Equilibrium Transition from Loss-Leader Competition: How Advertising Restrictions Facilitate Price Coordination in Chilean Pharmaceutical Retail

Yu (Jasmine) Hao

REVIEW 3 major objections 6 minor 9 references

Equilibrium Transition from Loss-Leader Competition: How Advertising Restrictions Facilitate Price Coordination in Chilean Pharmaceutical Retail

T0 review · 3 major / 6 minor · reviewed 2026-08-03 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A 2007 ban on comparative drug ads in Chile did not prevent coordination; it created it, by dismantling the economics of below-cost loss-leader pricing.

desk verdict A careful re-study of the Chilean pharmacy case whose central causal claim is not identified, because the spillover estimates are recovered under the very competitive-conduct assumption the paper is trying to overturn. read the letter →

arxiv 2512.22917 v2 pith:ZLIQ3KNH submitted 2025-12-28 econ.GN q-fin.EC

Yu (Jasmine) Hao This is my paper · ORCID

classification econ.GNq-fin.EC

keywords loss-leaderpricingadvertisingrestrictionspricecoordinationdemandspilloversequilibriumtransitionpharmaceuticalretailnestedlogitChileanpharmacies

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper argues that Chile's September 2007 restriction on comparative pharmaceutical price advertising did not simply make coordination easier to sustain; it destroyed the economic foundation of the pre-existing loss-leader equilibrium. Before the ban, the three chains priced 60 percent of products below cost because each drug customer brought 15,700–25,800 pesos of profit from other purchases, a cross-product spillover that made below-cost pricing rational. The ban severed that spillover, simultaneously making loss-leader pricing unprofitable for all chains and turning a coordinated price increase into each firm's static best response. Counterfactual simulations attribute the transition primarily to spillover disruption rather than to the drop in demand elasticity. A sympathetic reader would care because the result implies that a regulation intended to cool price comparisons can, in concentrated multi-product retail, unintentionally produce the coordination it was meant to prevent.

What carries the argument

The central object is the per-customer spillover bonus μ_i, the incremental profit a pharmacy earns on non-pharmaceutical sales for each pharmaceutical customer it attracts. In the static Nash-Bertrand pricing equations, μ_i enters the first-order condition as a negative marginal cost, so large μ_i rationalises prices below marginal cost; collapsing μ_i to zero makes loss-leader pricing unprofitable. Around this, the paper assembles a nested logit demand system with structural breaks at the ban, a dynamic Markov Perfect Equilibrium model with menu costs, a public trust stock, endogenous leader selection and follower compliance probabilities, and counterfactual comparisons that hold spillover

What would settle it

Store-level basket or loyalty-card data showing that cross-category purchase rates per pharmacy customer did not decline after the ban — i.e., customers kept buying the same non-drug items — would falsify the spillover-collapse mechanism. Similarly, if prices rose by comparable amounts in a control market or category that had no comparative price advertising and no loss-leader traffic, the advertising-ban channel would be called into question.

Watch

Extended reading notes

Core claim

Working from transaction-level data on 222 medicines, the paper reconstructs a coherent causal chain: comparative advertising broadcast who was cheapest, so undercutting paid; the September 2007 ban removed that broadcast, collapsing per-customer cross-product spillovers from thousands of pesos to near zero; with no traffic bonus to offset negative margins, 60 percent of products could no longer be rationally sold below cost; and the three chains sequentially raised prices 28–60 percent, led mainly by the smallest chain, with increases that never reverted. The paper's central claim is that this wave of coordinated increases was an equilibrium transition forced by demand-side disruption, not

Load-bearing premise

The load-bearing premise is that the three chains were playing static Nash-Bertrand competition both before and after the ban, so that matching predicted to observed prices recovers true spillovers; if the chains were already coordinating, the estimated collapse in spillovers is observationally equivalent to coordinated conduct and the mechanism is not separately identified.

Editorial extensions

If this is right

If spillover disruption is the mechanism, coordinated price increases should be permanent, not cyclically reversed; the paper reports no reversion.
Coordination should begin in the largest, most elastic, chronic markets, where lost spillover revenue was greatest; the Cox estimates show market size dominates timing.
The welfare effect is mostly a transfer: consumer losses of roughly CLP 13.6 billion versus firm gains of CLP 10.7 billion, with deadweight loss only 21–28 percent of consumer losses.
Antitrust authorities should ask whether a regulatory shock destroyed the mechanism sustaining a competitive equilibrium before reading observed coordination as cartel conduct.
The same logic predicts advertising bans in other multi-product retail settings will raise equilibrium prices fastest in categories that were previously loss leaders.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Editorial inference: the identification cuts the other way too — if prior work's 'safe market first' ordering is right, the spillover estimates here would be suspect; the paper's strongest evidence is the sequencing reversal, which is reduced-form and does not depend on the structural spillover recovery.
Editorial inference: the mechanism suggests a testable generalisation — in markets where below-cost pricing is funded by advertising-driven traffic, any policy that suppresses price signalling (online price-display bans, most-favoured-nation clauses, opaque pricing) should produce analogous coordination waves.
Editorial inference: the near-zero late-period spillovers may mix true spillover collapse with coordinated conduct; if regulators later find direct evidence of communication before December 2007, the causal story would shift from equilibrium transition toward standard collusion.
Editorial inference: welfare accounting excludes possible long-run harm from reduced entry and innovation in the chain segment; deadweight loss estimates are therefore a lower bound on total harm.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

This paper reinterprets the Chilean pharmacy price-fixing episode: the 2007 advertising ban destroyed cross-product spillovers that made loss-leader pricing profitable, so the coordinated price increases were a rational equilibrium transition rather than trust-building collusion. It does some things well. The demand estimation is careful: IV-GMM with post-lasso instruments, robustness across nesting bounds and fixed effects, and equilibrium price predictions that line up with observed prices. The Cox evidence that larger, more elastic markets coordinated first, and that Salcobrand led because it had the strongest exit incentive, is a genuine empirical challenge to Ale Chilet's trust-building story. The welfare accounting matches the actual fines, which is a nice external check. The paper is also transparent about its main identification problem.

The soft spot is load-bearing, and the paper names it. Section 5.2 concedes that observed prices can be rationalized either by high spillovers plus coordinated conduct or by low spillovers plus competitive conduct. The spillover series in Table 7 is then recovered by assuming static Nash-Bertrand pricing and matching predicted prices to observed prices. That means the dramatic collapse from 15,700–25,800 CLP to near-zero is exactly what you would expect if conduct changed from competition to coordination, regardless of what happened to true spillovers. So the central claim — that spillover disruption, not coordination, drove the transition — is not separately identified. The paper's own Appendix E makes this worse: the reduced-form regressions show no significant negative relationship between chronic-drug markups and non-pharmacy profit in the price-war period, which is the period where structural spillovers are supposed to be largest. The dynamic model also underpredicts Salcobrand's leadership share badly (28.3% vs. 74.6%), and the paper attributes the gap to firm-specific learning that is not in the model. There is also a tension between the disclaimer that the author takes no definitive stance on conduct and the conclusion that spillover disruption was the primary driver.

Net: this is a serious paper with a real empirical contribution in the descriptive findings and a sharp identification failure at the center of its causal story. It deserves a serious referee — not a desk reject — but the referee should push hard on Section 5.2 and Appendix E. If the author reframes the paper as a cautionary identification case study, or finds external instruments for spillovers, it could become valuable. As it stands, I would not cite the main mechanism.

Recommendation: send to peer review, with the expectation of major revision.

Referee Report

3 major / 6 minor

Summary. The paper studies the 2007–2008 Chilean pharmacy price-coordination episode. It argues that a ban on comparative price advertising eliminated cross-product demand spillovers that had sustained a loss-leader equilibrium, making below-cost pricing unprofitable for all three chains simultaneously and thereby triggering a rapid, non-collusive transition to coordinated higher prices. The empirical strategy combines a Cox hazard analysis of coordination timing, a nested logit demand model with structural breaks, spillover parameters recovered by inverting static Nash-Bertrand first-order conditions, and a dynamic model of sequential price leadership with belief updating. The paper concludes that spillover disruption—not reduced demand elasticity—was the primary driver of the equilibrium transition, with welfare losses of roughly CLP 13.6 billion to consumers and small deadweight loss.

Significance. If the central claim were identified, this would be a valuable contribution to the industrial-organization literature on equilibrium transitions, loss-leader pricing, and the unintended competitive effects of advertising regulation. The paper is transparent about many modeling choices, provides extensive robustness checks on demand estimation, and offers a rich dynamic framework that replicates several qualitative features of the episode. The descriptive findings—Salcobrand's leadership, the sequencing of coordination, the permanence of price increases—are interesting and potentially useful for future work. However, the main causal claim rests on spillover parameters that are not identified separately from the conduct regime. Because the paper itself acknowledges this observational equivalence, the structural counterfactual cannot support the conclusion that spillover disruption, rather than coordination, caused the observed price increases.

major comments (3)

[§5.2, Table 7] The spillover coefficients μ_i are recovered by minimizing squared deviations between Nash equilibrium prices and observed prices (Table 7 note). Under this inversion, higher observed prices mechanically produce near-zero μ_i if the model assumes competition; the model then 'predicts' that low spillovers lead to higher prices. The counterfactual in §7.4 (Table 12) uses these μ_i to attribute welfare losses to spillover disruption. But the paper itself states (§5.2) that high spillovers plus coordinated conduct and low spillovers plus competitive conduct are observationally equivalent. Therefore the estimated spillover collapse from 15,700–25,800 CLP to near-zero could simply reflect the change from competitive to coordinated conduct, not a genuine change in cross-product spillovers. The central claim is not identified without additional data or assumptions that distinguish conduct from s
[Appendix E, Table A5] The direct reduced-form evidence does not corroborate the timing of the claimed spillover collapse. The chronic-markup coefficient on non-pharmaceutical profit is -2.59*** in the pre-campaign period, but it is 0.41 (insignificant) during the price war (2006-11 to 2007-09)—precisely the period when the structural estimates in Table 7 imply the largest spillovers (15,700–25,800 CLP). The post-ban coefficient is -0.77 (insignificant). With only 11 observations in the price-war period and acknowledged endogeneity, this evidence cannot bear the weight of the structural estimates. At minimum, the paper should explain why the reduced-form and structural evidence diverge so sharply.
[§7.4, Table 12 and §8] The counterfactual decomposition that identifies spillover disruption as the primary driver is conditional on the estimated μ_i, which are only identified under the maintained assumption of static Nash-Bertrand competition. The comparison of 'Post-Ban' with 'Pre-Ban + T0' holds μ at pre-ban levels and re-solves the model; but if firms actually coordinated, the pre-ban μ estimates are not the relevant counterfactual spillover parameters, and the counterfactual equilibrium under alternative conduct would differ. Consequently, the conclusion in §8 that 'spillover disruption—not changes in demand elasticity—was the primary driver of equilibrium transition' is unsupported by the identified parameters. The paper's own qualification in §5.2 that estimates 'should not be interpreted as structural truth' is in tension with the strength of the abstract and conclusion.

minor comments (6)

[Table 7 note] The table note contains a LaTeX remnant: 'extbfPrice Tier'. Please fix the formatting.
[§3 vs Appendix C] The Cox results are inconsistent across tables. Table 4 reports a positive and significant Chronic coefficient (0.675**), while Appendix C Table A3 (Round 1) reports a negative and insignificant Chronic coefficient (-0.408). The paper should reconcile these differences or explain why the samples and specifications differ.
[§4.7] The text says 'Own-price elasticity increased by 2.65 percentage points' when the estimate changes from -6.70 to -4.06. This is a change in the absolute value of elasticity, not percentage points. Please rephrase.
[§7.2 and Tables 9-10] The model labels are inconsistent: 'Trust-Augmented' vs 'Standard MPE' in the text and Figure 4, but 'Trust-Building MPE' and 'Standard MPE' in Tables 9-10. Please standardize.
[§6.2] In the deviation-profit equation π^D_ijt(ℓ), the price sensitivity parameter is written as α_j without a period superscript, while other profit expressions use α^{Post}_j. Please clarify whether this is intentional and which elasticity is used for the deviating firm.
[Table 8 note] The units and conversion in Table 8 are confusing: menu costs are said to be in '000s CLP' but the main values are around 1,600, and the USD conversion is described as 'multiplied by 1000'. Please clarify the units and check the arithmetic.

Circularity Check

2 steps flagged · score 7.0 of 10

Core spillover estimates are a Nash-implied price residual; the counterfactual that 'spillover disruption' drove the price transition mechanically inverts that same residual, and the paper admits the competing coordination interpretation is observationally equivalent.

self definitional [Section 5.3 / Table 7 note; Eq. (8) in Section 5.1]
"Spillover coefficients µ_i estimated by minimizing squared deviations between Nash equilibrium prices and observed prices. ... Interpretation caveats: ... these estimates are identified under the assumption that firms play static Nash competition within each price tier. If actual conduct deviates from Nash (e.g., firms coordinate prices), estimated spillovers conflate true spillover effects with unobserved conduct."

Under the Bertrand first-order condition (Eq. 8), µ_i is the residual margin needed to make each observed price a Nash best response: µ_i = s/(-∂s/∂p) - (p-c). Before the ban, observed markups are negative (-15.8% to -12.1%), so µ_i must mechanically be large and positive; after the coordinated increases, observed markups are high, so µ_i must mechanically fall near zero. The paper's headline fact that spillovers 'collapsed from 15,700–25,800 CLP to near-zero' is therefore a restatement of the observed price increase through the assumed Nash conduct, not an independently measured traffic spillover.
fitted input called prediction [Section 7.3 / Figure 4 and Table 12; Section 8 conclusion]
"When I hold spillover at pre-ban levels ('Tier 0 Spillover'), the model predicts substantially lower average prices ... maintaining pre-ban spillover levels (15,700–25,800 CLP per customer) versus post-ban levels (7,700–10,600 CLP) reduces equilibrium prices by approximately 500-1000 CLP, or 7-14% relative to actual observed prices."

The counterfactual is the equilibrium of the very first-order condition used to define µ_i. Since µ_i was fitted as the per-unit subsidy that rationalizes pre-ban low prices, inserting it back into Eq. (8) must push prices down; the Table 12 result that consumer losses would be 36% lower is a transformed version of that fitted residual. Section 5.2 admits that high-spillover-plus-coordinated-conduct and low-spillover-plus-Nash-conduct are observationally equivalent, so this exercise cannot establish that spillover disruption, rather than a change in conduct, caused the transition. The Section 8 conclusion that 'spillover disruption—not changes in demand elasticity—was the primary driver of equilibrium transition' therefore depends on assuming exactly the Nash conduct whose change is the le

full rationale

The paper is unusually transparent about its core identification problem: Section 5.2 spells out the observational equivalence between high spillovers with coordinated conduct and low spillovers with competitive conduct, and the Table 7 note warns that estimates are valid only under static Nash competition. But the main quantitative claim is built by inverting that assumption. The spillover coefficient is recovered by minimizing the distance between Nash equilibrium prices and observed prices, so the pre-ban/post-ban change in µ_i is essentially the price markup residual after imposing Bertrand behavior. Calling that residual a 'demand spillover' and then simulating what happens if the residual is restored is a fitted-input-called-prediction exercise: the counterfactual is close to a mechanical implication of the estimating equation. The Appendix E reduced-form regressions offer some independent support for chronic-drug spillovers in the pre-campaign period, but the price-war coefficient is positive and insignificant, which undercuts the structural claim that spillovers were largest precisely during the period of largest estimated µ_i. The dynamic model and Cox sequencing results add real descriptive content, but they do not separately identify the spillover mechanism from conduct. No self-citation issue arises here; the prior work by Alé Chilet is external to this author. Overall, the central mechanism decomposition is substantially circular, though not a fully empty derivation, so a score of 7 is appropriate.

Assumptions & free parameters 8 free parameters · 7 assumptions · 1 invented entities

The central narrative rests on a chain of estimated objects: demand elasticities, spillover bonuses backed out from Nash price-matching, menu costs, belief parameters, and several imposed bounds (sigma<=0.9, alpha>=0.01, market-size normalization). The most consequential is mu_i, identified only under the Nash assumption the paper concedes is observationally equivalent to coordination. These choices, not external benchmarks, carry the main counterfactual result.

free parameters (8)

Demand spillover bonus mu_i = Tier 0: 16,105.5 CLP (homogeneous); heterogeneous CV 25,756.4, FA 16,074.7, SB 15,660.1. Tier 1: 8,138.7; Tier 2: -48.8
Recovered by matching Nash equilibrium prices to observed prices. This fitted series is the paper's central evidence that spillovers collapsed and caused coordination.
Menu costs kappa_i = CV 1,657.8; FA 2,376.8; SB 1,133.7 (000 CLP, Table 8)
Estimated in the dynamic model; Salcobrand's lower menu cost is used to rationalize its leadership role.
Belief parameters lambda0, lambda1 = tau(0)=0.00, tau(50)=0.84, tau(100)=0.86, tau(200)=0.89 (Table 8)
Logistic trust-formation parameters estimated with a penalty anchoring endpoints; drive follower compliance beliefs.
Targeting propensity psi_bar = 0.2 calibrated (0.25 used in Table 9)
Calibrated to match the empirical probability of market targeting; not estimated from first principles.
Nesting parameter sigma = 0.90 (upper bound binding in preferred specification)
Bound imposed to ensure random utility maximization; unconstrained estimates exceed 1, so the value is effectively fixed by assumption.
Market size normalization N_j = N_j = 7 * max_t Q_jt per product
Chosen normalization to define outside-good share; affects all elasticity and welfare magnitudes.
Discount factor delta = 0.95
Assumed, not estimated; used in all dynamic value functions.
Alpha lower bound and mean-utility trimming = alpha >= 0.01; trim to 0.2-0.8 quantiles
Imposed regularization in demand estimation (Section 4.5) that affects estimated elasticities.

assumptions (7)

domain assumption Nested logit demand with i.i.d. Type I extreme value errors and nesting parameter sigma in [0,1)
Section 4.1 specifies the RUM; the paper imposes sigma<1 to ensure consistency.
domain assumption Lagged prices and BLP-style characteristics are valid instruments, uncorrelated with contemporaneous demand shocks
Section 4.3 relies on this for IV-GMM identification; exogeneity is asserted, not tested.
domain assumption Firms behave as static Nash-Bertrand competitors when spillover parameters are estimated
Section 5.2 explicitly assumes Nash conduct to identify mu_i, while acknowledging the observational equivalence with coordinated conduct.
ad hoc to paper Market size normalization N_j = 7 * max_t Q_jt
Section 4.5 defines this scaling to ensure meaningful outside-good shares; it is a modeling choice, not derived from data.
ad hoc to paper Sigma bound (0,0.9) and price coefficient lower bound alpha>=0.01
Section 4.5 imposes these bounds for stability; they shape the elasticity estimates that feed the counterfactuals.
domain assumption Markov Perfect Equilibrium with logistic belief updating and non-decreasing trust stock
Section 6 models firms' dynamic behavior under MPE and equation (12) imposes the logistic belief structure.
standard math Proportional hazards assumption in Cox regressions
Section 3 uses Cox proportional hazards; the model assumes multiplicative baseline hazards.

invented entities (1)

Trust stock s_t
purpose: Latent state accumulating successful coordination events; drives follower compliance beliefs via logistic function (equation 11-12)
Not directly observed; constructed from hand-coded coordination successes and discretized. The dynamic model's leadership predictions fit the data poorly (Table 10), and the trust mechanism is not separately falsified.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Equilibrium Transition from Loss-Leader Competition: How Advertising Restrictions Facilitate Price Coordination in Chilean Pharmaceutical Retail." pith.science (2026). https://pith.science/paper/ZLIQ3KNH

@misc{pith2026251222917,
  author       = {Pith},
  title        = {Pith review of: Equilibrium Transition from Loss-Leader Competition: How Advertising Restrictions Facilitate Price Coordination in Chilean Pharmaceutical Retail},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZLIQ3KNH}},
  note         = {Machine review of arXiv:2512.22917}
}

read the original abstract

Between December 2007 and April 2008 Chile's three retail pharmacy chains coordinated price increases on 222 medicines, weeks after advertising restrictions ended the comparative-price war that drove prices below cost. I study the transition with a demand-grounded structural model. The mechanism has two parts. Store traffic: comparative-price ads broadcast who is cheapest, so undercutting pays, yielding a below-cost war. Belief: a coordinated increase holds only if rivals expect it matched. The advertising ban moves both: by collapsing price sensitivity it makes undercutting unprofitable for the inelastic majority of drugs, so the coordinated price becomes a static best response, and as a public event it shifts beliefs, releasing the wave. A dynamic model estimated by simulated method of moments reproduces the path--the war, the failed attempts, and the post-ban coordination. The harm is distributional: a transfer to supra-competitive rents, with small deadweight loss because post-ban demand is inelastic.

Figures

Figures reproduced from arXiv: 2512.22917 by the authors.

**Figure 1.** Sequential Price Coordination: Marvelon-20 Oral Contraceptive (a) First Price Increase (December 2007) (b) Second Price Increase (January 2008) Notes: Daily prices (lines) and quantities (bars) for Marvelon-20 across three chains. Salcobrand led both increases, with competitors following within 2–4 days. First increase: +130% (Dec 2007); second increase: +40% (Jan 2008). Source: Chilean Competition Authority [PITH_… view at source ↗

**Figure 2.** Prediction of Market Share and Equilibrium Price [PITH_FULL_IMAGE:figures/full_fig_p019_2.png] view at source ↗

**Figure 3.** Cumulative Coordination Events Over Time: Simulated vs. Actual [PITH_FULL_IMAGE:figures/full_fig_p030_3.png] view at source ↗

Figures from the paper (2 more)

**Figure 4.** Figure 4: Counterfactual Spillover Scenarios: Comparison of Trust-Augmented vs. Stan [PITH_FULL_IMAGE:figures/full_fig_p032_4.png]

**Figure 5.** Figure 5: Decomposition of Total Welfare Change: Consumer Surplus vs. Producer [PITH_FULL_IMAGE:figures/full_fig_p033_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

9 extracted references

[1]

Aguirregabiria, V. (1999). The dynamics of markups and inventories in retailing firms. The review of economic studies, 66(2):275–308. 36 Aguirregabiria, V. and Jeon, J. (2020). Firms’ beliefs and learning: Models, identifica- tion, and empirical evidence.Review of Industrial Organization, 56(2):203–235. Aguirregabiria, V. and Magesan, A. (2020). Identific...

1999
[5]

investigate how trust evolves and leadership patterns differ across firms—important questions about coordination mechanics. This paper investigates how regulatory changes to demand structure force 41 equilibrium transitions—a fundamentally different causal question requiring different theoretical framework and empirical strategy. Both approaches can be si...

2006
[6]

Table A2: Wholesale Cost Changes During Equilibrium Transition Period (Nov 2007– May

Prevalence of negative margins:Count of products with negative markup in each tier. Table A2: Wholesale Cost Changes During Equilibrium Transition Period (Nov 2007– May

2007
[7]

The contrast between rounds reflects optimal coordination sequence

shows chronic medications have substantially larger effects when interacted with market size (Mkt×Chronic: 2.498***), indicating chronic products in larger markets coordinated faster in round two. The contrast between rounds reflects optimal coordination sequence. In Round 1, firms target high-volume products where absolute profits are largest: each perce...

2007
[17]

Share (Post)

D.2 Institutional Context Al´ e Chilet (2016) provide extensive institutional analysis of this case, including detailed financial data showing negative margins in competitive period, internal communications confirming firms’ coordination intentions, description of the sequential leadership mech- anism, and timeline of regulatory proceedings and appeals. T...

2016
[59]

Nevo, A. (2001). Measuring market power in the ready-to-eat cereal industry.Econo- metrica, 69(2):307–342. N´ u˜ nez, J., Rau, T., and Rivera, J. (2010). Informe pericial sobre el requerimiento de la fne en contra de farmacias ahumada s.a. y otros, rol c nº184-2008. Technical report. [Peritaje t´ ecnico]. OECD (2010). OECD Reviews of Health Systems: Chile

2001
[2008]

Statistic Value Mean 3.33% Median 2.36% Std dev 5.83% 75th percentile 4.45% Notes:Percent changes computed from Salcobrand wholesale prices between November 2007 and May 2008 for the 222 products in the coordination sample. The modest wholesale cost increases (mean 3.33%, median 2.36%) contrast sharply with retail price increases averaging 28.4% (first ro...

2007
[2010]

Pharmaceutical pricing policies in a global market

OECD (2014). Pharmaceutical pricing policies in a global market. Policy Actions for Affordable and Accessible Pharmaceuticals Assessment. Rao, R. C. and Syam, N. (2001). Equilibrium price communication and unadvertised specials by competing supermarkets.Marketing Science, 20(1):61–81. Rust, J. (1987). Optimal replacement of gmc bus engines: An empirical m...

2014

Show all 9 references

[2018]

safer” (low-elasticity) markets to accumulate evidence that rivals will maintain coordinated prices, only then expanding to “riskier

provides the foundational empirical documenta- tion of the Chilean pharmacy coordination episode analyzed in this paper. Al´ e Chilet (2016) documents the gradual emergence of collusion among the three pharmacy chains, establishing that firms sequentially expanded coordination...

2016

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.

[1] [1]

Aguirregabiria, V. (1999). The dynamics of markups and inventories in retailing firms. The review of economic studies, 66(2):275–308. 36 Aguirregabiria, V. and Jeon, J. (2020). Firms’ beliefs and learning: Models, identifica- tion, and empirical evidence.Review of Industrial Organization, 56(2):203–235. Aguirregabiria, V. and Magesan, A. (2020). Identific...

1999

[2] [5]

investigate how trust evolves and leadership patterns differ across firms—important questions about coordination mechanics. This paper investigates how regulatory changes to demand structure force 41 equilibrium transitions—a fundamentally different causal question requiring different theoretical framework and empirical strategy. Both approaches can be si...

2006

[3] [6]

Table A2: Wholesale Cost Changes During Equilibrium Transition Period (Nov 2007– May

Prevalence of negative margins:Count of products with negative markup in each tier. Table A2: Wholesale Cost Changes During Equilibrium Transition Period (Nov 2007– May

2007

[4] [7]

The contrast between rounds reflects optimal coordination sequence

shows chronic medications have substantially larger effects when interacted with market size (Mkt×Chronic: 2.498***), indicating chronic products in larger markets coordinated faster in round two. The contrast between rounds reflects optimal coordination sequence. In Round 1, firms target high-volume products where absolute profits are largest: each perce...

2007

[5] [17]

Share (Post)

D.2 Institutional Context Al´ e Chilet (2016) provide extensive institutional analysis of this case, including detailed financial data showing negative margins in competitive period, internal communications confirming firms’ coordination intentions, description of the sequential leadership mech- anism, and timeline of regulatory proceedings and appeals. T...

2016

[6] [59]

Nevo, A. (2001). Measuring market power in the ready-to-eat cereal industry.Econo- metrica, 69(2):307–342. N´ u˜ nez, J., Rau, T., and Rivera, J. (2010). Informe pericial sobre el requerimiento de la fne en contra de farmacias ahumada s.a. y otros, rol c nº184-2008. Technical report. [Peritaje t´ ecnico]. OECD (2010). OECD Reviews of Health Systems: Chile

2001

[7] [2008]

Statistic Value Mean 3.33% Median 2.36% Std dev 5.83% 75th percentile 4.45% Notes:Percent changes computed from Salcobrand wholesale prices between November 2007 and May 2008 for the 222 products in the coordination sample. The modest wholesale cost increases (mean 3.33%, median 2.36%) contrast sharply with retail price increases averaging 28.4% (first ro...

2007

[8] [2010]

Pharmaceutical pricing policies in a global market

OECD (2014). Pharmaceutical pricing policies in a global market. Policy Actions for Affordable and Accessible Pharmaceuticals Assessment. Rao, R. C. and Syam, N. (2001). Equilibrium price communication and unadvertised specials by competing supermarkets.Marketing Science, 20(1):61–81. Rust, J. (1987). Optimal replacement of gmc bus engines: An empirical m...

2014