visory authority. I also so the major types of taxes specife, on the questioonide, sore Vas provided for repening any NMA neous types of taxes that had bee so. Va as deductions on the Federal tax return bo whe at ao fi the designated exterones. 201 estimates for any taxes 201 shown as deductible taxes on the Federal return but induced in other reHTELT EIKOSE Berounts as operis expenses. These latter would mente reber, sties Taxes, selective taxes on unity billings, and other known taxes paid in the price of goods or services purchased by the reporting bank.

For each category of taxes, respondents were asked to show sez arately the amounts paid within the home State of the bank and to all other States, identifying these States of this mommation was really available. In addition, where home State taxes had been significantly affected by changes in tax law a reguand since 1999, respondents were asked to estimate for each afected category what their 1909 taxes would have been had the change been effective throughout that year, To provide checks on the basis and comparability of reporting, respondents also were asked to mürite whether the income statement for 1969 had been prepared primam on a cash or accrual basis and whether or not a parent holding company had incurred tax habilides for 1969 in addition to those shown in the report, and to provide selected information relating to their 1969 Federal income tax returns.

In the course of its formulation, the reporting form was subjected to extensive review by the technical staffs of the Board of Governors of the Federal Reserve System, the Federal Reserve Banks, Federal Deposit Insurance Corporation, and the Comptroller of the Currency, as well as the Conference of State Bank Supervisors, the American Bankers Association, a group of commercial bank tax officers and comptrollers, several State tax administrators, and a number of independent tax consultants.

The survey was directed to a probability sample of 2.250 insured commercial banks out of a universe numbering close to 13,500. The sample was designed to provide reliable estimates of the amounts of the principal types of State and local taxes paid in each State by national and by State-chartered banks separately. Selection of the sample was on the basis of probability proportionate to size, with size measured by the bank's gross operating income in 1969.

As in the banking universe, a little over one-third of the banks in the sample had national charters and the remainder were Statechartered. The sample of national banks accounted for about 75 percent of the gross operating income of all national banks in 1969 and that of State banks for nearly 65 percent of the gross operating income of all insured State banks.

Since the purpose of the survey was to determine tax expenses for the year 1969, the sample banks were selected from the bank universe as it existed on December 31, 1969. Sample banks that had merged or changed their status after that date were requested to supply data for the particular unit as it existed in 1969. In a few instances where a bank chosen in the initial sample was unable to supply the requested information, a substitute bank was selected.

The survey was conducted by mailed questionnaire distributed through the Federal Reserve Banks during the latter part of December 1970. Usable reports were received from a total of 2,222 banks, or about 99 percent of the total sample.

Reporting Difficulties and Limitations of the Data

Tax expenses shown in the tables represent in every case the total for all insured commercial banks, estimated on the basis of survey responses, except that amounts derived from column C of the questionnaire (tax changes) represent actual responses, without adjustment to magnitudes that would represent the entire commercial banking universe.

Statistics derived from a survey on a sample basis are necessarily subject to some margin of error. However, in view of the relatively large sample and the virtually complete response, the resulting estimates of the U.S. totals for major categories of taxes are believed to be highly reliable. Within individual States, where greater variability is inevitable, the estimates appear to fall within acceptable tolerances.

The data were collected from respondents through the regional Federal Reserve Banks, where they were reviewed and edited before being forwarded to the Board for computer processing. Since the pattern of response within most States was fairly consistent, major errors and abnormalities for the most part were discernible and necessary corrections were worked out through telephone communication with the respondent.

Although information requested in the survey generally was readily available at the banks, difficulties were encountered by some respondents in determining which taxes to report and in selecting the appropriate category in which to report them. One major problem area was the distinction between a tax on bank shares, which in most States is a local ad valorem tax on shareholders but is paid and absorbed by the bank, and a tax on capital structure, which generally is a direct tax levied by the State government on the bank itself and is measured by a specified concept of capital. In view of widespread failure of respondents to distinguish these categories, the reported payments have been combined for analytical purposes.

Problems in reporting also arose where the legality of a tax was in question and the respondent bank had paid the tax under protest and then filed a claim for refund. Except where the refund had been received, the tax was considered as having been paid and the respondent was asked to report it. In other cases involving taxes of questionable applicability, respondents encountered no particular difficulty but a mixed reporting pattern emerged, since some banks had paid the tax while others had not.

Another part of the questionnaire which was troublesome for some respondents was column C, which called for reporting revised 1969 estimates for any taxes which had been subject to change by law or regulation since the December 1969 amendment to section 5219. In recognition of the difficulties that smaller banks, especially, might encounter in preparing these estimates, banks with total deposits of less than $15 million were advised that they need not fill in this part of the questionnaire. Some of them chose to respond to it, and their replies are included in the statistics.

Taxes whose legality was being questioned also gave rise to reporting problems in this part of the survey. One type of situation involved sales and use taxes that had been collected from national banks without statutory authority for all or part of 1969 and for which the bank

had filed for refund. Where a refund had been received, respondents were requested to report no 1969 tax in column A but to show in column C what the tax would have been if legally applicable. Where the tax had been paid in 1969 but no refund had been requested or received, the amount paid was to be reported in column A and no entry shown in column C. A similar problem given parallel treatment arose in one large city, where the authority of the city to collect a vault tax and a business occupancy tax was contested by some banks, which refused to pay the tax while others paid without protest. Patterns of Bank Taxation

Total State and local tax expense for all insured commercial banks in 1969 is estimated at $623 million, as shown in table 1.5 This total includes all separately identifiable State and local taxes of banks and their domestic subsidiaries for that year except State payroll taxes for financing unemployment compensation and disability insurance.

TABLE 1.-STATE AND LOCAL TAX EXPENSES OF ALL INSURED COMMERCIAL BANKS, 1969

General sales and use taxes (including amounts reported under miscellaneous and other)....

Taxable income for Federal income tax purposes.

Federal income tax...

1 Of this total, $8,274 thousand was reported by banks in Indiana and consisted almost entirely of a combination of that State's taxes based on deposits and share values (net of gross income tax credit for State banks).

2 Taxes not shown as deductible taxes on Federal income tax returns but included in the relevant category of other expenses, such as selective taxes on utility billings or known taxes included in the cost of purchased goods and services.

To some extent, the reported data probably understate the total State and local tax expense associated with domestic commercial banking activity for calendar 1969.6 Two possible sources of such understatement are suggested by the responses.

(1) Survey data show that over 800 banks were affiliated with a parent company that incurred tax liabilities on 1969 operations in

For the distribution of this total by States and charter-classes of banks, see table A at the end of this appendix. A detailed breakdown of each State total by type of tax and class of bank appears in table G below, pp. 56-72. 6 See below, p. 36, footnote 12, for a possible source of slight overstatement of taxes on domestic operations.

addition to those reported on the survey for the bank itself (see table B). In many cases, these taxes were associated with activities closely interrelated with those of the subsidiary bank or banks, particularly the tax liabilities incurred by most registered bank holding companies and the larger one-bank holding companies that recently had been created. However, the amounts of such bank-related taxes generally were small compared with the total of State and local taxes paid by the subsidiary banks.

(2) Survey data also show that the income statements which provided the base for derivation of the tax expenses reported on the survey had been prepared mainly on a cash rather than accrual basis by a substantial proportion of smaller banks-85 percent in the case of banks with total deposits under $15 million and 43 percent for those with deposits of $15-$100 million (table C, p. 48). With cash accounting, tax expenses would reflect a lag between tax accruals and payments, and given the uptrend in earnings, property values and other bases for taxation prevailing at that time, reported tax expense for 1969 would be less than the amount of tax accruing on 1969 operations. This may have been true in part even for net income taxes, which were especially affected in 1969 by large capital losses. Although all banks were asked to report such taxes for 1969 on a current basis, this was a transition year for a major revision in requirements for income reporting to Federal supervisory authorities, and it is possible that some banks were not able to complete the changeover.

The distribution of bank tax expense by type of tax is heavily concentrated, with taxes on net income, real property, and bank shares accounting for 86 percent of the total. The most important of these is the tax on net income which totaled $235 million in 1969 and accounted for 38 percent of the total. But taxes on real property and on shares or capital structure also were important; they amounted to $179 million and $121 million, respectively, or 29 and 19 percent of the total. Nearly half of the $88 million remainder was in the tax on bank deposits; the rest was distributed mainly among five categories-sales and use, gross receipts, tangible personal property, miscellaneous deductible, and other.

Under "miscellaneous," respondents were requested to report identifiable amounts of taxes that were deductible on line 17 of IRS Form 1120 in calculating Federal income tax and that did not fit into any of the other categories on the survey reporting form. About twothirds of the amounts reported in this category consisted of the combination deposits and shares tax levied in Indiana. Small amounts of selective sales and use, occupancy, and license or privilege taxes also were reported in this category.

Under "other taxes," respondents were asked to report any significant amounts of taxes that were not shown as deductible taxes in

7 Most banks in the $15-$100 million deposit group that reported mainly on a cash basis probably fell in the lower part of this size range, since every bank with total resources of $50 million or more at the end of 1968 was required to prepare its 1969 “Consolidated Report of Income" on the basis of accrual accounting. However, certain exceptions were permitted-for reporting trust department income and reporting particular accounts where the results would not be significantly different.

The estimates of net income taxes derived from the bank tax expense survey ($125.6 million for national banks and $109.5 million for State-chartered banks or a total of $235.2 million) are somewhat higher than the "provision for State and local income taxes" reported by banks on their "Consolidated Report of Income" ($116.0 million for national banks and $101.8 million for State-chartered banks, or a total of $217.8 million). See Annual Report of the Federal Deposit Insurance Corporation 1969, p. 280. These differences probably reflect mainly the reporting of final data on the tax survey, which was conducted late in 1970 after tax returns for 1969 had been filed, whereas the amounts of taxes reported on income statements, which had to be submitted early in the year, in many cases were preliminary estimates.

calculating Federal income tax liability but were included with other expenses. Nearly all the amounts reported were sales and use taxes. If sales and use taxes reported in the "miscellaneous" and "other" categories are combined with those separately reported, the total of such taxes rises to $23.6 million or nearly 4 percent of total bank tax expense. Excluding sales and use taxes, the combined total of residual categories falls to 2 percent of all taxes, instead of 4 percent.

The concentration of bank taxes in the net income, real property, and shares or capital structure categories was evident in all sizeclasses of banks (table 2). However, the relative shares accounted for by each tax category at the largest banks (total deposits of $500 million and over) differed markedly from the relative shares of these taxes at banks in the smaller size groups. The tax on net income was much more important for the largest banks than for others, while the shares tax was much less important.

TABLE 2. PERCENTAGE DISTRIBUTION OF 1969 STATE AND LOCAL TAX EXPENSES OF ALL INSURED COMMERCIAL BANKS, BY MEASURE OF TAX AND SIZE OF BANK

To a considerable extent, these differences reflect the locational distribution of large banks ielative to the varying tax structures of the States. The largest banks tend to be concentrated in States placing heavy reliance on a net income tax, particularly New York and California. These two States alone accounted for nearly three-fifths of the $235 million total of bank income taxes reported for the country as a whole. States placing heavy reliance on shares taxes, on the other hand, tend to be in nonindustrial areas, where relatively small banks predominate. An additional factor contributing to lower shares tax ratios for large banks is a tendency for high-value properties to be assessed at lower ratios of value than those of lesser value. While large banks also operate with thinner capital cushions than smaller banks, the effect of this difference would be slight.

Another significant tax structure difference related to bank size is the smaller importance of certain minor taxes at large than at smaller banks. The main taxes showing this difference were those on tangible personal property, gross receipts, and documents-types of taxes not authorized by section 5219 for application to national banks before December 1969. Since national banks account for the bulk of the resources of all banks in the largest size groups, their exemption from these particular types of taxes would tend to lower the ratios.