XI

COSTS: INTO THE STRATOSPHERE

Another way to look at the cost picture is to analyze the price rise of a single type. When McNamara studied the costs-benefits of reviving production of the piston A-1 for the Vietnam War, he was shocked at the cost. With no need for development, production of 300 A-1's was estimated to cost $800,000 apiece. They had cost $175,000 to $200,000 apiece in the late forties. The price of the A-1 had thus quadrupled in less than two decades; McNamara decided the plane was not "cost-effective." Over the same period of time, the average hourly earnings in aircraft and parts plants, including overtime, only doubled. The price of an F-4 more than doubled between 1963 and 1970 during a modest increase in weight in the Air Force version. This cost rise took place in spite of the learning curve on a lengthy production run.

Allowing for some increase in manufacturing productivity, something, then, has caused costs to more than double over what one would expect in the production phase alone.

FIGURE IV

COMPARISON OF GROWTH IN AIRLINER PRICES, PRODUCTIVITY, & THE CONSUMER PRICE INDEX

Sources:

• Airliner prices from Aviation Week and Wall Street Journal.
• Airliner productivity from Aviation Week, Jul. 11, 1966, p. 109.
• Consumer Price Index from Economic Almanac, 1967-1968 Business Factbook (New York: Macmillan, 1967) and The World Almanac and Book of Facts (New York: Newspaper Enterprise Assn., 1973).

In the Vietnam War the low cost-effectiveness of the Douglas A-1 meant the end of the piston combat aircraft. Courtesy McDonnell Douglas Corporation.

The broadest influence must be the society of which the industry is a part. Ours is a society which has kept its love of gadgets, but which has become disenchanted with technology, production, and the work ethic. It has become a service rather than a production society. In the latter half of the sixties the productivity of Europe rose 40 to 50 percent, that of Japan 90 percent, and that of the United States only 10 percent. And for the United States this was down by half from the two previous decades.

THE CUSTOMER AS THE CULPRIT

The huge increase in productivity which the jet engine gives over the piston is graphically shown by these airliner interior views. At the top is a Lockheed Super Constellation and below is a Lockheed TriStar. Courtesy Lockheed Aircraft Corporation.

Within the government, Congress has both economized and helped costs rise. Often Congress has been aggressive in trying to hold profits down, undertaking many effective investigations such as the one of Kaiser-Frazer in 1953; it has attempted to police procurement, it has used the GAO to make analyses, and it has scrutinized appropriations. The results of these efforts have been that Congress is the most effective government agency in holding down costs, despite the fact that it has also done much to increase them. Congress has been generous to defense, often appropriating more than was requested by the civilian leaders in the Defense Department. Perhaps the strongest expression of this attitude was given in 1972 by the influential F. Edward Hebert, chairman of the House Armed Services Committee. He said, "In war they don't pay off for second place. There's one bet and you've got to have the winner. I intend to build the strongest military we can get. Money's no question." Besides this attitude, there is politics. It cannot be attributed as a factor in any one contract, but its influence is apparent in the aggregate. Defense contracts are concentrated in areas represented by influential congressmen, and Congress strongly intervenes to promote social programs which may have long-run economic benefits, but not necessarily short-run ones. These social objectives include racial considerations and favoritism for small business and labor surplus areas. Defense contracts came to dominate the pork barrel over patronage and public works. In 1971 when the Pentagon tried to reduce contract costs by simplifying specifications for a project, it still produced a twenty-five-page document; two pages listed the design specifications and twenty-three listed the social and political requirements imposed by Congress.¹ It is probably impossible to measure the cumulative effect on costs over decades of these policies.

The executive branch of the government has more direct controls over costs and therefore bears heavy responsibility for them. Yet, from the early days of the cold war to 1970, the Bureau of the Budget, one of the principal governmental watchdogs, left the Defense Department alone.

The above are monitoring organizations for the most part; the Pentagon is the action agency and is therefore best placed to avoid costs. Its overall failure to do so is conspicuous. The French, Swedes, English, and Russians, all were producing aircraft for a fraction of American costs in the early seventies. Perhaps the main reason for the Pentagon's failure lay in its nature as a bloated and musclebound bureaucracy. In the early seventies it contained enough personnel for a World War II army corps. And to compound the problem, the services had more generals and admirals than in World War II. The Pentagon is a good example of Parkinson's Law. Since 1947 power has been increasingly centralized in the Pentagon, and within it in the Office of the Secretary of Defense. The evidence points to increasing paralysis as a result. In the early seventies even routine contract matters might call for fifty written concurrences. When Deputy Defense Secretary David Packard left the Pentagon he said, "There are a great many people in the department and . . . it just takes a long time to get anything done, even some of the most simple recommendations." Earlier he had told Congress that he would like to "give the contractor a contract without all of this damn red tape."²

The musclebound condition of the Pentagon has been enough to cause procurement inefficiency, but attitudes might prevent adequate cost controls even in a streamlined organization. Among the legions of bureaucrats, there has been only a handful of production experts, the group which is naturally cost-conscious. The only military officer famous for doing something about costs has been the maverick Vice Admiral Hyman Rickover, who was kept on active duty only by the will of Congress. The most famous civil-servant advocates of the belief that procurement costs could and should have been reduced, the Air Force's Deputy for Management Systems A. E. Fitzgerald and Navy Director of Procurement Control and Clearance Gordon Rule, obviously were regarded as troublemakers; the euphemism "not a team player" was used by top management about Fitzgerald.³ To be an empire builder, a spender, has been to be honored and promoted. To be sharp on costs has been to be ostracized. One reason for this has been a prevalent view that performance and time are of overriding concern and cost is not. Another reason was a belief that spending, in itself, was in the national interest. One Air Force officer, Major General "Zeke" Zoeckler, even said in the late sixties, "Inefficiency is national policy."⁴ This is consistent with the most common problem-solving approach in the services, where it has been believed that almost all difficulties can be removed with the application of masses of money and men, the counterpart to the engineers' solution of "more technology." The problems of Rickover, Fitzgerald, and Rule illustrate that service attitudes have fostered sycophantism and repressed unpopular judgments. Holding to unfashionable beliefs can jeopardize an officer's career.

Service practices which have impaired effective decision-making, and which have therefore meant higher costs, have been caused by working conditions. An efficient procurement job has been impossible because personnel have rotated too frequently. The problem has been compounded by the service belief that a senior officer is so competent that he can do any kind of job well, an opinion not substantiated by results, and by the illusion that to be assigned to a position confers competence upon the officeholder. An officer assigned to a position outside his competence is derisively called "the instant expert." Finally, there has been so much busy work that decision makers have found it difficult to devote the proper amount of time to their problems. This was even codified in the idea that a manager should spend only 15 percent of his time in being creative.

The centralization and the mass approach to problem solving led the Pentagon to impose its style of management upon the aerospace companies. The Pentagon and its procurement satellites became, in fact, the, managers in detail of the aerospace companies. The process began with a cost contract that required monitoring of a company to protect the government from over-payments. Bureaucrats were therefore in a position to expand their power and, quite naturally, did so. Too, bureaucrats have sought to protect themselves from review agencies by expanding controls. The aerospace companies acquiesced bit by bit, each time to get a new contract for which they were hungry. Another form of pressure to conform was put on North American. Air Force Major General Samuel C. Phillips, director of the NASA Apollo program in 1965, studied North American's management, and the general thrust of his conclusions was that the film's management concepts should be similar to the Air Force's. He did not say "Air Force" but merely described his own views.⁵ His report surfaced during the Apollo fire investigation in 1967; it probably damaged North American's reputation and may have caused the imposition of additional bureaucratic systems on the company.

The general result has been that the aerospace companies came to be run by a huge government bureaucracy, in which many individuals could obstruct action by their non-agreement. By 1971 defense procurement was regulated by over 14,000 documents. And not only was there the defense bureaucracy, but also there were the necessary counterparts within the aerospace company to serve the government officials. Professor Frederic M. Scherer said this has made the firms "arteriosclerotic."⁶ Reporting is naturally immense under such circumstances: Kelly Johnson said, "Every project we have ends up with tons and tons of needless paperwork";⁷ Martin had to send biweekly reports on 2,500 items on Titan III; 23 percent of the A-7's cost was attributed to paperwork. Costs also rise as engineering problems are solved by mass attack. The Air Force's reaction to problems on the Titan in 1960 was that the difficulty was caused by too few bosses.⁸ Similarly, when Skybolt had troubles the Air Force asserted that the problem was that the project needed 67 percent more engineers. The attitude has been a reflex in the Air Force.

The centralization which has proved to be an organizational problem has been compounded by a fashion in decision-making. In the services the old custom of briefing has been abused. Briefings are a most useful and efficient means for communicating simple or background information. Like the TV news, briefings are part showmanship and are particularly suitable for gimmickry. They are routinely used as propaganda and almost always present a Pollyanna approach to matters. Briefings have proved so popular that they have nearly driven the staff study, which is intended to be a thorough, objective, documented paper, out of use. They have come to be a major basis for decisions for the centralized top managers. In the important F-111 selection decision, most of the top military officers involved did not even glance at the written report furnished them; they made their decision solely on a briefing of a few minutes which emphasized Boeing's optimistic hopes of reaching certain performance levels, and which omitted the criticism contained in the written report.⁹

A related form of deception which protects waste has been concealment of the extent of financial errors or overruns; success in this tactic naturally obstructs efficiency. Another Pentagon process has been to use historical cost data to determine present costs. Fitzgerald has accurately described this as having a ratchet effect ensuring rising costs. Thus if one uses historical costs without analysis, every case of waste becomes incorporated in later estimates, so that costs have only one direction: up. Some of the waste has been provided by the annual rush in June to ensure that the whole budget was spent and no monies were saved.

Yet another difficulty in the Pentagon system is related to "buying-in." There are advantages to a service in "buying-in" on a program, whereby costs are understated in order to get a program started. Once money is spent, there is natural reluctance, even in Congress, to cut losses on sunk costs. Starting several programs in which the costs are relatively small at the beginning provides leverage to expand service budgets later, when sunk costs exist and the projects have advanced to more expensive stages. Somewhat related to this is the personal investment in a program which results from choosing a leader for it who has a chance for promotion. His opportunity is therefore directly dependent on the success of the program, a good policy for most tasks but not necessarily where it might be to the nation's benefit to drop the project. An officer in this circumstance normally will be a partisan for the program at all costs, in every sense of the word, instead of the objective manager he should be. This practice alone may be the major reason that the Air Force has developed a reputation for keeping unpromising projects alive far too long.

Another personal influence in the equation has been the relation between military officers in procurement positions and the aerospace firms. Many procurement officers have retired from the service and then gone to work for an aerospace company. This raises serious cost implications: how well does the officer protect the government's interests when dealing with a potential future employer? How much influence does the ex-officer exert on friends and acquaintances still in the service? Vincent Davis' study, The Admirals' Lobby, concludes that the ex-Navy officers' influence has been nil.¹⁰ Other evidence may be found in the cases of two former Air Force officers who became company presidents. One was General Oliver P. Echols, who had an undistinguished career with Northrop from 1949 to 1954; the company was in relative decline under him. The other was General Joseph T. McNamey, once commander of the Air Materiel Command when it was the USAF's procurement organization. He was hired as president of Convair, ostensibly to give General Dynamics the ability to think ahead of the Defense Department. However, McNamey was given "credit" for doing most of the selling on Atlas, the B-58, and the F-I02, so persuasion may have been his principal function.¹¹ He gave an inept defense before Congress in 1956 of the practice of a military procurement officer's becoming an executive of an aerospace company; it is probably a fitting commentary on the practice. Part of the exchange went as follows:

MR. [Congressman F. EDWARD] HEBERT:

. . . Your [McNARNEY'S] previous defense. . . of the B-36 and your consistent appearance in the front ranks of those advocating or recommending the continuation of the controversial B-36 contract with Consolidated has nothing to do with your almost immediate employment by the Consolidated Co.?

MR. McNARNEY:

That is correct sir.

MR. HEBERT:

Did you at any time ever consider that perhaps such a connection, an almost immediate connection, might make you suspect?

MR. McNARNEY:

No, I never considered that. I felt that my reputation in the services was such that a thing like that would never be charged against me.

MR. HEBERT:

We are getting nowhere fast. But I think your replies will indicate exactly what you have said. And I can understand now, General, how you saw nothing wrong or did not think anybody else would see anything wrong in your taking employment with a company like Consolidated in view of the previous happenings in connection with the B-36. Such naivete is refreshing, but I don't think practical.¹²

Some U.S. government practices have been even more directly connected to aerospace costs. It has already been said that aircraft selection has been made in the Air Force by a group interested principally in the big bomber. That such a limited view was a handicap in choosing fighters, attack aircraft, or transports should be apparent; the advanced technology of planes demands expert knowledge because of the degree of specialization in the types. One need only review the difficulties that airline experts have had in ordering the less complicated airliners to see that selection of fighters by bomber pilots was a gross error.

Related to the basic big bomber bias was the casual attitude toward weight, which is expensive, and the desire to add every gadget known to a military aircraft to increase its "performance," while neglecting the importance of weight as a factor in how well a plane flies. Perfection has been sought without regard for the high cost of attaining that final 5, or even only 1, percent increment. In the late fifties Northrop asserted that going for' the final 5 percent would double development costs, and it designed its successful F-5 without that last 5 percent.¹³ Perhaps the clearest example of the Air Force's attitude on weight is the B-66, which was originally designed for the Navy's carriers as the A-3. The Air Force liked it and ordered its own version. Since the structural requirements for a land-based aircraft are less than for a carrier-based one, the B-66 should have been lighter than the A-3. But after the Air Force had insisted on weight-producing additions, the B-66 ended up much heavier, slower, and costlier than the A-3. This implies that the Navy is free from the gadget and weight complex, but reference to the isolated case of the A-4 and general aircraft development indicates that the Navy has the same bias but to a lesser degree.

That gadget and weight increases raise costs greatly has been of little concern to the military, who for decades have been concerned with attaining the last ounce of operational ability at the earliest time. This has a rational basis, for a slender margin of equipment performance can be the source of victory. But that costly high quality can be overemphasized has so far not been accepted, nor has the fact that high costs may limit . the quantities that can be bought to inadequate levels. As indicator of a lack of concern for costs is use of letter contracts because they have the fewest controls. They originated as a means to expedite production regardless of cost during the panic days of June 1940. Because they can be useful in saving small amounts of time, they are still of potential value. But in 1968, a time of wartime equilibrium, they were used 1,500 times for contracts totaling $7 billion. This is not use but abuse.

The airlines have had the restraints of the profit system, yet they have contributed to the costs of airliners in their insistence upon product differentiation. Some of this is probably unavoidable, but it appears to have been carried too far. Boeing estimated that nearly one-third of its loss on the Stratocruiser was due to catering to the individuality of airline desires. The cost for individual interiors alone on the Stratocruiser was $8 million. For the 737 Boeing offered 100 standard options. The 727 had 6,000 customer-peculiar parts. For its entire commercial transport line there were 21 models with 9 lengths, 10 wing styles, and 21 different engines, for a total of 113 customer varieties or possible combinations.

THE COMPANIES AS THE CULPRIT

After the fatal Apollo capsule fire of 1967, North American was accused of gross inefficiencies and padding payrolls; the management of the space division was sacked. The Lockheed C-5A scandal appeared to reveal gross inefficiencies there as well, and there was a minor shakeup in Lockheed's staff. Both Lockheed, with its C-5A and Cheyenne contract losses, and Grumman, with its F-14, have obviously been shocked to be held to contracts in this new era of anti-militarism; and there is a basis for saying that the government even broke the original contracts. There is no doubt they have been treated roughly as compared to the manner to which they had become accustomed, and Lockheed was forced to take a $200 million loss despite the re-determination clause of the C-5A contract. The shock must have been all the greater to the companies because, by comparison with earlier programs in the industry that averaged 220 percent overruns, the C-5A and F-14 have been conservative in their overruns. A perusal of programs from Apollo to the F-14, therefore, reveals inefficiencies by both government and management, resulting in excessive costs.

Better evidence has come from the upheavals of the failed companies and those in recent crises, beginning with Republic in 1965. After Fairchild had taken over, it found Republic's overhead to be excessive. McDonnell has had the reputation of being the most efficient producer in the industry. Within six months of its takeover of Douglas in 1967, deep cuts had been made in the latter's engineering and indirect staff despite the need to expand jetliner output:- David S. Lewis, a production man, was put in charge of the Douglas operation, and changes were made in top and middle management. From January to August 1968, 10,000 man-hours were pared from DC-9 production, and the number of men on final assembly was cut from 2,400 to 1,600. Some of this reduction was made possible by growing worker experience and not just improved management. Earlier, the inefficient practice of work done late and away from the proper production station had been followed by over 200 manufacturing personnel.

In 1971 Lewis took over General Dynamics in the wake of its shipbuilding and Datagraphix woes. He found a lack of control over procurement and inventories. And this is what he said about his Pomona Division, which was manufacturing the Standard missile: "The production line looked like a supermarket with pieces scattered here and there. There was no flow. . . ."¹⁵

Similar actions were being taken at North American. As its business slid after the Apollo disaster, Chairman Willard F. Rockwell, Jr., sent in some men from the automobile industry to do something about costs at North American. They were Robert Anderson, with twenty years at Chrysler, and Wallace W. Booth, with twenty years at Ford. Other auto men were hired to work at lower levels. The auto men were surprised at the laxness towards costs they found. The company had been centralized but responsibilities were vague. Controls were inadequate and cost estimating was so casual that programs worth millions of dollars had been started without consideration for returns. The space division, for example, had launched one program for which no market was known. The auto men had cut the overhead from 1,100 men to 600 by 1972, and had slashed inventories, receivables, and plant by $200 million. McKinsey and Company, management consultants, were brought in to improve the avionics business. McKinsey found it to be top heavy. Seven subdivisions were cut to two, and a campus-like manufacturing plant was put up for sale. The North American experience well illustrates the utility to the aerospace business of an alliance with the auto industry.

When Boeing's sales dived, going into the seventies, it cut itself so much that the company believed it had carried through the deepest and most successful retrenchment in business history. In the process it learned that it had been carrying much fat and could produce aircraft with many fewer workers than believed previously. Delivery time on 727's and 737's was slashed from seventeen to eleven months, reducing inventories. Engines had been held for four months, and this time was reduced to two to four weeks. John E. Steiner, vice president and general manager of the 707/727/737 division, said, "You may ask why the hell didn1f we do that earlier, and the answer is I don't know. We never had to. We could have done better" (emphasis added).¹⁶ An analysis of production workers found that they were at their work place only 26 percent of the time because of associated activities such as obtaining approvals. Work and layout changes raised the percentage to 70. These experiences are not unusual for companies cutting back in a recession and discovering accumulated fat, but they seem excessive in the aerospace industry.

A major share of the rise in costs in companies must be assigned to the expanding share of overhead expenses. Much attention has been given to the industry's increasing technology and to the resultant increase in scientists and engineers compared to production men. From 1955 to 1961 there was a 113 percent increase in scientists and engineers. But in the same period overhead personnel rose by 163 percent. Finally, in their periods of rapid expansion, the aerospace companies have probably promoted many men beyond their capabilities.

The rank and file personnel of the aerospace companies have not been as productive as possible. Foremen overmanned their shops in the industry because they were more subject to criticism over missed deadlines than over escalating costs. Overmanning hurts morale and has a spiraling effect on inefficiency. Yet aerospace management has lacked diligence in facing up to the workers over work standards and other cost matters. Morale has probably fallen progressively as the intermittent layoff and rehire cycle has gone on and on. In 1966 one of Douglas' major problems was hiring skilled workers, and the inefficiency of the inexperienced workers who were hired, and their training costs, added greatly to Douglas' desperate financial condition. Five years later, in the midst of a recession which had hit the aerospace industry and Southern California particularly hard, Lockheed found it to be most difficult to hire even a modest number of workers. The problem has been a continuing one since World War II, but Lockheed's experience indicates that it may be getting worse. The cumulative numbers of "fed-up" workers who will not tolerate the risk of another layoff must represent hidden, but very real, costs for the aerospace companies. The problem will affect even those who are not laid off, for they cannot be sure of their jobs. The human response, according to William McDonald Wallace, a management economist consultant, is make-work; and he asserts that the white-collar workers in the aerospace industry have been an example of a make-work group.¹⁷

COMMON EXPLANATIONS

The rapid growth of technology is given as another reason: By this is meant the proliferation of gadgets, primarily electronic. Their impact, as discussed above, is not only in their own cost but in what they do to the rest of the airframe in adding structure, and in the fact that they make the airframe much more dense, increasing production costs. This argument has validity as well, but its problem as an explanation lies not only in its inapplicability to the A-1 case but also in its assumption that there is justification for the gadget proliferation. Many gadgets, this author strongly believes, are superfluous "gold plating." The firms themselves bear some responsibility for gadget proliferation: there is a saying in the industry that an elephant is a mouse designed and produced under a cost plus fixed fee contract to military specifications. Some arguments in favor of keeping the gadgets sound desperate. An Air Force colonel, H. J. Sands of the Air Research and Development Command, alleged in 1952 that the gadgets were necessary because inexperienced pilots could not fly simple aircraft.¹⁸ Undoubtedly, some of the gadgets are necessary, some are useful, but some have a marginal return which is less than their marginal cost. The growth of technology is, therefore, a valid partial cause but is not as important as claimed.

In some specific contracts change orders have been advanced as the reasons for cost overruns. Yet change orders have always been a fact of life in the aircraft industry. This argument assumes that change orders always increase costs; but some, at least, must represent better methods and therefore act to reduce expenses. Also, the basic problem with assessing the impact of change orders goes deeper. The services have used change orders as a device to bail out contractors who have incurred more costs than officially forecast. The excess costs have been loaded onto the next change order in these cases. The process is called "contract nourishment," and the practice makes change orders indeterminant as an impact on costs. Further, there remains the suspicion that the only basic reason for some changes has been to serve as a vehicle for contract nourishment. For example, in 1954 it was estimated that only 10 to 15 percent of design changes were mandatory. In 1968 alone the Pentagon issued 12,563 change notices costing $2.6 billion, and these orders were not nickel and dime matters. There were 3,000 change orders for the C-5A. In contrast the excellent T-38 had less than 60 engineering changes before entering service in 1961, and the successful C-141 project had only 25 design changes over a roughly comparable two and a half years. It seems likely that far too many change orders represent poor initial engineering, the addition of frills, the failure of the customer to adequately anticipate his wants, contract nourishment, or a combination. These are hardly causes for unavoidable rises.

Labor trouble and subcontractor inefficiencies have also been blamed. These factors add to costs, but there has been no indication that they have been other than of minor impact.

The Pentagon has blamed stretch-outs for cost rises. This is valid, when it occurs, for the stretch-out increases the share of cost going to overhead. Another excuse has been to say there was overoptimism early in the program, or an honest mistake. But, because of "buying-in" practices, brochuremanship, and the Pollyanna nature of the briefing system, the overoptimism must also be regarded as a culpable failing and not always an honest mistake.

One of the most valid alleged causes of excessive costs is the stop-and-go nature of some programs. Perhaps the most expensive example of this was the B-58, which suffered from a struggle in the Air Force between the big bomber generals and others who wanted the particular performance the relatively small bomber could give. Obviously, this problem is generated entirely by the government. Also, as the B-58 example shows, it is nothing new, and there has not been any noticeable recent increase in indecisiveness. Stop-and-go is sometimes a valid cause of increased cost, but not of the excessively rising rate.

Related to stop-and-go as a cause for higher costs are the shorter: production runs of today. Some airframes continue to be built in long production runs, but most do not. Short runs means production takes place entirely in the most expensive beginning stages.

Fitzgerald reports that the Pentagon acted defensively to set up a group which tried to find that the sharply rising costs were a result of "Inexorable Economic Process." They failed.¹⁹ Critics from outside the government and industry have often blamed at least part of the high costs on "duplication." By this is meant competitive, parallel development or production of the same type of weapon because separate services exist. What appears to be true is not so in this case. In the case of novel designs, deliberate parallel efforts at the beginning and cheapest stages of development improve estimates and provide the best chance of determining the ultimately optimum project. Frequently, the initial ideas are so dissimilar that it is reasonable to assume that some will not even be practical, but sometimes this cannot be determined at an early stage. Thus, conducting projects in parallel, to a point, is a major cost reduction device rather than a waste.²⁰

POSSIBLE CONTROL CASES

There may be no airframe firm superior to the exemplary Avions Marcel Dassault. Its work force has run between 3,000 and 8,000, a contrast to Boeing's 58,000 in 1972 after its cutbacks, and McDonnell's 86,700 in 1972. In the quarter-century since World War II this meager by American standards-work force has produced 200 Hying prototypes and 3,000 production aircraft of different designs, from fighters to supersonic strategic bombers. The production figures must take into consideration that Dassault subcontracts extensively. All designs have been technologically advanced. The excellent Mirage III was produced by only 25 engineers, 50 draftsmen, and 100 craftsmen in less than 13 months from contract to first flight. The Mirage IIIC variant cost only a little over $1 million. The Mirage G was developed to first flight in 16 months at a total cost of only $35 million. Evidently, Dassault's engineers do not spend their time designing brochures instead of airframes. It is significant that Dassault, too, thinks U.S. designs are "gold plated."²¹ The Russians use design teams like Dassault's.

Perhaps the best control case is the General Dynamics prototype, the 1965 Charger, built to compete with North American's OV-10. The Charger was built with company funds, and the OV-10 was a regular government program. The performance of the two aircraft differed little, but the development of the Charger cost only one-fifth that of the OV-10. The Defense Department bought the OV-10 because of the time and money sunk into it.

There was one isolated development designed to test the drawbacks of our heavy-handed bureaucracy, This was the 1961 development of Lockheed's Agena D space vehicle. The Agena D team was physically isolated for the development. Only six Air Force representatives were allowed into the work area, and they had highly unusual powers of decision in the Air Force areas of interest, including funds. The Lockheed counterparts were given direct access to their top management. Deliberately eliminated were the customary technical directive meetings, visits by groups, and the job of keeping persons informed who had little relation to the development. Detailed reports were not made. Although the project required a "modest" advance in technology, it used one-fourth of the engineers originally proposed, and was completed in half the scheduled time and at half the estimated cost. Despite the achievement, the method was not repeated. The Air Force maintained the success was due to a lack of change orders. As shown by the "contract nourishment" practice, the absence or presence of change orders means nothing in terms of the necessity for technical change. The Air Force judgment, therefore, although it cannot be disproven, seems inadequate.

Most of the above material concerns developmental costs, but there is one good comparison on production. It is old, but it is such an extreme case it probably still has validity. During the Korean War, the F-84F proved difficult for both Republic and General Motors to get into production. But by its last assembly run, General Motors had drastically reduced production time and costs, producing high profits. Most of its savings came from cutting production time and, therefore, overhead costs. Its learning curve was steeper than custom my in manufacturing airframes. As a final achievement, General Motors' F-84F's were superior in quality to those of Republic, the originator. This achievement by General Motors was a repeat of its performance in World War II. In December 1941 Republic's plant at Evansville began production on P-47's. One month later General Motors began on F4F's and TBF's in two different plants. It took Republic 24 months to produce its one thousandth aircraft at Evansville; General Motors took 21 months to reach its thousandth F4F, and 23 for the thousandth TBF. Since Republic had a high reputation for efficient airframe production in the war, the results may indicate superior management attitudes and methods in the auto industry. It appears that the aerospace firms are not doing as well as they could in reducing or holding down production costs.

General Dynamics' Charger. The development of this aircraft as a company project cost one-fifth that of the competitive North American QV-10 built under government contract. Aircraft performance was similar. Courtesy General Dynamics Corporation.

In 1965 Professor Rennan O. Steckler claimed that defense contractors had about 25 percent more costs because of an absence of adequate competition. Many estimates were made later. Fitzgerald thought it was 50 percent. Air Force Colonel A. W. Buesking, who had extensive experience in military procurement, told Congress he believed that costs were 30 percent to 50 percent excessive. Professor Frederic M. Scherer estimated 50 percent in 1972. Pierre Sprey, manager of the Systems Division of Enviro Control, Inc., told the Senate in 1972 that our fighters should cost half the price then current. Robert Perry of RAND told the Senate that the U.S. could reduce development costs by 25 percent to 50 percent. Leonard Sullivan, principal deputy director of Defense Research and Engineering, estimated 35 percent in 1973.²²

All in all, there seems to be no doubt that American practices have resulted in excessive costs through overmanagement and overmanning, caused by a wasteful government bureaucracy and acquiesced in by company managements.