Printer Friendly

The nonbank activities of bank holding companies.

In 1970, Congress amended the Bank Holding Company Act of 1956 to establish a new framework for the expansion of bank holding companies into nonbank activities. This article provides some historical background on the nonbank activities of bank holding companies. It also uses 1988 data from two relatively new bank holding company reporting forms to describe the extent of these nonbank activities and their contribution to the financial condition of bank holding companies. HISTORICAL BACKGROUND In the early 1900s, the bank holding company emerged as a new form of bank ownership. While the permissible business activities of banks were limited by state or federal law, bank holding companies were not banks and therefore were not subject to those limitations.

Relatively little is known about the extent of nonbank activities of early bank holding companies, but many of the companies appear to have been formed as a means of expanding geographically into areas in which branch banking was restricted, rather than as a means of engaging in activities prohibited to banks. Although a few bank holding companies owned major nonbank subsidiaries, most bank holding companies that expanded beyond banking combined a small bank with a company providing another financial service, such as an insurance agency. (1) The Bank Holding Company Act of 1956 Although the Glass-Steagall Act of 1933 made some provision for the regulation of multibank holding companies, the opponents of holding companies pressed for more restrictions. After numerous unsuccessful attempts during and after World War II, additional legislation to restrict the expansion of multibank holding companies was passed in 1956. (2) The motivation for this legislation, the Bank Holding Company Act of 1956, has been ascribed to various factors; most explanations focus on the fear of financial concentration resulting from a few multistate bank holding companies acquiring control over a large percentage of nationwide banking assets. This concern was addressed by the restrictions on interstate banking in the Douglas Amendment to the Bank Holding Company Act. The Douglas Amendment effectively prohibited interstate bank holding companies unless individual states specifically permitted their banks to be acquired by out-of-state holding companies.

In addition to the threat of concentration posed by extensive interstate banking, concentration also could result from the affiliation of major nonfinancial corporations with leading banking organizations. Many feared that huge banking and industrial conglomerates of this type would dominate the economic system. Therefore, restrictions were placed on the nonbank activities of bank holding companies.

Restrictions on the nonbank activities of multibank holding companies were based on several factors besides the concern with economic concentration. First, some believed that a bank holding company should not be permitted to perform activities that could not be performed directly by a bank. By this line of reasoning, if the activity was not safe or appropriate for a bank, then it was not safe or proper for an organization owning a bank. Second, many nonbank businesses feared that firms affiliated with banks would gain a competitive advantage over unaffiliated competitors in the same industry. These businesses were concerned that firms affiliated with banks would receive preferential credit treatment from the banks and would have access to low-cost funds provided by them from non-interest-bearing deposits. Third, there was a persistent fear that a bank would tie access to credit to the purchase of services provided by its nonbank affiliates. For example, if all of the bank's commercial borrowers were required, as a condition of obtaining credit, to buy their business travel services from the bank holding company's travel agency, the independent travel agencies would be unable to compete.

Given these concerns, the framers of the Bank Holding Company Act of 1956 restricted nonbank activities very severely and permitted mainly those activities incidental to banking or performing services for banks. Approved activities included ownership of the bank's premises, auditing and appraisal, and safe deposit services. In general, the law required that nonconforming nonbank businesses be divested over a period of years, but the Federal Reserve Board was given the power to allow retention of activities in the areas of banking, finance, or insurance if these activities were ". . . so closely related to the business of banking or of managing or controlling banks as to be a proper incident thereto. . . ."

As required by the 1956 legislation, the Federal Reserve Board evaluated applications to retain existing nonbank subsidiaries throughout the period between 1956 and the passage of the 1970 Amendments. Fewer than thirty applications were submitted, and most approvals were for insurance-related subsidiaries; these affiliates were either insurance agencies to be operated in conjunction with a bank or underwriters of credit-related insurance. The Board's analysis required that the nonbank subsidiaries be functionally or operationally related to the bank: The sale of insurance in conjunction with granting a loan was viewed as different from the general sale of insurance. The major denial during the period was Transamerica Corporation's application to retain Occidental Life Insurance Company.

The relatively restrictive provisions of the Bank Holding Company Act of 1956 effectively prohibited interstate banking and limited the expansion of nonbank activities. The number of multibank holding companies continued to grow, however. Forty-seven separate multibank holding companies controlling 7.5 percent of insured commercial bank deposits were in operation at the end of 1956. By the end of 1970, there were 111 multibank holding companies controlling 16.2 percent of commercial bank deposits. Only a few of the smaller multibank holding companies operating before the passage of the 1956 legislation divested banks after the law was enacted so as to become one-bank holding companies and avoid regulation. The one-bank holding companies, although not yet regulated, were a potential source of many of the same difficulties that their opponents associated with multibank holding companies. The One-Bank Holding Company and the 1970 Amendments In 1955, there were approximately 117 one-bank holding companies with $11.6 billion of insured commercial bank deposits, or approximately 6 percent of the total of such deposits. These organizations were generally small and combined small bank subsidiaries with a nonbanking activity; the nonbanking activity most typically was insurance. Because the one-bank holding companies were not subject to the Bank Holding Company Act of 1956, the way was open to use this type of organization to expand into nonbank activities. In the mid-1960s, the rush to form one-bank holding companies began; by 1968, there were 783 one-bank holding companies with $108.2 billion of deposits. Thirty-four of the nation's 100 largest banking organizations-including 7 of the top 10-had formed or announced plans to form one-bank holding companies.

By the end of 1970, one-bank holding companies owned nonbank subsidiaries operating in a wide variety of industries. Most frequently these firms participated in financial activities, such as insurance and real estate, but they were also engaged in businesses ranging from animal husbandry to water supply. In terms of the government's industrial classification system, 276 three-digit Standard Industrial Classification codes were represented on the list of nonbank activities being performed by one or more holding companies.

The long-standing concerns about aggregate concentration and the separation of banking and commerce were revived as the nation's largest banking organizations formed one-bank holding companies. The Congress responded by enacting the 1970 Amendments to the Bank Holding Company Act of 1956. The 1970 Amendments extended bank holding company regulation to one-bank holding companies and established rules for nonbank activities that were applicable to all bank holding companies.

The 1970 Amendments established a two-part test for the permissibility of proposed nonbank activities. First, a proposed nonbank activity must be closely related to banking. "Closely related to banking" has been interpreted by the Federal Reserve Board and the courts that have reviewed Board decisions to include the following: (1) activities in which banks have traditionally engaged, (2) activities that are so closely related to traditional activities that banks are well equipped to engage in the activity, or (3) activities that are integrally related to permissible bank activities. The second step in the test requires that the proposed activity be a proper incident to banking. For this test, the law established specific factors to be weighed by the Board in making its decision. In weighing the public benefits and costs, the Board was instructed to examine potential gains, such as greater public convenience, increased competition, and greater efficiency, and then to determine if these gains would outweigh any possible costs, such as increased concentration, decreased competition, unfair competition, conflicts of interest, or unsound banking practices.

The Congress also had to resolve the future status of those nonbank activities that were being conducted by bank holding companies before the passage of the 1970 Amendments. The law provided ten years for bank holding companies to divest those impermissible activities begun after June 30, 1968. Activities conducted before June 30,1968, were grandfathered, although the Federal Reserve was required to review the grandfather status of otherwise impermissible activities engaged in by a bank holding company with more than $60 million of bank assets.

Under the rules established by the 1970 Amendments to the Bank Holding Company Act of 1956, holding companies have continued to expand their role in the American banking system. More than 90 percent of domestic banking assets are now held by banks owned by bank holding companies: 70 percent of these assets are held by banks owned by multibank holding companies, and 20 percent by one-bank holding companies. Not all of these holding companies have been established for the purpose of engaging in nonbank activities; many were established for tax reasons or for interstate or intrastate geographic expansion by banks. Nearly all of the nonbank activities considered permissible for bank holding companies are also permissible for national banks. THE DATA COLLECTION SYSTEM FOR NONBANK ACTIVITIES Although considerable research has focused on the performance of banks owned by bank holding companies, little work has been done on the characteristics of the nonbank subsidiaries. In part, this lack of research is attributable to a lack of consistent data. Beginning in 1986, however, data on the nonbank activities of bank holding companies have been available from two new forms, the FR-Y11Q and the FR-Y11AS, which are filed by all bank holding companies with consolidated assets of more than $1 billion and by those with assets of more than $150 million and nonbank activities above certain levels.

The FR-Y11Q, a quarterly report form filed by each covered bank holding company, consists of a brief balance sheet and a profit and loss statement that consolidate data for all of the nonbank subsidiaries of each bank holding company, excluding Edge and agreement corporations and subsidiaries of the affiliated banks. The FR-Y11AS, which is filed annually by each covered bank holding company, also consists of a brief balance sheet and an income statement, but the data are aggregated by line of business. For reporting purposes, the bank holding company assigns each of its nonbank subsidiaries, according to the primary activity of that subsidiary, to one of eleven specified categories. The data are then aggregated within activity categories so that all consumer finance subsidiaries are reported as a unit, all commercial finance subsidiaries as a group, and so on. This grouping of data by activity permits an analysis of each category, although there can be some mixing of activities within a given subsidiary or within categories. For example, a holding company's commercial finance subsidiary may also engage in mortgage banking as a secondary line of activity, but all of its assets would be allocated to commercial finance.

The FR-Y11Q and FR-Y11AS are filed by a relatively small group made up mostly of very large bank holding companies. In 1988, 284 banking organizations filed these forms compared with 298 organizations in the previous year. This number of organizations in 1988 represented less than 3 percent of all U.S. banking organizations and only 4.8 percent of all bank holding companies. These 284 organizations, however, controlled $2.36 trillion of total assets, or 87 percent of the $2.72 trillion of assets (combined bank and nonbank) held by all 1,409 bank holding companies filing on the consolidated financial statement, the FR-Y9. BANK HOLDING COMPANY INVOLVEMENT IN NONBANK ACTIVITIES The 1988 share of nonbank assets in total assets for the 284 reporting bank holding companies is relatively small. However, this share has increased significantly since 1986, and a few large bank holding companies have large holdings of nonbank assets. An examination of the types of nonbank activities shows that many of the assets are held in what are considered to be traditional banking activities. Subsidiaries engaged in securities brokerage, a nontraditional activity, have experienced the largest rate of asset growth. Consilidated Nonbank Activities At the end of 1988, total nonbank assets held by the 284 bank holding companies reporting on the FR-Y11Q were $175.5 billion, which was 7.43 percent of the consolidated bank and nonbank assets of these firms (see table 1). Further, this amount represents an increase of 7.9 percent in the volume of nonbank assets held by 298 firms reporting on the FR-Y11Q at year-end 1987. In contrast, total bank holding company assets increased only 1.3 percent.

The values of nonbank assets reported on the FR-Y11Q overestimate those actually devoted to nonbank activities. The reported asset values include balances due to the nonbank subsidiaries from their parent company and their affiliated banks. To calculate net nonbank assets, these intrafirm balances should be subtracted from total assets, just as they are netted out in calculating the total assets of the consolidated bank holding company. The total volume of net nonbank assets held by the 284 reporting firms was $164.0 billion at the end of 1988, which represents 6.94 percent of consolidated bank and nonbank assets of these firms. Comparable volumes of net nonbank assets were $146.8 billion for the end of 1987 and $133.9 billion at the end of 1986, representing increases in the volume of net nonbank assets of 9.6 percent from 1986 to 1987 and 11.7 percent from 1987 to 1988.

The 1988 ratio of net nonbank to total bank holding company assets of 6.94 percent has grown significantly since 1986 and is substantial enough to warrant attention. In particular, the ratio of net nonbank assets to total assets for the reporting bank holding companies increased 10.3 percent from 1987 to 1988 and 6.6 percent from 1986 to 1987. By contrast, assets of the reporting bank holding companies increased only 4.1 percent over the 1986-88 period. Furthermore, large disparities exist among organizations in the amount of nonbank assets held, and total nonbank assets are very large for a few bank holding companies.

The disparity in the holdings of nonbank assets is apparent in comparing them for banking organizations ranked by net nonbank assets. As the data in table 2 indicate, the ownership of nonbank assets is highly concentrated. The top five firms (in terms of net nonbank assets) held 57.2 percent of the total held by the 284 firms. The top ten firms held 73.4 percent; the top fifty, 96.9 percent; and the top one hundred, 99.5 percent. The remaining one hundred eighty-four firms filing this report collectively held less than 1 percent of all net nonbank assets, and eleven of these firms reported that they had no nonbank assets.

The data in table 2 also show that the top five bank holding companies also averaged the largest totals of bank and nonbank assets. The average size in terms of total assets of the top five firms ranked by net nonbank assets was $106.7 billion; for the top ten, the average size was $67.5 billion, while the average size for the two hundred eighty-four reporting firms was only $8.3 billion. Further, those organizations with large dollar amounts of nonbank assets also had relatively high ratios of nonbank to total assets. The top five banking organizations in terms of total nonbank assets had 17.6 percent of their total assets in nonbank activities, and the top ten had 17.8 percent; whereas the comparable figure for all reporting firms was only 6.9 percent.

Considerable overlap exists between the fifty largest bank holding companies in terms of consolidated bank and nonbank assets and the fifty largest in terms of nonbank assets. The fifty largest in terms of total assets collectively held $154.4 billion, or 94.2 percent of the total net nonbank assets of $164.0 billion at year-end 1988. Three organizations had just over 20 percent of their total assets in net nonbank assets; seven organizations had more than 10 percent; and eight had more than 5 percent. Thus, only eighteen of the fifty largest bank holding companies in terms of total assets held a significant percentage of their total assets in nonbank activities. Many very large banking organizations had relatively small amounts of nonbank assets. By contrast, when all reporting organizations were ranked by net nonbank assets, thirty-nine organizations had more than 5 percent of their total assets in nonbank activities. Nonbank Activities by Type Table 3 shows the distribution of aggregate nonbank assets among eleven categories of nonbank activities. The aggregate of nonbank assets is defined as the sum of total nonbank assets across the eleven types of nonbank subsidiaries, including balances due from other types of nonbank subsidiaries and from the parent bank holding company and its affiliated banks. Aggregate nonbank assets in 1989 totaled $209.8 billion. Thus, based on consolidated nonbank assets of $175.5 billion, balances due from other types of nonbank subsidiaries amounted to $34.3 billion in 1988.

A large percentage of nonbank assets is invested in traditional banking activities that can be conducted within the bank. As the data in table 3 indicate, subsidiaries engaged principally in commercial finance, mortgage banking, consumer finance, and leasing account for about 47 percent of aggregate nonbank assets. If the "other depository institution" category is included, traditional banking activities account for 56 percent of aggregate nonbank assets.

The remainder of nonbank assets is accounted for primarily by securities brokerage (15.4 percent) and other nonbank (26.5 percent) subsidiaries. Subsidiaries classified under securities brokerage include those engaged principally in discount brokerage and permissible securities underwriting activities. Other nonbank subsidiaries are engaged mainly in the following activities: trust services; provision of general economic and financial information and advice; sponsorship, organization, or control of a closed-end investment company; investment to promote community welfare; courier services; management consulting; sale of money orders, travelers' checks, or savings bonds; personal property and real estate services; foreign exchange services; and acting as futures commission merchants. In addition, this category includes nonbank activities conducted before the 1970 Amendments that are still being conducted under grandfather rights. Finally, one major bank holding company finances all of its subsidiaries through a funding subsidiary; the funding subsidiary is included under other nonbank for purposes of this reporting system.

The data in table 3 also indicate that the other nonbank category is the one reported by the largest number of bank holding companies, reflecting in part their continuation of nonbank activities conducted before the 1970 restrictions. Large numbers of bank holding companies also report subsidiaries engaged in credit-related insurance underwriting, mortgage banking, and leasing.

Total nonbank activities have grown 7.4 percent since 1987, largely because of the growth in securities brokerage and commercial finance. The assets of securities brokerage subsidiaries increased 54 percent, from $21.0 billion in 1987 to $32.3 billion in 1988. This increase resulted primarily from the shifting of permissible government securities underwriting activities previously conducted in banks to the securities subsidiaries. The number of firms engaged in this activity, however, fell from ninety-nine in 1987 to eighty-three in 1988, reflecting the sale of brokerage subsidiaries by many bank holding companies. Assets in small business investment companies and leasing also increased; in contrast, assets in insurance agency subsidiaries fell 29 percent, and assets in data processing subsidiaries fell 9 percent.

Finally, table 3 also shows the percentage of assets in nonbank activities held by the top five firms in each category. In nearly all of the nonbank activity categories, but particularly securities brokerage and other depository institutions, the expected levels of concentration appear in that the largest five firms engaged in each activity held the vast bulk of the total assets of those bank holding company subsidiaries engaged in that activity. This result is consistent with the concentration of nonbank assets revealed by the previous analysis of the consolidated nonbank data. Only twenty-seven bank holding companies had nonbank subsidiaries that ranked among the five largest in one or more activity categories, and only five bank holding companies had nonbank subsidiaries among the five largest in more than three types of activities. PROFITABILITY OF NONBANK ACTIVITIES Data for 1986 through 1988 suggest that, in the aggregate, nonbank activities were relatively profitable. In 1986 and 1987, profit rates for nonbank activities exceeded profit rates for the affiliated bank subsidiaries and the consolidated bank holding company; in 1988, however, when bank profits were not under pressure from loan loss provisions for loans to developing countries, nonbank profits were lower than bank profits. Profit rates varied widely across nonbank activities and bank holding companies. The wide variation reflected in part the different mixes of nonbank activities engaged in by the firms, different formulas for the allocation of joint costs, and different management abilities. Consolidated Nonbank Activities The ratios of net income to assets and net income to equity are the commonly used measures of the profitability of banking organizations and are shown in table 4 for the consolidated nonbank subsidiaries of firms reporting on the FRY11Q. The average of the ratios of nonbank net income to assets was 0.69 percent and of nonbank net income to equity was 6.24 percent over the 1986-88 period. The data also show some variability in nonbank profits: The ratio of nonbank net income to assets was 1. 19 percent in 1986, 0.34 percent in 1987, and 0.61 percent in 1988. The ratios of nonbank net income to equity have shown similar variability.

For comparison, table 4 also presents the ratios of net income to assets and net income to equity for the affiliated bank subsidiaries and the consolidated holding companies. Based on the 1986-88 average ratio of net income to assets, nonbank subsidiaries appear to be fairly profitable relative to the affiliated bank subsidiaries and the consolidated bank holding company. Over the 1986-88 period, the average ratio of net income to assets was 0.69 percent for the consolidated nonbank subsidiaries compared with 0.38 percent for the affiliated bank subsidiaries and 0.39 percent for the consolidated bank holding company. Based on the 1986-88 average ratio of net income to equity, however, nonbank subsidiaries appear to be less profitable. The average ratio of net income to equity was 6.24 percent for the consolidated nonbank subsidiaries compared with 7.39 percent for the affiliated bank subsidiaries and 6.97 percent for the consolidated bank holding company. The differences between the ratios of net income to assets and of net income to equity reflect the higher equity capitalization of nonbank subsidiaries relative to bank subsidiaries.

Examining the two profit rates-the ratios of net income to assets and net income to equity capital-over the various years, the profit rates of the nonbank subsidiaries exceeded those for the affiliated bank subsidiaries in both 1986 and 1987; the profit rates for the nonbank subsidiaries in 1988, however, were lower than those of the bank subsidiaries as bank profits increased significantly from their 1987 lows. The relationship between bank profits and nonbank profits is also of interest. The data show that the profits of nonbank subsidiaries fell in 1987 from 1986 levels, and then rose in 1988, thus exhibiting the same pattern as bank profits. Given that the fall in bank profits in 1987 was attributed in large part to loan write-off provisions for developing-country debt, it is not clear why profits of nonbank subsidiaries also fell. Although only a few years' data are available for analysis, this pattern of profits raises questions about the potential gains from diversification resulting from currently allowed nonbank activities.

The profit rates for nonbank activities are also calculated separately for the top fifty firms based on consolidated bank and nonbank assets and for the remaining firms. These profit rates, shown in table 5, suggest that the profits of the top fifty firms are similar to rates for the entire sample of firms, as reported in table 4, because the top fifty firms account for 75 percent of total assets of the reporting bank holding companies and for 94 percent of the total nonbank assets reported. The data further suggest that the relationship between bank and nonbank profits for the largest fifty firms differs from the comparable relationship for the other firms. In particular, when ratios of nonbank net income to assets for the fifty largest firms fell substantially in 1987 to 0.26 from 1.15 in 1986, this ratio for the other firms fell only to 1.13 percent from 1.51 percent. In addition, the bank holding company ratio of net income to assets of the fifty largest firms fell in 1987 to -0.39 percent because of large loan write-off provisions by the bank subsidiaries, whereas this ratio for the other firms remained relatively high, at 0.44 percent. Nonbank Profits by Type of Activity Profit rates for the eleven categories of nonbank activities vary widely. Table 6 presents the median ratios of net income to assets and net income to equity in 1988 for the bank holding company subsidiaries engaged in the eleven nonbank activities, along with the profit rates at the first quartile (25th percentile) and at the third quartile (75th percentile). The table also reports 1986-88 median ratios of net income to assets and of net income to equity for the bank holding company nonbank subsidiaries. For the most part, activities that were relatively profitable in 1988 were relatively profitable over the entire period. Overall, securities brokerage firms were the least profitable and insurance agencies were the most profitable nonbank activities. The median securities brokerage firm earned 0 percent on assets and equity, and the median insurance agency earned 8.85 percent on assets and 20.4 percent on equity.

The variation in profit rates of the nonbank subsidiaries may reflect more than just differences in performance across activity categories. Profit rates may differ across activities based on whether the subsidiaries hold assets (such as commercial loans) or provide services (such as insurance). Profit rates may also vary because of differences among activities in their equity capitalization ratios. Although somewhat lower than bank profits in 1988, profit rates for the nonbank activities most like traditional bank activities--commercial finance, mortgage banking, consumer finance, and leasing-are similar to profit rates for banks. Collectively, the ratio of net income to assets for these activities was 0.68 percent in 1988; with other depository institutions included, the ratio was 0.65 percent. The ratio of net income to assets for the affiliated bank subsidiaries (shown in table 4) was 0.79 percent in 1988.

The median profit rates for the nonbank subsidiaries in 1988 are similar to the 1986-88 rates, except for the securities brokerage firms and small business investment companies. Based on data for three years, these subsidiaries appear to have the highest variability in income among the nonbank subsidiaries. Comparing 1986-88 median ratios across activities, small business investment companies were the least profitable, and insurance agencies were the most profitable. (3) SOME RISK MEASURES OF NONBANK ACTIVITIES The overall riskiness of the bank holding company is affected by the riskiness of its nonbank activities. The most frequently used measure of risk, the standard deviation of profits for the individual nonbank subsidiaries, cannot be used to assess risk in this case because data are available for only three years. As an alternative, some indirect measures of risk-the equity capitalization of the nonbank subsidiaries and the riskiness of loans made by them-provide some indication of the riskiness of nonbank activities. In addition, a measure of the probability of insolvency can be calculated assuming that the cross-section distribution of net income to assets is representative of the distribution of net income to assets for an average firm. Ratios of Equity Capital to Assets The equity capital-to-assets ratio is often used as an indicator of risk in banks and bank holding companies because high levels of capital provide protection against large declines in income. Thus, better capitalized organizations will, other things equal, incur less risk of insolvency because of loan losses, lower revenues, or higher costs. The use of this ratio as an indicator of risk across different types of organizations, however, is not straightforward. For example, differences in capital-to-assets ratios may simply reflect different ways in which firms in various activities are typically funded. Thus, higher capital-to-assets ratios may indicate activities with higher risk rather than lower risk because firms engaging in activities with high expected profits and high variability of profits need high levels of capital to protect against insolvency.

The data in table 7 indicate that nonbank subsidiaries appear to be better capitalized than affiliated bank subsidiaries are. Specifically, the capital-to-assets ratio for the consolidated nonbank subsidiaries, as reported on the FR-Y11Q forms, is 10.83 percent for 1988. In contrast, the capital-to-assets ratio for the 284 consolidated bank holding companies is only 5.84 percent, and for the aggregated bank subsidiaries, it is 5.44 percent. These ratios are similar to those for the 1986-88 period.

The capital-to-assets ratios shown in table 7 vary widely across the different types of nonbank subsidiaries. Mortgage banking, securities brokerage, and leasing subsidiaries have relatively low capital-to-assets ratios of about 5 to 7 percent. The high ratio for other depository institutions reflects the large amount of capital kept in the savings and loan associations owned by the single bank holding company that accounts for 90 percent of the assets in this category. Small business investment and insurance underwriting subsidiaries have very high capital-to-assets ratios of 67.7 and 50.5 percent respectively, which are more similar to norms in those industries than to bank ratios.

Capital-to-assets ratios are commonly used as indicators of risk. However, in the case of holding company subsidiaries, these ratios reflect any double leveraging practices of the parent bank holding company, whereby the holding company issues debt and invests the funds in the nonbank subsidiaries as equity. However, if capital-to-assets ratios are reliable indicators of risk in the comparison of holding company subsidiaries engaged in different activities, nonbank subsidiaries appear less risky than the affiliated bank subsidiaries. Further, rather than reflecting just risk and double leveraging, these differences in capital-to-assets ratios may reflect differences in typical funding patterns for firms engaged in various activities. Indeed, those subsidiaries engaged in so-called traditional banking activities-commercial finance, mortgage banking, consumer finance, and leasing-collectively have a capitalto-assets ratio of 8.0 percent, similar to capital-to-assets ratios of banks, although still somewhat higher. Measures of Loan Risk Given the difficulties associated with the use of capital-to-asset ratios as measures of risk, loan risk measures were also calculated for those nonbank subsidiaries that hold a large portion of their assets in loans. The loan-to-asset ratios in 1988 were 88.0 percent for the fifty-eight commercial finance subsidiaries; 58.4 percent for the one hundred ten mortgage banking subsidiaries; 89.1 percent for the fifty-one consumer finance subsidiaries; and 83.7 percent for the ninety-six leasing subsidiaries.

Net charge-off rates (charge-offs less recoveries as a percentage of total loans) and past due rates (loans past due 90 days or more and nonaccruing loans as a percentage of total loans) are calculated for the nonbank subsidiaries engaged in commercial finance, mortgage banking, consumer finance, and leasing that reported values of more than zero for loans and leases. The net charge-off and past-due rates for loans made by commercial finance subsidiaries are compared with net charge-off and past-due rates on commercial and industrial loans made by the bank subsidiaries of the same bank holding company. Comparable ratios for mortgage banking subsidiaries are compared with rates on the affiliated commercial banks' loans secured by real estate. The rates for consumer finance subsidiaries are compared with rates on commercial bank loans to individuals for household, family, and other personal expenditures, and the comparable rates for leasing subsidiaries are compared with rates on commercial bank lease financing receivables. Table 8 shows these rates.

As the data indicate, the 1988 net charge-off rates on loans made by nonbank subsidiaries are higher than the net charge-off rates on similar loans made by the affiliated bank subsidiaries, except for commercial finance subsidiaries. These net charge-off rates suggest that loans and leases made by the consumer finance, mortgage banking, and leasing subsidiaries are riskier than similar loans and leases made by the affiliated bank subsidiaries. The past-due rates suggest the same relationship. Specifically, the past-due rates are higher for the consumer finance, mortgage banking, and leasing nonbank subsidiaries and lower for the commercial finance subsidiaries than for the affiliated bank subsidiaries, although these results tend to be heavily influenced by the very poor performance of a few firms. Overall, loans made by mortgage banking, consumer finance, and leasing subsidiaries appear to be riskier than similar loans made by the affiliated bank subsidiaries. In contrast, commercial finance loans made by the bank subsidiaries appear to be riskier than those made by the nonbank subsidiaries. Probability of Insolvency Finally, the risk of nonbank subsidiaries can be measured by an indicator of the probability of insolvency. This conceptual measure assumes an independent firm, whereas in a holding company a parent organization would likely assist a subsidiary by providing additional capital. The measure is derived from the probability that income losses will exhaust capital and assumes that income returns are normally distributed. Specifically, risk can be measured by

g = [E(P/A) + C/A]/s. where E(P/A) = the expected ratio of net income to

assets C/A = the ratio of capital to assets s = the standard deviation of net income to

assets. The standard deviation of net income to assets is calculated based on the cross-section distribution of net income to assets, assuming that it is representative of the time-series distribution for an average firm. The g ratio is appealing because it incorporates the three financial variables most closely associated with the financial health of a firm. Those firms with higher C/A, higher E(P/A), and lower s will have a higher g measure and lower overall insolvency risk. As with the loan risk measures reported above, the g measure of risk is constructed for subsidiaries engaged in commercial finance, mortgage banking, consumer finance, and leasing. Only subsidiaries that were engaged in these activities in both 1987 and 1988 are included in the sample for purposes of calculating g so that start-up and exiting firms are not included when calculating the standard deviation of the net-income-to-assets ratio. These measures are then compared with g measures constructed for the bank subsidiaries of those holding companies with nonbank subsidiaries that make commercial and industrial loans, loans secured by real estate, loans to individuals, and leases. Table 9 shows these g measures. The calculated g ratios indicate that the commercial finance, mortgage banking, and leasing nonbank subsidiaries are riskier than their affiliated bank subsidiaries. For these three types of activities, the g ratios are lower (suggesting higher risk) for the nonbank subsidiaries than for the affiliated bank subsidiaries. In the case of consumer finance, bank subsidiaries may be more risky although the difference in the g ratios is small. These relationships are particularly interesting because these nonbank subsidiaries have higher capital-to-assets ratios and higher net-income-to-assets ratios (except for mortgage banking subsidiaries) than the affiliated bank subsidiaries do, which, other things equal, would suggest lower risk. However, the nonbank subsidiaries have a much higher standard deviation of net income to assets; the higher standard deviation offsets the risk-reducing effects of higher capital and higher profits. CONCLUSION To date, the new data provided by the FR-Y11Q and FR-Y11AS on nonbank activities of bank holding companies are only available for three years. Nevertheless, the available data indicate that some of the nation's largest banking organizations have large holdings of nonbank assets. Furthermore, the overall share of holding company assets devoted to nonbank activities has increased significantly since data collection began in 1986.

Available evidence on nonbank activities of bank holding companies suggests that nonbank subsidiaries are more profitable than bank subsidiaries, but nonbank profits have moved in parallel with bank profits over the 1986-88 period. Furthermore, nonbank subsidiaries appear to be better capitalized than bank subsidiaries, although this may reflect double leveraging. However, various measures of risk involving loans or insolvency appear to indicate that nonbank subsidiaries are riskier than bank subsidiaries and therefore may be more than a device to circumvent restrictions on geographic expansion. Thus, the growth in the relative share of nonbank activities could have important implications for the safety and soundness of banking organizations.

The future share of nonbank assets within the consolidated bank holding company is difficult to predict. As barriers to intrastate and interstate banking are reduced, lending activities conducted in nonbank subsidiaries may be moved to bank subsidiaries, thus reducing the amount of activities conducted in nonbank subsidiaries. On the other hand, any differences in the nature of the lending done by bank and nonbank subsidiaries may justify their continued existence and growth. Finally, bank holding companies may wish to expand their nonbank activities to include those that are even more removed from traditional banking activities. *

1. According to Klebaner, examples of bank holding company ownership of large nonbank enterprises were the holdings of Transamerica, including extensive insurance interests, a tuna and salmon canning plant, and a manufacturer of metal products; and the Trust Company of Georgia's ownership of a soft drink bottling company. See Benjamin J. Klebaner, "The Bank Holding Company Act of 1956," Southern Economic Journal, vol. 24 (January 1958), pp. 313-26.

2. General histories of the bank holding company movement include Gerald C. Fischer, Bank Holding Companies (Columbia University Press, 1961); Gerald C. Fischer, Robert A. Eisenbeis, and Joseph F. Sinkey, The Modern Bank Holding Company: Development, Regulation and Performance (Philadelphia: School of Business and Management, Temple University, 1986); Michael A. Jesse and Steven A. Seelig, Bank Holding Companies and the Public Interest: An Economic Analysis (Lexington Books, 1977); and Donald T. Savage, "A History of the Bank Holding Company Movement, 1900-78," in The Bank Holding Company Movement to 1978: A Compendium (Board of Governors of the Federal Reserve System, 1978), pp. 21-68.

3. The median profit rates shown in table 6 may differ substantially from weighted average profit rates because of the relatively small numbers of subsidiaries engaged in some nonbank activities, extreme performances by a few large subsidiaries, and because net income, which is a flow measure, is measured against assets or equity, which are stock measures. For comparison, the weighted average net income-to-assets ratios in 1988 were (in percent) the following: 1.43 for commercial finance, -0.71 for mortgage banking, 0.78 for consumer finance, -0.62 for securities brokerage, 0.51 for other depository institutions, 1.60 for leasing, -4.29 for data processing, 8.10 for insurance underwriting, 8.93 for small business investment companies, 6.71 for insurance agencies, and 1. 12 for other nonbank activities. In several cases, the relative profitability of the activities depends on whether profits are measured by the weighted average or median, suggesting that the profit data for the different types of nonbank subsidiaries should be used carefully.

NOTE. This article is an update of an earlier study by the same authors, New Data on the Performance of Nonbank Subsidiaries of Bank Holding Companies, Staff Studies 159 (Board of Governors of the Federal Reserve System, February 1990). Many of the more technical footnotes and data references in the earlier paper are omitted here. Interested readers can obtain a copy of the earlier paper from the authors.
COPYRIGHT 1990 Board of Governors of the Federal Reserve System
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1990, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Author:Savage, Donald T.
Publication:Federal Reserve Bulletin
Date:May 1, 1990
Words:6680
Previous Article:U.S. international transactions in 1989.
Next Article:The Federal Reserve in the payments system.
Topics:


Related Articles
New data on the performance of nonbank subsidiaries of bank holding companies.
Statements to the Congress.
Bank holding company investments for community development.
Statement by Susan M. Phillips, Member, Board of Governors of the Federal Reserve System, before the Subcommittee on Financial Institutions and...
Report on the condition of the U.S. banking industry: second quarter, 2004.
Report on the condition of the U.S. banking industry: fourth quarter, 2003.
Report on the condition of the U.S. banking industry: first quarter, 2004.
Report on the condition of the U.S. banking industry: fourth quarter, 2004.
Report on the condition of the U.S. Banking Industry: Second Quarter, 2005.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters