How And Why The Per Se Rule Against Price-Fixing Went Wrong

This document is available in two formats: this web page (for browsing content) and PDF (comparable to original document formatting). To view the PDF you will need Acrobat Reader, which may be downloaded from the Adobe site.

How and Why the Per Se Rule Against Price-Fixing Went Wrong


Sheldon Kimmel*

March 2006


Most scholars believe the Supreme Court dropped its per se rule against price-fixing in Appalachian Coals (1933), re-instituting that rule in Socony-Vacuum (1940), but that the rule ignored "reasonableness" until BMI (1979), and that Maricopa (1982) relied on Socony to step back from "reasonableness" again. However, the view that Socony's per se rule had nothing to do with "reasonableness" came from unreasonably ignoring Socony's comments on Appalachian Coals, which came from misunderstanding Appalachian Coals by ignoring the economic implications of the facts the district court found. Those implications show that Appalachian Coals, Socony, and BMI all gave the same price-fixing rule.

I. Introduction

As Broadcast Music, Inc. v. CBS (441 U.S. 1 [1979]) explains, the per se rule against price-fixing isn't to be taken literally. In BMI the Court noted that the defendant was clearly fixing prices "in the literal sense ... But this is not a question simply of determining whether two or more potential competitors have literally 'fixed' a 'price.' As generally used in the antitrust field, 'price fixing' is a shorthand way of describing certain categories of business behavior to which the per se rule has been held applicable." (441 U.S. 1, 9-10). It isn't clear how much of BMI survived Arizona v. Maricopa County Medical Society (457 U.S. 332 [1982 ]), but BMI's view that "price fixing in a literal sense, ... is not price fixing in the antitrust sense" seems to have survived since that language is still being cited (Texaco v. Dagher (547 U.S. ___ [2006]).

Although BMI's non-literal view of price-fixing has survived, the survival of its view on when to apply the per se rule is less clear.(1) In deciding whether to apply the per se rule, the BMI Court said that "our inquiry must focus on whether ... the practice facially appears to be one that would always or almost always tend to restrict competition and decrease output, and in what portion of the market, or instead one designed to 'increase economic efficiency and render markets more, rather than less, competitive.' " (441 U.S. 1, 19-20).

Unfortunately, three years later Maricopa rejected the defendants' claim that procompetitive justifications for their agreements meant that they were not illegal per se (457 U.S. 332, 351). In rejecting that claim, the Maricopa Court relied on its reading of the claim in U.S. v. Socony-Vacuum Oil (310 U.S. 150 [1940]) that "Whatever economic justification particular price-fixing agreements may be thought to have, the law does not permit an inquiry into their reasonableness. They are all banned" (457 U.S. 332, 351 quoting 310 U.S. 150, 224). As Maricopa used that quote, it seemed to say that "reasonableness" played no role in either the Socony Court's per se rule against price-fixing, or the Maricopa Court's rule. However, this paper shows that in fact Socony anticipated the standard that is commonly attributed to BMI (i.e., Maricopa's misreading of Socony is how the per se rule went wrong, a more reasonable reading of Socony may allow a return to a more reasonable use of the per se rule).

That Socony anticipated BMI may seem surprising since Socony is commonly misread by ignoring its views on Appalachian Coals v. U.S. (288 U.S. 344 [1933]), views that are widely ignored since they make sense only if we understand the economic implications of the facts the Appalachian Coals district court found (implications which have not been sufficiently spelled out until now). Thus, while the Socony Court clearly felt that "price-fixing agreements are unlawful per se" (310 U.S. 150, 218), and while it agreed with Appalachian Coals that 137 coal producer defendants created an exclusive selling agency "to determine the prices" of their coal (310 U.S. 150, 215)(2), nevertheless it felt that didn't violate the Sherman Act (310 U.S. 150, 215). Indeed, Appalachian Coals was part of the history that Socony summarized by saying that "for over forty years this Court has consistently and without deviation adhered to the principle that price-fixing agreements are unlawful per se under the Sherman Act and that no showing of so-called competitive abuses or evils which those agreements were designed to eliminate or alleviate may be interposed as a defense." (310 U.S. 150, 218, emphasis added). Thus, according to the Socony Court, both it and the Appalachian Coals court held all three of these positions (i.e., price-fixing is illegal per se, Appalachian Coals' 137 owners literally fixed prices, but that was not illegal per se). Taking all three of those positions seriously, that's just BMI: Socony and Appalachian Coals recognized that those 137 coal producers were fixing prices "in the literal sense", but concluded that wasn't the sort of price-fixing that is illegal per se.

The scholarly consensus is able to ignore this foreshadowing of BMI by taking two steps. First, Appalachian Coals is dismissed as being due not to facts or law but due to the shock of the Depression.(3) Second, the Socony Court's agreement with Appalachian Coals is dismissed as a failure to admit how terribly wrong Appalachian Coals was. While those steps have led to the scholarly consensus, in fact, as shown below, both of these Courts' claims about Appalachian Coals follow reasonably from the facts the district court found, so there is no excuse for not taking the Socony Court's claims seriously: it knew that Appalachian Coals intended to fix prices "in the literal sense", but it concluded that wasn't "price-fixing" in the antitrust sense. As shown below, both of those Courts focused their inquiry on the economic justification and effect of the practice before approving it. As BMI suggested, that inquiry into justification is what determines whether or not the behavior is covered by the per se rule (although as Socony says, once behavior is deemed to be covered by the per se rule, there is no [further] role for economic justification).(4)

Thus, this paper shows that the Appalachian Coals decision follows reasonably from the economic implications of the facts the district court found, so the Socony Court's agreement with Appalachian Coals is also reasonable, and the literalist (mis)reading of Socony flowed from the (mistaken) scholarly consensus about Appalachian Coals. Reading Socony as a whole, as this paper does, gives us exactly the view that is more commonly attributed to BMI: although Appalachian Coals clearly intended to fix prices "in the literal sense", the Socony Court understood that this was not price-fixing in the antitrust sense. Thus, properly understood, all three of these cases make the same point.

Which brings us back to Maricopa. As Justice Powell's dissent pointed out, in that case there was no agreement that literally restrained trade: while the agreements in Maricopa did control prices for any trade the parties chose to conduct under those agreements, the agreements left everyone free to also make whatever other arrangements they wanted to and to trade at any price they wanted to under any alternative agreements (457 U.S. 332, 360). Since the case came to the Court on a plaintiff's motion for summary judgment after only limited discovery, there is no basis to conclude that such agreements must necessarily restrain trade, nevertheless, the Maricopa Court felt that its literalist (mis)reading of Socony's per se rule constrained it to ban those agreements.(5) Thus, it remains of interest to see how Socony anticipated the per se rule that BMI merely repeated and why that rule should survive Maricopa's misreading of Socony, but to do that we first have to understand Appalachian Coals.(6)

In U.S. v. Appalachian Coals, Inc. (1 F.Supp. 339 [1932]) the district court held that it was illegal per se for an agent to sign 137 exclusive contracts for all the coal sold by mines that had produced 74% of the coal produced by non-captive mines in the "Appalachian territory" (which the district court defined to be 24 counties in Virginia, West Virginia, Kentucky, and Tennessee).(7) However, the Supreme Court dismissed the complaint against Appalachian Coals, overruling the district court. Since then, the scholarly consensus has been that Appalachian Coals was a temporary Depression era loss of faith in free markets in which the Court endorsed a price-fixing cartel, abandoning the per se rule against price-fixing until it came to its senses in Socony and effectively overturned its endorsement of price-fixing.

That consensus view also has to reject the Socony Court's claim that it distinguished between what it approved in Appalachian Coals and what is widely thought of as a virtually identical cartel that it disapproved in Socony.(8) Thus, the scholarly consensus is that Appalachian Coals was a mistake that the Court effectively overturned in Socony (even while saying it was just distinguishing between those cases). In fact, as shown below, Socony didn't overturn Appalachian Coals because Appalachian Coals wasn't a price-fixing case, but a rule of reason case about a firm that did not fix prices or intend to fix prices and had no market power (despite its high share of sales).

Despite all the courts have said on this case, this paper relies only on the facts that the Appalachian Coals district court found: we do not rely on any court's analysis of those facts (since those analyses are at best confusing and at worst inconsistent with the facts). Of course, the district court may have been wrong about its facts, but no higher court ever suggested that, so we have to assume that the Appalachian Coals and Socony Courts accepted those facts, and that those facts are therefore part of the context that might affect our understanding of what those Courts meant. Section II describes Appalachian Coals. Section III analyzes just the four dispositive facts (out of the 53 facts the district court found) and uses them to show first that the district court's legal analysis shows that the defendants were not engaged in per se illegal activity, and second that the defendants did not unreasonably restrain trade. Section IV takes the novel approach of considering the entire Socony decision seriously (instead of taking the traditional approach of ignoring that decision's comments on Appalachian Coals). That novel approach clarifies the consistent view of the per se rule against price-fixing that was given in Socony, Appalachian Coals, and BMI. Conclusions are in Section V.

II. A short summary of Appalachian Coals

The defendants were 137 small Appalachian coal producers that organized and owned Appalachian Coals and signed exclusive contracts allowing it to sell all their coal. As noted above, the predominant scholarly view is that the Supreme Court approved a price-fixing cartel in this case, but when it condemned price-fixing a few years later in Socony, it did not condemn Appalachian Coals (which scholars say approved price-fixing) because the Socony Court claimed that Appalachian Coals wasn't a price-fixing cartel.

Although there is much that is helpful in the court decisions, their insightful comments on Appalachian Coals are buried among comments that are unclear, irrelevant, or just wrong. This paper focuses on the facts that the district court found since the various court's analyses are either too cryptic (i.e., the Supreme Court) or too full of error (i.e., the district court). An analysis of four of those facts shows that the defendants did not violate the law. It should not be too surprising that the defendants did not violate the law given the consumer testimony: the only relevant comments that the Appalachian Coals district court's Findings of Fact ("Findings of Fact") has on "Opinion of Consumers" is that Appalachian Coals "will not restrain competition in the purchase of coal" and similar opinions.(9) What is surprising is that the evidence supporting the consumer testimony has been buried in these courts' decisions all these years, obscured by the lengthy documents that these courts (and generations of scholars) have produced. Ignoring all that, we now consider just the four dispositive facts that the district court found.

III. Four facts that show that the defendants didn't violate the law.

Although the district court's decision and 100 pages of findings contain much of interest, this section pulls out just the four dispositive facts. The first key fact is in Section III.A: the district court found that the agency was an attempt to realize economies and eliminate waste. That section goes on to show that proper consideration of this fact should have led the district court to a rule of reason analysis. The three key facts in section III.B show that the agency likely had no ability to raise prices (i.e., no market power).(10) Combining these four facts, we conclude that the district court should have conducted a rule of reason analysis and found that the defendants collectively had no market power and therefore didn't violate the law.

  1. Fact #1: the agency was an attempt to relieve the wasteful problem of distress coal

The first key fact is that the district court found that the agency would have been able to mitigate what the court found to be the wasteful practice of shipping unsold coal(11) and was an attempt to do that. As shown below, this fact will imply that the district court should have applied the rule of reason, but to see all that, we first need to understand the problem of distress coal.

  1. "Distress coal" in theory

Of course it is tempting to guess that "distress coal" is just what a cartel calls sales at less than the cartel price. Indeed, that's what "distress" gasoline seems to be in Socony (310 U.S. 150, 171). Posner and Easterbrook (1981) have a different guess about "distress coal", but they also suspect foul play. Their view seems to be (p.125) that "Since the distress coal was a by-product of making larger sizes, any reduction in the supply of distress coal entailed a reduction in the supply of all coal." Although Posner and Easterbrook recognize the importance of the heterogeneity of coal sizes, they seem to see "distress coal" as part of a standard cartel story: the cartel reduces sales of "distress coal" in order to reduce all sales and increase price. However, the Appalachian Coals district court made it clear that the issue of "distress coal" is much more complicated than that: while Appalachian Coals did want to reduce their sales of "distress coal", and while it would have been anti-competitive to form a joint venture to reduce such sales by reducing production, as the district court found that's not the way the defendants planned to reduce their sales of "distress coal."

  1. What the district court found about "distress coal"

The 11th of the district court's Findings of Fact describes the problem of "distress coal". The court found that coal could be produced only in a mixture of different lump sizes, but that mixture was commonly screened to produce carloads of uniform sized pieces of coal, which is what was most consumers wanted. Producers shipped out coal in sizes they had orders for, and the unordered sizes accumulated until producers shipped them unsold as "distress coal" which was then subject to storage fees unless they were sold quickly.

After calling the shipment of unsold coal "one of the worst practices in the coal industry" and "One of the chief problems of the industry", Findings of Fact #11 went on to say that "A regional sales agency, being in a position to obtain orders for a large volume of coal, would be able by the allocation of orders to mitigate this situation, and reduce the amount of distress coal on the market, because eventually the demand for the various sizes of coal will absorb the production." Thus, for a district court writing in the 1930s, Findings of Fact #11 shows a surprisingly good understanding of the Law of Large Numbers (the fact that in many circumstances, an average is more stable than the individual components it is composed of).(12) What the court found was that a joint sales agency can do better than 137 independent producers each getting a few orders for coals of particular sizes, producing coal (which necessarily comes in many coal sizes), and then urgently trying to sell pieces of coal in sizes that they didn't happen to have orders for. Instead, a joint sales agency would receive a large number of orders (so that orders for each coal size would be a relatively stable fraction of the total). From that large total, the agency can allocate each producer a relatively balanced demand for different coal sizes, thus reducing inventory costs, and shipments of unsold coal. The key point being the economy of scale: the district court found that the agency could reduce inventory costs.

The following example may clarify the pro-competitive effects of this economy of scale. We note that "The greater part of the demand is for particular sizes of coal such as nut and slack, stove coal, egg coal, and lump coal."(13) Therefore, suppose coal mines can produce only a mixture of coals in these 5 different sizes, and that mines screen their production to get 5 different piles of coal, each pile having lumps of only one size. While a joint sales agency might sell those 5 sizes of coals at prices p1, p2, ... p5 such that the market would clear if each firm produced at full capacity, 137 independent firms can't do that. An independent firm that gets an order for a particular size of coal can produce coal to fill that order, and then be stuck with an unsold inventory of coals of sizes that it has no orders for. With orders coming in over time, each of the 137 firms had to either pay to store their unsold inventory while waiting for orders for sizes they had in stock, or they could dump their excess on the market as distress coal. Since a single joint sales agency allows a reduction in inventories (and therefore inventory costs) due to economies of scale in jointly managing inventories (due to the Law of Large Numbers), that clearly allowed Appalachian Coals to reduce the amount of distress coal and its costs for any output level.(14) However, it is no "restraint of trade" for a joint venture to direct consumers to already existing inventories of the sizes of coals they want (increasing both consumer surplus and total quantity sold) even if that reduces the residual quantity of what would otherwise have been sold in the bargain basement at low prices to consumers who don't much value their purchases (at least on the margin).(15) Similarly, if a single firm invests in better inventory management tools so that it sells a higher fraction of its output at prices at or above its costs, that may hurt some consumers who would otherwise have bought the product at scrap prices, but there is no presumption that this restrains trade, and there wouldn't be such a presumption even if a group of firms jointly developed a better inventory management tool (although for a group we might have concerns about effects). Thus, the first key fact is that the district court found that the agency was an attempt to mitigate the wasteful practice of shipping unsold coal and could in fact have had that effect.

Of course, in theory, the problem of distress coal could have been solved without resort to a joint sales agent: independent producers could simply have swapped loads of different sizes of coals. However, there are clearly economies of scale in matching orders to inventories of different sizes of coals, and there would have been substantial transaction costs in bartering five different sizes of coals of different qualities among 137 independent producers. The relevance of those transactions costs is shown by the fact that the industry had been distressed about this problem for years, and apparently did not solve the problem by barter: "One of the chief problems of the industry is, therefore, the practice of a larger number of operators of producing different sizes of coal even though orders are on hand for only one size, and the necessity of marketing all sizes." (Findings of Fact #11).

  1. Appalachian Coals was a rule of reason case

Both under the current legal standard we now have and under the legal standard at the time of the decision, the district court should have applied the rule of reason. The current standard is simpler, so we begin with that.

"When 'persons who would otherwise be competitors pool their capital and share the risks of loss as well as the opportunities for profit . . . such joint ventures [are] regarded as a single firm competing with other sellers in the market.' Arizona v. Maricopa County Medical Soc., 457 U. S. 332, 356 (1982). As such, though Equilon's pricing policy may be price fixing in a literal sense, it is not price fixing in the antitrust sense." (Dagher, 547 U.S. ___ [2006]).

Since the district court found that all of the common stock of Appalachian Coals had been sold to the 137 coal producers who had organized it (Findings of Fact #8), we would now consider Appalachian Coals to be a single firm and that while its "pricing policy may be price fixing in a literal sense, it is not price fixing in the antitrust sense." Thus, Appalachian Coals would now be analyzed under the rule of reason.

As to the state of the law at the time of the Appalachian Coals decisions, the district court responded to the defendants' argument that this was a rule of reason case by saying that it saw "a vital distinction in principle, however, between a bona fide corporate organization" to which the rule of reason applied "and an agreement eliminating competition between independent dealers". That distinction being that

"Corporate organization is ordinarily the product of natural economic forces ... internal economies and the elimination of duplication and waste ... competition either actual or potential can be depended upon to hold prices within a range that is fair ... Combinations of independent producers, on the other hand, organized to fix uniform prices for the sale of their products or to eliminate competition among themselves ... are artificial agreements ... Where the parties to such a combination control a substantial portion of the trade, however, the unified control arising from such combination will necessarily affect prices". (1 F.Supp. 339, 344).

Thus, the district court distinguished between combinations of firms driven by "natural economic forces" of "internal economies" that such combinations generate (such combinations being analyzed under the rule of reason even though the combined firm may charge only one uniform price), and "Combinations of independent producers ... organized to fix uniform prices" (which are illegal per se).(16)

Both a merged firm and a combination of independent firms may charge a uniform price, but the district court saw internal economies as the goal of a merger, the uniform price being just a side effect (which it would therefore analyze under the rule of reason). In contrast, the district court saw a combination of independent firms as forming just to collude on price, the price-fixing being naked (and thus illegal per se).(17) Thus, the district court's legal analysis should have led it to apply the rule of reason in Appalachian Coals because, as we saw above, the district court's Findings of Fact #11 shows that the district court found the agency to be an attempt to mitigate a wasteful practice, exactly the sort of "internal economies and the elimination of duplication and waste" that the district court's decision uses to justify judging mergers under a rule of reason analysis. By applying the district court's first key fact to its legal analysis, we conclude that the district court should have found that Appalachian Coals wasn't a per se case because the agency was "an attempt to ... relieve the deplorable conditions resulting from ... wasteful trade practices" (1 F.Supp. 339, 342) which they were trying to mitigate by generating "internal economies and the elimination of duplication and waste".

  1. Facts #2, #3, and #4 show the defendants likely had no market power

The second key fact concerns the firms that were in the relevant market. Although the courts considered statistics for many different regions (e.g., "Appalachian territory", east of the Mississippi, ...), the firms in the market would have to include at least the producers in "the Appalachian and surrounding territory".(18) The third key fact is that the defendants' coal production was barely half as large as the excess coal production capacity held by atomistic non-defendants in the relevant market.(19) The fourth key fact is that the industry's excess productive capacity was "one of the causes of its depressed condition."(20)

From just those three facts, we reach the following conclusions. Since the excess capacity was a cause of the industry's depressed condition, that capacity must have been readily available to the market at prevailing prices. Since even in the smallest possible geographic market the defendant's competitors held an amount of this readily available excess capacity that dwarfed what the defendants produced, the defendants had no market power: insofar as the defendants might have tried to raise price above competitive levels, their non-conspiring atomistic neighbors could have easily displaced the defendants.(21) Since the defendants had no power to raise price above competitive levels, they had no "power to fix prices" as the courts used that expression.(22)

  1. The district court was wrong in presuming market power

"While there is substantial competition to be met in each market, and defendants will not through their sales agency be able to exercise monopolistic control of the market, it is clear that the control in the selling agency of the vast quantity of coal which they sell will necessarily affect prices." (1 F.Supp. 339, 348)

In fact, the district court was right that allowing an agent to gain control over a vast quantity of coal can affect prices in some cases, but as shown above, that wasn't the case here. As the USDJ and FTC Horizontal Merger Guidelines put it (§2.22), a merged firm may find it profitable to raise price unilaterally, but that's unlikely if the merged firm's customers have economical alternative sources that can supply the same product at the old price. In fact, the district court found that there were such alternatives in this case given that the industry's excess productive capacity was depressing its price and given the abundant excess capacity held in Appalachia by atomistic competitors of Appalachian Coals. Thus, the district court had no basis for presuming that the agency had market power.(23)

IV. Socony: the Supreme Court's look back at Appalachian Coals

The only time that the Supreme Court has looked extensively back at its Appalachian Coals decision was seven years later when the Supreme Court tried to distinguish the facts in Appalachian Coals from those in Socony. In Socony, the Court accepted (without much further explanation) what it had earlier said in Appalachian Coals, that "the plan did not either contemplate or involve 'the fixing of market prices' " (310 U.S. 150, 215). As explained above, the district court in Appalachian Coals should have reached that conclusion because it found that the agency was formed to produce efficiencies and because its legal analysis concluded that behavior aimed at producing efficiencies was judged under the rule of reason. However, the traditional reading of Socony (which ignores Socony's approval of Appalachian Coals) rejects Socony's view that 137 producers that set their prices jointly weren't price-fixing, relying on quotes like the one near the beginning of this paper: "Whatever economic justification particular price-fixing agreements may be thought to have, the law does not permit an inquiry into their reasonableness. They are all banned" (310 U.S. 150, 224).

The flaw in that traditional reading of Socony is that while Socony clearly forbids an inquiry into the reasonableness of a price-fixing agreement, Socony does not clearly forbid the inquiry into the reasonableness of an agreement that has not yet been condemned as a price-fixing agreement (even if it literally fixes prices), and Socony approves of Appalachian Coals (where the reasonableness of the agreement kept it form being condemned as price-fixing). Thus, the Court had before it enough information to conclude that the agreement in Appalachian Coals was reasonably aimed at producing efficiencies, and it had before it the district court's legal analysis where such efficiencies were the key difference between conduct that is held to be illegal per se and conduct that is analyzed under the rule of reason. Since Socony and Appalachian Coals make sense if and only if we assume that they understood all this and since the Court presented the key points(24), it seems reasonable to credit the Court with understanding all this.(25)

Indeed, we should note a key sentence in Socony that makes no sense at all unless as an implicit reference to the point that Section III of this paper makes explicitly: in distinguishing Appalachian Coals, Socony approvingly quoted that earlier decision: "The intelligent conduct of commerce through the acquisition of full information of all relevant facts may properly be sought by the co-operation of those engaged in trade, although stabilization of trade and more reasonable prices may be the result." (310 U.S. 150, 215 quoting 288 U.S. 344, 374). As Section III above explained, the economy of scale that the Appalachian Coals district court pointed out came from managing 137 inventories as a single inventory. Such management doesn't require physically holding all the coal in a single place, it's sufficient for the agency to acquire "full information of all relevant facts" so that a single manager knows what's in all the inventories and use that information in the "intelligent conduct of commerce". Thus, this key sentence from Socony makes perfect sense as a description of Section III above, but it makes no sense at all if one views Appalachian Coals as just a cartel that colludes on price.

One might object that this approach anachronistically reads BMI's ideas into a much earlier decision, but the alternative is ignoring the fact that Socony explicitly endorsed what was literally price-fixing by Appalachian Coals. If we are to take Socony seriously, as a whole, and in the absence of an alternative explanation for why they felt that Appalachian Coals wasn't a price-fixing case, it seems reasonable to credit the Socony Court with this reading as well.

The Socony Court also said that Appalachian Coals had no power or intent to fix prices (310 U.S. 150, 216), just as the Appalachian Coals Court did (288 U.S. 344, 375), but again it failed to explain how it knew that. Section III.B above has explained that they had no power to fix prices, we now show that there was no intent.

  1. No intent to fix prices

We know two things about the intentions of the Appalachian Coals defendants. First, we know that the district court found that the defendants intended to avoid unduly restraining competition. The district court's Findings of Fact #21 says that in setting up the agency the producers considered allowing it to control as much as 90% of the commercial tonnage of the territory, but rejected that, setting the maximum that the agency could control at 80%. Findings of Fact #21 goes on to say that "The maximum of 80% was adopted because a majority of the producers felt that an organization with a greater degree of control might unduly restrain competition in local markets." Thus, the court found an intent to avoid this conduct "because ... the producers felt that [it] ... might unduly restrain competition". A more skeptical court might have said "because ... the producers felt that [it] ... might appear to unduly restrain competition". However, the court did not find that this decision was made for the sake of appearances; the court found that the decision was made out of a genuine concern of the defendants that competition not be unduly restrained.

The second thing we know about intent is that while the defendants intended to fix the prices they charged "in the literal sense", in fact they had no power to fix prices in the antitrust sense (as shown in Section III.B. above). Having no power to fix price and having the intent that their efforts at eliminating waste not unduly restrain competition, it seems reasonable to conclude that they did not intend to try to do what they had no power to do (i.e., that they had no intent to fix prices).

  1. What Socony added

The Court added two new views in Socony. First, instead of just saying that the Appalachian Coals defendants intended to do many good things, in Socony the Supreme Court said that "The occasion for the formation of the agency was the existence of certain so-called injurious practices and conditions in the industry." (310 U.S. 150, 214). Of these "injurious practices", the one Socony singled out was "the problem of 'distress coal' ". Thus, the procompetitive justifications for Appalachian Coals were claimed to be more causal in Socony's look back than they had been in the original Appalachian Coals decision itself. Second, in distinguishing the plan in Socony from the one in Appalachian Coals, the Court said that "Unlike the plan in the instant case, the plan in the Appalachian Coals case was not designed to operate vis-à-vis the general consuming market and to fix the prices on that market. Furthermore, the effect, if any, of that plan on prices was not only wholly incidental but also highly conjectural." (310 U.S. 150, 216).

Combining its two new views of the Appalachian Coals case, the Socony Court said that "the occasion for the formation" of the Appalachian Coals agency was the existence of market failures, and that any effect of the agency on price was "not only wholly incidental but also highly conjectural." Arguably, such a conclusion could be based on Section III.B above (showing that the defendants had no market power) and the district court's conclusion that Appalachian Coals was "an attempt to organize the coal industry and to relieve the deplorable conditions resulting from overexpansion, destructive competition, wasteful trade practices, and the inroads of competing industries." (1 F.Supp. 339, 341 also 288 U.S. 344, 359). Thus, as in the Appalachian Coals case, the Socony Court seems to have described the facts well, although (apparently) not persuasively.

Thus, the Socony Court made two points. First, that price-fixing agreements are illegal per se regardless of whether they are reasonable or not (310 U.S. 150, 224). And second, (even though Appalachian Coals was clearly planning to fix prices "in the literal sense") that Appalachian Coals was not engaging in price-fixing because "the occasion for the formation" of Appalachian Coals was the existence of market failures that it meant to address and the effect of their plan on prices was "wholly incidental" to that legitimate purpose. This non-literal approach to price-fixing, focusing on effects and purposes is exactly the approach taken in BMI (441 U.S. 1, 9). We also note that the district court's "vital distinction in principle" (between per se illegal naked price-fixing and the rule of reason for ventures that eliminate waste) would argue for this same rule of reason approach to an attempt to ameliorate a wasteful practice that might also have an incidental effect on prices.

V. Conclusion

According to Socony, price-fixing agreements are unlawful per se regardless of any justification (310 U.S. 150, 218). However, if we read Socony as a whole (including its views on Appalachian Coals), we wind up with BMI's view of the per se rule. Thus, Socony approved of Appalachian Coals even though Appalachian Coals was an agreement to fix prices (in the literal sense) for the 137 separate companies that formed it: "The plan contemplates that prices are to be fixed by the officers of the Company at its central office" (288 U.S. 344, 358). Ignoring what Socony says about Appalachian Coals has led to an unreasonable reading of Socony's per se rule. However, (as shown above) Appalachian Coals was not price-fixing (in the antitrust sense) because (as the district court found) the agency was an attempt to relieve the wasteful problem of distress coal and (as the district court concluded) the rule of reason applies to agreements that produce economies and eliminate waste. Applying the rule of reason to an agency structured to avoid undue restraints on competition and producing only a fraction of what its atomistic competitors held as excess production capacity (which was so readily available at prevailing prices that it was "one of the causes of [the industry's] depressed condition."), it's clear that the Appalachian Coals defendants had no power to raise price above competitive levels, and therefore had no "power to fix prices" as the Court had earlier defined that expression (see footnote 22).

Thus, contrary to the scholarly consensus, Appalachian Coals was a case about firms with "no intent or power to fix prices" working together to reduce waste and the Socony Court reasonably followed earlier precedent in refusing to call that "price-fixing". Insofar as the Socony Court was following that earlier precedent (or taking BMI's approach of focusing "on whether the effect and ... the purpose of the practice are to threaten the proper operation of our predominantly free-market economy") "reasonableness" had a big role to play in the decision on whether or not to label behavior as "price-fixing" even though if the Court found that it was price-fixing, from then on "Whatever economic justification particular price-fixing agreements may be thought to have, the law does not permit an inquiry into their reasonableness." (310 U.S. 150, 224).

Unlike the consistent approach to the per se rule taken by Appalachian Coals, Socony, and BMI, Maricopa County seems to be based on the popular misreading of Socony. Although the NCAA Court accepted a less literal approach two years after Maricopa County, it may have accepted that non-literal approach only for the most extreme circumstances. Nevertheless, that may be a harbinger of a return some day to the reasonable approach to the per se price-fixing rule that the Court took for almost 50 years, from Appalachian Coals to BMI.


Appalachian Coals, Inc. v. U.S., Findings of Fact, U.S. District Court for the Western District of Virginia, October 3, 1932.

Bauer, Joseph P. and William H. Page, Kintner Federal Antitrust Law, Volume II, Practices Prohibited by the Sherman Act: Anderson Publishing Co., Cincinnati, 2002.

Coase, Ronald H., "The Nature of the Firm." Economica, 4 n.s. (1937), 386-405.

Kovacic, William E. and Carl Shapiro, "Antitrust Policy", Journal of Economic Perspectives, (Winter 2000), pp .43-60.

Posner, Richard A. and Frank H. Easterbrook, Antitrust Cases, Economic Notes and Other Materials, 2nd edition: West Publishing, 1981.

U.S. Department of Commerce, Fifteenth Census of the United States, Mines and Quarries: 1929, U.S. Government Printing Office, Washington, 1933.

U.S. Department of Justice and Federal Trade Commission, Horizontal Merger Guidelines, April 1997.

Viscusi, W. Kip, Joseph E. Harrington, Jr., John M. Vernon, Economics of regulation and antitrust, 4th edition, Cambridge, Mass. : MIT Press, 2005.

Williamson, Dean V. "Organization, Control and the Single Entity Defense in Antitrust." [forthcoming]

Wood, Diane P., "U.S. Antitrust Laws in a Global Context." Colum. Bus. L. Rev. (2004), 265-281.


*Economic Analysis Group, Antitrust Division, U.S. Department of Justice (e-mail: This paper does not purport to reflect the views of the Department of Justice. Craig Conrath, Russ Pittman, Alex Raskovich, and Greg Werden provided helpful comments on an earlier draft, but any remaining errors are the author's.

1. As Justice Powell's dissent explained, the Maricopa Court didn't clearly distinguish its facts from BMI (457 U.S. 332, 365). Thus, the extent to which Maricopa limited BMI isn't clear.

2. Appalachian Coals was more to the point: "The plan contemplates that prices are to be fixed by the officers of the Company at its central office" (288 U.S. 344, 358 (emphasis added)).

3. Viscusi et al (2005) say Appalachian Coals was "clearly in opposition to the per se rule" and "explained perhaps by the 'deplorable' conditions of the industry". Judge Wood (2004) saw it as endorsing an exception to the per se rule for industries in distress. Bauer and Page (2002) say that language the Court paraphrased from Appalachian Coals "appeared to establish that the presence of bad market conditions for sellers, such as overproduction and consequent low prices, could justify cooperative actions." FTC Commissioner Kovacic and ex-Deputy AAG Shapiro (2000) saw it as "a Depression-era aberration" in which the Court "lost faith in free market competition". Judges Posner and Easterbrook (1981) saw it as being due to the Depression shaking the Court's "Faith in the policy of competition".

4. The word "further" is in brackets to emphasize that Socony doesn't explicitly say "further", although (as shown below) that's the only way to read Socony as a whole consistently.

5. However, the 'illegal per se' approach does not apply to "an industry in which horizontal restraints on competition are essential if the product is to be available at all." (NCAA v. Board of Regents, 468 U.S. 85, 100-101 [1984]).

6. Of course, other decisions also view price-fixing literally (e.g., Citizen Publishing Co. v. U.S. (394 U.S. 131) although one can hope that case won't survive the recent Dagher decision), but Maricopa relied on its misreading of Socony. This paper simply shows the only way to read Socony consistently as a whole, in hopes that those that want to rely on Socony will rely on what it actually said (which, one might hope, would lead us back to BMI's approach to price-fixing).

7. That territory is about 300 miles long (from Marion County TN to Kanawha County WV) and about 100 miles wide. The district court did not define a relevant market, possibly because it saw this as a case of per se illegal price-fixing. Instead, the district court just reported what shares the defendants had in four different universes (1 F.Supp. 339, 339-340). Although the district court's decision didn't discuss choosing among these universes, it seemed to favor the 74% quoted above.

8. In dissent in Socony (310 U.S. 150, 261-262), Justices Roberts and McReynolds saw no difference between the price-fixing cartel in Socony and the plan proposed by Appalachian Coals.

9. Customer views on competition were discussed only in Findings of Fact #49 (which was just summarized). Although Findings of Fact #28 lists 10 customers that testified for the U.S. on relevant product market, nothing in the Findings of Fact or the district court's decision suggests that those customers had views on whether Appalachian Coals would be good or bad for anyone. Of course, consumers can all be wrong (e.g., U.S. v. Ivaco, 704 F. Supp. 1409), but it's still worth noting the unanimous customer opinion supporting this paper's conclusions.

10. This is stronger than the Court's view that bad price effects had not been shown (288 U.S. 344, 373). In fact, the district court's findings imply that bad price effects were unlikely.

11. The agency was "an attempt ... to relieve the deplorable conditions resulting from ... wasteful trade practices" (1 F.Supp. 339, 341) and shipments of unsold coal (a.k.a. "distress coal") was "one of the worst practices in the coal industry." (Findings of Fact #11).

12. The Supreme Court has mentioned the Law of Large Numbers only once (German Alliance Insurance Co. v. Ike Lewis, 233 U.S. 389, 394 [1914]). The first district court case to mention this Law is Holt v. Richardson (240 F. Supp. 724, 728 [1965]).

13. 288 U.S. 344, 362 and also (but at greater length) district court Findings of Fact #11.

14. This economy of scale doesn't require physically combining the inventories, it's enough for the agency to know about each firm's inventories so that when it receives an order for a coal size that some firms have in inventory, it can allocate sales to such firms rather than to firms that would have to mine more coal to be able to fill the order. That knowledge is the key here may explain why both Courts emphasized that: "The intelligent conduct of commerce through the acquisition of full information of all relevant facts may properly be sought by the co-operation of those engaged in trade, although stabilization of trade and more reasonable prices may be the result." (Appalachian Coals 288 U.S. 344, 374; similarly in Socony 310 U.S. 150, 215)

15. As a more recent example, some products that used to be sold for scrap are now redirected through the internet to purchasers that pay more than scrap value and presumably value those goods even more. The price increase for such products that are now sold by eBay and others (instead of going for scrap at lower prices) is a good thing for society.

16. From an economic standpoint, the district court's "vital distinction" is not at all clear. For example, suppose a firm is just a bundle of contracts with factors of production where the factor agrees that in return for some payment, it will "obey the directions of an entrepreneur within certain limits." (Coase, 1937 p.391). Since Appalachian Coals signed contracts with coal producers requiring them to obey its directions in shipping the coal they produced, Appalachian Coals fits Coase's definition of a firm as well as any other corporation does. For a modern view on the distinction that the district court saw as "vital", see Williamson (forthcoming).

17. Although the above quote from the district court inappropriately included market power in a discussion of the per se standard, we note that its market power concern was unjustified: as Section III.B will show, Appalachian Coals didn't have any market power.

18. "The coal produced in the surrounding territory is the same kind of coal as that produced in the Appalachian territory and is suitable for the same purposes and available to the same markets, generally on the same freight rates, and for all practical purposes might have been included in the territory described as Appalachian territory." (Findings of Fact #29, also quoted in 288 U.S. 344, 357; emphasis added). That finding also noted that captive mines sell commercially only what their owners can't absorb (i.e., they sold commercially during the Depression, but "with a return to normal business conditions their output will not be a material factor in the commercial market."). That such expansion of the captive mine owners' businesses would absorb their excess capacity shows how limited their excess capacity was (indeed less than 10% of the mines were captive (Findings of Fact #51)). That and footnote 19 will show their irrelevance.

19. Non-defendant excess capacity is just total capacity minus defendants' capacity minus non-defendant production. In million tons of coal, total capacity was 245.2 of which the defendants had 86.6 (Findings of Fact #47a) for "southern West Virginia, Virginia, eastern Kentucky, and Tennessee" (whose coal is just in the "Appalachian territory" and its immediate surroundings (pp. 259-61, U.S. Department of Commerce, 1933)). The district court's figures implied that (in million tons of coal) non-defendants produced 49 (i.e., total coal production in and immediately surrounding Appalachia was 107 of which the defendants produced 58, 1 F.Supp. 339, 339). Thus, non-defendants held excess capacity (in million tons) of 245.2 86.6 49 = 109.6, which is almost twice as much as the 58 that the defendants produced.

20. Instead of the language quoted above (from the district court's Findings of Fact #9c), the Supreme Court quoted Findings of Fact #47a: " 'This excess capacity over actual production,' the court said, 'could be brought into production at moderate expense and with reasonable promptness.' " (288 U.S. 344, 369). While that language doesn't say what prices would be needed to get firms to make that "moderate expense", as shown below, the language in the text above shows that that capacity was readily available at prevailing prices.

21. In the Appalachian territory there were 130 non-defendant commercial coal producers (Findings of Fact #51; 288 U.S. 344, 371) and also 20 captive mines (Findings of Fact #51).

22. The claim in the text is just the contrapositive of (and therefore logically equivalent to) the following: "The power to fix prices, whether reasonably exercised or not, involves power to control the market and to fix arbitrary and unreasonable prices." (U.S. v. Trenton Potteries (273 U.S. 392, 397 [1927]).

23. The 53rd and last of the district court's Findings of Fact lists five "General Findings of Fact" that it concluded from the 52 foregoing facts. The fourth of those "General Findings of Fact" is "That the effect of the plan of defendants will be to eliminate free competition among a large group of producers of coal ... and that, because they control so substantial a part of the coal sold in the United States, this elimination of competition and concerted action will ... have a tendency to ... raise prices" (emphasis added). Since the district court found no evidence of the defendants conspiring with anyone outside of Appalachia (Findings of Fact #24) and since the previous subsection showed that the facts the district court found imply that there wouldn't be such price effects in any market, it was ridiculous for the court to suggest a national market. Since this General Findings of Fact is a conclusion about economics (which seems beyond the district court's competence) this paper concludes that finding was erroneous.

24. While this paper relies on Finding of Fact #11 (which the Court did not specifically mention), the Court did point to the district court's similar (albeit vaguer) statement that "it is but due to defendants to say that the evidence in the case clearly shows that they have been acting ... in an attempt to ... relieve the deplorable conditions resulting from ... wasteful trade practices" (288 U.S. 344, 359 and also 1 F.Supp. 339, 341).

25. Since Section III.A.3 above showed that Dagher supports the outcome in Appalachian Coals, one might suggest that as an alternative explanation of the reasoning in Appalachian Coals (i.e., one might say that Appalachian Coals just said that it's not per se illegal for a joint venture to set its price). However, that alternative explanation is untenable since the Court did not consider Appalachian Coals and its 137 owners as a single unified entity (288 U.S. 344, 376).

Updated June 25, 2015