Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The growing nostalgia for past regulatory misadventures and the risk of repeating these mistakes with Big Tech

The growing nostalgia for past regulatory misadventures and the risk of repeating these mistakes... Abstract Digital markets are increasingly described as the ‘railroads’ of the 21st century. Extending that metaphor, some commentators argue we should revive stale railroad-era economic regulations and adapt them to the digital age. This enthusiasm appears to be buoyed by both a sudden nostalgia for railroad and airline regulations once administered by the Interstate Commerce Commission (ICC) and the Civil Aeronautics Board (CAB) and an equally sudden amnesia of the enormous harm those regulations caused to consumers. ICC and CAB regulations are indeed an apt metaphor, as they illustrate perfectly how sectoral regulations sold to the public as simple, clear, and cheap can go awry. Ultimately, a bipartisan consensus emerged to disband those agencies and deregulate those industries. After deregulation, prices fell, output expanded, and firms innovated. Proposals to regulate Big Tech today in a similar fashion forget these important lessons. We should know better than to do the same thing again today and expect a different result. I. INTRODUCTION Heavy sectoral regulation is back in vogue. In the USA, Senator Elizabeth Warren argued last year for ‘break[ing] up Big Tech’ by emulating ‘Progressive Era’ regulations that ‘required a structural separation between the network and other businesses, and also demanded that the network offer fair and non-discriminatory service’.1 In Europe, the European Commission recently implemented the General Data Protection Regulation (GDPR)2 and continues to consider additional digital regulations, including competition rules presuming certain business practices or transactions involving digital platforms to be unlawful unless the platforms can prove otherwise.3 Australia, Japan, and the UK are studying heightened regulations for digital markets,4 and the Competition Commission of India recently heard (and rejected) a significant case involving two digital platforms.5 In the USA, this discussion has induced comparisons to—and an odd nostalgia for—the comprehensive regulations that once governed transportation. In 2017, one commentator argued Amazon was ‘a 21st-century version of the 19th-century railroads’ and compared Amazon’s practices to those of ‘the railroads of yore’.6 Her then-employer, a left-leaning Washington think tank, has adopted this thread as its own, proclaiming online platforms to be ‘the railroad monopolies of the 21st century’.7 And two opinion articles in the Financial Times argued that ‘the internet is the railroad of our times’8 and that the ‘obvious’ solution to today’s problems is to recycle yesteryear’s railroad regulations and adapt them to 21st-century technologies.9 Many of these proposals implicitly or explicitly favour imitating regulations once administered by the US Interstate Commerce Commission (ICC) and the Civil Aeronautics Board (CAB). Congress created the ICC in 1887 with a mandate to regulate railroads;10 at its peak, its jurisdiction also included barges, bus lines, trucks, and pipelines. The ICC also briefly regulated airlines,11 although it lost this authority in 1938 when Congress created a new regulator that soon became the CAB.12 According to the Financial Times columnist, the formation of the ICC ‘ushered in a period of prosperity’,13 and its purported success means ‘[i]t is time to think hard about whether we need another ICC – an Internet Commerce Commission – to do the same’ today.14 Others similarly extoll the virtues of the old ICC and its focus upon structural separation and ‘fair and non-discriminatory’ access.15 The sudden enthusiasm for the old ICC and CAB—which Congress disbanded in 1996 and 1985, respectively—stands in stark contrast to the epitaphs these agencies earned from those who actually knew them. The CAB was deeply unpopular in the years before it was disbanded. For example, a Senate subcommittee lead by Sen. Ted Kennedy in 1975 found that ‘present CAB practice is fundamentally deficient’ and ‘Board regulation has not effectively brought about the low-fare service that is technically feasible and that consumers desire’.16 In 1978, CAB Chairman Alfred Kahn famously testified that he often wondered whether deciding thousands of picayune decisions—from whether an airline may buy a new plane to whether a supplemental (charter) airline may transport a flock of sheep to England—was ‘what my mother raised me to do’.17 Two successive Presidential administrations—one Republican, the other Democratic—sought and eventually succeeded in disbanding the CAB.18 The ICC fared no better. In the late 1960s, a study prepared for the Brooking Institution conservatively estimated that ICC regulations cost consumers at least $500 million per year (in 1960s dollars) and that ending ICC regulation ‘would create benefits far in excess of costs’.19 In 1975, President Ford described ICC (and CAB) regulations as something ‘begun as a protection for consumers’ that ‘now guarantees that in many cases they will pay higher prices than a free market would call for’.20 In 1980, President Carter enthusiastically signed the Staggers Rail Act, remarking that it ‘will benefit’ both shippers and consumers ‘[b]y stripping away needless and costly regulation’ by the ICC.21 And in 1995, upon signing the bill disbanding that agency, President Clinton remarked that he was ‘disappointed’ by a bill that only disbanded the ICC because the bill ‘falls short of my administration’s much bolder proposal for extensive deregulation of transportation industries’.22 With the passage of time, fewer and fewer of today’s policymakers were personally involved in those events. As a new generation takes their place, ideas that were tried and failed are once again gaining purchase, albeit now as a solution to perceived problems with Big Tech. Because ‘[t]hose who cannot remember the past are condemned to repeat it’,23 we should remember the reality: railroad-style regulations distorted competition in the marketplace, reduced economic efficiency, hindered innovation, and harmed the very consumers they ostensibly protected.24 More specifically, the American experience with the ICC and CAB illustrates three important lessons. First, provisions that sound simple in theory, like a ‘non-discrimination’ requirement, seldom prove simple in practice. Secondly, effective regulation requires clear instructions, particularly regarding the ‘what’, ‘why’, and ‘how’. This kind of clarity is needed not just at the outset, but also over time, as changed circumstances and mission creep may later confuse matters and lead to perverse and unintended results. Thirdly and finally, given the substantial consumer benefits that flowed from the elimination of ICC and CAB regulations, regulations—however well-intentioned—can impose very real costs on consumers. Or, in three words, simplicity, clarity, and costs. Many of today’s proposals to regulate Big Tech forget these important lessons. Proposed digital regulations are sold to the public as simple, obvious, and beneficial. Behind the curtain, however, there is little clarity on what the problem is, let alone why a given proposal is the best solution or how the agency would operationalize it. Nor is there any recognition that burdensome new regulations may distort markets and harm consumers. In short, given the tarnished history of the ICC and CAB, we should know better than to do the same thing again today and expect a different result. II. THE AMERICAN EXPERIENCE WITH THE ICC AND CAB Like today’s calls for heavier regulation of digital markets, both the ICC and the CAB were created to regulate perceived problems with dynamic industries. Both were given mandates to regulate their respective industries ‘in the public interest’, a broad, vague, and malleable standard. Presented with an amorphous problem and vague instructions, both agencies constructed complex and increasingly anti-competitive regulations. A broad political consensus eventually emerged to dismantle the regulations and disband the regulators, leading to widespread consumer benefits. This section traces the evolution of each agency. The ICC In the late 19th century, the Age of Steam, the railroad was considered both cutting edge technology and big business. In the USA, ribbons of steel opened the American West to development, with farms, mines, towns, and factories developing in their wake. With new development came new economic patterns. What we might now call geographic markets expanded, as products could be shipped in from far away. For example, wheat grown in North Dakota was shipped by rail to Minneapolis, where it was milled into flour, and shipped by rail again to bakeries in Chicago, St. Louis, and even New York City. Given the important role that railroads played in connecting buyers to sellers, they naturally elicited complaints from all sides. Farmers complained that they often had only one choice, whichever railroad ran closest to their property, on which to ship their produce, thereby forcing them to pay supra-competitive rates. Many merchants made similar complaints. Both groups called for new regulation at both the state and federal level to ensure that this service—which they deemed essential—was offered at fair rates on non-discriminatory terms.25 The railroads, too, were dissatisfied. When more than one railroad served a given route, as occurred in the important grain routes from Chicago to ports on the East Coast, the railroads themselves complained of ‘ruinous competition’.26 Railroads often formed ‘pools’—which today we would simply call cartels—to allocate traffic and prop up prices. And like today’s cartels, each firm had an incentive to cheat by giving powerful customers—like Standard Oil—secret rebates to secure additional traffic.27 Eventually, both camps struck upon the idea of regulation to cure all that ailed them. Customers envisioned comprehensive state and federal regulations that would set the terms of service and prohibit price discrimination. The railroads envisioned targeted regulations that would use the coercive power of the state to end secret rebating and thereby ensure the stability of their cartels.28 So in 1887 the US Congress passed the Act to Regulate Commerce, better known today as the Interstate Commerce Act.29 To those of you following today’s digital markets proposals, its provisions may sound eerily familiar. Section 1 of the Act required railroads to charge rates that were ‘reasonable and just’,30 whereas sections 2, 3, and 4 prohibited ‘unjust discrimination’ in the setting of rates or terms of service.31 Using language that echoes the Sherman Act, which was enacted three years later, section 5 banned ‘any contract, agreement, or combination … for the pooling of freight’.32 The Act also created an independent agency, the ICC, to enforce the Act.33 Yet public optimism soon evaporated; new regulation was not the magic elixir all had hoped. The answer, obviously, was more regulation. So in 1903 Congress passed the Elkins Act, which banned railroad rebates to large industrial customers but not small ones, and therefore supposedly levelled the playing field.34 This step, too, was insufficient, so in 1906 Congress passed the Hepburn Act,35 which strengthened the ICC’s rate-setting powers and banned vertical integration. In 1910 Congress granted the ICC the authority to suspend railroad rates during an investigation,36 and in 1920 Congress added the authority to set minimum rates for the explicit purpose of protecting financially weaker railroads from stronger ones.37 The 1920 Act also allowed firms to pool traffic—in other words, to form a cartel—if the agency determined that service would be improved and competition not ‘unduly’ restrained.38 And when new forms of transportation began to compete with the railroads, Congress granted the ICC the authority to regulate most of them, as well.39 Over time, the ICC acquired jurisdiction over many other forms of transportation, from trucking and barge traffic to natural gas pipelines, leading to conflicting priorities. Airlines and international shipping eventually escaped, regulated instead by their own specialist agencies. Three anecdotes illustrate the scope of the ICC’s regulatory misadventures. First, as previously described, the 1906 amendments prohibited railroads from vertically integrating. Populists at that time were concerned that railroads—which were the only economical way to ship coal—were vertically integrating into the coal mining business, thereby hurting less efficient independent producers. (Indeed, proponents of new regulations today have returned to this example.40) So Congress considered legislation, which eventually would become the Hepburn Act, to bar railroads from vertically integrating into other industries (a practice that was already commonplace, particularly for timber and coal). The prohibition in the draft bill was soon broadened over the objection of some of its supporters,41 although the Senate also rejected an even broader amendment that would have banned common carriers from producing coal and other products for their own use.42 In what became known as the Commodities Clause, this additional provision made it a criminal offense for a railroad to carry any product for sale43—other than timber, whose lobbyists got a special exemption—that it either had produced or in which it had an economic interest.44 All sides soon realized how economically damaging such a ban would be. In early litigation involving this provision, the Supreme Court adopted a narrow interpretation that rendered it inapplicable in most cases.45 The railroads also learned to adapt their corporate form in ways that sidestepped the Commodities Clause.46 In a subsequent case against US Steel, the provision was further cabined, in essence applying only to the coal industry.47 And in those rare instances when the US government did assert the clause, it was typically subsidiary to a Sherman Act claim, thereby suggesting that the antitrust laws were sufficient to correct any abuses.48 Congress apparently agreed, time and again rejecting legislative attempts to override the narrow interpretation adopted by the Supreme Court.49 Secondly, in the 1950s, the Southern Railroad developed a new kind of railcar for grain that was more than twice as efficient as earlier methods.50 When the railroad sought to lower rates by 60 per cent to reflect these new efficiencies, the ICC denied the request pending its investigation. It ultimately rejected the railroad’s proposed rate on the grounds that the rate was so low that it would harm shipments by barge, instead authorizing a 53.5 per cent reduction.51 The case—generally known as the ‘Big John’ matter—went to the Supreme Court, which sided with the railroad, but only after customers had been forced to pay the higher ICC-imposed rate for four years.52 In addition, scholars have estimated that the Southern Railroad incurred between $20 million and $40 million in expenses seeking to overturn the ICC’s initial order,53 an amount that the National Academy of Sciences found ‘significantly undermines such incentives to innovate’.54 Thirdly, in 1965, two railroads lowered their rate for shipping steel between Pittsburgh and a downstream factory in Kentucky to match the competing rate charged by barges and trucks. In a proceeding called the Ingot Molds decision, the ICC disallowed the rate as too low to both cover its costs and produce what the ICC believed to be a ‘fair’ profit.55 The Supreme Court affirmed the ICC’s Order.56 Remarkably, it also commended the ICC for ‘properly’ avoiding ‘a rate war’ because the competing barge-truck rate was not ‘in need of being driven down by competitive pressure’.57 Decrying these and other obvious mistakes, in 1971, a professor at the Massachusetts Institute of Technology aptly summarized the situation in an op-ed in The New York Times: ‘The heavy hand of regulation, which [Congress] has laid on the industry through the Interstate Commerce Commission, has stifled innovation, discouraged industry, [and] driven many of the capable and imaginative men out of it.’58 That year, 30 US senators called for the abolition of the ICC.59 Soon after, politicians of all stripes concluded that deregulation was necessary. The rail and trucking industries were significantly deregulated starting in the late 1970s, and the ICC was disbanded entirely at the end of 1995 by a Democratic President and a Republican Congress.60 Freed from restrictive price and service rules, railroads optimized their networks. They pared unproductive routes and reduced labour costs, increasing efficiency markedly on their remaining routes.61 Rail rates eventually fell system-wide,62 and declines were particularly steep for bulk commodities like grain, lumber, and coal that could be moved much more effectively using methods that ICC regulations had heretofore discouraged.63 Trucking rates also fell precipitously, at least in those areas where state-level regulations did not survive. For example, a 1989 study by the US Federal Trade Commission found that trucking rates in deregulated markets were roughly 40 per cent lower than the rate on a regulated, but otherwise equivalent, route in Texas.64 Depending upon the type of trucking service, prices and costs fell between 35 and 75 per cent,65 yielding consumer benefits of ‘more than $18 billion’ in 1996 dollars.66 A 2010 joint study by the Federal Trade Commission and the Antitrust Division of the US Department of Justice concluded that ‘deregulation of the trucking industry … has been entirely beneficial for consumers’.67 Milton and Rose Friedman aptly summarized the arc of the agency in their 1980 book Free to Choose: The ICC illustrates what might be called the natural history of government intervention. A real or fancied evil leads to demands to do something about it. A political coalition forms consisting of sincere, high-minded reformers and equally sincere interested parties. The incompatible objectives of the members of the coalition (e.g., low prices to consumers and high prices to producers) are glossed over by fine rhetoric about “the public interest”, “fair competition”, and the like. The coalition succeeds in getting Congress (or a state legislature) to pass a law… . Success [by interested parties] breeds its problems, which are met by broadening the scope of the intervention… . In the end the effects are precisely the opposite of the objectives of the reformers and generally do not even achieve the objectives of the special interests. Yet the activity is so firmly established and so many vested interests are connected with it that repeal of the initial legislation is nearly inconceivable. Instead, new government legislation is called for to cope with the problems produced by the earlier legislation and a new cycle begins. The ICC reveals clearly each of these steps … .68 Or, as the Editorial Board of The New York Times put it in 1996, the ICC was ‘[i]nitially created to protect farmers and communities from the monopoly power of railroad barons’, but ‘later ganged up on the consumers it was supposed to protect, siding instead with industries—especially trucking—that it was supposed to regulate.’69 The CAB The story of airline regulation in the USA is much the same. Airline regulation began in 1926 with the formation of the Aeronautics Branch of the Commerce Department, which focused principally on air safety.70 After briefing transferring some functions to the ICC,71 in 1938, Congress created a special regulator just for airlines.72 Thus, the CAB was born. Like the ICC, the CAB was directed to place other ‘public interest’ values ahead of competition, which was allowable ‘to the extent necessary to assure the sound development of an air-transportation system properly adapted to the needs of the foreign and domestic commerce of the United States’.73 Given this directive, once one airline served a given route, like Chicago to Memphis, the CAB typically refused to allow any other carriers to fly the same route on the grounds that the competition was not necessary to develop that route.74 Between 1950 and 1974, the Board received 94 applications for new airlines to provide ‘trunk-line’ air service between major US cities—and denied every last one of them.75 Although the CAB occasionally allowed incumbents to add new routes, it did so only reluctantly; remarkably, it took Continental Airlines more than eight years to receive approval to fly a new route between Denver and San Diego,76 and only then because the D.C. Circuit vacated the Board’s denial of the application and directed the Board to approve a carrier.77 When multiple airlines were allowed to compete on the same route, the CAB rarely allowed the carriers to cut their prices and the carriers rarely asked.78 These high prices depressed demand, with airlines often flying planes that were less than 60 per cent full.79 Indeed, the CAB set pricing on the assumption that the average plane would fly at only 55 per cent of capacity.80 To fill those empty seats, carriers increasingly competed on service. Air carriers that could not compete on price instead competed by offering carved chateaubriand on rolling silver carts, piano lounges on the upper decks of Boeing 747s, or free alcoholic drinks.81 Yet the CAB increasingly intervened here as well to avoid any changes that might permit ‘unfair’ competition on quality. At the top end, competitor complaints caused the CAB to deny Trans World Airlines’ request to add lie-flat sleeper seats to its trans-Atlantic first class service.82 On the bottom end, the agency initially protected first-class flights by permitting coach-class flights only at off-peak times.83 The airlines’ industry group also got into the act, making industry-wide agreements that, as economist Alfred Kahn described them, ‘prescribe[ed] the maximum allowable knee-room …, dictat[ed] that meals be limited to sandwiches … and requir[ed] a uniform supplementary charge for in-flight motion pictures’.84 As with the ICC, the inefficiencies built until the public could bear them no longer. In the early 1970s independent economists found that fares on intrastate carriers not regulated by the CAB were substantially lower than the fare on an interstate carrier regulated by the CAB—sometimes even for the exact same route.85 In 1974, the CAB ‘first encouraged experiments with discount fares’ that it heretofore had prohibited.86 Senator Edward Kennedy and then-Senate counsel Stephen Breyer began a series of hearings in 1975 questioning the efficacy of CAB regulations.87 That same year, the ‘Pulsifer’ CAB staff study found that ‘the present regulatory regime has produced undesirable economic effects, is not justified by the structural economic characteristics of the industry …, and seems likely to produce even more objectionable results in future years’.88 Because ‘[t]hese undesirable effects outweigh the benefits of such regulation’, and because these inefficiencies were the product ‘of the law itself’, rather than how the CAB administered it, the CAB staff recommended repealing the statutes authorizing these regulations.89 Chairman John E. Robson surprised observers in 1976 by announcing that the CAB Commissioners unanimously agreed that Congress should reduce the CAB’s authority and loosen regulations.90 In 1977, President Carter appointed Alfred Kahn, a leading critic of the CAB, as its Chairman. Kahn pushed for radical deregulation, and policymakers agreed. With the passage of the Airline Deregulation Act of 1978,91 the CAB began the process of disbanding, closing at the beginning of 1985.92 The liberalization of airline markets quickly reduced airfares and increased output. By 1990, 12 years after deregulation, economists estimated that airline deregulation was already increasing US consumer surplus by more than $6 billion per year, or $20 billion annually in 2019 dollars.93 In the 1990s and 2000s, the Government Accountability Office (GAO) similarly found large airfare reductions, which ‘resulted in large part from increased competition’,94 and in particular a significant increase in the number of airlines that serve each airport.95 And the benefits of airline deregulation continue to grow. Stephen Breyer, by now a Justice of the US Supreme Court, noted in 2011 that ‘the number of airline passengers increas[ed] from 207.5 million in 1974 to 721.1 million’ in 2010.96 During that same time, ‘[a]irline revenue per passenger mile … declined from an inflation-adjusted 33.3 cents … to 13 cents’, and the cost of a flight from New York to Los Angeles fell, again in inflation-adjusted terms, from $1442 to $268.97 Given the overwhelming evidence that airline deregulation benefitted consumers, almost all economists, lawyers, and policymakers view the dissolution of comprehensive airline regulation as the right decision. For example, one leading scholar noted that ‘[t]he simplest prediction of economists about airline deregulation, and one of the few on which nearly all economists agreed, was that deregulation would improve consumer welfare in comparison to continued price and entry regulation’ and, ‘[f]ourteen years later, nearly all economists still agree on this.’98 Similarly, the then-General Accounting Office reported to Congress in 1999 that ‘[o]ver the years, our work has consistently shown that airline deregulation has led to lower fares and better service for most air travelers’ and that these benefits are ‘largely due to increased competition spurred by the entry of new airlines into the industry and established airlines into new markets’.99 The sceptics occasionally point to two purported exceptions,100 both of which actually reinforce the consensus that airline deregulation benefitted consumers. First, former CAB lawyer Paul Stephen Dempsey published a paper in 1990 arguing that airline prices would have fallen even faster under regulation than under deregulation.101 Setting aside the methodological weaknesses in his analysis,102 Dempsey subsequently testified before Congress that he does not support the imposition of new regulations.103 He testified that ‘[w]herever possible competition is preferable to Government as regulator of the market’ and that he is a ‘proponent[] of competition, … not regulation’.104 He also proposed limited new legislation premised upon the belief ‘that the competition unleashed by airline deregulation has been beneficial to large segments of the consuming public, but that competition should be further advanced so that a larger universe of Americans can enjoy the benefits of airline deregulation’.105 Ironically, some critics now cite his 1990 study as ‘Exhibit A’ for the purported failure of airline deregulation106 and forget his subsequent sworn testimony to the contrary. Secondly, former CAB economist David B. Richards wrote an article in the mid-2000s arguing that ‘the grant of pricing freedom to the airline industry has generally resulted in average prices being higher than they would have been had regulation continued’.107 He recognizes that this conclusion is inconsistent with ‘most other studies’.108 Yet he fails to mention that his conclusion is also inconsistent with some of his own data, which he says elsewhere show ‘considerable passenger savings’ in the early 2000s.109 Indeed, on a weighted average basis, he found that the average US airfare in 2005 was more than 15 per cent lower than it would have been if the 1979 version of the fare regulations (the so-called ‘SIFL’) had remained in effect.110 This estimate is likely conservative, as other studies point out that fares had already been partially deregulated by 1979,111 so comparisons using that year as the base may understate the overall effect of dismantling the CAB regime.112 Moreover, Richards’ data (if not his conclusion) are consistent with several other studies the US GAO relied upon when it concluded that ‘reregulating the airline industry would likely reverse consumer benefits’.113 These studies estimate that airfares are approximately 20–30 per cent lower than they would be if the 1979 regulations were still in effect.114 III. LESSONS LEARNED Given this history, Americans have learned—the hard way—three lessons from earlier efforts to require ‘just’, ‘reasonable’, and ‘non-discriminatory’ pricing and service in network industries. First, no matter how simple the regulatory principle, implementing the regulations day-to-day is almost always devilishly complex. In the ICC context, the framers of the original 1887 Act thought their commandment of ‘just’ and non-discriminatory rates would be straightforward. Yet policymakers felt it necessary to clarify and augment the Commission’s authority in major new statutes in 1903, 1906, 1910, and 1920.115 Similarly, the CAB found that regulating entry alone was insufficient, leading it to regulate prices and (with amusing but likely inefficient results) service levels.116 Secondly, effective regulation requires a clear ‘what’, ‘why’, and ‘how’. More specifically, we must first ask what problem regulation is necessary to solve. Once a problem is clearly identified, we must then ask why the proposed regulation is the best option. And once those items are clear, we must finally ask ourselves how the new agency may achieve this solution and, at least as importantly, how it may not. Or, to put it another way, how we design institutions and what we ask them to do matters a great deal.117 For example, the ban on vertical integration embodied in the Commodities Clause was not the best way to solve perceived abuses in the coal industry, as most agree that antitrust enforcement forced the industry to change its practices. Similarly, even if there had been some legitimate need for ICC regulation, it should not have been permissible for the agency to use rate-setting policies to protect less efficient producers—like the barges in Ingot Molds—from more efficient producers. As the ICC’s expanding jurisdiction illustrates, mission creep—often driven by technological change—only exacerbates these problems. And thirdly, as our experience after deregulation shows, even the most well-intentioned regulations come at a steep cost. Removing ICC and CAB regulations significantly reduced consumer prices and increased output, generating billions of dollars in consumer surplus. Many other students of regulation have distilled similar principles that recognize the importance of simplicity, clarity, and alternatives to regulation. For example, future Justice Stephen Breyer distilled an exhaustive study of regulatory failings in the 1970s into three ‘rules of thumb’.118 ‘First and most obviously, modesty … in one’s approach to regulation’,119 by which he means that regulation should be reserved only for ‘serious defects for which regulation offers a cure’ and that we must consider the potential difficulties of ‘regulatory alternatives’ to ‘an unregulated status quo’.120 Secondly, ‘regulators ought to aim at worst cases and, in attacking such cases, they should strive for simplicity’.121 And thirdly, given ‘[t]he problems accompanying classical regulation’, it ‘ought to be looked upon as a weapon of last resort’.122 DG-Comp official Cyril Ritter embeds similar concepts in his proposed five-part test for deciding whether regulation is well-suited to address a problem, like what he calls ‘digital matters’.123 IV. APPLICATION TO TODAY’S PROPOSALS Reading the headlines today, some appear to have forgotten these valuable and hard-won lessons. This amnesia is particularly alarming given recent calls to replicate our experience with the ICC and CAB by imposing similar regulations upon Big Tech. Many of these proposals forget the first lesson, selling railroad-style regulations as simple and beneficial. For example, Lina Khan argues that ‘the best way to preserve fair and open competition’ is ‘simply to completely ban any network monopolist from owning businesses that place it in competition with the companies that depend on it to reach [the] market’, which ‘is what previous generations did with railways’.124 Similarly, based on what she believes to be the success of previous railroad regulations, Senator Elizabeth Warren would like to impose similar ‘structural separation’ and demand ‘that the network offer fair and non-discriminatory service’.125 To be clear, this enthusiasm is not universally held; for example, the UK’s Furman Report notes that many have called for regulating digital markets ‘in the same way as electricity, gas, or railway networks’, but found that it was ‘too early to conclude’ that such regulation would be necessary or appropriate.126 Other proposals forget the second lesson, arguing for the re-imposition of regulations like the ICC’s old Commodities Clause without considering whether it is the best option available. For example, an opinion piece published in the Financial Times last week traced the early history of the Commodities Clause and argued that ‘[i]t seems obvious … that we should apply these same standards to the digital giants of today.’127 Lina Khan similarly touts the potential of imposing a regime like the Commodities Clause on technology platforms,128 although she at least acknowledges doing so may ‘sacrifice certain cost savings, resulting in higher prices’.129 Khan’s honesty—that imposing heavy new regulations on Big Tech may result in higher prices—is refreshing but rare. Indeed, most have also forgotten the third lesson. For example, in 2017, a fellow Open Markets Institute commentator argued that we should ‘stop Amazon from selling groceries’—which sounds suspiciously like the kind of entry restrictions the CAB imposed on airlines—because banning Amazon from the market would somehow increase competition.130 The same commentator also argued the Federal Trade Commission must ‘ban Amazon from engaging in any price discrimination in food products, anywhere, ever’,131 another failed approach the CAB employed. And the Open Markets Institute argues that CAB regulation was preferable to today’s marketplace, because ‘regulation by the CAB prevented airlines from abusing their market power while also ensuring that citizens in cities of comparable size received roughly equal service, in terms of both quality and price’.132 That claim is strictly true, at least in the sense that a much smaller number of customers paid the same much higher price and ate chateaubriand. Given the Institute’s populist instincts, this claim is also ironic, as it would return us to an age when flying was only for the very wealthy.133 V. CONCLUSIONS In conclusion, those who cannot remember history are condemned to repeat it. More than 100 years ago, Congress decided the platform industry of the day—railroads—must charge rates that are ‘reasonable and just’ and eschew ‘unjust discrimination’.134 These apparently simple requirements turned out to be highly complex, prompting the growth of a regulatory leviathan that banned price and non-price competition alike on the grounds that competition might harm smaller, less efficient firms. The credo also spread to airlines. Ultimately, academics and policymakers found that those regulations, however well-intentioned, were costing consumers tens of billions of dollars a year. Broad political support quickly emerged in favour of deregulation. Today, many of those lessons have been forgotten. Some now believe—apparently sincerely—that railroad and airline regulations were simple and beneficial, and therefore a great model for how to regulate Big Tech. Taking it a step further, some now argue that imposing railroad-style regulations upon Big Tech firms is simply ‘obvious’.135 And perhaps worst of all, some now argue that banning firms from the market and imposing strict pricing rules will surely increase competition and reduce prices.136 The USA spent more than 100 years trying, and failing, to make these ideas work as intended. Many have now forgotten that history and seem intent on repeating it. Acknowledgement This article is based on a speech presented by Commissioner Wilson on 28 June 2019 at the British Institute for International and Comparative Law and a related speech presented by Commissioner Wilson on 30 June 2019 at The Antitrust Enforcement Symposium, University of Oxford. The views expressed herein are solely those of the authors and do not necessarily reflect the views of the US Federal Trade Commission or any other Commissioner. While in private practice, Commissioner Wilson represented airlines in various regulatory and competition matters around the world. The authors thank Neil Chilson, Brian Keating, Jeremy Sandford, D. Daniel Sokol, and Joshua Wright for helpful comments. Of course, all errors are our own. Footnotes 1 Elizabeth Warren, ‘Here’s How We Can Break Up Big Tech’ (Medium, 8 March 2019) <https://medium.com/@teamwarren/heres-how-we-can-break-up-big-tech-9ad9e0da324c> accessed 15 October 2019. 2 EU GDPR: Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation) [2016] OJ L119/1. 3 Jacques Crémer and others, European Commission, Directorate-General for Competition, ‘Competition Policy for the Digital Era’ (2019) 4 <http://ec.europa.eu/competition/publications/reports/kd0419345enn.pdf> accessed 15 October 2019 (arguing, among other things, ‘‘[t]he specific characteristics of many digital markets have arguably changed the balance of error cost and implementation costs, such that … one may want to err on the side of disallowing potentially anticompetitive conducts, and impose on the incumbent the burden of proof for showing the pro-competitiveness of its conduct’’). 4 See eg Jason Furman and others, ‘Unlocking Digital Competition: Report of the Digital Competition Expert Panel’ (March 2019) 2 <https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/785547/unlocking_digital_competition_furman_review_web.pdf> accessed 15 October 2019 [hereinafter Furman Report] (‘Instead of just relying on traditional competition tools, the UK should take a forward-looking approach that creates and enforces a clear set of rules to limit anti-competitive actions by the most significant digital platforms while also reducing structural barriers that currently hinder effective competition.’); Australian Competition & Consumer Commission, ‘Digital Platforms Inquiry’ (July 2019) 3 <https://www.accc.gov.au/system/files/Digital%20platforms%20inquiry%20-%20final%20report.pdf> accessed 15 October 2019 (‘The ACCC considers that now is the time to consider the current and likely future issues associated with digital platforms and their business models and to put in place frameworks that enable adverse consequences to be addressed and that reduce the likelihood of new issues arising.’); Japan Fair Trade Commission, ‘Study Group on Improvement of Trading Environment surrounding Digital Platforms’ (12 December 2018) Interim Discussion Paper 7 <https://www.jftc.go.jp/en/policy_enforcement/survey/index_files/190220.2.pdf> accessed 15 October 2019 (‘In order to facilitate a sound development of digital platform businesses in our country, it would be necessary to consider whether we should review each industry-specific regulation.’). 5 Competition Commission of India, Case No 20 of 2018, Order under s 26(2) of the Competition Act, 2002 (11 June 2018) <https://www.cci.gov.in/sites/default/files/20-of-2018.pdf> accessed 15 October 2019. 6 Lina M Khan, Op-Ed, ‘Amazon Bites Off Even More Monopoly Power’ The New York Times (New York, 21 June 2017) A23 <https://www.nytimes.com/2017/06/21/opinion/amazon-whole-foods-jeff-bezos.html> accessed 15 October 2019 (‘Think of Amazon as a 21st-century version of the 19th-century railroads that connected consumers and producers … . In several key ways, Amazon uses its power as the railroads did.… And like the railroads of yore, Amazon dictates terms and prices to those dependent on its rails.’). 7 Open Markets Institute, ‘Free Press & Platform Monopolies’ <https://openmarketsinstitute.org/issues/free-press-platform-monopolies> accessed 15 October 2019. 8 Rana Foroohar, ‘Big Tech is America’s New ‘Railroad Problem’’ Financial Times (London, 16 June 2019) <https://www.ft.com/content/ec3cbe78-8dc7-11e9-a1c1-51bf8f989972> accessed 15 October 2019 [hereinafter, Railroad Problem]; see also Rana Foroohar, ‘Why We Need to Regulate the Tech Platforms’ Financial Times (London, 5 November 2017) <https://www.ft.com/content/84f402ac-bfc0-11e7-b8a3-38a6e068f464> accessed 15 October 2019 [hereinafter, Why We Need to Regulate]. 9 Railroad Problem, ibid (describing the ICC’s so-called Commodities Clause and arguing ‘[i]t seems obvious to me that we should apply these same standards to the digital giants of today’). 10 Act to Regulate Commerce of 1887 [Interstate Commerce Act], Pub L 49-104 (1887) 24 Stat 379 (1887). 11 See Stephen G Breyer, Regulation and Its Reform (Harvard UP 1982) 29 (describing the ICC’s short-lived authority to administer the air mail subsidy system in the 1930s). 12 See eg Edward M Kennedy, ‘Airline Regulation by the Civil Aeronautics Board (1975) 41 J Air L & Com 607, 609 (reprinting the Summary of Report of the Senate Subcommittee on Administrative Practice and Procedure) (describing the divestment of the ICC’s air mail authority in 1938). 13 Why We Need to Regulate (n 8). 14 ibid. 15 See eg William A Nitze, ‘The Tech Giants Must Be Stopped’ The American Conservative (Washington, DC, 16 April 2018) <https://www.theamericanconservative.com/articles/the-tech-giants-must-be-stopped> accessed 15 October 2019 (‘Measures that might help to achieve this result include requiring the large technology companies to deal with their customers or competitors on the basis of greater equality just as AT&T did with new telecommunications companies after its court-ordered break-up, or oil refiners did with railroads after the creation of the Interstate Commerce Commission.’); Lina Khan, ‘The Separation of Platforms and Commerce’ (2019) 119 Colum L Rev 973, 980 (‘This Article argues that these combined problems of discrimination and information appropriation invite recovering common carriage’s forgotten cousin: structural separations.... Structural prohibitions … have been applied as a standard regulatory tool and key antitrust remedy in network industries, often to prohibit a dominant intermediary from competing with the businesses that depend on it to get to market’.). 16 Kennedy (n 12) 614, 633 (report of the subcommittee); see also ibid 607 (forward by Sen. Kennedy noting he ‘fully subscribe[s] to the views expressed in this report’). 17 See Alfred E Kahn, ‘A Grateful Response and Supplement’ (Silicon Flatirons, University of Colorado, 5 September 2008) <https://siliconflatirons.org/documents/conferences/2008.09.05-207/KahnResponse.pdf> accessed 15 October 2019; see also Alan Greenspan, The Age of Turbulence – Adventures in a New World (1st edn, Penguin 2007) 71–72 (cited approvingly by Kahn); Susan E Dudley, ‘Alfred Kahn 1917-2010: Remembering the Father of Airline Deregulation’ (Spring 2011) Regulation 8, 10 (recounting the same sheep example). 18 See David Burnham, ‘Ford to Seek a Reduction of C.A.B. Regulatory Role’ The New York Times (New York, 7 February 1975) 61 <https://www.nytimes.com/1975/02/07/archives/ford-to-seek-a-reduction-of-cab-regulatory-role.html> accessed 15 October 2019 (describing a Ford Administration plan to introduce deregulatory legislation); Ernest Holsendolph, ‘Backers of Airline Deregulation Gird for Expected House Battle’ The New York Times (New York, 23 April 1978) A16 <https://www.nytimes.com/1978/04/23/archives/backers-of-airline-deregulation-gird-for-expected-house-battle.html> accessed 15 October 2019 (describing similar efforts by the Carter Administration to pass the Airline Deregulation Act of 1978). 19 Ann F Friedlaender, ‘The Dilemma of Freight Transport Regulation’ (background paper prepared at the request of the Brookings Institution for a conference held in 1967, Brookings Institution, Washington DC 1969, reprinted 1975) 164–65. 20 David Burnham, ‘Issue and Debate: Regulatory Agencies and Competition’ The New York Times (New York, 20 September 1975) 37 <https://www.nytimes.com/1975/09/20/archives/issue-and-debate-regulatory-agencies-and-competition.html> accessed 15 October 2019. 21 Statement of Pres Carter on Signing S 1946 Into Law (14 October 1980) <https://www.presidency.ucsb.edu/documents/staggers-rail-act-1980-statement-signing-s-1946-into-law> accessed 15 October 2019. 22 ‘Clinton Signs ICC Bill’ (United Press, 30 December 1995) <https://www.upi.com/Archives/1995/12/30/Clinton-signs-ICC-bill/8245820299600/> accessed 15 October 2019. 23 George Santayana, The Life of Reason: Reason in Common Sense (Dover 1980) vol 1 (1905) 284. 24 Thomas K McCraw, Prophets of Regulation: Charles Francis Adams; Louis D. Brandeis; James M. Landis; Alfred E. Kahn (Belknap Press of Harvard UP 1984) (reviewing the deregulatory work of Kahn). 25 See Richard D Stone, The Interstate Commerce Commission and the Railroad Industry: A History of Regulatory Policy (Praeger 1991) 5 (summarizing the underlying literature). 26 See ibid 5–6. 27 See Standard Oil Co of New Jersey v United States (1911) 221 US 1, 32–33, 42–43. 28 See Stone (n 25) 6 (identifying James J Hill, the famed railroad ‘Empire Builder’, as a particularly hopeful proponent of the ICC). 29 Pub L 49-104 (1887) 24 Stat 379. 30 ibid, § 1 (‘All charges made for any service rendered or to be rendered in the transportation of passengers or property as aforesaid, or in connection therewith, or for the receiving, delivering, storage, or handling of such property, shall be reasonable and just; and every unjust and unreasonable charge for such service is prohibited and declared to be unlawful.’). 31 ibid, § 2 (banning ‘unjust discrimination’ among passengers or freight transported contemporaneously ‘under substantially similar circumstances and conditions’); ibid, § 3 (banning ‘undue or unreasonable preference or advantage to any particular person’ and requiring common carriers to interchange traffic); ibid, § 4 (banning carriers from charging a higher rate, ‘under substantially similar circumstances and conditions, for a shorter than for a longer distance over the same line, in the same direction, the shorter being included within the longer distance’). 32 ibid, § 5 (‘That it shall be unlawful for any common carrier subject to the provisions of this act to enter into any contract, agreement, or combination with any other common carrier or carriers for the pooling of freights of different and competing railroads, or to divide between them the aggregate or net proceeds of the earnings of such railroads, or any portion thereof; and in any case an agreement for the pooling of freights as aforesaid, each day of its continuance shall be deemed a separate offense.’). 33 Some scholars believe incumbents favoured the creation of the ICC because they believed it would allow them to engage in rent-seeking. See eg Thomas W Gilligan, William Marshall and Barry Weingast, ‘Regulation and the Theory of Legislative Choice: The Interstate Commerce Act of 1887’ (1989) 32 J L & Econ 35. 34 57 Cong Ch 708 (1903) 32 Stat 847. 35 59 Cong Ch 3591 (1906) 34 Stat 584. 36 61 Cong Ch 309 (1910) 36 Stat 539 (Mann-Elkins Act). 37 66 Cong Ch 91 (1920) 41 Stat. 456 (Transportation Act of 1920). 38 ibid, § 5. 39 See eg Motor Carrier Act, 74 Cong Ch 498 (1935) 49 Stat 543. 40 See eg David Dayen, ‘How to Think about Breaking Up Big Tech’ (The Intercept, 1 April 2019) <https://theintercept.com/2019/04/01/elizabeth-warren-tech-regulation-2020/> accessed 15 October 2019 (explaining that Sen. Warren’s proposal to break up Big Tech in general, and Amazon in particular, ‘is merely following a long history’ that began with the Commodities Clause in the Hepburn Act). 41 See LC Marshall, ‘The Commodities Clause’ (1909) 17 J Pol Econ 448, 449–51; Thomas LeDuc, ‘Carriers, Courts, and the Commodities Clause’ (1965) 39 Bus Hist Rev 57 (describing Sen. Tillman’s role in shepherding the bill through the Senate and his failed efforts to ‘tighten’ the prohibition). 42 Marshall (n 41) 451 n10 (describing the Culberson substitute, which was rejected by the Senate committee of the whole by a vote of 62 to 11). 43 That is, railroads were allowed to continue producing and transporting coal they intended to consume themselves. 44 Much has been written about the so-called Commodities Clause. See eg Marshall (n 41) (summarizing the adoption of the clause and early Supreme Court precedent and questioning in conclusion ‘[w]hether that clause, as popularly interpreted, employed a wise method’); Eliot Jones, ‘The Commodity Clause Legislation and the Anthracite Railroads’ (1913) 27 QJ Econ 579 (decrying early Supreme Court precedents narrowly interpreting the Act but nonetheless concluding that effective enforcement of the Clause against anthracite railroads would not restore competitive conditions, but instead would induce the newly independent miners to organize ‘a coal trust, or at least some form of agreement among the coal companies to restrict output or to fix prices’, which would ultimately lead to additional regulation up to and including nationalization). 45 See eg Marshall (n 41); LeDuc (n 41) (tracing the history of the clause, noting that for its first thirty years ‘the government invoked the commodities clause of the Hepburn Act only against anthracite carriers’, and that the failed attempt to expand its scope in the New Deal-era U.S. Steel case effectively spelled its end). 46 See Edwin C Goddard, ‘The Commodities Clause of the Hepburn Act’ (1915) 14 Mich L Rev 49, 51 (note and comment) (noting that ‘[a]fter each of these decisions the coal-carrying roads were busily engaged in rebuilding their corporation plant so as to make it conform to the requirements laid down by the court’ and describing various partial spin-offs and leasing maneuvers). 47 See United States v Elgin, Joliet & Eastern Ry Co (1936) 298 US 492; see also ibid 512 (Stone, J, dissenting) (observing that ‘[i]f the commodities clause permits control such as is exhibited here, one is at a loss to say what scope remains for the operation of the statute’); United States v South Buffalo Ry Co (1948) 333 US 711; LeDuc (n 41) 72 (recounting these cases as the only times the commodities clause was asserted against an industrial firm that had vertically integrated by purchasing a common carrier). 48 See LeDuc (n 41) 59–60 (‘Doubt may be expressed whether the drastic prohibition embodied in the commodities clause was needed to protect either the consumer or the remaining independent producers. Existing law, had it been enforced, afforded adequate remedies…. [E]vents had already shown that the Sherman Act could be effectively invoked against the railroads, and later suits would show that it provided the chief weapon in dealing with the anthracite carriers.… As will later be apparent, the courts in deciding the anthracite cases relied more on the antitrust laws than on the Hepburn Act.’); Eugene V Rostow and Arthur S Sachs, ‘Entry into the Oil Refining Business: Vertical Integration Re-Examined’ (1952) 61 Yale LJ 857, 875 (arguing ‘the primary basis for the Court’s determination [in the first Anthracite case] was its construction of the Sherman Act, without reference to the policy of the commodities clause’). But see Robert Bork, ‘Vertical Integration and the Sherman Act: The Legal History of an Economic Misconception’ (1954) 22 U Chi L Rev 157, 168–69 (arguing that in the last two Anthracite cases, ‘[t]he evasions of the law were not evasions of the Sherman Act but of Pennsylvania law and the ‘Commodities Clause’ of the Interstate Commerce Act’). 49 See Jones (n 44) 600, 605 (describing early efforts to amend the Commodities Clause after Supreme Court decisions). 50 See eg JY Smith, ‘Ex-Southern Railway Head D.W. Brosnan Dies’ The Washington Post (Washington, DC, 19 June 1985) <https://www.washingtonpost.com/archive/local/1985/06/19/ex-southern-railway-head-dw-brosnan-dies/f88678b9-ef01-487b-8ae8-965f3a491364/> accessed 15 October 2019 (noting that the ‘Big John’ cars developed for Brosnan were ‘capable of carrying more than 100 tons, twice the capacity of earlier grain cars’). 51 Grain in Multiple-Car Shipments—River Crossings to the South (1963) 318 ICC 641, 683; Grain in Multiple-Car Shipments—River Crossings to the South (1965) 325 ICC 752, 758–59, 770–76; ‘Big John: ICC Cuts Against the Grain’ Railway Age (22 July 1963) 32; Alfred E Kahn, The Economics of Regulation: Principles and Institutions (MIT Press 1970; reprinted 1988) vol 1, 165 n12 [hereinafter, Kahn vol 1]; Alfred E Kahn, The Economics of Regulation: Principles and Institutions (MIT Press 1971; reprinted 1988) vol 2, 23 [hereinafter, Kahn vol 2]. 52 See Arrow Transp Co v Southern Ry Co (1963) 372 US 658, 659–62 (summarizing the four-year history). 53 See National Academy of Sciences, Innovation in Transportation: Proceedings of a Workshop (1980) 116–17 (‘While there has never been an explicit accounting from the Southern, it is estimated that the cost related to the workings of the regulatory scheme were in the range of $20 to $40 million in the case of Big John’.) 54 ibid 117. 55 Ingot Molds, Pa to Steelton, Ky (1965) 326 ICC 77, 82, 85; see also Kahn vol 1 (n 51) 165 n12; Kahn vol 2 (n 51) 24. 56 American Commercial Lines, Inc v Louisville & Nashville R Co (1968) 392 US 571. 57 ibid 593–94. 58 A Scheffer Lang, ‘Point of View: For Congress and the Railroads, a Last Chance’ The New York Times (New York, 10 January 1971) NER10 <https://www.nytimes.com/1971/01/10/archives/for-congress-and-the-railroads-a-last-chance.html> accessed 15 October 2019. 59 ibid. 60 See eg ‘The I.C.C. Dies’ The New York Times (New York, 3 January 1996) A14 <https://www.nytimes.com/1996/01/03/opinion/the-icc-dies.html> accessed 15 October 2019. 61 James M MacDonald and Linda C Cavalluzzo, ‘Railroad Deregulation: Pricing Reforms, Shipper Responses, and the Effects on Labor’ (1996) 50 ILR Rev 80, 90 (concluding that because of deregulation, ‘shippers switched to low-cost methods of transport, productivity measures grew sharply, shipping rates fell, and carrier profits grew’). 62 Wesley W Wilson, ‘Market-Specific Effects of Rail Deregulation’ (1994) 42 J Indus Econ 1, 20 (‘The evidence suggests that the majority of commodities prices initially rose under deregulation, reflecting greater market power and modest cost savings. By 1988, however, deregulation produced lower prices in most commodity classifications and did not increase prices in other classifications, suggesting that advances in productivity have dominated any adverse market power effects…. With price decreases and cost savings from deregulation, welfare gains from deregulation are likely positive.’). 63 See MacDonald and Cavalluzzo (n 61) 83 (‘Railroads moved aggressively in the post-Staggers [Act] era to offer volume discounts, and particularly large discounts to unit trains, to shippers of ‘bulk’ products, such as grain, lumber, coal, and other minerals. Because shippers must invest in specialized loading and storage facilities for unit trains, they will not do so without lower rates or improved services.’); Stone (n 25) 52 (‘Other ICC rate policies forced the railroads to postpone the use of unit trains (a whole trainload of cars permanently coupled and shuttling back and forth from producer to consumer – most frequently from coal mines to power plants) in the East for years.’). 64 See Submission of the United States to the Ibero-American Competition Forum (September 2007) 2–3 <https://www.ftc.gov/system/files/attachments/us-submissions-oecd-2010-present-other-international-competition-fora/ibero-trucking.pdf> accessed 15 October 2019 (citing Letter from Thomas Carter, Regional Director, Federal Trade Commission, to Raymond Bennett, Transportation/Gas Utilities Division, Railroad Commission of Texas (2 October 1989)). 65 ibid 4 (citing Clifford Winston, ‘U.S. Industry Adjustment to Economic Deregulation’ (1998) 12 J Econ Persp 89, 94). 66 ibid 5 (citing Stephen Morrison and Clifford Winston, ‘Regulatory Reform of U.S. Intercity Transportation’ in Gomez-Ibanez and others (eds), Essays in Transportation Economics and Policy (Brookings Institution Press 1999) 477). 67 ibid 8. 68 Milton Friedman and Rose Friedman, Free to Choose: A Personal Statement (Harcourt Brace 1980, reprinted 1990) 201. 69 ‘I.C.C. Dies’ (n 60). 70 See Fed Aviation Admin., United States Dep’t of Transportation, A Brief History of the FAA <https://www.faa.gov/about/history/brief_history/> accessed 15 October 2019 (tracing the history of US air regulation). 71 See above nn 11–12 and accompanying text. 72 The legislation initially created two separate regulators, the Civil Aeronautics Authority and the Air Safety Board. These agencies were reorganized in 1940, with many functions transferring to the newly-created CAB. See FAA (n 70). 73 Civil Aeronautics Act of 1938, formerly codified at 49 U.S.C. § 1302d (1964). 74 See Kennedy (n 12) 614–15 (describing how ‘entry into the industry has been effectively blocked’ by the CAB’s route entry policy). 75 See Civil Aeronautics Board Special Staff on Regulatory Reform, Regulatory Reform: Report of the C.A.B. Special Staff (July 1975) 49 n1 (on file with authors) [hereinafter Report of the CAB Special Staff] (‘According to information supplied by the Board to the Subcommittee on Administrative Practice and Procedure …, 94 applications for trunkline authority have been received from outsiders since 1950, and none has succeeded.’); Bradley Behrman, ‘Civil Aeronautics Board’ in James Q Wilson (ed), The Politics Of Regulation (James Q. Wilson ed., Basic Books 1980) 88 (‘Between 1950 and 1974, for example, the Board turned down all ninety-four of the applications that would-be entrants to the industry submitted in pursuit of domestic ‘trunk-line’ authority.’). 76 See John E Robson, ‘Airline Deregulation: Twenty Years of Success and Counting’ (Spring 1998) Regulation 17-8 (‘In a widely-cited example of the CAB at its worst, it took the board eight years to give Continental Airlines permission to fly between San Diego and Denver.’). 77 Continental Air Lines, Inc v Civil Aeronautics Board (DC Cir 1975) 519 F 2d 944, 959–60. 78 Kahn vol 2 (n 51) 210–11. 79 ibid 212 (citing Ronald E Miller, Domestic Airline Efficiency: An Application of Linear Programming (MIT Press 1963) 108–14). 80 Behrman (n 75) 98 (‘[I]n 1971 the CAB announced that all subsequent proposals for fare adjustments would be evaluated according to whether airlines would earn a reasonable return if they attained an average load factor of 55 percent.’ (citing CAB Order 71-4-54 (9 April 1971)); David B Richards, ‘Did Passenger Fare Savings Occur After Airline Deregulation?’ (2007) 46 J Transp Res Forum 73, 78 (describing the Domestic Passenger Fare Investigation pricing model developed by the CAB in 1974, and administered by Richards himself, which required ‘normal coach fares … to be offered on a formula rate’ based, among other things, ‘on a 55% full-fare load factor’ (emphasis omitted)). 81 Madhu Unnikrishnan, ‘A Law that Changed the Airline Industry Beyond Recognition (1978)’ Aviation Week (4 June 2015) <https://aviationweek.com/blog/law-changed-airline-industry-beyond-recognition-1978> accessed 15 October 2019 (chateaubriand and piano lounges); Behrman (n 75) 93 (alcoholic drinks); see also Kahn vol 1 (n 51) 211 (‘In part because the doors to price competition are closed, airline companies compete very strenuously among themselves in the quality of service they offer’.). 82 Trans World Airlines Siesta Sleeper-Seat Service (1958) 27 CAB 788, 790 (1958) (Opinion); see also Kahn vol 2 (n 51) 215 (discussing same). 83 Kahn vol 2 (n 51) 214 (citing Richard E Caves, Air Transport and Its Regulators (Harvard UP 1962)). 84 ibid 212. 85 See Behrman (n 75) 91–94 (citing William A Jordan, Airline Regulation in America: Effects and Imperfections (1970) 279–81, 285); Simat, Helliesen, and Eichner, Inc., An Analysis of the Intrastate Air Carrier Regulatory Forum (January 1976) vol 1, App Ex 1 (prepared for the US Department of Transportation)) (listing examples in California and Texas where PSA and Southwest, intrastate carriers not regulated by the CAB, competed on the same routes as CAB-regulated interstate carriers but charged much lower fares); see also Michael E Levine, ‘Is Regulation Necessary? California Air Transportation and National Regulatory Policy’ (1965) 74 Yale LJ 1416, 1430–39 (note) (providing examples from California, where intrastate carriers entered the Los Angeles–San Francisco route at prices well below the CAB-certified interstate carriers and where the unregulated price in 1965 ($11.43 one way) remained far below the CAB-set price on the otherwise comparable Boston-Washington DC route ($24.65 one way)). 86 Clifford Winston, ‘U.S. Industry Adjustment to Economic Deregulation’ (1998) 12 J Econ Persp 89, 89. 87 See above n 12 and accompanying text. 88 Report of the CAB Special Staff (n75) 252; see also ibid, Executive Summary 3, attached to Letter from Roy Pulsifer, Special Advisor to the CAB on Regulatory Reform, to the CAB, 22 July 1975 (on file with authors) (‘[I]t is possible that regulation will become more restrictive and thereby increase inefficiencies. Real improvement in economic efficiency can come only if protective regulation is eliminated or materially reduced.’). 89 ibid, Executive Summary 1–2. 90 Behrman (n 75) 104; Robson (n 76) 17–18 (‘The watershed reform came in April 1976 when the CAB unanimously announced its support for deregulation.’). 91 Pub L 95-504 (1978) 92 Stat 1705. 92 Irvin Molotsky, ‘C.A.B. Dies After 46 Years; Airlines Declared ‘on Own’’ The New York Times (New York, 1 January 1985) 9 <https://www.nytimes.com/1985/01/01/us/cab-dies-after-46-years-airlines-declared-on-own.html> accessed 15 October 2019 (describing the final closure of the CAB, an event that ‘was mandated in the airline deregulation enacted in 1978’). 93 Steven A Morrison and Clifford Winston, ‘The Dynamics of Airline Pricing and Competition’ (1990) 80 Am Econ Rev 389, 390 (‘On average, deregulated fares are lower than regulated fares by 18 percent, amounting to an average annual savings to travelers of roughly $6 billion (1988 dollars).’). 94 US General Accounting Office, Domestic Aviation: Changes in Airfare, Service, and Safety Since Airline Deregulation (25 April 1996, GAO/T-RCED-96-126) 4. 95 See ibid (‘The overall trend toward lower fares since deregulation has resulted in large part from increased competition, spurred in many cases by the entry of new airlines. The average number of large airlines serving the medium-sized-community airports in our sample, for example, increased by over 50 percent between 1978 and 1994, while the average number of commuter carriers serving these airports increased by about 40 percent.’); US Government Accountability Office, Airline Deregulation: Reregulating the Airline Industry Would Likely Reverse Consumer Benefits and Not Save Airline Pensions (June 2006, GAO-06-630) 4 [hereinafter, GAO Airline Deregulation] (‘As predicted by the framers of deregulation, airline markets have become more competitive and fares have fallen since deregulation. For consumers, airfares have fallen in real terms since 1980 while service has generally improved. Overall, median fares have declined in real terms by nearly 40 percent since 1980.’); US General Accounting Office, Airline Deregulation: Changes in Airfares, Service Quality, and Barriers to Entry (March 1999, GAO/RCED-99-92) 4 (‘Since deregulation, numerous new airlines have started operations, while established airlines have expanded into new markets.… In 1990, we reported that from 1979—the earliest year for which reliable data on fares were available—through 1988, the average fare per passenger mile, adjusted for inflation, declined by 9 percent at airports serving small communities, 10 percent at airports serving medium-sized communities, and 5 percent at airports serving large communities.’). 96 Stephen Breyer, ‘Airline Deregulation, Revisited’ Bloomberg (New York, 20 January 2011) <https://www.bloomberg.com/news/articles/2011-01-20/airline-deregulation-revisitedbusinessweek-business-news-stock-market-and-financial-advice> accessed 15 October 2019. 97 ibid. On Breyer and airline deregulation, see Robert H Nelson, ‘The Economics Profession and the Making of Public Policy’ (1987) 25 J Econ Lit 49, 62, and Breyer (n 11) chs 11 and 16. 98 Severin Borenstein, ‘The Evolution of U.S. Airline Competition’ (1992) 6 J Econ Persp 45, 45. 99 US General Accounting Office, Airline Deregulation: Changes in Airfares and Service at Buffalo, New York (20 September 1999, GAO/T-RCED-99-286) 1. 100 Open Markets Institute, Airlines & Monopoly <https://openmarketsinstitute.org/explainer/airline-monopoly/> accessed 15 October 2019 (‘Indeed, after adjusting for changes in energy prices, a 1990 study by the Economic Policy Institute concluded that airline fares fell more rapidly in the 10 years before 1978 than during the subsequent decade. Another study finds airfares would have fallen still faster after 1978 if, instead of letting carriers charge whatever the market would bear, the CAB had been allowed to continue enforcing its long-standing formulas for setting maximum fares.’ (citing Dempsey and Richards studies, discussed below)). 101 Paul Stephen Dempsey, ‘Airline Deregulation and Laissez-Faire Mythology: Economic Theory in Turbulence’ (1990) 56 J Air L & Comm 305. 102 Most glaringly, economists have pointed out that ‘the significant weakness’ in Dempsey’s analysis is that he controls for only two factors, inflation and fuel prices, whereas the ‘Morrison and Winston study does a much more complete comparison and finds substantial price decreases relative to regulation’. Borenstein (n 98) 46 n1. Dempsey controls only for fuel and inflation despite acknowledging that fuel accounts for a relatively small portion of overall airline costs and criticizing the Morrison and Winston study because it ‘does not hold all other factors constant’. ibid 356 (emphasis in original); see ibid 350 (recognizing that ‘fuel constituted anywhere from twelve percent of costs … to thirty percent’ during the regulatory period); see also Richards (n 80) 75 Fig. 3 (showing that non-fuel costs rose significantly after 1979 while fuel costs remained relatively flat until 2004). Because the proxy that Dempsey constructs does not control for all of the factors he deems relevant, let alone estimate the air fares that would have been set had regulation continued, it therefore cannot support his claim that ‘[t]he twenty-eight percent fall in real yields that has occurred since deregulation would have occurred under regulation as well’. ibid 355. In addition, Dempsey takes the same aggressive tack as Richards, counting the significant benefits of limited deregulation between 1974 and 1979 as part of the ‘regulation’ sample and incorporating the large price reductions during that era in his trend line. See ibid 348–54 and 353 tbl. 1. As explained below at nn 111–112, significant deregulation of fares began in 1974, and indeed Dempsey’s data shows real yields fell 20 per cent between 1974 and 1979. ibid 353 tbl. 1 (from 116.2 to 94.2). 103 Airline Competition: Special Hearings Before the S. Comm. on Appropriations, 105th Cong. 56 (1998) (statement of Paul Dempsey, Vice Chairman, Frontier Airlines and Professor, University of Denver/College of Law) <https://www.govinfo.gov/content/pkg/CHRG-105shrg53117/html/CHRG-105shrg53117.htm> accessed 15 October 2019. 104 ibid 56–57. 105 ibid 64 (Preamble to Dempsey’s proposed ‘Airline Competition Act of 1998’). 106 David Morris, ‘Airline Deregulation: A Triumph of Ideology Over Evidence’ Huffington Post (New York, 13 December 2013, updated 19 April 2017) <https://www.huffpost.com/entry/airline-deregulation-ideology-over-evidence_b_4399150> accessed 15 October 2019 (citing the 1990 Dempsey paper as primary evidence for the argument that ‘regulation worked better’ than competition); Open Markets Institute (n 100) (citing Dempsey paper to support the claim that CAB regulation ‘on balance … worked well’ because ‘airline fares fell more rapidly in the 10 years before 1978 than during the subsequent decade’). 107 Richards (n 80) 87. 108 ibid 85. 109 ibid 87. 110 ibid 80 tbl. 2 (estimating a total weighted average ‘ratio of average fares to SIFL’ of 0.861 in 1979 and 0.726 in 2005, representing a decline in the index of more than 15 percent); ibid 81 (explaining that ‘any ratio less than those shown in 1979’—such as the 0.726 in 2005 versus 0.861 in 1979—‘indicates a general reduction in fare level’). 111 See eg Winston (n 86) 89 (‘Economic deregulation of the airline industry began in 1974 when the Civil Aeronautics Board first encouraged experiments with discount fares. It was completed in 1983 when all regulations on fares, entry, and exit were eliminated.’). 112 Severin Borenstein and Nancy L Rose, ‘How Airline Markets Work…or Do They? Regulatory Reform in the Airline Industry’ in Nancy L Rose (ed), Economic Regulation and Its Reform: What Have We Learned? (NBER 2014) 63, 78 (discussing the Richards paper and noting that Richards chose as his regulated reference point a date by which ‘significant relaxation of fare controls had already occurred’). 113 GAO Airline Deregulation (n 95) ‘Highlights’. 114 Borenstein and Rose (n 112) 63, 78 (finding ‘actual fares were about 26 percent lower than SIFL formula fares in 2011, suggesting a consumer welfare increase in the range of $31 billion in that year’ and noting factors suggesting that this figure may be either understated or overstated); see also GAO Airline Deregulation (n 95) 24 (summarizing papers by (i) Winston and Morrison (1995) and (ii) Rose and Borenstein (2005) that estimated deregulation reduced average airfares by between 22 and 33 per cent). 115 See above nn 34–37 and accompanying text. 116 See above nn 81–84 and accompanying text. 117 See eg D Daniel Sokol, ‘Explaining the Importance of Public Choice for Law’ (2011) 109 Mich L Rev 1029, 1033. This principle particularly holds true in the area of competition policy. See William E Kovacic, ‘The Institutions of Antitrust Law: How Structure Shapes Substance’ (2012) 110 Mich L Rev 1019. 118 Breyer (n 11) 184. 119 ibid (emphasis in original). 120 ibid. 121 ibid (emphasis in original). 122 ibid 185. 123 See Cyril Ritter, The Interface between Competition and Regulation in EU Law (paper presented at the Antitrust Enforcement Symposium 2019, CCLP(S)165) 21–22 (on file with authors) (‘As regards regulation, one could consider a five-part test for determining whether regulation may be appropriate to address a particular issue, (a) There is a clearly definable issue …. (b) The issue is widespread or recurrent … (c) The issue is usually bad, or always bad, so that it does not require a case-by-case assessment … (d) The regulatory rule takes competition principles into account … (e) The regulatory framework would implement a reliable enforcement mechanism …’. (emphasis in original)); see ibid 3 (noting that ‘[t]he debate on the respective merits and roles of competition enforcement and regulation is particularly intense when it comes to digital matters’). 124 Kevin Carty, Leah Douglas and Lina Khan, ‘6 Ideas to Rein in Silicon Valley, Open Up the Internet, and Make Tech Work for Everyone’ NY Magazine (11 December 2017) <http://nymag.com/intelligencer/2017/12/open-markets-institute-antitrust-for-silicon-valley.html> accessed 15 October 2019 (Idea #6, authored by Lina Khan). 125 Warren (n 1). 126 Furman Report (n 4) paras 2.4–2.5 (‘Some assessments see digital markets as raising no particular issues and needing nothing beyond existing competition frameworks to continue delivering the benefits of well-functioning markets. Others have argued they are natural monopolies that need regulating in the same way as electricity, gas or railway networks. The Panel’s assessment is that, while they share some important characteristics with natural monopolies, it is too early to conclude that competition within and for digital markets cannot be achieved.’). 127 Railroad Problem (n 8). 128 Khan (n15). 129 ibid 1085. 130 Carty, Douglas and Khan (n 124) (Idea #5, Leah Douglas) (‘Just as the giant corporation has used its power to engage in predatory pricing and to avoid paying sales tax to drive thousands of retail stores across America out of business, [Amazon] could now do the same to many local and regional groceries. This would result both in greater concentration of power over food retailing, and even fewer physical stores…. To prevent these harms, Amazon should not only be blocked from future grocery acquisitions but its purchase of Whole Foods should be unwound. And while regulators at the Federal Trade Commission are taking care of this business, they should also ban Amazon from engaging in any price discrimination in food products, anywhere, ever. Without these safeguards, we risk handing over a huge swath of our food economy to one giant corporation, and having that giant harm our well-being in fundamental ways.’). 131 ibid. 132 Open Markets Institute (n 100). 133 See eg Dudley (n 17) 8 (noting that the author’s flight to Massachusetts for the weekend ‘would have been an unthinkable luxury’ during the CAB era because back then ‘air travel was reserved for businessmen or the wealthy’). 134 See above nn 30–31 and accompanying text. 135 Railroad Problem (n 8). 136 See above n 130 and accompanying text. Published by Oxford University Press 2019. This work is written by US Government employees and is in the public domain in the US. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Antitrust Enforcement Oxford University Press

The growing nostalgia for past regulatory misadventures and the risk of repeating these mistakes with Big Tech

Journal of Antitrust Enforcement , Volume Advance Article – Oct 1, 2008

Loading next page...
 
/lp/oxford-university-press/the-growing-nostalgia-for-past-regulatory-misadventures-and-the-risk-FuRJlIiWJx
Publisher
Oxford University Press
Copyright
Published by Oxford University Press 2019. This work is written by US Government employees and is in the public domain in the US.
ISSN
2050-0688
eISSN
2050-0696
DOI
10.1093/jaenfo/jnz029
Publisher site
See Article on Publisher Site

Abstract

Abstract Digital markets are increasingly described as the ‘railroads’ of the 21st century. Extending that metaphor, some commentators argue we should revive stale railroad-era economic regulations and adapt them to the digital age. This enthusiasm appears to be buoyed by both a sudden nostalgia for railroad and airline regulations once administered by the Interstate Commerce Commission (ICC) and the Civil Aeronautics Board (CAB) and an equally sudden amnesia of the enormous harm those regulations caused to consumers. ICC and CAB regulations are indeed an apt metaphor, as they illustrate perfectly how sectoral regulations sold to the public as simple, clear, and cheap can go awry. Ultimately, a bipartisan consensus emerged to disband those agencies and deregulate those industries. After deregulation, prices fell, output expanded, and firms innovated. Proposals to regulate Big Tech today in a similar fashion forget these important lessons. We should know better than to do the same thing again today and expect a different result. I. INTRODUCTION Heavy sectoral regulation is back in vogue. In the USA, Senator Elizabeth Warren argued last year for ‘break[ing] up Big Tech’ by emulating ‘Progressive Era’ regulations that ‘required a structural separation between the network and other businesses, and also demanded that the network offer fair and non-discriminatory service’.1 In Europe, the European Commission recently implemented the General Data Protection Regulation (GDPR)2 and continues to consider additional digital regulations, including competition rules presuming certain business practices or transactions involving digital platforms to be unlawful unless the platforms can prove otherwise.3 Australia, Japan, and the UK are studying heightened regulations for digital markets,4 and the Competition Commission of India recently heard (and rejected) a significant case involving two digital platforms.5 In the USA, this discussion has induced comparisons to—and an odd nostalgia for—the comprehensive regulations that once governed transportation. In 2017, one commentator argued Amazon was ‘a 21st-century version of the 19th-century railroads’ and compared Amazon’s practices to those of ‘the railroads of yore’.6 Her then-employer, a left-leaning Washington think tank, has adopted this thread as its own, proclaiming online platforms to be ‘the railroad monopolies of the 21st century’.7 And two opinion articles in the Financial Times argued that ‘the internet is the railroad of our times’8 and that the ‘obvious’ solution to today’s problems is to recycle yesteryear’s railroad regulations and adapt them to 21st-century technologies.9 Many of these proposals implicitly or explicitly favour imitating regulations once administered by the US Interstate Commerce Commission (ICC) and the Civil Aeronautics Board (CAB). Congress created the ICC in 1887 with a mandate to regulate railroads;10 at its peak, its jurisdiction also included barges, bus lines, trucks, and pipelines. The ICC also briefly regulated airlines,11 although it lost this authority in 1938 when Congress created a new regulator that soon became the CAB.12 According to the Financial Times columnist, the formation of the ICC ‘ushered in a period of prosperity’,13 and its purported success means ‘[i]t is time to think hard about whether we need another ICC – an Internet Commerce Commission – to do the same’ today.14 Others similarly extoll the virtues of the old ICC and its focus upon structural separation and ‘fair and non-discriminatory’ access.15 The sudden enthusiasm for the old ICC and CAB—which Congress disbanded in 1996 and 1985, respectively—stands in stark contrast to the epitaphs these agencies earned from those who actually knew them. The CAB was deeply unpopular in the years before it was disbanded. For example, a Senate subcommittee lead by Sen. Ted Kennedy in 1975 found that ‘present CAB practice is fundamentally deficient’ and ‘Board regulation has not effectively brought about the low-fare service that is technically feasible and that consumers desire’.16 In 1978, CAB Chairman Alfred Kahn famously testified that he often wondered whether deciding thousands of picayune decisions—from whether an airline may buy a new plane to whether a supplemental (charter) airline may transport a flock of sheep to England—was ‘what my mother raised me to do’.17 Two successive Presidential administrations—one Republican, the other Democratic—sought and eventually succeeded in disbanding the CAB.18 The ICC fared no better. In the late 1960s, a study prepared for the Brooking Institution conservatively estimated that ICC regulations cost consumers at least $500 million per year (in 1960s dollars) and that ending ICC regulation ‘would create benefits far in excess of costs’.19 In 1975, President Ford described ICC (and CAB) regulations as something ‘begun as a protection for consumers’ that ‘now guarantees that in many cases they will pay higher prices than a free market would call for’.20 In 1980, President Carter enthusiastically signed the Staggers Rail Act, remarking that it ‘will benefit’ both shippers and consumers ‘[b]y stripping away needless and costly regulation’ by the ICC.21 And in 1995, upon signing the bill disbanding that agency, President Clinton remarked that he was ‘disappointed’ by a bill that only disbanded the ICC because the bill ‘falls short of my administration’s much bolder proposal for extensive deregulation of transportation industries’.22 With the passage of time, fewer and fewer of today’s policymakers were personally involved in those events. As a new generation takes their place, ideas that were tried and failed are once again gaining purchase, albeit now as a solution to perceived problems with Big Tech. Because ‘[t]hose who cannot remember the past are condemned to repeat it’,23 we should remember the reality: railroad-style regulations distorted competition in the marketplace, reduced economic efficiency, hindered innovation, and harmed the very consumers they ostensibly protected.24 More specifically, the American experience with the ICC and CAB illustrates three important lessons. First, provisions that sound simple in theory, like a ‘non-discrimination’ requirement, seldom prove simple in practice. Secondly, effective regulation requires clear instructions, particularly regarding the ‘what’, ‘why’, and ‘how’. This kind of clarity is needed not just at the outset, but also over time, as changed circumstances and mission creep may later confuse matters and lead to perverse and unintended results. Thirdly and finally, given the substantial consumer benefits that flowed from the elimination of ICC and CAB regulations, regulations—however well-intentioned—can impose very real costs on consumers. Or, in three words, simplicity, clarity, and costs. Many of today’s proposals to regulate Big Tech forget these important lessons. Proposed digital regulations are sold to the public as simple, obvious, and beneficial. Behind the curtain, however, there is little clarity on what the problem is, let alone why a given proposal is the best solution or how the agency would operationalize it. Nor is there any recognition that burdensome new regulations may distort markets and harm consumers. In short, given the tarnished history of the ICC and CAB, we should know better than to do the same thing again today and expect a different result. II. THE AMERICAN EXPERIENCE WITH THE ICC AND CAB Like today’s calls for heavier regulation of digital markets, both the ICC and the CAB were created to regulate perceived problems with dynamic industries. Both were given mandates to regulate their respective industries ‘in the public interest’, a broad, vague, and malleable standard. Presented with an amorphous problem and vague instructions, both agencies constructed complex and increasingly anti-competitive regulations. A broad political consensus eventually emerged to dismantle the regulations and disband the regulators, leading to widespread consumer benefits. This section traces the evolution of each agency. The ICC In the late 19th century, the Age of Steam, the railroad was considered both cutting edge technology and big business. In the USA, ribbons of steel opened the American West to development, with farms, mines, towns, and factories developing in their wake. With new development came new economic patterns. What we might now call geographic markets expanded, as products could be shipped in from far away. For example, wheat grown in North Dakota was shipped by rail to Minneapolis, where it was milled into flour, and shipped by rail again to bakeries in Chicago, St. Louis, and even New York City. Given the important role that railroads played in connecting buyers to sellers, they naturally elicited complaints from all sides. Farmers complained that they often had only one choice, whichever railroad ran closest to their property, on which to ship their produce, thereby forcing them to pay supra-competitive rates. Many merchants made similar complaints. Both groups called for new regulation at both the state and federal level to ensure that this service—which they deemed essential—was offered at fair rates on non-discriminatory terms.25 The railroads, too, were dissatisfied. When more than one railroad served a given route, as occurred in the important grain routes from Chicago to ports on the East Coast, the railroads themselves complained of ‘ruinous competition’.26 Railroads often formed ‘pools’—which today we would simply call cartels—to allocate traffic and prop up prices. And like today’s cartels, each firm had an incentive to cheat by giving powerful customers—like Standard Oil—secret rebates to secure additional traffic.27 Eventually, both camps struck upon the idea of regulation to cure all that ailed them. Customers envisioned comprehensive state and federal regulations that would set the terms of service and prohibit price discrimination. The railroads envisioned targeted regulations that would use the coercive power of the state to end secret rebating and thereby ensure the stability of their cartels.28 So in 1887 the US Congress passed the Act to Regulate Commerce, better known today as the Interstate Commerce Act.29 To those of you following today’s digital markets proposals, its provisions may sound eerily familiar. Section 1 of the Act required railroads to charge rates that were ‘reasonable and just’,30 whereas sections 2, 3, and 4 prohibited ‘unjust discrimination’ in the setting of rates or terms of service.31 Using language that echoes the Sherman Act, which was enacted three years later, section 5 banned ‘any contract, agreement, or combination … for the pooling of freight’.32 The Act also created an independent agency, the ICC, to enforce the Act.33 Yet public optimism soon evaporated; new regulation was not the magic elixir all had hoped. The answer, obviously, was more regulation. So in 1903 Congress passed the Elkins Act, which banned railroad rebates to large industrial customers but not small ones, and therefore supposedly levelled the playing field.34 This step, too, was insufficient, so in 1906 Congress passed the Hepburn Act,35 which strengthened the ICC’s rate-setting powers and banned vertical integration. In 1910 Congress granted the ICC the authority to suspend railroad rates during an investigation,36 and in 1920 Congress added the authority to set minimum rates for the explicit purpose of protecting financially weaker railroads from stronger ones.37 The 1920 Act also allowed firms to pool traffic—in other words, to form a cartel—if the agency determined that service would be improved and competition not ‘unduly’ restrained.38 And when new forms of transportation began to compete with the railroads, Congress granted the ICC the authority to regulate most of them, as well.39 Over time, the ICC acquired jurisdiction over many other forms of transportation, from trucking and barge traffic to natural gas pipelines, leading to conflicting priorities. Airlines and international shipping eventually escaped, regulated instead by their own specialist agencies. Three anecdotes illustrate the scope of the ICC’s regulatory misadventures. First, as previously described, the 1906 amendments prohibited railroads from vertically integrating. Populists at that time were concerned that railroads—which were the only economical way to ship coal—were vertically integrating into the coal mining business, thereby hurting less efficient independent producers. (Indeed, proponents of new regulations today have returned to this example.40) So Congress considered legislation, which eventually would become the Hepburn Act, to bar railroads from vertically integrating into other industries (a practice that was already commonplace, particularly for timber and coal). The prohibition in the draft bill was soon broadened over the objection of some of its supporters,41 although the Senate also rejected an even broader amendment that would have banned common carriers from producing coal and other products for their own use.42 In what became known as the Commodities Clause, this additional provision made it a criminal offense for a railroad to carry any product for sale43—other than timber, whose lobbyists got a special exemption—that it either had produced or in which it had an economic interest.44 All sides soon realized how economically damaging such a ban would be. In early litigation involving this provision, the Supreme Court adopted a narrow interpretation that rendered it inapplicable in most cases.45 The railroads also learned to adapt their corporate form in ways that sidestepped the Commodities Clause.46 In a subsequent case against US Steel, the provision was further cabined, in essence applying only to the coal industry.47 And in those rare instances when the US government did assert the clause, it was typically subsidiary to a Sherman Act claim, thereby suggesting that the antitrust laws were sufficient to correct any abuses.48 Congress apparently agreed, time and again rejecting legislative attempts to override the narrow interpretation adopted by the Supreme Court.49 Secondly, in the 1950s, the Southern Railroad developed a new kind of railcar for grain that was more than twice as efficient as earlier methods.50 When the railroad sought to lower rates by 60 per cent to reflect these new efficiencies, the ICC denied the request pending its investigation. It ultimately rejected the railroad’s proposed rate on the grounds that the rate was so low that it would harm shipments by barge, instead authorizing a 53.5 per cent reduction.51 The case—generally known as the ‘Big John’ matter—went to the Supreme Court, which sided with the railroad, but only after customers had been forced to pay the higher ICC-imposed rate for four years.52 In addition, scholars have estimated that the Southern Railroad incurred between $20 million and $40 million in expenses seeking to overturn the ICC’s initial order,53 an amount that the National Academy of Sciences found ‘significantly undermines such incentives to innovate’.54 Thirdly, in 1965, two railroads lowered their rate for shipping steel between Pittsburgh and a downstream factory in Kentucky to match the competing rate charged by barges and trucks. In a proceeding called the Ingot Molds decision, the ICC disallowed the rate as too low to both cover its costs and produce what the ICC believed to be a ‘fair’ profit.55 The Supreme Court affirmed the ICC’s Order.56 Remarkably, it also commended the ICC for ‘properly’ avoiding ‘a rate war’ because the competing barge-truck rate was not ‘in need of being driven down by competitive pressure’.57 Decrying these and other obvious mistakes, in 1971, a professor at the Massachusetts Institute of Technology aptly summarized the situation in an op-ed in The New York Times: ‘The heavy hand of regulation, which [Congress] has laid on the industry through the Interstate Commerce Commission, has stifled innovation, discouraged industry, [and] driven many of the capable and imaginative men out of it.’58 That year, 30 US senators called for the abolition of the ICC.59 Soon after, politicians of all stripes concluded that deregulation was necessary. The rail and trucking industries were significantly deregulated starting in the late 1970s, and the ICC was disbanded entirely at the end of 1995 by a Democratic President and a Republican Congress.60 Freed from restrictive price and service rules, railroads optimized their networks. They pared unproductive routes and reduced labour costs, increasing efficiency markedly on their remaining routes.61 Rail rates eventually fell system-wide,62 and declines were particularly steep for bulk commodities like grain, lumber, and coal that could be moved much more effectively using methods that ICC regulations had heretofore discouraged.63 Trucking rates also fell precipitously, at least in those areas where state-level regulations did not survive. For example, a 1989 study by the US Federal Trade Commission found that trucking rates in deregulated markets were roughly 40 per cent lower than the rate on a regulated, but otherwise equivalent, route in Texas.64 Depending upon the type of trucking service, prices and costs fell between 35 and 75 per cent,65 yielding consumer benefits of ‘more than $18 billion’ in 1996 dollars.66 A 2010 joint study by the Federal Trade Commission and the Antitrust Division of the US Department of Justice concluded that ‘deregulation of the trucking industry … has been entirely beneficial for consumers’.67 Milton and Rose Friedman aptly summarized the arc of the agency in their 1980 book Free to Choose: The ICC illustrates what might be called the natural history of government intervention. A real or fancied evil leads to demands to do something about it. A political coalition forms consisting of sincere, high-minded reformers and equally sincere interested parties. The incompatible objectives of the members of the coalition (e.g., low prices to consumers and high prices to producers) are glossed over by fine rhetoric about “the public interest”, “fair competition”, and the like. The coalition succeeds in getting Congress (or a state legislature) to pass a law… . Success [by interested parties] breeds its problems, which are met by broadening the scope of the intervention… . In the end the effects are precisely the opposite of the objectives of the reformers and generally do not even achieve the objectives of the special interests. Yet the activity is so firmly established and so many vested interests are connected with it that repeal of the initial legislation is nearly inconceivable. Instead, new government legislation is called for to cope with the problems produced by the earlier legislation and a new cycle begins. The ICC reveals clearly each of these steps … .68 Or, as the Editorial Board of The New York Times put it in 1996, the ICC was ‘[i]nitially created to protect farmers and communities from the monopoly power of railroad barons’, but ‘later ganged up on the consumers it was supposed to protect, siding instead with industries—especially trucking—that it was supposed to regulate.’69 The CAB The story of airline regulation in the USA is much the same. Airline regulation began in 1926 with the formation of the Aeronautics Branch of the Commerce Department, which focused principally on air safety.70 After briefing transferring some functions to the ICC,71 in 1938, Congress created a special regulator just for airlines.72 Thus, the CAB was born. Like the ICC, the CAB was directed to place other ‘public interest’ values ahead of competition, which was allowable ‘to the extent necessary to assure the sound development of an air-transportation system properly adapted to the needs of the foreign and domestic commerce of the United States’.73 Given this directive, once one airline served a given route, like Chicago to Memphis, the CAB typically refused to allow any other carriers to fly the same route on the grounds that the competition was not necessary to develop that route.74 Between 1950 and 1974, the Board received 94 applications for new airlines to provide ‘trunk-line’ air service between major US cities—and denied every last one of them.75 Although the CAB occasionally allowed incumbents to add new routes, it did so only reluctantly; remarkably, it took Continental Airlines more than eight years to receive approval to fly a new route between Denver and San Diego,76 and only then because the D.C. Circuit vacated the Board’s denial of the application and directed the Board to approve a carrier.77 When multiple airlines were allowed to compete on the same route, the CAB rarely allowed the carriers to cut their prices and the carriers rarely asked.78 These high prices depressed demand, with airlines often flying planes that were less than 60 per cent full.79 Indeed, the CAB set pricing on the assumption that the average plane would fly at only 55 per cent of capacity.80 To fill those empty seats, carriers increasingly competed on service. Air carriers that could not compete on price instead competed by offering carved chateaubriand on rolling silver carts, piano lounges on the upper decks of Boeing 747s, or free alcoholic drinks.81 Yet the CAB increasingly intervened here as well to avoid any changes that might permit ‘unfair’ competition on quality. At the top end, competitor complaints caused the CAB to deny Trans World Airlines’ request to add lie-flat sleeper seats to its trans-Atlantic first class service.82 On the bottom end, the agency initially protected first-class flights by permitting coach-class flights only at off-peak times.83 The airlines’ industry group also got into the act, making industry-wide agreements that, as economist Alfred Kahn described them, ‘prescribe[ed] the maximum allowable knee-room …, dictat[ed] that meals be limited to sandwiches … and requir[ed] a uniform supplementary charge for in-flight motion pictures’.84 As with the ICC, the inefficiencies built until the public could bear them no longer. In the early 1970s independent economists found that fares on intrastate carriers not regulated by the CAB were substantially lower than the fare on an interstate carrier regulated by the CAB—sometimes even for the exact same route.85 In 1974, the CAB ‘first encouraged experiments with discount fares’ that it heretofore had prohibited.86 Senator Edward Kennedy and then-Senate counsel Stephen Breyer began a series of hearings in 1975 questioning the efficacy of CAB regulations.87 That same year, the ‘Pulsifer’ CAB staff study found that ‘the present regulatory regime has produced undesirable economic effects, is not justified by the structural economic characteristics of the industry …, and seems likely to produce even more objectionable results in future years’.88 Because ‘[t]hese undesirable effects outweigh the benefits of such regulation’, and because these inefficiencies were the product ‘of the law itself’, rather than how the CAB administered it, the CAB staff recommended repealing the statutes authorizing these regulations.89 Chairman John E. Robson surprised observers in 1976 by announcing that the CAB Commissioners unanimously agreed that Congress should reduce the CAB’s authority and loosen regulations.90 In 1977, President Carter appointed Alfred Kahn, a leading critic of the CAB, as its Chairman. Kahn pushed for radical deregulation, and policymakers agreed. With the passage of the Airline Deregulation Act of 1978,91 the CAB began the process of disbanding, closing at the beginning of 1985.92 The liberalization of airline markets quickly reduced airfares and increased output. By 1990, 12 years after deregulation, economists estimated that airline deregulation was already increasing US consumer surplus by more than $6 billion per year, or $20 billion annually in 2019 dollars.93 In the 1990s and 2000s, the Government Accountability Office (GAO) similarly found large airfare reductions, which ‘resulted in large part from increased competition’,94 and in particular a significant increase in the number of airlines that serve each airport.95 And the benefits of airline deregulation continue to grow. Stephen Breyer, by now a Justice of the US Supreme Court, noted in 2011 that ‘the number of airline passengers increas[ed] from 207.5 million in 1974 to 721.1 million’ in 2010.96 During that same time, ‘[a]irline revenue per passenger mile … declined from an inflation-adjusted 33.3 cents … to 13 cents’, and the cost of a flight from New York to Los Angeles fell, again in inflation-adjusted terms, from $1442 to $268.97 Given the overwhelming evidence that airline deregulation benefitted consumers, almost all economists, lawyers, and policymakers view the dissolution of comprehensive airline regulation as the right decision. For example, one leading scholar noted that ‘[t]he simplest prediction of economists about airline deregulation, and one of the few on which nearly all economists agreed, was that deregulation would improve consumer welfare in comparison to continued price and entry regulation’ and, ‘[f]ourteen years later, nearly all economists still agree on this.’98 Similarly, the then-General Accounting Office reported to Congress in 1999 that ‘[o]ver the years, our work has consistently shown that airline deregulation has led to lower fares and better service for most air travelers’ and that these benefits are ‘largely due to increased competition spurred by the entry of new airlines into the industry and established airlines into new markets’.99 The sceptics occasionally point to two purported exceptions,100 both of which actually reinforce the consensus that airline deregulation benefitted consumers. First, former CAB lawyer Paul Stephen Dempsey published a paper in 1990 arguing that airline prices would have fallen even faster under regulation than under deregulation.101 Setting aside the methodological weaknesses in his analysis,102 Dempsey subsequently testified before Congress that he does not support the imposition of new regulations.103 He testified that ‘[w]herever possible competition is preferable to Government as regulator of the market’ and that he is a ‘proponent[] of competition, … not regulation’.104 He also proposed limited new legislation premised upon the belief ‘that the competition unleashed by airline deregulation has been beneficial to large segments of the consuming public, but that competition should be further advanced so that a larger universe of Americans can enjoy the benefits of airline deregulation’.105 Ironically, some critics now cite his 1990 study as ‘Exhibit A’ for the purported failure of airline deregulation106 and forget his subsequent sworn testimony to the contrary. Secondly, former CAB economist David B. Richards wrote an article in the mid-2000s arguing that ‘the grant of pricing freedom to the airline industry has generally resulted in average prices being higher than they would have been had regulation continued’.107 He recognizes that this conclusion is inconsistent with ‘most other studies’.108 Yet he fails to mention that his conclusion is also inconsistent with some of his own data, which he says elsewhere show ‘considerable passenger savings’ in the early 2000s.109 Indeed, on a weighted average basis, he found that the average US airfare in 2005 was more than 15 per cent lower than it would have been if the 1979 version of the fare regulations (the so-called ‘SIFL’) had remained in effect.110 This estimate is likely conservative, as other studies point out that fares had already been partially deregulated by 1979,111 so comparisons using that year as the base may understate the overall effect of dismantling the CAB regime.112 Moreover, Richards’ data (if not his conclusion) are consistent with several other studies the US GAO relied upon when it concluded that ‘reregulating the airline industry would likely reverse consumer benefits’.113 These studies estimate that airfares are approximately 20–30 per cent lower than they would be if the 1979 regulations were still in effect.114 III. LESSONS LEARNED Given this history, Americans have learned—the hard way—three lessons from earlier efforts to require ‘just’, ‘reasonable’, and ‘non-discriminatory’ pricing and service in network industries. First, no matter how simple the regulatory principle, implementing the regulations day-to-day is almost always devilishly complex. In the ICC context, the framers of the original 1887 Act thought their commandment of ‘just’ and non-discriminatory rates would be straightforward. Yet policymakers felt it necessary to clarify and augment the Commission’s authority in major new statutes in 1903, 1906, 1910, and 1920.115 Similarly, the CAB found that regulating entry alone was insufficient, leading it to regulate prices and (with amusing but likely inefficient results) service levels.116 Secondly, effective regulation requires a clear ‘what’, ‘why’, and ‘how’. More specifically, we must first ask what problem regulation is necessary to solve. Once a problem is clearly identified, we must then ask why the proposed regulation is the best option. And once those items are clear, we must finally ask ourselves how the new agency may achieve this solution and, at least as importantly, how it may not. Or, to put it another way, how we design institutions and what we ask them to do matters a great deal.117 For example, the ban on vertical integration embodied in the Commodities Clause was not the best way to solve perceived abuses in the coal industry, as most agree that antitrust enforcement forced the industry to change its practices. Similarly, even if there had been some legitimate need for ICC regulation, it should not have been permissible for the agency to use rate-setting policies to protect less efficient producers—like the barges in Ingot Molds—from more efficient producers. As the ICC’s expanding jurisdiction illustrates, mission creep—often driven by technological change—only exacerbates these problems. And thirdly, as our experience after deregulation shows, even the most well-intentioned regulations come at a steep cost. Removing ICC and CAB regulations significantly reduced consumer prices and increased output, generating billions of dollars in consumer surplus. Many other students of regulation have distilled similar principles that recognize the importance of simplicity, clarity, and alternatives to regulation. For example, future Justice Stephen Breyer distilled an exhaustive study of regulatory failings in the 1970s into three ‘rules of thumb’.118 ‘First and most obviously, modesty … in one’s approach to regulation’,119 by which he means that regulation should be reserved only for ‘serious defects for which regulation offers a cure’ and that we must consider the potential difficulties of ‘regulatory alternatives’ to ‘an unregulated status quo’.120 Secondly, ‘regulators ought to aim at worst cases and, in attacking such cases, they should strive for simplicity’.121 And thirdly, given ‘[t]he problems accompanying classical regulation’, it ‘ought to be looked upon as a weapon of last resort’.122 DG-Comp official Cyril Ritter embeds similar concepts in his proposed five-part test for deciding whether regulation is well-suited to address a problem, like what he calls ‘digital matters’.123 IV. APPLICATION TO TODAY’S PROPOSALS Reading the headlines today, some appear to have forgotten these valuable and hard-won lessons. This amnesia is particularly alarming given recent calls to replicate our experience with the ICC and CAB by imposing similar regulations upon Big Tech. Many of these proposals forget the first lesson, selling railroad-style regulations as simple and beneficial. For example, Lina Khan argues that ‘the best way to preserve fair and open competition’ is ‘simply to completely ban any network monopolist from owning businesses that place it in competition with the companies that depend on it to reach [the] market’, which ‘is what previous generations did with railways’.124 Similarly, based on what she believes to be the success of previous railroad regulations, Senator Elizabeth Warren would like to impose similar ‘structural separation’ and demand ‘that the network offer fair and non-discriminatory service’.125 To be clear, this enthusiasm is not universally held; for example, the UK’s Furman Report notes that many have called for regulating digital markets ‘in the same way as electricity, gas, or railway networks’, but found that it was ‘too early to conclude’ that such regulation would be necessary or appropriate.126 Other proposals forget the second lesson, arguing for the re-imposition of regulations like the ICC’s old Commodities Clause without considering whether it is the best option available. For example, an opinion piece published in the Financial Times last week traced the early history of the Commodities Clause and argued that ‘[i]t seems obvious … that we should apply these same standards to the digital giants of today.’127 Lina Khan similarly touts the potential of imposing a regime like the Commodities Clause on technology platforms,128 although she at least acknowledges doing so may ‘sacrifice certain cost savings, resulting in higher prices’.129 Khan’s honesty—that imposing heavy new regulations on Big Tech may result in higher prices—is refreshing but rare. Indeed, most have also forgotten the third lesson. For example, in 2017, a fellow Open Markets Institute commentator argued that we should ‘stop Amazon from selling groceries’—which sounds suspiciously like the kind of entry restrictions the CAB imposed on airlines—because banning Amazon from the market would somehow increase competition.130 The same commentator also argued the Federal Trade Commission must ‘ban Amazon from engaging in any price discrimination in food products, anywhere, ever’,131 another failed approach the CAB employed. And the Open Markets Institute argues that CAB regulation was preferable to today’s marketplace, because ‘regulation by the CAB prevented airlines from abusing their market power while also ensuring that citizens in cities of comparable size received roughly equal service, in terms of both quality and price’.132 That claim is strictly true, at least in the sense that a much smaller number of customers paid the same much higher price and ate chateaubriand. Given the Institute’s populist instincts, this claim is also ironic, as it would return us to an age when flying was only for the very wealthy.133 V. CONCLUSIONS In conclusion, those who cannot remember history are condemned to repeat it. More than 100 years ago, Congress decided the platform industry of the day—railroads—must charge rates that are ‘reasonable and just’ and eschew ‘unjust discrimination’.134 These apparently simple requirements turned out to be highly complex, prompting the growth of a regulatory leviathan that banned price and non-price competition alike on the grounds that competition might harm smaller, less efficient firms. The credo also spread to airlines. Ultimately, academics and policymakers found that those regulations, however well-intentioned, were costing consumers tens of billions of dollars a year. Broad political support quickly emerged in favour of deregulation. Today, many of those lessons have been forgotten. Some now believe—apparently sincerely—that railroad and airline regulations were simple and beneficial, and therefore a great model for how to regulate Big Tech. Taking it a step further, some now argue that imposing railroad-style regulations upon Big Tech firms is simply ‘obvious’.135 And perhaps worst of all, some now argue that banning firms from the market and imposing strict pricing rules will surely increase competition and reduce prices.136 The USA spent more than 100 years trying, and failing, to make these ideas work as intended. Many have now forgotten that history and seem intent on repeating it. Acknowledgement This article is based on a speech presented by Commissioner Wilson on 28 June 2019 at the British Institute for International and Comparative Law and a related speech presented by Commissioner Wilson on 30 June 2019 at The Antitrust Enforcement Symposium, University of Oxford. The views expressed herein are solely those of the authors and do not necessarily reflect the views of the US Federal Trade Commission or any other Commissioner. While in private practice, Commissioner Wilson represented airlines in various regulatory and competition matters around the world. The authors thank Neil Chilson, Brian Keating, Jeremy Sandford, D. Daniel Sokol, and Joshua Wright for helpful comments. Of course, all errors are our own. Footnotes 1 Elizabeth Warren, ‘Here’s How We Can Break Up Big Tech’ (Medium, 8 March 2019) <https://medium.com/@teamwarren/heres-how-we-can-break-up-big-tech-9ad9e0da324c> accessed 15 October 2019. 2 EU GDPR: Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation) [2016] OJ L119/1. 3 Jacques Crémer and others, European Commission, Directorate-General for Competition, ‘Competition Policy for the Digital Era’ (2019) 4 <http://ec.europa.eu/competition/publications/reports/kd0419345enn.pdf> accessed 15 October 2019 (arguing, among other things, ‘‘[t]he specific characteristics of many digital markets have arguably changed the balance of error cost and implementation costs, such that … one may want to err on the side of disallowing potentially anticompetitive conducts, and impose on the incumbent the burden of proof for showing the pro-competitiveness of its conduct’’). 4 See eg Jason Furman and others, ‘Unlocking Digital Competition: Report of the Digital Competition Expert Panel’ (March 2019) 2 <https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/785547/unlocking_digital_competition_furman_review_web.pdf> accessed 15 October 2019 [hereinafter Furman Report] (‘Instead of just relying on traditional competition tools, the UK should take a forward-looking approach that creates and enforces a clear set of rules to limit anti-competitive actions by the most significant digital platforms while also reducing structural barriers that currently hinder effective competition.’); Australian Competition & Consumer Commission, ‘Digital Platforms Inquiry’ (July 2019) 3 <https://www.accc.gov.au/system/files/Digital%20platforms%20inquiry%20-%20final%20report.pdf> accessed 15 October 2019 (‘The ACCC considers that now is the time to consider the current and likely future issues associated with digital platforms and their business models and to put in place frameworks that enable adverse consequences to be addressed and that reduce the likelihood of new issues arising.’); Japan Fair Trade Commission, ‘Study Group on Improvement of Trading Environment surrounding Digital Platforms’ (12 December 2018) Interim Discussion Paper 7 <https://www.jftc.go.jp/en/policy_enforcement/survey/index_files/190220.2.pdf> accessed 15 October 2019 (‘In order to facilitate a sound development of digital platform businesses in our country, it would be necessary to consider whether we should review each industry-specific regulation.’). 5 Competition Commission of India, Case No 20 of 2018, Order under s 26(2) of the Competition Act, 2002 (11 June 2018) <https://www.cci.gov.in/sites/default/files/20-of-2018.pdf> accessed 15 October 2019. 6 Lina M Khan, Op-Ed, ‘Amazon Bites Off Even More Monopoly Power’ The New York Times (New York, 21 June 2017) A23 <https://www.nytimes.com/2017/06/21/opinion/amazon-whole-foods-jeff-bezos.html> accessed 15 October 2019 (‘Think of Amazon as a 21st-century version of the 19th-century railroads that connected consumers and producers … . In several key ways, Amazon uses its power as the railroads did.… And like the railroads of yore, Amazon dictates terms and prices to those dependent on its rails.’). 7 Open Markets Institute, ‘Free Press & Platform Monopolies’ <https://openmarketsinstitute.org/issues/free-press-platform-monopolies> accessed 15 October 2019. 8 Rana Foroohar, ‘Big Tech is America’s New ‘Railroad Problem’’ Financial Times (London, 16 June 2019) <https://www.ft.com/content/ec3cbe78-8dc7-11e9-a1c1-51bf8f989972> accessed 15 October 2019 [hereinafter, Railroad Problem]; see also Rana Foroohar, ‘Why We Need to Regulate the Tech Platforms’ Financial Times (London, 5 November 2017) <https://www.ft.com/content/84f402ac-bfc0-11e7-b8a3-38a6e068f464> accessed 15 October 2019 [hereinafter, Why We Need to Regulate]. 9 Railroad Problem, ibid (describing the ICC’s so-called Commodities Clause and arguing ‘[i]t seems obvious to me that we should apply these same standards to the digital giants of today’). 10 Act to Regulate Commerce of 1887 [Interstate Commerce Act], Pub L 49-104 (1887) 24 Stat 379 (1887). 11 See Stephen G Breyer, Regulation and Its Reform (Harvard UP 1982) 29 (describing the ICC’s short-lived authority to administer the air mail subsidy system in the 1930s). 12 See eg Edward M Kennedy, ‘Airline Regulation by the Civil Aeronautics Board (1975) 41 J Air L & Com 607, 609 (reprinting the Summary of Report of the Senate Subcommittee on Administrative Practice and Procedure) (describing the divestment of the ICC’s air mail authority in 1938). 13 Why We Need to Regulate (n 8). 14 ibid. 15 See eg William A Nitze, ‘The Tech Giants Must Be Stopped’ The American Conservative (Washington, DC, 16 April 2018) <https://www.theamericanconservative.com/articles/the-tech-giants-must-be-stopped> accessed 15 October 2019 (‘Measures that might help to achieve this result include requiring the large technology companies to deal with their customers or competitors on the basis of greater equality just as AT&T did with new telecommunications companies after its court-ordered break-up, or oil refiners did with railroads after the creation of the Interstate Commerce Commission.’); Lina Khan, ‘The Separation of Platforms and Commerce’ (2019) 119 Colum L Rev 973, 980 (‘This Article argues that these combined problems of discrimination and information appropriation invite recovering common carriage’s forgotten cousin: structural separations.... Structural prohibitions … have been applied as a standard regulatory tool and key antitrust remedy in network industries, often to prohibit a dominant intermediary from competing with the businesses that depend on it to get to market’.). 16 Kennedy (n 12) 614, 633 (report of the subcommittee); see also ibid 607 (forward by Sen. Kennedy noting he ‘fully subscribe[s] to the views expressed in this report’). 17 See Alfred E Kahn, ‘A Grateful Response and Supplement’ (Silicon Flatirons, University of Colorado, 5 September 2008) <https://siliconflatirons.org/documents/conferences/2008.09.05-207/KahnResponse.pdf> accessed 15 October 2019; see also Alan Greenspan, The Age of Turbulence – Adventures in a New World (1st edn, Penguin 2007) 71–72 (cited approvingly by Kahn); Susan E Dudley, ‘Alfred Kahn 1917-2010: Remembering the Father of Airline Deregulation’ (Spring 2011) Regulation 8, 10 (recounting the same sheep example). 18 See David Burnham, ‘Ford to Seek a Reduction of C.A.B. Regulatory Role’ The New York Times (New York, 7 February 1975) 61 <https://www.nytimes.com/1975/02/07/archives/ford-to-seek-a-reduction-of-cab-regulatory-role.html> accessed 15 October 2019 (describing a Ford Administration plan to introduce deregulatory legislation); Ernest Holsendolph, ‘Backers of Airline Deregulation Gird for Expected House Battle’ The New York Times (New York, 23 April 1978) A16 <https://www.nytimes.com/1978/04/23/archives/backers-of-airline-deregulation-gird-for-expected-house-battle.html> accessed 15 October 2019 (describing similar efforts by the Carter Administration to pass the Airline Deregulation Act of 1978). 19 Ann F Friedlaender, ‘The Dilemma of Freight Transport Regulation’ (background paper prepared at the request of the Brookings Institution for a conference held in 1967, Brookings Institution, Washington DC 1969, reprinted 1975) 164–65. 20 David Burnham, ‘Issue and Debate: Regulatory Agencies and Competition’ The New York Times (New York, 20 September 1975) 37 <https://www.nytimes.com/1975/09/20/archives/issue-and-debate-regulatory-agencies-and-competition.html> accessed 15 October 2019. 21 Statement of Pres Carter on Signing S 1946 Into Law (14 October 1980) <https://www.presidency.ucsb.edu/documents/staggers-rail-act-1980-statement-signing-s-1946-into-law> accessed 15 October 2019. 22 ‘Clinton Signs ICC Bill’ (United Press, 30 December 1995) <https://www.upi.com/Archives/1995/12/30/Clinton-signs-ICC-bill/8245820299600/> accessed 15 October 2019. 23 George Santayana, The Life of Reason: Reason in Common Sense (Dover 1980) vol 1 (1905) 284. 24 Thomas K McCraw, Prophets of Regulation: Charles Francis Adams; Louis D. Brandeis; James M. Landis; Alfred E. Kahn (Belknap Press of Harvard UP 1984) (reviewing the deregulatory work of Kahn). 25 See Richard D Stone, The Interstate Commerce Commission and the Railroad Industry: A History of Regulatory Policy (Praeger 1991) 5 (summarizing the underlying literature). 26 See ibid 5–6. 27 See Standard Oil Co of New Jersey v United States (1911) 221 US 1, 32–33, 42–43. 28 See Stone (n 25) 6 (identifying James J Hill, the famed railroad ‘Empire Builder’, as a particularly hopeful proponent of the ICC). 29 Pub L 49-104 (1887) 24 Stat 379. 30 ibid, § 1 (‘All charges made for any service rendered or to be rendered in the transportation of passengers or property as aforesaid, or in connection therewith, or for the receiving, delivering, storage, or handling of such property, shall be reasonable and just; and every unjust and unreasonable charge for such service is prohibited and declared to be unlawful.’). 31 ibid, § 2 (banning ‘unjust discrimination’ among passengers or freight transported contemporaneously ‘under substantially similar circumstances and conditions’); ibid, § 3 (banning ‘undue or unreasonable preference or advantage to any particular person’ and requiring common carriers to interchange traffic); ibid, § 4 (banning carriers from charging a higher rate, ‘under substantially similar circumstances and conditions, for a shorter than for a longer distance over the same line, in the same direction, the shorter being included within the longer distance’). 32 ibid, § 5 (‘That it shall be unlawful for any common carrier subject to the provisions of this act to enter into any contract, agreement, or combination with any other common carrier or carriers for the pooling of freights of different and competing railroads, or to divide between them the aggregate or net proceeds of the earnings of such railroads, or any portion thereof; and in any case an agreement for the pooling of freights as aforesaid, each day of its continuance shall be deemed a separate offense.’). 33 Some scholars believe incumbents favoured the creation of the ICC because they believed it would allow them to engage in rent-seeking. See eg Thomas W Gilligan, William Marshall and Barry Weingast, ‘Regulation and the Theory of Legislative Choice: The Interstate Commerce Act of 1887’ (1989) 32 J L & Econ 35. 34 57 Cong Ch 708 (1903) 32 Stat 847. 35 59 Cong Ch 3591 (1906) 34 Stat 584. 36 61 Cong Ch 309 (1910) 36 Stat 539 (Mann-Elkins Act). 37 66 Cong Ch 91 (1920) 41 Stat. 456 (Transportation Act of 1920). 38 ibid, § 5. 39 See eg Motor Carrier Act, 74 Cong Ch 498 (1935) 49 Stat 543. 40 See eg David Dayen, ‘How to Think about Breaking Up Big Tech’ (The Intercept, 1 April 2019) <https://theintercept.com/2019/04/01/elizabeth-warren-tech-regulation-2020/> accessed 15 October 2019 (explaining that Sen. Warren’s proposal to break up Big Tech in general, and Amazon in particular, ‘is merely following a long history’ that began with the Commodities Clause in the Hepburn Act). 41 See LC Marshall, ‘The Commodities Clause’ (1909) 17 J Pol Econ 448, 449–51; Thomas LeDuc, ‘Carriers, Courts, and the Commodities Clause’ (1965) 39 Bus Hist Rev 57 (describing Sen. Tillman’s role in shepherding the bill through the Senate and his failed efforts to ‘tighten’ the prohibition). 42 Marshall (n 41) 451 n10 (describing the Culberson substitute, which was rejected by the Senate committee of the whole by a vote of 62 to 11). 43 That is, railroads were allowed to continue producing and transporting coal they intended to consume themselves. 44 Much has been written about the so-called Commodities Clause. See eg Marshall (n 41) (summarizing the adoption of the clause and early Supreme Court precedent and questioning in conclusion ‘[w]hether that clause, as popularly interpreted, employed a wise method’); Eliot Jones, ‘The Commodity Clause Legislation and the Anthracite Railroads’ (1913) 27 QJ Econ 579 (decrying early Supreme Court precedents narrowly interpreting the Act but nonetheless concluding that effective enforcement of the Clause against anthracite railroads would not restore competitive conditions, but instead would induce the newly independent miners to organize ‘a coal trust, or at least some form of agreement among the coal companies to restrict output or to fix prices’, which would ultimately lead to additional regulation up to and including nationalization). 45 See eg Marshall (n 41); LeDuc (n 41) (tracing the history of the clause, noting that for its first thirty years ‘the government invoked the commodities clause of the Hepburn Act only against anthracite carriers’, and that the failed attempt to expand its scope in the New Deal-era U.S. Steel case effectively spelled its end). 46 See Edwin C Goddard, ‘The Commodities Clause of the Hepburn Act’ (1915) 14 Mich L Rev 49, 51 (note and comment) (noting that ‘[a]fter each of these decisions the coal-carrying roads were busily engaged in rebuilding their corporation plant so as to make it conform to the requirements laid down by the court’ and describing various partial spin-offs and leasing maneuvers). 47 See United States v Elgin, Joliet & Eastern Ry Co (1936) 298 US 492; see also ibid 512 (Stone, J, dissenting) (observing that ‘[i]f the commodities clause permits control such as is exhibited here, one is at a loss to say what scope remains for the operation of the statute’); United States v South Buffalo Ry Co (1948) 333 US 711; LeDuc (n 41) 72 (recounting these cases as the only times the commodities clause was asserted against an industrial firm that had vertically integrated by purchasing a common carrier). 48 See LeDuc (n 41) 59–60 (‘Doubt may be expressed whether the drastic prohibition embodied in the commodities clause was needed to protect either the consumer or the remaining independent producers. Existing law, had it been enforced, afforded adequate remedies…. [E]vents had already shown that the Sherman Act could be effectively invoked against the railroads, and later suits would show that it provided the chief weapon in dealing with the anthracite carriers.… As will later be apparent, the courts in deciding the anthracite cases relied more on the antitrust laws than on the Hepburn Act.’); Eugene V Rostow and Arthur S Sachs, ‘Entry into the Oil Refining Business: Vertical Integration Re-Examined’ (1952) 61 Yale LJ 857, 875 (arguing ‘the primary basis for the Court’s determination [in the first Anthracite case] was its construction of the Sherman Act, without reference to the policy of the commodities clause’). But see Robert Bork, ‘Vertical Integration and the Sherman Act: The Legal History of an Economic Misconception’ (1954) 22 U Chi L Rev 157, 168–69 (arguing that in the last two Anthracite cases, ‘[t]he evasions of the law were not evasions of the Sherman Act but of Pennsylvania law and the ‘Commodities Clause’ of the Interstate Commerce Act’). 49 See Jones (n 44) 600, 605 (describing early efforts to amend the Commodities Clause after Supreme Court decisions). 50 See eg JY Smith, ‘Ex-Southern Railway Head D.W. Brosnan Dies’ The Washington Post (Washington, DC, 19 June 1985) <https://www.washingtonpost.com/archive/local/1985/06/19/ex-southern-railway-head-dw-brosnan-dies/f88678b9-ef01-487b-8ae8-965f3a491364/> accessed 15 October 2019 (noting that the ‘Big John’ cars developed for Brosnan were ‘capable of carrying more than 100 tons, twice the capacity of earlier grain cars’). 51 Grain in Multiple-Car Shipments—River Crossings to the South (1963) 318 ICC 641, 683; Grain in Multiple-Car Shipments—River Crossings to the South (1965) 325 ICC 752, 758–59, 770–76; ‘Big John: ICC Cuts Against the Grain’ Railway Age (22 July 1963) 32; Alfred E Kahn, The Economics of Regulation: Principles and Institutions (MIT Press 1970; reprinted 1988) vol 1, 165 n12 [hereinafter, Kahn vol 1]; Alfred E Kahn, The Economics of Regulation: Principles and Institutions (MIT Press 1971; reprinted 1988) vol 2, 23 [hereinafter, Kahn vol 2]. 52 See Arrow Transp Co v Southern Ry Co (1963) 372 US 658, 659–62 (summarizing the four-year history). 53 See National Academy of Sciences, Innovation in Transportation: Proceedings of a Workshop (1980) 116–17 (‘While there has never been an explicit accounting from the Southern, it is estimated that the cost related to the workings of the regulatory scheme were in the range of $20 to $40 million in the case of Big John’.) 54 ibid 117. 55 Ingot Molds, Pa to Steelton, Ky (1965) 326 ICC 77, 82, 85; see also Kahn vol 1 (n 51) 165 n12; Kahn vol 2 (n 51) 24. 56 American Commercial Lines, Inc v Louisville & Nashville R Co (1968) 392 US 571. 57 ibid 593–94. 58 A Scheffer Lang, ‘Point of View: For Congress and the Railroads, a Last Chance’ The New York Times (New York, 10 January 1971) NER10 <https://www.nytimes.com/1971/01/10/archives/for-congress-and-the-railroads-a-last-chance.html> accessed 15 October 2019. 59 ibid. 60 See eg ‘The I.C.C. Dies’ The New York Times (New York, 3 January 1996) A14 <https://www.nytimes.com/1996/01/03/opinion/the-icc-dies.html> accessed 15 October 2019. 61 James M MacDonald and Linda C Cavalluzzo, ‘Railroad Deregulation: Pricing Reforms, Shipper Responses, and the Effects on Labor’ (1996) 50 ILR Rev 80, 90 (concluding that because of deregulation, ‘shippers switched to low-cost methods of transport, productivity measures grew sharply, shipping rates fell, and carrier profits grew’). 62 Wesley W Wilson, ‘Market-Specific Effects of Rail Deregulation’ (1994) 42 J Indus Econ 1, 20 (‘The evidence suggests that the majority of commodities prices initially rose under deregulation, reflecting greater market power and modest cost savings. By 1988, however, deregulation produced lower prices in most commodity classifications and did not increase prices in other classifications, suggesting that advances in productivity have dominated any adverse market power effects…. With price decreases and cost savings from deregulation, welfare gains from deregulation are likely positive.’). 63 See MacDonald and Cavalluzzo (n 61) 83 (‘Railroads moved aggressively in the post-Staggers [Act] era to offer volume discounts, and particularly large discounts to unit trains, to shippers of ‘bulk’ products, such as grain, lumber, coal, and other minerals. Because shippers must invest in specialized loading and storage facilities for unit trains, they will not do so without lower rates or improved services.’); Stone (n 25) 52 (‘Other ICC rate policies forced the railroads to postpone the use of unit trains (a whole trainload of cars permanently coupled and shuttling back and forth from producer to consumer – most frequently from coal mines to power plants) in the East for years.’). 64 See Submission of the United States to the Ibero-American Competition Forum (September 2007) 2–3 <https://www.ftc.gov/system/files/attachments/us-submissions-oecd-2010-present-other-international-competition-fora/ibero-trucking.pdf> accessed 15 October 2019 (citing Letter from Thomas Carter, Regional Director, Federal Trade Commission, to Raymond Bennett, Transportation/Gas Utilities Division, Railroad Commission of Texas (2 October 1989)). 65 ibid 4 (citing Clifford Winston, ‘U.S. Industry Adjustment to Economic Deregulation’ (1998) 12 J Econ Persp 89, 94). 66 ibid 5 (citing Stephen Morrison and Clifford Winston, ‘Regulatory Reform of U.S. Intercity Transportation’ in Gomez-Ibanez and others (eds), Essays in Transportation Economics and Policy (Brookings Institution Press 1999) 477). 67 ibid 8. 68 Milton Friedman and Rose Friedman, Free to Choose: A Personal Statement (Harcourt Brace 1980, reprinted 1990) 201. 69 ‘I.C.C. Dies’ (n 60). 70 See Fed Aviation Admin., United States Dep’t of Transportation, A Brief History of the FAA <https://www.faa.gov/about/history/brief_history/> accessed 15 October 2019 (tracing the history of US air regulation). 71 See above nn 11–12 and accompanying text. 72 The legislation initially created two separate regulators, the Civil Aeronautics Authority and the Air Safety Board. These agencies were reorganized in 1940, with many functions transferring to the newly-created CAB. See FAA (n 70). 73 Civil Aeronautics Act of 1938, formerly codified at 49 U.S.C. § 1302d (1964). 74 See Kennedy (n 12) 614–15 (describing how ‘entry into the industry has been effectively blocked’ by the CAB’s route entry policy). 75 See Civil Aeronautics Board Special Staff on Regulatory Reform, Regulatory Reform: Report of the C.A.B. Special Staff (July 1975) 49 n1 (on file with authors) [hereinafter Report of the CAB Special Staff] (‘According to information supplied by the Board to the Subcommittee on Administrative Practice and Procedure …, 94 applications for trunkline authority have been received from outsiders since 1950, and none has succeeded.’); Bradley Behrman, ‘Civil Aeronautics Board’ in James Q Wilson (ed), The Politics Of Regulation (James Q. Wilson ed., Basic Books 1980) 88 (‘Between 1950 and 1974, for example, the Board turned down all ninety-four of the applications that would-be entrants to the industry submitted in pursuit of domestic ‘trunk-line’ authority.’). 76 See John E Robson, ‘Airline Deregulation: Twenty Years of Success and Counting’ (Spring 1998) Regulation 17-8 (‘In a widely-cited example of the CAB at its worst, it took the board eight years to give Continental Airlines permission to fly between San Diego and Denver.’). 77 Continental Air Lines, Inc v Civil Aeronautics Board (DC Cir 1975) 519 F 2d 944, 959–60. 78 Kahn vol 2 (n 51) 210–11. 79 ibid 212 (citing Ronald E Miller, Domestic Airline Efficiency: An Application of Linear Programming (MIT Press 1963) 108–14). 80 Behrman (n 75) 98 (‘[I]n 1971 the CAB announced that all subsequent proposals for fare adjustments would be evaluated according to whether airlines would earn a reasonable return if they attained an average load factor of 55 percent.’ (citing CAB Order 71-4-54 (9 April 1971)); David B Richards, ‘Did Passenger Fare Savings Occur After Airline Deregulation?’ (2007) 46 J Transp Res Forum 73, 78 (describing the Domestic Passenger Fare Investigation pricing model developed by the CAB in 1974, and administered by Richards himself, which required ‘normal coach fares … to be offered on a formula rate’ based, among other things, ‘on a 55% full-fare load factor’ (emphasis omitted)). 81 Madhu Unnikrishnan, ‘A Law that Changed the Airline Industry Beyond Recognition (1978)’ Aviation Week (4 June 2015) <https://aviationweek.com/blog/law-changed-airline-industry-beyond-recognition-1978> accessed 15 October 2019 (chateaubriand and piano lounges); Behrman (n 75) 93 (alcoholic drinks); see also Kahn vol 1 (n 51) 211 (‘In part because the doors to price competition are closed, airline companies compete very strenuously among themselves in the quality of service they offer’.). 82 Trans World Airlines Siesta Sleeper-Seat Service (1958) 27 CAB 788, 790 (1958) (Opinion); see also Kahn vol 2 (n 51) 215 (discussing same). 83 Kahn vol 2 (n 51) 214 (citing Richard E Caves, Air Transport and Its Regulators (Harvard UP 1962)). 84 ibid 212. 85 See Behrman (n 75) 91–94 (citing William A Jordan, Airline Regulation in America: Effects and Imperfections (1970) 279–81, 285); Simat, Helliesen, and Eichner, Inc., An Analysis of the Intrastate Air Carrier Regulatory Forum (January 1976) vol 1, App Ex 1 (prepared for the US Department of Transportation)) (listing examples in California and Texas where PSA and Southwest, intrastate carriers not regulated by the CAB, competed on the same routes as CAB-regulated interstate carriers but charged much lower fares); see also Michael E Levine, ‘Is Regulation Necessary? California Air Transportation and National Regulatory Policy’ (1965) 74 Yale LJ 1416, 1430–39 (note) (providing examples from California, where intrastate carriers entered the Los Angeles–San Francisco route at prices well below the CAB-certified interstate carriers and where the unregulated price in 1965 ($11.43 one way) remained far below the CAB-set price on the otherwise comparable Boston-Washington DC route ($24.65 one way)). 86 Clifford Winston, ‘U.S. Industry Adjustment to Economic Deregulation’ (1998) 12 J Econ Persp 89, 89. 87 See above n 12 and accompanying text. 88 Report of the CAB Special Staff (n75) 252; see also ibid, Executive Summary 3, attached to Letter from Roy Pulsifer, Special Advisor to the CAB on Regulatory Reform, to the CAB, 22 July 1975 (on file with authors) (‘[I]t is possible that regulation will become more restrictive and thereby increase inefficiencies. Real improvement in economic efficiency can come only if protective regulation is eliminated or materially reduced.’). 89 ibid, Executive Summary 1–2. 90 Behrman (n 75) 104; Robson (n 76) 17–18 (‘The watershed reform came in April 1976 when the CAB unanimously announced its support for deregulation.’). 91 Pub L 95-504 (1978) 92 Stat 1705. 92 Irvin Molotsky, ‘C.A.B. Dies After 46 Years; Airlines Declared ‘on Own’’ The New York Times (New York, 1 January 1985) 9 <https://www.nytimes.com/1985/01/01/us/cab-dies-after-46-years-airlines-declared-on-own.html> accessed 15 October 2019 (describing the final closure of the CAB, an event that ‘was mandated in the airline deregulation enacted in 1978’). 93 Steven A Morrison and Clifford Winston, ‘The Dynamics of Airline Pricing and Competition’ (1990) 80 Am Econ Rev 389, 390 (‘On average, deregulated fares are lower than regulated fares by 18 percent, amounting to an average annual savings to travelers of roughly $6 billion (1988 dollars).’). 94 US General Accounting Office, Domestic Aviation: Changes in Airfare, Service, and Safety Since Airline Deregulation (25 April 1996, GAO/T-RCED-96-126) 4. 95 See ibid (‘The overall trend toward lower fares since deregulation has resulted in large part from increased competition, spurred in many cases by the entry of new airlines. The average number of large airlines serving the medium-sized-community airports in our sample, for example, increased by over 50 percent between 1978 and 1994, while the average number of commuter carriers serving these airports increased by about 40 percent.’); US Government Accountability Office, Airline Deregulation: Reregulating the Airline Industry Would Likely Reverse Consumer Benefits and Not Save Airline Pensions (June 2006, GAO-06-630) 4 [hereinafter, GAO Airline Deregulation] (‘As predicted by the framers of deregulation, airline markets have become more competitive and fares have fallen since deregulation. For consumers, airfares have fallen in real terms since 1980 while service has generally improved. Overall, median fares have declined in real terms by nearly 40 percent since 1980.’); US General Accounting Office, Airline Deregulation: Changes in Airfares, Service Quality, and Barriers to Entry (March 1999, GAO/RCED-99-92) 4 (‘Since deregulation, numerous new airlines have started operations, while established airlines have expanded into new markets.… In 1990, we reported that from 1979—the earliest year for which reliable data on fares were available—through 1988, the average fare per passenger mile, adjusted for inflation, declined by 9 percent at airports serving small communities, 10 percent at airports serving medium-sized communities, and 5 percent at airports serving large communities.’). 96 Stephen Breyer, ‘Airline Deregulation, Revisited’ Bloomberg (New York, 20 January 2011) <https://www.bloomberg.com/news/articles/2011-01-20/airline-deregulation-revisitedbusinessweek-business-news-stock-market-and-financial-advice> accessed 15 October 2019. 97 ibid. On Breyer and airline deregulation, see Robert H Nelson, ‘The Economics Profession and the Making of Public Policy’ (1987) 25 J Econ Lit 49, 62, and Breyer (n 11) chs 11 and 16. 98 Severin Borenstein, ‘The Evolution of U.S. Airline Competition’ (1992) 6 J Econ Persp 45, 45. 99 US General Accounting Office, Airline Deregulation: Changes in Airfares and Service at Buffalo, New York (20 September 1999, GAO/T-RCED-99-286) 1. 100 Open Markets Institute, Airlines & Monopoly <https://openmarketsinstitute.org/explainer/airline-monopoly/> accessed 15 October 2019 (‘Indeed, after adjusting for changes in energy prices, a 1990 study by the Economic Policy Institute concluded that airline fares fell more rapidly in the 10 years before 1978 than during the subsequent decade. Another study finds airfares would have fallen still faster after 1978 if, instead of letting carriers charge whatever the market would bear, the CAB had been allowed to continue enforcing its long-standing formulas for setting maximum fares.’ (citing Dempsey and Richards studies, discussed below)). 101 Paul Stephen Dempsey, ‘Airline Deregulation and Laissez-Faire Mythology: Economic Theory in Turbulence’ (1990) 56 J Air L & Comm 305. 102 Most glaringly, economists have pointed out that ‘the significant weakness’ in Dempsey’s analysis is that he controls for only two factors, inflation and fuel prices, whereas the ‘Morrison and Winston study does a much more complete comparison and finds substantial price decreases relative to regulation’. Borenstein (n 98) 46 n1. Dempsey controls only for fuel and inflation despite acknowledging that fuel accounts for a relatively small portion of overall airline costs and criticizing the Morrison and Winston study because it ‘does not hold all other factors constant’. ibid 356 (emphasis in original); see ibid 350 (recognizing that ‘fuel constituted anywhere from twelve percent of costs … to thirty percent’ during the regulatory period); see also Richards (n 80) 75 Fig. 3 (showing that non-fuel costs rose significantly after 1979 while fuel costs remained relatively flat until 2004). Because the proxy that Dempsey constructs does not control for all of the factors he deems relevant, let alone estimate the air fares that would have been set had regulation continued, it therefore cannot support his claim that ‘[t]he twenty-eight percent fall in real yields that has occurred since deregulation would have occurred under regulation as well’. ibid 355. In addition, Dempsey takes the same aggressive tack as Richards, counting the significant benefits of limited deregulation between 1974 and 1979 as part of the ‘regulation’ sample and incorporating the large price reductions during that era in his trend line. See ibid 348–54 and 353 tbl. 1. As explained below at nn 111–112, significant deregulation of fares began in 1974, and indeed Dempsey’s data shows real yields fell 20 per cent between 1974 and 1979. ibid 353 tbl. 1 (from 116.2 to 94.2). 103 Airline Competition: Special Hearings Before the S. Comm. on Appropriations, 105th Cong. 56 (1998) (statement of Paul Dempsey, Vice Chairman, Frontier Airlines and Professor, University of Denver/College of Law) <https://www.govinfo.gov/content/pkg/CHRG-105shrg53117/html/CHRG-105shrg53117.htm> accessed 15 October 2019. 104 ibid 56–57. 105 ibid 64 (Preamble to Dempsey’s proposed ‘Airline Competition Act of 1998’). 106 David Morris, ‘Airline Deregulation: A Triumph of Ideology Over Evidence’ Huffington Post (New York, 13 December 2013, updated 19 April 2017) <https://www.huffpost.com/entry/airline-deregulation-ideology-over-evidence_b_4399150> accessed 15 October 2019 (citing the 1990 Dempsey paper as primary evidence for the argument that ‘regulation worked better’ than competition); Open Markets Institute (n 100) (citing Dempsey paper to support the claim that CAB regulation ‘on balance … worked well’ because ‘airline fares fell more rapidly in the 10 years before 1978 than during the subsequent decade’). 107 Richards (n 80) 87. 108 ibid 85. 109 ibid 87. 110 ibid 80 tbl. 2 (estimating a total weighted average ‘ratio of average fares to SIFL’ of 0.861 in 1979 and 0.726 in 2005, representing a decline in the index of more than 15 percent); ibid 81 (explaining that ‘any ratio less than those shown in 1979’—such as the 0.726 in 2005 versus 0.861 in 1979—‘indicates a general reduction in fare level’). 111 See eg Winston (n 86) 89 (‘Economic deregulation of the airline industry began in 1974 when the Civil Aeronautics Board first encouraged experiments with discount fares. It was completed in 1983 when all regulations on fares, entry, and exit were eliminated.’). 112 Severin Borenstein and Nancy L Rose, ‘How Airline Markets Work…or Do They? Regulatory Reform in the Airline Industry’ in Nancy L Rose (ed), Economic Regulation and Its Reform: What Have We Learned? (NBER 2014) 63, 78 (discussing the Richards paper and noting that Richards chose as his regulated reference point a date by which ‘significant relaxation of fare controls had already occurred’). 113 GAO Airline Deregulation (n 95) ‘Highlights’. 114 Borenstein and Rose (n 112) 63, 78 (finding ‘actual fares were about 26 percent lower than SIFL formula fares in 2011, suggesting a consumer welfare increase in the range of $31 billion in that year’ and noting factors suggesting that this figure may be either understated or overstated); see also GAO Airline Deregulation (n 95) 24 (summarizing papers by (i) Winston and Morrison (1995) and (ii) Rose and Borenstein (2005) that estimated deregulation reduced average airfares by between 22 and 33 per cent). 115 See above nn 34–37 and accompanying text. 116 See above nn 81–84 and accompanying text. 117 See eg D Daniel Sokol, ‘Explaining the Importance of Public Choice for Law’ (2011) 109 Mich L Rev 1029, 1033. This principle particularly holds true in the area of competition policy. See William E Kovacic, ‘The Institutions of Antitrust Law: How Structure Shapes Substance’ (2012) 110 Mich L Rev 1019. 118 Breyer (n 11) 184. 119 ibid (emphasis in original). 120 ibid. 121 ibid (emphasis in original). 122 ibid 185. 123 See Cyril Ritter, The Interface between Competition and Regulation in EU Law (paper presented at the Antitrust Enforcement Symposium 2019, CCLP(S)165) 21–22 (on file with authors) (‘As regards regulation, one could consider a five-part test for determining whether regulation may be appropriate to address a particular issue, (a) There is a clearly definable issue …. (b) The issue is widespread or recurrent … (c) The issue is usually bad, or always bad, so that it does not require a case-by-case assessment … (d) The regulatory rule takes competition principles into account … (e) The regulatory framework would implement a reliable enforcement mechanism …’. (emphasis in original)); see ibid 3 (noting that ‘[t]he debate on the respective merits and roles of competition enforcement and regulation is particularly intense when it comes to digital matters’). 124 Kevin Carty, Leah Douglas and Lina Khan, ‘6 Ideas to Rein in Silicon Valley, Open Up the Internet, and Make Tech Work for Everyone’ NY Magazine (11 December 2017) <http://nymag.com/intelligencer/2017/12/open-markets-institute-antitrust-for-silicon-valley.html> accessed 15 October 2019 (Idea #6, authored by Lina Khan). 125 Warren (n 1). 126 Furman Report (n 4) paras 2.4–2.5 (‘Some assessments see digital markets as raising no particular issues and needing nothing beyond existing competition frameworks to continue delivering the benefits of well-functioning markets. Others have argued they are natural monopolies that need regulating in the same way as electricity, gas or railway networks. The Panel’s assessment is that, while they share some important characteristics with natural monopolies, it is too early to conclude that competition within and for digital markets cannot be achieved.’). 127 Railroad Problem (n 8). 128 Khan (n15). 129 ibid 1085. 130 Carty, Douglas and Khan (n 124) (Idea #5, Leah Douglas) (‘Just as the giant corporation has used its power to engage in predatory pricing and to avoid paying sales tax to drive thousands of retail stores across America out of business, [Amazon] could now do the same to many local and regional groceries. This would result both in greater concentration of power over food retailing, and even fewer physical stores…. To prevent these harms, Amazon should not only be blocked from future grocery acquisitions but its purchase of Whole Foods should be unwound. And while regulators at the Federal Trade Commission are taking care of this business, they should also ban Amazon from engaging in any price discrimination in food products, anywhere, ever. Without these safeguards, we risk handing over a huge swath of our food economy to one giant corporation, and having that giant harm our well-being in fundamental ways.’). 131 ibid. 132 Open Markets Institute (n 100). 133 See eg Dudley (n 17) 8 (noting that the author’s flight to Massachusetts for the weekend ‘would have been an unthinkable luxury’ during the CAB era because back then ‘air travel was reserved for businessmen or the wealthy’). 134 See above nn 30–31 and accompanying text. 135 Railroad Problem (n 8). 136 See above n 130 and accompanying text. Published by Oxford University Press 2019. This work is written by US Government employees and is in the public domain in the US. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Journal

Journal of Antitrust EnforcementOxford University Press

Published: Oct 1, 2008

There are no references for this article.