My Years With General Motors

by Alfred P. Sloan Jr.

  The next step after the K 2 Rig therefore consisted in building two experimental Cadillac cars . . . These had two different independent front suspensions . . . [One of these was that developed by Mr. Dubonnet; the other, the "wishbone" type, we had developed.] An independent rear suspension was also used, as we had in mind that, as soon as possible, we should also get rid of the conventional rear axle (a change which in my own opinion is now several years overdue).

  On these cars which were ridden by many of the Corporation engineers, it was evident that we had something very special in the way of improved ride and handling. We also ran into our usual share of troubles. The chief of these was the steering, which, especially on the wishbone suspension, was not free from shimmy.

  We had to redesign the steering mechanism several times . . .

  Finally, by March of 1933, we were ready for a full-dress demonstration. Early in March the General Technical Committee met at the Cadillac Engineering Building to ride our two experimental cars, and a Buick car without independent front suspension, but with an I.V. [infinitely variable] transmission . . .

  I recall that [you] and Mr. Grant were riding one of the [wishbone-type] cars, when Ernest Seaholm and I, in one of the accompanying cars, pulled up alongside [you] at the traffic light in River Rouge. We could see [you] smiling widely at Dick Grant [vice president of sales] in the rear seat, and moving the flat of [your] hand up and down [and] horizontally. Within two miles from the Cadillac plant the flat ride had sold itself!

  After the run to Monroe and back on the three cars, the Committee sat at the Cadillac plant, and Seaholm and I, in the background, awaited the verdict, with the pious hope that Cadillac would be granted a clear year's run on the new suspension, ahead of the other divisions.

  O. E. Hunt [vice president of engineering], I recall, led off by asking Mr. Grant what he thought of the new automatic transmission.

  You will recall that in March of 1933 there was not a bank open in the United States, and anyone who owned a farm was thankful that at least he could eat. Under these circumstances Dick Grant's reaction was not surprising. He turned down the [automatic] transmission, and the hundred dollar cost that went with it, as something that a Buick buyer could very well do without. "But", said he, "if I could have a ride like you've shown us, for a matter of fifteen bucks, I'd find the money somehow."

  Dutch Bower [chief engineer] at Buick had already put in his claim for the new front suspension, and the Oldsmobile and Pontiac engineers also seemed determined that they would show it in New York next November.

  Then finally Bill Knudsen [the general manager of Chevrolet] declared in words of one syllable that Chevrolet [was] not going to be left out. O. E. Hunt tried to persuade him that there were not enough center-less grinding machines available in the United States to grind the wire for the coil springs for Chevrolet. But Knudsen was adamant, saying that the machine tool industry had been in a bad way for years, but they were going to be busy for the next year at least. And Chevrolet actually made the New York Show in November with their 1934 model on the Dubonnet suspension. Pontiac also inherited this suspension from Chevrolet, while the three other divisions adopted the wishbone suspension.

  This meeting stays in my mind because it was such a tremendous demonstration of American enterprise in action. In the face of the conditions then existing, the millions of expenditure to which the Corporation was committing itself argued a type of courage which was new in my experience. I still remember Ket's statement, "It seems to me we can't afford not to do it."

  We thus introduced simultaneously two different types of independent front-wheel suspension. However, after some further improvements on the wishbone type, it became apparent that it was cheaper and easier to manufacture and more trouble-free in operation, and soon all our lines of cars adopted it.

  Duco

  One of the striking scenes of America today viewed from the air in the daytime is the splash of jewel-like color presented by every parking lot. The colors are of an enormous variety, and the finishes are nearly indestructible.

  All this is in contrast to the appearance of automobiles in the early twenties, when Ford, Dodge, Overland, and General Motors were using only black enamel on high-volume jobs. The external finish was then a subject of general complaint. The practices of the carriage industry had been carried over into automobile manufacturing without much change; automobiles for the first twenty five years of their existence wore carriage paint and varnish. The customer could not understand why the finish of a carriage lasted for a long time, while when he bought a car the paint would sometimes soon peel off. The fact, of course, was that the carriage and the motorcar were very different mechanisms. The automobile was subject to much harder service; it was used in more kinds of weather, and the heat of the engine produced temperature changes in parts of the car—with a resultant disastrous effect on the finish.

  We dreamed of what a wonderful thing it would be if a finish could be developed which would last even if the car stood out in all kinds of weather. We also began to realize that a good, fast-drying finish could revolutionize our time schedules and the consequent cost of production.

  The finishing process at that time, using paint and varnish, was slow and cumbersome. Between the time a car was ready to be finished and the time the job was completed, something like two to four weeks went by, depending among other things on the temperature and humidity. It can readily be seen that this created a terrible inventory problem.

  For a while many automobile manufacturers shifted from paint and varnish to oven-dried enamels, in an effort to deal with some of these problems. The Dodge Brothers' open car, for example, was wholly oven finished with no paint or varnish. This was a black Gilsonite enamel which was very durable. However, oven finishing was only a transition—there was a better and cheaper answer to the problem.

  On July 4, 1920, more by accident, I think, than by intention, a chemical reaction was noted in one of the du Pont laboratories which led to the development of a nitrocellulose lacquer eventually called Duco. It was observed that a lacquer base could be created which would carry more color pigment in suspension, and produce more brilliant colors. Three years of experiment and development were required to get the bugs out of the new product. This was a co-operative project of the General Motors Research Corporation under the direction of Mr. Kettering and the du Pont laboratories. A Paint and Enamel Committee was organized in General Motors in 1921 (ironically both paint and enamel were soon to be superseded), and the first body finished in the new lacquer came off the production line in 1923. It was the "True Blue" Oakland of the 1924 line.

  The new lacquer product, under the trade name Duco, was made available to the entire motorcar industry in 1925. There were still many problems to be solved, and research continued in the du Pont and General Motors research laboratories. A very important part of this work was the development of undercoats, for Duco as first developed was not very adhesive and sometimes stripped from the metal. Duco also required the use of natural resins, which were limited in quantity and of variable quality. In time, the invention of synthetics relieved us of dependence on these variable natural products.

  Color had always been available in automobile finishes, both in the paint-and-varnish period and in the enamel period that followed it, but it was expensive and the range was limited. Duco, by reducing the cost of color finishes and increasing enormously the range of color that could be economically applied to cars, made possible the modern era of color and styling. Furthermore, its quick drying removed the most important remaining bottleneck in mass production, and made possible an enormously accelerated rate of production of car bodies. Today a car can be finished in an eight-hour shift, compared to the two-to-four-week period of the paint-and varnish age.

  Consider the saving in space alone: a production of 1000 cars a day once required space for 18,000 cars in process, since three weeks on the average were needed for the finishing work—that is, twenty acres of cover
ed indoor space. Think of what this would mean at today's production rates of 15,000 or more cars per day.

  Since the introduction of nitrocellulose lacquers in the twenties, there has been continuous study to improve them and to reduce the cost of application. In 1958 General Motors introduced a new line of finishes based on the acrylic resins. These again were the product of over eight years of research in our laboratories in co-operation with resin manufacturers. The acrylics are even more durable than the nitrocellulose lacquers and are capable of producing even more pleasing colors.

  There were many other important improvements in which General Motors played a key role. Crankcase ventilation in the 1920s eliminated one of the main causes of deterioration of the engine. "Internal" crankcase ventilation, which reduced air pollution, was pioneered by General Motors in 1959 and made available to the industry in 1962. The development of four-wheel and hydraulic brakes contributed greatly to the safer and more effective use of the motorcar. Four-wheel brakes were not an exclusive General Motors development, but we participated in improving them, helped develop volume production, and created a special division to manufacture them for our cars. The corporation also took a leading role in the development of power brakes, power steering, car air-conditioning, and innumerable other refinements of the automobile. These are only a few important selections from the results of the ingenious and untiring labor of many thousands of research workers, engineers, and others who have given their professional interest to the development of efficient and comfortable individual transportation.

  Chapter 13 - The Annual Model Change

  Annual car models are now such a natural and accepted part of American life that few persons, I would imagine, have thought about the vast effort of management that lies behind them. The procedure we follow in designing a typical American passenger car differs significantly from that followed for foreign cars and specially designed domestic cars.

  Each year we must produce a line of cars which embodies advanced engineering and styling features, and which will be competitive in price and meet the demands of the retail customer. The cars in this line must have some common styling features, giving them all a "General Motors look," but at the same time they must be clearly distinct from one another. They must also complement one another in price, which means that their own cost elements as well as the trend of competitive prices must be estimated well in advance of production.

  In General Motors there are thousands of persons—in addition to production workers—involved in the creation of the new models: they include style artists and engineers; scientists; financial and marketing experts; members of the technical staffs of the various divisions; and the general executives and staff technicians of the corporation, not to mention our outside suppliers. The problem of co-ordinating their varied activities is extremely complex.

  On the average about two years elapse between the time we make the first decisions on the new models and the time the cars appear in dealers' showrooms. Ordinarily, the sequence of events during these two years is determined principally by the requirements of body production. Body changes, of course, are usually substantial from one year to the next, and the body work takes the most time. There are continual changes in chassis components, too, of course, but only occasionally in any one year do we introduce changes in all the chassis units—the frame, engine, transmission, front and rear suspensions.

  To generalize broadly, the first year of the model development is devoted to laying out the basic engineering and styling characteristics of the new model; and the second year is devoted mainly to the engineering problems entailed in bringing the cars into full production. It is extremely difficult to get either of these jobs done in much less than a year's time. If we compress the time given to setting the basic style concepts we increase the danger of "locking ourselves in" with a product which will not meet the approval of the retail customer. And if we compress the engineering-production time we pay extraordinary overtime charges, create inventory problems, and possibly delay the time when we can start production—which in turn might mean delays in a car's announcement date and loss of sales.

  On the other hand it would be unwise to lengthen the time taken to produce the new models. There is, of course, no reason in principle why we could not begin to plan our model changes three or even five years in advance—and, in fact, we do some thinking that far ahead—but there is the practical difficulty that the planners are then remote from the realities of the market place in which their work will be tested. Even the two-year period usually required now imposes a severe strain on the corporation's ability to gauge the market correctly. The problem may be viewed this way: General Motors, like other automobile companies, is obliged to invest millions of dollars to devise new products, which cannot, however, be sold until a long period of time has elapsed. Meanwhile the consumers' taste, income, and spending habits may all have changed radically. For that matter, we cannot even be certain that the new model is "right" at the time it is first conceived. Responses to sketches, and to survey questions, are often undependable. It is an axiom of marketing research that automobile customers never know whether they like the product well enough to buy it until they can actually see the real thing. But by the time we have a product to show them, we are necessarily committed to selling that product because of the tremendous investment involved in bringing it to market. Every automobile manufacturer has on occasion been caught off base by the consumer. Nevertheless, in the nature of things, we must plan and co-ordinate our efforts in order to get to market with a new model.

  This very special kind of co-ordination has evolved out of the planning experience of many years. I have described the near disaster that General Motors experienced in 1921-22 because there was no established co-ordination procedure which would enable the several distinct management groups to work together on a new-model program. After that experience we gradually put system and method into the introduction of the corporation's models. In 1935, we set down on paper for the first time, I believe, a procedure governing the production of new models. It was a manual designed "to provide a definite and orderly method for submitting the essential data required, in order that the economic, financial, engineering and commercial position of proposed new products may be evaluated; and, second, that their progress, from the time of approval to production, may be established, for the information of all concerned." The product-approval procedure was substantially revised in 1946, and to some extent is continually changing. It should be emphasized that these written procedures are not an exact "timetable" to which our model runs must conform.

  In stating that the model-development periods average perhaps two years, I do not mean to imply that we start each model from scratch at the beginning of this period. The Styling Staff, for example, is continually experimenting with new designs for distant-future models, and at any point in time there is sure to be a sizable backlog of new styling ideas at our disposal, some of them quite conventional, some revolutionary. And each car division is continually engineering a variety of new features, mainly for the chassis. Some of these features may have been taken over from the Research Laboratories and the Engineering Staff or perhaps from the accessory divisions and refined in the car divisions to the point at which it seems feasible to introduce them into production models; others may have been developed entirely within the car divisions' own engineering shops and laboratories.

  Usually, the first formal meeting on a new model is preceded by many informal discussions. For example, the car-division management and the Styling Staff review the advantages and disadvantages of past production programs, they examine customer-research reports and market analyses, and discuss the general package size and styling concepts for the new cars to be designed. Some of these issues may be taken up with the central-office Engineering Staff and with Fisher Body Division, as well as with the chief officers of the corporation.

  Even though some important work on future models is always going on in the corpo
ration, most of us have come to think of each new model program as "beginning" with a meeting called by the Engineering Policy Group. The reader will recall that this group reports directly to the Executive Committee and includes the corporation's chairman, president, and the principal central-office executives. The group's chairman is the vice president in charge of the Engineering Staff. Since the group concerns itself with broad corporate policy, the membership does not include the general managers of the car divisions or Fisher Body, although these men and the chief engineers of those operations are often invited to attend the group's meetings for a review of the programs in which they are involved.

  The main business of this first meeting is to determine the outline of our styling and engineering programs, that is, to determine the cars' general appearance and size characteristics, and to indicate the direction of further styling and divisional development. The desired seat widths, head and leg room, as well as exterior height, width, and length are all considered. The Styling Staff displays full-size styling drawings, so that those present can get a feeling about the appearance, size, and roominess characteristics. Along with the drawings, we generally show full-size, dummy seating arrangements, commonly referred to as "seating bucks," which are constructed to simulate the proposed car interior. This "buck" allows us to check entrance-room conditions, vision, roominess, and seating position. The members of the group look over, you might say, what the stylists have to offer.

  In line with the ideas developed at this "kickoff" meeting, the Styling Staff progressively develops several series of full-size styling drawings, as well as full-size clay models and seating bucks for each kind of car in our line. In order to achieve the desired objectives of the program, and to keep abreast of tooling and manufacturing requirements, the Styling Staff must work closely with the car divisions and Fisher Body for many months following this first meeting. In general, it is the Styling Staff's responsibility to set the basic appearance of each kind of car. That is, the staff works out the basic appearance of the General Motors sedan, coupe, hardtop, station wagon, and convertible—and usually the work is done in that order. Each division has its own studio within the Styling Staff, and these studios are responsible for giving each car line its own distinctive features—for example, the features which distinguish the Chevrolet and the Pontiac lines.