We should, I believe, confine our discussions in these meetings to problems of common interest [a] fleeting all divisions. Realizing that all of you men are extremely busy we will try to keep away from details — dealing only with the basic problems. We will do everything in our power to make the sessions business-like and to the point. No time will be taken to prepare papers, etc. unless in some instances you may wish it that way. Mr. [B. G.] Koether [director, Sales Section] will serve as Secretary of the Committee. He has a Staff which can be expanded if necessary, and whose services are entirely at your command.
We have not developed any definite programs for these meetings because we want to leave such matters to you, realizing that you are in a better position to know just what problems require the most urgent attention and while we may suggest a number of things from time to time it is entirely up to you to act upon such suggestions as you may see fit . . .
The chairmanship of the General Sales Committee was later given to Donaldson Brown, vice president of finance, because of the bearing of statistical and financial controls on production and sales problems. Co-ordination in sales thus extended to the Financial Staff.
After a study of the inter-divisional type of committee by Mr. Pratt in late 1924 confirmed in the minds of everyone that this was the best form of co-ordination we had found up to that time, it was made more or less official and was extended to works managers and the power and maintenance staff. Something of the same sort of co-ordination was extended to the very top level of management —but with a difference.
The reader will recall that under Mr. Durant the Executive Committee was composed largely of division managers who campaigned there for the interests of their respective divisions. When we formed the new temporary Executive Committee of four, we placed the former members, mainly the division managers, in an advisory operations committee. For some time, while the emergency was being liquidated, this advisory committee was not regularly active. After I became president and the Executive Committee was enlarged again, it included at different times one or two division managers, depending on circumstance, or motivated by the thought that the largest car division should have representation there. But these were exceptions, not the rule, for I believed in principle that the top operating committee should be a policy group detached from the interests of specific divisions. In other words, it should contain only general executives. Holding this view, after I became president I felt that something should be done to bring the general managers into contact in a regular way with the members of the top operating policy group. I therefore reactivated the Operations Committee and had placed on it all the general operating officers on the Executive Committee and the general managers of the principal divisions, thus making it the major point of regular contact between the two types of executives. The Operations Committee was not a policy-making body but a forum for the discussion of policy or of need for policy. The Operations Committee would receive a full set of data on the performance of the corporation and would review that performance. The word "forum" may suggest something idle, but I assure you that in this case it does not mean that. In a large enterprise some means is necessary to bring about a common understanding. It is perhaps sufficient to note that, with all of the members of the top operating policy group present, an agreement on a policy, suggested say by a division manager, would be tantamount to acceptance on the operating side of the corporation.
In sum then, the whole picture of co-ordination in 1925 and for a number of years thereafter was as follows: The inter-divisional relations committees gave a measure of co-ordination to the functions of purchasing, engineering, sales, and the like. The Operations Committee, including the general managers, appraised the performance of the divisions. The Executive Committee, with contacts in all directions, made policy. It sat at the head of operations, responsible to the board of directors—indeed it was a committee of the board—but beholden to the Finance Committee for its larger appropriations. On the operating side the Executive Committee was supreme. Its chairman was the president and chief executive officer of the corporation; and he had all the authority he needed to carry out established policy. This was the new General Motors scheme of management from which developments down to this day, through much evolution, have been derived.
Chapter 8 - The Development Of Financial Controls
The development of co-ordination by committee in the early 1920s was accompanied by co-ordination of another order, namely that of financial control. General Motors' progress, I believe, was largely the result of advances the corporation made in this area of management, along with those in organization and product policy. Our modem financial policies, like those in organization, came out of the ruins of 1920.
For the leading members of the new administration that took over the corporation, the necessity of providing new forms of financial control was doctrine. The question was what they should be and how to put them into effect. The specific forms of financial control in General Motors were introduced in good part by Donaldson Brown, who came to General Motors from the du Pont Company at the beginning of 1921, and Albert Bradley, his young associate who came to General Motors in 1919 and who was to succeed Mr. Brown as the top financial officer and later to succeed me as chairman of the board. Their contributions to financial thought have long been recognized. They wrote papers on the subject which are classics of the 1920s, and at the same time put their concepts into practice in General Motors. Frederic G. Donner, present chairman and chief executive officer; George Russell, executive vice president, and other gifted members of the Financial Staff rose out of this great school of finance to make their own contributions in their long service to the corporation. Although I wrote on the subject of finance in the form of reports, particularly in connection with inter-divisional business and appropriations, my experience was mainly from the operating side. My responsibility involved the application of financial method, for finance could not exist in a vacuum but had to be integrated with operations.
I think I have made clear that Mr. Durant had no systematic financial methodology. It was not his way of doing business. And yet modern financial concepts were brought into General Motors during his administration. Mr. Durant was instrumental in arranging for the du Pont executives to come on the Finance Committee and take the responsibility for that aspect of the corporation's affairs. I believe the outstanding benefit General Motors derived from the du Pont association—apart from their general position as a responsible shareholder serving on the board of directors—was in the financial area. A number of du Pont men experienced in accounting and finance came to General Motors in the early years and assumed key positions.
Mr. Brown was one of these men. A word on his background, as he has told it to me: He was in the sales department of the du Pont Company for a number of years during the early part of this century. In 1912 he was taken into the office of one of the du Pont general managers as an assistant without portfolio. That was when Coleman du Pont was president of the du Pont Company. The general manager of this department was handicapped seriously by bad health and had to be out of the office for a period. At that time the du Pont Executive Committee was seeking a realistic report on the efficiency of the operating departments of the company, then engaged almost wholly in manufacturing explosives: blasting powder, dynamite, and the like. Mr. Brown took upon himself the job of developing a method to reveal the desired facts about the several activities under the general manager. The method he chose emphasized the importance of capital turnover as well as profit margin in calculating return on investment. Mr. Brown passed his report on to the chief executives and it made such an impression on Coleman du Pont that he recommended that Mr. Brown be transferred to the Finance Department. Pierre S. du Pont was then treasurer, and Mr. Raskob was assistant treasurer. Mr. Raskob made Mr. Brown the junior assistant treasurer—Very junior" Mr. Brown says. I guess he was, with Mr. Raskob around. But eventually Mr. Raskob succeeded Mr. du Pont as treasurer, and Mr. Brown su
cceeded Mr. Raskob in that post when Mr. Raskob went into General Motors. Mr. Brown brought economists and statisticians into the du Pont Company, a practice that was unusual in those days. Thereafter when the du Pont Executive Committee met with the du Pont general managers, Mr. Brown displayed charts on the efficiency of divisional performance, a technique of presentation which he initiated.
At Mr. Raskob's request Mr. Brown came to General Motors on January 1, 1921, as vice president in charge of finance. He and I shared similar views on the value of detailed, disciplined controls in the operation of a business. From the time of his arrival in the corporation we recognized this affinity and began a long and congenial relationship.
The du Pont group, after coming into the corporation in 1917, had made an effort to apply the principle of return on investment in appropriating funds to the operating side of the corporation. Yet Mr. Raskob, though he had the right idea in general, was not prepared with the instrumentation for General Motors. I have described in an earlier chapter how, during the expansion of 1919, difficulties arose owing to the loose manner in which appropriations were made, and how the inventory runaway and shortage of cash brought crisis to the corporation in the slump of 1920. These three emergency problems—overruns on appropriations, inventory runaway, and the resulting cash shortage—exposed the lack of control and co-ordination in the corporation. It was in the effort to meet these specific emergency problems that new methods of financial co-ordination and control were developed in General Motors.
Financial method is so refined today that it may seem routine; yet this method—the financial model, as some call it—by organizing and presenting the significant facts about what is going on in and around a business, is one of the chief bases for strategic business decisions. At all times, and particularly in times of crisis, or of contraction or expansion from whatever cause, it is of the essence in the running of a business. The situation in 1920 proved this negatively and we were to prove it positively in the subsequent critical years.
I have related how, in 1919 and 1920, in the absence of a system for control of appropriations, each division manager got his maximum request satisfied, without real effort on the part of the corporation to evaluate the request or to reconcile the total amount of all requests with the available funds. This, together with overruns on appropriations and the inventory rise, represented a drain on available funds which had to be met in some way. To get the money we sold common stock, debenture stock, and preferred stock, though not so easily or in such amounts as we expected; and before the year 1920 was out we had to borrow about $83 million from banks. From then through 1922 we charged against income of the corporation about $90 million for extraordinary write-offs, inventory adjustments, and liquidation losses, an amount equal to about one sixth of the total assets of the corporation. Financial control at this juncture was not merely desirable, it was a necessity. To survive we had to pull back from the brink and find a general solution.
The story of how we did it falls into two parts. The first part concerns how the corporation curtailed the excessive freedom of the divisions, which had gone so far as to jeopardize the survival of the corporation, and how we established control over them. The immediate remedies inevitably were centralizing in character, for the corporation could not afford to let its divisions continue to make the kind of mistakes they were making: the weak divisions threatened the existence of the strong ones, and the strong ones themselves were operated more for their own than the corporation's interest. These centralizing remedies—largely operating controls —created a temporary distortion in our general policy which later had to be corrected in order to return to a workable decentralization. The second part of the story involves the development of the financial instruments which made it possible to establish decentralization with co-ordinated control.
Effecting Corporate Control: Appropriations for Capital Spending
Just before the economic collapse of 1920, that is, in June of that year, the Committee on Appropriation Request Rules, which had been formed at the end of 1919 and of which I was chairman, gave its report to the Executive Committee. This report, prepared by Mr. Pratt, Mr. Prentis, and myself, marks a historic turning point in the development of appropriations procedures in General Motors.
The core of our concept lay in the determination of the propriety of proposed projects. Four principles were to be satisfied, which we stated as follows:
a. Is the Project a logical or necessary one considered as a commercial venture?
b. Has the Project been properly developed technically?
c. Is the Project proper, considering the interest of the Corporation as a whole?
d. What is the relative value of the Project to the Corporation as compared with other Projects under consideration, from the standpoint not only of the return on the necessary capital to be invested, but of the need of the particular Project in supporting the operations of the Corporation as a whole?
Bearing down on the main weakness of the corporation in this area at that time, we said in the report:
... a very careful consideration of the subject on the part of your Committee leads it to the inevitable conclusion that at least so far as the larger Projects are concerned, an independent impartial review and checking of all phases of [the] proposed project outside of the Division or Subsidiary itself, is essential and will be found more and more so as time develops and the operations of the Corporation become more interwoven and complex.
This procedure required a review by an appropriations committee before submission of a request to the Executive Committee or Finance Committee for approval, and a policy review by the latter committees. We defined the scope of their review as follows:
It appears to your Committee that the members of such committees [Executive and Finance] should be interested in the Projects as a matter of general policy and that their passing of such Projects should be a matter of financial returns or necessity of the Projects to the general development of the Corporation rather than from the standpoint of the particular type of lathe or milling machine, or how many such lathes or milling machines are essential to the proper development of the proposed Project.
Following this line of reasoning, we allowed certain small amounts of expenditure to be authorized by the general manager of a division on his own. For larger amounts, we proposed a detailed procedure on the development and follow-up of supporting data, and, in this connection, we proposed to bring together the two main branches of the corporation: "Your Committee recognizes the necessity of determining proper co-ordination between the Financial and Operating Staffs as regards expenditures . . ." And to be specific we recommended that an appropriations manual be developed for the corporation setting forth in detail the kind of information the divisions and subsidiaries should present to demonstrate the desirability of a proposed expenditure both from an engineering and an economic standpoint.
The Executive Committee approved our recommendations in September 1920 and requested that the manual be prepared. This manual, approved by the Executive and Finance committees in April 1922, established the first well-defined capital-appropriation procedure in General Motors. It called for an appropriations committee, functioning under both the Finance and Executive committees, to have general charge of all appropriation matters and coordinate programs involving more than one division. The divisions were to make monthly reports of construction in progress to the appropriations committee, which in turn would present a combined report each month to the Finance Committee. Each appropriation request was to receive consideration and analysis from a corporation as well as a divisional standpoint before any commitment was made. Proper records were to be kept of expenditures and approvals for expenditures, and uniform treatment was to be given to appropriation requests throughout the corporation. In short, we were for the first time to get accurate and orderly information. After that it would be a matter of business judgment whether to grant a request. Changes in this procedure have been made from
time to time, and an appropriations committee as such was discontinued long ago. However, in its essentials, this is still the way capital appropriations are approved in General Motors.
Cash Control
We were short of cash in 1920 because we spent a lot of money on the future and did not take in enough in the present. Hence the bank borrowing, which reached a maximum of about $83 million at the end of October of that year. For some time afterward cash was a question of conservation.
The way cash was handled at that time is almost unbelievable. Each division controlled its own cash, depositing all receipts in its own accounts and paying all bills from those same accounts. Since only the divisions sold products, none of these cash receipts flowed directly to the corporation itself. We had no effective procedure for getting cash from the points where we happened to have some to the points where we happened to need some. When the corporation, as an operating company, had to pay dividends and taxes, and such items as rent, salaries, and other expenses of the general staff, the usual procedure was for the treasurer to request cash from the divisions. That was not so simple as it sounds, however, for the divisions, operating independently, tried to keep their cash balances high enough to satisfy their own peak requirements. Therefore, when they had more cash than they needed at the moment, they were not eager to turn it over to the corporation.
I remember that Buick, for example, at that time was very loath to give up its cash. This profitable division was, of course, the most prolific source of cash for the corporation, and long experience had made Buick's financial staff highly adept at delaying its report of the cash they had on hand. Buick made a practice of maintaining large cash balances in its factory sales branches. The amounts of these balances were not ascertainable at headquarters until Buick had submitted its monthly financial statement for the division as a whole—and this was usually a month or two after the fact. When the corporation needed cash, the treasurer, Meyer Prentis, would try to guess how much Buick actually had and how much of it he could probably get from them. Then he would go to Flint, discuss whatever other questions might be outstanding between Buick and headquarters, and at last casually bring up the subject of cash. Buick's financial people would invariably express surprise at the size of Mr. Prentis' request and occasionally would try to resist the transfer of such a large amount. Naturally, this cat-and-mouse game did not result in the most efficient utilization of funds, especially when some divisions had more operating cash than they needed, at the same time that other divisions were short of operating cash.
My Years With General Motors Page 15