Also, as a matter of procedure, even if you had a compromising Civil Aeronautics Board, five economists whose sole objective was to maximize economic efficiency, to the extent they would be changing precedent and moving in very new directions, I would imagine there would be a significant protest of such activity. I am not an expert on law, but I am not confident this kind of activity would survive court

tests.

Senator KENNEDY. Let me just ask you from the point of view of the Council whether you feel from that vantage point that the idea of greater competition in terms of rates and entry would be beneficial from an overall economic point of view for our economy or from the consumer point of view.

Mr. MILLER. I think it is the major objective of the Council to recummend policies or present analyses which focus on economic efficiency questions as they pertain to regulation and other types of governmental controls. This, of course, counts the consumer to a very large measure. After all, Senator, the ultimate objective of production is consumption. Senator KENNEDY. Could you develop your thoughts briefly on the question of cross-subsidization. Then secondly, it is really an unrelated question, with this proliferation of new airlines and different prices, how will the consumer know where to go to get a ticket, will we have mass confusion or do you think this is a manageable problem Would you talk on those two questions?

CHAOS

Mr. MILLER. I will be glad to. Let me take the latter first. Information is a scarce resource, and when consumers know a producer charges a specific rate, and these rates are not likely to be changed, the producer's reputation is enhanced. Consumers find out this information; producers advertise as well. In a freely competitive airline market, consumers would learn the prices that are charged and the carriers that provide good service-the same way as in existing competitive, unregulated markets.

CROSS-SUBSIDY (SMALL TOWN SERVICE)

Now, on the question of cross-subsidy, we essentially model the airlines in the following way: Because there is a lot of competition todaymost markets are served by two or more carriers the Board is really in the business of setting the quality of service. They set the price and carriers compete on the frequency of schedules and other means so as to bring the cost level up to the price level. For example, transcontinental markets were referred to a few minutes ago where the prices are such that carriers can break even at 40 to 45 percent load factors. Carriers do, in fact, schedule up to the point where they are breaking even at those load factors. If the Board lowered the price, then the carriers' break even point would be higher. They would restrict scheduling. This would drive up load factors. So in these markets the carriers are earning neither excess profit nor losses.

On short-haul markets where the fare taper is such that the pricevis-a-vis the cost of a constant load factor service is too low, the carriers simply cut back on scheduling and provide fairly poor service.

Again, in these markets the carriers tend to earn neither excess profits nor losses.

So our general conclusion, based on this and additional information, is that there is not much cross-subsidy actually taking place. Senator KENNEDY. Very fine. Thank you very much.

[The prepared statement submitted by Mr. Miller for himself and Mr. Seevers follows:]

PREPARED STATEMENT OF GARY L. SEEVERS, MEMBER, COUNCIL OF ECONOMIC ADVISERS AND JAMES C. MILLER, SENIOR STAFF ECONOMIST

Economic Effects of Regulation of the Domestic Air Carriers by the CAB

Mr. Chairman, members of the Committee: We are pleased to appear before you today to discuss the economic effects of regulation of the domestic air carriers by the U.S. Civil Aeronautics Board.

For a number of years the Council has questioned the efficacy of airline regulation and suggested certain reforms. As a matter of fact, our most recent economic report,* published just two days ago, contains a section which discusses airline regulation and which implies that certain reforms are needed. A copy of that section of the report is attached as an appendix to this testimony.

In the remainder of this testimony we outline what we consider to be the major costs of CAB regulation, we examine the economic performance of a hypothetical deregulated air carrier market, and we indicate the kinds of regulatory reforms the Administration now has under review.

THE COSTS OF AIRLINE REGULATION

As they pertain to the airlines, economic conditions today are very much different from what they were in 1938, when airline regulation was established. At that time, the U.S. Government was attempting to promote an “infant” industry through an inefficient system of airmail subsidy. Basically, the Government granted contracts to air carriers and prevented competition on those routes where contracts were granted. Recognizing the potential for excess profits on passenger services then or in the future, carriers would "buy-in" on these contracts for extremely low rates.1 This perfectly rational economic behavior on the part of the air carrier firms was then cited as evidence of "destructive competition" in the airline industry and thus a need for governmental intervention to "rationalize" competition. It was also said that governmental controls were needed to assure safety of operations.

Today, the domestic airline industry is no longer an infant industry in need of promotion; having increased in size since 1938 some 250-fold, by most standards it is now truly "mature". Mail contracts are no longer the vehicle for subsidy, and as a percent of total domestic revenue subsidy has declined from 31.6 percent in 1939 to less than 1 percent today. Except for minor payments to Northeast Airlines in the mid-1960's, the trunk carriers have been completely off subsidy since 1959.3 Air safety, which until 1958 was a primary CAB concern, is now vested with the Federal Aviation Administration of the Department of Transportation (DOT).*

Another important change in the nature of the industry and its regulation is that the principal city-pair markets today are served by two or more airlines.5

Economic Report of the President, Washington, GPO, February 4, 1975.

1 See Richard E. Caves, Air Transport and Its Regulators: An Industry Study, Cambridge, Harvard University Press, 1962, p. 124.

2 Note that the legislative "Declaration of Policy" admonishes the Board to create "Competition [only?] to the extent necessary to assure the sound development of an airtransportation system .. [Original (1938) language now contained as section 102(d) of the Federal Aviation Act of 1958, as amended.]

3 For a discussion of the existing subsidy mechanism and its deficiencies, see George C. Eads, The Local Service Airlines Experiment, Washington, The Brookings Institution, 1972.

Responsibility for investigating air accidents was transferred from the CAB to the National Transportation Safety Board in 1966.

5 Note that from 1955 to 1971 the percentage of total revenue passenger miles attributed to markets where 2 or more carriers each accounted for at least 10 percent of the market rose from 55.6 percent to 76.6 percent. See, George W. Douglas and James C. Miller III, Economic Regulation of Domestic Air Transport: Theory and Policy, Washington, The Brookings Institution, 1974, p. 114.

Thus, in addition to no longer regulating so as to promote an infant industry and no longer regulating air safety, the Board finds itself no longer preoccupied with regulating monopoly. Instead, its primary activity is regulating competition. But under CAB regulation this competition is of a rather special sort: It is manifest almost totally in dimensions other than price.

A potential price-cutter in a CAB-regulated market faces significant costs in carrying out such an initiative. First, the carrier must announce the new rate at least 30 days in advance, thereby alerting its competition to the intended action.* Second, there is the simple cost of publishing the new tariff with the Board, as legally required. Third, any fare decrease is likely to be protested by competitors as being unreasonably low, discriminatory, preferential, prejudicial, or simply an instance of "unfair competition." Thus, the price-cutter nearly always must make an affirmative case before the Board that the new rate is justified, and this, of course, costs money.

The new rate may be rejected outright or set down for investigation. If it is rejected, then of course any advertising by the carrier about the prospective lower rates is lost, and perhaps on balance creates ill will because the carrier is unable to deliver. If the rate is suspended, the carrier may either withdraw the initiative or pursue it further. If the rate reduction is pursued, then significant procedural costs must be absorbed by the initiating carrier as the rate travels through various steps: prehearing conference, hearing, briefs to the Administrative Law Judge, possibly briefs to the full Board, oral argument, and possibly, in the end, even court challenges.

Because of these impediments, one observes little price competition in the airlines. When rates are lowered they are usually done so in the interest of the whole industry, as for example, introducing discount fares to enlarge total revenues and to make the airlines more effective in competing with other modes of common-carrier transportation, such as intercity buses. Another ramification of this constraint system is that from the standpoint of an individual airline it makes little sense to change the rate in just one market. In an attempt to "spread the cost," airlines which propose rate reductions usually do so on fairly large chunks of traffic, although such a strategy inevitably reduces the likelihood of ultimate approval by the Board and of course raises litigation costs.

The Board's statutory authority to control price and its procedures for implementing that control thus have rendered price competition in the airlines all but non-existent. On the other hand, there are other means that carriers have for attracting and competing for passengers over which the Board exercises little or no direct control. Such non-price competition takes various forms, including costlier meals, "free" drinks, expensive advertising, flashy interior and exterior color schemes, "VIP" airport lounges, on-board lounges, pianos, bars, and the like. But much more important, in terms of its ultimate cost to the consumer, is the scheduling form of non-price competition. As will be discussed below, scheduling additional flights is the most effective means that individual carriers have of attracting additional passengers. Notably, except for its power to grant antitrust immunity and to orchestrate capacity agreements among carriers, the Board is prohibited from controlling schedule competition."

For any price that is approved or, in essence, "set" by the Board, the market has a "break-even" load factor, which we define so as to include a normal return on investment. If actual load factors are below the break-even level, the carriers will be earning less than a normal profit, or even accounting losses and will cut back on capacity. Since market demand is inelastic with respect to capacity,' load factors will rise, and the process of capacity curtailment will continue until actual load factors have risen to the break-even level, at which point the incentive to reduce capacity unilaterally will disappear.10 On the other hand, if actual load factors exceed the break-even level, individual airlines have profit incentives to increase capacity. Actual load factors will fall and the process will

Thus, there is no such thing as a conventional "sale" in the airline business.

7 Section 401 (e) (4) of the Federal Aviation Act states that, "No term, condition, or limitation of a certificate shall restrict the right of an air carrier to add or change schedules.

8 "Load factor" is the proportion of seats filled, usually expressed as a percentage. That is, the percentage change in total traffic in the market is less than the percentage change in the market's total capacity.

10 Short of the break-even load factor an individual carrier can safety assume that it reduces capacity its competitors will also. However, once equilibrium has been reached, a carrier reducing capacity unilaterally may not assume that its competitors will do likewise.

continue until capacity increases have reduced actual load factors to break-even, at which point there is no more incentive to add to capacity."

An extremely important aspect of this non-price competition is that there is a whole range of prices which the Board may choose and still enable competitive returns to the individual carriers, or at least to the carriers as a group. If the Board chooses a "high" price, the break-even load factor will be "low" and, in equilibrium, so will be the actual load factor. If the Board chooses a "low" price, the break-even load factor will be "high" and, in equilibrium, so will the actual load factor. The nature of this trade-off between fare and average load factor is displayed in figure 1.12

(For illustration purposes only; not drawn to scale.)

11 When market load factors are above break-even, an individual carrier can make more profit by expanding its own capacity provided other carriers do not also expand their capacity. The evidence suggests that carriers act as though they make such an assumption. A variant explanation of observed behavior is that since an individual carrier may not assume that its competitors will not increase their capacity it must increase its capacity just to maintain its market share. In any event, when the break-even load factor is reached, there is no incentive to increase capacity, since the market load factor will fall below break-even and each carrier may assume that its competitors will follow a policy of restraint.

For a more thorough description of this nonprice-competing behavior and evidence on same, see Douglas and Miller, ibid., chapter 4 and the papers cited therein by De Vany, Douglas, Miller, Straszheim, Yance, Barnekov, Eads, Milward, and White. For an empirical analysis of the relationship between market shares and capacity shares, see James C. Miller III. "Airline Market Shares vs. Capacity Shares and the Possibility of Loss Equilibria" (processed, 1974) and CAB Docket 22908 (Capacity Reduction Agreements Case) DOT-T-1 through 5 (1974).

12 Notably, the Board would appear to have accepted this model of regulated-carrier behavior. In a decision in its recent domestic passenger fare investigation (CAB Docket 21866), the Board said:

"We find. that the higher the fare level in relation to cost, the more capacity carriers will offer and the lower load factors will be; and, conversely, the lower the fare level, the less capacity carriers will operate and the higher load factors will be." [CAB Order 714-54 (April 9, 1971), p. 23.]

Over a fairly wide range of prices, carriers, in equilibrium, will earn normal profita-and thus, arguably, the choice of price is not material to them. However, the passenger's cost of service is greatly dependent upon the price and load factor option chosen by the Board. In essence, the passenger's "fuli cost" of travel is the ticket price plus the "cost" of delays he, or she, ineurs in waiting for a flight. We see in figure 1 the rather obvious proposition that as the average load factor rises the associated break-even fare falis. If this were the only element in the passenger's cost of service, public policy would dictate a fare consistent with load factors of near 100 percent. However, as load factor falls delay cost increases. Passengers find it more difficult to secure accommodations on the desired departure and flights are fewer. with more time in between departures. When translated into money terms this delay cost is as characterized in figure 1. The passenger's full cost of service is thus the sum of these two types of cost, i.e., ticket price pins delay, and, given these two curves, for some average load factor level the "full cost" is at a minimum, i.e., ALF.

A recent Brookings publication by Professors Douglas and Miller came to the conclusion that the Board has chosen too low a load factor standard, i.e.. 55 percent as opposed to 60-65 percent, and consequently is promulgating fares which are too high. This means that the typical passenger is paying an "excess fare" which exceeds the value of the reduction in delay. This in turn means a higher full cost of service with no offsetting higher profits to carrier. Thus, there is regulation-induced excess capacity which represents a deadweight loss to society.” Douglas and Miller estimate that during 1969 air passengers paid excess fares to domestic trunk carriers ranging between $366 million and $538 million, for which they received quality improvements valued at between $118 million and $182 million. This leaves a deadweight welfare loss in trunkline service for 1969 of between $248 million and $356 million.

13

Since 1969 the Board has established target load factors of 55 percent as opposed to the then-prevailing levels of approximately 50 percent. However, the recent increases in fuel prices have raised the optimal average load factor to approximately 65-70 percent, so the present configuration of service is still characterized by efficiency costs on the same order of magnitude. Based on total domestic trunk revenues of $9,316 million for the year ending September 1973, this implies a current annual welfare cost for trunk service ranging between $355 million and $509 million.

18

There are additional costs of airline regulation. First, there is evidence that the relationship between the Board and the industry has resulted in a level, and structure, of fares which maximizes total capacity rather than one which maximizes total passenger traffic. This is illustrated in figure 2.1 Since some costs are "external" to the airlines and their passengers, this behavior has quite likely resulted in excessive investments in airport and airway facilities as well as excessive consumption of fuel.

18

Second, the Board's policy of protecting existing carriers from competition by preventing the entry of new carriers not only means that the public has been denied lower price-quality options, but that potentially more efficient carriers have not been able to test the efficiency of existing carriers. Whether new carriers would have significantly lower costs is subject to considerable debate, but evidence on relative carrier costs and the evidence from unregulated markets certainly raises this possibility. There are two significant problems with this approach, however. First, a regulator is inherently less capable of administering resources "correctly" than is an individual competitive entrepreneur. The regu

19

13 In the DPFI the Board announced its intention of in effect setting fares at levels which would cover costs (plus a reasonable return on investment) on the basis of an industry wide average load factor of 55 percent. 14 The analysis only briefly summarized here can be found in Douglas and Miller, ibid., chapter 6.

15 Miller and Douglas, ibid., p. 172.

18 See Arthur S. De Vany, "Effects of Price and Entry Regulation on Airline Output, Capacity and Efficiency," Bell Journal of Economics and Management Science, (forthcoming Spring 1975; and Douglas and Miller, ibid., pp. 60 and 176-77.

17 Rather than choosing fare level F*, the Board has chosen fare level F**. Figure 2 is adapted from De Vany, ibid.

18 Since regulation was established in 1938, not a single new trunk carrier has entered the market, and not a single trunk has exited the market except through merger.

19 See Robert J. Gordon, "Airline Costs and Managerial Efficiency" in Transportation Economics: A Conference, Columbia University Press for the National Bureau of Economic Research, 1965, pp. 61-94; Theodore E. Keeler, "Airline Regulation and Market Performance," Bell Journal of Economics and Management Science, Autumn 1972, pp. 399-424; William A. Jordan, Airline Regulation in America: Effects and Imperfections, Baltimore, The Johns Hopkins Press, 1970, chapter 11; and Douglas and Miller, ibid., pp. 141-9.