy objectives rather than rank all remedies by one unadjusted price average.

Source audit and status

Farrell (2003) is resolved to the verified Berkeley working-paper version [S1], avoiding an unverified book-chapter edition. [S2] is Kwoka's 2015 retrospective monograph. "FTC merger-retrospective literature" is not a complete citation; [S3] supplies one explicit 2017 staff study without claiming it uniquely exhausts that phrase. These references contain existing remedy evidence. The target population and restoration test are editorial extensions, with causal and generalization limits stated explicitly.

References

    },leftmargin=*]

  1. Joseph Farrell (2003). Negotiation and Merger Remedies: Some Problems. UC Berkeley Competition Policy Center working paper, August 2003. https://escholarship.org/uc/item/9p72k8fn

    Source role: Economic analysis of remedy negotiations; working-paper version used. Locator: Abstract and article metadata.

  2. John Kwoka (2015). Mergers, Merger Control, and Remedies: A Retrospective Analysis of U.S. Policy. MIT Press. https://mitpress.mit.edu/9780262028486/mergers-merger-control-and-remedies/

    Source role: Synthesis of merger retrospectives. Locator: Abstract and article metadata.

  3. Federal Trade Commission, Bureaus of Competition and Economics (2017). The FTC's Merger Remedies 2006--2012: A Report of the Bureaus of Competition and Economics. FTC staff report, January 2017. https://www.ftc.gov/reports/ftcs-merger-remedies-2006-2012-report-bureaus-competition-economics

    Source role: Explicit replacement for unspecified FTC retrospective literature. Locator: Report overview and case-study methodology.

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-620-A1 · Exploration · LLM-generated · gpt 6 astra ultra ·

GPT 6 Astra Ultra attempt on Econ-620

Effectiveness of antitrust remedies Buyer capacity, durable restoration, and an identification boundary

Question and scope

The recorded problem asks which divestiture or conduct remedies restore competition over a stated post-merger horizon, relative to the market that would have existed without the merger. Restoration differs from improving on an unremedied merger. Remedies also include buyer capabilities, asset maintenance, monitoring and implementation costs. The primary sources [farrell,ftc] motivate these distinctions but do not identify universal causal remedy effects. This attempt solves a capacity-and-cost model of divestiture, derives finite-horizon durability conditions, compares a monitored price rule, and proves a limit to identification of case-level restoration probabilities.

A divestiture must create sufficient effective capacity

There is a homogeneous product with inverse demand $P=A-BQ$, where $B>0$. Production is at constant marginal cost and there are no fixed costs or quality effects. Set $D=A-c>0$. Without the merger two independent firms have cost $c$ and no capacity limits. The merger combines them into a cost-$c$ monopolist, with no efficiency gain. A structural remedy transfers assets to an independent entrant whose cost is $c+\Delta$, $\Delta\ge0$, and whose usable capacity is $K\ge0$. The remaining incumbent retains cost $c$ and no capacity constraint. The two firms then play a simultaneous quantity game.
Theorem (Exact divestiture equilibrium and restoration): The unique equilibrium has entrant output and incumbent output \[ q_E=\min\left\{K,\left[\frac{D-2\Delta}{3B}\right]_+\right\}, \qquad q_I=\frac{D-Bq_E}{2B}, \] and price $P=c+(D-Bq_E)/2$. The no-merger benchmark has $q_0=D/(3B)$ for each firm and $P_0=c+D/3$, while the unremedied monopoly price is $P_M=c+D/2$. Thus \[ P-P_0=\frac B2(q_0-q_E)\ge0. \tag{1} \] The divestiture strictly improves price relative to monopoly if and only if $q_E>0$. It exactly restores the no-merger price if and only if $\Delta=0$ and $K\ge q_0$.
Proof: The best-response maps are $q_I=[(D-Bq_E)/(2B)]_+$ and $q_E={\operatorname{clip}}_{[0,K]}((D-\Delta-Bq_I)/(2B))$. They are each Lipschitz with constant $1/2$, so their composition has constant at most $1/4$; this gives at most one fixed point. Any equilibrium entrant output is at most $D/(2B)$, since $\Delta,q_I\ge0$. Thus incumbent output is strictly positive and its response has the unclipped expression above. Substituting it into the entrant response shows that the unconstrained positive fixed point is $(D-2\Delta)/(3B)$. If this is nonpositive, zero is the fixed point; if it exceeds $K$, $K$ is the fixed point. These cases give the stated formula and verify existence. The benchmark quantities follow from the same symmetric first-order conditions and the monopoly profit maximum. Subtracting prices gives (1). Since $\Delta\ge0$ implies $q_E\le q_0$, equality in (1) requires both $\Delta=0$ and adequate capacity, and those conditions suffice. ∎

The entrant's cost wedge can represent missing complementary assets, loss of personnel, or a financing disadvantage. A nominal divestiture package with positive physical capacity can have $q_E=0$ if its cost wedge is too high. Conversely, merely observing some entry does not establish restoration.

Theorem (A finite-horizon durability threshold): Suppose periods $t=1,\ldots,T$ are independent quantity games with unchanged demand, while entrant capacity and cost deteriorate as \[ K_t=K_0(1-\rho)^{t-1},\qquad \Delta_t=\Delta_0+\eta(t-1), \] where $0\le\rho<1$, $\Delta_0,\eta\ge0$. For $0\le\varepsilon<D/6$, the remedy keeps $|P_t-P_0|\le\varepsilon$ in every period if and only if \[ K_0\ge\frac{q_0-2\varepsilon/B}{(1-\rho)^{T-1}}, \qquad \Delta_0+\eta(T-1)\le3\varepsilon. \tag{2} \] These are finite-horizon conditions; no claim about unbounded durability follows.
Proof: Entrant output is nonincreasing in $t$ by the preceding theorem, so the largest price gap occurs at $T$. Equation (1) makes the desired condition equivalent to $q_{E,T}\ge q_0-2\varepsilon/B$. The right side is strictly positive by the bound on $\varepsilon$. A minimum of capacity and a positive-part cost expression is at least this threshold exactly when each term is. The capacity inequality gives the first part of (2); the cost inequality $(D-2\Delta_T)/(3B)\ge D/(3B)-2\varepsilon/B$ simplifies to the second. Reversing these steps proves sufficiency. ∎

Thus asset quantity and buyer viability impose separate restrictions. With a zero tolerance, exact restoration over multiple periods requires no effective cost deterioration and sufficient surviving capacity. This illustrates why a first-year price estimate need not answer the durability question.

A monitored conduct alternative

Instead of divesting, require the monopolist to charge no more than $P_0$. The firm can comply and meet demand, or choose a higher price. A violating firm is detected with probability $\mu\in[0,1]$ and pays a fixed fine $F>0$, independent of the size of its violation. Detection is absent for compliance; a compliant action is selected when the firm is indifferent. Monitoring uses real resources $m(\mu)$, with $m(0)=0$ and $m$ nondecreasing. Fines are transfers, while monitoring is a resource cost.
Proposition (Exact compliance and cost comparison): Compliance is optimal if and only if \[ \mu F\ge\frac{D^2}{36B}. \] It is implementable by this monitoring technology exactly when $F\ge D^2/(36B)$. The least required monitoring probability is $\mu^*=D^2/(36BF)$. Relative to the unremedied monopoly, the gain in consumer surplus net of minimum monitoring cost is \[ \frac{7D^2}{72B}-m(\mu^*). \] If a perfect divestiture has real implementation cost $\Gamma$ and no further distortions, the two remedies have identical prices and consumer surplus; on this consumer-surplus-minus-resource-cost criterion the cheaper of $\Gamma$ and $m(\mu^*)$ is preferred, provided the conduct remedy is feasible.
Proof: Monopoly profit is $\pi_M=D^2/(4B)$. Under the cap the profit-maximizing compliant price is $P_0$, since monopoly profit increases up to $P_M>P_0$. It yields quantity $2D/(3B)$ and profit $\pi_C=2D^2/(9B)$. Among violating prices the fixed expected fine changes no ranking, so the best deviation is $P_M$. Its excess gross profit is $\pi_M-\pi_C=D^2/(36B)$. The incentive inequality and feasibility follow. Since $m$ is nondecreasing, $\mu^*$ minimizes monitoring resources among feasible implementations. Linear-demand consumer surplus is $BQ^2/2$, giving the difference $2D^2/(9B)-D^2/(8B)=7D^2/(72B)$. The final comparison follows because perfect divestiture and compliance both yield the no-merger quantity and price. ∎

This comparison is intentionally restricted. It excludes evasion that cannot be detected, quality reductions, market entry, innovation, merger efficiencies, uncertain demand and endogenous fine collection. It also declares a consumer criterion; a total-surplus or distributionally weighted criterion requires the corresponding producer and financing terms. The calculation does not treat collected fines as a free social benefit.

Even randomized marginal effects do not identify individual restoration

Let a binary market outcome $Y(r)$ describe a remedy and $Y(0)$ the no-merger counterfactual. For $0\le\varepsilon<1$, restoration means $|Y(r)-Y(0)|\le\varepsilon$, equivalently equality. The notation $0$ here refers to no merger, not to an unremedied merger.
Theorem (Sharp restoration bounds from randomized marginals): Suppose a hypothetical experiment or valid identification design identifies $p={\mathbb{P}}(Y(r)=1)$ and $q={\mathbb{P}}(Y(0)=1)$, but imposes no cross-counterfactual dependence restriction. The sharp identified interval for the probability of restoration is \[ \left[|1-p-q|,\ 1-|p-q|\right]. \tag{3} \] The same failure of point identification applies to the all-period restoration event for paths that are constant over a fixed finite horizon.
Proof: Write $u={\mathbb{P}}(Y(r)=1,Y(0)=1)$. The four cells of the joint distribution are $u,p-u,q-u,1-p-q+u$. Their nonnegativity is equivalent to $\max\{0,p+q-1\}\le u\le\min\{p,q\}$. Equality of the two outcomes has probability $1-p-q+2u$. Substituting the endpoints gives (3). Every intermediate $u$ defines a valid joint distribution with exactly the observed marginals, proving sharpness. Setting each potential path equal to its corresponding binary variable makes restoration in all periods the same equality event, proving the path claim. ∎

Mean remedy effects therefore require less identification than the probability that a particular market is restored to its own counterfactual path. Actual remedy datasets add selection into enforcement and selection among remedies. To resolve the original empirical question one needs a defensible target distribution of cases, a no-merger comparison, monitoring and buyer data, and a design addressing that selection; estimating the preceding structural parameters is an additional route with additional assumptions. The explicit model supplies testable capacity and cost thresholds, not evidence that real remedies meet them.

Completion Estimate

56%

This is a post hoc, heuristic estimate for the full catalogue target. The attempt proves exact divestiture capacity and cost thresholds for finite-horizon restoration, a conduct-monitoring comparison, and sharp restoration-probability bounds from randomized marginals. Full remedy effectiveness still needs credible case selection and no-merger counterfactuals, measured buyer and monitoring performance, richer quality and entry responses, and joint outcome paths for the target case population.

References

Go to contribution →