Second, concerning broker-dealers to which factors 2, 3, and 4, above, are applicable the following formula is used: For a specific period of time we compute for each broker-dealer (a) 100 percent of the analysis profit on his deposit balances, (b) 50 percent of the interest paid by him on loans, and (c) 10 percent of his rental payments. The sum of these three figures forms the dealer's base. We then allocate the orders during the ensuing period among the broker-dealers in proportion to their respective bases.

Non-exchange transactions in bonds, unlisted securities, and certain listed stocks traded "off the board" are executed on a net price basis. Such transactions are handled separately from our commission business and are awarded on the basis of best price and execution without reference to whether the broker-dealer is among those mentioned above. In the event of a tie bid or offering, however, the business is awarded under the same criteria to a brokerdealer as described above.

Some of the responses in this area pointed out that for over-thecounter trades of unlisted as well as listed stocks, the institution tends to go directly to the dealers making the markets for the stocks. The comment of one bank was:

The [order department] has authority to select broker-dealers and types of execution through normal market channels. In the case of an unlisted security, the [order department] makes inquiries of various dealers known to maintain a trading market in that security, and enters into the transaction with the dealer offering the best price. In the case of listed securities, the [order department] executes orders in the over-the-counter market or on an exchange, whichever is found to be more favorable to the Trust. In the case of orders executed on an exchange, the [order department] places the order with a broker/dealer selected from a list of member firms approved by the bank's [brokers' relationship department]. This list (referred to hereafter as "Approved Names") is compiled on the basis of such factors as overall relationships with the bank, credit standing, competence, and the extent to which investment suggestions developed by research staffs of the broker/dealers are furnished to the bank. Approved Names at the present time are over 340 in number and names are added or deleted from time to time in accordance with the above-mentioned criteria for approval. There is, of course, no difference to the Trust in the commissions charged by the respective brokers/dealers. No Approved Names are assigned by the bank specifically to this Trust and normally orders are spread among the Approved Names. The bank's [brokers' relationship department] also makes checks as to the credit standing and competence of dealers with whom the bank. transacts business in connection with unlisted securities and listed securities where the trade is made over the counter.

Another bank stated:

*For a specific order or transaction, the criteria used in placing business have been the most efficient execution and, where applicable, the most expeditious execution at the most advantageous price. An example of the latter would be a broker-dealer making an active primary market in size for an unlisted issue.

There is no objection to execution of orders on regional exchanges or over the counter, provided the execution is at the same level as an execution on the New York Stock Exchange or better. For sale orders, the broker-dealer is asked to check other regional exchanges for simultaneous (same price) sales executions in order to save the expense of the New York State transfer tax on stocks which are transferable outside New York State. For listed securities which are also traded over the counter, the daily "pink sheets" of the National Daily Quotation Service are examined to determine if such stocks are traded over the counter.

Other institutions made similar comments. A university said:

Where the decision to purchase or sell is primarily the result of a strong recommendation of a broker, he receives all or most of the order, provided, of course, he is in a position to perform adequately.

Where a transaction is contemplated primarily as a result of internal efforts, our first concern is the greatest net advantage to the university. In the case of a listed security, this might result in the selection of an over-the-counter dealer

when he maintains a market. If the equity is unlisted, we go to a broker who maintains a market in this particular issue.

A foundation made the following comment:

The manner in which an order is executed will depend upon the marketability of the stock, the size of the order and in some cases is governed by a specific limit objective. Thus, both limit orders and market orders are used. In some cases the broker involved is shown the whole order and in some cases only part of the order at a time. The order department is under instructions to make the best possible execution and will, therefore, effect transactions in the over-thecounter market through the broker-dealer making a primary market. In addition, "off-board" markets are used occasionally where listed securities are traded on a net basis when this is advantageous to the foundation. Otherwise, the order department is directed to place brokerage with a listed group of brokers who provide research information and other services to the foundation.

Life insurance companies mentioned that they try to allocate business to those broker-dealers who, as agent for the issuer or as principal, bring them private placements of various types of securities or give them participations in underwritings. Purchase of insurance from the company did not appear to be a significant factor. For colleges, consideration of the "old school tie"-the interest of the broker in the college and the help he gives it-plus the providing of opportunities for private placements were the most important factors mentioned. Open-end (load) investment companies most frequently mentioned the sale by the broker of the fund's shares as a reason for allocating brokerage business. Other services listed were the provision of free wire services and of quotations to assist in computing the fund's net asset value. One fund described its policy for allocating commission business among broker-dealers selling the fund's shares as follows:

While there is no agreement to do so, it is the practice of [the fund], consistent with seeking the most favorable prices and execution of orders, to place orders to execute purchase or sale transactions in portfolio securities with competent broker-dealers who sell its shares or whom broker-dealers who sell such shares have requested receive portfolio orders. No specific proportion of brokerage business or of brokerage commissions is allocated to such broker-dealers. As a general practice, however, and subject to the availability of orders in the normal conduct of the [fund's] business and the competence of the brokerdealers, such broker-dealers receive orders having a principal value approximating the net asset value of shares of the [fund] which they have sold. The amounts of commissions on such orders vary, of course, with the prices of the securities involved, but over a period of time they tend to average about 1 percent of the principal value of transactions. Variations from this general rule may occur if broker-dealers are found not to be competent in the handling of orders or in the case of broker-dealers who are especially competent in performing the brokerage functions, who perform special services such as offering or making bids for blocks of stock, or who have shown by experience that they produce particularly good quality sales of the [fund's] shares. Dealers who are not capable of performing the brokerage function receive no orders to execute portfolio transactions but may receive selling group concessions upon such occasions when the [fund] buys securities in a registered public offering. * *

Another fund stated:

While the fund has made no commitment to any broker-dealer, it is the practice of the fund, where possible, while endeavoring to obtain the most favorable prices in the execution of orders, to place a part of its portfolio transactions with broker-dealers or to permit them to participate in commissions thereon, using their relative sales of shares of the fund or estimates of the value of their services as factors in the allocation of such portfolio business. Commissions and/or give-ups are allocated, in the case of reciprocal business, as evenly as is practicable among the broker-dealers in relation to their sales of fund shares. Currently, the reciprocal rate is slightly in excess of 1 percent. This may vary

considerably depending on portfolio activity at any particular time. Furthermore, it should be noted that certain broker-dealers who do business only in the over-the-counter market may not receive any reciprocal business except through special arrangements such as participating in the selling groups of underwritings or secondaries or where a stock exchange permits give-ups to nonmember firms. A third fund expressed its practice in this area as follows:

Distribution of portfolio commissions and business * * * in recognition of broker-dealer efforts to distribute investment company shares, is not arrived at according to any specific system, formula, or ratio. However, as a general rule of thumb, an effort is made to keep these commissions at a level of approximately 2 percent of the particular broker-dealer's gross sales of the fund's shares.

No-load funds mentioned, as reasons for allocating commission business, research information, the provision of quotations and wire service to assist in computing and publishing the fund's net asset value and offering and redemption prices, and ownership of stock in the fund. As for closed-end investment companies, the general factors of brokerdealer affiliation and the provision of research and analytical material tended to apply. Some of the closed-end companies, however, have given orders to broker-dealers which also sell shares of, or provide quotation services to, an affiliated open-end (load) company or companies.

The IN-4 respondents were also asked 101 to name the 20 brokerdealers to whom they paid the largest amounts of gross brokerage commissions on their portfolio transactions in common and preferred stocks for the year 1961, and to supply, for each such broker-dealer and for all of the other broker-dealers with whom they did commission business during 1961, the gross commissions paid to them by the respondent, the give-ups paid or received by them at the direction of the respondent, and the net commissions retained by them on the respondent's business. The analysis of these responses is contained in tables VIII-39 through VIII-41. The tables reveal the following:

1. The institutions in the IN-4 sample paid $41.7 million in gross commissions 102 during 1961 on their portfolio business in stocks. More than half of this was generated by the investment companies in the sample. The lion's share of the institutions' gross commissions (more than two-thirds) went to NYSE member firms.

2. For the institutions other than investment companies, gross commissions paid averaged only 0.1 percent of the average value of their stock portfolio for the year. For investment companies the figures were three to five times as big, reflecting these companies' higher

turnover rates.

3. For all institutions combined, give-ups paid averaged 8 percent of total gross commissions. The only significant users of the give-up device, however, were the open-end (load) companies and life insurance companies. For the open-end (load) companies, give-ups averaged 12 percent of gross commissions and for life insurance companies, 8 percent.

4. On the average, each open-end (load) fund gave commission business to 251 broker-dealers, a significantly higher number than used by the other institutions, where the average was close to 60. When the number of broker-dealers is related to the size of the institutions'

101 See question 8 of Questionnaire IN-4, app. VIII-A.

102 The figure for "commissions" excludes, as indicated by the wording of question 8 of Questionnaire IN-4, the cost of executing most principal transactions.

equity portfolios, however, a greater equality results. Thus, the openend (load) funds used 0.68 broker-dealers per million dollars of stock assets compared with around 0.43 for the other institutions.

5. The institutions, although using a large number of firms, tend to concentrate their brokerage executions among a relatively few. Thus, on the average, over 55 percent of an institution's gross commission business was given by it to its favorite 10 firms.103 To the extent the give-up device was used, it served to divert some of these commissions to other firms: on the average the individual institution's favorite 20 firms accounted for 88 percent of the give-ups paid, but only 20 percent of give-ups received.

5. PORTFOLIO TURNOVER

The IN-4 respondents were asked to report their total dollar purchases and sales of stocks during 1960 and 1961, and the market value of their stockholdings at the beginning and end of each of those years.104 From these figures the turnover of each institution's stock portfolio for each year was computed.105 The results of these computations appear in table VIII-42. For each institutional group and for each year, there are shown the weighted average turnover rate,106 the unweighted average turnover rate,107 the median turnover rate, 108 and the range between the lowest and highest turnover rates reported by institutions in each category.

The picture which emerges from the analysis is that the investment companies had strikingly higher turnover rates than any of the other institutional groups. The three investment company categories combined had weighted average, unweighted average, and median turnover rates, respectively, of 11.1, 15.6, and 11.8 percent for 1960, and 15.2, 17.1, and 13.1 percent for 1961. On the other hand, the comparable rates for all other institutions combined were 2.9, 4.1, and 2.3 percent for 1960 and 4, 5.8, and 3.5 percent for 1961.109

By way of comparison, the turnover rate on the New York Stock Exchange for all stocks listed there 110 was 12 percent in 1960 and 15

103 The favorite 10 of course varied from institution to institution; but even so, the 10 firms receiving the largest amounts of commission business from the total IN-4 sample received 40 percent of the total commissions.

104 Question 12 of Questionnaire IN-4, app. VIII-A.

105 The formula used for this purpose was the lesser of purchases or sales divided by the average market value (beginning value plus ending value divided by 2) of stockholdings for the year. Use of the lesser of purchases or sales as the numerator in the formula is based on the assumption that the excess of purchases over sales represents the investment of money newly devoted to stocks, not the turnover of stock investments; and that any excess of sales over purchases représents elimination of stock investment rather than turnover.

The same formula was used (along with two other more elaborate formulas) in the Wharton School report on mutual funds. "A Study of Mutual Funds," H. Rept. 2274, 87th Cong., 2d sess., pp. 230–234 (1962).

108 The weighted average turnover rates for institutional groups and for all groups combined were computed by weighting the rate for each fund by the average value of its stock portfolio for the year involved.

107 The unweighted average turnover rate is the average of the turnover rates reported by the individual funds in the category. 108 The median turnover rate is the median of the turnover rates reported by the individual funds in the category.

109 Figures for the year 1935 contained in the Commission's "Report on Investment Trusts" showed a similar differential in turnover rates between investment companies and common trust funds, the only other institutional group studied. The weighted average turnover rate for the total portfolios of the common trust funds for that year was only about half that for the investment companies. SEC. "Report on Investment Trusts and Investment Companies," pt. 2, ch. VIII, p. 633 (1939); id., "Commingled or Common Trust Funds Administered by Banks and Trust Companies," p. 19 (1939).

110 This was computed by dividing the total value of stocks sold on the NYSE by the average value of stocks listed on the NYSE during the year.

percent in 1961. While these rates are higher than the average turnover rate of the institutions in the IN-4 sample as a whole, the two sets of figures are not strictly comparable. Thus, the institutions' net purchases or net sales of stocks are excluded from consideration. in their turnover rates, but these purchases or sales, in addition to their turnover, contribute to the liquidity in the trading markets.

A further point which emerges from the analysis is the variation of turnover ratios within each of the institutional groups. Thus, the median turnover rate for the pension funds in 1960 was 1.8 percent, but one of the funds had a turnover rate of 22.9 percent, while all of the other respondents in the pension fund sample had rates 10 percent or under. In 1961, the median turnover rate for the pension fund group was 3.6 percent, but five of the individual funds had rates over 10 percent and three of these had rates over 18 percent. For the common trust funds, the median turnover rates were 3 and 5 percent for the 2 years; all of the common trust funds had turnover rates under, and most well under, 9 percent, except for two funds administered by a single bank which had turnover rates of 34.5 and 39.6 percent in 1960 and 24.9 and 26.4 percent in 1961. Among the investment companies, turnover rates for even the biggest of the open-end (load) funds varied from 4.3 percent for one fund to 30.9 percent for another.

6. RECOMMENDATIONS BY INSTITUTIONS ABOUT THE SECURITIES MARKETS

The institutions in the IN-4 survey were asked to state any suggestions they might have for changes in the practices, procedures, or structures of the various securities markets which in their opinion would make them better adapted to their use and needs.111 Fourteen of the institutions responded to this question, with comments covering a number of different subject matters. The most important of these comments are quoted in the various segments of the report to which they are most pertinent, and they are summarized only briefly here.

The most frequent suggestion was for a volume discount or lower commission rate for large blocks on the New York Stock Exchange; several of these comments are quoted in chapter VI.I. Two institutions (joined by a number of other persons, in correspondence or interviews) discussed the inadequacy of the present Exchange ticker system, one of them specifically pointing out that tape delays place individual and institutional investors at a "substantial disadvantage" compared with a specialist or floor trader. One institution commented on the "thinness" of the market for most stocks-attributed to institutional activity plus the unwillingness of investors to pay the capital gains tax on profits-and the resulting difficulty and expense of acquiring a substantial block on the Exchange. Another pointed out that some specialists were "not active enough" in their assigned stocks and questioned whether the Exchange's standards for judging a specialist's fulfillment of his responsibilities are sufficient; see chapter VI.D.

The same institution suggested that it might be desirable for the NASD to designate certain dealers as "specialists" in major over-the

111 Question 14 of Questionnaire IN-4, app. VIII-A.