....... ...H . a... ... . ... ......2... . .. _ . .1... ......... .. . . . , . .. Lyn? m mm... .H. . . h ...u ; y. M :1». ... ., .. . . .... . ...“.WWWRUWUW ..v . . . . i», . ._ .... . . ... .. x . 4.0.,” i «...... . . , . .. . w5..,....... . . T. . . . ... . .1 . H . .» . . I . . . a, £2. $3.0. ’ ABSTRACT THE INVESTMENT PERFORMANCE OF STATE RETIREMENT FUNDS BY Daniel Richard Vellenga Currently, state retirement systems have over seven and one-half million members and more than sixty billion dollars invested in debt and equity issues. How this money is invested is important to the funding of the pension plans. That is, as investment returns on the pension funds increase, the lower will be the necessary contributions into the fund to provide a given level of benefits; or, for the same level of contributions higher benefits can be paid to the retirees. This study was designed to measure the invest- ment performance of state retirement funds. It is signi- ficant in that it measured the rate of return on these funds by utilizing market values instead of book values. Previous studies and annual reports have generally shown the investment return on the portfolios as the percentage of investment income (dividends and/or interest) to the book value or average book value of the funds. Conse- quently, in periods of rising markets return is under- stated and in falling markets overstated. Daniel Richard Vellenga This study is also significant in that it utilized data from forty state retirement systems from twenty-nine different states. These forty funds held almost 60 percent of the total assets of all state retirement systems. In the past, studies of state pension funds have largely been qualitative in nature and limited to individual states or geographical regions. The investment analysis used the Dietz formula in calculating the rate of return on the funds over the years 1967 through 1972.1 The average overall rate of return on the forty funds over the six years was -.24 percent. The highest average overall return was achieved in 1971 with 10.63 percent and the lowest average overall return was experienced in 1970 with a -7.88 percent return. Further statistical analysis was performed with the following results: a) There was no statistical significance between the size of the funds and the rate of return earned on the funds. b) In 1967 there was a significant positive correlation between portfolio turnover and rate of return. In 1969 a significant nega- tive correlation existed between turnover and rate of return. For the remaining four years no significant relationship was found. Daniel Richard Vellenga c) The degree of liquidity of the funds did not have an appreciable effect upon the rates of return earned on the portfolios. d) There was a significant positive relationship between the percentage of equity held and portfolio rates of return in three of the six years--1967, 1968 and 1971. In the remaining three years there was no such relationship. e) No significant relationship between the rate of return and degree of portfolio diversi- fication (measured by the number of issues held) was found. f) There was a significant difference indicated between rate of return and type of management making the investment decisions. However, further analysis showed no consistent year to year pattern of one type of management out- performing the others. 1P. O. Dietz, Pension Funds: Measuring Invest- ment'Performance, The Free Press (Division of the Machllan Co.), New York, N.Y., 1966, pp. 50-51. THE INVESTMENT PERFORMANCE OF STATE RETIREMENT FUNDS BY Daniel Richard Vellenga A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Accounting and Financial Administration 1974 ©Copyright by DANIEL RICHARD VELLENGA 1974 ACKNOWLEDGMENTS I would like to thank my committee of Professors Alan E. Grunewald, Chairman, Thomas L. vwnufl< and Gary K. Stone for their suggestions while writing this dissertation. Also, thanks are due to Professors Alden C. Olson, my doctoral advisor, and Gardner M. Jones, Chairman, Department of Accounting and Financial Administration for their encouragement while completing this project. My sincere appreciation is given to David L. Mosconi, DiviSion Head,and Arthur Mason, Jr., Dean of the College of Business, at the University of Denver for their support and considerations given while writing. I would also like to thank my parents, my wife's parents, and all the relatives and friends who provided encouragement and help during this time. Finally, my deepest thanks are due to my wife, Cornelia, and children, Julie and Carl, for their patience, encouragement and consideration in completing the dissertation. ii LIST OF CHAPTER I. II. III. IV. TABLES . INTRODUCTI Purpose Need for TABLE OF CONTENTS ON 0 O O O O I O O O D O of Study . . . . . . . . Study . . . . . . . . . Data Availability . . . . . . . Sample S Time Per Organiza ize . . . . . . . . . . iod Covered . . . . . . tion of the Study . . . THE INVESTMENT POLICIES OF STATE RETIREME Dollars Investme Legal Re Investme Investme RATE OF RE NT SYSTEMS . . . . . . . to Invest . . . . . . . nt Objectives . . . . . strictions . . . . . . . nt Holdings . . . . . . nt Responsibility . . . TURN C O l O O O O O O 0 Measuring Performance . . . . . Current Rate of Overall Rate of Average Rate of Techniques . . . . . . . Return Measure Used . . Rate of Return . . . . . Return Ranges . . . . . Compound Rates of Return Return on Equity . . . . INVESTMENT ANALYSIS . . . . . . . Size of Correlation Between Size and Return Turnover Portfolio . . . . . . . Correlation Between Turnover and Return . . . . . . . . . . . . Liquidity . . . . . . . . . . . Page 11 13 15 16 17 17 18 21 27 30 36 36 37 41 43 48 51 55 62 62 63 66 69 71 CHAPTER Equity Holdings Equity Holdings Diversification Diversification and Rate and Rate Type of Management . V. SUMMARY AND CONCLUSIONS Investment Performance Analysis of Performance Availability of Data Future Areas of Research Conclusions . APPENDIX . . . . . . . BIBLIOGRAPHY . . . . . iv of Return of Return Page 75 77 80 83 84 89 89 93 95 97 99 105 115 LIST OF TABLES Table Page I—l. Cash and security holdings of state retirement systems for selected years (in billions of dollars) . . . . . 2 I-2. Dollar value (in billions) of all retirement systems in the United States as of 1972 . . . . . . . . . . . 11 I—3. States included in the sample to be studied . . . . . . . . . . . . . . . . l4 II-l. Dollars (in billions at market value) invested by the forty state retirement systems from 1967 to 1972 . . . . . . . . . . . . . . . . . . l7 II-Z. Investment objectives of selected state retirement systems by frequency . . . . . . . . . . . . . . . l9 II-3. Investment restrictions on state retirement funds by number of states . . . . . . . . . . . . . . . . . 24 II-4. Legal limits on the amount of stock which can be held by state funds . . . . 26 II—S. Types of investment holdings for state retirement systems for selected years in billions of dollars . . . . . . . . . . . . . . . . 29 II-6. Analysis of "Other Securities" category for state retirement funds for selected years in billions of dollars . . . . . . . . . . 31 II-7. Investment responsibility for investing state pension fund assets . . . . . . . . . . . . . . . . . 33 III—l. Methods of calculating rate of return on state retirement funds . . . . . . . . . . . . . . . . . 39 V Table Page III-2. Overall rate of return in percent in state retirement funds for years 1967 through 1972 . . . . . . . . 44 III-3. Average compound rates of return in percent earned on state retirement funds . . . . . . . . . . . . 52 III-4. Rate of return in percent on equity portion of state retirement funds . . . 56 IV-l. Fund size ranges and averages in $000's at the end of each fiscal year . . . . . . . . . . . . . . . . . . 63 IV—2. Ranges and average rates of return in percent earned on funds grouped in quartiles by dollar value . . . . . . 64 IV-3. Pearson correlation coefficients between size of fund and rate of return . . . . . . . . . . . . . . . . . 65 IV—4. Average turnover rates and ranges in percent for state pension funds . . . 68 IV—S. Pearson correlation coefficients between turnover rates and rate of return . . . . . . . . . . . . . . . . . 69 IV-6. Liquidity, in percent of state retirement funds 1967-72 . . . . . . . . 72 IV-7. Pearmmqcorrelationcoefficients between liquidity and rate of return . . . . . . . . . . . . . . . . . 74 IV—8. Market value in $000's of equity holdings of state retirement funds . . . . . . . . . . . . . . . . . 76 IV-9. Equity holdings in percent of state retirement systems . . . . . . . . . . . 77 IV-lO. Pearson correlation coefficients between equity holdings and rate of return . . . . . . . . . . . . . . . 78 IV—ll. Correlation between rate of return on state fund equity and market averages . . . . . . . . . . . . . . . . 79 vi Table Page IV-12. Diversification by number of issues held by state retirement funds . . . . . . . . . . . . . . . . . 82 IV-13. Correlation coefficients between rate of return and degree of diversification . . . . . . . . . . . . 83 IV-14. Ranking of average rate of return by type of management . . . . . . . . . 87 V-1. Average overall rates of return and equity rates of return in percent on state retirement funds . . . . . . . 90 A1. Legal restrictions on the investment of state pension funds . . . . . . . . . 105 vii CHAPTER I INTRODUCTION The growth in cash and security holdings of state retirement or pension systems over the past two decades has been dramatic. In 1952 the dollar value of these assets totaled $3.844 billion. As of June 30, 1973, these assets totaled $58.499 billion (see Table I-l). This fifteen-fold increase in assets held by the state retirement systems is due to a number of factors. First, the general population growth of the United States led to an increased demand for state government goods and services. As the states hired more employees to provide these goods and services, the dollar contribu- tions to the pension systems likewise increased. Secondly, the right of a guaranteed education through at least twelve grades coupled with the post World War II baby boom, led to a rapid increase in the numbers of school employees and teachers. Since a large number of school employees and teachers are covered by state pension systems, the contributions toward their pension plans helped swell the dollar value of these pension fund assets. 2 Table I-l. Cash and security holdings of state retire- ment systems for selected years (in billions of dollars). Year Cash and Security Holdings Pgfigigg 1952 $ 3.844 - 1957 _ 8.050 109 1962 15.546 93 1967 27.666 78 1972 51.158 85 1973 58.499 14 Sources: U.S. Bureau of the Census, Employee Retirement Systems of State and Local Governments, Vol. IV, No. l of the 1957 Census of Governments. U.S. Bureau of the Census, Emplgyee Retirement Systems of State and Local Governments, Vol. VI, No. l of the 1962 Census of Governments. U.S. Bureau of the Census, Employee Retirement Systems of State and Local Governments, Vol. 6, No. 2 of the 1967 Census of Governments. U.S. Bureau of the Census, Emplgyee Retirement Systems of State and Local Governments, Vol. 6, No. l of the 1972 Census of Governments. U.S. Bureau of the Census, Finances of Emplgyee- Retirement Systems of State and Local Governments in 1972-73, March, 1974. 3 Third, over time, salaries and wages tend to rise in the United States economy. This rise may be due to such factors as seniority, collective bargaining, merit raises or an attempt to keep salaries in line with a rising cost-of-living. The majority of states gear contributions to the pension funds as a percentage of employee earnings. Consequently, as salaries and wages rise in an attempt to keep up with inflation or other reasons, the dollar contributions also increase. Fourth, through the process of incorporation or merger of systems and centralization of government ser- vices many former city and local government employees are now covered by state retirement systems. The states have taken over the financing of these former local sys- tems and have absorbed the financial assets of these smaller units. Finally, demands for increased retirement benefits by state employees mean that additional contributions must be made by the states and/or employees in order to finance these higher levels of benefit payments. Table I-l also shows the interesting fact that for each five-year period between 1952 and 1972 the assets of the retirement systems almost doubled. Purpose of Study This study is designed to measure and test the investment performance of state retirement systems. 4 The rate of return, using market value of fund assets, earned by the retirement systems will be measured. Then the rate of return will be statistically tested to see if various investment strategies have had any significant affect on the rate of return earned on the portfolios of investments. The rate of return earned on the pension fund assets is important for a number of reasons. First, an increased rate of return earned on pension assets enables the cost of operating the pension system to decrease from 4 percent to 6 percent for each one-fourth of 1 percent increase in rate of return.1 Another way of looking at this is if the yield on the portfolio rises from 4 percent to 6 percent per year on a normal life fund, the long-run costs of the fund will be about 65 percent of what they would have been without the increased yield.2 This decrease in costs, while maintaining a given level of retirement benefits, enables tax revenues to be used for other government services and programs. Similarly, if the earnings on the funds rise, increased benefits can be paid to retirees without an 1Joseph J. Melone and Everett T. Allen, Pension Planning, (Homewood, I11.: Richard D. Irwin, Inc., 1966), p. 233. 2Daniel L. Schneid, "On Measuring Pension Fund Performance," Burroughs Clearing House, April, 1971, p. 61. 5 attendant increase in the cost of operating the retire- ment system. Thirdly, as state government expenditures rise at a faster rate than their revenues, the pressure for greater investment return on retirement system assets increases. That is, the government would like the em- ployee retirement system to be self-sustaining so that general tax revenues would not have to be diverted to pay retirement system benefits. However, as the number of retirees increase, the states may find that insuf- ficient financing of the system will necessitate using general tax revenues to pay benefits in the future. Thus, any increase in investment income will tend to reduce the need for utilizing general tax revenues to finance pension payments. Also, since many of the retirement systems cal- culate benefits as a percentage of the last few years (or highest salary years if different from the last few years salary in some cases) earnings, there is a normal tendency for benefits to increase over time as salaries rise for merit and/or economic reasons. There- fore, if earnings rise on the pension fund assets, it will enable these increasing benefits to be paid to re- tirees without too much financial strain on the system. In addition, the rampant inflation rates being experienced in the United States (with little assurance 6 that these rates will be controlled in the near future) puts increased pressure upon retirement systems to pro- vide higher benefits. That is, retirees will demand greater pension payments in order to keep up with the rising cost-of-living.3 In a recent settlement between the United Steel- workers and the aluminum industry the workers pension payments included a partial cost-of-living escalator. Those who retire will receive periodic pension adjust- ments of about 65 percent of the increase in the Consumer Price Index.4 Various estimates of the added cost to the pen- sion plans for tying benefits in with the Consumer Price Index are: 1. Costs would range from 25 to 50 percent more. 2. Costs would increase from 8 to 10 percent for every 1 percent increase in the cost- of-living.5 It will only be a matter of time before this inno- vation in the private sector will be carried over to the 3"Texas State Employees," Wall Street Journal, May 28, 1974, p. 1, column 5. 4"The Price of Aluminum Peace," Business Week, February 9, 1974, p. 44. S"Different Pattern in Pensions," U. S. News & World Report, February 18, 1974, pp. 90- 91. 7 public sector. Once this happens, the costs of public pensions will rise and an increase in the rate of return on the state retirement system assets will help offset some of the increasing costs. Another recent development which will have long run implications for pension fund financing is zero popu- lation growth. Relatively fewer workers in the future contributing toward the pension benefits of a growing num- ber of retirees will put increasing pressure on the funds to earn a greater rate of return if the funds are on a pay-as-you-go basis. Also, the trend toward early retirement increases the cost of pension plans if benefits are not reduced.6 With fewer people working a shorter number of years, an additional rate of return on assets would help increase the states ability to pay the greater dollar amount of benefits. Finally, as state funds attempt to earn a higher rate of return they may accept greater risks in order to obtain the higher return. Investment alternatives which once were not allowed in the portfolio because they were thought to be too risky (for example, common stock) might now be legally acceptable as investment possibilities. However, in spite of legal restrictions or in- creased risk acceptance by the portfolio managers, there 6"Why the Big Swing to EarLyRetirement," U-S- News & World Report, May 13, 1974, p. 59. 8 is no guarantee as to a higher rate of return on the portfolios. That is, loss in value and/or decrease in the income stream generated by the investment is possible. Those states holding shares of Equity Funding and Penn Central, just to cite a couple of examples, can readily testify to the fact that no return is guaranteed. Once the rate of return at market value is com- puted for all the funds in the sample for the years 1967 through 1972, meaningful comparisons among the funds and the market averages are possible. Also, since rate of return on all the funds is calculated in the same way, interfund comparisons can be made. The major hypothesis to test is that the state retirement systems have not been able to outperform the market in general, as reflected in the Dow Jones Indus— trial Average or the Standard and Poor's 500 Common Stock Index. Then statistical tests will be run to determine if the following factors significantly affected the rate of return: 1. Size of the pension fund. 2. Turnover rate of the pension fund. 3. Liquidity of the fund. 4. Type of management of the fund. 5. Percentage of equity investments in the fund. 6. Degree of diversification of the portfolio. Need for Study Previous studies of the investment performance of state retirement systems have been incomplete. A major- ity of the studies have been limited to a qualitative study of the investment aspects of state retirement plans.7 These studies primarily concentrate on areas such as: l. The legal restrictions on types of investments in which the states can invest. 2. The aggregate portfolio holdings by class of investment--i.e., Federal Government and Agency bonds, corporate bonds, mortgages, common and preferred stock. 3. The investment objectives of the various funds such as emphasis on liquidity; safety of principal; long-term growth of securities; large current income or as a hedge against inflation. Furthermore, when these studies do attempt to show a rate of return earned on the assets, it is usually 7For examples see: Bruce Davie, Investment Prac- tices of Public Employee Retirement Systems (Boston: Féderal Reserve Bafik of Boston, 1959): Elizabeth Deran, State and Local Employee Pension Systems (New York: Tax Foundation, Inc., 1969); W. W. Schmid, Retirement Systems of the American Teacher (New York: Fleet Academic Edi- tions, Inc., 1971); J. R. Tucker, State and Local Pension Funds 1972--Digest of Authorized Investments and Actual Investments (New York: Securities Industry Association, 1972). 10 expressed as the percentage of dividend and interest in- come divided by the average 922k value of the assets of the fund. Unfortunately, these measures do not account for changing values in the assets of the funds as reflected in their market prices--whether it be capital appreciation or depreciation. Another limitation of previous studies is that they focus only upon single state retirement systems or at best on regions of the United States.8 In contrast, this study will be national in scope and should overcome some of the limitations of making general conclusions from regional data. A further reason for this study is that state re- tirement fund assets account for almost one—fifth of the total pension fund assets in the United States (see Table I-2). 8For examples see: J. L. Cooper, "Investment Policies of State and Local Pension Funds," The Southern Journal of Business, Vol. 6, No. 2, Athens, Georgia, April, 1971; R. M. Soldofsky and E. V. Zuber, The Invest- ment Policies of the Iowa Public Employees Retirement System (Iowa City, Iowa: Bureau of BusIness and Economic Research, University of Iowa, 1964); Investment Manage- ment Performance of the New York State Teachers Retire- ment System (Albany, New York: New York State Teachers Retirement Board, January, 1964); Management Survey of the State Teachers Retirement Systems (Sacramento, Cali- fornia: Peat, Marwick, Mitchell and Company, Joint Legislative Retirement Committee, 1967); and Staff Report No. 58, Major State Retirement Systems (Columbus, Ohio: Ohio Legislative Service Commission, 1965). 11 Table I-2. Dollar value (in billions) of all retirement systems in the United States as of 1972. Type of Pension Fund Dollar Holdings Ogeggigf Private $ 152 54 State 51 18 Local Government 17 6 Federal Government 60 22 Total $ 280 100 Source: United States Securities and Exchange Commission, Statistical Series Number 2581 of April 4, 1972. Data Availabiligy At the onset of this study the naive assumption was made that since state pension funds are funded by public tax revenues (either in the form of state and/or local government contributions directly to the fund and/or through employee contributions in the form of tax paid salary deductions) there would be no problem in ob- taining the security holdings and annual reports of these funds. However, in collecting the data, several states obviously did not want to provide information regarding their investment holdings. Six states did not bother to respond to the original request for data nor to any of the three follow-up letters. Another thirteen states sent 12 insufficient or incomplete data which made it impossible to calculate market values of the portfolios for this study. Typical responses received from those states sup— plying minimal information are as follows: This report does not contain a listing of our se- curities holdings. . . . We do not attempt to pro- vide a listing of our holdings for anyone. The information you requested is not filled out. It will take too long to research. Our annual reports and portfolio lists are not in any way secret. However, we do not have extra copies which can be distributed to interested parties. I shall be glad to discuss our holdings with you if you care to call me on the phone. We do not have reports available for distribution. The State of does not have back data which would permit an accurate evaluation of its pension performance over the period which you men- tion. It is the duty of the state retirement board to furnish once a year to each member currently making deposits, a statement of his account together with appropriate explanatory material. The additional paragraphs defined a member precisely-- obviously not the researcher. You may come to our office and sight the material but may not make any copies of it. In a couple of extreme cases, members of the sys- tem attempted to get detailed data regarding their own pen- sion plan portfolios but were unable to do so. Furthermore, two state retirement systems were in the process of reorganization or merger so they were 13 excluded as potential members of the sample to be stud- ied. On a more positive note, twenty-nine states did provide data on their retirement system portfolios. Sample Size The states currently administer about 175 differ- ent retirement systems.9 In order to keep the sample size manageable, only the major retirement systems of each state will be studied. The major retirement systems of each state are those covering state employees and those covering teachers and/or school employees (in some states non-teaching school employees are covered by the state employees system). There are thirty-three states which have both employee and teacher (or school employee) systems--a total of sixty- six. The remaining seventeen states cover all state em- ployees, including teachers, under one system. Therefore, the sum of the major state retirement systems is eighty— three. Even though this is only 47 percent (83 out of 175 systems) of the number of state retirement systems, these eighty-three funds account for 93 percent ($54.372 billion) of the $58.499 billion in state pension fund assets.10 9U.S. Bureau of the Census, Finances of Employee- Retirement Systems of State and Local Governments in 10 Ibid., pp. 8-21. 14 Next, of the eighty-three state systems, the com- bination of six states not supplying information, thirteen states supplying minimal information,and two states in the stage of reorganization means that thirty-two of the major systems are excluded from the study. Of the remaining fifty-one (83-32) major state systems, only forty could ultimately be examined. This was true because among the twenty-nine states providing data, in eleven cases there were sufficient gaps between years to make continuous analysis of the data across six years impossible. This means that for eleven of the twenty—nine participating states only one of the two state retirement systems can be examined. However, these forty major retirement system funds left in the sample account for $34.066 billion out of total state pension assets of $58.499 billion--almost(H)percent (see Table I-3). Table I-3. States included in the sample to be studied. Arizona Iowa North Carolina Arkansas Kentucky Oklahoma California Louisiana Pennsylvania Colorado Michigan Rhode Island Connecticut Minnesota Texas Florida Montana Utah Georgia Nevada Vermont Idaho New Jersey West Virginia Illinois New Mexico Wisconsin Indiana New York 15 Time Period Covered Originally, this study was to cover a ten year period. However, in collecting data it soon became ap- parent that very few states could provide portfolio data on the retirement systems for this length of time. Sub- sequently, data for fiscal years ending in 1967 through 1972 were collected. These years were deemed suitable for analysis of the investment performance of the forty state funds for several reasons. First, this period of time represents both per- iods of significant growth and decline in the securities markets. The investment performance of the funds can then be measured in both periods of upturn and downturn. Secondly, many states were legally authorized to invest in equity issues for the first time during this time frame. The purpose of investing in equity issues was to increase the rate of return on the portfolios through capital appreciation of the common stock. The impact upon rate of return of adding stocks to the portfolios will be observed. Thirdly, as in most cases of financial reporting, there is a time lag between the end of a fiscal year and the time a report for that period is published. In one extreme case there was a four year lag on a consistent basis between the end of the fiscal year and the annual 16 report. The year 1972 represents the last year for which relatively complete data for the sample could be obtained. Ogganization of the Study Chapter II is a coverage of the investment poli- cies of the state retirement systems. It will include the legal restrictions imposed on the funds; the responsible agent for making investments in the state; the invest- ment objectives of the various funds and the security holdings of the funds. It willsnunvhow these factors have changed over the years 1967 through 1972. Chapter III is devoted to the measurement of the rate of return on each portfolio in the sample for the six years of the study. It includes a comparison of the rate of return earned on the retirement system assets with the market averages. In Chapter IV an analysis of the rate of return with the specific factors previously mentioned is made. This will indicate if various investment policies and strategies have had any significant affect on the rates of return of the funds. A summary of the findings of this study is the subject of Chapter V. CHAPTER II THE INVESTMENT POLICIES OF STATE RETIREMENT SYSTEMS A description of the investment policies of the state retirement funds is presented in this chapter. Dollars to Invest The forty state retirement funds included in the sample being studied have responsibility for investing billions of dollars. Table II-l show the dollars invested each year by these funds. Table II-l. Dollars (in billions at market value) invested by the forty state retirement systems from 1967 to 1972 Percent Change Year Dollars Invested From Previous Year 1967 $ 14.428 12 1968 16.144 12 1969 17.645 9 1970 18.634 5.6 1971 23.492 26 1972 26.903 14.5 Source: Annual reports of the forty state re- tirement systems. Note the slow down in the rate of growth of the funds in 1969 and 1970. This was due to rising interest 17 18 rates during this time period and the corresponding gen- eral decrease in stock prices. With depressed stock markets and bond prices on outstanding issues (to reflect the higher yields in the current market) the total market value of the portfolios naturally decrease. Similarly, in 1971 and 1972 the growth rate of the portfolios accelerated, reflecting lower interest rates in the economy with the attendant rise in stock prices and also outstanding issues of debt to reflect lower cur- rent yields. When going from the depressed base of 1970 to a high period of market activity in 1971, the change is par- ticularly dramatic. Investment Objectives The investment objectives of the pension funds are best seen by examining the responses of the fund managers as to what they perceived to be the most important objectives of their particular funds. In order to accomplish this, the fund directors were asked to indicate for their system the degree of im- portance of seven given objectives. Ten funds responded specifically to the question of investment objectives and these responses are summarized in Table II-Z. Judging from these limited responses, it appears as if safety of principal is of primary importance. This .meOHUwHHmv OCH—M OUMUM 0U “Gmw OHHMGCOHumOSO "OUHDOW 19 OH 0 m v N H COHumHmcH umsHmom omoom .5 0H m H m m 0 mchm Hmuwmmo u0\psm onH> Hwakuchmm humuocos mmumq .o OH H H m m o 6385 39.8 .m OH m m m o H meoosH ucwuuso mmnmq .v OH o m m N m meuHHsomm mo nusoum Enmunmcoq .m 0H 0 o H m 4 H885... mo >uowmm .N 0H N H ... H 0 58363 .H Hmuos ucmuuomEHco mHnmuHmoQ ucmuuomEH ucmuuomeH HmuH> 0>Huomnbo >Hm> .>ocm5qmum wn mEmummm ucmfimuHumu mumum UmuomHmm mo mmbHuomnno ucmEumm>sH .mnHH mHQMB 20 is not too surprising when one considers that these are publicly contributed funds and therefore, a considerable public trust and fiduciary responsibility exists for the fund managers. Also of concern, as indicated in Table II—Z, are long—term growth of securities, stable income and liquid- ity. The need for long term growth of securities is looked upon as a hedge against inflation as well as a means of keeping up the ability to pay increasing benefits over time.11 Liquidity and stability of income is important so that the fund has sufficient cash flows to meet its re- quirements. That is, it is able to pay benefits to retirees; lump sum payments to those withdrawing from the system; and administrative expenses in the normal nay-to» day Operations of the system. Most funds are now in a situation where the cash inflows to the fund exceed the outflows so liquidity is not a problem at this time. This is further substantiated by the relatively low importance placed on large current income by most of the fund managers who did respond to the questionnaire. Other objectives <1f funds as expressed by their directors and not included in the above questionnaire are as follows: 11Barbara A. Patocka, "Public Funds: The Hercu— lean Task is Under Way," Pensigns, May/June 1973, Vol. 2, No. 2, pp. 33-34. 3. 21 Increased use of equities as a percentage of the total fund. Increased turnover of the fund to get bet- ter investment performance. Increased emphasis on quality growth stocks. These fund managers seem to agree that increased use of equities in their portfolios coupled with more aggressive management of the assets (higher turnover) will increase the rate of growth of their funds. A more detailed analysis of the effect of turn- over rates and increased use of equities on the rate of return of the portfolios will follow in Chapter IV. Legal Restrictions The various state retirement funds are limited as to what types of investments they can participate in by the reSpective state legislatures. Statutes abound as to: 1. The type of investment in which the fund can invest--bond, equity, mortgage, savings certificate or real estate. The agency issuing the security to be pur- chased by the fund-—state, local and federal government or corporation. 12"A Report on Ten Other State Pension Funds," The Institutional Investor, February, 1970, pp. 45-47. 12 22 3. The quality of the security being procured—- measured by national bond and stock rating services. 4. The person, committee or agency having the authority to invest the system funds. Consequently, the objectives as stated above by fund managers may be tempered by the legal environment in which they must operate. Some state funds must invest in debt issues only, whereas others may invest up to 100 percent in equity if they so desire. Several states specify the exact percent- age of equity funds to total funds that can be utilized. Other state restrictions are not as specific and require only that the investment policy follow the "prudent man rule" or those prescribed for state banks and/or insurance companies. See Appendix A for a detailed listing of the legal restrictions on investments for the major state funds. Basically all states allow investments in United States Government Securities, United States Government backed securities (such as agency bonds), their own state, local or municipal bonds and high grade corporate bonds. The major legal restrictions by state apply to mortgages, investment funds (mutual funds), corporate preferred and common stock and short term investments such as certi- ficates of deposit, commercial paper and savings shares. v: r: "so 23 Table II-3, on the following page, shows a summary of the investment restrictions on the major state systems. All states are authorized to invest in United States Government and U.S. Government backed bonds. Six states limit the purchase of state and local government bonds to only those bonds of their own state. The other forty-four states allow purchases of state and local government issues of any of the United States. Seventeen states have broadened the investment rules to include Canadian Government and/or Canadian Province Bonds. A11 fifty states authorize the funds to invest in corporate bonds with the major provision being that these bonds be rated in the top three ratings by one or two national bond rating services. This is in contrast to 1955 when only ten states authorized the funds to invest in corporate bonds.13 Nineteen states allow the funds to invest in railroad equipment trust certificates in addition to corporate bonds. All but three states authorize investments in mort- gages. Most states limit mortgage investments to those 13A. E. Grunewald, "Investment Requirements of State and Local Pension Funds," Municipal Finance Offi- cers Association of the United States and Canada, Spe- cial Bulletin 1957A, Chicago, Illinois, January 16, 1957, p. 4. 24 Table II-3. Investment restrictions on state retirement funds by number of states. Restriction Number of States State and local bond purchases restricted to own state issues only 6 May not invest in mortgages 3 May not invest in common stock 7 May not invest in preferred stock 13 May not invest in mutual funds 12 May invest as prudent man does 12 May invest as savings banks within the state do 7 May invest as domestic life insurance companies do 11 May not invest in Canadian or Canadian Province Bonds 33 May not invest in equipment trust certificates 31 May not invest in real estate 38 May not invest in short term assets such as C.D.'s Commercial 7 paper and Savings Shares Sources: John H. Harper, State and Local Pension {Ends 1970, Investment Bankers Association of America, WashIngton, D.C. J. Richard Tucker, State and Local Pension Funds 1972, Securities Industry Association, Washington, D.C. Fund annual reports. 25 backed by the V.A. and the FHA. On the other hand, only twelve states permit investments in income producing real estate such as apartments, office buildings and shopping centers. There has been a significant change in equity investments since 1955. In that year only fourteen states authorized their funds to purchase common stocks.14 Today, forty-three states permit the purchase of common stocks by the funds. However there is a wide range in the amounts of common stock which the various state funds may hold. Most states specify the maximum percent- age of equity to total fund assets which may be held at any given time. Table II-4 gives the legal limits on the amount of stock authorized to be held by the funds. In addition, thirty-seven states authorize invest- ment in preferred stocks provided they have a sound his- tory of dividend payments. Twelve states allow the funds to purchase shares in investment companies or mutual funds. At one time a couple of states purchased shares of common stock oriented mutual funds in order to obtain equity growth at a time when they were legally unable to buy common stock directly. However, these states now authorize investment in equity and thus the attractiveness of investing in mutual fund shares has been diminished. 14Ibid., p. 4. 26 Table II-4. Legal limits on the amount of stock which can be held by state funds. Limit Number of States No stock may be held 7 Up to 10% in stock may be held 4 More than 10% but no more than 25% in stock may be held 15 More than 25% but no more than 40% in stock may be held 7 More than 40% but no more than 75% in stock may be held 8 No percentage limit specified 12 Total 53* *Three states had different restrictions on different type funds. Sources: John H. Harper, State and Local Pension Funds 1970, Investment Bankers Association of America, Washington, D.C. J. Richard Tucker, State and Local Pension Funds 1972, Securities Industry Association, Washington, D.C. Fund annual reports. 27 Thirty-three states permit investment in short term securities such as certificates of deposit, com- mercial paper, mutual savings bank shares and savings and loan shares. Many states availed themselves of this opportunity to invest in short-term securities during the credit crunches of 1966 and 1969 when short term interest rates were high and much uncertainty existed as to the future of the stock market. Other states limit the investment alternatives of the funds to the same investment categories in which domestic life insurance companies and/or savings banks may participate. Twelve states restrict the funds to the same investments as domestic life insurance companies. Another seven states authorize investing in the same alternatives available to savings banks within the state. Of these seven states, three states also authorize invest- ments in what life insurance companies may invest. That is, for these three states it is a situation of investing in the securities authorized for insurance companies and/ or savings banks. Finally, eleven states restrict the funds to investing as a man of prudence would do. This is known as investing according to the Prudent Man Rule. Investment Holdings The change in types of portfolio holdings of state retirement systems has been significant over the past 28 two decades. Table II—S shows this change over time. Cash and near cash items have leveled off at 1 percent of total assets since 1964. This indicates that most funds are normally fully invested and not allowing idle assets to accumulate. Investment in Federal Government Securities has dropped markedly over the twenty year period, from 55 percent of total assets to only 4 percent last year. This is due to the fact that other newly authorized in- vestments, primarily corporate bonds and stocks, were more attractive investment vehicles for the fund managers. Also, the dollars invested in state and local securities has dropped from 17 percent of total assets in 1954 to a mere 1 percent in 1969 and subsequent years. It is ironic that states have invested in tax-free secur- ities when there is no tax advantage to the states when investing in these securities. The advantage of pur- chasing tax-exempt securities accrues primarily to indi- viduals in high personal income tax brackets. Some reasons offered for the funds buying state and local bonds are: political pressure, patriotic appeal at the state/local level and to support the price of existing bonds.15 158. A. Patocka, op. cit., p. 36. 29 .mucmscum>oo mo msmcmo NROH may no H .02 .O .Ho> .mucmEcuw>oo Hmooq paw mumum mo meummm ucmfimuHumMIOmhonEm .msmcmu on» «0 smmusm .m.D .mucmscuw>00 mo mamcwu thH on» no N .02 .0 .Ho> .mucwscum>oo Hmooq pom mumum mo meumwm unmEmnHuwmnomhonEm .msmcmu on» NO smmusm .m.D .mOINOmH was .HOIOONH .onlmomH .molmomH .wonbme .omnmmmH .monvmmH ”GH massacum>oo Hmooq 0cm oumummfiumEoummm ucmEmHHummlmoonmEm mo mwocmsflm .msmcmu can no smmnsm .m.D .vmmH CH mpcsm wmwdeEm oHHnsm mo mucmEumm>cH paw Ammo .msmcwo mnu Mo smmusm .m.D "mwousom OOH m.mm Om vv.mm H 4m. 4 OH.N H Om. msOH OOH mH.Hm Om Hm.nv H OO. O O~.m H NO. NHOH OOH Om.mv mm sv.Hv H mm. O mm.m H Om. HsOH OOH OO.mm OO Om.mm H Om. O Om.m H mm. OsOH OOH -.mm mm H0.0m m Om. HH Hm.m H ON. OOOH OOH OO.Hm Om N0.0N O OO. mH MH.v H mm. OOOH OOH so.sm OO OH.- m On. RH mm.v H ON. sOOH OOH ms.OH OO mO.mH O OH.H Om Os.v H mH. OOOH OOH sm.m Om mv.H RH OO. mm OO.N O OH. vaH a w w m a m a w a m Hmuoe meuHusomm mmHUHHsowm ucmscum>oo mmHuHHSme muHmommo pom ammo Mom» Hmnuo Hmooq can madam unmesum>oo Hmumpmm mchHom mo muse .mHmHHOU mo mcowHHHQ CH whom» pmuomem How mEmumxm ucmEmHHumu mumum Mom mmsHUHon usmEumm>cw mo monks .mIHH mHQMB 30 On the other hand, it is readily seen that the majority of new funds have been invested in other secur- ities consisting primarily of corporate bonds, corporate stocks, mortgages and miscellaneous investments such as real estate, mutual funds, certificates of deposit, com— mercial paper, mutual savings banks shares and savings and loan shares (see Table II-6). The majority of funds in the "Other Securities" category are invested in corporate bonds. This reflects liberalized investment rules for state pension funds and also an effort by fund managers to obtain a greater re- turn with corporate bonds than could be obtained on government securities. However, as legal categories of investments come to include equity, more funds shifted to stock investments to obtain long-term growth and dividend income to supplement the interest income of the funds. In the last decade corporate stocks increased by more than twelve times as a percentage of the "Other Securities" category of investments. The effect of the changing portfolio composition on the rate of return will be analyzed more closely in Chapter IV. Investment Responsibility The investment responsibility for the funds dif- fers from state to state. In some states the retirement 31 .mucmecum>oo mo msmcmo NOmH 0:» mo H .02 .m .Ho> Hmooq pom oumum mo mEoumOm unwEmHHuwmlmmOonEm .msmcmu on» mo smwusm .mucwficum>oo mo mamcmu OomH on» no N .02 .o .Ho> Hmooq pom cumum mo mEmumOm ucwEmHHummummOOHmEm .wsmcmo on» no smmusm .mucchum>oo .m.D . mHCOECHO>OU . m . D .MOINOmH can HOIOOmH .OOummmH .molmomH .monOomH .mm|¢omH “CH uschHm>ou Hmooq was mumum mo mEoummm ucmEmHHummuwmmddem mo moocmch .msmcmo 0:» mo smmusm .m.D "moousow mv.mm m mm.m HH mm.m NN mo.NH Ho om.mm MOmH Nm.Ov o om.N mH qH.m mH HN.m No Om.mN NOmH Ov.Hv m ON.N mH No.0 OH mm.m mo wN.mN HOmH mm.mm m 0N.N 0H Hm.m VH MH.m we mo.NN OOmH Ho.0m O OH.N 0H Hm.v NH mo.m mo Om.mH mmmH No.0N m Ho.N OH Om.v OH Om.N mo Om.oH mme VH.NN O mo.H mH 0N.v m Hm.H mm Nm.vH OomH mm.MH O OO. mH mm.N O Nm. mo mv.m cmmH Hmuoe w msomcmHHmumHz w mommwuuoz w xooum m mpcom How» mumuomuou mumuomuou .mHmHHoo mo mcoHHHHn CH munch pmuOMHmm mom mpcsm ucmEmuHumH mumum How Onomwumo =mmHuHusomm Magoo: mo mHmOHmcd .muHH mHQMB 32 system administration and investment functions are car- ried out by separate units, whereas in other states the same unit carries out both tasks. Some states have pro- fessional investment service while in others it is a function for the State Treasurer or System official(s). Many of the states utilize the advisory services of professional investment firms in selecting securities. This information serves as an input into the investment decision process, but final authority rests with the person or committee shown in Table II-7. Two noteworthy exceptions are Oregon and Idaho which have delegated the actual investment function to an outside group. Oregon uses three investment firms to invest its equity funds. The three firms are evaluated on their performance and those that do well are retained. If a firm does not do well in comparison to the others it can be replaced by a new investment house. In effect, firms compete to be the manager of the equity portion of Oregon's fund.l6 Idaho has delegated the investment responsibility of its fund to a New York bank and a local bank. The effect of the type of fund management on the rate of return earned by the fund will also be analyzed in Chapter IV. 16"Oregon Blazed the Pension Trail," The Insti- tutional Investor, February 1970, pp. 41-44. 33 Table II-7. Investment responsibility for investing state pension fund assets. Fund investment is the responsibility of: State Board of Administration, Board of Trustees, Retirement Board or Board of Control Alabama Arizona (with consultation of investment advisor) Arkansas California (delegates authority to fund executive officer) Colorado Delaware Florida Georgia Idaho (selects trustee - currently a New York City bank and an Idaho bank) Indiana Kansas Kentucky (appoints an investment committee) Louisiana Maine Mississippi Maryland Missouri Montana Nevada New York (with investment counselor who is State Comptroller) North Carolina Ohio Oklahoma Pennsylvania Rhode Island South Carolina South Dakota (can transfer funds to a funding agent to invest for them - cor- porate or individual trustee or insurance company) 34 Table II-7. (continued) Tennessee Texas Virginia Washington West Virginia Wyoming Fund investment is the responsibility of a State Official Alaska - Commissioner of Revenue Connecticut — State Treasurer Iowa - State Treasurer, after consulting with Advisory Board Massachusetts - State Treasurer subject to Investment Committee approval Michigan - Deputy State Treasurer New Hampshire - State Treasurer Vermont - State Treasurer Fund investment is the responsibility of a State Board of Investment (Board usually invests for all the funds of the state) Illinois Minnesota Nebraska New Jersey - (part of the Treasury Department) New Mexico North Dakota Oregon (turns over part of the fund to profes- sional investor on competitive basis) Wisconsin Sources: John H. Harper, State and Local Pen- §i9n Funds 1970, Investment Bankers Association of America, Washington, D.C. J. Richard Tucker, State and Local Pension Eflflds 1972, Securities Industry Association, Washington, Fund Annual Reports. Personal correspondence with fund officials. 35 Before analyzing the effects of different invest- ment policies and strategies, the rate of return on the funds will be calculated in the next chapter. Chapter III RATE OF RETURN Measuring Performance The task of measuring the investment performance of the retirement funds presents a dilemma between taking a theoretically sound approach or a practical approach. There is general agreement that a discounted rate of return technique is appropriate for determining the rate of return on an investment portfolio. However, a distinction is made between the geometric average rate of return (time weighted return) and the internal rate of return (dollar weighted return). The internal rate of return is useful in measuring the overall results of a retirement fund, whereas the time weighted measure is more suitable for measuring the performance of the investment board or trustees. This is true because the trustees or board has no control over the timing of contributions and withdrawals from the fund and therefore should be evaluated only on how they did with the funds given to them. Using an internal rate of return would penalize the managers in that they 36 37 would be held responsible for factors over which they 17 have no control. Furthermore, the time weighted average rate of return requires an evaluation of the portfolio at each time a contribution is.made. Obviously, this information is not provided in annual reports and makes it impossible to calculate a geometric average rate of return for this study even if one desired to do so. Current Techniques Currently, states do not measure their portfolio returns on comparable bases. That is, there appears to be almost as many techniques for measuring return as there are states. Consequently, a 6 percent rate of return reported for one fund may really be similar to a 4 percent return achieved on a second fund if both were calculated in the same way. Several states only report a rate of return on their investments but offer no explanation of how this return was obtained. This, of course, is quite confusing to the reader who is unfamiliar with these different calculating techniques. 17For a discussion of these issues see: J. Peter Williamson, Investments—-New Analytic Techniques, Praeger Publications, New York, 1970, pp. 23:24 and Catherine A. Higgins "Calculating Rate of Return," Pensions, July/ August 1973, Vol. 2, No. 3, pp. 82-83. 38 Examples of the different techniques utilized by states in calculating rate of return are shown in Table 111-1 on the following page. Note that of the twenty-nine states included in the sample, twenty-four mention some type of return in their annual report. However, only three states go on to give the formula utilized in calculating the particular rate of return. Consequently, no meaningful comparison among states can be made utilizing these measures because we would be comparing equals. Also, all of the return measures are based on book values and therefore ignore any market value increases or decreases. Basically, the dividend and interest income is divided by total assets or average assets on hand at book value for the year. The percentage given as the rate of return on the portfolio is really quite meaningless and should not be taken at face value. There are two major reasons why states use book value in the annual reports. First of all, the investment portfolios have traditionally been all debt instruments. The investment income was interest only and the bonds were held to maturity so there was no concern about capital appreciation or depreciation. It was standard operating procedure to carry debt issures at par value and if any portfolio included stocks, they were carried at cost value. No explicit 39 Table III-1. Methods of calculating rate of return on state retirement funds. Number of States Name of Return Utilizing Formula Measure This Measure Given Average Annual Rate of Return 1 No Average Return on New Debt Investment 1 No Average Weighted Yield 2 No Average Yield 1 No Effective Interest For 1 of the 2 Rate 2 States only Net Yield 2 For 1 of the 2 States only Return on Cost 1 No Return on New Assets in Excess of Current Liabilities 1 No Yield 6 For 1 of the 6 States only Yield on Cost 1 No Yield to Maturity 1 No No Yield Given 5 Totals 24 3 formulas given Source: Annual Reports of Funds. 4O recognition of capital gains or losses was generally given in the annual reports. A second reason for the enchantment with book value by the state funds is that book value was used for actuarial purposes and can readily be carried over into investment reporting. That is, for actuarial purposes capital gains and losses are accounted for only when realized and unrealized gains and losses are ignored. However, as states began to authorize the funds to invest in equity to achieve higher returns on the portfolios, it became necessary to measure the market values of the portfolios or at least the equity portions. Therefore, beginning in the mid 19605 a few states included the market value of the securities in their annual reports. However, it was not until the late 19605 and early 19705 that several billion dollar funds actually'began reporting market values. Other states Still. have not switched their reporting techniques and Conthiue to only give book values for their portfolio holdings. In an effort to obtain both better performance measSurement and a comparison among funds, some states haVEE begun subscribing to professional investment evaldiation services. 41 As an example, several states have subscribed to the portfolio evaluation services of A. G. Becker Company and a national investment banking firm and also a consultant in the field of measuring investment performance of pension funds. A. G. Becker does not manage the portfolio in these cases; instead it measures the investment performance of funds on a comparable basis. It is important to note that funds are classified such that those with similar size, objectives and restrictions are compared. This precludes comparing fund performance erroneously. Ohio and Idaho were the first two states subscribing to professional investment measurement service and as of this writing another four states are in the process of joining the service. The performance results are not publicly reported by the consulting firm. Instead, the state is given its results and a comparison of its fund performance in relation to similar funds. It is up to the discretion of the state fund administration to publish the results in its own annual report. Rate of Return Measure Used A practical approach to calculating the rate of return on the state pension assets had to be found. This calculation had to recognize any changes in the market value of the portfolio while at the same time not require 42 monthly or quarterly data input. The latter was true because only data on an annual basis could be obtained from the funds. The technique ultimately found which meets the above criteria is the formula developed by Peter O. Dietz:18 Mz-Ml-C Ml + 1/2C R = where, R = rate of return for the year earned on the portfolio, M = dollar market value of the portfolio at the end of the year, M = dollar market value of the portfolio at the beginning of the year, C = net dollar contributions to the fund for the year (net means contributions less payments/ withdrawals). The Dietz formula assumes that dividend and/or interest income is reinvested when received. This assumption is quite valid for state pension funds since the contributions to the funds exceed the withdrawals and payments so that the funds do not have to rely upon 18Peter O. Dietz, Pension Funds: Measuring Investment Performance, The Free Press TDIvision of the MacMillan Co.), New York, u.y., 1966, pp. 50-51. 43 investment income to meet expenses and pay benefits. That is, the income can realistically be reinvested as it is received. Also, most states limit the rate at which the composition of the portfolios as between debt and equity can be changed. Therefore the assumption of reinvestment of income in the same proportion as the existing portfolio is quite realistic. Overall Rate of Return The rate of return on each of the forty funds was calculated for the years 1967 through 1972. Changes in both the market value of equity and debt issues are included in this measure. The ending market value of each year also reflects the reinvested value of dividends and interest received by the funds. Table III-2 on the following pages shows the overall rate of return on the forty state pension funds included in the sample for each of the years 1967 through 1972. Most of the states are on a fiscal year ending June 30. Therefore, the rate of return is calculated at that date for each fiscal year. As a risk surrogate the mean absolute deviation 19 (MAD) is used. A higher MAD implies a greater degree l9 . . . _ For a discuSSIOn of risk measures see: J. C. Francis and S. H. Archer, Portfolio Analysis, Prentice- Hall, Inc., Englewood Cliffs, New Jersey, pp. 212—223. 44 vm.m mm. Ho.m Hv.mN 0N.NHI Om.wl mm.va mm.NI OH Nm.v Om.HI mo.v Nm.m mo.ml NO.OI OO.MI mN.HI mH oo.m 0O. wH.mH mN.0H mO.MHI MO.HI mm.vl vm.HI mH vo.m mo. mv.m mw.¢H Hm.0HI mO.mI NN. mm.1 vH 0O.HH Hm.m mm.m mH.mN mv.mHI mm.vl mv. No.m MH mm.O mv.H om.v mo.mH mO.HHI mo.ml Nm.HI mm.v NH ON.m mw.H mv.m Hm.m mm.m| mO.wl Om.m mN.N HH Om.m mm.HI Nv.N ON.HI mm.m vv.OHI mm.m| ON.NI 0H mm.m mm.| Nv.m «H.m mm.ml mO.OI ON.mI «H. m Om.m Hm.H1 mm.m HH. vm.OH Om.mNI mm.m1 mo.HI m oo.v mm.HI HN.H mm.m 0O.ml OH.OI 0H.m1 mo.N| O vN.v Hm.NI NO.N mO.v HO.OI mm.ml mm.vu oO.mI w OH.m mH.HI mm.m mm.m ON.OI HN.mI qv.m| MN.NI m Ho.v 0H.Hl om.H mm.v ON.OI ON.mI 0O.HI m.N O OO.@ mH.mI mO.m No.m mH.vHI mm.OI mv.OI mm.HI m wo.m om.| vH.v Nm.NH vO.mI wm.m1 mm.OI vm.l N mo.m mm.l mm.m Nm.HH MH.HHI Om.mI NN. mN.HI H Hmummouusm xmwmv x NOmH HOmH OOmH mme momH OmmH bash COHumH>mo musHOmnd smmz .NOOH assess» OmmH whom» MOM mpcsm usmEmuHuwu mumum so usmoumm cw cusumu mo mush HHmum>o .NIHHH mHnt 45 mm.m vm.H| mm.H mN.m mm.ml Oo.ml mm.MI oo.HI mm mo.m @H.HI mo. Om.o wm.VI mw.wl HH.MI mm.| Nm m0.0 mv.l mo.m om.MH mH.HHI mo.ml mo.ml NN.HI Hm mH.v mO.I vo.v vm.m ow.¢l om.ml mo.ml Ho.l om mO.m HH. mO.m om.vH Nm.m1 mv.vl Nm.NI mH.mI mN vm.O Hm. mv.O mv.mH 0N.MHI wH.mI mm.N Hm.m1 mN om.o om.| mH.0H Nm.m mo.HHI mm.ml mv.wl mm.l ON mm.m ON.HI mO.H Hm.O Om.vl mo.ml Hm.HI ON.HI 0N vo.m vo.NI Om.HI vo.NH mm.ml NN.0HI mm.N1 OH.HI mN mm.m ov. mo.mH mm.O mm.0HI mo.OI ON.OI mm.m vN Hm.O mm.H MH.m mH.0N mH.mI 0N.ml Om.l mo.HI MN wo.w mH. vm.m Hw.NH Nm.ml mO.mI ma.H NH.NI NN Ov.m ov.H| VN.v m.m mN.OHI OO.mI Nm.HI mm.H| HN mw.o vm.N mm.MH mv.HH vm.l ON.vI mm.v1 No. ON Om.m Om.HI vo.m ma.m NO.0HI mm.OI HN.NI mm.N mH MO.m Om. mm.m om.HN MH.VHI mO.mI mN.HI vH.HI mH Amummouusm memv x NOmH HOmH OOmH mmmH mmmH OmmH Undh coHumH>wo OOOHOmnd com: H.U.UCOOV NIHHH OHQMB 46 vm.m mm.OH mm.OI mO.o| mv.NI om.l m Hm.w mm.H OO.H om.mH mm.ml mm.ml mH.NI oo.N ow HN.m Hm. mO.m mw.m Hm.l mO.mI vo.vl HH.m mm mm.v NN. mm.m mv.OH MN.VI HN.mI mm.HI mm.H mm mv.v mo. mm.m mv.O ON.vI mo.¢l Nm.ml oo.HI Om Nm.m mO.I mm.v mo.v mv.wl mm.ml vm.HI mm.NI mm mm.o mo.| Hm.O mm.HH mo.OHI mo.ml mN.I Ho. mm N0.0 OH.N mv.w OO.HH mo.MHI Om.NI Nw.HH om.| Om Amummouusm memv x NOmH HOmH OOmH mmmH mme OmmH wash coHumH>mo musHOmnd smmz H.U.UCOOV NIHHH OHQMB 47 of risk than does a lower MAD. In this sense, risk is the volatility in rate of return over the six-year time span. Therefore,:h1analyzing rate of return one must also look at the risk involved in obtaining a given rate of return. Other things being equal, one would select from among two portfolios having the same rate of return, that portfolio with the least MAD. That is, less risk associated with this portfolio for a given return. From Table III-2 it can be seen that in four of the six years covered by this study there was a negative average rate of return on the forty funds. The overall average rate of return for all the funds for six years was -.24 percent. When using market values, the funds on the average about broke even over the six years. This is in sharp contrast to most annual reports in which the rate of return is calculated on book value. This method cannot help but show a positive rate of return because most states calculate an average rate of return by dividing interest plus dividend income (if any) by the book value (or average book value) of the fund. However, before being too critical of the performance achieved by these funds and coming to unfounded conclusions, one must keep several things in mind. 48 First of all, this is a six-year time span which is relatively short when dealing with pension funds. Pension fund managers say that they should be judged over the long run. However, finding agreement as to what constitutes the long run (for example, twenty years or thirty years) is even more difficult. Secondly, the time period covered excludes much of the growth in the securities markets in the early 19605. However, since data on this period was not available the study had to begin in fiscal year 1967. Finally, the market fluctuated wildly during the 1967-1972 time period. Credit crunches in 1966 and 1969 led to depressed market prices for both debt and equity issues. However, at the same time, new debt issues with 8 percent-10 percent coupons looked quite attractive and secure to the fund managers. Thus, in evaluating the rates of return, one should keep in mind what the market was doing. From this perspective, a 5 percent loss in a time period when the market falls 10 percent would not be considered too bad. Rate of Return Ranges Table III-2 shows that in 1967 the highest rate of return, 9.02 percent, was earned by fund number thir- teen. The lowest rate of return, -3.76 percent, was earned on fund number six. The average rate of return 49 for all forty funds in 1967 was —.3 percent. Twenty- eight of the forty funds achieved a negative return in 1967. Fund thirteen was heavily invested in equity issues, whereas fund six was almost 100 percent in debt issues. Fund thirteen selected stocks which appreciated:h1value, whereas the fund six bonds fell in value as interest rates increased in 1966 with some carryover into 1967. In 1968, fund thirty-four led with a rate of return of 11.42 percent. The lowest rate of return, —7.54 percent, was earned on fund three. The average loss of the forty funds in fiscal year 1968 was -2.45 percent. Thirty-three of the funds experienced a negative rate of return in 1968. Again, the high rate of return for fund thirty- four was because it was highly invested in equity (about one-third) and these holdings appreciated in value considerably. On the contrary, fund three was totally invested in debt and its holdings were largely depressed in price. Fund eight was the largest loser in 1969 with a -28.67 percent rate of return. The least lost was experienced by fund fifteen with a 1.73 percent depreciation in value. All forty funds experienced a loss in 1969 with the average loss of -6.78 percent. 50 These losses were the result of the credit crunch of 1969 when both stock and debt prices fell as interest rates rose in the United States economy. The reason for the large 28.67 percent loss on fund eight was that its portfolio consisted of a large number of state and local debt issues as well as low coupon (3-4 percent) older corporate debt issues. The fall in market price of these bonds was quite severe. In 1970, the high rate of return was 17.54 percent and the lowest return was -19.45 percent. The average rate of return for the fiscal year was —7.88 percent. Thirty-eight of the forty funds experienced a loss as the credit crunch spilled over in 1970. A turnaround took place in 1971 as all but one fund experienced positive rates of return. The average gain for the forty funds was 10.63 percent with 28.18 percent for fund thirteen the high and -1.20 percent for fund ten the low. This turnabout was the result of the end of the credit crunch and the attendant fall in interest rates which caused the market price of both debt and equity issues to increase. In going from a low base of 1970 the rise in 1971 was even more dramatic. In 1972 there was a continuation of the positive gains of 1971 although at a more modest rate. 51 Again, all but one fund showed a positive rate of return with the average fund achieving a 5.34 percent return. The high return was 15.66 percent and the low was a -1.87 percent earned on fund twenty-five. Average Compound Rates of Return Another way to look at the return earned on these state retirement funds is to calculate the average compound rates of return over various time horizons. Table III-3, on the following pages, shows the compound rates for the funds for two years through six years. For the two-year period 1967-1968, the average compound rate of loss was -l.4 percent. In the three- year period 1967-1969, the average compound rate of decrease was -3.27 percent. With the credit crunch of 1969 and 1970 the average compound rate of return for the 1967-1970 period fell to -4.48 percent. As the market improved in 1971, the average compound rate for the five years 1967 through 1971 for all the funds increased to -1.66 percent. For the six-year period 1967—1972 the average compound growth rate for all the funds was -.59 percent. 52 mo.H No.H Ho.v| No.H Hw.v MH oo.H «N. OO.mu mm.| mH.H NH mm.H HO. mH.HI om. vo.¢ HH wm.HI Nm.NI NN.m1 mv.m| Hm.N| 0H mN.H| MN.mI mm.mu mm.v1 NH.m1 m OO.NI vm.v| Nv.mu mo.NHI Hm.N| m mo.NI NO.N| mm.vn vO.vu mm.m| O mO.NI mO.mn Hm.m| OH.mI Hm.v| m mm.H| mm.Nn mo.m| om.v| mm.m| m mH.Hu OO.H| HN.mu Hm.Hn OO. O mv.m| mm.v| vm.OI mm.m| VO.vI m OO.: HO.HI vo.mu mO.mI om.m| N mH.H| OH.N1 mN.m| NN.mn mm.| H NOmHnOmmH HOmHuOomH OOmHIOomH mmmHnOmmH mmmHlOmmH cash usmEmuHumH mumum co pwsumm usmoumm CH ansumu mo mmumu UCDOQEOO mmmum>< .mIHHH OHQMB 53 0H.HI HN.mI vm.ml OH.¢I mm.m| ON mm.HI wm.HI mH.¢I Ho.vl mm.HI mN mN.NI mm.NI mm.ml Om.vl OO.HI mN no.1 Hm.NI mv.ml OO.MI mo.NI vN mm. vH. Nm.¢I mm.NI mm.l MN HH.I NN.HI mv.vl mo.m| mo.l NN mm.HI OO.NI Om.m| mm.ml mv.HI HN mN.N mH. mv.NI mo.ml mv.NI 0N Om.HI ov.N| mm.v| mm.NI mH. mH mm. H@.1 MO.mI mO.NI 0N.HI mH VN. mm.l om.wl om.vl mv.m1 OH Ov.HI vm.NI mO.v1 om.vl vm.NI mH mN. mq.NI «v.ml mv.NI mm.NI mH mN.I Om.: mm.v1 Nm.NI Hm.l vH NOmHIOmmH HOmHIOmmH OOmHImeH mmmHIOmmH mmmHlOmmH Gash H6130": mnHHH «Hams 54 mm.| mm.HI mw.vl ON.MI ov.HI m Om. Nm. mv.m1 mm.Hl HH.I 0v mN. vv.l vv.HI Hm.HI mm.l mm we. 00.1 mH.m1 mm.NI mo.| mm mo.l NN.HI ON.MI mm.NI Nv.NI Om vm.l Hm.HI mv.ml mv.NI vm.HI mm mm.| wm.HI mm.¢l 0N.m1 HH.I mm mO.H Om. mo.HI «v.N mo.m vm mv.HI mo.NI Ho.vl mm.m1 mv.NI mm HN.HI ov.HI mm.ml OO.NI Hm.HI Nm vm.l HO.NI mm.o| mm.v| mm.m1 Hm Om.l mm.HI NO.m1 mm.ml mm.HI om ON.1 mm.HI No.mu mv.ml mm.NI mN OO.! Hm.HI em.vl Ho.NI mm.I mN NOmHIOmmH HOmHImeH OOmHImeH mmmHlOmmH mmmHlmeH chm ..O.ucoov msHHH mHOma 55 One can see that over the six-year period of this study the average fund lost ground in that the average compound rate of return was about -.6 percent. This means that even though millions of dollars of interest and dividends were received by the funds, the depreciation in market values of the portfolios was even greater resulting in net losses for the typical fund. However, twelve funds did achieve positive average compound rates of return over the six-year period. These positive rates ranged from .04 percent to 2.29 percent. Rate of Return on Equity Another way to analyze performance is to compare the rate of return on the equity section of the portfolios with the market averages as reflected in the Dow Jones Industrial Average (DJIA) and Standard and Poor's 500 (S & P 500). Table III-4 which follows shows the rate or return on the equity portion of those funds authorized to invest in common stocks and also the market performance as indicated by the market indexes. In 1967 the average fund earned 3.17 percent on its equity portfolio as compared to 5.81 percent on the S & P 500 and -1.93 percent on the DJIA. The range of return figures was -13.52 percent on fund V to 21.98 percent on fund J. Seven of the twenty-five 56 ms.m NH.HH HH.~H mm.Om- mm.m- HO.H~ ~O.O s OH.O HO.OH mm.sm mm.m~u HO.H- OO.O OO.H~ O OH.O OO.~H ms.mm ~4.m~- mm.m- OH.HH mm.s H om.HH mm.H OH.OO HH.OH- OO.HH- ON.H mO.H- m NO.OH HH.HH NH.HH mv.O~- HO. Om.v- mv.H O mm.~ NO.m Hm.ON mH.OH- HO..- mH.m Hm.H- m OO.H Om.m NO.m~ OO.OH- mm.m- ma.m O~.m- m HH.H Hm.a HO.N¢ Hm.HN- OO.H- OO.NH .a.z O mm.H- mo.m HH.O~ HO.NN- OO.HH- .<.z .¢.z O H~.H- OO.m O~.mm .O~.Hm- HO.HH- Hm.~- .<.z m HH.H HO.H ms.om HO.ON- HO.H- m~.m Hm.H « m NHmH HHOH OHOH OOOH OOOH HOOH pass coHuuom OHHDOO CO mpasu usmEmuHumu mumum mo usmoumm :H cusumu mo mumm .quHH mHomB 57 .muwmm N ch04 mo.HI vw.v Ov.MN mm.vNI HN.wI mo.OH Nm.mHI > mo.l vm.ml mm.Hv om.HmI Ow.NHI Nm.m MH.m D mO.m Oo.OH NH.Hv mN.MNI wm.v1 mm.m mN.N B mm.v Nm.m VH.vm mv.MNI mm.HI vo.OH ov. w OO. OH.m om.mm OO.MMI mm.vHI mm.O Om.m m o~.mu mo. vm.mH mN.HNI mo.OI om.mNI «H.H O mH.NI mm.v mm.O HO.om1 Om.v1 oo.NH .4.z m mv.m H0.¢H Hm.mv oo.mNI mo.NI mN.m em.N O MO.m Hm.m HN.mm No.MNI mm.ml Hm.m .¢.z z «No.NH mm.m mo.OH .<.z .¢.z .¢.z .m.z Z 0H.v mH.m mw.Nm mv.ONI mo.ml Om.NH vv.m H x NOmH HOmH OOmH mmmH mmmH OmmH wash H.p.ucoov vnHHH wama 58 .mumwa m OHGO«« pmNHuonusa uoz u .¢.z OO.~ Hm.m ~0.0~ OH.H~1 HH.mu m~.e mm.Hu «HOO 4H.m O0.0 OO.mm HO.m~u OO.mu O0.0 HO.m oom mam O0.0 OO.~m mO.~Nu mm.mu HN.O OH.m m OO.H HO.N OO.vm mm.OHu Ov.mu Om.m HO.~H mm mm.m N0.0 mm.OH H0.0H- mv.m Om.m H0.0 mm OO.m H0.0 OH.O OO.m H0.0- H~.OH Om.su OO OO.m Hm.~H OH.OH mm.mH- Om.v- OO.m Om.OH oo 4~.OH 40.0H m~.H~ H0.0~- OH.~| Hm.mm «4.4- mm ..Om.~ mm.~ OH.Hm N0.0~- .«.z .<.z .<.z «a ma.m mm.mH Hm.O mm. OO.HH- O0.0 NH.NH N mm.O O~.HH OH.OH OO.¢~I OH.m- O0.0 N0.0 H ~O.H HH.O H0.0m Om.O~u OH.mu O~.O OH.m x OH.» HH.N HO.~m Hm.m~u NO.mu O~.m Om.mu z m NOOH HOOH OOOH OOOH OOOH HOOH pass H.U.ucoov quHH mHQme 59 funds outperformed the S & P 500 (5.81 percent) rate of return. In 1968, the average fund earned 8.24 percent as compared to 9.89 percent on the S & P 500 and 4.23 percent on the DJIA. The highest return was fund K (21.94 percent) with thelrwvbeing -29.5 percent on fund Q. In 1969, the credit crunch took its toll on all but two of the equity portfolios. The average loss was 5.55 percent as compared to about a 3 percent decrease in the market indexes. The greatest loss of 15 percent was experienced by fund R. In 1970 the market downtown proved to be quite severe with the average fund losing 22.75 percent. This figure was between the market averages of -23.71 percent and -21.15 percent. Only two funds had a positive rate of return for this year. The year 1971 showed an aboutface as the market rebounded from the 1970 lows. The average rate of return was 32.77 percent which was slightly below the S & P 500 but better than the DJIA figure. In 1972, the upward trend continued with the average rate of return being about 7 percent. Again ‘Ehis rate was between those of the two market indexes ILsed for this study. Only one fund (U) experienced a loss in 1972. 60 If the S & P 500 index is used, the average fund outperformed the market measure in only one year-— 1970. The average fund dropped 22.75 percent while the S & P 500 dropped 23.71 percent. In using the DJIA as the market measure, the average fund outperformed the market in four years-- 1967, 1968, 1971 and 1972. In 1969 and 1970 when the stock market dropped the average fund dropped by a larger amount than the market as measured by the DJIA. Consequently, depending on which market index is used there is some difference as to how well the pension funds did in relation to the general market. If one uses the S & P 500 as an index, the funds did poorer than the market. If one uses the DJIA as a standard, then the funds come out a little better in up markets and poorer in decreasing markets. Generally speaking, the performance of the stock portfolios fluctuates widely and often large gains (losses) are reversed in subsequent years so that the long term growth is not particularly dramatic. Also, one must consider that in the majority of cases the equity portfolios are a very small portion of the total portfolio. Therefore, even large gains or decreases in the stocks held have relatively little effect on overall portfolio return. 61 In the next chapter, an analysis of various factors which may affect the rate of return will be performed. CHAPTER IV INVESTMENT ANALYSIS In this chapter the rate of return on the portfolios will be examined with respect to size, turnover, liquidity, equity h01dings, diversifica- tion and type of management of the funds. Size of Portfolio The size of the portfolio could be an important factor on the rate of return earned if one subscribes to the argument that a small fund is more likely to earn a greater rate of return than a larger fund because the fund manager has a greater degree of flex- ibility in changing his portfolio holdings. If this were true, we would expect a significant inverse relationship between rate of return and the portfolio size. The size of the portfolios included in this study ranged widely. Table IV-l shows the average- size fund and the range of sizes for each of the six years studied. 62 63 Table IV-l. Fund size ranges and averages in $000's at the end of each fiscal year. .— f * Year Smallest fund Largest fund Average fund 1967 21,900 2,435,960 340,874 1968 30,714 2,687,034 382,145 1969 31,198 2,936,023 419,091 1970 33,371 3,137,605 450,216 1971 44,034 3,949,933 526,575 1972 56,878 4,672,985 642,434 Source: Annual reports. Similarly, the rate of return earned on the funds can be related to the size of the portfolios as shown in Table IV—2 on the following page. As seen, Group IV funds had the highest rate of return on the average in years 1967, 1969, 1971 and 1972. Group II funds did best in 1968 and 1970. On the other hand, Group IV did worst in 1968 and 1970. Group II did worst in 1967 and 1969. Correlation Between Size and Return In order to determine if any significant rela- tionship exists between rate of return and fund size, Pearson correlation coefficients were calculated for the six years. 64 Table IV—2. Ranges and average rates of return in percent earned on funds grouped in quartiles by dollar value. Quartile 1967 1968 1969 1970 1971 1972 I H 2.80 -.23 -3.88 -4.64 19.90 7.31 (Ten largest E.'3°15 -5.44 —9.09 -10.26 4.33 .05 funds) X -.51 -2.53 -6.32 -7.41 0.50 3.26 II H 2.59 5.87 -4.20 17.54 26.41 13.68 £.'2-33 -6.48 -28.67 -12.26 -1.20 2.42 X -.57 -l.69 -9.37 -4.88 10.31 6.18 III H 4.39 11.42 -1.73 —.91 20.18 15.16 S_-3.76 -6.66 -10.22 -l3.79 3.65 -1.87 X -.53 -2.22 -5.74 -8.80 10.44 4.67 IV H 9.02 2.63 ~3.54 -4.20 28.18 15.66 (Ten smallest .S -l.95 -7.45 -8.22 -l9.45 4.68 3.33 funds) X .41 -3.34 -S.69 -lO.43 12.28 6.77 H = highest rate earned in that quartile £_= lowest rate earned in that quartile x: average rate earned in that quartile 65 This was done by ranking the forty funds by dollar value from highest to lowest value by year. Then the rates of return earned on the funds were ranked from high to low for each of the six years. The corre- lation coefficients calculated are shown in Table IV-3 below. Table IV-3. Pearson correlation coefficients between size of fund and rate of return. Year Pearson correlation coefficient 1967 -.12 1968 -.05 1969 .07 1970 .09 1971 -.10 1972 -.19 In years 1967, 1968, 1971 and 1972 there was a negative correlation between fund size and the rate of return. In 1969 and 1970 there was a slight positive correlation between the two. However, the Pearson correlation coefficients for all six years, whether positive or negative, are not statistically significant.20 Consequently, no relation- ship between size and rate of return existed for this 20 than .05. The cut off for significance levels was greater 66 group of state retirement funds over the six year time span. Turnover The next factor studied was the portfolio turn- over rate. It is assumed that portfolio changes are made to ultimately enhance the rate of return earned on the portfolios. Managers seek to replace lower yielding securities with higher yielding ones as well as select equity issues which will appreciate in value. Thus a higher turnover should result in a higher rate of return. During this time period several states actively began to switch their portfolio holdings from low coupon debt issues to high coupon debt issues in a process commonly called bond swapping. This was par- ticularly true in 1966-67 and 1969-70 when the credit crunches produced record interest rates on high grade corporate bonds and mortgages. Funds sold their low yielding state, local and corporate bonds and replaced them with new debt bearing coupon rates of 7-9 percent. In doing so, the funds sold their existing low coupon securities at a loss. However, these fund managers felt that the additional gain in interest income over the next ten to twenty years would more than offset the loss in principal incurred 67 on the old bonds. This process was encouraged by many state legislatures who authorized the loss in principal to be amortized over the future life of the bonds bought to replace the low yielding bonds. Other states were prohibited from bond swapping in that the accounting regulations for these states required the entire loss in principal to be written off in the year it occurred--not amortized over the future. These states, there- fore, had very low turnover rates over the 1967-1972 period.21 Insufficient data could be Obtained to accurately measure the impact of the bond swapping programs on the overall fund rate of return. The turnover rates experienced by the state funds over the six years varied considerably. In many cases there was no turnover at all, while in a few extreme cases it was close to 100 percent. Table IV-4 shows the ranges and average turnover rates for the state retirement funds during 1967- 72. The most dramatic examples of turnover which are also documented elsewhere are those for the states 21See Sidney Homer and M. L. Leibowitz, Inside the Yield Book—-New Tools for Bond Management Strate (Englewood Cliffs, N.J.: Prentice-Hall, Inc., I972), pp. 100-101. 68 of Louisiana and Connecticut.22 In 1968 and 1969 there was more than a one-third turnover in the assets of the Louisiana Teachers' Retirement System. In Connecticut the portfolio turnover rates approached nearly 100 percent as old debt issues were replaced with higher yielding debt issues or by common stock. Table IV-4. Average turnover rates and ranges in percent for state pension funds. Highest Lowest Average Year turnover rate turnover rate turnover rate 1967 46 0 5.5 1968 40 0 6 1969 43 0 6.1 1970 45 0 5.2 1971 70 0 7.6 1972 99 0 14.4 Source: Annual reports Calculating Turnover Rates In order to calculate the turnover rate of each portfolio the following formula was utilized: 22For Louisiana see: T. P. Bleakney, Retirement Systems for Public Employees, Pension Research Council (Homewood, Illinois: Richard D. Irwin, Inc., 1972), p. 141. For Connecticut see: Part I "Axe for the Buttonwood?", Barron's, May 7, 1973, pp. 3 ff., and Part3I%f"Treasurer's Report," Barron's, May 14, 1973, pp. . " 69 Turnover = S where, S = sales of securities during the year Vl = total value of the fund at the beginning of the fiscal year V2 = total value of the fund at the end of the fiscal year Correlation Between Turnover and Return Once the turnover rate for each fund was calcu- lated they were ranked from high to low for each year. Similarly, the rate of return on the funds was ranked from highest rate of return to lowest rate of return for each year. Next a Pearson correlation coefficient was cal- culated for each year to determine if there was a signi- ficant relationship between turnover rates and fund rate of return. The coefficients are shown in Table IV-5 below. Table IV-S. Pearson correlation coefficients between turnover rates and rate of return. Year Pearson correlation coefficient 1967 .64 1968 .08 1969 -.36 1970 .09 1971 .08 1972 .13 70 The correlation coefficients for 1967 and 1969 are the only significant ones. The 1967, .64 coefficient is significant with a p value of <.001. That is, it is highly significant and says that those funds with the highest turnover had the highest rate of return. How- ever, this result should be tempered by the fact that the one fund with the highest rate of return of all the funds in 1967 also had the highest turnover rate. Therefore, this combination may have unduly influenced the overall rate of return and turnover correlation analysis. For 1969, the -.36 coefficient is significant with a p value of <.05. This is not as significant as the 1967 coefficient but nevertheless does show a nega- tive relationship between the turnover rates and return earned on the funds. The remaining coefficients are very slightly positive but not statistically significant in showing any relationship between the turnover rates and return on investment. These results largely dispel any continued bene— fits of portfolio turnover. Stronger findings were made in the Institutional Investor Study of the Securities and Exchange Commission in which a negative correlation between turnover and return was established.23 23J. A. McQuown, "Technical and Fundamental Analysis and Capital Market Theory," Journal of Bank Research, Spring 1973, p. 11. 71 Analysis by the A. G. Becker evaluation service shows that on the average, the greater the turnover in recent years, the lower the realized returns for managed portfolios.24 In summary, there does not appear to be a defin- ite relationship between turnover rates and rate of return on the average for the forty state retirement funds examined during the 1967-72 time span. Liquidipy The third variable to examine is the liquidity of the retirement funds. Liquidity will be defined as the percentage of cash, deposits and short-term invest- ments (such as certificates of deposit, Treasury Bills and commercial paper) to total fund value. Generally speaking, the rate of return on funds with a high percentage of liquidity should be less than those funds with a lower degree of liquidity. There are two major reasons for this. First of all, cash, includ- ing checking accounts, receive no return and are in essence idle assets. Secondly, bank deposits and short- term investments normally have a lower yield than other 24William L. Fouse, "Measuring Investment Per- formance," Investment Systems Division Newsletter, Financial Analysis Department, Wells Fargo Bank, San Francisco, California, N. 11, Oct-Dec 1973, pp. 5-6 (from a paper presented at the 49th National Convention of the Bank Administration Institute, San Francisco, Oct. 28-31, 1973). 72 investment alternatives because of less risk exposure involved. Consequently, one would expect a negative correlation coefficient to exist between liquidity per— centages and the rates of return on the funds. Those funds with more idle assets or more lower yielding assets naturally produce a lower return than those funds fully invested in higher yielding investments. The liquidity for each of the funds was deter- mined in each year. The results appear in Table IV-6 below. Table IV—6. Liquidity, in percent of state retirement funds 1967-72. Highest Lowest Average Year liquidity liquidity liquidity 1967 28 0 2.5 1968 12 0 1.9 1969 10 0 2.0 1970 9 0 2.5 1971 10 0 3.0 1972 17 0 3.4 Source: Annual reports The average liquidity ranged from a low of 1.9 percent in 1968 to a high of 3.4 percent in 1972. This shows that in most cases the funds were almost fully 73 invested. The overall average liquidity for the forty funds over the six year period was 2.5 percent. One might expect higher percentages in 1966— 67 and 1969-70 as uncertainty over the future of the capital markets prevailed and fund managers were reti- cent to put money into securities—-particularly equity issues which were falling in price. However, this tendency may have been negated by the extremely high yields on corporate debt issues and mortgages. Many fund managers deliberately suspended their policy of adding to the stock portfolios and instead purchased high yielding debt issues which in many cases locked in a 7-9 percent rate of return for future years. In order to measure the relationship between liquidity and rate of return, Pearson correlation co- efficients were again calculated for each year. This was done by ranking the liquidity percent- ages fICmI high to low for each year. Then the rate of return was ranked from high to low for each of years 1967-72. The correlation coefficients are those in TafleIWJ. The results show no clear relationship between liquidity and rate of return. For three of the six years (1967, 1971 and 1972) there was a positive rela- tionship between the two factors. In the other three years a negative relationship existed. However, for 74 all six correlation coefficients, there was no statis-- tical significance associated with liquidity and rate of return. Table IV-7. Pearson correlation coefficients between llQuidity and rate of return. r Year Pearson correlation coefficient 1967 .25 1968 -.09 1969 -.30 1970 -.05 1971 .02 '1972 .21 One possible explanation for some of the coefficients being positive rather than negative is that during this period inverted yield curves existed in the money markets from time to time. That is, short-term interest rates were greater than long-term rates. COn- sequently, if fund managers increased their short-term investments (greater liquidity) during these time periods, the rate of return would be higher rather than lower. Another explanation for positive coefficients rather than negative ones is that in years 1967, 1971 and 1972 the credit crunches of 1966 and 1969-70 were easing and interest rates in general fell and stock prices rose. Therefore, portfolios appreciated in 75 market value and in many cases this percentage increase in value more than offset the higher liquidity rates of the funds. In conclusion, the degree of liquidity of the forty state funds did not appear to have any significant import on the rate of return earned on these funds. Equity Holdings As mentioned in Chapter II, many fund managers expressed a strong desire to add to the equity holdings of their portfolios in hopes of achieving a greater rate of return. State legislators were under pressure to authorize state funds to invest in equity or, for those already holding stock, to increase the maximum percent- age of equity that such funds could hold. The release, in 1968, of the famous Lorie and Fisher study of stock prices-~"Rates of Return on Invest- ments in Common Stocks"--which showed that the average annual rate of return for all common stocks listed on the New York Stock Exchange from 1926 through 1965 had been 9.3 percent, probably provided a great impetus for 25 the greater interest in equity investment in state funds. Also, the highly attractive growth figures of the stock 25Lawrence Fisher and J. H. Lorie, "Rates of Return on Investments in Common Stock: The Year-by-Year Record, 1926-65," Journal of Business, XLI, No. 3 (July, 1968). PP. 291-316. 76 market in the late 1950's and early 1960's added credence to the arguments by fund managers that they be authorized to invest more in equity issues. The fact that these efforts to increase the equity holdings of state retirement funds were successful can be seen in Table IV-8 below. Table IV-8. Market value in $000's of equity holdings of state retirement funds. Market value Number of Year of equity funds 1966 701,292 22 1967 985,069 27 1968 1,448,928 ‘ 29 1969 1,891,012 30 1970 2,166,408 31 1971 3,835,700 32 1972 5,484,485 32 Source: Annual reports The market value of equity holdings of the state funds increased from $701 million to $5,484 million-- an eightfold increase--from the end of fiscal year 1966 to the end of fiscal year 1972. Similarly, the number of funds in the sample holding equity issues grew from 55 percent in 1966 to 80 percent in 1972. 77 Another way to View this growth is to express the equity holdings as a percentage of the total port- folio value. The equity holdings of the funds grew from an average of 7.5 percent equity per fund in 1967 to an average of 19.7 percent equity per fund in 1972. The overall average percentage of equity held by the funds over the six years was almost 13 percent (see Table IV-9). Table IV-9. Equity holdings in percent of state retirement systems. Highest equity Lowest equity Average equity Year percentage percentage percentage 1967 41 0 7.5 1968 46 0 9.3 1969 39 1 11.4 1970 54 0 13.5 1971 63 0 16.3 1972 73 0 19.7 Source: Annual reports Sguity Holdings and Rate of Return Since the objective of adding to the equity hold- ings of the portfolios is to increase the overall rate of return, a positive correlation should exist between the percentage of equity holdings and the rate of return obtained by the fund. 78 Pearson correlation coefficients were again calculated to determine if any such relationship existed. This was done by ranking the portfolios from highest percent of equity to lowest and also by rate of return from highest to lowest. The coefficients can be seen in Table IV-lO. Table IV-10. Pearson correlation coefficients between equity holdings and rate of return. Year Pearson correlation coefficient 1967 .54 1968 .67 1969 .16 1970 -.45 1971 .65 1972 .15 In three of the six years there was a signifi- cant positive relationship between the two factors. In 1967 (.54), 1968 (.67) and 1971 (.65) the correlation was highly significant with a p<.001. For the remaining three years, the coefficients were not statistically significant although a -.45 coefficient in 1970 strongly hints that portfolio losses are the result of large stock market drops. That is, holding equity can work against the fund as well as for it. 79 The general conclusion that can be made from the above results is that when the stock market in general is rising the earnings potential of the fund rises too. However, when the market falls so does the earnings on the funds and holding equity certainly does not guaran- tee a higher rate of return on the funds. To further substantiate the claim that the rate of return on the equity portion of state portfolios is highly related to the market in general, a simple corre- 1ation was run by state between its equity return and both the Standard and Poor's 500 (S & P 500) Index and Dow Jones Industrial Average (DJIA) as market barom- eters. The results of this correlation analysis are summarized in Table IV-ll. Table IV-ll. Correlation between rate of return on state fund equity and market averages. Correlation Number of states Number of states coefficients with DJIA with S & P 500 .95 - 1.00 17 ll .85 - .94 6 12 .70 - .84 5 5 Less than .70 4 4 The return on equity for seventeen of the funds was between .95 and 1.00 correlated with the Dow Jones Industrial Average. Six of the funds were in the .85- .94 correlation range with the DJIA. 80 Similarly, eleven of the funds had equity returns related to the Standard and Poor's 500 Index with corre- lation coefficients between .95 and 1.00. Twelve funds were in the .85-.94 correlation range with the S & P 500. A total of twenty-three of the funds had a return on equity of between .85 and 1.00 correlated with both of the market indexes. Likewise, five of the funds were between the .70 and .84 correlation coefficients of both indexes and only four state funds were less than .70 correlated with the DJIA and the S & P 500 Index. Consequently, one would expect the performance Of the equity portion of the state retirement system portfolios in general to closely parallel that of the market averages. Diversification The state retirement fund portfolios are com- posed of a large number of debt issues and where author- ized common stock issues. Frequently, state legal restrictions governing these funds limit the ownership of the securities of any one firm to a maximum percent— age of the outstanding securities issued by that firm. The intent of these regulations is to ensure that the fund assets are not exposed to undue risk. In other words, the funds should "not have all their eggs in one basket." The primary concern is that if a large share 81 of the fund assets are invested in a small number of issues, the performance of the fund is too dependent on that particular group of securities. This is especially damaging when these securities all decline in value. The requirements for diversifying the portfolios as to limits on ownership of any one security or as to between debt and equity issues are designed primarily to protect the funds from significant losses over time. On the other hand, the diversification requirements also tend to limit gains on the upside of a market, but the primary emphasis as far as the legislatures are concerned is to prevent losses which would impair the funds abil- ity to pay benefits to the retirees. To measure the degree of diversification the number of issues of both common stock and corporate debt was calculated. Table IV-12 summarizes these calculations. The number of debt issues clearly outweighs the number of equity issues. This is to be expected how- ever, as most funds are limited as to how much in common stock they can hold. The average number of issues of both common stock and debt held by the funds almost doubled over the six year period. This shows that greater diversification of the portfolios, as measured by number of issues, took place over time. 82 Table IV-12. Diversification by number of issues held by state retirement funds. Year Common Stock Corporate Debt 1967 H 101 718 S 21 3 X 28 183 1968 H 107 872 L 14 18 X 33 177 1969 H 125 1045 S 13 32 X 41 248 1970 H 131 1049 L 15 44 X 47 276 1971 H 137 1069 E 20 63 X 51 299 1972 H 144 1081 S 27 71 X 50 326 H = High S = Low X = Average Source: Annual reports 83 Diversification and Rate of Return Diversification is said to reduce portfolio loss on the downside and also limit gain on the upside of the market. That is, when the market rises the gain on a highly diversified portfolio will approximate the market gain and the benefit of a few issues which out-perform the market by a wide margin will not have an appreciable effect on overall return. Similarly, on the downside, a large loss incurred on a few isSues will not have a noticeable effect on the overall rate of return. In order to determine if the degree of diversi- fication of the portfolio had any effect on the rate of return, correlation analysis was performed. Table IV-13 shows the correlation coefficients between rate of return and number of equity issues held in the portfolio. Table IV-13. Correlation coefficients between rate of return and degree of diversification. Year Correlation coefficient 1967 .26 1968 .18 1969 .04 1970 .02 1971 -.01 1972 -.08 84 The coefficients for all six years are not significant and indicate that no relationship exists between rate of return and diversification for these state funds. This finding can further be supported by the fact that all of the funds are diversified to such a high degree that little variation in return should be expected. Once the funds hold more than about sixteen randomly selected issues, very little further diversi- fication occurs as more and more securities are added to the portfolio.26 In effect, most of the portfolios are most likely over diversified and little risk from concentration of issues exists. Type of Management The final factor to consider in relation to return on investment is type of management. This is an attempt to determine if one type of portfolio manage- ment produces better returns than other management types. Three different management types are represented by the forty funds in the sample. First of all, two states have delegated the management of the portfolio to a professional investment advisor. 26For examples of studies relating number of issues to degree of diversification see: H. A. Latane and W. E. Young, "Test of Portfolio Building Rules," The Journal of Finance, 24 (September, 1969), pp. 595- 612 and Lawrence Fisher and J. H. Lorie, "Some Studies of Variability of Returns on Investments in Common Stocks," Journal of Business, XLIII, No. 2 (April, 1970), p. 117. 85 Twenty-four states have retained external advisors and even though the states do have final investment responsibility, these advisors basically make the deci- sions as to what the funds will invest in. The remaining fourteen states are largely internally administered. They may receive input from investment advisors but the fund personnel actually decide what securities will be purchased. In several cases these states have formed State Investment Boards to handle the portfolios of all the retirement funds in that state. The rates of return earned by the funds over the six years were studied in relation to the type of management to see if any differences existed. An analysis of variance was run over the years 1967-72. The analysis showed that there is a significant difference in overall rate of return on the funds over the six years. The F value was 33.11 which is significant ‘with p<.0005. This is to be expected though as the re- turn depends on the market conditions from year to year. The type of management was also found to be Significant with F = 28.51 and p<.0005. The analysis also shows that the interaction between the two (year and management type) is significant with F = 24.52 and P<.0005. 86 Even though differences in management were significant, there is no consistent pattern from year to year of any one management type always outperforming the others. This is seen in Table IV-l4. In 1967, 1968, 1971 and 1972 the two funds who delegated their portfolios to outside management did better on the average than the other two types of manage- ment. The internally managed funds ranked second in 1967, 1968, 1970 and 1971 and first in 1969. Finally, those funds who relied largely on the recommendations of investment counselors for their portfolio decision-making finished in third place in four of the years and ranked first in 1970 and third in 1971. At first glance it appears as if the portfolios delegated to outside management did particularly well since they were ranked first in four of the six years. However, the significance of this management type may be due to the fact that the sample size consisted of only two funds and both of these funds were highly invested in equities. The stock market went up in 1967, 1968, 1971 and 1972 which are precisely the years in which this type of management ranked first. Stock portfolios held in the other two management categories may have performed equally as well but the results were averaged down by portfolios in their own group holding high levels of debt instruments. 87 m N N H N N «H stumucH N m H m m m «N HomH>om nuH3 HmcuousH H H m N H H N Oonuso omummmHmo NOmH HOmH OOOH OmmH mmmH OomH mossm mo ucmfimmmcmz mo omxe HmnEsz .usoEmmmamE mo mama On cusumu HO mums mommm>w mtg mCchmm .vHu>H OHQMB 88 Furthermore, with the stock market decline of 1969—70, the outside managed portfolios dropped to second and third in the rankings. Thus it appears that the results of the two outside managed were dependent on fluctuations in the market in general rather than any particular type of management. The fact that these managers selected stocks that moved in concert with the market could be viewed as a fortuitous event. After all, had the small sample of two funds been in- vested in equities which happened to move opposite to the market, the performance would have been ranked poorly. The following chapter will summarize the find- ings of this study as well as point out future topics of concern for state retirement system investments. CHAPTER V SUMMARY AND CONCLUSIONS This study was designed to measure the invest- ment performance of state retirement funds. It is significant in that it measured the rate of return on state funds by utilizing market values instead of book values. Previous studies and fund annual reports generally showed the investment return on their funds as the percentage of investment income (interest and/or dividends) to book value of the portfolio holdings. These types of calculations do not account for any market fluctuations in the values of the stocks and bonds held by the retirement funds. Consequently, in periods of rising markets, return is understated and in falling markets overstated. Once the rate of return on each of the port- folios was calculated in a consistent manner, utilizing market values, valid comparisons among funds and analysis of investment performance could be done. Investment Performance The investment performance of the funds was calculated for years 1967-1972. These results are summarized in Table V-l on the following page. 89 90 Table V-l. Average overall rates of return and equity rates of return in percent on state retire- ment funds. Average overall Average return Year rate of return on equity 1967 -.3 3.17 1968 -2.45 8.24 1969 -6.78 —5.55 1970 -7.88 -22.75 1971 10.63 32.77 1972 5.34 6.99 Source: Tables III-2 and III-4. The average return on the bond portfolios of the funds over the six years was -l.05 percent. This low return was due to the decrease in market prices of outstanding bonds held by the funds as interest rates in general rose during this period of time. The average return on the equity portfolios of those states authorized to invest in common stock was 3.81 percent over the same six years. However, since the bond portfolios comprised almost 83 percent of total assets on the average over the six years, the average overall rate of return for the forty funds from 1967-1972 was -.24 percent. During the six year period the average return on the Standard and Poor's 500 Index was 5.14 percent 91 and the Dow Jones Industrial Average was 2 percent. Since most of the funds holding stocks have more than the thirty issues comprising the Dow Jones Industrial Average, the Standard & Poor's 500 is probably a better index against which to measure equity performance.27 If we use the Standard & Poor's 500 average of 5.41 percent over the six years then the average retirement fund performance of 3.81 percent is poorer. For those funds whose equity holdings closely approximate the blue chip composition of the Dow Jones Industrial Aver- age then the Dow's 2 percent return would be the stan- dard of comparison. In looking at overall rates of return, it is more difficult to evaluate the performance. One can make relative interfund comparisons, or comparisons to the average fund. However, there is difficulty in comparing the performance to a generally accepted market index which measures bond portfolio performance.28 The A. G. Becker Company has a study which shows that for 2600 pension funds (including some state funds in the last half of the time period) for the 1962-1972 time period, half failed to achieve a 7.5 percent return 27"The Popular Indexes Don't Tell the Whole Story," Business Week, July 7, 1973, pp. 74-77. 28See "Bond Men Seek a Uniform Index," Business 1623):. September 8, 1973, p. 65. —‘———' 92 on their total investments--including fixed income securities. During this same period the Standard & Poor's 500 Index returned 10 percent. Similarly the A. S. Hansen (an actuarial con- sultant) Company report for the same 1962-72 time period showed that the equity pension funds managed by thirty of the top banks yielded an average return slightly below the Standard & Poor's 500 Index.29 One should exercise caution in making direct comparisons with the state funds in this study and the funds analyzed in the Becker and Hansen reports above. First of all, the legal restrictions on invest- ment alternatives are different. Secondly, these reports covered ten years instead of six and therefore this study does not reflect the growth of the market in the period of 1962-1966. Finally, both the Becker and Hansen reports utilize time weighted rates of return as contrasted to the Dietz method for this study. Therefore, direct comparisons of the numbers should not be made. However, the predominant theme in the Becker and Hansen studies as well as the findings of this study is that it is difficult for a typical pension ‘ 29See Dana L. Thomas, "Divide and Multiply-- Pension Consultants Like to Spread Assets Around," Egrron's, Nov. 5, 1973, p. 7. 93 fund to consistently out-perform the market averages. This may lead to many state funds reassessing their strategy of trying to out-perform the market and instead be content to invest in unmanaged portfolios. An un- managed portfolio is one invested in an average such as the Standard & Poor's 500. Instead of trying to out-perform the market, the portfolio returns would duplicate the market. At the same time professional management and advisory fees would be substantially reduced.30 Further support for unmanaged portfolios is found in a recent study on the performance of commingled bank managed equity funds during the 1962-1970 time Span.31 Analysis of Performance The investment results of the forty state retirement systems studied were analyzed to see if the return was affected by fund size, turnover rate, liquidity, equity holdings, diversification, and type of management. 30See "Unmanaging the Portfolio," Business Week, Nov. 17, 1973, p. 78. . 31See Edward Malca, Bank-Administered Commingled Pens10n Funds-~Performance and CharacteriStics 1962-70, Lexington Books (Lexington, Mass.: D. C. Heath & Com- pany, 1973). 94 The findings are as follows: 1. There was no statistical significance between the size of the fund and the rate of return. 2. In 1967 there was a significant positive correlation between turnover and rate of return. In 1969 a less significant negative correlation existed for the funds. For the remaining four years there was no significant relationship between the two variables. Thus, it appears that turnover of the portfolio does not particularly enhance rate of return. 3. The degree of liquidity of the funds did not have a significant effect upon the rates of return earned on the portfolios. 4. There was a significant positive relation- ship between the percentage of equity held in three of the six years covered--l967, 1968 and 1971. However, the stock market was rising in these years so it is obvious that return should be better for those funds holding equity issues. However, in the three other years no significant relationship between equity holdings and return existed. This implies that holding equity does not ensure a better rate of return. 5. No significant relationship between the rate of return and portfolio diversification was found. Most of the portfolios are so highly diversified (measured by number of issues held) that it is not a significant factor with respect to rate of return. 95 6. The effect of type of management of the state funds on the rate of return was analyzed. To do this, the funds were divided into groups on the bases of whether: a) the management of the fund was delegated to an outside firm, b) the fund made most of the investment decisions based on the recommendations of investment advisors, or c) the fund was largely internally admin- istered. An analysis of variance over the six years between rate of return and type of management showed there was a significant difference between the variables. However, further analysis showed no consistent pattern of the relationship between return and management type from year to year. Also, the small sample size of two for delegated management types may have unduly influenced the analysis of variance. Availability of Data The most disappointing aspect of this study was the data form and data availability of state retirement system investments. Of the eighty-three major state systems only forty systems could eventually be analyzed. These forty systems were from twenty-nine different states 96 and represented almost 60 percent of the assets of the major state retirement funds. Twenty-one states were not included in the study for various reasons. Six states did not send any infor- mation. Another thirteen sent data which was either insufficient or incomplete and subsequently had to be excluded from the study. Two state funds, Kansas and Delaware, were in the process of reorganization and therefore were eliminated from the sample. In many cases, there was a definite reluctance to provide portfolio information for this study. Many fund administrators gave the impression that this in- formation was somehow secretive in nature. When one considers that the funds are financed with tax revenues it was surprising that fund managers should view their particular portfolios as being privileged information. Rumors of retirement fund mishandling can only be per- petuated in this type of environment. On the other hand, twenty-nine states did pro- vide data and the fund administrators were extremely cooperative in providing information for this study. Another problem encountered was that of data format. Many states still carry debt issues at par value and equity issues at book value. This has been standard operating procedure and is largely the result of actuarial reporting where capital gains and losses are not accounted for until actually realized. 97 This meant that many of the portfolios had to be converted from book to market values before investment performance and analysis could be undertaken. It is encouraging to note that the fund annual reports for more recent years are reporting the market value of securities as well as book values. Finally, timeliness in reporting for retirement fund should be emphasized. In most cases, the fund annual reports were produced shortly after the end of the fiscal year. However, in a few states, the annual reports were often a year or more late in coming out. In the extreme case, one state fund was consistently four years late in turning out its annual report. Future Areas of Research Next some possible future areas of research will be discussed. The Treasurer of the State of Connecticut has purchased a seat on the PBW Stock Exchange in order to cut the commission costs to the state of trading secur- ities on its pension funds. The Securities and Exchange Commission (SEC) has attempted to halt this action and subsequently the State of Connecticut has instituted a suit against the SEC to enable it to hold its seat on the PBW Exchange. Other states have joined Connecticut in pursuing this issue. If successful, it will be interesting to see if portfolio turnover rates are 98 affected by states owning seats on the exchange. It will also be interesting to see by how much portfolio trading costs are reduced.32 A second issue to examine is whether or not bond swapping (riding the yield curve) is an effective method of improving the rate of return on state retirement system bond portfolios. Currently, a limited number of states authorize their funds to amortize bond losses over the life of the new high coupon bonds purchased to replace the low coupon bonds held in the portfolio. If accounting regulations are changed to allow more states to amortize the loss over time, instead of writing off the entire loss in the year it occurs which effec- tively prohibits states from bond swapping, the benefits of this process can be tested. Finally, as many states attempt to get more and more equity issues in their portfolios, there may be funding difficulties for the retirement systems in the future. The retirement systems have assumed a rate of return for actuarial purposes related to the funding requirements of the pension plan. As funds added equity to their portfolios, anticipating higher rates of return on this common stock, they assumed funding 32"Axe for the Buttonwood?", op. cit., pp. 3 ff. 99 would be less of a burden. However, as the rates of return on equity over the six years of this study indi- cate, many of the funds may have overstated their assump- tions regarding earnings on their investment portfolios. Furthermore, with the benefits of hindsight one can project significant further losses in the stock market for 1973 and 1974. With little prospect of a quick reversal of the drop in stock prices there may be fund- ing problems for some state retirement programs. That is, overstating the rate of return estimates earned on the equity portfolios, coupled with severe and pro- longed market drops may lead to funding difficulties for some state retirement systems. More of current tax revenue may have to be diverted to pay retirees their benefits. Conclusions The growth of state retirement funds over the past two decades, even though dramatic, has received very little publicity. The cash and security holdings of these funds have grown from slightly over four billion dollars in 1953 to over sixty billion dollars today. Currently, over seven and one-half million people belong to these retirement systems. As these funds grow, they become a more signifi- cant factor in the money and capital markets of the United States. With sixty billion dollars to invest, 100 what the fund managers do can have noticeable effects on the value of individual securities or groups of corporate stocks and bonds. These funds have also become significant suppliers of funds in the market for mortgages. However, as the funds have grown and become an important factor in the money and capital markets, there has been limited reporting of their portfolio holdings and investment performance. Reports showing the rate of return based on market values of the portfolios are even more scarce. It has been standard operating procedure to merely report the debt issues at par value and the equity holdings at cost. When rate of return is reported, it is generally expressed as the percentage of interest and dividend income to book value or average book value. It ignores any unrealized capital gains or losses which have taken place since the securities were purchased. This may be fine for actuarial reporting but is not for investment reporting. In periods of rising values the returns are understated and in downturns the rate of re- turn is overstated. Consequently, it is recommended that the state retirement fund administrators do a better job of report- ing their investment activities. At a minimum, the reports should include the market value of their security holdings, intereSt and dividend income, contributions 101 to the fund, and any realized gains or losses of secur- ities traded during the year. One could use the Texas Teacher Retirement System annual investment report as an example of outstanding public reporting. Also, the states of Arizona, Arkansas, California, Idaho, New Mexico and Wisconsin have fairly complete annual reports. Within the past few years the states of Ohio, Illinois and Montana have begun publishing detailed investment reports. The above mentioned states all include detailed port- folio holdings in the reports. Unfortunately, other states still View the funds investment activities as being secretive in nature and are hesitant to provide much information about them. It is this type of situation which may eventually lead to forced outside intervention upon the funds. This could be in the form of educational associations, teacher unions or state legislative bodies who may choose to investigate more closely the investment actions of the funds. If one does not think this is a possibility, he need only look at the recent Congressional legisla- tion signed into law on Labor Day 1974 governing private pension plans. It is up to the funds to take the initi- ative in reporting their investment activities or other- wise requirements will be forced upon them. Many states have pointed with pride to the growth in assets of their retirement systems. The impression 102 is given that this growth is often due to the invest- ment performance of the funds. However, in most cases the growth in the fund size is due to increased levels of contributions rather than any significant investment returns. As the number of state employees grows to meet rising demands for government services there are added contributors. Sim- ilarly, as inflation continues, salaries rise too. Since contributions to the funds are based on a per- centage of salaries, the contributions cannot help but increase. Therefore, increased inflow from the members and states rather than investment returns has accounted for most of the growth of the pension funds. This is substantiated in this study which shows the overall rate of return on the forty funds from 1967-1972 was -.24 percent. The fact that the investment return was low for the six year time period covered by this study shows that holding equity issues does not guarantee high rates of return for the funds. Many fund managers were of the opinion that adding equity issues to their portfolios ‘would greatly enhance the rate of return on these funds. Unfortunately, the market downturns of 1966-67 and 1969- 70 showed otherwise. This is not said to encourage the discontinuance of equity investments in the portfolios but only to point out that a different investment strategy may be needed. 103 In the past, formula timing plans and dollar- cost averaging programs were widely used by investors. However, with the sustained growth of the stock market through the 19505 and continuing into the mid 19605 the use of these plans decreased-~particularly the formula timing plans. The reason for the abandonment of formula timing plans was that the market moved above the pro- jected levels of the plan users and stayed there. Con- sequently, it was not thought advisable to continue ‘with the plans.33 Only two states in the sample used in this study indicated a dollar-cost averaging plan was being ‘utilized. In both cases this plan was suspended in 1969-70 as high interest rates in the economy made the "guaranteed" 8-10 percent return on high grade corporate loonds an overwhelming choice over riskier equity issues. However, with the wide market fluctuations during ‘the period of this study, 1967-1972, and with the con- ‘tinued wide fluctuations of the market in 1973 and 1974 (coupled with a highly uncertain future outlook for (duplicating the high sustained growth of the 1950-1965 —_..._._. 33D. H. Bellemore and J. C. Ritchie, Invest- {pgnts Principles/Practices/Analyses (Cincinnat1, gHuxn South-Western Publishing Co., 1974), pp. 113- 17. 104 period, formula timing and dollar—cost averaging programs may become useful again.34 In order to facilitate these plans though, some states will have to pass legislation authorizing the funds to amortize losses on securities sold at a loss over the future rather than requiring the entire loss to be written off in the year it occurs. This will enable them to effectively manage their bond portfolios as well as their equity portfolios. One thing is certain, the funds will continue to have larger sums to invest and increasing the returns eearned on these funds will ease the funding burden on the retirement plans. 34For a study showing the unique reasons for ‘the sustained growth of the 1950-1965 period see: A. E. <3runewa1d and Robert C. Klemkosky, "If You Believe (Erowth is Dead, Try the Formula Timing Plan," SSS Eyusiness Topics, Summer 1972 (E. Lansing, Mich.: Blichigan State University), pp. 59-65. APPENDIX 5 0 l mumwmm moss“ on» NO ON Ommuxm was ucmEumm>cH on one .moumnm cosEOO muH «0 am Ommoxm uo: OMS OcmmEOU madam :H NOmH O>Huoomum .xooum omuummmum 5H am use was cH unusumm>cH mxcmn mmcH>mm now no meow pom xooum GOEEOO cH me OHCHOMHHOO BHHM woo has mo xooum 5H mummmm mo mm can» whoa uoz .Hmmm mOOH>uwm mcHusu HmomHm OCHOOOOOHQ How photomum o3u ummmH um mEOOCH ucoeumm>sH no mo mcoHHMOHMHmmmHO ummann unsoam HmHHOO 0» oouHEHH moan» on» cHnuHs coach on mmmH xooum mo Ommnousm Hmscc< umsE mOHuHHSOOm mumuomuoo o>Huommmo 1 umoo um OOH mmmcmxua cOHumuomuoo so Oocmmm mocon can xooum mumuomuoo one no OuHHsowm cH mummmm Odd mommmuuoe .m.ou .mmumo HOOH coon mo mm :dnu whoa uoz uHmHuuwo umsuu ucmsmHsom O>Huommwm I umoo um woo mcoand mscm>mm mo umconmHe meson pom xooum mumuomuoo 1500 On UONHHOQODM md .mmumnm H m m .moommuuoz uHEHH mmmucmoumm oz mxmde meson 0cm xooum mumuomuoo tam mommmuuos I moosHOcH Hm>oummm Houucoo .mchdmsoo mucousmcH mo venom ou pummnsm ONHH OHummaoo mm meow msHm> xoon um wON MEMQMH4 mucmaumm>cH mooon HmooH pcsm mumum so OGOHuOHHumOH Honuo new oumum .Hmuoomu con» umsuo pOuHuonusm mucmaumo>cH mo menus '1 .l‘ Hmuou mo unmuumm m as meson OuHsUO mo undead ESSmez i'”) ll as... 5.28 35. .5 5.5.5.... 2: .5 835..me H53 .3 .3... 106 xooum moo >56 cH mummmm m.o:5m «0 mm cmgu whoa Hos EuHu 050 was NO xooum OCHocmumuso Hmuou mocon new Rooum mumuomuoo .mmumnm H w w .moumu OHso xooum «O OH cans muoa uoz IHmHuuoo anon» useEQHsom omumHH 1 xoon um am\H1mm mHocHHHH oumom ucmfimuHuom Onu On Oummmmooc md mHsu awe ucmosum uHEHH OOmusoouom oz OQMOH mocon pom xooum mumuomuoo .mmmmmunoz xoon pm wow HHmzmm xooum cOEEOO pom powwow Imam .mocon mumuomuoo mOHGMQEOO mucousmsH OOOH .mmmmouuoa moosHocH meH OHumOEOO mm meow m>Huommmm xoon um «om MHOHOOO mOOH>Hmm mcHumH mocon HmcoHumc 03» umme an an mumuomuoo :H team mo wom ¢¢¢ so «a .d Omumu on song muoE uoc can OHcO umss meson coHumuomnOO mocom HmmHOHcsz moHuon ow3OHHm OOHSUO Oz moHuon venom mo cOHumuomHo uHEHH mmmucmoumm oz oum3mHoo mHsu one pompous uHEHH ommucmouwm Oz usOHuomcCOO ocsm on» «O msHm> mocon pom xooum mum xoon may mo wOm ommuxm luomuoo .mmmmmuuoe mmOSHO voodoo OuHsom OucH IcH 1 mHsu one unmozum mHnHuum>coo no OuHsom oomuOHOO mucoaumm>cH mocon HMOOH team mumum so mcoHuoHHummu umsuo can mumum .Hmumpmm cmnu nonuo omNHuonusm mucwEumm>cH mo mwmaa Hmuou mo ucmuumm a no mocsm OuHsUO mo pandas azfimez 1.5581 2 .3... 107 OCmmEOO 0C0 NCO CH OHHouuuom qufiumm>CH mo OOH mo EdEmez .msmmH OCO OCC CH oHHOmuuom UCOEUmO> ICH OCu mo ON mo ECEmez mumumm Home CH mummmm Hmuou mo am Can» whoa oz .mmHCmmEOO OOCM IHCmCH mm mmHsu memo on pomnnsm mommsousm xooum use» OCO OCH CH mxooum CH woumO>CH OOCOE 36C mo me mo ECEmez zuHusomm COHumuomuoo mCo OCM CH Com um 0Com Hmuou mo wOH mo ECEmez OHCH CME uCoosum mommmuuos .mocon OCm xooum mumuomuoo .mwumo 1HMHuumo unsuu uCoEmHsom mmHCmmEOO moCm nusmCH OHumOEOO mm meow mOHCmmEOO OOCm IHCmCH OHumOEOO mm meow mommmuuoe .mOCon OCH xooum mumuomuoo .mmumnm H w m .mpCOQ OUCOOM no uCoECuO>ow .m.D CH OCCM wCu mo me Con» mmmH uoz CHEHH OOCCCOOHOQ oz OCsm OOOOHQEO How wom .OCsm nonummu How me OCCN mmeOHQEm How owonHm zuHsom oz .OCCM umnomou Mom amN wom OomH m>HuomuMm OOH OCCM mmOOHmEm mumum Mom OmsoHHm OOHCUO oz .OCCH mesons» How OOH OCHmz MCmHmHCOH OxOCuCOM mmmsz m3OH mCmHUCH mquEumo>CH Co mCoHOOHnumwu umCuo mOCon HmooH OCC mumum .Huuooou CmCu MOCuo oouHuonusm muCOEumO>CH mo mamas 0Com Hmuou mo quoumm o no OOCCM OuHCOO mo uCCOEM ECEmez mumum H.U.uCoov HC OHQMB 108 OCHUHH kuuon HO C no mOCon OHHHOQHOU COHHHCOQ cuoo OCO OCH mo xooum OCHHO> mo am mo ESE 1Htz .xOOum OHHHOQHOO OCO OCH CH muwmmH OCCH mo wN\H1H mo ECEHtz .HHOO OCO OCH CH xooum OOHHOHOHQ no COESOO OuCH usm On CHO mummmH HHuou mo am no ECEHtz HHOO Mom xooum CH muommH mo mm mo ECEHtz HOCOQ HOH>HOm OHHme CH OCCH mo wov «0 EsEHtz .mOCon OCOCQOHOH CH OCCH mo amN mo ECEHtz .mOCoo OHOHHHHH CH OCCm «0 wON mo ECEHtz OHCO mxooum OCHOHQ OCOOH>HQ mOOHO uuuoe .mOCon HuHHomuoo mxuoum OCH mOCOQ OHHHOQHOO mOHCHmEoO OOCHHCmCH HHHH OHumOsOO mH wEHm meHQ mmCH>Hm mH mEHm mOHCHmaoo OOCHHsmCH OMHH oHumOEOO mH HEHm szoHHH OHHCUO oz wmv me mxooum OOCHHCHCH OCH xCHn mo me me HmmHmmHmmHz HuomOCCHz CHOHCOHC mqumCCOHmmHz OCHHOHHS muCOEumO>CH Co mCoHHOHHummH umcuo HOCOQ HHOOH OCH OuHum .HHHOOOH CHCu umnuo OONHHOCHCH mquaumO>CH mo momma OCCm HHuou mo uCOOHOm H mH mOCCm OHHCUO mo quOEH ECEHtz mumum ..H.ucooc Ha mHnHe 109 OCHQEOO OCO OCH mo xooum OCHHO> OCH mo OOH mo ECEHtz HOCon HHHHu IHCOCH OCH OuHHHu: OCH HHOHOHHOE .mouHOHmHuHHO Oman» uCHEQHCOH HOOCHOCH mOCon OCH xooum HOCHHsm 1CH OCH OuHHHuC OHHEHHQ OCH .mHmHmuHOE mOOCHOCH OCHQEOO OCO OCH CH mummmH HHuou HO OH HO ECEHtz OOH>uOm OCH IHHH mo mOOHHO OHHHHCO HmonOHn owns» chqu mHHHCm H H m HOOCHOCH mOCon OCH xooum HOHHOQHOO .HHOHOHHOS mOHCHQEOO OOCHHCHCH HHHH OHHHHEOO mH HEHm HOHCHQEOO OOCHHCHCH OHHH OHumOEOO mH HEHm HOCOC OCH xooum OHHHOQHOO .HOOHOHHOS mOCOn OCH xooum HOHHOQHOO .mHOHOuHoz mwuHonHuumo umsuu uCOEmHCoH .HOOHO 1uuos .mOCon HHHHOQHOO mOHCHmEoo OOCHHCHCH OHHH OHumOEOO mH HEHm OCsm mo HHuou mo me meoxm CH0 muCOE 1umm>CH mo mmHHo 0C0 oz Omuummmum OCH COEEOO Cuon How soon um OOH sooum COEEOO How xoon OH «OH OHEHH OOHuCOOHmm oz «ON mom O03OHHH OHHCUH oz uHEHH OOHHCHOHOQ oz OOHtz 3oz OOmHHO 3oz HuHCmmsHm 382 HOH>Hz meHHnHz HCHuCoz HusommHz muCOsumH>CH CO mCoHuOHuummH HOCHO mOCon HHOOH OCH OHHum .HHHOOmm CHCu HHCHO OHNHHOCHCH muCHEumO>CH mo mOmOB OCCm HHuou mo HCOOHOQ H OH mOCsu OuHCoH mo uCCOEH EsaHtz wumum H.O.uCoov H< HHQHB 110 xooum OOHHOMOHQ How OOHHCOOH umOu OOCHCHHm OOH>HOO OCHHHH HHCOHOHC H On HOCOHC HO O OOHHH OOCom COHHHHOQHOO H On OCHOCHHO 1uso OCH OOOOOH mOHHCm mo HOOEOC HHuou OCH «0 Om mo ECEHtz .xooum O.COHuHHomHOO OCO OCH CH OCsm mo ON\H1H m0 ...2H Itz .mOOH>HOm OCHHHH HHCOHHHC o3» umHOH uH mo mOCHuHH umOCOHC OOHCu OCH CHCuH3 mOCom COHuHHom IHOO OCO OCH mo OHHCUO OCHOCHumuoo OCH OOHCOOH HHuou mo Om CHCu OHOE uoz .xooum m.OCHm IEOO OCO OCH CH OCsm HHuou HO OH HO ECEHtz mOCon OCH xooum OuHHomuoo .OOOHOHHOZ mOCon OCH xooum OOHHOQHOO .OOOHmuuoz mOCon OHHHOQHOO .xooum C08 1500 OCH OOHHOHOHQ mom OOHHCUOH umOu mOCHCuHm mOCon OCH xooum OuHhomuoo .OOOHOuuoa mOOsHOCH .meHn mOCH>Hm OH OEHw umoo uH xooum OOHHOH 1OHm OCH Cossoo Mom Omm uHEHH OOHHCOOHOQ oz xooum OOHHOMOHO OCH COEEOO you OmH won oHOO HuoxHO nunoz HOHHOCHO nuuoz xHOO 3Oz muCOEumO>CH CO mCoHuOHHumOH HOCuo mOCon HHOOH OCH OuHum .HHHOOOH CHCu HOCuO OONHHOCHCH muCOEumO>CH mo OOQOB OCsm HHuou mo uCOOHOm H OH OOCsm OHHCOO mo HCCOEH EOEHtz OuHum A.O.uCoov Hd OHDHB 111 mOCsm HHCOCE .mOCon OCH xuoum OuHuom IHOO .mOuHOHMHuHOO umsuu uCOEQHCUm mm mOOCHOCH OCHQSOO OCO OCH mo xooum OCHOCHOOHCO Ho OOCOOH OCH mo ON\H1N CHCu OHOE oz .HHOO OCO CH xooum CH OOumO>CH OC OHE OCCm mo muOmmH HHuOu Ho .xoon. .N H0 asemez OOCOC OCH Cooum OuHuomuoo .OOOHOHHOE mOOCHOCH OOCOC HHOOH OCH OuHum ..m.D CH OmO mo ECEHtz .OOH> IHOO OCHHHH OONHCOOOOH On mxooum OOHHOHOHQ OCH OOCOC mo mCoHuHOHm IHOOHHO OOuCu umOCmHC CH OOm mo ECEHtz .OOOHO 1uuoe umuHm OOOHCHHHCO uCOECHO>oo .m.D CH OCCm mo OmN mo ECEHtz OHCH CHE uCOOsum OCH OCCHC mOCH>Hm OH OEHm OOOHOOHOE OCH OOCOQ OuHHomHOO How 1 OHCH CHE uCOOCHm mH OEHm OHCH CHE CCOOCHO mucon OOHHHus OHHnsm 1 OHCH CHE UCOOCHO HHEHH OOHUCOOCOQ oz NOmH O>Huommmm .H> no COO OCH On OOHCOCH OCH OuHumO HHOH .CCOm CO OCHOH OOHOHHOE CH uCOfiumO>CH OCCH mo uCCOEH HHuou OCH mo OOm OOOOxO 0» OOC Coon OH OOH HOmH 0>Huomwwm OmN OmN HcHHmH mHoCm HHCH>HOOCCOO COOOHO HaoCHHxO muCOEumO>CH Co mCoHuuHuumOH HOCHO mOCOC HHOOH OCH OHHHO .HHHOOOH CHCu uOCuo OONHHOCOCH muCOsumO>CH mo OOQOB OCCm HHuou mo uCOOHOm H OH mOCCu OOHCUO mo uCCOEH ECEHtz mumum A.O.uCooV HG OHQHB 112 OOH>HOm OCO On OCHOHH HOHHOC Ho 1 CH CuHs xuoum OOHHOMOHO .mOOH>HOm HHCOHuHC 03» On OCHuHH HOHCOC Ho 4 mo mOCon OuHHomuoo HOHuCOO 0mm ou COOHCCO COOHO OOHHOMOHQ OCH COEEOO OOHOOOHOOH OHHHCoHqu OOHOCOOH HHCOHOHC 03» umHOH OH HO mCOHuHOHMHmmHHO ummanC mmpnu may cHnqu OOCHOHH OCOC O>HC CODE mCOHuHHomHOO HCHHOHHO Cusom .mOOH>HOm OCHCHH OONHCOOOOH OHHHCOHOHC 03C umHOH OH HO OOCHOHH OOOCOHC 03» OCu CHCOH3 OC CODE OOCOC OuHuomuoo OOCOC OCH xooum OHHHOQHOO mOOsHOCH .mOHCHmEOO OOCHHCOCH OHHH OHOOOEOO OH OEHm mOCon OCH Cooum OUHHOQHOU OOHHCO H a m .mOCOC OuHuomuou OCHH> xoon pH OON OOOOOCCOB OOm HuoxHo Cusom OO3OHHH OHHCUO oz HCHHOHHO Cusom muCOsumO>CH CO OCOHCOHHCOOH HOCHO mOCon HHooH OCH OuHum .HHHOOOM CHCu COCOO OONHHOCuCH muCOBumO>CH mo OOQOB OCCH OuHum HHuou mo uCOOHOm H OH OOCCM OHHCUO mo uCCOEH ECEHtz A.O.HCOOV H4 OHQHB 113 HOHOCOOH OCHHHH 03H OC OOCHHHH umwnOH: Csou cHsuHs HCOO OHHHomHOO .xOOHm OOOHCHHHCO Ho OOHHOH Imam CH Om mo EsstHz OOH>HOO OCHHHH OCO HHHOH HO HO mCoHHHOHMHOOHHO HOO uann HHHOH mg» :HsuHs On HOCE mOCOC OCH COOHm COHHHHomuOO OCO OCH mo COOHO OCHHO> OCH «0 Om mo ECEHXHZ .COHHHHOQ IHOO OCO OCH mo COOHO CH OCsm mo HCOOCV OH HO ECaHtz .mOCon HHm 1HOHCCS OCH HHHOOOH CH OOHOO>CH xoon HH OCsu mo OmN CHCH OOOH Hoz OOHCHQEOO OOCHHCOCH OHHH OHHOOEOO mH OEHm OHsu CHE HCOOCHm OCH OOHCHQEOO OOCHHCmCH OHHH OHHmOEOO OH OEHm mOmHO IHHOE Ham OCH OOCOC OCH xOOHm OHHHOQHOU OOCOC OCH xOOHm OHHHOQHOO OOOCHOCH 1 OHCH CHE HCOOCHm OCHmEOO HOCHH Ho CCHC H mo COOHO OOHHOMOHQ OCH COEEOO How Hmoo HH OON Omm HHEHH OOHHCOOHOQ oz OOHHHHCOOm HHmHOHCCz HCOECHO>OO CH OOHOO>CH mH OO0.000.00mm CHCH mmOH NH OCCH HOCOHOH How Om\H1mm HQOOXO HHEHH OOHHCOOHOQ oz HHCHOHH> HCOEHO> CHHO meOB mHCOEHmO>CH Co mCOHHOHHHmOH HOCHO mOCon HHOOH OCH OHHHm .HHHOOOH CHCH HOCHO OONHHOCHCH OHCOEHOO>CH mo OOQOB OCCH HHHOH mo HCOOHOQ H OH mOCsu OHHCOO mo HCCOBH ECEHCHS OHHHm 2.. see H... 2.5 114 .mHHHonuo OCCm Cqu HUCHOCommmuuoo OCH «mmuHum HHHHCQOMQQH may we manommu HHCCCC “thH ..U.o .CoumCHCmHz .COHuHHOOmmC muumCOCH mmHuHusowm .NHmH .mocsm concmm Hmooq Ham mumum “obma ..u.a .coumcHnmmz .HUHHHEC mo CoHUHHUOHmC mumeHm UCmeumm>CH .oan .mOCCm ConCmm HHooA OCH muHum u mOUHflpOm OCCM quEwuHumu HHuou mo AHCHH> xoonv woo meoxm HOCCHU mOCon muHuomuoo mmmHauuos OCH mOCon OCH xuoum muHHomuoo mwOCHoCH m00H>Hmm mCHuHu HHCOHuHC o3» umeH uH an umuumn no 4C OmuHH ma umsfi mOCOQ wuHuomuoo xooum mCo aCH How OCCM mo umoo uH am mo xCHn HHCoHuHC H how Om>oummH macs» ou OmuoHuummn mOCon muHuomuou moHCHmEoo HUCHquCH HHHH uHummEOO mH mEHm mmmHmuuoe «mm OmHC .mOCon muHuomuoo wuHsUm oz mmHuHHCuwm auHsvm CH :AHHHHBHHQ: Omumm>CH 0n CH0 OCCm auHCCCH HHnHHMH> was .omm Om3oHHH >uHCwm oz mCHEo>3 CHmCoomH3 HHcHouH> ummz ECEHtz .CoHuHComuoo H mmmmmuuoa ‘mmuHo mmHCHmEoo UCmeumw>CH How mCHOCHumuCo mmHHCm IHMHuuwo umsuu uCHEmHCUm OleCmmo OCH .Omuummoum CQEEOU mo mm «o ECEHtz .mOCon OCH xooum muHuomuoo .COEEoo How umou uH wmm CoumCHCmHs mqueumm>CH mOCon HHooH OCCm wuHum Co mCoHuoHuummu umnuo OCH wuHum .HHumOwu CHCu umnuo OHNHCOCCCH mquEumH>CH mo mwmwa HHHOH mo quoumm H mH mOCCm auHCUm mo uCCOEH ECEHxHS ..H.ucoov Hg manna BIBLIOGRAPHY BIBLIOGRAPHY Books Bellemore, Douglas H. and Ritchie, John C. Investments-- Principles, Practices, Analyses. 4th ed. Cincin- nati: South-Western Publishing Company, 1974. Bleakney, Thomas P. Retirement Systems for Public Em lo ees. Homewood, Illinois: Richard D. Irwin, Inc., I972. Davie, Bruce. Investment Practices of Public Employee Retirement Systems. Boston: Federal Reserve Bank of Boston, 1959. Deran, Elizabeth. State and Local Employee Pension Systems. New York: Tax Foundation, Inc., 1969.~ Dietz, Peter O. Pension Funds: Measuring Investment Performance. New York: The Free Press'TA DiVision of the Macmillan CompanY). 1966. Francis, Jack C. and Archer, Stephen H. Portfolio Analysis. Englewood Cliffs, New Jersey: Prentice- Hall, Inc., 1971. Harper, John H. State and Local Pension Funds 1970. Washington, D.C.: Investment Bankers Association of America, 1970. Homer, Sidney and Leibowitz, Martin L. Inside the Yield Book--New Tools for Bond Market Strategy. Englewood Cliffs, New Jersey: Prentice-Hall, Inc., 1972. Latané, Henry A. and Tuttle, Donald L. Security Analysis and Portfolio Management. New York: The RonaId Press Company, 1970. Malca, Edward. Bank-Administered Commingled Pension Funds-—Performance and Characteristic841§62-1975. Lexington, Massachusetts: Lexington Bodks, D. C. Heath and Company, 1973. Melone, Joseph J. and Allen, Everett T., Jr. Pension Planning. Homewood, Illinois: Richard D. Irwin, Inc., 1966. 115 116 Schmid, W. William. Retirement Systems of the American Teacher. New York: Fleet Academic Editions, Inc., 1971. Soldolfsky, Robert M. and Zuber, Ernest V. The Invest- ment Policies of the Iowa Public Employees Retirement System. Iowa City: Bureau of Business and Economic Research, University of Iowa, 1964. Tucker, J. Richard. State and Local Pension Funds 1972. Washington, D.C.: Secufities Industry Association, 1972. Voorheis, Frank L. Bank Administered Pooled Equity Funds for Employee Benefit Plans. East Lansing, Michigan: Bureau of Business and Economic Research, Michigan State University, 1967. Williamson, J. Peter. Investments New AnalytiggTech- nigues. New York: Praeger Publishers, 1970. Articles and Periodicals "A Report on Ten Other State Pension Funds," The Insti- tutional Investor (February 1970), 45-47. "Axe for the Buttonwood?" Barron's (May 7, 1973), 3. "Bond Men Seek a Uniform Index," Business Week (Septem- ber 8, 1973), 65. Cooper, Jack L. "Investment Policies of State and Local Pension Funds," The Southern Journal of Business, Vol. 6 No. 2 (April 1971?, 29-43. "Different Pattern in Pensions," U.S. News & World Report (February 18, 1974), 90-92. Fisher, Lawrence and Lorie, James H. "Rates of Return on Investments in Common Stock: The Year-by-Year Record, 1926-65," Journal of Business, XLI, No. 3 (July 1968), 291-316. . "Some Studies of Variability of Return on Investments in Common Stocks," The Journal of Business, XLIII, No. 2 (April 1970), 99-134. Grunewald, A. E. "Investment Requirements of State and Local Pension Funds," Municipal Finance Officers Association Special Bulletin 1957A (January 16, 1957T1 1’7. 117 . "Investment Policies of Public Pension Funds: Present and Prospective," Municipal Finance Officers Association Special Bulletin 1957C TJune 1, 1957), 1-8. , and Klemkosky, Robert C. "If You Believe Growth is Dead Try the Formula Timing Plan," MSU Business Topics (Summer 1972), 59-65. Higgins, Catherine A. "Calculating Rate of Return," Pensions (July/August 1973), 82-83. Latané, H. A. and Young, W. E. "Test of Portfolio g Building Rules," The Journal of Finance, 24 (Sep- 1 McQuown, J. A. "Technical and Fundamental Analyses and ;_ Capital Market Theory," Journal of Bank Research J (Spring 1973), 8-17. . Patocka, Barbara A. "Public Funds: The Herculean Task is Under Way," Pensions (May/June 1973), 33-48. Schneid, Dan L. "On Measuring Pension Fund Performance," Burroughs Clearing House (April 1971), 22. "Texas State Employees," Wall Street Journal (May 28, 1974). l. ‘ "The Popular Indexes Don't Tell the Whole Story," Business Week (July 7, 1973), 74-77. "The Price of Aluminum Peace," Business Week (February 9, 1974), 44. Thomas, Dana L. "Divide and Multiply," Barron's (November 5, 1973), 5-8. "Treasurer's Report," Barron's (May 14, 1973), 3. "Unmanaging the Portfolio," Business Week (November 17, 1973). 78. "Why the Big Swing to Early Retirement," U.S. News & World Report (May 13, 1974), 59-60. Other Sources Fouse, William L. "Measuring Investment Performance." An address at the 49th National Convention of the Bank Administration Institute. San Francisco, California, October 28-31, 1973. 118 New York State Teachers Retirement Board. Investment Management Performance of the New York State Teachers Retirement System. Albany: January 1964. Ohio Legislative Service Commission. Staff Report No. 58, Major State Retirement Systems. Columbus: February 1963. Peat, Marwick, Mitchell and Company. Management Survey of the State Teachers Retirement System. Sacramento: Joint Legislature Retifement Committee, 1967. Vertin, James R. "Practical Applications of Modern Investment Theory." An address to the Investment Management Program, Stanford University Graduate School of Business. July 26, 1973. United States Government Documents U.S. Department of Commerce. Bureau of the Census. Finances of Employee—Retirement Systems of State and Local Governments in 1964-65. Washington, D.C.: Government Prihting Office, January 1966. . Finances of Employee-Retirement Systems of State and Local Governments in 1965-66. Washington, D.C.: Government Printing Office, January 1967. . Finances of Employee-Retirement Systems of State and Local Governments in 1967+6B. Washington, D.C.: Government Printing Office, February 1969. . Finances of Employee-Retirement Systems of State and Local Governments in 1968-69. Washington, D.C.: Government Printing Office, March 1970. . Finances of Employee-Retirement Systems of State and Local Governments in 1969-70. Washington, D.C.: Government Printing Office, March 1971. . Finances of Employee-Retirement Systems of State and Local Governments in 1970-71. Washington, D.C.: Government Printing Offibe, March 1972. . Finances of Employee-Retirement Systems of State and Local Governments in 1972-73. Washington, D.C.: Government Printing Office, March 1974. 119 . United States Census of Governments: 1967. Vol. 6, No. 2, Employee-Retirement Systems of State and Local Governments. Washington, D.C.: Govern- ment Printing Office, 1968. . United States Census of Governments: 1972. Vol. 6, No. l, Employee-Retirement Systems of State and Local Governments. Washington, D.C.: Government Printing Office, 1973. MTIT/WITHWHENflWIIIfil'flfl('tfllflflfll1fllfllWs 3 1293 0317