Division of Research Graduate School of Business Administration The University of Michigan January 1975 COMPETITION AND ENTRY INTO BANKING MARKETS Working Paper No. 99 by Thomas G. Gies Graduate School of Business Administration The University of Michigan ~ The University of Michigan 1975 FOR DISCUSSION PURPOSES ONLY None of this material is to be quoted, por-'reprbduced..withdout: the express permission of the Division of Research Sponsored by the Brookings Institution as part of the Study of Potential Competition in the U.S. Economy.

I. INTRODUCTION There has recently been a strong renewal of interest in competitive behavior of commercial banks, as well as other financial institutions, 1/ which follows a lapse of almost three decades.- The Keynesian revolution focused attention strongly on macrodimensions of banking, and both theoretical and empirical effort concentrated on the relationship of monetary institutions to national economic aggregates. Beginning in the early 60s, however, researchers again directed attention to microdimensions of bank behavior. These studies have been directed in the main toward structural characteristics of the market as they bear upon competitive behavior -- dimensions of the market, number of firms, relative size, economies of scale, and concentration of control. Ease of entry and exit has received somewhat less attention. Role of management Very little effort, however, has been directed toward investigating the effect of nonstructural variables such as management upon the behavior of banks -- not because economists deny the relevance of management attitudes and philosophy in economic performance, but perhaps because it is felt that management philosophy can only be regulated indirectly through altering market structure and organization. I would like to introduce a different argument here. I present a case study of a banking group, the COMAC Group, which introduced a l/ Lester Chandler,'s "Monopolistic Elements in Banking," Journal of Political Economy, Feb, 1938, was virtually the last analysis of the bank as a firm until the decade of the sixties. -1 -

-2 - highly aggressive type of management in certain Michigan banking markets during the 1960s. First, the actions taken by this ownermanagement group are spelled out and the results for the banks under their control identified. Second, we attempt to sort out the impact which these actions produced in three banking markets -- where the group operated -- Lansing, Muskegon, and Kalamazoo. Third, we view the reactions of the banks affected, and, finally, we review the - responses of government regulatory authorities to the disequilibrating group. The lessons regarding competition and regulation have some force for the normal run of banking markets. Toward a banking market definition It has long been recognized that banking has oligopoliStic tendencies, but only recently has work been directed towards defining a banking market. A group of studies [1, 3, 7, 8, 9, 12, 13, 14] show that banking is local in nature, with only the largest firms having banking mobility. Most customer affiliation with banks tends to be based on convenience and quality rather than price. Large businesses on the other hand, are more mobile in their bank selection, and the primary determinants of their choice are the financial condition of the bank, the location, the number and quality of services offered, and credit availability. It is interesting to note that even in this market segment, considered to be the most price-elastic, price factors were not of primary importance. This is not to say that large businesses are indifferent to price differences, but that they also consider factors other than price in choosing their banks.

-3 -The product markets in which price factors are important are those in which bank customers have nonbank as well as bank alternatives. For example, savings and loan associations compete for time deposits, the money market competes for commercial deposit money, and nonfinancial corporations compete for government securities [5]. On the other hand, such products as demand deposits and business loans come under little pressure from nonbank financial intermediaries. Competition for these products sometimes takes a nonprice format, with emphasis being placed on new services, product differentiation, and mass promotion. Numerous studies have been conducted to determine the effect of bank structure on performance, assessing the effects of new unit banks, de novo branches, mergers, and holding company acquisitions. Although there are numerous differences between the studies, the following four works will illustrate the principal directions which these analyses have taken. Furthermore, these four are of special interest, for in each instance they touch on problems related to the banking markets under review here. Paul Jessup [6] has analyzed the performance of a group of banks where significant shift in ownership has occurred to determine whether operating results were measureably altered by the restructuring of ownership. Robert Lawrence [10, 11] conducted two studies on bank holding companies, which were aimed at identifying differences in performance between banks affiliated with bank holding companies and comparable independent banks.

-4 - The last of the four precedent studies was conducted by Talley [15], and used Lawrence's methodology. Talley's study shows that the major effect of holding companies on operating performance is to alter the portfolio composition of acquired banks,!I;; PRICE AND OUTPUT DETERMINATION The banking system It is essential to recognize that output and price determination to commercial banking markets bear less relationship to the number of banking firms in the market than is the case with industrial firms. The difference springs from the fact that increasing or decreasing the number of banks has no direct or immediate effect on the capacity of the banking industry as a whole to create credit; the monetary authority has, at least in the short run, effective control over the total quantity of money and bank credit. Because credit unquestionably is the most important product banks produce in terms of revenue dollars generated, the insensitivity of output to number of firms means that the price of bank output also is not directly 2/ and immediately related to the number of banking firms in the industry. This single fact is of paramount importance in judging questions of organization and structure, of entry and exit, and of stability in ownership and management in banking markets. The underlying reason for the irrelevance of number of firms to output 2/ Whether number of firms in the banking industry per se affects the attractiveness of bank credit vis-V-vis nonbank credit will not be debated here, though the likelihood seems remote to this author.

-5 -of bank credit in turn rests on two facts: first, government policy, as administered through the monetary authority, establishes maximum output of bank credit for the entire economy, though not for individual market areas nor for individual banks; and, second, the output quota established by the monetary authority is virtually always below the profit-maximizing equilibrium for the banking industry and therefore actual output normally approximates the maximum permitted output. The phenomenon of full utilization of bank reserve positions is the normal condition of the banks and follows from the profit-maximization incentive, driving the banks to strain against the upper limit on output presented by monetary controls. The rationale for monetary control is the need to regulate the overall level of employment, output, income, price level, and growth in the economy. The firms composing the banking sector merely represent the vehicle for execution of monetary policy, and the number of firms is irrelevant to accomplishment of monetary goals. The addition or subtraction of one banking firm will not alter central bank decisions regarding the optimum quantity of bank credit at a given time or the optimum growth rate of bank credit. If it is accepted that the total supply of bank credit is outside the discretion of the commercial banking industry and no decision by the industry can alter maximum output, attention can be directed to the demand for bank credit. In an aggregate sense, demand is a function of demographic and economic factors, such as number of households, businesses, and governments and the scale and timing of their financial deficits. Clearly, these considerations are also outside the discretion of the banking industry.

-6 -If neither the level of supply and demand nor changes in supply and demand are controllable by banks, it must follow that banks have little potential for controlling the yields on securities traded in open markets -- e.g., the yield on Treasury issues. Indeed, the price of securities is a decision made at the level of the monetary authority and is one of the principal "proximate"l objectives of monetary policy, the ultimate objective of course being control of price and output behavior in goods and service markets. The whole question of competitive relationships in such a market -- the significance of number of competitors, their stability and duration of life, and their management policies -- takes on a different light with these conditions in view. The first rule of monopolistic behavior, that output should be reduced from the competitive market equilibrium, may require modification when applied to a market in which maximum permissable output is already decreased to substantially below the profit-maximizing level even for a monopoly firm. The modern central bank has exclusive control over creation of legal reserves for the banks and by this means can adjust 3/ the quantity of bank credit up or down. - The presence or absence of vigorous competition in banking markets would not necessarily-lead to any difference in output, where output of the industry is severely inhibited by monetary control. Specifically, if the quantity of bank reserves limits creation of bank credit to an amount below the profit-maximizing level, 3/ The central bank's control over supply and price of bank reserves is in fact one of the few modern day cases of virtually perfect monopoly.

-7 -then, regardless of market organization, there will be no further overall restriction of output at the commercial bank level, even where there is only one commercial bank in the market. Regional and local banking markets Determination of output and pricing policy within regional and local banking markets arises from considerations at the system level. The monetary authority is responsible for establishing the total quantity and rate of growth of bank credit, but it does not specify the proportion of that total which will become available to a regional or local market. However, as with the system of banks, that determination is not a function oft the number of banks in the regional or local market, in the sense that adding a bank does not.add to the supply of bank credit in a given geographic or functional market. This is not to say that structural factors, such as the number of bank offices in the market and the relative yields on substitute assets, do not affect demand for deposits. Additional factors -- income, wealth, amount of business located in the market, and proximity to other financial centers -- also influence the volume of deposit funds. But before we discuss these forces, it may be helpful to clarify our concept of a banking market. There are two dimensions to a banking market -- geography and product. The geographic market is generally local in nature, although for some customers (e.g., very large businesses) the market may be national or international. A considerable volume of studies testify to this fact.

I I -8 -There is much less agreement on what the appropriate product market is. Briefly, the two alternative viewpoints are the single-product concept and the multiproduct concept. The single-product, or customer relationship, concept emphasizes the tying nature of bank products. It sees the commercial bank as providing a bundle of services and consequently as not in direct competition with other financial institutions. Proponents of this view feel that the central service is providing a means of payment with the other services tied to this essential service. The multiproduct or institutional-investor concept sees the commercial bank as providing different products, each distinct from the other -- e.g., mortgage credit, instalment credit, etc. In each product area the bank may encounter competition from other financial institutions. In this respect commercial banks are seen as not unique. Actually, the controversy appears to concern degree more than kind, Alhadeff and Kreps, both proponents of the multiproduct approach, agree that business customers of the bank are often required to keep compensating balances, which is a method of tying products. Even though some of the other bank customers are not required to do all their business with one institution this might still be their normal behavior. Certainly banks prefer to lend to customers they already know. On the other hand, some credit products, such as residential mortgages, are quire competitive between banks and nonbanks. Thus, elements of both theories seem to have validity. Our purpose, however, is not to evaluate the literature on bank markets but is merely to point out the relevant dimensions in identifying

-9 -the markets in the three areas under study. Geographically, Lansing, Kalamazoo, and Muskegon are distinguishable, isolated markets. The Kalamazoo Standard Metropolitan Statistical Area makes up one county (Kalamazoo County). Approximately 48 per cent of the county population is within the city limits. The Muskegon SMSA also makes up only one county (Muskegon County). Approximately 31 per cent of the county population is within the city limits. Although the Lansing SMSA is comprised of three counties (Clinton, Ingham and Eaton), about 36 per cent of the SMSA population is within the Lansing city limits, and 71 per cent is within Ingham county. Because of the nature of this study, the product delineation appears to be of subordinate importance. Given the funds in the market and thus the potential size of the local banking sector, total credit will be a function of the amount of reserves needed to cover deposit liabilities and the amount of excess reserves the bank desires to keepo The composition of this credit will be a function of such variables as capital requirements, relative yields, perceived risks, liquidity, deposit variability, management preferences, legal restrictions, market quality of assets available,and tax considerations. As Hodgman [16] has pointed out, the customer relationship is extremely important in its effects on loan composition since the bank is likely to favor lending to its major depositors, Looking at the variables that affect output, one can see that individual bank management can control to a large extent both the total quantity of credit extended and the composition of this credit.

I I -10 -Granted that individual banks have substantial control over output decisions, how much control do they have over pricing? Without becoming too specific, one can make certain generalizations about the pricing policies for different credit outputs. Except for small, locally placed municipal issues, the individual bank has no control over pricing for marketable securities. These rates are determined on the open market and are exogeneous to the individual bank. On the other hand, commercial banks in general have power to set 4/ prices on certain assets, e.g., customer loans.4/ For large business loans, the market is demonstrably national in scope, which means that, besides the large number of banks to choose from, national firms also have the money market as a viable alternative to short-term borrowings from commercial banks. For small businesses, however, credit sources are much more restricted geographically, and the local commercial bank has more autonomy in setting rates. The market for mortgage credit is divided between commercial and residential real estate. Commercial real-estate loans often draw lenders nationally -- as for example, a large regional shopping center -- while single family mortgage credit is largely restricted to local markets. Residential mortgages are relatively homogeneous, prices are typically close for all suppliers of this type of credit, and, consequently, banks do not have much freedom to set rates. The increased willingness of certain 4/ - It is not clear whether they exercise this power. Studies on the relationship between concentration and loan rates are inconclusive.

-11 -nonbank mortgage lenders, such as life insurance companies, to lend in a national market through local mortgage brokers has further unified mortgage rates. Consumer instalment credit demand is regarded as inelastic to price in the aggregate, but this may be less true with respect to the individual bank. -Some borrowers will go to one bank and accept whatever terms are offered,, thers shop around. They may be willing to pay 20 percent- per year for credit, but they are not insensitive to price differences, Recognizing this, banks appear to discriminate among classes of borrowers on the basis of competitive conditions rather than on cost alone. To summarize, it can be said that, while total deposits in the market are determined mainly by factors outside the bank's control, an effective liability-management program may enable the bank to increase its supply of funds and thus its credit output. Further, by increasing its willingness to take risk, the bank can further minimize its liquid reserves and increase credit output. In addition to increasing total credit output, the bank can change the composition of this output. By shifting from low-yield credits to higher return uses, the bank may be able toaffect its revenues favorably. The important point is that both the increase of total output and the change in its composition are within the scope of management powers of the individual commercial bank, Static vs. dynamic market organization for banks The foregoing considerations form the basis for arguing that the structure of the market may have less influence on price and output in

-12 -the case of bank credit than -t does in the case of industrial products. Rather than sheer number of competitors, the important dimension of bank markets may be (1) ease of entry and exit and (2) managerial philosophy. The magic quality needed to assure vigor in exploring new products and new methods may be more intimately related to change in the number of competitors than to the particular number, whether that number is large or small. In other words, fewness of banks may be less disadvantageous than is generally thought, providing the membership of the small group is capable of being changed readily or has a genuine potential for change. Beyond this, the management philosophy of the dominant owner-manager group may be a critical behavior dimension that is not revealed in standard market structure models. It is common observation that local retail merchants become accustomed to each other's presence -- their respective policies concerning pricing, quality and service, etc. -- and feel little threat from the rival's operation. But let word get around among pharmacists, for example, that a new drug store is going to open, and deep concern is aroused: What will the new store offer? Will it cut prescription prices? Be open on Sunday? Where will it locate? And so fortho The addition of a new firm is disturbing, whether the existing number of firms is one, two, or ten. The principal direction of effort to establish control in a market may not be to wipe out existing competitors, but rather to prevent new competitors from entering. Existing competitors are known quantities and in many cases have come to accept the very human approach to life of "live and let live," But no one knows exactly what a new competitor will do -- and therefore conventional oligopoly wisdom would be to control entry.

-13 - The banking business controls entry through legally established and sustained barriers. For reasons long established and well known, entry to the banking business requires obtaining.a license from the state. And in most jurisdictions, existing banks are provided extensive opportunity for objecting to new applications for entry. The existing firm or firms can be counted on to allege that they are "adequately serving" the market and that there is '"no necessity" for additional facilities. In many jurisdictions, "necessity" is interpreted as certainty that no existing bank will be injured. And in one recent instance, the question involved c6nsideration of whether the new'entrant might injure future growth potential of existing firms (banks). III. — EMPIRICAL STUDIES OF BANK MARKETS A review of some of the empirical work on bank market entry provides insight into why entry occurs and what impact new competitors have. Two studies have examined the reasons for entry into a particular banking market. Chandross [24] looked at the performance in certain banking markets before and after new banks entered these markets. The performance ratios of these entry markets were then compared to an average ratio composed of all nonmember banks in that particular state and revealed that earnings were higher in the entry markets than in the nonentry markets for the period preceding entry. Further, he found that loan-to-assets ratios were lower in the entry markets, From this it might be hypothesized that the reasons for entry were two-fold: to take advantage of the already greater-than-average profits in the market and to increase these profits further by increasing the below-average loan-to-assets ratio.

I I -14 -Another study by Fraser and Rose [4] asked whether markets that experienced entry of a new bank had performance characteristics different from markets that had no new entry, the intent being to find out why new banks wanted to enter these markets. Entry and nonentry markets were compared to determine if the two groups had different performance characteristics. The results of this study conflict in some respects with the results of the Chandross study. Although the loan-to-assets ratio was found to be lower in the entry markets, there was no difference in,profitability between the entry and nonentry markets. In addition, Fraser and Rose found that the entry markets had a higher ratio of U.S. government securities to total assets and a lower ratio of time deposits to total deposits. Neither of these studies looked at the situation of entry through the holding company or chain banking mechanism, but we feel that some general'i'zations maybe made concerning motivation for bank entry. From both of the studies it can be seen that banks entered the markets either because of the high level of-profits or the potential for profits that could be realized through new management policies (e.g., portfolio shifts). Fraser and Rose [4] conclude that "... entry of new banks into Southwestern communities did not appear to result from excessive profits carried by the previously operating institutions. Rather it appeared to stem from the anticipation of the probable streams of profits arising from an expansion of credit in the local area." There are a number of studies that have attempted to determine the impact on bank performance of a change in market structure. Of interest here are three studies noted earlier, which looked at the impact of holding

company acquisition and one that looked at the impact of changes in bank ownership and managemento Jessup [6] looked at the situation in which banks had a new majority ownership and new senior management. He compared banks which experienced these changes to control groups of other banks both before and after change. Hypothesizing that important differences in performance are associated with new individual ownership of banks, Jessup was able to confirm this by statistical tests. In general, the study banks experienced higher loansto-assets ratios, a higher percentage of consumer loans, and no change in the prices of bank services, Thus it appears that the supply of credit to the local economy was increased without additional cost to customers. Lawrence conducted two studies concerning the effects of bank holding company acquisitions on performance [10, 11]. In the first he used the paired method of analysis. Banks that had been acquired by holding companies were compared to similar nonaffiliated commercial banks both before and after the acquisition. The results indicated that differences in performance were minimal. Holding company affiliates, however, did have a higher ratio of municipals than nonaffiliated banks. Furthermore, a smaller percentage of their assets were committed to U.S. government securities and to balances due from domestic banks. The breakdown for loans shows that instalment loans increased more than any other category. From these and other statistical results, Lawrence concludes that bank holding company affiliates tend to be more aggressive lenders than independent banks and thus make more credit available to the community. It is interesting to note that there was no significant difference

-16 - between affiliates and nonaffiliates on interest paid to savings depositors or interest charged on loans, although affiliates charged higher service charges on demand deposits. Finally, profitability differences were found to be insignificant. Lawrence's second study [11] examined the operating policies of the multibank holding companies towards their banks. By means of an extensive questionnaire distributed to all registered bank holding companies, Lawrence found that there were, in fact, important differences in the companies'operating policies. The degree of centralization of decision-making differed among holding companies. These differences could not be attributed to any economic or structural factors. Instead, Lawrence suggests "... that a holding company's policies can only be determined by investigating the management philosophy of the senior officers of the particular company." [ 11, p. 32 1 But holding companies exert more influence in some areas of the individual bank's function than in others. Securities investments, federal funds transactions, and bank correspondent relationships were among the most influenced functions. Pricing policies, decisions on the composition of the loan portfolio, and decisions on individual loans were among the functions least influenced by holding companies. Finally, Talley [15] updated the first Lawrence study on holding companies, using the paired method of analysis as in that previous study. Talley found that affiliates had significant differences from independent banks. After acquisition, for example, affiliates reduced their cash-tototal-assets ratio and their U.S.-government-securities-to-total-assets ratioo More funds were shifted into state and local government securities

-17 -and into loans, particularly instalment loans. Consequently, it was hypothesized, affiliates served the credit needs of the local community better than independents. Little change was found in the amount of capital provided and the profitability of the affiliates. Although Talley found service charges on demand deposits to be lower, the result was not significant at the.05 level. This conflicts with Lawrence's finding that service charges on demand deposits rose. As in the Lawrence study, prices for savings accounts and loans did not change significantly. Thus, except for the service charges, the results of the Talley and the Lawrence studies are similar. To sum up the results of these studies we can say that in general we would expect acquired banks to reduce their liquidity as measured by cash and U.S. government securities and increase their investment in state and local securities and loans. We would expect to find little change in capital, prices, and profitability. IV. CRITICISM OF CONTEMPORARY BANK MANAGEMENT The American banking system has been changed significantly in the past 25 years by a number of important factors, such as the growth of nonbank competitors, as well as by changes in American industry. On the other hand, a number of observers, among them Dr. George Mitchell, Member of the Board of Governors of the Federal Reserve System, have noted that the banking industry itself has not been an innovative force for change. Dr. Mitchell has written that some firms and industries have produced a product or service, generated public awareness and acceptance for it,

-18 -and, using generative, adaptive, and creative forces from within, have accomplished a role and importance for their own enterprise. The result is an identifiably new business or industry. By contrast, in the case of banking, Mitchell feels that there has been an absence of innovative quality: Banking is not such an enterprise or industry. It has had a pattern of traditional services, an imposed molecular structure, and a pedestrian operating technology, none of which it could call its own. It has not innovated its service products nor shown much adaptive ingenuity in their promotion. Its favorite image has been a passive conformity to the moves of its better customers. Its competitive aggressiveness has been schizophrenic, with large sectors of the industry advocating of supporting publicly administered price ceilings for time deposits, public prohibitions against the absorption of exchange, and a variety of regulatory devices or postures that by sanction or promise dilute competitive ingenuity. [19] Another author has described banking as follows: "Aggressiveness and profit motivation became synonymous to recklessness; backwardness unwittingly became synonymous to conservatism and prudence. The management of a commercial bank is not on the whole very difficult. It calls for prudence, probity, adherence to routine and system, and large acquaintance in the business community." [18] While considerations other than inadequacies in bank management share responsibility for the relatively slow growth of banks, the fact remains that commercial banks have declined in relative importance, and studies of bank attitudes at about the time the COMAC group was formed tend to bear out the picture of banks as being conservative and traditional in their approach to the provision of credit and other financial services

-19 -commonly used by the American economy. Federal Reserve studies of growth of financial intermediaries reported in the Flow of Funds Accounts indicate that commercial banks held approximately 70 percent of the assets of all major financial intermediaries in 1945 but that 25 years later, in 1970, that share had declined to 54 percent. Whitledge surveyed the attitudes of commercial banks in the early 1960s with a view to identifying bank attitudes toward marketing against competitive institutions and other measures of management aggressiveness. In reporting his findings it is of interest to note that 30 percent of banks felt that encouragement of mass use,of instalment credit by individuals for purchase of consumer goods was inconsistent with encouragement of personal integrity and good charactero [22] "A bank which has an aggressive consumer credit instalment loan policy wholly for the purpose of making profit for the bank is not violating any duty to the public, the banking industry, ownership or customers." A full 50 percent of bankers surveyed felt that this statement was not valid, Nearly one-fourth of the bankers Whitledge surveyed indicated that consumer credit was not consistent with good banking, bank safety, and the public service required of a bank. The reported views refer only to the Whitledge sample, of course, and may not be fully representative of all banks. Nevertheless, it seems ironic that as recently as the last decade. a sizeable portion of any sample of bankers considered consumer lending to be a questionable form of credit. After all, a bank's business is lending money, and the fact of the matter is that the need for credit on the part of the household

-20 -sector of the American economy has grown so vastly as a result of a number of influences -- including the extraordinary quantity of funds needed for the construction of new housing, the financing of new cars, and of durable household goods -- that the enormous increase in credit demand from this sector has been perhaps the most striking feature of the American financial scene during the past quarter century. Yet the banking industry was slow to recognize the opportunity and lost the momentum of growth to the savings and loan industry, the sales finance companies, and the credit unions. In another area of banking surveyed by Whitledge it is interesting to note the results of a question that relates directly to the matter of competitive attitude of bank management. About new business solicitation of commercial customers banks were asked: "Is there a general policy against raiding competitor bank customers?" Fully 28 percent of the group stated that their bank avoided raiding other banks' customers. An additional 7 percent replied that they "didn't know." Thus more than a third of the group either had a definite policy against aggressive competition, or no policy at all or at least no policy that was known. This seems especially significant, for commercial customers represent the market in which banks have been the specialists. Certainly one could not call the above attitudes aggressive. It seems a fair assumption that the decline in relative importance of commercial banks over the last 25 years can be only partially explained by the changing economy. It also should be recognized that banks have failed to respond fully to the changes that have occurred in the economy

-21 - and have thus made their decline greater than it might otherwise have been. Banks have not pursued new business aggressively until someone else demonstrated the value of such business. David Rockefeller, Chairman of the Board of Chase Manhattan, has noted this characteristic: But as the years went by and the demands became increasingly complex, many banks did not respond as alertly as they might have. They held back from new fields, turned away from new services. Inevitably, therefore, specialized institutions such as credit unions and savings and loan associations moved in to fill the vacuum, Subsequently they also became major competitors of the commercial bank. [21] V. COMAC Origin and functions The COMAC Company, an acronym for Comprehensive Management Company, was formally brought into existence in 1967 for the purpose of systemizing the management functions in a rapidly growing conglomerate engaged mainly in banking and real estate. In an informal sense, the group of partners who formed COMAC had been in operation for some five to six years before that time, but in 1967 the need for formalizing the group arose from the complex planning, coordination, and direction required for the successful operation of a group of ten banks, a number of large commercial real estate properties, and, eventually, three overseas investment offices. The COMAC partners were interested in acquiring investments in which they would have a controlling interest and in which they would have active management participation. The foctis of their interests for the most part lay in the financial area with side interests in real estate markets. As

-22 -time proceeded, the domain of their operations spread so widely and included such vast numbers of individual business firms, both bank and nonbank, that it became awkward if not impossible for the informal group of a half-dozen to control directly and personally, the individual firms that composed this federation. In 1967 when COMAC was formed the employees numbered less than a dozen, who together with the original six partners did the active managing and directing of the properties. The employees' duties and skills lay for the most part in accounting and financial planning. Rapid expansion from 1967 to 1970 was accompanied by a correspondingly rapid expansion in the number and variety of staff employees and skills represented in the COMAC Company, however, and by the end of 1969 COMAC had 350 employees. Their skills were enormously varied and included not only the original accounting type of skills but the whole range of management techniques -- personnel, managers, purchasing agents, economists, investment analysts, mortgage specialists, and capital market specialists. The COMAC group functioned as a management consulting firm not entirely different from McKinsey or Booz, Allen, Hamilton, although the principal clients of the COMAC Company were firms in which the COMAC partners themselves had strong equity interests. In addition to such clients, however, COMAC also had nonaffiliated clients, such asMichigan Consolidated Gas Company, Michigan Bell Telephone, and perhaps as many as 50 or 60 other nonrelated firms. Nonetheless, the major portion of COMAC revenues were derived from management contracts with firms in which COMAC partners had majority ownership.

-23 - Partnerships Holding Majority Stock of the Operating Entity Managing Partners [- -[ "X" P I / - _N COMAC II ing "Y" Ptsp I "Z" Ptsp Etc. I Etc. Company -- -- 31 EX W1[S~~~~~~~~~~~~~~~~~~~~~~ I i I I I Bank Lans: /Etc. i Ec Etc. Etc Ownership Management contract but no ownership Fig. 1, Functional relatiQnshipa within tfhe. CQMAG group,

-24 -Such clients purchased a variety of management services under a contract system with COMAC. A bank client normally would purchase planning services, accounting services, asset management services, liability management services, and economic forecasting services. The experience acquired in development of new deposit programs for one bank could be carried over to other client banks via the management services of the COMAC group. Favorable results discovered in a given loan or investment opportunity were carried over from one bank to another. On the cost side, it was common for COMAC to find that a new bank client would be carrying certain types of expense, not because they were a necessary part of the bank operation but simply as a matter of tradition. In many instances, COMAC was able to effect substantial cost savings in the operating patterns of client banks. Management gains While the COMAC Company employed some 300 to 350 employees directed by a dozen partners, the fact of the matter was that COMAC in virtually all respects embodied the management aims and personal drive of Donald H. Parsons, Chairman of the COMAC Company. Therefore, it was correct to say that COMAC management aims were fundamentally the management ambitions of the chairman. The strategy by which the general management aim was to be accomplished was fairly simple and straightforward. The most severe competition in the American economy appeared then as now to be in the areas of heavy manufacturing and high technology firms. The manufacturing sector has

-25 -been a high growth area over the past 50 years, while the high science firms represented the most dramatic, exotic, and potentially profitable section of American industry during the decade of the 1960s, COMAC was determined to shun both these areas as objects of investment activity. The reason was simple. They were attractive to so many highly skilled, competitive, owner-management organizations that competition in such industries was likely to be intense. Instead, COMAC elected to specialize in those areas of the economy which had substantial growth potential but which were receiving relatively little attention from such aggressive firms as Gulf and Western, Xerox, IBM, Dow, and McDonnell. At the same time it was necessary to direct attention to those areas where it was financially possible to develop an acquisition plan. For example, the mutual organization of the savings and loan and insurance industries madeit difficult if not impossible to develop a system for taking control of these institutions. The logical institution was one in which there was stock ownership but in which stock ownership was sufficiently concentrated that it was possible to assemble effective control through a stock-offer plan. Commercial banks, particularly modest-sized commercial banks, represented the ideal object of acquisition. Furthermore, it was quite evident even in the late 1960s that banks were not developing loan, investment, and deposit volume up to the potential that was present in many markets. This was evident not only in country bank markets but even in metropolitan areas where controlled entry assured existing firms a satisfactory if not spectacular result. However, COMAC attention was not limited to banks, but included a

-26 - chartered airline, considerable commercial real estate, a ranch in Australia, and an oil field supply company in Singapore. In brief, COMAC explored bank holding company areas off limits to registered holding 5/ companies. The execution of the COMAC strategy was directo It consisted of systematically searching through banking markets, particularly Michigan banking marketing, identifying a bank -- usually in a market where there were at least two other banks -- in which the existing ownership-management group was not bringing the operation up to the potential that the market afforded. Identification of this combination of circumstances could be made from a review of deposit volume, the growth rate of deposits, the loan-to-deposit ratio, the tax position of the bank, and the operating costs of the bank. In short, the profit position of the bank was a revealing measure of the effectiveness of management and ownership in achieving the potential that the market afforded. The ideal situation was one in which there were three or four banks in a market, one of which, frequently a long time member of the market group, lagged in terms of size, growth, service to customers, and profitability to owners. Such a bank might be available at a bargain price -- bargain price in such a case being 125-150 percent of book value of the stock. In many instances, 5/ Eisemann [25] has recently developed a Markowitz type of model to analyze the potential for maximizing gains while minimizing risk in the nonbank subsidiaries of a bank holding company. The model is then used as a measure of investment efficiency in a number of successful bank holding companies.

the existing management-owner group was following traditional pricing practices, traditional marketing prices (i.e., none), and traditional cost control practices (i.e., none), which jointly produced low profits on equity capital, frequently 5-6 percent, or about the same as on a good grade of corporate bonds. Specific markets: Lansing, Kalamazoo, Muskegon The COMAC operation eventually penetrated some twelve banking markets and directed the activities of nineteen banks including two overseas banks, For purposes of this analysis, however, three market areas, Lansing, Kalamazoo, and Muskegon, have been identified as exemplofying the COMAC operation. They were chosen because they are far enough removed from Detroit geographically that they are not subject to strong influence from the Detroit banks and yet they represent markets of sufficient size and complexity to enable us to analyze the impact of COMAC management efforts, It will be of interest to compare the behavior of the three banks in these three market areas as they performed, grew, developed, and changed under COMAC direction, compared to the development and growth of rival banks in their respective communities during the period 1966 to 1970. We might ask: If the three banks had continued under their former ownership and management, would they have followed approximately the trend of development of all banks in the state of Michigan? Using the performance of all Michigan banks as a proxy for the independent growth and development path which these three banks would have taken, we can make some specific comparisons.

I I -28 -All Michigan banks during the 1966,-70 period 'ifcreased in total footings by some 30 percent. Secondly, asset allocations changed significantly for all Michigan banks during this period. Government securities declined from 23 percent of total assets of Michigan banks in 1966 to 13.6 percent by 1970. Obligations of states and political subdivisions, that is, tax-exempt municipals, increased from 11.7 percent in 1966 to 12.6 percent in 1970. Loans as a percentage of assets increased from 53 percent at all Michigan banks in 1966 to nearly 59 percent in 1970. Cash assets declined very slightly from 10.6 percent to 9.9 percent. The ratio of time deposits to total deposits increased from 65 percent in all Michigan banks to 67 percent in 1970. The capital ratio of all banks in the state increased very slightly from 7.8 percent of total assets to about 8.1 percent. Income after taxes as a ratio to capital accounts was 8.9 percent for all Michigan banks in 1966; by 1970 this ratio had increased to 10.25 percent. These changes in the distribution of assets and in the sources of funds and in the operating returns parallel the performance of all banks in the United States. The banking industry as a whole increased in size, shifted asset allocations toward more loans and smaller amounts of Treasury securities and cash, and increased its return on equity capital by modest amounts. Bank of Lansing. Comparing these changes with the changes at the COMAC banks, however, reveals very sharp distinctions, and a summary of the changes which occurred at those banks is revealing. Bank of Lansing for example, increased its total assets from the beginning of 1966 to

the end of 1969 by approximately 150 percent, that is from $66 million to $154 million. This surprising change was funded largely from increases in time and savings deposits. Time and savings deposits were some $33.5 million on the earlier date and increased to more than $70 million by the end of 1969; deposits of states and political subdivisions increased from about $2.5 million to over $16 million; demand deposits of individuals partnerships, and corporations increased from $23 million to $35 million. In addition to these deposits sources, Bank of Lansing employed external markets for sourcing asset acquisitions. It is apparent from condition reports that Bank of Lansing was heavily involved in Federal Funds purchases ($6.75 million) at year end 1969. In addition, by the end of 1.969, Bank of Lansing had developed Eurodollar lending sources. Lastly, capital accounts provided some $4 million at the beginning of 1966 whereas at the end of 1969 total capital accounts amounted to $7.5 million. In addition to these increased sources of funds, asset allocations were shifted drastically during the three-year period under review. In 1966 Treasury securities were nearly 30 percent of Bank of Lansing s asset allocations, whereas at the end of 1969 Treasury securities had decreased to a nominal percentage of total assets, Changes also took place in the Kalamazoo and Muskegon bank situations. In the case of Industrial State Bank of Kalamazoo the increase in total assets between 1965 and 1970 was some 65 percent, while in the case of Muskegon Bank and Trust the change during the corresponding period was also 65 percent in terms of total assets. The major change that enabled

-30 - these banks to grow roughly two to three times the rate of other banks in the state was an aggressive program of fund raising, both through the use of time and savings deposit incentive programs and through the use of nondeposit sources such as the Fed-Funds market and the Eurodollar market. Muskegon Bank and Trust. Muskegon Bank and Trust was acquired by COMAC for the Muskegon market. This bank, smaller than National Lumbermants and Hackley-Union Bank, was number three in size in the Muskegon market, and it probably had the most conservatively designed portfolio. For example, Muskegon Bank and Trust held liquid assets consisting of roughly 8 to 9 percent of cash plus nearly 38 percent in Treasury securities. In addition, the bank held approximately 9 percent of assets in municipal securities. Its total cash and securities as a ratio to total assets in 1961 was more than 56 percent; that is, the bank represented in a sense a large pool of cash. Its loan portfolio was just under 40 percent{ and its capital ratio was approximately 5 percent. This represented a very conservatively managed bank even for 1961. The asset ratios at this bank were fairly stable, cash and Treasury securities comprising close to 40 percent of assets up to the end of 1965 and declining only to about 32 or 33 percent in 1966 and early 1967. Following acquisition by COMAC in 1967 this portfolio policy was changed abruptly. Cash holdings were reduced moderately from about 8 percent to around 6 percent but government securities were reduced very sharply -- from holdings that averaged 30 to 40 percent up to 1963 and averaged generally above 25 percent until 1967 -- to a 10 to 12 percent

-31 - range. By the end of 1969, holdings of government securities were below 5 percent. Incidentally, at that time, holdings of cash were only slightly above 4 percent. On the other hand, state and local government securities were acquired in large quantities to replace the reduced holdings of cash and governments. (In the early 1960s, state and local government securities comprised something less than 10 percent of the portfolio of Muskegon Bank and Trust.) While the management of the bank prior to the COMAC purchase was gradually increasing its holdings of tax-exempt securities as a proportion of total assets, this process was greatly accelerated following acquisition. By the end of 1967 tax-exempt municipals comprised more than 17 percent of the assets of the bank and by the end of 1969 this ratio had risen to 31 percent. Among the consequences of these adjustments in cash and security holdings was, first, their overall reduction from about 56 percent to approximately 40 percent between the early 1960s and the end of the 1960s. However, the change in the composition of the cash and government securities holdings was equally marked. The portion of it that had been made up of cash and Treasury securities was largely eliminated. In 1965 cash and Treasury issues accounted for 80 percent of the bank's security holdings, whereas by 1969 cash and governments represented only 15 percent. State and local government securities had replaced Treasury issues and represented 75 to 80 percent of total security holdings. As noted above, in 1961 the loan volume at this bank was slightly below 40 percent of total assets. This ratio also was changed following

-32 - acquisition by the COMAC group. Although there had been a tendency for the loan ratio to rise during the 1960s and it can be noted that in contrast to the 41 percent ratio in 1961 loans had risen to 43 percent by 1963 and to 45 percent by 1965, under COMAC ownership and management the loan ratio moved upward even further. By the end of 1967 it comprised 49 percent of total assets and exceeded 50 percent on each of the call dates following that. The maximum ratio to total assets was 57 percent in mid 1970. Incidentally, it is interesting to note that the sum of cash, securities,.and loans on the June 1970 call date amounted to some 97 percent of total deposits of the bank, an accomplishment which probably was matched at certain other major metropolitan banks, but certainly was unusual among small-to-medium sized banks in the Midwest. Industrial State Bank and Trust: The Kalamazoo market. Industrial State Bank and Trust was acquired by the COMAC group (actually, of course, the acquisition was made by the Kalamazoo Investment Company) in 1964, and the standard reconstitution of the bank's asset structure was initiated: the municipal portfolio was sharply increased, cash assets and U.S. securities were substantially reduced, and a program of time-deposit expansion was undertaken. The time-deposit program was based upon the growing awareness among savers that interest on bank deposits was worth having, and the savings and loan industry was succeeding on a largely price-based competition. ISB adopted a program of soliciting both Passbook savings accounts and Certificate of Deposit accounts, generally setting the pace in the

-33 -Kalamazoo market insofar as rates were concerned. A major part of the marketing effort was aimed at individual depositors, but business and government deposits were not shunned. The results were striking. Private time deposits -- mainly individuals' deposits -- increased from $20 million to $58 million between 1964 and 1970, while public time deposits increased from less than $5 million to more than $16 million. Efforts were made to increase demand deposits also and with a certain amount of success. During the six years between 1964 and 1970, demand deposits show growth of some 50 percent. However, demand deposits cannot be attracted on a straight price basis, as is the case with time deposits within the limits of Regulation Q. Price bidding is evidenced indirectly by the introduction of low-minimum balance checking account plans. The overall plan for developing a bank's potential market was to seek out all sources of funds available to the bank, proceeding from the least costly funds (demand deposits) to the most costly funds (equity capital). The purpose was not to minimize the average cost of money, the presumed aim of the industrial financial manager, but to continue acquiring funds as long as the cost was below the net yield on marginal assets. Following this philosophy meant developing sequentially the several sectors of the bank funds market. Demand deposits are the least costly source of funds, followed by passbook savings. Certificate of Deposit funds were the third segment of the deposit market, and development of this segment resulted in substantial enlargement of public deposits as noted above. (The ratio of publid deposits to total deposits at ISB increased from 13 to 18 percent between 1964 and 1970.)

-34 - In addition to solicitation of deposit funds, the COMAC banks purchased (borrowed) funds in the Federal Funds and, when possible, in the Eurodollar market. Federal Funds purchases, never in evidence at ISB prior to the change in ownersmanagement in 1964, were regularly in evidence by 1966. The amounts varied, and call date figures probably are not representative, but it is clear that as much as 9 percent of total funds in use at the bank were purchased from the Federal Funds market at certain times. Eurodollar borrowings are not identified as such in the bank's periodic financial statement. but presumably appear under the heading "other liabilities." Amounts up to $3 million were successfully borrowed in the Eurodollar market. On the asset management side of the balance sheet, ISB showed the results of COMAC planning promptly after the acquisition was completed. The first change was a sharp build-up in the bank's loan portfolio, as so-called loan packages were sold by other COMAC entities to ISBo At year end 1964, ISB's call statement showed a very conservative 41 percent loan-to-total-assets ratio. (This at a time when Michigan banks were around 46 percent and Detroit banks around 50 percent.) But six months later, the loan-to-assets ratio had jumped to 65 percent and continued at between 60 and 90 percent through 1970. The second change was a sharp decrease in the ratio of cash assets and U.So Securities to total assets. Cash and Governments constituted 43 percent of assets in mid-1964 and 48 percent at year-endo Six months

-35 -later this ratio had declined to 29 percent and a year later to 21 percent. After mid-1966, a build-up in municipal holdings became evident and at year-end represented 19 percent of ISB assets, compared with 7 to 8 percent in earlier periods. VI. COMAC AND THE REGULATORY AUTHORITIES Regulatory authorities dealing with COMAC banks changed gradually from a posture of uncertainty and "wait and see" to a position of warm support at the time that the Bank of the Commonwealth took over the defunct Public Bank assets and liabilities. Later, however, the regulators gradually moved to a position of doubt and uncertainty and, eventually, to very strong hostility. The change was brought about in part by tangible, objective differences in management viewpoint. These differences in views obiStasset-liability management in part represent the divergence between a new, different outlook and a more traditional one, but in part the differences were also the result of an extremely abrasive and cavalier manner adopted by personnel of the COMAC group. This writer feels that to no small degree the difficulties in which the COMAC group eventually found itself were caused by failure to maintain good relationships and full communication with -- and~ thus to elicit sympathetic support from -'- the regulatory agencies. Probably no single act on the part of the COMAC group triggered the eventually negative attitude of the regulatory authorities. It is more likely that a variety of individual management decisions by COMAC disenchanted them.

-36 - During 1968, and more intensively during 1969, regulatory authorities sought to develop a change in the manner of operation of the COMAC banks, First and most importantly, the authorities sought to limit further acquisitions of banks by the COMAC group. Secondly, regulatory authorities sought to modify major policy characteristics of the COMAC banks, including their heavy dependence upon the Federal Funds market noted earlier and their tendency to deemphasize the need for raising new capital. None of these considerations were in fact ignored by COMAC. There was active, urgent concern to achieve a higher capital ratio and a greater degree of liquidity. But at no time did the COMAC group bend far enough either to satisfy regulatory authorities or to reassure them that there would be full accommodation and adjustment at some time in the future. The final outcome of the increasing divergence of feeling and the divergent opinions as to the appropriateness of COMAC management recommendations to their client banks occurred on March 31, 1970, when the Federal Reserve announced a denial of application by one of the COMAC banks, Bank of the Commonwealth, to establish a foreign branch in Nassau and an Edge Act Corporation. This was the first public refusal by the Federal Reserve to approve an off-shore branch application. There may have been de facto disapprovals in the form of a suggestion that the applicant withdraw its application in order to avoid disapproval, but the Bank of the Commonwealth disapproval was the first case in which the Federal Reserve had explicitly denied an application for a foreign branch. As important as the denial was the language used by the Board in pronouncing it:

-37 -In the circumstances, the Board has concluded that the general character of the Bank of the Commonwealth management and the bank's financial history and condition, including its liquidity and capital position, mitigate against approval of the present applications; that the adverse effects of the establishment of the proposed branch would outweigh any benefit accruing to the convenience and needs of the bank,o, that the approval of the application will be contrary to the public interest. The impact of this statement can hardly be overestimated. It called into question the veracity of the management of the bank, the responsibility of the management of the bank, and eventually it produced question as to the actual viability of the bank. This had repercussions on all COMAC banks, for the Bank of the Commonwealth was without any question the major banking element in the entire COMAC group. The Federal Reserve's harsh criticism of the management of the Commonwealth was followed by a reduction in the bank's demand deposits by some 18 percent between January 1 and April 30, 1970, By June 30 the bank's deposits remained down by some 7 percent. During that same six-month period other Detroit bank deposits declined only 4 percent. Even more sensitive than demand deposits, however, was the attitude of banks which had been lending funds to the COMAC group via the Federal Funds market. They clearly were made very uneasy by the Federal Reserve statement. While one cannot say that the Federal Reserve was responsible for the ultimate demise of the COMAC operation and for the dissolution of the COMAC banking group, there does seem to be a link between the Federal Reserve's statement in early 1970 and the problems which ultimately became catastrophic for COMAC in their magnitude.

-38 - The central characteristic of the COMAC Banking Operation is that it was designed around the holding company plan without employing the legal structure of the holding company. At the time the COMAC banking group was formed, holding companies were not provided for under Michigan corporate law, and hence it would not have been permissable to form a corporate holding company. Instead, the COMAC Company represented a management partnership which performed a supervisory, shepherding function over some nineteen banks. Each bank was owned in major part by an independent partnership, the partners in which were identical or had a major crossover with the COMAC partners. Hence, the total apparatus provided effectively for a system which managed, owned, and controlled a number of banks, without infringing on Michigan banking law. It was this structural characteristic, embodying many of the features of the bank holding company despite the fact that legally COMAC was not a holding company, which made the COMAC operation distinctive among Michigan banks. However, this was not its only distinctive characteristic, for COMAC engaged in asset and liability management planning much more aggressively than typical midwestern banks. Basically, the asset-management plan was designed to maximize profits, regardless of whether the resulting asset portfolio of the bank was conventional in format. For example, COMAC would select an application of funds, that is, a portfolio application, which represented the best and most profitable use of funds at the margin. This might in fact mean that there would be no accommodation of customer loan requirements during certain periods when

-39 - it was more profitable to invest bank funds in open-market securities, such as municipal bonds, commercial paper, or Treasury securities. This asset-management plan ran counter to the traditional position of banks that they should be institutions which accommodate to local market credit needs. To understand the extent of COMAC's departure from conventional banking moves, one must realize that profit maximization has always been a difficult problem for bank management to deal with. For example, Whitledge in his survey of bank officers reports the responses to the two following questions: I. Is the objective of your bank to serve the public in the manner and spirit of a quasi public corporation? 2. Is it the objective of your bank to make as much net profit over as long a term as possible? The answer to the first question was affirmative among 55 percent of the bank officers queried, while responses to the second question were negative in almost two out of three cases [22]. On the basis of these responses, one is led to believe that a majority of bankers tend to reject profit as their primary goal. Other researchers have also tried to assess bank management aims, including Anderson and Burger in a recent study sponsored by the Federal Reserve Bank of St, Louis [23]. That study postulates that if banks are profit maximizers they will respond to changes in market variables differently than will banks which are loan accommodators. The study then attempts to predict how banks will

change their preferred holdings of excess reserves, borrowings, and loans under alternative hypotheses. The results of the Anderson-Burger study suggest that bank managements may have begun to exhibit more aggressive behavior both among themselves and toward other financial intermediaries in recent years but that the "accommodation" principle is still conventional wisdom among many banks. A second important concept in the COMAC philosophy is the idea that a bank should be aggressive in the retail savings market, the conviction being that if a large proportion of total deposits was in the individual savings category the bank would have an extremely stable deposit base. It was felt that this type of deposit was somewhat insensitive to interest rate differentials and that, to the extent that there was a sensitivity, the problem could be minimized by paying maximum allowable interest rates and using aggressive marketing techniques. If the above philosophy is correct two things seem obvious: First, the strength of or need for strong customer relationships is diminished. Second, there is need for high gross income to pay for the high proportion of relatively high-cost time money. This philosophy of management has implications for problems of liquidity management. If the requirement to maintain the relationship has minimized, then the need for liquidity arises from three sources instead of the usual four. Liquidity to meet future loan demands is no longer a big factor, but liquidity is still needed for the uncertainties of: (a) future deposit levels, (b) degree of loan defaults, and (c) changes in the value of investments.

-41 - Liquidity is the ability to cope with reductions in asset values or withdrawal of deposits. From the asset side, need for liquidity can be accommodated by short-term investments, by maintenance of a highquality loan portfolio, and by holding large amounts of cash. However, as noted earlier, these policies tend to be costly in terms of profit. From the liability side, a bank can also gain liquidity by minimizing deposit volatility. Secondly, a bank can accomplish liquidity by developing potential sources of funds that can be tapped as occasion demands. Hence, the two basic methods of achieving liquidity are by having assets that are convertible to cash and by having a source of funds that can be developed to replace lost deposits or to meet needs for incremental deposits. Traditional methods of meeting liquidity needs have concentrated on the first technique. A large proportion of bank assets have traditionally have concentrated in what are called primary and secondary reserves. A decade back nearly fifty percent of all bank assets would have been in the form of cash and government securities. While these very high proportions have been reduced in recent years, it is still true that many banks depend heavily upon a liquid portfolio in order to meet the liquidity requirements. The COMAC philosophy on the other hand stressed the concept of a bank's liquidity being met by the simultaneous creation of an asset and a liability. A part of this philosophy was the idea that the holding of municipal securities was not needed for liquidity purposes. This is an extension of the nonaccommodation posture in portfolio design described earlier. There are two main points in the COMAC approach to liquidity management: First, a bank does not

-42 - need to maintain a great proportion of short-term liquidity in its asset accounts. The second aspect of the COMAC liquidity management is use of a bank's ability to create claims against itself to meet immediate liquidity demands. Placing primary emphasis on this approach frees all but an absolute minimum of cash in governments for pure investment purposes. Besides these two techniques, COMAC used the more generally accepted tool of liquidity management, asset sales and participations. Earlier in this discussion it was pointed out that the bank using liability creation for liquidity had to insure itself of constantly being able to attract new funds. It has been pointed out that because of interest rate ceilings a bank could not always depend upon being able to attract new deposits. Thus, it had to be assured of being able to obtain loans on the Federal Funds market. To insure this source of liquidity, the COMAC banks followed what is believed to be a unique practice in the Federal Funds market. COMAC arranged intermediate-term agreements with nonrelated banks to supply a specified level of Federal Fund loanso This amounted to a secured line of credit in the Federal Funds market. The COMAC banks paid fees for these lines of credit, but the lines of credit provided what the banks needed: an assured source of funds to meet liability drains. It became evident through the late 1960s that this technique met a very important need, for, as deposit rates reached regulation Q ceilings and it became difficult to raise funds in the deposit market, time deposits as a new source of funds for the COMAC-related banks tended to diminish, and the Federal Funds market dramatically increased as a source of funds. The volume of such Federal

-43 - Funds borrowing attained importance to the degree that it could truly be said that some of the COMAC banks were borrowing their total reserve requirements. Reviewing the distinctive characteristics of the COMAC-related banks, they were as follows: First, loans were a relatively less prominent part of the bank's portfolio than is traditionally the case; second, municipal securities, particularly long maturity issues, were an extremely prominent part of the COMAC banks': portfolios; third, time deposits as opposed to demand deposits were an important source of funds to the banks; and finally the Federal Funds. market was an immensely greater source relatively for COMAC banks than for banks generally. In addition to these asset and liability management characteristics, it should be noted that the general growth in size of the COMAC banks caused their capital ratios to diminish to a level that would be regarded as near the lower margin for Michigan banks. In addition there is an intangible factor which needs to be stated even though its exact importance in conjunction with the other characteristics of COMAC management is difficult to assess. The quality of the personal relationships with regulatory authorities came to be well known during the period of COMAC activity. Most banks adopt a respectful, though somewhat restrained, posture toward the regulatory authorities. Bank examiners are handled with a good deal of tact and diplomacy. And while a regulated bank may not fully accommodate to all of the requests of the regulatory agency, whether that agency is the comptroller of the

-44 - Currency, the Federal Reserve, or the Federal Deposit Insurance Corporation or the state banking commissioner, the bank would be very careful not to develop hostile relationships between itself and regulatory personnel. An alternative hypothesis Some colleagues have suggested an alternative explanation for the resistance which COMAC provoked. It is useful to state this hypothesis in clear and extreme form, so that we can then consider where the truth may lie. I will present the hypothesis as a series of points. 1. COMAC did achieve entry by takeover, with great speed and effectiveness, This process parallelswhat bank holding companies are able to do in larger markets, though they rarely do so. 2. COMAC did not, however, replace managers in large numbers, preferring instead to set new policies. Its takeovers were aimed at new behavior, with a minimum of upset to prior arrangements. 3. Many of the target banks were classic cases of high inefficiency, offering dramatic increases in profitability under new policies. It is not clear that any such degree of inefficiency exists widely in larger metropolitan banks. 4. COMAC s changes were: a shift in portfolio toward state and local bonds and a small degree of price competition to attract savings deposits. 5. The effect on earnings was dramatic.

-45 - 6. The effect on other banks was not clear. Given a longer time to work, COMAC might have forced more competitive behavior on its rivals. But this process apparently takes more than two or three years. 7. COMAC was belled by public regulators, in a series of specific acLions. 8. This intervention was not justified by the economics of the COMAC member banks, for generally they were in a good position to survive the 1969 credit crunch. 9. There are two possible and alternative interpretations of the regulators' action: eitherj(a CCOMAC was so abrasive that regulators sought revenge and a reassertion of their authority along conventional channels, or (b) the regulators acted as agents of the conventional banking establishment, to which COMAC was a threat. If so, COMAC could not have escaped by being more diplomatic. 10. The opportunities which COMAC exploited are not widely available in large-scale banking, though they may exist in many smaller localities. 11. Still, takeover can powerfully affect banking behavior. 12. Unless regulators are at pains to be neutral or even to promote takeover, they will easily slide into the role of stopping takeover in -- or as if they were acting in -- the interests of established banks.

-46 - This alternative hypothesis is highly negative, and yet it accords with the most acute recent interpretations of regulatory behavior, especially of bank regulation. It is in banking markets that the protective role of regulation has become perhaps most fully developed. It may serve valid economic goals of security and efficient banking, But in this case it did terminate a sound and efficient experiment in modern banking, which was using the traditional market processes to improve performance. Both hypotheses can be supported by all or most of the facts about the whole COMAC episode. I prefer the first, milder alternative. But I cannot reject the second one. VII. SUMMARY The preceding sections of this paper describe the origin and development of the difficulties which the COMAC group experienced during the period between 1966 and 1970. At this point we should return to a review of the overall philosophy of the COMAC group and the relationship and success which this philosophy evidently achieved, taking each aspect of the philosophy and briefly analyzing its practicality and effectiveness. The cornerstone of the COMAC philosophy was the idea that a stable base of time deposits could be attracted to the banks. It was felt that by making this type of deposit the major source of its funds, a bank could substantially reduce its short-term liquidity problems. Because of insufficient data it is not possible to determine if each of the banks was successful in this respect, but they did attract more than their share of

-47 - total time deposits in the three markets which we have reviewed. It is not possible to determine what that proportion was of the class of deposits from which those banks had been drawing. It is hypothesized here that two things occurred with respect to deposits during the period 1966-69. First, it seems reasonable to assume that the banks were initially successful in attracting the desired deposits and that they were in fact a stable source of funds. The banks apparently went through the credit crunch of 1966 without serious problems. Very little use, for example, was made of the Federal Reserve Discount facility, Further, it is apparent that the banks were able to continue expanding their sources of funds through the period of credit tightness in 1968 and 1969. Even in 1969, when the credit crunch was having its major impact on all banks in the United States, COMAC banks experienced significant net growth. The market for large certificates of deposit was a dependable source of funds and provided a stable deposit base so long as the regulation Q ceiling provided enough room for aggressive banks to compete on a price basis. The banks that were examined in this paper are banks which were mainly price competitors, despite the effort to carry out the creation of competitive services in other areas. But by the end of 1969 regulation Q ceilings were having a very severe damping effect upon time-deposit growth and consequently upon banks such as the COMAC group, which depended upon price competition to attract funds. It appears then, that the cornerstone of the COMAC philosophy may have been well founded, but the bank was unable to adhere to it 100 percent

-48 - because of regulation -- i.e., a ceiling on deposit rates, which, in a tight credit period, prevented the group from continuing to be vigorous, aggressive competitors. Another major aspect of the COMAC philosophy was profit maximization in asset management. Inherent in this approach was the breaking down of the traditional customer relationship. It seems possible to hypothesize two things that resulted from this asset-management plan. First, the weakened customer relationships made COMAC banks susceptible to severe deposit drains. COMAC banks- did.suffer from heavier than average drains in 1969, although this does not seem to have been true in 1966. Second, it has been suggested that because of the COMAC banks' investment policy they were able to establish substantial customer relationships with municipalities. However, in 1969-70 when the real test came, these relationships proved unstable. Municipal and state treasuries were hard pressed to meet the rising demands for funds and grew sensitive to maximizing income at all points. Thus as market rates climbed above the Q ceilings, municipal treasurers could not continue to justify bank deposits. To gain the added income they had to break the customer relationship which had been established with COMAC banks. In all probability this phenomenon served to increase the rate of disinteremediation which the COMAC banks experienced. The COMAC philosophy specifically indicated that municipal bonds would not be used as a source of liquidity. Accordingly, COMAC-managed banks tended to have relatively long municipal portfolios because the longer end of the term structure afforded

-49 - a significantly higher yield. Specifically, if the COMAC banks had preserved 25 percent of their municipal bond account in liquid form, 50 percent for reacting to market changes (i.e., pure profit maximization), and 25 percent invested permanently in long-term issues, they would have had a distinctively better liquidity position at the end of 1969 and the start; qf 1970. It is not possible to say that this would have enabled the COMAC banks and the COMAC groups to have survived the first half of 1970, but it surely would have relieved some of the pressure on borrowings. There is nothing to suggest that borrowing short-term liquidity (liability management) was wrong. The facts show no problems with the approach until about March of 1970. The technique was obviously the immediate cause of the COMAC debacle, but this was mainly because other parts of the philosophy had proved unworkable in their implementation. There would be certain macro-economic problems if the approach were to be used universally, of course, but this is not to say that it is not workable for an individual bank or banking group.

-50 - References 1. Bowers, Robert D. "Businesses, Households and Their Banks." Business Review (Federal Reserve Bank of Philadelphia) fIar. 1969: 14-19. 2. Chandler, L.V. "Monopolistic Elements in Commercial Banking." Journal of Political Economy, Feb. 1938: 1-22. 3. Fortune Magazine Research Department. Business and Banking: a Fortune Survey, New York: Time, Inc., 1965. 4. Fraser, Donald R., and Rose, Peter S. "Bank Entry and Bank Performance." Journal of Finance, Mar. 1972: 65-78. 5. Guttentag, Jack, and Herman, Edward. Banking Structure and Performance, New York University of.Institute of Finance Bulletin No. 41-3. New York: By the Institute, 6. Jessup, Paul F. Changes in Bank Ownership: The Impact on Operating Performance. Washington, D.C.: Board of Governors of the Federal Reserve System, 1969. 7. Johnson, H.L., "Commercial Bank Markets: A Case Study." University of Missouri, Business and Government Review (July-Aug. 1966): 15-21. 8. Kaufman, G.G. Customers View Bank Markets and Services: A Survey cf Elkhart Indiana, Chicago.: Federal Reserve Bank of Chicago, 1967. 9. Kreps, Clifton H., Jr., "Character and Competitiveness of Local Banking: A Summary." Unpublished paper, Federal Reserve Bank of Richmond, 1965. 10. Lawrence, Robert J. The Performance of Bank Holding Companies. Washington, D.C.: Board of Governors of the Federal Reserve System, 1967. 11. _ Operating Policies of Bank Holding Companies Part I. Washington, D.C.: Board of Governors of the Federal Reserve System, 1971. 12. Schmid, S. "The Size of Bank Loan Markets." (Board of Governors of the Federal Reserve System), Staff Economic Comment, No. 26, Oct. 1966: 14-45. 13. Staats, William F. "Corporate Treasuries and Their Depositories," Business Review (Federal Reserve Bank of Philadelphia), Mar. 1969: pp. 4-13.

-51 - 14. Stiles, T.A. Busieneses View Banking Services: A Survey of Cedar Rapids, Iowa. Chicago: Federal Reserve Bank of Chicago, 1967. 15. Talley, Samuel H. The Effect of Holding Company Acquisitions on Bank Performance. Washington, D.C.: Board of Governors of the Federal Reserve System, 1971. 16. Hodgman, Donald R. Commercial Bank Loan and Investment Policy. Champaign, 1963. 17. Goldfield, S. Commercial Bank Behavior and Economic Activity. Amsterdam: North Holland Pub. Co,, 1966. 18. Garg, K. Principles of Banking. Calcutta: Kitab Mahal Ltd., 1961. 19. Mitchell G.W. "Exogeneous Forces in the Development of Our Banking System." Law and Contemporary Problems, Winter, 1967. 20. Board of Governors of the Federal Reserve. Flow of Funds Accounts, 1945-72. Washington, D.C. 1973. 21. Rockefeller, David. Creative Management in Banking. (By the author, 1971). 22. Whittledge, L. Vitalizing Banks. Boston: Bankers Publishing Co. 41964. 23. Anderson, L.C., and Burger, A.E. "Asset Management and Commercial Bank Portfolio Behavior." Journal of Finance, May 1969..3 -24, Chandross, 25. Eisemann, Peter C. A Test of the Diversification Motive for Congeneric Bank Holding Companies. Ann Arbor: University of Michigan Microfilms, 1973. 26. Alhadeff, David, "Monopoly and Competition in Banking," The University of California Press, Berkeley, 1954. To be completed in a later draft.