My Years With General Motors

by Alfred P. Sloan Jr.

  It is obvious that, with stable material costs and wage rates, if our bench mark showed that unit costs were high in relation to price, efficiency had been reduced. Because of competitive resistance to price increases, profits could be maintained only by reductions in unit costs. If there was a general rise in what industry paid for materials and labor, competitive forces might allow prices to rise. This was likely if consumer demand for cars was strong. Without higher prices under the circumstances, the automobile industry could not long continue to supply the cars that the market wanted. However, even in this case the individual manufacturer was under pressure to reduce his unit costs, since competition would probably not allow price increases sufficient to recover the full cost increase.

  An alternative approach to our standard-volume policy would have been to evaluate prices strictly in terms of actual unit costs at actual or anticipated production levels. Because our fixed costs were so large, this would have meant that unit costs would drop in times of high volume and increase during periods of low production. Any attempt to raise prices during periods of low volume, even if competition permitted, to recover the higher unit costs could have deflated sales still further, with the result of still lower profits, less employment, and a generally depressive effect on the economy. Operating as we did in a highly cyclical industry, the use of the actual unit-cost type of evaluation would have been socially and economically unsound. I want to make clear, however, that in any given year our income, which had to reflect all actual costs, was importantly affected by the volume attained. Fixed costs had to be met whether business was good or bad. If our volume fell below standard, only a portion of our total fixed costs could be allocated to the unit cost of production, but the unallocated remainder would have to be deducted in arriving at reported income. Conversely, if volume was above standard, total income would be increased because the fixed costs would be distributed over a greater number of units produced.

  It should be apparent from the foregoing that profit is residual, based upon the ability of the manufacturer to keep his costs below the sales price established in a competitive market. That is, profit represents the difference between the price which can be obtained in the competitive market, and total cost. And it is substantially affected by volume. We can estimate very closely what our profit per unit should be at standard volume, but this is not the same as knowing what our actual profit will be when actual volume is realized. Profit is a variable, and a volatile one, in the automobile business.

  The need for financial controls grew out of crises. Controls were brought in to ensure that crises did not recur. Their effectiveness was demonstrated particularly in the depression year 1932. The corporation's U.S. and Canadian unit volume in that year was 50 per cent less than that of 1931, and 72 per cent below the high of 1929. But the corporation was not demoralized as it had been in 1920 and it stayed in the black. Not many corporations did as well.

  Financial control as worked out by General Motors gave the corporation a review of operations that reduced the need to administer operations from the top. Central-office management was able to know whether the decentralized management was operating well or poorly and had a factual basis for a judgment regarding the future of any particular part of the business. We had the fundamentals of this system worked out just in time for one of the greatest changes that has ever taken place in the automobile market.

  Chapter 9 - Transformation Of The Automobile Market

  By the middle of the 1920s General Motors had accomplished some things, but apart from survival and reorganization, they were more in the realm of the mind than of reality. We knew, as I have related, the strategy with which we proposed to approach the car business, how we proposed to manage the enterprise financially, and the relationships we wanted to establish among persons in different roles. But by the end of 1924 little of this was reflected in our activities in the automobile market. That our volume of business had increased after the slump of 1921—and especially in 1923—could be attributed less to our own wits than to the improvement in the general economy and the rising demand for automobiles. While internally we had made much progress, externally we had marked time. But the time had come to act.

  Now it so happened—luckily for us—that during the first part of the 1920s, and especially in the years 1924 to 1926, certain changes took place in the nature of the automobile market which transformed it into something different from what it had been all the years up to that time. (Seldom, perhaps at only one other time in the history of the industry—that is, on the occasion of the rise of the Model T after 1908—has the industry changed so radically as it did through the middle twenties.) I say luckily for us because as a challenger to the then established position of Ford, we were favored by change. We had no stake in the old ways of the automobile business; for us, change meant opportunity. We were glad to bend our efforts to go with it and make the most of it. We were prepared, too, with the various business concepts which I have described, though I must say we saw them as merely our way of doing business and not as having any general application or logical involvement in the future of the industry.

  To set the scene, let me divide the history of the automobile, from a commercial standpoint, into three periods. There was the period before 1908, which with its expensive cars was entirely that of a class market; then the period from 1908 to the mid-twenties, which was dominantly that of a mass market, ruled by Ford and his concept of basic transportation at a low dollar price; and, after that, the period of the mass market served by better and better cars, or what might be thought of as the mass-class market, with increasing diversity. This last I think I may correctly identify as the General Motors concept.

  All three of these periods have in common the long-expanding American economy, the horizon of each period having been formed by the respective degrees of that rise and its spread through the population. The willingness of the relatively few who could afford them to buy expensive though unreliable cars—by today's standards—enabled the industry to get going. Then when a large number of individuals were able to afford a few hundred dollars of expenditure, they made possible the development of the inexpensive Model T (it is possible that such a market was long waiting for the offering of a car like the Model T). As the economy, led by the automobile industry, rose to a new high level in the twenties, a complex of new elements came into existence to transform the market once again and create the watershed which divides the present from the past.

  These new elements I think I can without significant loss reduce to four: installment selling, the used-car trade-in, the closed body, and the annual model. (I would add improved roads if I were to take into account the environment of the automobile.) So embedded are these elements in the nature of the industry today that to conceive of the market without them is almost impossible. Before 1920 and for a while thereafter the typical car buyer was in the situation of buying his first car; he would buy it for cash or with some special loan arrangement; and the car would be a roadster or touring car, most likely of a model which was the same as last year's and could be expected to be the same as next year's. This situation was not to change for some years and the change would not be sudden except at its climax. For each of the new elements of change had a separate beginning and rate of development before they all interacted to cause complete transformation.

  Installment selling of automobiles in regularized form first appeared in a small way shortly before World War I. This form of borrowing, or inverse saving, when placed on a routine basis, enabled large numbers of consumers to buy an object as expensive as an automobile. The statistics of installment selling in those days were very poor, but it is clear that it grew from some very low level in 1915 to around 65 per cent for new cars in 1925. We believed that with rising incomes and the expectation of a continuance of that rise, it was reasonable to assume that consumers would lift their sights to higher levels of quality. Installment selling, we thought, would stimulate this trend.

&nbs
p; As the first car buyers came back for the second round and brought their old cars as down payments, the custom of trading was established. That the industry was engaged in a trading business had revolutionary significance not only for dealer arrangements but for manufacturing and the whole character of production, since dealers usually had to sell to a man who already had a car with mileage left in it.

  The statistics for used-car trade-ins before 1925 are as poor as those for installment selling. It stands to reason, however, that there was some kind of upward curve in used cars traded from World War I on, if only because there were relatively few cars in existence before that time. Until some unknown date in the early 1920s, the majority of car buyers were buying their first car. The total number of passenger cars in operation in the United States from 1919 through 1929 rose by years in millions approximately as follows: 6, 7.3, 8.3, 9.6, 11.9, 13.7, 15.7, 16.8, 17.5, 18.7, 19.7. The industry, on the other hand, produced in those years passenger cars for domestic and export markets in approximate millions as follows : 1.7, 1.9, 1.5, 2.3, 3.6, 3.2, 3.7, 3.7, 2.9, 3.8, 4.5. (Note 9-1.) This production was enough to cover both the growth in numbers and the scrappage. The used car was traded perhaps two or three times on the way to the scrap heap. So I assume there must have been a rising curve of used-car trade-ins.

  The closed body was a specialty and mainly a custom-job affair before World War I. In the years 1919 through 1927, in round numbers by years, the industry sold closed cars in the following uninterruptedly rising percentages: 10. 17 . 22. 30. 34, 43, 56, 72, and 85

  Of the annual model I shall say more later; suffice it to say here that in the early twenties it was not a formal concept as we know it today, except as it was negatively expressed in Ford's concept of a static model.

  We were not unconscious of the unfolding of these four elements when the administration of General Motors changed in 1921. We started GMAC in the installment financing field in 1919. We had an interest in Fisher Body, which made closed bodies. As large sellers of medium and high-price cars, we met the used-car trade-in early. And we tried to make our models more attractive each year. Yet we did not see the movement— especially the interaction—of these elements in the whole automobile market as I can see it today looking back. We saw them then as uncertainties, unknown and trends, in the form of figures to study at a desk. However, the plan of campaign laid down in the product program of 1921 logically fitted better and better the unfolding situation.

  It was that plan, policy, or strategy of 1921—whatever it should be called—which, I believe, more than any other single factor enabled us to move into the rapidly changing market of the twenties with the confidence that we knew what we were doing commercially and were not merely chasing around in search of a lucky star. The most important particular object of that plan of campaign, which followed from its strategic principles, was, as I have said, to develop a larger place for Chevrolet between the Ford car below and the medium-price group above, a case of trying; to widen a niche. That was all, in the beginning, despite the completeness of the plan with regard to the whole market.

  There was the pause while we settled the copper-cooled-engine matter, in which we gave up die commercial-mindedness of our original strategic plan to pursue an engineering dream. We were rescued from that folly by the four-million car-and-truck year of 1923, which absorbed some 450.000 Chevrolets. and we saw the illusions of the upward swing; of that year dashed in the recession of 1924. It was thus made clear to us that the plan of 1921 would have meaning only if meaning were given to it in the design of the product itself.

  Certain facts of failure in particular were impressed upon us. During the year 1924, while the industry's passenger-car sales in the United States fell 12 per cent, General Motors' sales fell 28 per cent. Of the industry's decline in sales of about 439,000 car units, almost half was represented by the decline in General Motors' car sales. Our share of the passenger-car market in units dropped from 20 per cent to 17 per cent, while Ford's share went up from 50 per cent to 55 per cent. Some of the General Motors decline was in Buick and Cadillac, as was to be expected of higher-priced cars in a period of economic recession. (Olds increased. Oakland was unchanged.) But most of it was in sales of the Chevrolet, which fell 37 per cent while sales of its opposite number, the Ford, fell only 4 per cent. Of course, what happened was not due entirely to the events of 1924, including some bad management, but to the recession of that year combined with earlier events. The lag between automotive design and production is a peculiar feature of the automobile industry. The events in a current year are always in part due to decisions taken from one to three years earlier. Hence the extent of the Chevrolet slump of 1924 could properly be laid to the retarded development of Chevrolet's design during the previous three years. Among other things, it had an infamous rear end; but there is no use specifying its deficiencies. The curious thing was that there we were with a plan that rested upon the concept of better and better cars, with a bigger package of accessories and improvements beyond basic transportation, and the concept of a Chevrolet at a higher price that would be so compellingly attractive as to draw buyers away from the Model T. It would be difficult to find a wider margin between aspiration and realization than that represented by the plan of 1921 and the Chevrolet of 1924. Nevertheless, we did not alter the original plan, perhaps because we knew better than anyone the causes of our decline.

  Indeed from the time the copper-cooled-engine program was abandoned, in the summer of 1923, Chevrolet's engineers, headed by Mr. Hunt, had worked intensively on redesigning the old car into a new model, known as the K Model, for the 1925 model year. The K Model had among its new features a longer body, increased leg room, a Duco finish, a one-piece windshield with automatic wipers on all closed cars, a dome light in the coach and sedan, a Klaxon horn, an improved clutch, and a sound rear-axle housing in place of the old one which had given so much trouble. It was far from being a radically new car but it was much better than it had been, and in the particulars noted above it gave the first real expression of what we had in mind to do. The K Model came on a rising market in 1925 and recovered Chevrolet's position sharply with factory sales of 481,000 cars and trucks, a 64 per cent gain over 1924 and a level 6 per cent above the 1923 peak.

  Ford's sales held about even in 1925 with a volume of about two million cars and trucks. But since the market as a whole in that year rose substantially over 1924, Ford's share declined relatively from 54 to 45 per cent, a sign of danger, if Mr. Ford had chosen to read it. Yet he still held almost 70 per cent of the low-price field, and his touring car, priced at $290—without a starter or demountable rims—seemed unbeatable in that area. The Chevrolet touring car in 1925 was selling at $510, though with its extras it was not exactly comparable with the Ford. The Ford sedan—with starter and demountable rims—then sold at $660; the Chevrolet K Model at $825. Chevrolet's dealer discount was larger than Ford's, which made a difference in trading.

  Chevrolet's internal statement of policy at this time was that it was our objective to get a public reputation for giving more for the dollar than Ford. As a matter of fact, when the Ford and Chevrolet were considered on a comparable-equipment basis, the Ford price was not far below that of Chevrolet. On the quality side we proposed to demonstrate to the buyer that, though our car cost X dollars more, it was X plus Y dollars better. Too, we proposed to improve our product regularly. We expected Ford, generally speaking, to stay put. We set this plan in motion and it worked as forecast.

  Nevertheless, despite the success of the K Model Chevrolet, it was still too far from the Ford Model T in price for the gravitational pull we hoped to exert in Mr. Ford's area of the market. It was our intention to continue adding improvements and over a period of time to move down in price on the Model T as our position justified it.

  As we said in our product policy of 1921, any given car was related to other cars that impinged upon it below and above in price and engineering design. Hence when looking at the Chevrolet in relat
ion to the Ford below it, it was logical to consider equally what might happen to Chevrolet as a result of similar actions by competitors above it. This question was very much on our minds while the Chevrolet 1925 K Model sedan was being prepared during the year 1924, and for good reason.