these Mach 2.8 aircraft would earn about $100 per flight hour, about three times as much as the present DC-8 captain.

Improvements in aircraft, plus a host of complementary innovations in all phases of airline operations, have produced a rapid growth in manhour output: about 7 percent per annum since 1950. This growth rate has been outpaced, however, by the increase in traffic, and the employment trend has therefore been upward. Thus, the airlines employed 16,000 in 1939, 81,000 in 1949, 161,000 in 1959, and 190,000 at the end of 1964. Flight crews have been about 10 percent of total employment: 15,000 pilots and 4,000 others, mostly flight engineers, at the beginning of 1965.

The pilots, as one might expect, have exhibited considerable bargaining power. I would suggest, as major factors in pilot bargaining strength: the competitive environment, in which each carrier has been anxious to maintain public goodwill and to avoid loss of traffic to rivals; the pilots' successful resistance to carrier efforts to develop multiemployer bargaining arrangements; Federal subsidy policies which, especially during the 1940's and early 1950's, were important offsets to the costs of the pilots' economic gains; bargaining by the pilots as a separate occupational group; militant pilot support for ALPA (and, at American Airlines since 1963, for the independent Allied Pilots Association); sympathetic intervention by Congress and the Civil Aeronautics Board on pilot matters; and the pilots' remarkable success, earlier before Congress and later before Emergency Boards, in enforcing certain productivity criteria as a traditional basis for minimum economic gains.

Hours versus Earnings

The rapid activation of jet aircraft caused pilot employment to fall from 1961 to late 1962, and only in 1964 did it become clear that the airlines were again hiring pilots and that the number of pilot jobs was destined to keep increasing for the foreseeable future. It was in this changing employment context that the carriers and their respective pilot groups found themselves dividing into two camps: the American pattern (July 1963) of a flat reduction to a maximum of 75 hours of flight pay and credit per month; and the United pattern (January 1965), which not only retains the

traditional 85-hour maximum but permits earning pay for credited hours above this maximum.

The Federal Aviation Agency (FAA), which determines safety requirements and crew complements, requires large aircraft to have a three-man crew of two pilots and a flight engineer. The FAA does not require the flight engineer to be a pilot, but ALPA policy since 1956 has been to insist that the flight engineer on jets must have at least a commercial and instrument (C & I) pilot rating. The Flight Engineers International Association (FEIA)-the union representing flight engineers at carriers where this group had been drawn from mechanically oriented nonpilot sources-insisted that the jet flight engineer have an aircraft and engine (A&E) mechanic license. American Airlines, in December 1958, anxious to avoid any work stoppage that would delay its activation of jets, announced that it would fly with a four-man crew, thus accommodating both ALPA and FEIA. Other carriers dealing with FEIA (including Eastern, Pan American, Trans World, and Western) followed American's costly example. At carriers where the flight engineer was pilot qualified, however, as at United, the jets were flown by three-man crews from the outset.

Contract negotiations between American and ALPA started in March 1961. American was determined to eliminate the "featherbird", but it was not willing to meet its pilots' economic demands and also provide C & I training to its 600 nonpilot engineers. In July 1962, the pilots' negotiating committee (composed entirely of American pilots) initiated a break with national ALPA policy by advising American that it would not require C & I training for nonpilot engineers on jets if American would grant its pilots major objectives. American agreed, but ALPA would not permit this deviation from its established national policy. American's pilots then withdrew from ALPA and created a new independent union, the Allied Pilots Association (APA), to serve as their bargaining representative.

On July 9, 1963, American and APA signed a bargaining agreement which authorized a threeman jet crew that could include a nonpilot engineer. This contract abandoned the 85-hour monthly maximum for flight pay and credit and, effective on March 2, 1964, established a 75-hour maximum for turbojets and an 80-hour limit on propeller aircraft. The minimum monthly guar

antee was placed at 80 percent of the applicable monthly maximum for the equipment at half-day and half-night rates. All pilots on the seniority list as of July 1, 1963, were guaranteed full protection against both layoff and any pay loss deriving from the transition to three-man jet crews. In fact, there was no furlough problem because the hours and rules changes had been expressly designed to "create" the 280 jobs that were immediately at stake.

American agreed to provide the same pay for 75 hours on jets and 80 hours on propeller aircraft that had previously been given for 85 hours, and to contribute an additional 9 percent of each pilot's earnings to an improved retirement and disability program. Many pilots were to benefit very promptly from the upgrading opportunities generated by the cut in hours, which included 100 new first pilot jobs.

The American-APA example became the pattern for other pilot contracts. ALPA and National adopted an average 80-hour maximum on June 15, 1964; ALPA and Delta established a 75-hour limit for jets in their contract of July 1, 1964 (effective December 31, 1965); and ALPA and Trans World agreed on November 12, 1964, to move to 75 hours by December 1, 1966.

Although not every ALPA contract signed between July 1963 and December 1964 provided a reduction in monthly flight pay and credit, it did appear that the path toward a 75-hour "pay cap" had been so well blazed that the rest of the industry would inevitably accept it. But the improving economic climate, evidenced by increased pilot employment in and after late 1964, set the stage for a significant change of pace initiated at United Air Lines.

The United Pattern

The 1963 United-ALPA contract was scheduled to terminate on January 1, 1965. Intensive bargaining did not take place until late 1964, when United had already launched a substantial long-term pilot hiring program. This bargaining environment was a sharp contrast with the static employment and jet crew reduction faced by American in 1963. The new United-ALPA agreement, settled by direct negotiations on January 18, 1965, demonstrated that United's pilots, in the absence of any layoff worries, preferred

higher earnings opportunities to an hours reduction and, in addition, were willing (for a price) to accept certain modifications on traditional seniority rights that would save United substantial training costs during the ensuing growth period.

United had proposed to keep its pilots' total duty hours and total trip (away from home) time within the limits imposed by American's duty and trip ratios, provided that the ceiling of 85 credited flight hours was not only retained but eased by an averaging or "banking" formula.

ALPA accepted this proposal, as well as a "pay bank” arrangement under which a pilot completes his monthly schedule even when this causes him to exceed 85 credited hours (provided only that he does not exceed 85 actual hours); the pay and credit above 85 in any month is banked, the credits to be drawn upon in subsequent months to bring lower pay and credit up to 85 hours' worth; and any "balance" as of each December 5 is paid off each December 15. The major significance of this pay bank is that it enabled United to increase by a full 2 hours per month (from 81 to 83) the average scheduled pay and credit of its pilots' monthly schedules.

The 1965 United-ALPA agreement also contains a major departure from the traditional pilot pay formula, namely, uniform pay within each of three equipment groups: large jets (DC-8 and B-720); small jets (Caravelle and B-727) ; and all propeller equipment (DC-6 series, Convairs, Viscounts). Within each group, the pilot is paid as if he is flying the type with the highest pay yield: but, in return, the pilots have agreed to forgo the right to bid for other aircraft types within the same group at the same domicile (unless a promotion is also involved). United's net savings will be substantial, since, for example, the cost of transition training for an experienced three-man jet crew is about $35,000 and the nonavailability of the crew while training adds up to about $15,000.

Pan American and its pilots had been engaged in serious bargaining since November 1963 along the lines of the American (and later the TWA) 75hour precedent. But then the Pan Am pilots examined the new United-ALPA formula in January 1965, realized its earnings implications, and shifted course. While different in many details because of Pan Am's international route structure, the 2-year Pan Am-ALPA contract signed on

April 10, 1965, left the 85-hour limit intact. Negotiations at Eastern began 9 days later, and Eastern's management immediately proposed "a United contract". Again, apart from differences in detail, the 2-year contract signed by Eastern on July 30, 1965, was patterned on the United example.

Meanwhile, American and APA were again engaged in contract bargaining, and American warned that failure to obtain relief from the rigid 75-hour limitation it had granted in 1963 would, in view of the spreading United pattern, jeopardize American's 1966 expansion plans and permit its competitors to acquire a larger share of the growing market. This argument-and, I suspect, the willingness of many American pilots to earn more money when the additional flight hours would not cause layoffs-was persuasive enough to produce a substantial relaxation of the 75-hour maximum for 1 year starting August 7, 1965. A Letter of Agreement permits American to schedule one duty period each calendar month for which the pilot is paid at regular rates "but for which he shall be given no flight time credit toward the applicable monthly maximum."

The recent and current developments affecting

pilot working hours offer, I am convinced, an illuminating opportunity to analyze the incomeleisure choice in a context of much relevance to the world of tomorrow, which is likely to have many organized employee groups with common joboriented interests and real incomes comparable to today's airline pilots. In addition to the job-security context that I have mentioned, for example, I am examining other determinants of carrier attitudes (such as the impact of shorter hours on training and fringe benefit costs) and of pilot attitudes (such as the extent of nonflying moonlighting, compulsory retirement at age 60, progressive income taxation, and the impact of hours changes on the status and pay of less senior pilots). But I will hazard a forecast, applicable only to subsonic air transportation: that monthly hours of flight pay and credit will tend toward 85 in periods of expanding pilot employment, towards 75 in periods of declining pilot employment; and I will not be surprised if such arrangements are formalized by contractual provisions that establish hours of, say, 80 or 85, but with the proviso that no pilots may be furloughed until hours have first been reduced to, say, 70 or 75.

Future of Maritime Manpower

AARON W. WARNER*

THE POTENTIAL SUPPLY of qualified and experienced seamen has for many years exceeded the estimated number of men needed, by a very large margin. Roughly calculated (and using data from separate sources), the excess of men over jobs in 1962 appears to have been of the order of 20,000 to 21,000 licensed officers for approximately 11,000 jobs and 69,000 to 70,000 unlicensed men for approximately 34,000 jobs. There is no recent evidence, however, on the extent of job dispersion.

On the basis of these estimates, the ratio of the active labor force to the number of jobs appears to be approximately 2 to 1. This might indicate a serious problem of adjustment; however, a number of important qualifications must be made in the interpretation of the overall data. There can be considerable fluctuation in the number of jobs available, depending on the emergency needs of the Government. Under present circumstances, as a result of the Viet Nam crisis, vessels carrying Government-sponsored cargoes are finding it difficult to staff all positions.

A second qualification is that what appears to be an excess of personnel in certain categories is in fact merely the existence of large numbers of men whose connection with the industry is only casual. Thus the engineers, with a surplus of trained men, are really in short supply. Even in normal periods, ships not infrequently must sail without their full complement of engineers, and in periods of emergency this shortage becomes critical. Similarly among the unlicensed categories, there is in practice no marked surplus of stewards, who are able to find lucrative part-time work ashore in the hotel and culinary trades. There is, however, considerable unemployment among deck officers, and the Masters, Mates, and Pilots (MM&P) has recently claimed that of the union's 11,000 members, 25 percent cannot obtain berths. Still, it must be pointed out that a substantial portion of this unemployment consists of older men who no longer wish to spend time at sea and will accept only relief work in port. The major burden of the unemployment, then, falls on the unlicensed groups other than stewards, and particularly on the less

skilled workers for whom shoreside opportunities are not easily available.

Of the men employed as radio, deck, and engine officers during the period from April 1, 1962, to March 31, 1963, only 13 percent of the deck officers, 9 percent of the licensed engineers, and 6 percent of the radio officers had entered the industry since the end of the Korean conflict. Among the unlicensed personnel, on the other hand, the entry of new men into the unskilled categories had been fairly substantial, although the skilled categories among the unlicensed personnel have a higher average age and show a much reduced rate of recent entry.

The National Academy of Sciences has estimated that in conjunction with low entry rates, net attrition rates for deck officers, engineering officers, unlicensed deck men, and unlicensed engine men average approximately 5 percent a year. On the basis of this estimate, projections indicate that by 1972 all of these groups will be reduced to half of their 1962 size. For engineering officers, whose services are already in short supply, the proj ¿ctions indicate that by 1972, when only half the present labor force will still be active, approximately 70 percent of the engineers will have been in service for 25 years or more.

Need for Planning

It becomes apparent that the solution to the unemployment problems lies not merely in expansion of the fleet but more importantly in manpower planning. This is made more urgent by the introduction of automation, which intensifies the existing imbalance by further reducing the jobs for the unskilled, who are at present in greatest oversupply, while at the same time placing a premium on highly trained personnel, who will increasingly be in short supply.

Manpower planning so far has taken the form of collective bargaining arrangements to deal with existing and future unemployment. Over the past decade, for example, pension benefits have been greatly improved. Training and automation funds have also been established in certain unions, and these presumably will be adopted in other sectors of the industry. These arrangements, cou

*Professor of Economics, Columbia University.

pled with additional employment opportunities arising from expansion of the fleet and measures to improve the distribution of manpower, should assist the industry in coping with technological displacement. Since automation for the most part will require the construction of new ships, the transition to an automated fleet will necessarily be gradual. The transition will also be eased, at least in the immediate future, by the increased need for manpower resulting from the Viet Nam crisis.

On the basis of these considerations, it becomes possible to consider the policies that will be required to restore a proper balance between manpower needs and the existing labor supply. The major limitation to expanding the fleet, even an automated one, is the imminent shortage of licensed personnel. Any present oversupply of manpower among the licensed officer groups will be self-correcting within a relatively short time because of attrition and lack of new entry. The major problem will then become one of recruiting and training new officers for the increasing number of automated ships.

The situation for deck officers is somewhat paradoxical. Although this group has the highest attrition rate and will rapidly become inadequate for future needs, the presence of older men on the shipping list has blocked the entry of younger recruits. As a result, those trained as deck officers in the various maritime academies must look elsewhere for careers. One manifestation of the lack of career opportunities is the large proportion of deck officers who sail with license ratings higher than the rank they fill.

The shortage of licensed engineers, which already exists in fact, is due to another set of factors. The training of engineers in the maritime academies is sufficiently broad to equip them for engineering positions in industries other than maritime; many prefer it. Men who leave the maritime academies frequently spend a short time at sea, during which they become members of the union, then leave for other employment. The present rate of intake of licensed engineers into the maritime industry is approximately 159 a year, most of whom have been trained in the maritime academies. Of this number, approximately 44 percent leave the industry within the first year, and another 18 percent during the second

year.

To attract the caliber of men needed as chief officers on the new technologically advanced ships, and to retain them, will require more than financial inducements. In view of the many opportunities ashore for highly trained personnel, attention will have to be given not only to training, but to career opportunities within the maritime industry, including opportunities for further study and perhaps for a continuing career after the initial 20-year period leading to retirement has been served. This may require the establishment of a definite career pattern for advancement from service aboard a ship into managerial or professional positions in the shore-side operations of the industry. In addition, it may become necessary to admit a larger number of potential officers to the maritime academies in order to increase the intake into the industry, and also to modify the curriculum to include the teaching of managerial and technical skills that will be needed on automated vessels.

Achieving a Balance

With the shortage of officers as a limiting factor, any transition to a larger fleet will necessarily be gradual, even apart from the need to build new ships. There remains the difficult problem of the oversupply of unlicensed men, particularly among those who are less skilled. To the extent possible, this group should be permitted to grow smaller through attrition, since shipboard automation will inevitably reduce the number of unskilled jobs regardless of the size of the fleet. It may be necessary in addition to take positive measures to reduce the size of this group, including the hastening of retirement for older men and the retraining of younger men for possible transfer to shore-side jobs. One possibility is that a part of this labor force can be shifted to maintenance work ashore. Once the proper balance of men to jobs in this category has been achieved, it will then be possible to establish a rational policy for admitting men to the industry as they are needed, with suitable provisions for training and promotion.

A further opportunity for balancing the labor force exists in the promoting of unlicensed men to become licensed officers. One important reason for the low rate of entry into the licensed officer categories is the falling off in the number of men