then, both these and other reasons may be applicable. It is only proper to point out that the CAB lowered its approved rates starting in 1962 and continuing to 1967-68, before inflationary pressures reversed the trend. Round-trip passenger rates, for example, fell to 1.75 cents per passengermile for standard jets and 1.66 cents per passenger-mile for stretched jets. With some adjustments for the relatively low inflation of that 8-year period, it can be seen that these rates are significantly lower than the competitive-bid rates of 1959. It should also be recognized, however, that these rates were achieved with unusually high traffic flows over extremely long-hauls to Vietnam and other points in the Pacific, and that jet aircraft have experienced substantially lower operating costs than the piston aircraft that provided most contract services prior to the early 1960's. With regards to this latter point, it should also be noted that CAB-approved rates for round-trip passenger service on piston aircraft have never fallen below 2.55 cents per passenger-mile." Summary It is remarkable how consistent the evidence from these four cases has been. The large amount of cross-sectional evidence from the California experience is supported by similar evidence from Texas and Canada, as well as by time-series evidence from the most recent years in California, from Canada, and from military airlift procurement. Where entry has been possible and airlines have been able to introduce new or lower fares without regulatory constraint, airline fares have been low. Where regulation has served to close entry and fares have been controlled, fares have been much higher. Indeed, in most cases regulated fares have been 50 percent or more above nonregulated fares, and differences of over 100 percent have been common. These consistently large fare differences are not explained by factors such as whether, airport congestion, traffic density, and so forth. Of course, cold and snow for 5 months or so each year increase the resources required for ground handling ; and fog, snow and rain delay flights at various times, as does airport congestion. These factors all increase costs somewhat, but not by 50 to 100 percent over costs in nonregulated city-pairs which also experience weather and congestion problems. However, should weather and airport congestion influence costs appreciably in particular interstate city-pairs, and if CAB-authorized fares are supposed to be cost based, why aren't CAB-authorized fares lower for city-pairs where weather is generally good and airports are free of congestion? The fact is that these cost differences are relatively small and thus it is infeasible to adjust the present rigid fare structure to reflect them. Traffic density is certainly a relevant factor in average costs per passengermile and thus fares. This was reflected in the higher fares charged by the intrastate carriers in minor California city-pairs as opposed to the major city-pairs. But beyond a certain threshold level of, say, 100,000 true O&D passengers per year, differences in costs per passenger-mile are small. Here again, it should be noted that the CAB makes no allowances for traffic density in its basic fare formula. In the new cost-based coach-fare formula to be implemented on April 29, 1975, the CAB has established a terminal charge of $13.85 and various line-haul charges starting at 7.79 cents per passenger-mile for the first 500 miles.56 Implicit in this "cost-based" formula is the belief that it costs the average CAB-regulated airline $13.85 merely to attract a passenger, process his reservation and issue a ticket, enplane him, takeoff and then immediately land at the same or an adjacent airport, and then deplane the passenger. All this regardless of whether the airport is large, small, congested, has bad weather, or whatever. In sharp contrast PSA will do all this plus fly a passenger 159 miles (Fresno-San Francisco) for $12.73, that is, for $1.12 less than the CAB's terminal charge.57 Similarly, in Texas, South 53 CAB, supra, note 1, at 597-99. 54 Id. The current round-trip passenger rate for jet aircraft is 2.848 cents per passengermile. CAB press release 75-15 (Jan. 17, 1975). 55 The threshold traffic level for a city-pair depends on many related factors such as distance, aircraft size required to operate efficiently over that distance, existence of alternative transportation services, the number of other flights serving each airport in the citypair, the amount of complementary connecting traffic, and so on. 100,000 true 0&D passengers per year would yield an average of 274 per day which, alone, would provide a 60-percent load factor for four one-way flights provided with 115-seat aircraft. In 1972, the 100th largest city-pair was New York-Norfolk with 186,000 true O&D passengers (appendix B). 56 CAB order 74-12-109 (Dec. 27, 1974). 57 PUC decisions 83814 (Dec. 10, 1973), 83918 (Dec. 30, 1974) and 83939 (Dec. 30, 1974). west Airlines will provide terminal services and then fly a weekend or nightcoach passenger from 192 to 253 miles for $13.89 ($15.00 including tax), almost identical to the alleged cost of a CAB-regulated airline's brief "sightseeing" flight. There are indeed fundamental differences between CAB-regulated airlines and the successful intrastate carriers; and these differences stem from regulation which serves to close entry, provide a mechanism through which agreements can be reached regarding fares and other matters, and facilitate the low-cost (to the airlines) enforcement of such agreements. CAB REGULATION HAS DECREASED AIRLINE EFFICIENCY If the fares and rates of the certified airlines have been increased so much by CAB regulation, why haven't these airlines made consistently large profits? Of course, at various times since 1938, most of the trunk carriers have enjoyed substantial profits and a few of them, such as Delta, Northwest and Western, have had quite good and relatively consistent profit records. On the whole, however, the profits of the entire regulated industry have been less than spectacular and certainly lower than would be expected from the higher fares resulting from regulation. The fundamental reason for the surprisingly poor profit performance of the certificated airlines is that, while increasing airline prices, CAB regulation has also greatly increased airline costs. Three of the major sources of these higher costs appear to be: 1) imperfections in CAB regulation which promote intense service-quality (rather than price) rivalry among the airlines; 2) regulatory entry restrictions which require each of the relatively few certificated airlines to provide a diversity of services instead of being able to specialize in an operationally homogeneous service pattern; and 3) the assistance that entry restrictions give monopolistic or oligopolistic airline suppliers in capturing larger portions of the airlines' gains by increasing the prices the airlines must pay for their inputs (especially labor). These three factors-carrier rivalry, decreased specialization, and increased input prices-will be examined below. Carrier rivalry 50 While the CAB has full discretionary power over entry and exit, under the Federal Aviation Act of 1958 (as well as the earlier 1938 act) it has had only limited direct power over the quality of service provided by the airlines. In addition, it has been slow to influence service quality through indirect means such as promoting capacity agreements and conditioning its approval of fares on modifications of service quality." Thus, CAB regulation has been asymmetric or "imperfect," with complete control in some areas and little control in other areas. The effects of this asymmetric regulation have been amplified by the Board's failure to allocate predetermined shares of industry revenues and profits to each carrier. Specifically, it has authorized two or more airlines to operate over every major interstate route and between every major city-pair without assigning traffic shares in each city-pair. Therefore, great rivalry has developed between the certificated carriers as they seek to retain or increase their traffic shares." Since CAB regulation requires cooperation and agreement in setting prices, this rivalry has long been channeled toward the area of service quality. Operationally, this has meant that each carrier in a multi-carrier city-pair has endeavored to operate more frequent flights with the fastest, most modern and comfortable aircraft, while providing elaborate in-flight services and superior ground services. The following analysis indicates some of the effects that such service-quality rivalry can have on the costs of regulated airlines. An airline's most important physical asset is its aircraft and related equipment. As of June 30, 1974, flight equipment accounted for over 84 percent of the nondepreciated value of the total operating property and equipment of all CABcertificated route air carriers. Obviously, if CAB-induced service-quality rivalry CAB. supra, note 1, at 411. 62 William A. Jordan, supra. note 1, at 2-4. William A. Jordan, Airline Capacity Agreements: Correcting a Regulatory Imperfection, 39 Journal of Air Law and Commerce 193-205 (spring 1973). In addition to the effects of multi-carrier authorizations on service-quality rivalry, there is no indication that the CAB has made route awards to the carriers able to provide the optimal level and quality of service at the lowest marginal cost I n'ess lowcost carriers are authorized to provide new services. and average costs of airline operations are increased over minimum attainab 62 CAB, Air Carrier Financial Statistics 53 (Jur ww results in excessive numbers of aircraft being purchased and then being underutilized, the costs of regulated airline operations will be increased. Evidence on this point can be obtained by comparing the aircraft utilization by the CABregulated trunk carriers with that of the California intrastate carriers between 1949 and 1965. The California carriers could, of course, undertake rivalry both by reducing fares and by increasing service quality, while the regulated trunk carriers were largely constrained to service-quality rivalry. Not surprisingly, the record of this period demonstrates that price rivalry was indeed very important within California in sharp contrast to the emphasis on service-quality rivalry by the CAB-regulated airlines.63 The following four measures can be used to determine the overall intensity with which aircraft are utilized: 1. The average number of revenue hours per day that each aircraft is operated (a measure of airframe utilization). 2. The number of seats installed in each aircraft type for a given class of service (indicating the extent to which the aircraft's interior is utilized). 3. The average passenger load factor (which measures the degree to which the installed seats are utilized). 4. The total number of years each aircraft is operated (measuring the length of time this resource is utilized, as well as indicating the flow of resources required to maintain some given stock of aircraft over time). It happens that the successful California intrastate carriers operated their aircraft about the same number of hours per day as the trunk carriers (and somewhat more than the local service carriers), so there is little difference in efficiency in this respect, unless detailed adjustments are made to compensate for the intrastate carriers' shorter stage lengths. Significant differences did exist, however, in the number of seats installed per aircraft, in passenger load factors, and in aircraft life. 64 Two factors are relevant with regard to numbers of seats installed. First, no intrastate carrier operated first-class service from 1949 to 1965 and, therefore, never provided the low-density seating associated with that high-quality service. Second, when comparing identical aircraft types in all-coach configurations, one finds that the intrastate carriers generally managed to install more seats. Usually the differences were in the order of 5 to 10 percent, but there was one case of a 24 percent increase and a few others with differences of only 1, 2 or 3 percent. Combined, these two factors have resulted in the intrastate carriers utilizing their aircraft appreciably more than the CAB-regulated airlines, simply by putting more seats in each aircraft. 65 Even with more seats per aircraft, the California intrastate carriers usually managed to fill an appreciably larger percentage of them and thus have higher coach load factors than the certificated airlines. The exceptions all occurred between 1949 and 1953 when the trunk carriers' coach service was very limited and was mainly provided in long-haul city-pairs while, the intrastate carriers operated only in short-haul city-pairs and were experiencing their initial period of extensive entry and exit (with failing carriers having low-load factors). For that five-year period the trunk carriers' average coach load factor was 73.9 percent compared with 69.8 percent for the intrastate carriers. In every subsequent year, however, the California intrastate carriers' annual average load factor exceeded the coach load factor of the trunk carriers by amounts ranging from 1.0 percentage point in 1954 to 17.9 percentage points in 1963. From 1955 through 1964, the annual average load factor for all intrastate carriers combined never fell below 70 percent and it reached a high of 80.6 percent in 1957. Their 12-year weighted average load factor was 71.2 percent compared with only 59.1 percent for the trunk carriers' coach operation. Some might argue that these differences resulted from the high traffic densities in the three major California city pairs. But a detailed comparison of the scheduling practices of the two carrier groups in those city pairs demonstrates that the successful intrastate carriers' higher load factors were due to management decisions. Specifically, the intrastate carriers' adjusted their schedules to conform more closely with traffic fluctuations than did the trunk carriers. 63 William A. Jordan, supra, note 1, at 34–72. 64 Id. at 197-99. 65 Id. at 199-201. 66 Id. at 200-3. 67 Id. at 203-9. 67 Finally, during much of this period the California intrastate carriers mainly operated older, obsolescent aircraft that they purchased from the trunk carriers or from governments surplus stocks. Thus, they served to extend the productive lives of these aircraft, thereby increasing their utilization and economic efficiency. This increased efficiency was reflected in the low prices the intrastate carriers paid for their aircraft which, in turn, served to lower their costs. Had the CAB-regulated airlines been allowed to adopt prices lower than those authorized by the CAB, they too could have used such aircraft in low-fare service, but so long as price rivalry was prevented by regulation it was desirable for the trunk carriers to concentrate on high-quality service and thus retire their aircraft relatively early.68 The above experience is what happened with piston-powered aircraft through the 1950s and into the early 1960s. The turbine-powered aircraft innovation was different in that these aircraft provided a happy combination of lower operating costs and superior service quality. This indicates that such aircraft would have been adopted quite early by all airlines regardless of regulation. Not surprisingly, PSA inaugurated turbine-powered Electra service less than eleven months after the first trunk carrier, and it adopted the medium-range Boeing 727-100 fifteen months after that aircraft first flew in scheduled service. It was those aircraft that provided somewhat higher service quality at higher costs (such as the DC-7, the L-1649) that were fathered by CAB-regulation and which would probably not have been developed without such regulation. 69 The implications of these differences in aircraft utilization can be illustrated by the following example. The trunk carriers' 1973 system load factor was 52.4 percent and, with the operational adjustments related to the fuel crisis, it rose to about 56 percent in 1974.70 Also, the service life of a turbojet aircraft is commonly thought to be around 14 years. Now, the California experience indicates 70 percent load factors are feasible without regulation and that aircraft with low operating costs would have longer lives without regulation than with regulation. A 70-percent load factor is 25 percent higher than the 56-percent load factor of 1974 (and 34 percent above the 52.4-percent load factor in 1973), and applying a similar 25-percent extension to a 14-year aircraft service life would bring it to 17.5 years. Increasing the average load factor by 25 percent means that the trunk carriers' present traffic volume could be carried by 80 percent of their existing fleet, that is, by about $9.8 billion worth of aircraft rather than the $12.25 billion (undepreciated value) in existence as of June 30, 1974."1 17.5 years, the annual replacement rate would be in the order of $560 million ($9.8 billion-÷17.5) rather than the $875 million rate required to replace the current fleet every 14 years ($12.25 billion÷14). This would be a saving of roughly $315 million per year, or about $1.90 per passenger enplanement.72 The above calculations are quite rough and ignore the fact that increased seat density without regulation would result in further reductions in the fleet size required to carry existing traffic. Furthermore, while important, aircraft are just one item of airline costs. A 25-percent increase in load factor also implies a 20-percent decrease in the number of flights which means fewer flight personnel, less fuel consumption, lower maintenance costs, smaller landing fees, and so forth. Clearly the cost differences would be large. One can begin to see how emphasizing service-quality rivalry increases the costs of the CAB-regulated airlines. One can also begin to see why the successful intrastate carriers have been able to survive and profit under fares much lower than those authorized by the CAB. Decreased specialization When relatively unregulated, the intrastate carriers have been characterized by operational simplicity and homogeneity. Generally speaking, each carrier has operated only one aircraft type at any point in time (except for brief transitional periods), offered only one class of service (coach), and operated between city-pairs that provided traffic flows and stage lengths compatible with the chosen aircraft type. Some of the carriers, such as Mercer Enterprises and Holiday Airlines, have served recreational communities or activities where low 68 Id. at 209-10. 69 Id. at 40-44, 55-56. Also. CAB, supra, note 1. at 552. 70 CAB, Air Carrier Traffic Statistics 1 (Dec. 1973), and 2 (Sept. 1974). 71 CAB. supra, note 62, at 55. Furthermore, if this smaller number (value) of aircraft were replaced every 72 Total system trunk revenue passenger enplanements for the year ended September 36, 1974, was 166,730,000. CAB, supra, note 70, at 2. traffic densities exist. Others, like PSA and Southwest, have operated between large cities having high traffic densities. To this day, PSA, the largest of the intrastate carriers, serves only 11 airports while, prior to 1965, it never served more than five airports at one time. Similarly, Air California now serves eight airports, Holiday seven, and Southwest three.73 Overall, the intrastate carriers have customarily been quite small. Only one (PSA) has ever operated more than a dozen aircraft at one time (three to six aircraft has been the usual fleet size), and they have consistently had very simple fare structures and procedures. All this has allowed each employee to become intimately familiar with his work. Pilots, for example, spend large parts of their professional lives flying only one aircraft type between just a few airports. In the same way, the work of employees engaged in maintenance, engineering, scheduling, reservations, accounting, etc., are all simplified, and managers can become intimately acquainted with the carrier's traffic and operational features. Contrast this to the trunk carriers at the end of 1972. The smallest number of U.S. and foreign cities served by any carrier was 37 (Western) while the largest number was 89 (United). The number of certificated route miles ranged from 10,821 (Western) to 38,968 (TWA), and the trunk fleets ranged from 60 aircraft (Continental) to 388 (United) with each carrier operating several different types of aircraft." Even the local service carriers were geographically large and diverse in comparison with the intrastate carriers. In December 1972 each of them between 50 and 95 cities connected by 3,660 to 8,295 certificated route miles, and each operated 33 to 133 aircraft.75 Even brandnew Air New England has been authorized to serve 14 cities located throughout New England. The operational problems and complexities faced by these certificated airlines are much greater than those of the intrastate carriers, and the associated decrease in specialization translates into less output per employee. 78 Rough indications of the effects of specialization on airline productivity may be obtained by comparing the 1965 average annual output per employee of the total trunk carriers as a group and of Western Air Lines (one of the more productive of these carriers) with that of PSA. 1965 was, of course, the last year of relatively unregulated operations in California, and it was also the year that PSA introduced jet aircraft and thus operated a fleet whose composition was very similar to that of the trunk carriers." Two of the productivity measures will be available seat-miles (ASM) per employee and revenue passenger-miles (RPM) per employee. It happens that these two measures are biased against the certificated carriers because they provided relatively more cargo service than PSA which served to inflate their number of employees without increasing their ASM and RPM figures. Therefore, to provide balance, operating revenues per employee will also be given with the recognition that this measure is biased against PSA since its lower fares required more physical output per dollar of revenue. The data for Western Air Lines are particularly useful, both because they indicate the range of trunk carrier productivity and because a substantial portion of its total operations were entirely or partially within California which may yield increased comparability with PSA's data. The comparisons are given in table 8. TABLE 8.-AVERAGE ANNUAL OUTPUT PER EMPLOYEE (PSA COMPARED WITH TOTAL TRUNK CARRIERS AND WESTERN AIR LINES), 1965 Source: William A. Jordan, Airline Regulations in America, 215, 217, 219 (1970). 73 PUC, Present and Proposed Route Structures of California Certificated Air Carriers and Routes Onerated by CAB Certificated Air Carriers as of July 1, 1974. 1A, 10, 15A (n.d.). Also, Southwest Airlines, Third Quarter Report (nine months ended Sept. 30, 1974). 70 CAB press release 75-18 (Jan. 23, 1975). 77 1965 was neither the best nor the worst year for PSA's productivity relative to the certificated airlines. While its jet operations increased total output, between April and August its total number of employees increased from 575 to 740 persons. William A. Jordan, supra, note 1, at 339-40. |