The next point in the study was how to carry this philosophy into action. I wrote:
Having established the above principles as fundamental, and it is believed that all interests within the Corporation agree as to such principles, the definite objects which it is hoped to attain by this study, are enumerated as follows:—
1. To definitely determine the functioning of the various divisions constituting the Corporation's activities, not only in relation to one another, but in relation to the central organization.
That was a big chew, but it is correct. If you can describe the functions of the parts and the whole, you have laid out a complete working organization, for by implication the apportionment of responsibility for decisions at various levels is contained in the description.
I continued with the second objective:
2. To determine the status of the central organization and to coordinate the operation of that central organization with the Corporation as a whole to the end that it will perform its necessary and logical place.
This is a restatement of the first point, but in reverse—that is, looking from the top down.
The third objective:
3. To centralize the control of all the executive functions of the Corporation in the President as its chief executive officer.
Decentralization or not, an industrial corporation is not the mildest form of organization in society. I never minimized the administrative power of the chief executive officer in principle when I occupied that position. I simply exercised that power with discretion; I got better results by selling my ideas than by telling people what to do. Yet the power to act must be located in the chief executive officer.
The fourth and fifth points speak for themselves:
4. To limit as far as practical the number of executives reporting directly to the President, the object being to enable the President to better guide the broad policies of the Corporation without coming in contact with problems that may safely be entrusted to executives of less importance.
5. To provide means within each executive branch whereby all other executive branches are represented in an advisory way to the end that the development of each branch will be along lines constructive to the Corporation as a whole.
In brief, the study presented a specific structure for the corporation as it existed at that time. It recognized the form of the divisions, each of which was a self-contained group of functions (engineering, production, sales, and the like) . It grouped the divisions, according to like activities, and, as I said in my letter to Mr. du Pont, proposed to place an executive in charge of each group. The plan provided for advisory staffs, which would be without line authority. It provided for a financial staff. It distinguished policy from administration of policy, and specified the location of each in the structure. It expressed in its way the concept that was later to be formulated as decentralized operations with co-ordinated control.
The principles of organization in the study thus initiated for the modern General Motors the trend toward a happy medium in industrial organization between the extremes of pure centralization and pure decentralization. The new policy asked that the corporation neither remain as it was, a weak form of organization, nor become a rigid, command form. But the actual forms of organization that were to evolve in the future under a new administration—what exactly, for example, would remain a divisional responsibility and what would be co-ordinated, and what would be policy and what would be administration—could not be deduced by a process of logic from the "Organization Study." Even mistakes played a large part in the actual events, as I shall show; and if our competitors —Mr. Ford among them—had not made some of their own of considerable magnitude, and if we had not reversed certain of ours, the position of General Motors would be different from what it is today.
Although the plan was officially adopted in 1920, expediency was for some time to rule the corporation. The formation of the new Executive Committee was the first outstanding example of this. Its four members, who were charged with the guidance of the corporation, had never before had the responsibility of producing an automobile. The great automotive producers in General Motors had been Mr. Durant, Mr. Nash, and Mr. Chrysler. They had already, by 1921, made their mark in the leadership of the automobile industry, and owing to the turns of fortune which I have described, they were then, or were about to be, numbered among our competitors. Mr. Durant, soon after leaving General Motors, formed another company, Durant Motors, which in its time was to produce several cars, the Durant, the Flint, the Star, and the Locomobile (which he took over). Mr. Chrysler was currently engaged in rescue operations at Willys-Overland and at Maxwell—forerunner of the Chrysler Corporation—and Mr. Nash was operating the enterprise that bore his name.
On the other hand, look at the new management of General Motors. Pierre S. du Pont in his first five years as chairman of General Motors had left the operating end of the business briefly to Mr. Nash and thereafter to Mr. Durant. Mr. Raskob was a financial man. Mr. Haskell's contact with the business was brief and not directly in divisional operations; and he soon dropped out of intimate participation in the new operating responsibility. He died on September 9, 1923. I came nearest to having automotive experience, and though I had spent my life in the industry, I was still undeveloped in car operations. So we were four amateurs, so to speak, in comparison with Mr. Nash, Mr. Chrysler, and Mr. Durant; soon, on the active end, we were three; and since Mr. Raskob was in finance, the highest responsibility for operating the corporation fell upon two of us, Mr. du Pont and myself as his principal assistant. Mr. du Pont and I worked closely together, traveled together, and met with operating executives in Detroit every two weeks. After six months I became a sort of executive vice president in charge of all operations, reporting to Mr. du Pont. But there was no straight and clear line in this matter; at one point, for example, Mr. du Pont added to his burdens and to the complexity of the improvised management by causing himself to be appointed general manager of Chevrolet along with his chairmanship and presidency of the corporation.
If we lacked experience in operations, we did not lack energy in overcoming this deficiency. The Executive Committee worked without respite throughout 1921. We met during that year exactly 101 times in formal session. Between sessions, individually and together, we were absorbed in the innumerable problems of the emergency and of the future, and were constantly on the go visiting the divisions and their plants in Detroit, Flint, Dayton, and elsewhere.
And so, if I were to take stock of the situation as it stood three or four months after the change of administration, I would say that, although we were short on experience, we were long on logic and energy, and we were getting control of the runaway elements of the business, particularly the inventory. Furthermore, we recognized that General Motors had no explicit policy as to the line of cars to be produced, and that that was the next order of business.
Chapter 4 - Product Policy And Its Origins
After the two great expansions of 1908 to 1910 and 1918 to 1920 — perhaps one should say because of them — General Motors was in need not only of a concept of management but equally of a concept of the automobile business. Every enterprise needs a concept of its industry. There is a logical way of doing business in accordance with the facts and circumstances of an industry, if you can figure it out. If there are different concepts among the enterprises involved, these concepts are likely to express competitive forces in their most vigorous and most decisive form.
Such was the case in the automobile industry in 1921. Mr. Ford's concept of a static model at the lowest price in the car market, expressed in the Model T, dominated the big-volume market then as it had for more than a decade. Other concepts were present, such as the one implied in about twenty makes of cars calculated to have low volume and very high price, and those behind the various cars in intermediate price brackets. General Motors then had no clear-cut concept of the business. It is true, as I have shown, that Mr. Durant had established the
pattern of variety in product expressed in the seven lines: Chevrolet (in two very different models with different engines, the "490" standard, and a higher priced "FB"), Oakland (predecessor of the Pontiac), Olds, Scripps-Booth, Sheridan, Buick, and Cadillac. Of these, only Buick and Cadillac had clear divisional concepts, Buick with its high quality and fairly high volume in the high middle-price bracket, and Cadillac with its permanent endeavor to present the highest quality at a price consistent with a volume that would make a substantial business; and in fact Cadillac and Buick had long been the industry leaders in their price classes.
Nevertheless, there was then in General Motors no established policy for the car lines as a whole. We had no position in the low price area, Chevrolet at that time being competitive with Ford in neither price nor quality. In early 1921, the Chevrolet was priced about $300 above the Model T (when an adjustment is made for comparable equipment), hence, out of sight from the viewpoint of competition. The fact that we were producers of middle and high price cars, so far as I know, was not a deliberate policy. It just happened that no one had figured out how to compete with the Ford, which had then more than half the total market in units. It should be observed, however, that no producer at that time presented a full line of cars, nor did any other producer present a line as broad as General Motors' line.
The spacing of our product line of ten cars in seven lines in early 1921 reveals its irrationality. Our cars and their prices at that time (priced from the roadster to the sedan, F.O.B. Detroit) were as follows:
Chevrolet "490" (four-cylinder) $795 — $1375
Chevrolet "FB" (four-cylinder) $1320 — $2075
Oakland (six-cylinder) $1395 — $2065
Olds (four-cylinder "FB") $1445 — $2145
(six-cylinder) $ 1450 — $2 145
(eight-cylinder) $2100 — $3300
Scripps-Booth (six-cylinder) $1545 — $2295
Sheridan (four-cylinder "FB") $1685
Buick (six-cylinder) $1795 — $3295
Cadillac (eight-cylinder) $3790 — $5690
Superficially this was an imposing car line. In the previous year, 1920, we had sold 331,118 U.S.-produced passenger cars, of which Chevrolet accounted for 129,525 and Buick for 112,208, the remaining 89,385 being distributed among the other cars in the line. In total output of vehicle units and in dollar sales, General Motors in 1920 was second to the Ford Motor Company. In the United States and Canada we sold 393,075 cars and trucks as compared with Ford's production of 1,074,336. The total industry factory sales were about 2,300,000 cars and trucks. Our net sales totaled $567,320,603 as compared with Ford's total of $644,830,550.
From the inside the picture was not quite so good. Not only were we not competitive with Ford in the low-price field—where the big volume and substantial future growth lay—but in the middle, where we were concentrated with duplication, we did not know what we were trying to do except to sell cars which, in a sense, took volume from each other. Some kind of rational policy was called for. That is, it was necessary to know what one was trying to do, apart from the question of what might be imposed upon one by the consumer, the competition, and a combination of technological and economic conditions in the course of evolution. The lack of a rational policy in the car line can be seen especially in the almost identical duplication in price of the Chevrolet "FB," Oakland, and Olds. Each division, in the absence of a corporation policy, operated independently, making its own price and production policies, which landed some cars in identical price positions without relationship to the interest of the enterprise as a whole.
The presence of Sheridan and Scripps-Booth in the line was, to my mind, without any justification. Neither car had its own motor. The Sheridan, assembled in a single plant in Muncie, Indiana, had the four-cylinder "FB" motor. The Scripps-Booth, made in Detroit, had an Oakland six-cylinder motor, which, I might add, was then no attraction. Both had only modest dealer organizations. Singly or together they added nothing but excess baggage to the General Motors car line. Why then were they there? Scripps-Booth stock had come into the corporation with the acquisition of Chevrolet's assets in 1918. But the car had not developed important volume (about 8000 in 1919 and the same in 1920) and had no reasonable place in General Motors' fine. The presence of the Sheridan is a mystery to me. Mr. Durant caused General Motors to acquire it in 1920, doubtless with something special in mind. I am uncertain what. It did not have a strong organization or demand or recognizable purpose in our fine.
As for Oakland and Olds, not only were they competing at nearly identical prices, but both of them were growing rapidly obsolescent in design. Take the Oakland. At a meeting in my office on February 10, 1921, Mr. Pratt described the problem of this car as follows: "Oakland is spending [its] efforts in trying to improve [its] product. Some days they produce ten cars and some days they produce 50 cars. The situation is this—they turn out a lot of cars that are not what they should be and then they have to fix them up . . . The power plant has been the great trouble . . ." At the same meeting I said: "There is a lot that enters into this problem. At the present time we are getting 35 to 40 H.P. out of the Oakland motor and the crankshaft is too light for this rate of speed [power], and we have had a lot of poor workmanship together with other things, and the Oakland Motor Car Company over a year ago decided that they would put in a new motor. A new motor plant was authorized a year ago but we had to hold it up when we curtailed our development program ... It is really a question of management to get this motor in the Oakland so it will pass inspection and be right . . ."
Oakland had sold its high of 52,124 cars in the boom year, 1919; it sold 34,839 in 1920, and, as it turned out, was to sell only 11,852 cars in 1921.
So much for Oakland.
Olds was only a little better off. It had sold 41,127 vehicles in 1919, 33,949 in 1920, and would sell 18,978 in 1921. It would take a new design just to save it.
Cadillac made 19,790 unit sales in 1920. In 1921 it would sell 11,130, and with the big price deflation that had taken place in the United States it would have to find a new optimum of cost, price, and volume.
The hard fact was that all the cars in the General Motors line, except Buick and Cadillac, were losing money in 1921. The Chevrolet Division that year lost about half of its 1920 volume. Its dollar losses at one point in 1921 reached approximately $1 million a month, and for the year as a whole it lost nearly $5 million. So strongly did I feel about the situation that, when someone proposed making changes in Buick's management, where Harry Bassett was successfully carrying on Walter Chrysler's old policy, I wrote to Mr. du Pont: "It is far better that the rest of General Motors be scrapped than any chances taken with Buick's earning power." If that seems like an overextended argument, consider Buick's position. Its sales dropped only moderately from 115,401 in the 1919 boom to 80,122 in the 1921 slump, and what's more, it continued to produce an income. It was Buick that made any kind of General Motors car line worth talking about.
This situation reflected in good part the poor quality and unreliability of the other cars in the line, as compared with the high quality and reliability of Buick and Cadillac; the effect of these factors was intensified by the stress of the general economic slump. Given the fact of the slump and the unavoidability of a general decline in sales, the relative decline of one division as compared with another was a question of management.
The slump had the effect of showing up all kinds of weaknesses, as slumps usually do. General Motors in 1920 had enjoyed 17 per cent of the U.S. car and truck market; in 1921 we were on our way down to 12 per cent. Ford, on the other hand, was in the course of rising from 45 per cent of the market in units in 1920 to 60 per cent in 1921. In other words, Mr. Ford, whom no one had dared seriously to challenge in the low-price field since 1908, was tightening his grip while we were losing in unit volume as well as in the profitability of most of our divisions. All in all, with no position in the big-volume low-price field and no concept to guide our actions, we were in a b
ad situation. It was clear that we needed an idea for penetrating the low-price field, and for the deployment of the cars through the line as a whole; and we needed a research-and development policy, a sales policy, and the like, to support whatever we did.
In view of these circumstances, it is hardly surprising that on April 6, 1921, the Executive Committee set up a special committee of the Advisory Staff, made up of experienced automobile men in management, to look into our product policy. This task was to be one of the most significant in the evolution of the corporation. The members of the committee were C. S. Mott, then group executive for car, truck, and parts operations; Norval A. Hawkins, who had been chief of Ford sales before joining General Motors; C. F. Kettering of General Motors research; H. H. Bassett, general manager of Buick; K. W. Zimmerschied, newly appointed general manager of Chevrolet, and myself from the Executive Committee. Since I was in charge of the Advisory Staff when the special committee was formed and the senior member of the committee, its work came under my jurisdiction. About a month later we had completed our study, and on June 9 I presented our recommendations to the Executive Committee, where they were approved and became the adopted policy of the corporation. The recommendations outlined the basic product policy of the corporation, a market strategy, and some first principles; all together they expressed the concept of the business.
The general historical circumstances described above had much to do with the nature of the recommendations. And there were other circumstances in the internal situation in General Motors which influenced what we had to say. In the first place the Executive Committee had instructed the special committee that the corporation intended to enter the low-price field—that is, that it intended to make a competitive challenge to the dominance of Ford. The Executive Committee asked the special committee for advice on this question, and suggested that cars be designed and built in two low price ranges, the lower of which would compete with Ford. They also asked for a discussion at some later time of other price areas. They excluded, however, any changes in the successfully established positions of Buick and Cadillac.
My Years With General Motors Page 8
