RETURNING MATERIALS: IV1ESI_J P1ace in bodk‘arop to LIBRARIES remove this ChECkOUt from “ your record. FINES win be charged if book is returned after the date stamped below. MARKET PERFORMANCE OF THE 5A0 PAULO STOCK EXCHANGE By Jorge Queiroz de Moraes, Jr. 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 1981 F u. a J a C ABSTRACT MARKET PERFORMANCE OF THE sAO PAULO STOCK EXCHANGE By Jorge Queiroz de Moraes, Jr. There were four main Objectives of this research: (1) to build a machine-readable data base of security-price information for securities traded at the Sao Paulo Stock Exchange in the past decade; (2) to analyze the rates of return for common and preferred stocks traded for the period covered by the study; (3) to reproduce the tests of validity of the capital-asset-pricing model developed by Black, Jensen, and Scholes and by Fama and MacBeth with new data and to com- pare and contrast the findings; and (4) to determine the relationship between inflation and returns from risky assets in an indexed economy. Information about 395 stocks was collected from 203 companies for a lO-year period (1970-79). From the comparison between a value-weighted index and an equally weighted index, the conclusion was reached that small com- panies did better than larger ones. The BOVESPA index, despite its construction bias, seemed to be a reasonable representation of the market. Common stocks did better than preferred, as was exepcted, except during the bull market of l97l. An analysis of rates of L w Sit fl)“ r+ Jorge Queiroz de Moraes, Jr. return for 210 investment strategies showed that investment in the stock market was not a good hedge against inflation. The results of the test procedure of Black, Jensen, and Scholes did not show a linear relationship between risk and return. However, because of bias introduced in this test, it was not possible to conclude that the relationship between risk and return was not linear. Nor was it possible, from the results of the procedure of Fama and MacBeth, to reject the hypothesis that the relationship between risk and return was linear. Neither could the hypothesis be rejected that beta was a complete measure of risk. Despite the fact that the coefficients of beta were greater than zero, they were not statistically significant. Therefore, no statistically observable positive relationship between expected real return and risk was found. In addition, no statistically significant relationship was found between real return and anticipated inflation, unanticipated inflation, or total inflation as could be expected in an indexed economy. Copyright by JORGE QUEIROZ DE MORAES, JR. 1981 Those friends who live share with me the happiness of this achievement. Therefore, this work is dedicated to the memory of my friend, Waldemar Rodrigues Alves. ii dcct sert assi and ACKNOWLEDGMENTS I wish to thank Dr. Myles S. Delano, my adviser for my entire doctoral program at Michigan State University and Chairman of my dis- sertation committee. His counseling was always precise, helpful, and full of encouragement. I wish also to express my thanks to Dr. Richard R. Simonds and Dr. Kelly Price, my teachers during the doctoral program and members of my dissertation committee, for their assistance during the deveTOpment of this research. In addition to the members of my committee, many other people have helped me in different ways. I wish to thank each of these people and especially: Dr. Hladimir Antonio Puggina, for his encouragement, support, and friendship; The Fundacfio Getulio Vargas, the Centro de Aperfeicoamento de Pessoal de Nivel Superior, and the Fundacao de Amparo a Pesquisa do Estado de 550 Paulo, for the financial support that I received during my studies in the United States; Francisco Vidal Luna, from the $50 Paulo Stock Exchange, for his encouragement and perceptive help, without which this study could not have been conducted; Adriana Beneducci Costa and Heloisa Helena Albero Nigro, research assistants, for their careful work of data collection; Flavio Marcondes Bojikian and Roberto Lang, research assistants, for tabulating the data; The people of the data-processing department of Fundacao Getfilio Vargas for the computer programs that were indispensable for carrying out this research; My best friend, Alberto José Rodrigues Alves, for his logis- tical support; Jose Luiz Mfiller de Godoy Pereira, for his specific help at a time when I greatly needed it; Zélia Maria Goes, for her careful review of the dissertation proposal; Ann Carroll, for her careful editing, and Sue Cooley, for her precise and beautiful typing; Fernando and Beatriz Luce, for their help and kindness to me and my family during my years in the United States; My sister-in-law, Maria Elvira Barros do Amaral, for her generous friendship during all these years; My parents, Jorge and Aracy Queiroz de Moraes, for the example that their lives have provided to me and for their unfailing support in my life. At the last, my thanks are due to my wife, Maria Regina. Her understanding and joy have been constant throughout the difficulties of this work, and indeed throughout my life. She has been always at my side. For all of this, she will have my eternal gratitude. iv TABLE OF CONTENTS Page LIST OF TABLES ......................... vii LIST OF FIGURES ......................... ix LIST OF APPENDICES ....................... x Chapter I. THE CAPITAL MARKET OF BRAZIL .............. l Historical Background ................. l Reforms of the Capital Market ............. 4 Reform Performance .................. 8 Indexation . . .................... l3 The Stock Exchanges .................. l7 The $50 Paulo Stock Exchange ............. 2l Research Objectives .................. 23 Organization of the Study ............... 25 II. THE DATA FILES ..................... 27 Period Covered in This Study ............. 27 Sample Size ...................... 28 Computation of Monthly Security Returns ........ 29 Inflation ....................... 34 Computation of Real Returns .............. 35 Sources of Information ........... ‘ ..... 35 Limitations on the Data-Collection Procedure ..... 36 Division of the Data Into Sectors ; .......... 37 Number of Stocks and Representativity in the Market . . 37 Information Provided by the Magnetic Tape ....... 39 Final Comments .................... 4l III. REVIEW OF LITERATURE .................. 43 Measuring Market Rates of Return ........... 43 The Value-Weighted Index ............... 44 The Equally Weighted Index .............. 44 Real-Market Returns .................. 45 The Capital-Asset-Pricing Model ............ 46 The Single-Factor Model ............... 46 The Two-Factor Model ................ 51 The Capital-Asset-Pricing Model and Inflation ..... 53 Empirical Tests .................... 55 The Test Developed by G. W. Douglas ......... 55 The Study of J. Lintner .......... ' ..... 58 The Study of Miller and Scholes ........... 60 The Study of Black, Jensen, and Scholes ....... 66 The Study of Fama and MacBeth ............ 72 The Behavior of Other Stock Markets .......... 74 The Critique of Richard Roll ............ 76 Returns From Risky Assets and Inflation ........ 77 Empirical Evidence ................. 77 IV. ANALYSIS OF RATES OF RETURN ............... 80 Why Use Two Deflators? .......... . ..... 80 Market-Performance Analysis .............. Bl The Value-Weighted Index Is for the Total Market . . Bl The Equally Weighted Index I's for the Total Market . 82 The BOVESPA Index .................. 86 The Value-Weighted Indices for Common and Preferred Stocks ................. 95 Sector Analysis ................... 99 Conclusions ............. . ........ ll9 V. CAPITAL-ASSET-PRICING MODEL: EMPIRICAL TESTS ...... l20 The Risk-Free Rate of Interest in Real Terms ..... l20 The CAPM Test Based on the Procedure of Black, Jensen, and Scholes ................ l22 The Time- Series Test of the Model .......... 124 Cross- Sectional Tests of the Model ......... 127 The CAPM Test: Based on the Procedure of Fama and MacBeth. . ................... l29 The Methodology of This Procedure .......... 130 The Results of the Fama-MacBeth Procedure ...... 131 VI. THE RELATIONSHIP BETWEEN INFLATION AND STOCK PRICES . . . T43 Anticipated Inflation ................. l43 Results ..... . . ................. l44 VII. SUMMARY AND CONCLUSIONS ................. l48 Summary ........................ l48 Conclusions ...................... l55 APPENDICES ........................... l57 BIBLIOGRAPHY .......................... l93 Ti J5 (I! LIT Table —l O O O 0 01 b no N O O 0 o.» N -' o o o .p .h h b N N N O O O O O O O b (.0 N —-‘ .b o 01 o 4.9. 5.1. 5.2. LIST OF TABLES Gross National Product .................. General Price-Index Variation .............. General Price-Index Variation .............. Growth in Gross National Product ............. Trade Balance ...................... Volume Traded ...................... Sample of Information Provided on the First Tape ..... Sample of Information Provided in the Second Tape The Value-Weighted Index 15 for the Total Market ..... The Equally Weighted Index I's for the Total Market The BOVESPA Index and the Value-Weighted Indices ..... The Value-Weighted Indices for Common and Preferred Stocks ......................... The Value-Weighted Indices by Sectors: lOOOO Industry . . The Value-Weighted Indices by Sectors: 20000 Trade . . . The Value-Weighted Indices by Sectors: 30000 Services The Value-Weighted Indices by Sectors: Industry, Trade, Services, and Miscellaneous .............. Value-Weighted Index for the Total Market: GPLI Deflated--Semi-Annual Average Rates of Return ..... Risk-Free Rate of Interest--Real Return ......... Time-Series Tests--Statistics .............. vii Page 2 3 l0 ll l3 39 40 42 84 89 93 97 TO] 106 109 Table 5.3. Page Relationship Between Risk and Return: Cross- sectional Test ..................... l29 Portfolio Formation, Estimation, and Testing Periods . . . l32 The Month-by-Month Record Of the Relationship Between Return and Risk .................... l33 The Month-by-Month Record of the Relationship Between Return, Risk, and Average Squared Risk ......... l34 The Month-by-Month Record of the Relationship Between Return, Risk, and Average Standard Deviation ...... l36 The Month-by-Month Record of the Relationship Between Return, Risk, Average Squared Risk, and Average Standard Deviation ................... l38 Tests of the Two-Parameter Model ............. l4l Variations in the GPLI .................. l45 Relationship Between Inflation and Stock Prices ..... l46 viii Figure 1.1. 1.2. 3.1. 5.1. LIST OF FIGURES The National Financial System .............. Organization of the $50 Paulo Stock Exchange ...... The Efficient Frontier, the Investor's Utility Function, and the Capital-Market Line ......... Value-Weighted Index Is for the Total Market ...... Equally Weighted Index I'S of the Total Market: Nominal and Real ................... Value-Weighted and Equally Weighted Indices of the Total Market, GPLI Deflated .............. The BOVESPA and the Value-Weighted Indices for the Total Market .............. . ...... Value-Weighted Indices for Common and Preferred Stocks: Real Terms .................. Value-Weighted Indices for Industry Subsectors: GPLI Deflated . . . .................. Value-Weighted Indices for Trade Subsectors: GPLI Deflated ..................... Value-Weighted Indices for Service Subsectors: GPLI Deflated ..................... Value-Weighted Indices by Sectors: GPLI Deflated Relationship Between Risk and Return: Cross- sectional Test .................... ix Page 24 47 83 87 88 92 96 100 105 108 112 LIST OF APPENDICES Appendix Page A. METHOD OF COMPUTING THE ORTN VALUE ........... 158 B. METHOD OF COMPUTING THE GENERAL PRICE-LEVEL INDEX . . . . l63 C. DIVISIONS AND CODES OF THE SECTORS ' ........... l66 D. TYPES OF STOCKS ..................... l69 E. COMPANIES INCLUDED IN THE TAPE ............. l7l F. INDEXED NATIONAL TREASURY BONDS (ORTN) AND GENERAL PRICE-LEVEL INDEX VALUES ............... l84 G. THE STOCK MARKET IN l97l ................ l88 H. PARTICIPATION OF LARGE COMPANIES IN THE TOTAL VOLUME TRADED ..................... 191 CHAPTER I THE CAPITAL MARKET OF BRAZIL Historical Background The l964 political revolution in Brazil caused deep altera- tions in the economy, particularly in the capital market. In order to give a background for this reorganization, a description of the market in the period before l964 is presented below. The Brazilian economy grew very quickly after the Second World War. From l947 to l956, growth rates of the gross national product were about 6.4 percent per annum, and the industrial sector expanded about 8.2 per- cent per annum. After l956, economic growth accelerated because of a policy of rapid industrialization in the sectors of consumer and capital goods. The gross national product showed a growth rate of more than 8 percent per annum from 1957 to l961, and the industrial sector expanded by more than l0 percent per annum (see Table l.l).1 Although the economy grew rapidly, lack of an adequate structure in the capital market provoked severe distortions. There were no savings-transference mechanisms between savers and investors. Inflation was quite high, though not as high as at the present time lFrancisco Vidal Luna and Thomaz de Aquino Nogueira Neto, "Indexation and Capital Market" (paper presented by the $50 Paulo Stock Exchange to the Fifth General Assembly of the Ibero-American Federation of Stock Exchanges, Buenos Aires, Argentina, 1978). (see Table 1.2). A usury law existed that forbade interest on savings of over 12 percent per annum. Therefore, real interest paid to savers was negative, provoking a severe disincentive to investing funds in fixed-income paper. Table 1.1.--Gross national product. 6 . . Industrial RURAL; WNW (smug 1947 8,047 ... 1948 8,657 7.5 1949 9,237 6.7 ... 1950 9,831 6.4 11.3 1951 10,413 5.9 6.4 1952 11,327 8.8 5.0 1953 11,606 2.5 8.7 1954 12,780 10.1 8.6 1955 13,655 6.8 10.6 1956 14,092 3.2 6.9 1957 15,225 8.0 5.7 1958 16,397 7.7 16.2 1959 17,292 5.4 11.8 1960 18,952 9.6 9.6 1961 20,949 10.5 10.6 1962 22,049 5.2 7.8 1963 22,306 1.2 1.9 1964 23,055 3.3 5.1 SOURCE: Fundacao Getfilio Vargas, Conjuntura EconOmica 27 (December 1973). Table l.2.--Genera1 price-index variation. Year Vargaégon 1947 11.89 1948 7.01 1949 7.09 1950 11.11 1951 16.51 1952 11.81 1953 14.79 1954 26.99 1955 16.42 1956 19.91 1957 14.19 1958 13.03 1959 37.08 1960 29.18 1961 37.05 1962 51.65 1963 75.36 1964 90.50 SOURCE: Fundacao GetGlio Vargas, Conjuntura EconOmica 27 (December 1973). In this period, several ways of avoiding the usury law evolved such as the bill of exchange, which was sold at a discount. The bill of exchange was the sole paper issued by the private sector. It had a medium-term maturity. At this time, some federal and state government papers were issued in the market in a compulsory way since there were no mechanisms for voluntary fund raising.2 The federal government resorted to printing money to finance its long-term investments. In the private sector, long-term financing was done mainly through self-financing (retained profits), rarely through external borrowing. Existing inflationary pressures, aggra— vated by this printing of money, led to a rate of price increases of 51 percent in 1962 and 75 percent in 1963. The average during the 19505 had been less than 20 percent. This inflationary climate, together with a government inclined to the left, created an unbearable political situation. In 1964, the revolution brought into being a new government that proposed to reor- ganize the country, reinstate economic growth, and reduce inflation. Reforms of the Capital Market The most important changes in the capital market occurred under the first government after the 1964 revolution. To create a solid structure for the capital market, it was necessary to provide a real return to savers. Law 4.357 of June 16, 1964, created the indexed national treasury bonds, paper indexed as a function of varia- tions in the currency-purchasing power.3 2Francisco Vidal Luna, "Capital Market and the Brazilian Economy" (paper presented at the $50 Paulo Stock Exchange, 1979). 3As shown later, the indexed national treasury bonds were adjusted according to alterations in the purchasing power of the currency until the mid-19705. From 1975 to 1981, adjustment was substantially lower. Law 4.595 of December 31, 1964, the Bank Reform Law, intro- duced a complete reformulation of the national financial system. This law defined what was meant by the national financial system, specifying its components as follows: Conselho Monetario Nacional (CMN) (National Monetary Council), Banco Central da Repfiblica do Brasil (BACEN) (Brazil Central Bank), Banco do Brasil (Bank of Brazil), Banco Nacional de Desenvolvimento Econ6mico (BNDE) (National Economic Development Bank), and other private and public financial institutions. The National Monetary Council is the primary entity that for- mulates currency and credit policies. No executive function is assigned to the National Monetary Council. It follows directions of the president of Brazil. The Banco Central do Brasil (Central Bank) is linked to the Ministerio da Fazenda (Finance Ministry). It is an executive entity and is responsible for fulfilling its legal requirements and the rules issued by the National Monetary Council. Among its functions are tO issue currency, to control foreign capital, to control financial institutions, to buy and sell federal paper, and to represent Brazil with foreign financial institutions. The Banco do Brazil (Bank of Brazil) has a double role. First, it is the federal government's finance agency and primary undertaker of the industrial and rural credit policies. Second, it acts as a commercial bank. The Banco Nacional de Desenvolvimento Econ6mico (National Economic Development Bank) is a public corporation linked to the presidents' planning department (Secretaria de Planejamento da Presidencia da Republica). It is the primary institution that carries out the federal government's investment policies. It offers support to strategic investment necessary to the country's development, par- ticularly to strengthening national private companies.4 The Capital Market Law (Law 4.728 of July 16, 1965), which assigned to the National Monetary Council the regulation of the capi- tal market and to the Central Bank its control, is another fundamen- tal part of the organization of the financial system. This law created new agents, organized the capital-market distribution system, and determined the institutions that could act in this market. The market-distribution system was carried out by the stock exchanges, brokerage companies, over-the-counter dealers (distribuidoras), and investment banks. Some of these institutions already existed at that time, although in other forms. Up to that time, the stock markets were public institutions. Through this law, they became private institutions belonging to the recently created brokerage companies. Until that time there were only public-funds brokers, individuals who represented the public. According to this new law, the public- funds brokers had the right to turn themselves into brokerage com- panies (corporations or partnerships). Other people were also allowed to constitute brokerage companies. One hundred thirty-four brokers' companies were created. 4For more details regarding the functions of the national finance system components, see Introducfio ao Mercado de Acoes [Introduction to the Stock Market] (Comissao Nacional de Bolsas de Valores [National Commission of Stock Exchanges], 1979). The main activities of these companies are to operate exclusively at the stock exchange of which they are members; to buy, sell, and dis- tribute stocks and shares for third parties; to make new share issues; to administer investment portfolios; to undertake custody of invest- ment paper; and to deal in the money market. The other two institutions created by Law 4.728 are the investment banks and over-the-counter dealers (distribuidoras). Investment banks are private institutions specializing in equity or medium- or long-term financing through the investment of their own resources. They also specialize in raising, intermediation, and investment of third-party funds. Over-the-counter dealers (distribui- doras) are firms organized as corporations or limited-liability companies, or even as single-name firms whose authority to operate is given by the Central Bank. Their basic activities were determined by Act 76 of November 22, 1967, issued by the Central Bank. These activi- ties include subscribing and intermediating placement of market issues, selling third-party paper, and dealing in the money market. Over and above these alterations, an attempt was made to provide investments for the housing program. The Sistema Financeiro da Habitacfio (Housing Finance System) was started by Law 4.380 of August 1964. The system gave functions to the Banco Nacional da Habitacao (BNH) (National Housing Bank), as the housing system's central bank. A real estate bill was issued that set up means for raising voluntary funds for the housing system. These instruments were indexed in accordance with the rate of inflation to provide a real return on capital. Funds were also raised through savings accounts, which were regulated in 1968. These savings accounts are issued by representatives of the Sistema Financeiro da Habitacfio (Housing Finance System), which consists of savings institutions, savings and loan associations, and real estate credit societies. Thereafter, the Central Bank regulated the recently created capital market. In 1976, the ComissSo de Valores Mobiliérios (CVM) (Stock Exchange Commission) was created. It undertook most of the regulat- ing functions of the stock market. The functions of the stock market are discussed in detail in later sections. Thus the Brazilian capital market is made up of two sub- systems: the normative subsystem composed of the National Monetary Council, the Stock Exchange Commission, the Central Bank, the Banco do Brasil, the National Economic Development Bank, and the National Housing Bank; and the operating subsystem composed of the financial institutions. The present organization of the national financial system is presented in Figure 1.1. Reform Performance As previously mentioned, the government that came to power in 1964 proposed to introduce an economic policy that would promote a decrease in inflationary rates and a return to economic growth. In the first years after 1964, inflation was reduced at the cost of a serious slowdown in economic growth. Although the inflation rate was reduced to levels lower than 40 percent per annum in 1965 and 1966 (see Table 1.3), the gross national product increased only 2.7 percent in 1965 and 3.8 percent in 1966 (see Table 1.4). During the period I. v REGULATING SUBSYSTEM OPERATING SUBSYSTEM NATIONAL FINANCIAL SYSTEM BANKING COMMERCIAL BANKSs SAVINGS BANKS NATIONAL MONETARY COUNCIL (CMN) SECURITIES COMMISSION CENTRAL BANK BRAZILIAN BANK NATIONAL DEVELOPMENT NATIONAL HOUSING BANK BANK T“ FINANCIAL INSTITUTIONS NONBANKING INVESTMENT BANKS, DEVELOP- MENT BANKS, DEVELOPMENT CORPORATIONS, FINANCE COMPANIES, REAL ESTATE CREDIT COMPANIES, SAVINGS ASSOCIATIONS, INSURANCE COMPANIES ADDITIONAL INSTITUTIONS: STOCK EXCHANGE-BROKERAGE COMPANIES, OVER-THE-COUNTER DEALERS Figure l.l.--The national financial system. (From Introducfio ao Mercado de AcOes [Introduction to the Stock Market], 2nd ed. (ComisEEO Nacional de Bolsas de Valores [National Commission of Stock Exchanges], 1980),;L 35.) 10 1964-66, the industrial sector was severely affected by the slowdown of activities. Therefore, in 1967 an industrial crisis of large pro- portions occurred. Table 1.3.--General price-index variation. Year Varzaégon 1964 90.5 1965 56.8 1966 38.0 1967 28.2 1968 24.2 1969 20.7 1970 19.8 1971 20.4 1972 17.0 1973 15.4 1974 28.7 1975 27.7 1976 41.3 1977 42.7 1978 40.8 1979 77.2 1980 110.2 SOURCE: Fundacfio Getulio Vargas, Conjuntura Econ6mica, March 1981. 11 Table 1.4.--Growth in gross national product. Year ngwég 1964 3.3 1965 2.7 1966 3.8 1967 4.8 1968 11.2 1969 10.0 1970 8.8 1971 13.3 1972 11.7 1973 14.0 1974 9.8 1975 5.6 1976 9.2 1977 4.6 1978 6.0 1979 6.4 1980 N/A SOURCE: Fundacfio Getdlio Vargas, Conjuntura Econ6mica, March 1981. In 1967, the revolution's second government came to power. It tried to stimulate the economy; reduction of inflation was a sec- ondary target. Several fiscal incentives were developed, among them export incentives that included the use of a flexible exchange rate.5 A price-control system was established through the organization of the 5The flexible exchange rate had as its objective cruzeiro devaluations occurring at shorter periods of time. 12 Conselho Interministerial de Precos (CIP) (Interministerial Price Council). An energetic public investment plan was started. The economic policy during 1967-68 aimed to create an effec— tive demand as much through public investment as through renewal of private investment and export growth. Generally speaking, this econ- omic policy marked the beginning of a period of fast economic growth that continued to 1973. During the period 1967-73, the gross national product grew at an average annual rate of 10.5 percent. Inflationary rates were gradually Contained, decreasing from 28.2 percent in 1967 to 15.4 percent in 1973 (see Tables 1.3 and 1.4). Exports went from US$1.7 billion in 1976 to US$6.2 billion in 1973, with favorable trade balances during five of these seven years (see Table 1.5). The strong external pressures to increase prices, particu- larly because of the increase in Oil prices, caused a return to the growth in inflation (see Table 1.3). By 1975, inflation was already 28.7 percent; it reached 110 percent in 1980. Together with this growth in inflation, Brazil was faced with a chronic deficit in its trade balance, particularly due to increases in the price of oil (see Table 1.5). After 1973, economic growth continued, but at a slower rate (see Table 1.4). 13 Table 1.5.--Trade balance (in US$ billion). Year Exports Imports Balance 1964 1.430 1.086 344 1965 1.596 941 655 1966 1.741 1.303 438 1967 1.654 1.441 213 1968 1.881 1.855 26 1969 2.311 1.993 318 1970 2.739 2.507 232 1971 2.904 3.245 -341 1972 3.991 4.235 -244 1973 6.199 6.192 7 1974 7.951 12.641 -4.690 1975 8.670 12.210 -3.540 1976 10.128 12.348 -2.218 1977 9.264 9.016 248 1978 12.659 13.683 -l.024 1979 15.244 17.961 -2.717 1980 20.132 22.960 -2.828 SOURCE: Central Bank, Annual Report, 1980. Indexation Inflation and the orderly development of the capital market are said to be mutually exclusive phenomena. Inflation destroys confidence in economic agents and in the economy's future performance. Brazilian experience in the period prior to 1964 seems to confirm this statement. At that time, the capital market in Brazil did not go beyond short-term operations in the commercial bank area and an 14 atrophied long-term market. However, the Brazilian experience after 1964 seems to show that high inflation rates may be associated with an orderly development of the capital market. The compensating fac- tor seems to be indexation. The origin of indexation was in Law 4.357 of 1964, which, as mentioned above, created the indexed national treasury bonds and, a posteriori, institutionalized indexation. This same law applied the indexation principle to overdue fiscal debts and the correction of companies' fixed assets. Afterwards, through Law 4.380 of 1964, which created the Banco Nacional da Habitacéo (National Housing Bank), the concept of indexation, a posteriori, was extended to loans pro- vided by the national housing system. This principle was then extended to several other sectors including savings, loans, debts, rents, insurance, interest, wages, the real estate sector, and the exchange rate. Parallel to the above, a comprehensive system to fight high prices was organized based on the same indexation principle. [From a financial point of view, indexation makes the economy extremely flex- ible. However, on the productive side, when a large variety of goods and services have their prices indexed, these prices cease to reflect relative scarcity and may jeopardize the effective allocation of economic resources. In spite of the authorities' initial efforts, the private financial institutions did not work with indexed securities because of the risk involved, as when, expecting an increase in the t1 15 inflationary rate, borrowers and investors turn to different sectors.6 0n the part of investors, there is a massive transference of resources from nonindexed monetary or financial assets to indexed assets. 0n the other hand, borrowers try to obtain nonindexed liabilities. To overcome this situation, prefixed monetary correction was created, which is in reality a nominal interest rate. It is adminis- tered by the Central Bank. This monetary correction enabled the formation of a wide capital market; however, its development was not independent of inflation. Thus it is necessary to clarify the rela- tionship between the capital market and inflation. While inflation was decreasing in the country from 1964 to 1973, the mechanism of indexation accomplished its aim, making possible the compatibility of inflation with the capital market. The relationship between monetary and nonmonetary holdings, which was 2.7 in 1968, decreased to 0.76 in 1973. The majority of the nonmonetary holdings exhibited exceptional real growth in that period as did bills of exchange, fixed-term deposits, and savings accounts.7 However, after 1973, when inflation began to increase, savers began a massive transference of funds from nonindexed assets to those that included monetary correction. Initially, investors tried to reduce part of their balances in monetary assets, particularly in 6Paul Beckerman, "The Trouble With Index—Linking: Notes on the Recent Brazilian Experience" (1978). (Mimeographed.) 7Luna and Nogueira Neto, Indexation and Capital Market. RTE 16 sight deposits, provoking a liquidity crisis in the banking system. They also transferred their funds from financial assets that were paid at a nominal prefixed rate, such as exchange bills and term deposits, to indexed assets. Until 1973, those assets with no indexation or with prefixed indexation were remunerated at a higher rate than those assets with a posteriori indexation. This situation changed in 1974 when infla- tion began to increase, a tendency that was reinforced by regulations that, at the time, established ceilings for the remuneration of those assets with no a posteriori indexation. Consequent to this movement of funds, which was already being felt by 1974, the private financial sector, which had not used indexa- tion to any great extent, went into a severe liquidity crisis. As a result, several financial institutions went bankrupt. A similar crisis was felt by those companies that had borrowed indexed funds. The increase in the inflation rates was reflected by an increase in the companies' liabilities at a time when the funds were locked into investments. The companies were unable to generate cash in the same proportion as the increase in their corrected borrowings, and finan- cial problems arose. To reduce the inflationary impact on borrowers' loans and to make savings in indexed assets less attractive, the for- mula used for the calculation of the index was modified so that only part of the effective inflation rate was indexed. 17 The state assumed control of a growing share of internal savings and the compulsory savings that it administers.8 0n the other hand, the state instituted a generalized system of subsidized credit through which it loaned the funds obtained. The Brazilian economy is thus in a situation in which the state controls an extremely high prOportion of financial savings, taking in the funds through compulsory savings schemes or through assets yielding interest and subject to monetary adjustment. These assets offer some of the high- est returns in the capital market with no risk and excellent liquidity. On the other hand, the state offers extensive credit lines at subsi- dized interest rates. In this way, the state operates with a negative spread, provoking income redistribution among various segments of society and creating serious distortions in the economy. The private financial sector remains unstable with funds being switched rapidly between indexed and nonindexed assets. As a result of inflation and government-guaranteed indexed assets, the activity of the equity mar- ket decreased because of lack of available voluntary funds. The Stock Exchanges The development of the Brazilian stock market is closely linked to the national financial system that was implemented in 1964. To understand how the stock market operates, it is necessary to 8The two largest sources of compulsory savings administered by the state are Fundo de Garantia por Tempo de Servico (FGTS) (Fund to Guarantee the Time of Work), a kind of unemployment insurance, through which every company is required to deposit monthly 8 percent of its wage bill, and the Programa de Integracfio Social (PIS) (Program of Social Integration), through which companies have to deposit 0.5 percent of their sales. 18 clarify certain factors. In terms of funding, the state agencies hold more than 60 percent of the domestic financial savings, of which a substantial part are compulsory. Thus the state agencies have much influence in the application of these funds. The agencies have, on the whole, invested the funds in loans to the private sector through an ample and intricate system of selective credit, usually at subsi- dized rates.9 In this way, the most important financial agents work with a negative spread that imposes limits on the growth of the stock market. In terms of demand for stocks, it is clear that investors usually prefer an investment with a real return and with low risk compared to one with high risk and poor liquidity. 0n the supply side, a company usually prefers to go into debt at subsidized interest rather than offer an issue of new shares to the market. There are other factors that impair the growth of the stock market. The larger national private companies, which should provide the rationale for the stock markets, are often family owned and generally look to the state for funds and guidance in their investment programs. 0n the other hand, the Brazilian industrial sector is domi- nated by state and multinational companies. One asks oneself how it is, then, that the stock market has remained alive; the answer, once again, lies with the state. The most important incentive to the equity market arose as a result of Law 157 of 1967, which instituted the so-called fiscal 9In 1980 the government sought a return to economic reality and substantially reduced its loans at subsidized rates. u. +' :i:'J.'J COJS this inve AER 1111;. to to Zac R Ger! van 19 mutual funds. Law 157 came at a time of extreme corporate difficulty caused by the recession through which the country was going. With this new law, the government attempted to channel funds to companies as well as to develop among individuals the habit of making equity investments, at that time an unusual choice of investment. The mechanism of the fiscal mutual funds in their original 10 form can be explained as follows. The private, individual taxpayer was allowed to put 10 percent of his income-tax bill toward the pur- chase of quotas in the fiscal funds (sometimes called 157 funds). Corporate taxpayers were also, initially, benefited by a 5 percent reduction in their income-tax bills through a similar investment plan. Shortly afterward, this benefit was denied to corporate tax- payers. Those financial institutions authorized tO deal in the stock market were able to create fiscal mutual funds and sell quotas, plac- ing the funds raised in diversified stock portfolios. The shares purchased had to come from the primary market, in other words, from new stock issues. Companies, to obtain the funds raised, had to invest them in working capital and thus to improve the ratio of their equity to their total liabilities. Thus the basic objective of Law 157, in its original form, was to strengthen the capital structure of companies, directing the funds to the primary market. This philosophy suffered innumerable changes 10Francisco Vidal Luna, "E1 Papel de la Bolsa en la Capitali- zacion de la Empresa Privada: E1 Caso Brasileiro" [The Effect of the Stock Exchange on Capitalization of Private Enterprise: The Brazilian Case] (paper presented by the $50 Paulo Stock Exchange at the Seventh General Assembly of the Ibero-American Federation of Stock Exchanges, Santiago, Chile, 1980). 20 over the years. The secondary market was benefited when the fiscal funds were allowed to acquire part of the shares in their portfolios through the secondary market by buying shares through the stock exchanges. The companies authorized to raise equity through the fis- cal funds have also varied over the years. It is only since 1974 that the fiscal funds achieved a relative stability in terms of their objectives. These objectives are strengthening national private companies, developing the equity market, and educating individuals to invest in the stock market. At present, individual taxpayers may deduct from 12 to 24 percent of their tax bills (depending on their gross incomes) and invest this money in the purchase of fiscal-fund quotas. The adminis- tration of these funds may be carried out by banks or brokerage com- panies. The proceeds obtained by the funds are invested in diversified portfolios that must obey rigorous norms: 1. Eighty percent of the value of the fund must be invested in national listed companies. 2. The total invested in any one company may not exceed 4 percent of the fund's worth. 3. No more than 10 percent of the voting stock of any com- pany may be held by the fund. Taxpayers may only realize their investment after five years. At this time, 50 percent may be sold; after the sixth year, the bal- ance may be sold. The fiscal funds, by reason of the volume and constancy of the resources that they operate, are among the most important investors in SIC cie S; (7 21 in the equity market. However, one of the most serious problems of the fiscal-fund system is the concentration of a large volume of stocks with relatively few institutions. This can impair the effi- ciency of the market. Other important institutional investors in the market are insurance companies, which are required by specific legislation to maintain part of their reserves invested in the stock market; private 1] and mutual funds. pension funds; Thus the Brazilian equity market is of recent origin and as such is still maturing. The $50 Paulo Stock Exchange This research is specifically concerned with the behavior of share prices on the S50 Paulo Stock Exchange. Thus it is necessary to describe the characteristics of this institution. As a result of the initiative of Emilio Rangel Pestana on August 23, 1890, a Free Exchange was created in Sao Paulo for the negotiation of stocks and other papers by public auction following the standards of the European exchanges. At that time, the town of S50 Paulo was of little significance within the young republic and was far from having the economic activity capable of sustaining a market for stocks. However, a number of busi- nessmen, with the lucidity and vision characteristic of all pioneers, foresaw the future development of the city that is today the country's 1]These funds normally complement the government-benefits scheme available to all workers. 22 economic capital. Against incomprehension and indifference, they maintained their efforts to convince the state government to give official recognition to the embryonic Free Exchange. On January 25, 1895, the city's anniversary, the Syndicate Of Brokers of Sao Paulo was formed with the support of the Commercial Association. In the same year the Free Exchange was transformed, by an act of the state government, into the Official Exchange for Bills in Sao Paulo, which was to operate in accordance with public laws and regulations. At the start of this century, due to the development of the growth of coffee, Sao Paulo began to appear on the economic map of the country. The exchange kept pace with this growth, raising funds for new enterprises and gradually transforming its own administrative and functional roles. In 1935, the exchange, renamed the Official Stock Exchange of $50 Paulo, was considered a semiautonomous entity, directly subordinate to the SEO Paulo State Secretary for Finances. The exchange remained unchanged until new legislation was passed to regulate the capital market and to provide for its develop- ment. On March 7, 1967, by an Extraordinary General Meeting, the Official Stock Exchange became the present Bolsa de Valores de sac Paulo, a civil, nonprofit association. The stock exchange has over one hundred brokerage companies as members, who, in the form of a General Assembly of Associates, constitute the exchange's highest deliberative authority. The Admin- istrative Council, responsible for operational and social policies, consists of eight officials; six of these are elected from among the info the l 5100 by tr 23 owners or directors of member brokerage companies, one represents the companies whose stocks are registered in the exchange, and one is the general superintendent. The president of the council is chosen from among the repre— sentatives of the brokerage companies and has the support of techni- cal and communication assessors who deal with the press and publish information to the public as a whole. The general superintendent, who is named by the Administrative Council, is responsible to the council for the implementation of the policies established by it. The general superintendency has reporting to it the executive super- intendencies: operations, systems, administration, and finance. (See Figure 1.2.) Research Objectives The objectives of this research are as follows: 1. To build a machine-readable data base of security-price information for securities traded at the $50 Paulo Stock Exchange in the past decade; 2. To analyze the rates of return for common and preferred stocks traded at the $50 Paulo Stock Exchange for the period covered by the study; 3. To reproduce the tests developed by Black, Jensen, and Scholes and by Fama and MacBeth with a new data base and to compare and contrast the findings;12 12F. Black, M. C. Jensen, and M. Scholes, "The Capital Asset Pricing Model: Some Empirical Tests," in Studies in the Theory of Capital Markets, ed. Michael Jensen (New York: Praeger, 1972); ‘ > ‘__v.~ ‘4‘..< .<: :L :‘\ 24 .wmcmsoxm xuopm opzma omm ecu mo cowam~wcemsoiu.m.p mszmwu Lounge: Lounge: .mccomgma : a o 4 E Lounge: Lounge: gmumcmz Leanne: seamen: zovumeummcwsu< oucacPJ :owuuauosa weavcsuwh memumam _ _ _ _ uuzm ">uzmozmhznmma2m m>~haumxm T ">uzmozuhz~mum3m m>~h2umxm _ mach~o=< +1 Lamaze: Loosen: Cowman: mcomuugono umxgm: mmwccasou _ _ alilll mzo_puzmozmhzumuaam m>_h=umxm L hzuzwozmpz~mmaam A_h4mzmmm< 44 The 33 or Na = Nde (Fr + FS - l) (11) where: Na = New number of shares after events Nb = Number of shares before events Fd’ Fr’ Fs = Adjustment factors already defined However, there were cases in which a stock dividend or stock split was given over the resulting amount of equity after subscrip- tion. So, if all the events were on a different basis, the adjust- ment was: N6 = Nb'Fd'Fr'Fs (12) After the computation of these adjustments, it was possible to determine the index for each security. As in Equation 1, N = 1000 0 PO Therefore, the starting index was expressed as: NOPo = 1000 = ID (13) where: N0 = Number of shares at the beginning of the first month for which the security return was computed P0 = Price at the beginning of the first month Io = Starting index The index at the beginning of each month was expressed as: It = NtPt (l4) 34 where: It = Index at the beginning of month t Nt = Number of shares (after adjustments) at the beginning of month t Pt = Price of share at the beginning of month t ('1' 1| (1.2....120) = number of months Finally, the monthly returns were computed as: I. - I. Rjt = ltI Jt-‘ ('5) jt-l where: Rjt = Return on security j during month t Ijt-l = Index of security j at time t-l Ijt = Index of security j at time t Inflation The returns computed according to Equation 15 were nominal rates of return instead of real rates of return. Since inflation in Brazil was very high in the past decade, it was important to know the real rates of return that occurred during the period. It was necessary to select a good deflator. For several reasons, which are discussed in Chapter IV, two deflators were chosen. The first was the General Price-Level Index (GPLI) published monthly by the Getfilio Vargas Foundation. It is widely used and is the least—questionable deflator. The second deflator was the monthly variation in the value of the ObrigacOes Reajustaveis do Tesouro Nacional (ORTN) (Indexed National Treasury Bonds), which is an offi- cial index and is largely used as the basis for indexation Of the Brazilian economy. 35 The method for fixing the values of the ORTN is presented in Appendix A. The method of computing the GPLI is presented in Appen- dix B. Computation of Real Returns Through use of two deflators, two sets of real returns for each stock were obtained. These real returns were computed as follows: 1. The inflated index was computed as: = 9121 - pt IIt It __TEF—__' (16) where: IIt = Inflated index of time t to January 1, 1980 It = Stock index computed according to Equations 13 and 15 0121 = Value of the deflator on January 1, 1980 pt = Value of the same deflator at time t 2. The real return was expressed as: II. - II. RR.t = JEI Jt" (17) J jt-l where: RR. = Real return of security j in period t jt Sources of Information The data were obtained from the following sources: 1. The $50 Paulo Stock Exchange The $50 Paulo Stock Exchange has a relatively well-organized file with information about all securities traded. Since 1972. a 36 great amount of information has been recorded on magnetic tapes. Unfortunately, not all of those files are complete; thus complemen— tary material from other sources was used. In addition, several pub- lications of the $50 Paulo Stock Exchange were consulted.5 2. IBEMEC Publications The publications of the Instituto Brasileiro de Mercado de Capitais (Brazilian Institute on the Capital Market) were used to supplement the data. 3. S.N. Publications Organizacfio S.N. Consultores Financeiros (S.N. Financial Consultants) is a private statistical-information system that pub— lishes analyses of companies in the market. 4. Others In a few cases, it was necessary to consult other sources of information such as records of stockholders' meetings or stockholders' departments. Limitations on the Data-Collection Procedure Some variables that should be taken into consideration were neglected because of practical problems. These variables were as follows: 5The most important publications systematically consulted were Anuario da Bolsa (Annual Bulletin of the Stock Exchange) and Boletim Diario de InformacOes (Daily Information Bulletin). a“? J":- 37 1. Commissions No commission for buying or selling any stock in the market was considered. This was because the value of the commission is a function of the amount traded. 2. Income tax Income taxes were not considered because of the enormous complexity of the legislation in this area. In addition, it was impossible to determine a typical investor and to compute his typi- cal income tax. 3. Tax incentives The Brazilian government has designed a very complex system of tax incentives by which investors may save income taxes by invest- ing in stocks. Again, it was impossible to determine the average investor; thus it was necessary to omit the benefits from the incen- tives. Division of the Data Into Sectors The companies were classified by sectors. Although any sector classification is, to some extent, arbitrary and subject to criticism, to facilitate the work the classification system used by the $50 Paulo Stock Exchange was used. The divisions and codes of the sectors are presented in Appendix C. Number of Stocks and Representativity in the Market Information about 395 stocks was collected from 203 companies. At this point, it is important to mention that normally each company 38 has at least two types of stocks being traded: common and preferred. In the United States, preferred stocks are closer to debt. In con- trast, in Brazil, preferred stocks are similar to common stocks, with the following basic differences: 1. Preferred stocks do not carry voting rights; 2. Preferred stocks bring the minimum dividend stated; 3. Preferred stocks have priority of reimbursement in case of liquidation. On the other hand, preferred stocks share equally with common stocks in all the profits of the companies. In this sense, it is logical to suppose that preferred stocks carry less risk than common stocks. It is also possible that the company might have more than one type of preferred stocks. In this case, the basic difference between them is generally the stated dividend. The tape codes used to iden- tify the types of stock are presented in Appendix D. The stocks included in the data file represented about 90 per- cent of the volume (measured in cruzeiros) traded on the SEO Paulo Stock Exchange. The total volume traded, the volume included in the tape, and the percentage of the total volume that is included in the tape for each of the past ten years are shown in Table 2.1. All of the companies included in the tape are listed in Appendix E. 39 Table 2.1.--Volume traded. Total Volume Volume on the Tape Percentage of Year Cr$ x 1000 Cr$ x 1000 the Total 1970 1,430,414 1,316,805 94 1971 9,110,730 8,316,338 91 1972 9,741,488 8,218,340 84 1973 9,086,236 7,815,188 86 1974 5,574,355 5,223,067 94 1975 9,248,010 8,187,748 89 1976 9,329,982 8,335,246 89 1977 12,744,707 11,757,107 92 1978 32,770,861 28,439,769 87 1979 32,863,168 29,323,889 89 Average for the entire period = 89.3% Information Provided by the Magnetic Tape_ Two tapes were available. The first gave basic information, with no further computations. This information was as follows: 1. 01 h on N I o o a Code of the company Sector classification of the company Name of the company Type of security First date: for shares traded before January 1, 1970, this date was January 1, 1970; for shares initially traded after this date, the first day of the month fol- lowing the first trading was used Information date Period number Quantity of stocks held 40 9. Stock price 10. Total number of stocks of the company in the market As a sample, the information for one company provided by the first tape as printed by the computer is presented in Table 2.2. Table 2.2.-~Samp1e of information provided on the first tape. Name-~ACESITA Number--001 Sector--13072 Type--02 23:2: 532:2? Tgizlkluibiaoa" 1/1/70 1 1.05 952.3810 194,495 1/2/70 2 0.98 952.3810 194,495 1/3/70 3 1.30 952.3810 194,495 1/4/70 4 1.17 952.3810 194,495 1/5/70 5 1.17 952.3810 194,495 1/6/70 6 0.95 952.3810 194,495 1/7/70 7 0.95 952.3810 194,495 1/8/70 8 0.99 952.3810 194,495 1/6/79 114 1.20 4562.9485 2644.990 1/7/79 115 1.03 5183.1551 2644.990 1/8/79 116 0.84 5183.1551 2644.990 1/9/79 117 1.03 5183.1551 2644.990 l/lO/79 118 1.30 5183.1551 2644.990 1/11/79 119 1.40 5183.1551 2644.990 1/12/79 120 1.20 5183.1551 2644.990 1/1/80 121 1.10 5183.1551 2644.990 41 In addition to the information available on the first tape, the second tape provided the following: the nominal wealth, the nomi- nal monthly return, the real wealth assuming the ORTN as the proper deflator, the real monthly return assuming the ORTN as the proper deflator, the real wealth assuming the GPLI as the proper deflator, and the real monthly return assuming the GPLI as the proper deflator. As a sample. the information for one company provided by the second tape as printed by the computer is presented in Table 2.3. Final Comments Four persons worked on a full-time basis during nine months in a special room at the SEO Paulo Stock Exchange to collect the data. In addition, the people from Getfilio Vargas Foundation Data Processing Department worked three months to process the data and to put them in the proper form. Now, the information is available to the public. The writer expects to update these tapes annually. 42 .np cayenne“ o» mcwugouuo emcwmuao use: Hana agape; pew; can zhmo cgzumg Fama .u x_ucmaa< cm umucmmmsn use mmzpm> mace can zbmo expo c .op copuoaam o» ucwugouuo umcwmuno mew: ~48o gapemz peas use zhao zupumz _ommu .upw; xooum on» an woven xuoum oz» mcwxpavu_=e an vwcvuuao no: cupmwz poemsozm .mp copumzam a? umumuwucw me kusasoo we: genome FucwEoz a <\z mome._on <\z coke._o~m oea.eee~ <\z ocue._oam _mm_.mm_m op.m _~_ om\_\_ mem_.o- Ame“.moee Nap_.o- “mom.m~ee oom.eee~ wao.o- .ews.m_~e mep.mmpm o~._ om, m~\~_\_ oe~o.o mweo.-Nm mafip.o- _-~.mmmu omm.eee~ Nme_.c- p~_e.emmfi _mmp.mm_m oe.. o_p ah\.P\_ cewo.o Neme.eo_m ~o~o.o oooo.mee~ oaa.eeem aeko.o _ecp.mMNe pmmp.mm_m cm._ mPF a~\o_\_ . . . . . . . . . . . . . . . . . . . . . . empo.o- mmeo.sommp oooo.o- ammm.me_~_ mae.eop oooo.o mmm~.e_P_ o_mm.~mm ~_.F e oh\e\_ _¢o_.o memo.mmo_~ cop—.o- Ammm.m~8m_ mae.emp memo.o- mmao.mm~p cpmm.~ma om._ m o~\M\F mmom.o epm..mo_op eoom.o meap.m_mop moe.emp meNm.o emmm.mmm o_mm.~mo mm.o N o~\~\_ hauo.c- eeam..oe~_ ammo.o- wwoo.a_m._ moe.ea_ eee¢.o- oooo.ooop o_wm.~mm mo.P _ cN\p\_ eeswwAaw _3au ezhao ezpmo coo. x axeeum seesaw“ e;u_eez ape: spate ee.eea memo a .eea eupeez poem eezuem .eem eepeez .eez Ce Leasez .euee .ee_eez .eeeeez axeeum Seeem . No--maap Naomp--toeuem _oo--eeee=z oea comumagowcw Go mpasmmii.m.~ mpnmh CHAPTER III REVIEW OF LITERATURE Measuring Market Rates of Return Two basic problems were presented when the market rates of return were measured. The first problem was to choose the proper weighting system. Fisher and Lorie chose to weigh each stock equally (the equally weighted index).1 An alternative was to weigh according to the total value of each company in the market (the value- weighted index). Both indices had their advantages and shortcomings, although the value-weighted index was considered theoretically more sound.2 In this study, indices were computed according to both assumptions. The second problem was related to the reinvestment of dividends and proceeds of rights. As discussed in the previous chapter, the dividends were considered as reinvested in the same stock, and the proceeds of rights were treated as in Equation 5. Following Puggina's notation, IS represented the value-weighted index and I'S represented the equally weighted index. 1Fisher and Lorie, "Rates of Return on Investments in Common Stocks: The Year by Year Records." 2See H. A. Latané, D. L. Tuttle, and W. E. Young, "How to Choose a Market Index," Financial Analyst Journal, Sept.-0ct. 1971, pp. 75-85, for a discussion of the development of market indices. Puggina (p. 58) showed the shortcomings of a value-weighted market index in the case of Brazil. 43 44 The Value-Weighted Index In this index. investment in every listed security was pro- portional to the total value of outstanding securities. This index was computed as follows: N P. N. jt+l jt+l . P. N'. 3:1 Pit “it at at RMt - N (18) X P. N'. j=l Jt Jt or M N. .73 P. “NI—’1” N'jt RMt _ j-lN jt+1 jt (19) Z P N'. t=l 3" 3" where: N'jt = Number of shares outstanding for security j at time t t and Ist = 100 tgl RMt (20) where: ISt = The market portfolio index at time t The Equally Weighted Index In this index, equal investment was supposed to have been made in every listed security at the beginning of each month, and the port- folio was supposed to be sold at the end of a one-month period. At that time. a new portfolio was formed, and so on. Mathematically, this index was expressed as: RM. .9 PJ'ET' <2» j=1 jt at where: RMt = Market return in period t = Number of securities in the market portfolio Pjt+l = Price of security j at time t+1 The market index was expressed as: t _ n RM (22) where: I = The market-portfolio index at time t st Real-Market Returns After these two market indices were computed, it was possible to compute the real monthly market returns as follows: _ p121 ‘ pt IIst — Ist —_——?F___— (23) and II + RRMt = lit 1 (24) st where: IISt = Inflated market index at time t RRM = Real-market monthly return Value of the deflator at time t 46 The Capital-Asset-Pricing Model The capital-asset-pricing model is a general-equilibrium model of the behavior of the pricing of capital assets. The most important assumptions of the model are as follows:3 1. That an investor maximizes his utility function under a single-period context; 2. That there are perfect and competitive capital markets; that there are no taxes. commissions, or other transaction costs; that assets are perfectly divisible and the quantities of all assets are given; that all investors are price takers. 3. That there is a risk—free security; that all investors can borrow or lend unlimited amounts at the risk-free interest rate; 4. That all investors have homogeneous expectations about returns, variances, and covariances; 5. That all investors evaluate their portfolios based on two parameters: expected return and standard deviation of returns; 6. That investors are considered risk averters.4 The Single-Factor Model According to the above assumptions. an investor's utility is a function of expected return and risk. This utility is represented as: 3M. C. Jensen. “Capital Markets: Theory and Evidence,“ Bell Journal of Economics and Management Science 3 (October 1972): 358-59. 4M. E. Rubinstein, "A Mean-Variance Synthesis of Corporate Financial Theory," Journal of Finance 28 (March 1973): 168-81. 47 u = ul) and consistently positive for the low-risk portfolios (b 0. To test Equation 47, Fama and MacBeth suggested the follow- ing stochastic model ~ -~ ~ ~ 2 ~ ~ Rjt ' Yot + Yit bj 1 th bj l Y3t0(ei) I ejt (70) where: Rjt = One period return on security j In Equation 70, Yot and Ylt may vary stochastically from period to period. As mentioned in the third testable implication, the expected value of the risk premium Ylt’ which is the slope [E(RM) - E(Rz)], is positive. Thus Em.) = E(R...) - E(th) > o 21 Tests." Fama and MacBeth, I'Risk. Return, and Equilibrium: Empirical 73 The variable b3.2 is included in Equation 70 to test linearity. Therefore, the condition for the first testable implication is that E(TZt) = O. The expression for the second testable implication is that E(I3t) = 0 because beta should be a complete measure for risk. The disturbance term Ejt is assumed to have zero mean and to be inde- pendent of all other variables. The data used were obtained from the Center for Research in Security Prices of the University of Chicago. The methodology was very similar to that of Black, Jensen. and Scholes. They also used earlier data to rank securities to be entered in the portfolios, thus avoiding the regression phenomenon. 0n the other hand, the grouping of data in portfolios reduces substantially the bias introduced due to errors in the measurement of beta. The authors formed 20 port- folios based on the ranked value of bj of securities computed from previous data. To compute the estimate of b for the portfolios, Fama and MacBeth used an expression developed by Blume:22 N b = 2 x.b. (71) p j=1 J .1 where: xi = the weight of security j in portfolio p They used seven years of data to form portfolios. For the subsequent five years, they computed initial values of the independent variables 22M. E. Blume, "Portfolio Theory: A Step Toward Its Practical Applications," Journal of Business 43 (April 1970): 152-73. 74 and the risk-return regression given by Equation 70, fitted month- by-month for the subsequent four-year period. The results of the regressions supported all the three testable implications of Equa- tion 47. The Behavior of Other Stock Markets Pogue and Solnik presented results of tests of the market model for seven European countries and compared these results with a sample of 65 American securities.23 The data base consisted of 229 stocks from these seven countries. On the whole, the results did not show substantial differences between the United States and the four major European markets (Great Britain, France, Germany, and Italy). Some evidence indicated that the three smaller markets (Belgium, the Netherlands, and Switzerland) were less efficient. Sharma and Kennedy tested the applicability of the random- walk hypothesis in a less-developed country, India, and compared this behavior to that of stock markets in advanced countries, the United States and England.24 The methodology included analysis of runs and a test for independence. The sample covered 132 monthly observations in the index for each market from 1963 to 1973. The conclusion was that stocks on the Bombay Stock Exchange obeyed a random walk and 23Gerald A. Pogue and Bruno H. Solnik, "The Market Model Applied to European Common Stocks: Some Empirical Results," Journal of Financial and Quantitative Analysis 9 (December 1974): 917-44. 24J. L. Sharma and Robert E. Kennedy, "A Comparative Analysis of Stock Price Behavior on the Bombay, London and New York Stock Exchanges," Journal of Financial and Quantitative Analysis 12 (Septem- ber 1977): 391-413. 118' U) ml tc Ir 75 were equivalent in this sense to the behavior of stock prices in the market of advanced industrialized countries. The primary objective of Lau, Quay. and Ramsey was to deter- mine, first, whether the capital-asset-pricing model was applicable to the Tokyo Stock Exchange and, second, to estimate the degree of dependence between the Tokyo Stock Exchange and the New York Stock Exchange.25 The results indicated that this model was applicable to the Tokyo Stock Exchange for the five-year period from October 1964 to September 1969 and that the two markets had a relatively low correlation coefficient.26 The methodology used to test the model was very similar to that used by Black. Jensen, and Scholes. Altman, Jacquillat. and Levasseur applied the market model to the French stock market and compared the results with equivalent studies of markets in the United States.27 They analyzed a sample of 316 common stocks traded on the Paris Bourse for the period 1964-71. They concluded that the risk parameter beta was as stationary and stable in the French market as in the U.S. market and that the market model explained approximately the same amount of variability in 250. Sheila Lau, Stuart R. Quay, and Carl M. Ramsey, "The Tokyo Stock Exchange and the Capital Asset Pricing Model," Journal of Finance 29 (May 1974): 507-14. 26However, after Richard Roll's article, "A Critique of the Asset Pricing Theory's Tests," Journal of Financial Economics 3 (March 1977): 129-76. these conclusions cannot be considered as definitive. Roll's article is discussed in more detail below. 27Edward I. Altman. B. Jacquillat. and M. Levasseur, "Com- parative Analysis of Risk Measures: France and the United States," Journal of Finance 29 (May 1975): 1495-1511. 76 individual firms' rates of return in France as in the United States. Officer searched for seasonality in the Australian stock market.28 Based on the performance of the forecasting model, he concluded that the Australian share market was seasonal but the presence of seasonality in share prices was not sufficient for reject- ing the efficient-market hypothesis. Although other studies could be cited, it seems that there is no reason to believe that the behavior of the Brazilian stock market is different, in terms of efficiency, from stock markets of other countries. In particular, it is reasonable to expect that the results from the tests that are described in this chapter will be similar for data from both Brazil and the United States. The Critique of Richard Roll Roll showed that the only economic conclusion of the capital- asset-pricing model is that the market portfolio should be mean- variance efficient.29 The so-called tests of the model, based on a market-portfolio proxy. whether mean-variance efficient or not, have no relevance as tests of the capital-asset-pricing model. Neither of the models mentioned previously nor the capital- asset-pricing model in its original form can be tested for validity. 28R. R. Officer, "Seasonality in Australian Capital Markets, Market Efficiency and Empirical Issues," Journal of Financial Eco- nomics 2 (March 1975): 29-51. 29Richard Roll. "A Critique of the Asset Pricing Theory's Tests," Journal of Financial Economics 3 (May 1977): 129-76. 77 The purpose of the empirical evidence that is presented, which duplicates tests developed in the American market. is to compare and contrast the findings, not with regard to their validity as tests of this model, but as new evidence bearing on the issue. Returns From Risky Assets and Inflation The classic models of Fisher and Williams presented the effects of inflation on unlevered firms.30 The assumption of these models was that the real value of unlevered equity is unaffected by inflationary or deflationary changes in general price levels. The same models, adjusted for a company's net creditor-debtor position, implied significant additional gains from unanticipated inflation when the company was in a net debtor position. Empirical evidence seems to show that this classic position is biased. Most of the studies of the relationship between inflation and returns from risky assets showed a negative relationship. Some of the findings of these studies are mentioned here. Empirical Evidence Michael Keran concluded that an increase in expected corporate earnings leads to a higher level of stock prices and that expectations of increasing inflation tend to lower the level of stock prices.31 30Irving Fisher, Appreciation and Interest (New York: Macmil- lan, 1896, reprint ed., New York: Augustus M. KETley, 1965), The_ Theory of Interest (New York: Macmillan, 1930); and John Burr Williams, Theory of Investment Value (Cambridge: Harvard University Press, 1938 . 31Michael W. Keran, "Expectations, Money and the Stock Market," Review: Federal Reserve Bank of St. Louis, January 1971. 78 Expectations of inflation increase both corporate earnings and the interest rate at which earnings are discounted. Changes in inflation expectations exert much greater influence on the rate of discount than on corporate earnings. John Lintner found that "a 5 percent deflation would reduce stock prices by 16.5 percent and 5 percent inflation reduces them by 3.35 percent."32 Charles Nelson concluded that the evidence appeared to suggest that ex ante as well as ex post returns on stocks are correlated negatively with current and past changes in the consumer- price index.33 Jaffe and Mandelker analyzed the relationship between a stock 34 index and the rate of change in the consumer-price index. The regression estimated was: Rmt = .0168 - 3.014 pt + E t=2.50 t where: Return on the stock-price index at time t Rmt Rate of change in the consumer-price index 32John Lintner, “Inflation and Common Stock Prices in a Cyclical Context," National Bureau of Economic Research, 53rd Annual Report, 1973. 33Charles Nelson, "Inflation and Rates of Return on Common Stocks," Journal of Finance 31 (May 1976): 471-87. 34Jeffrey F. Jaffe and Gershon Mandelker, "The 'Fisher Effect' for Risky Assets: An Empirical Investigation," Journal of Finance 31 (May 1976): 447-58. 79 The regression of the real return (rm) on the one-period lagged consumer-price index was estimated as: rmt = .015 - 3.338 It_] + at t=(-2.75) Using the Fama measure of anticipated inflation, the authors esti- mated the regression of the stock market returns on unanticipated inflation as: R = .003016 - 2.518 (pt - p't) + 6t t=(-1.642) mt where: p't = One month treasury-bill rate Jaffe and Mandelker concluded that nominal returns are nega- tively correlated with unanticipated inflation. However, Cagan and Lintner found that inflation was positively related to stock prices.35 35Phillip Cagan, "Stock Values and Inflation," National Bureau of Economic Research, September 1973; and Lintner, "Inflation and Common Stock Prices in a Cyclical Context." CHAPTER IV ANALYSIS OF RATES OF RETURN In this chapter, the performance of the market as a whole in the past decade is measured. Equation 14 showed how to compute the stock index. With this index, it was possible to compute market indices based on different methods. Equations 20 and 22 showed how to compute the value-weighted index and the equally weighted index. Equation 24 showed how to compute the real market return. As dis- cussed in this chapter, the deflated market index is computed by using two different deflators. The first is the variation in value of the indexed national treasury bonds, and the second is the General Price-Level Index. Why Use Two Deflators? As was shown in Chapter I, the Brazilian economy is indexed based on a government index called the indexed national treasury bonds (ORTN). Recently this index was used as a tool to reduce infla- tion. Therefore, the variation in this official index has been smaller than the real inflation. For the entire decade. the varia- tion of the official index was 35 percent smaller than the variation of real inflation measured by the General Price-Level Index (GPLI). On the other hand, government and private bonds and long- term loans were indexed according to a variation of the ORTN. The 80 81 official index represents an alternative opportunity of investment. The official index followed inflation closely until the middle of the decade; then it started to lag behind real inflation. Because the variation of the ORTN values represents an alter- native type of investment and is the basis for indexation of the economy, a comparison is presented between the performance of the market and the performance of indexed bonds. In this study, indexed government bonds were viewed as risk-free securities. However, as discussed above, in the second half of the past decade this invest- ment had inflation risk and could not be considered as risk free.1 Market-Performance Analysis The Value-Weighted Index Is for the Total Market In nominal terms, the market index IS increased from 103 to 1490 or 1390 percent during the decade. On the other hand, as deflated according to the GPLI, the market index decreased from 101 in the first period to 84 in the last, representing a real loss of 16 percent. Deflated according to the ORTN, the market index showed a gain for the decade, increasing from 100 to 129 or a 29 percent increase. At this point it is interesting to analyze the 1971 market boom. The market index, deflated according to the GPLI, increased from 335 in July 1971 or a 235 percent increase in real terms in only 18 months. Since then, the market decreased, although with a few periods 1The term "risk free" is used here as in the textbooks of finance. In reality, there is no pure risk-free security. 82 of small recoveries. At the end of the decade, the total market, in real terms, was only 25 percent of its value in July 1971. This means that an investor who bought the entire market (a portfolio of all the stocks in the same proportions as in the market) in July 1971 had, at the end of the decade, only 25 percent of his original purchasing power. To the extent that explanation of the market is possible, the implications of the boom market of 1971 are presented in Appendix G. The behavior of the market as measured by the value-weighted index for the period of this study is presented in Figure 4.1 and Table 4.1. The Equally Weighted Index 115 for the Total Market The behavior of the equally weighted index I'S was substan- tially superior compared with the behavior of the value-weighted index. In nominal terms, the equally weighted index increased from 100 to 12405 for the decade. When deflated according to the GPLI, the index increased from 99 in the first period to 704 in the last, showing a real gain for the decade of 1.65 percent a month. When deflated according to the ORTN value variations. the index increased from 98 to 1076. showing a real gain for the period of about 2.02 percent a month. This difference between the two indices was confirmed by Puggina.2 Although he was working with another set of data, his 2Puggina, p. 111. INDEX 1600- 1400‘ LMXF 1000* 600‘ 500- 400‘ 300‘ 200‘ 83 NOUNNAL REAL-GPLI DEFLATED I974 Figure 4.1.--Value-weighted index 15 for the total market. 84 amp wmp me— amp mmp NNP mmp «mp mmp «up cmp om— mnmp omp omp mcp Pmp nmp amp «my cop oc— cop MNF pop mum? omp mm— Fmp mmp mep wmp pmp omp mqp new oep mep sump nmp «mp pup map «mp cop mmp pep “up sup Rep mm— mmmp mep mnp emp mop nmp mmp cop mmp omp amp cup NNF mnmp oup mpp mm mm mop cpp mop —FF spy mmp mm— mm— ¢mmp mmp Fmp mmp Bop Nmp mmp oep Pm? exp wmp op— mmp mmmp Fmp mm— omp mmp any cop mwp com com FNN New mum «mop mew mom pmm mum Nam oem mpm 5mm new «pm . Nmp mmp Pump mmp mpp mop m—p mop om ow ow mm mm mop oo— ommp Azemov eeeepeeo omcp opep ommp weep ommp umpp mmpp ewPF mqpp mmop woop «mop mxmp coop omop mmop mwop mmop mmop «cop omoF Nam mom omop mum mmmp mew cow cum NNw mun my“ woo woo use «no Fmo pmo “map «mm Pom mom mum owe Foo mmm mmm Rom mow mem cam mum? ace Noe “we Nae “me mom New mom Nmm Fem cum Npm mump mFm mmm mmm mmm ecu emu me mmm mmm new you New enmp mmm mew —m~ mom mum mum me New Rom mum amp NFN mum— omm omm emu mmm mmm mom cam ppm opm vmm me 5mm Num— pom saw mum mmm mmm pme Npe ome Fpm mom emu NFN Pump mop mm_ mm_ «mp mpp mop mm mm em mop mo_ mop sump Fm:_soz .uma .>oz .poo .pamm .a=< x—zn mesa x6: Pwsa< seem: .nom .cmw newsma .eexeee _eeoe 8:8 284 we xeeew eeuemees-e=_e> wee--._.e e_eee 85 em mm mop mm mm mm mm pop pop mm em mop mum, mop mop cpp pup cup sup pup sup wm— Fmp ocp Pmp mmmp cup amp sup amp mmp mpp o—F mpp mpp pmp mpp mmp mxmp app Npp “op mmp mmp wmp pmp Ncp amp Fmp omp mmp mum, mm- mwp mep nmp mvp mop mmp mw— amp cup omp mpp mmmp mmp mpp mm mm mop sop mm mm cop omp mmp NNP exa— PNF omp mm— “op mm? mmp Pep mmp sup amp mpp “up mmmp mmp Pep amp mm— Pwp Om— map com mom mmm «em me Nump NmN ppm emu mum omm mmm mom mum New F—N owp mmp pump amp mpp mop opp mm mm om ow mm mm mop pop okay AHseev eege.eee .uwo .>oz .uuo .uamm .m:< apsa mesa an: Fwea< nose: .amm .cma cowema .eeeewpeeu--._.e 8,884 86 real I's index increased 111 percent between March 1970 and December 1972, whereas the index used in this study increased 107 percent for the same period. On the other hand, Puggina's real 15 index increased 52 percent, whereas the index in this study increased 49 percent for the same period. In Figure 4.2 and Table 4.2, the behavior of the market as measured by the equally weighted index is shown. The 15 and 1's indices are compared in real terms in Figure 4.3. It is possible to come to one important conclusion based on the analyses presented in these figures and table. If the same weight was given to all companies, the market index was substantially superior to the market index weighted according to the value of the companies. This means that smaller companies did better than larger ones. The reason for this can be found in the fact that a small num- ber of low-performance large companies represented a large percentage of the total volume traded. In Appendix H the names of these companies and their percentage of the total volume traded are presented. The BOVESPA Index The BOVESPA index is the official index of the S50 Paulo Stock Exchange. This index assumes the reinvestment of dividends and sale of rights in the same stock. The basic characteristics of this index are as follows: 1. A group of stocks most actively traded in a certain period is selected to form the BOVESPA Portfolio. 2. The weighting system is based on the volume of trading. 87 WIDE X 90004 IOUIML ——. REAL GPLI DEFLATED 1000‘ , ' -‘ v 1 v i 1 1 1 I ' 12 24 36 48 60 72 84 96 108 I20 IONTNS I9TC 1971 1972 1973 1974 1975 1976 1977 1978 1979 YEARS Figure 4.2.--Equa11y weighted index I; of the total market: nominal and real. 88 NOE): 000- 600- I 200 a (I VA LUI II ICNYID 1970 1971 an I973 1974 I975 1976 1977 I078 1979 Figure 4.3.--Va1ue-weighted and equally weighted indices of the total market, GPLI deflated. wmop weo— muop «For eom Few mmw enw mow mun wow wow mnmp wa 0mm me wew www owe wow now wmm mwm ewm wpm wmmp use oee m—e mmm wmw eem mmm wwm me oom owm mom Kemp mmw wem mwN wem ewm wmm mew wmm wmm FNN eNN wpm wmmp mom mmm mom ecu Now mom wow map mn— «we mop mop mmwp mwp wwp mop ewe owe map may wow wow mwm wNN mum efime mmw mmm mmm eem mew NNN wpm NPN pmm wpm emp wmp mum” Now Pom me wow «cu —m_ map mom mom mom wmm mmm NNmF mmm emu nmm cum mum wow cum wa oem owp wmp Nee PumF app mop mop mop pop wm —m pm mm wm cop wm omme Azemov eeeeeeeo 89 moem— ommpp oomp— wewop moww mom“ oewn .wmuu pmmw mpow eumm wmwm mmmp mmom wwwe muse wwwe mpee epwe ommm ewwm mpnm upmm emem mwmw wNm_ uwww pmew swww mppw meow wwnp wwwp mowp mpmp wwmp owwp ONFF new, wo_p weep yum wum wpop Pwm mww wa wmn Nee mm“ emw wxmp new mww mww wow wwm mwm nwm mmm owe awe mwe awe mum, wwe mpe mam emm wpe wwe o_e woe owe mme wee owe eump mwe wwe owe mee wee ope own cum eoe mwm www mmw mump mmw wmw wwm www mwm wow wow Pwm owm mFm wwm wwe NumF wwm mww Nww enm mum woe mwe mee wow eww Nap wmp Femp we— mwp wpp opp NPF woe mm um wm mop mo_ cop sump Pecwsoz .umo .>oz .uoo .uamm .m:< xpze menu has Fwen< seen: .20; .cee nowema .eexeee .8383 6e» co m.H xeeee eeeemeez appeeee wee--.m.e epeee 90 eon wow mmn oon www New www www wpw wew wow oww muo— opw www wpw wpw ppm woe ewe wwe wne nwe pwe owe wmop eow www wew eww ewm eow wmw wuw eww oew www www uno— www wpw mow mpw www eww wpw wow mop wop wow mop wmop Pop pop eop eop wop wop pow owp ww— “we wwp wwp wmop wwp wwp Pwp wwp mwp Pwp wup wu— wwp now wpw wpw emo— o—w www www eew new oww wpw epw www e—w mop oow mnop wow eow wow Few wow mop mop opw pew opw mww wmw wmo— uww oww oww an www mow mew wem wmw wwp ewe wep Fume opp wop pop wop wo wo om Fo mm mm po— mo omop Aeoeuo eeee_eeo. .uwo .>oz .poo .uamm .o=< apao mcao we: pwea< nose: .nmu .cmo noeemm .eeaeepeoo--.~.e e_eee 91 3. If the trading volume of one stock increases, it is included in the index; if the trading volume decreases, the stock is excluded. 4. The index is reviewed periodically, normally every three months. Puggina felt that this index is biased. He stated: "The index is biased twice, in the sampling process and in the weighting system."3 In Figure 4.4, the evolution of the BOVESPA index and the IS index 'Hs presented. both in nominal terms. In Table 4.3, the monthly values of this index are presented having 100 as a base for January 1, 1971. The 15 index in nominal terms for the total market is also pre- sented. Assuming that the value-weighted index 15 is the best market representation, it can be seen in Table 4.3 that, during the bull market of 1971, the BOVESPA index overstated the market. For all the other periods, the BOVESPA index understated the same market. In any case, the difference between the value-weighted index and the BOVESPA index was not extremely large. The reason for this behavior can be found in the fact that larger companies were more actively traded than small ones. Thus the weighting systems of the two indices were similar. The BOVESPA index can be considered a reasonable market representation. 3Puggina, p. 29. ”It“? .13: ‘ fir 0V6- ‘0: 4 ‘5: 1 2:: l 3. ... at: . Fig 92 INDEIl 1400 ‘ 1300 < / moot . r. VAUJE WEIGHTEO / _——. 1100 ‘ 10004 / 600 ‘ 400 " 200 ‘ 0 —fi 1 l2 1 ‘ T T T T ' I ' I 1 T 2 x 48 60 Te 04 96 108 120 ms 1970 197 I 1972 1973 1974 1973 1975 1977 1978 I979 VIMS Figure 4.4.--The BOVESPA and the value-weighted indices for the total market. 93 opop mwpp Pwop mew oww oww oew www www .mw www eww owop Fww Pow Pow oew www www www eow www www wow wow wwop mew new new oww www wa www www me ooe ooe pwe wwoF owe wow wwe opw owe mwm wwe eoe eoe ewe oee www wwop pwm mom wwe oom Foe —Fe me ewm www www www www mwo_ emw Pop oow oww mmw pow wow mew www www wa woe ewop opw oww oew New www www oww www mop wwp wwp owp mwop wop wwp oww wew pow www Fww pow www wow wow Fem wwo_ www www www oww eee ooe wme www www eow Pow owp Fwop wep ow— wmp ow_ mwp moF FPF oop wpp wpp wo_ mop owoF omumpmmmuipecmEoz ow meem wwww owew wome ewpe wwpe owee wepe womm owew epwm oeww owop wwmm eeww mpoe mooe owoe wwwm epom wam mmoe mope pwwm womm wwop wwwm wmew wwew wepw ewww wwww Poww waw oeww Fwww wwww opew wwop Poww owpw moww wwww mpww woww owew wwpw prw owew wmmw wwop wwop oeow mppw owww wwow wpew oopw www, wwep mmep omep owep wwep wwop wap Fwo_ owop www, wewp wwop oopp wwpp opep wwwp ewpp wwo— ewop Pwpp wmwp emmp wwmp wwwp wow— mmwp wwmp wmop peo moo opop mwop wwop woo wwpp wowp wwop wewp oeep womp owe_ wewp owwp wwwp wwop owm— wmwp ooop owop wwmw oePw wpew womp owwp moo— wwop www _wop wow wow oww eew www www wow www Pow wow wow omm owo_ Pecwsoz ow .umo .>oz .uuo .uomm .o=< »_:o mono on: Fpeoe sage: .amm .cmo oowemo .meeeee_ eeeeoees-e=_e> ego eee xeeee eemm>om eee--.m.e epeee 94 ooep weep oww— woep owwp owee wwpp ewpp weep wwop woop wwop omo— eoop wwop wwop wwop ooop wwop woop wwop woo woo owop wwo wmoe wew oww www www www wpn www www www www eww me wwo, eow Fww wow www www pow www www mow woe oew oew wmop owe owe owe woe owe wow wwe oww www pew eww wew wwop wew wow www www wew eww oww oww www www pww wew ewop www wew pww mow www www Fww oww mom www oop wpw wwo? oww oww eww www www www oow ppm wew eww pww www wuop eww wow www www wow ewe wpe owe ppm www eww wpw pso— wwp wmp wwp ewe opp wop ww ww eo wop woe mop omop FenwEoZTTxoecH eoonowoziozpe> .ooo .>oz .ooo .ooom .oo< o_=o onzo we: Fwsoe nose: .noo .neo oowsoo .eeeeeoeeo--.m.e esees 95 The Value-Weighted Indices for Common and Preferred Stocks The value-weighted index was also computed with two different sets of data. In the first, only common stocks were included, and, in the second, only preferred stocks were used. In Figure 4.5, a comparison is presented of the two indices on a real basis, deflated according to GPLI deflators. In Table 4.4, the value of this index in nominal and real terms is given. It is evident from the data presented in Figure 4.5 and Table 4.4 that the common—stock index was almost always ahead of the preferred-stock index in nominal and real terms. The bull market of 1971 was the only point at which the preferred-stock index was higher than the common-stock index. Since preferred stocks are less risky than common stocks, it would be expected that the return on preferred stocks would be less than the return on common stocks. Except for the bull market of 1971, this was true. This was the first evidence of irrationality of the market during 1971. During this period, the index for common stock in real terms reached a peak of 342 in June; the same index for preferred stocks reached a peak of 375 in May. After this period, common stocks showed a better resistance in the bearish market than preferred stocks. At the end of the decade, the index for common stocks was 106, showing a real gain for the lO-year period of 6 per- cent. On the other hand, the preferred-stock index closed the decade at 52, representing a real loss of 48 percent. 96 INDEX 400 . 1 ‘1 350 < 1' ‘ 1 l ‘ g 1 1 , l 300 ‘ l \ 1 \\ 1‘ \ \ l \ 250 4 1 \ 1 \ 1 \ \ 200 i \ \ \ \\ \ COMMON 150 ‘ \ ' \ 1 \ I \ A I ’IA‘x I \ ll \' ‘\ I] \ ’IA 100 4 [I \ ’/ \.’ ‘\\ \ / \\ ll \ \ 7 y \‘k PREFERRED V ‘\‘*”.‘-———\\‘/ \“--.‘\ so i 7' 0 T T T T T T fi T T T T T T T T ' T T l ' 12 24 x 48 50 72 84 96 108 120 norms 970 1971 1972 1973 I974 1975 1976 1977 1978 I979 YEARS Figure 4.5.--Value-weighted indices for common and preferred stocks: real terms. 97 wo— o—P wwp opp opp epp pr ww— wwp wpp wpp owe ouop owp wwp eee wep wwp wwp wep nwp wwp owe owe owp wnop pr we— pwp wwp we, ww— wwp wwp oep eep oep eep hoop wep wwp ewp wep owp owp esp wwp oep oep owe ewe wmop owp wwp wwp www wwp owp woe oep owe www ewe pr wwo, wwp o—p oo oo wop opp Pop ooe eoe owp wee wpp enop ope wwp we_ owe ewp wwp wee wep woe wwp owp ow_ wwop ewp wep “we owe ww— oop owp oow w—w oww oew www wwop eww pew www eww wow wew now oww oww wpw wwp owe poop oep wpp ooe wee oo wo ow ow oo wo wo— Pop onop Feomiixooom noEEoo Fwwp wwwp omo— ewwp pewp wuep owe— oeep oeep wwwp wwwp ewwp onop wwwp wwwp ewwp www. wwwp wewp owwp wwwp owa oowp wwwp owpp wmop ewop wwo wwo ono wpo www wow wow wow pow new pen wwop won www www eww wpn wow wwo www www www —ww wpw wwop www opw oww eew woe oew wow ooe pww wew www oww wwo— www wow wew Pew www oww www oww www eww wew oww eno— www pew www www oww www www www oew wa eow wpw wnop www pww www www eow www wow wpw www wew oww oew wwop eww wow www wow oow owe ooe owe wow www www wew Poop wwp wwp wwp uwp epp mop ww ww eo wop woe woe omop penesoziinooom noesoo .ooo .>oz .ooo .ooom .o:< opoo onoo we: pwso< nose: .nom .neo oowsoo .mxooom eossoeoso one noEEoo sow mooeenw eoonowozioepe> onhui.e.e aneH 98 ww ow ww ow ww ew ww Pw pw ww ww ww onop ew ow ow so wm ow no ow Fw ww oo ew wwop on go ew ow ww ww wn ww wk ow oo ww “mop ow ww en ow wo wo woe ooe oo wo mop oop wwo— mo woe woe wpp wpp owp owe wo woe wo_ wo wo wwop ooe wo ww ow pw ww ww ew ww woe ow- ep— eno— wpp wpp mop wnp owp Pwp opp wep oep PPF wo woe wwop wpp pwp ewp wwp eee wep owp owp wwp woe woe wow wnop www wow wpw oww oww oww epw wow www wop wwp wwp Poop we? oo ww wo eo no on on ow oo woe wo omop Peomiinooow oossoeoso wpo www who www eew oow oow oww wow oww www wew owop pww pww oww pow www www www wew oww New eww wow wwop wew oow new www www wow ewe eee owe wee ewe wwe wwop wow www wew ope wee w—e wee woe eww www wow www who? oww eew www eww eew www oww oww www eww www www wwop oww eww owe owp wop wow wop eop —ow www oew www ewop wew w—w wop new www www www oww oww Pop wwp wm— wwo, wwp mop oow e—w www oww new eew eww oww wow Pow wnop oww www —ow wew eww www wpe woe oww eew oo— owp Fmop oep wp— wop sop sop wo ww ww ww wop wop oop onop FenPEOZTTnooom eossoooso .ooo .>oz .ooo .ooom .o=< opoo oceo we: pesoe nose: .aoo .ceo oowsoo .eeseeoeeo--.e.e eseee 99 Sector Analysis In Appendix C, the division of the companies on the exchange into sectors is shown. The behavior of the major sectors and sub- sectors is analyzed in this section according to the following order: Sector Subsector llOOO--Construction lOOOO--Industry 12000--Mining 13000--Manufacturing 21000--Domestic 20000--Trade 22000--0i1 Dealers 3lOOO--Financial Institutions 3000O--Services 32000--Public Utilities 90000--Miscellaneous Thus. seven subsectors and four sectors are analyzed: 1. Industry Subsectors: Construction, Mining, and Manufacturing In Figure 4.6, the behavior of the three subsectors in real terms is compared. In Table 4.5, the value-weighted indices in real and nominal terms for the lO-year period for the three subsectors of the industry sector are shown. During the bull market of 1971. the mining subsector reached the highest index level in real terms, 930 points in September 1971. After that, the index for mining declined, closing the decade at 32, showing a loss of 68 percent in real terms for the entire period. 100 ‘— 11 000 CONSTRUCTION ---------- 12000 MINING -- --1.3000 Tum-mow INDEX 000 500 200 100 8m :97: '9" '973 1974 I975 1976 1977 (970 )9 Figure 4.6.--Va1ue-weighted indices for industry subsectors: GPLI deflated. 101 em Fpp eo ww ew ww wpp wwe wwe ow— oo wo omop wop wp— ewe . pwp wwe wep wwe ow— ewp owe wwp owp wwo? wwe wwe wpp wpp oop eop wo wo wo ww on on woo? ew ww ww wn ew ww nw ww ww ww ew wm wwop ow ww ww ww ww ww es en ww ow ow ow wwo? ww pw ww ew ow ow om ww ww wo opp wpp enop pwp wee wwe wwe wwp Pop mop wwp on— esp wwe —wp wuop —ow oow woe wow www www eew www wow wow www eww wwop oww eww eww wow wpe owe oww www www wow wee wa emo— oww ow— Pep wwp ooe ow pw ow ow ow ww ow omop Peee--eeoee=someoo ooop_ wowp pww— owe— onwp ompp woop weep wwe, wwwp eoop eoop owo omop wwop wwop wepp owpp NFPF wpwp owpp wwwp pwwp wopp wpwp owpp wmop eeo www www own www www www www oww ewe ewe ooe moop oww wew oow www www www www wew eww oww oow www wmop www ow— ooe wop eoe oow wew wpw wwe wwp Pwp owe wnop Rep owp wwe ww— eep ewe wwp ewe wwe wew www oww emop www eww pow eww o—w wow eew www ppw oow www wow wmop www www oww eww pww www www Foe wee wee wwe wwe wNoF wwe woe now oew www www eow www wen www wpw oww Poop www oop ewp oep wpp no ow ww ww ww ww Pw onop seeeeoz--eeeoe=someoo coop. .ooo .>oz .ooo .ooow .o=< :_:o onoo we: Fwso< nose: .noe .neo oowsoo .xsomeocw oooop "msoooom on moowenw eoonowozioepe> oneii.w.e opnep 102 ww ow ww ow ww ww ew ww ww ww ww Fw owo— wp up up op ww ww pw ew mw Pw ww ew wwo— ew ww we oe ww ww ow we ew ww ww ww wwo? ww ow ww ow om Pw ww ww ww ew mo wo— wmop oo wpp ww— oo woe owe wwe wwp wwe owp wpp mop wnop wpp wm eo ww wop uwp wwp wp— wwe ewp wwp oep emoF wee owe ee— wwp wwp owp wwe onp Pow esp owp wop wmop wwe mwp wwp onp eow wew —ow www oww www wow npe wmop New www www owo Pew new won oow eww wwe wee www poop epw www oww eww ewe ww pop eo pop eoe eop wop ouop Pemmuuocecpz ooowp eww woe Fww www oow wow wow wow wow eww oww eew omop owe ewe pr wwe wop pop wwe woe wpw oew oww oew wwo— www www www wew www www eww eww epw wew wow www Buoy www www pew oew www eww www www pww opw eww www wwo— www www woe www www oww oww www www www wpw www wnop —ow owp www wpw wew Pow oow www www wow wew mow emop pww eew www Pow wow oww www epw oww oow eow wwe woo” opw www pow www eww oww oee www wow oow Fwe wow wwo— ewu ewo ewop wwwp wopp woop wwo oeop won www www owe Poop wow wow wow wow wow wo pep Pop mop wop mop wop omop Pecmsoziuocwce: ooowp .ooo .>oz .ooo .ooow .o=< apoo ocoo we: Peso< nose: .nou .ceo oomsoo .eeseeoeeo--.m.e e_eee ww pm ow ow no no ow ow ow ow ww wo owop wo oo eop we" ewe wwp epp opp wpp oFP wwp wwp wwop mop woe wop wop mop oo ww po wo no wo ooe wwoF wo eo ww wop wpp wpp wwe wwe wpp wpp wep mep wwoe nwp wwp wwe owp eee wwe owe Pwp wwe wwe wpp Pep wmop wwp wwe on ow ow wo ww Pw ww wop wop oop emop oo oop ewp wwe wwp wwe wwe ewe wwe opp oop wee wnop wwp owp wwp wep Pup —wp wwp wwp woe wow oww wow woo? oww owe Pop oww wew eow eww oww oew wwe wwp wo— Poop ww pm ww ww we ow ow ww ww wo wo oo owo? .eee--oess=oeeo=eez ooom_ 103 wowp wwpp owwp wwwp wwop woo wwo oeop oeop wwo www owo omop woo wwo ewo wpop wwop wwop oeo weo oww woo ~wo oow wwop eww www oww www oww wow www www eww www wow epw wwo— pwe ewe woe wwe mew www www eoe wwe wwe eww wew wmop wwe wwe eee wwe wee opw wwe www oww www epw eow wnop opw w—w woe woe oow oww wop wwp woe www er woe ewop wow wow oww eww oew oww www www eww wow wwe woe wnop pow ppw mow eww eww www www www eow wow oww oow wnop wpw eww www wew eww wow www www eww wew wwp wwe Poop wo ew ow ww ww ww en we ww wo ooe Fo omop Penwsoznion_s:ooee:ne: ooowp .ooo .>oz .ooo .oeow .o:< opoo oczo we: Pesoe nose: .noo .neo eowsmo .eeeeooeeo--.m.e eseee 104 The construction subsector index also increased dramatically during the bull market of 1971, reaching 636 in May. After that, the index decreased, reaching its lowest level in December 1974. From 1977, the index recovered, showing values superior to 100. It closed the decade at 74, with a 26 percent real loss for the 10-year period. The manufacturing subsector showed the smallest variations. During 1971, the index reached 294 in June. It closed the decade at 73, showing a loss of 27 percent for the entire period. 2. Trade Subsectors: Domestic Equipment and Oil Dealers The behavior of the subsectors in real terms is shown in Figure 4.7. The value-weighted indices in real and nominal terms for the trade subsectors are presented in Table 4.6. During the bull market of 1971, the increase in the deflated value-weighted indices was the smallest of all the subsectors. The index of oil dealers reached its highest point in June at 209. The value of the index for domestic equipment was even lower, reaching its peak in June at 162. The index for the oil dealers reacted only after 1977, clos- ing the decade at 219. The index for domestic equipment closed the decade at 59, showing a loss of 41 percent for the entire decade. 3. Services Subsectors: Financial Institutions and Public Utilities The value-weighted indices in real terms for the service sub- sectors are shown in Figure 4.8. The value-weighted indices for the two subsectors in real and nominal terms are given in Table 4.7. During the bull market of 1971, the value-weighted index deflated, reaching 415 in July for financial institutions and 329 in 105 WIESTC EOUPIENT - -- ---— OIL DEALERS INDEX 4mm 800 " Qm-i Figure 4.7.--Value-weighted indices for trade subsectors: GPLI deflated. 106 ow ew ew ew ww we oe we we ee ew eo oeoe ew ew ww wo eo wo oo eo eoe ow oo we weoe ee we ww ew ww ew ow we ew ww ow ow eeoe ww ow we ow ee we ow ow ee ee ew we weoe we . ow ow Fe we we ee we ww ow ew ew weoe ew ww ww ew ww ee we we we ww ew oe eeoe ww ww ow ww ww ww ww ww ow ew we we weoe ee ee —e we we we ee ow ww ww ew ow weoe ww ww ww ow oo owe wwe wwe eee e—e oo wo eeoe ww we Fe oe ee ww ww ow oo ow eo wo oeo— _eee--eeeEee=ee eeomee°e ooopw wwoe wwee oeoe wwo w—o wwo eww www www www eoo ewo oeoe woe eew oow www www eww wee wwe eww oew oww www weoe oww oow wwe wew www oew oow www wew wow oew oww eeoe oww oww oew wwe e—w oow oew wow ewe ew— ewe ww— weo— ewe owe wwe ew— owe wwe ewe ow— wo ww ew ww weoe ow oo ow ew wo woe wo ooe Foe ewe ewe ope eeoe wee ope wee eee owe eee oo oo woe oo ee we weo— ee we ew ow we we ww oo eo ew wo ooe weoe ewe eee eee wwe wwe wew wew oww wwe eee wwe wee eeoe eoe oo ww wo ww eo Po wo wo wo ooe wo oeoe Peeseez--oee28e=ee eoomeeee ooepw .ooo .>oz .ooo .ooom .o=< eeoo oneo ee: Pesoe nose: .noo .neo ooesmo .oeeso oooow "msoooom on mooeene eoonoeoziooee> oneui.w.e opnee 107 oew www www woe wee wwe wee wee owe ewe wwe wee oeoe wwe oee wee owe owe woe ewe owe eee wee wee oo weoe eo eo ww ee ww ww we we ee ww ww ew eeoe ww ew ew ew —w ww ew ww ww ew oe Fe weoe ww oe ee we we we ew oe we ww ww ew weoe ww ww ww oe oe ew oe we ee ee we ww eeoe ww oe ee ee we we we we ww ww ow ww weoe we ow we we we ww we wo eo ow woe eoe weoe wwe wee owe ewe ewe owe oow wow owe wwe oo ew eeoe we ee ww ow ow wo ww ww ow woe wo_ wo oeoe eeowuumsoeeoo eeo oooww wwww wwww wwew owww ooww ewee ewee eoee wewe owwe wwwe weep oeoe owwe ewwe eowe wwwe eowe eewe owwe ewwe www eww eww eoe weoe wew eow oww oee oew www www eww eew eow www wew eeoe www owe wee eee oee wee wee wwe owe wee we— we— weoe owe ewe ewe wwe owe owe owe eee owe wee owe wwe weoe eee wwe ewe wo wee ewe eee eoe ooe eoe ow we eeoe ww ee ee ww oe ee we we eee oo eoe eee weoe ewe wee ewe ewe wee woe wee oee wwe ewe wwe wwe weoe owe wwe wwe owe wow wew wew wew wwe eee wwe wo eeoe ww ew we eoe woe woe eo wo eo ooe woe wo oeoe Feceeoziimsoeeoo eeo oooww .ooo .>oz .ooo .ooow .o:< weoo onoo we: Pesoe nose: .aoe .neo eoesoo .eeeeeoeoo--.e.e eseee 108 FINANCIAL INSTITUTIONS - - - — - --- PUBLIC UTILITIES INDEX 1000 . 3004 I2 24 36 ‘8 (0 72 04 96 I08 120 “has I970 197 I 1972 I973 1074 1975 1976 1977 1978 1979 YEAR. Figure 4.8.--Va1ue-weighted indices for service subsectors: GPLI deflated. 109 ooe eoe wwe eoe wo eo eo ooe oe— ooe woe ooe oeoe wee eee wwe ewe ewe ewe ewe wwe oee wwe wwe owe weoe wwe eoe ewe eoe wwe wee eee wwe wwe ewe ww— owe eeoe eoe eee owe eoe wew eow www wow ewe ewe eee owe weoe wwe eee eee woe wee ewe eee ewe ope wwe wwe wwe weoe ewe eoe eee wo wee wee woe woe wee oee ewe ewe eeoe ewe wwe wee www eoe wow eee ooe www ww— ewe eee weoe wwe oee ww— eoe eww oow eww www oww www wow eow weoe oww eww eww ooe oww wee oew ewe oww pow oew wew eeoe oew eoe owe wo— eee ewe wo wo wo woe oee wee oeoe eeem--meeee=oeomeH Feeoeeeoe ooe—m owoe eeoe ewoe eewe eowe wwwe woep ewwe wewe oeee wwe— eoee oeoe wwee Pope ewe— weep ewoe owoe wwo— wwee wwee eew— ewee wwee weoe wwee eewe wewe owwp oeee weoe eeoe ewoe owoe eeoe woo owo eeoe eeo wew oee pow wwo www owo wew www eww eew eww weoe oww eew oww Pew eww www wow www eww eew eew oew weoe wew www www oww oww www eew oww eww eow oww oow eeoe wow wow oew wwe wew www eow eww www oww wew eew weoe oew eew www wow www www www wow oew wwe wee eee weoe wwe wow ewe www eww www ewe oew eew oww wew www eeoe eow www oew eww wwe ewe woe ooe ooe wee eee owe oeoe _eeeeez--meeoo=oeomeH _eeoeeeee eoopm .ooo .>oz .ooo .ooow .oee weeo onoo we: Pesoe nose: .noe .neo ooesoo .mooe>som oooow "msoooom we moo_e:e ooonoeoznooee> oneui.e.e oenee 110 wo wo eee ope wee wwe ewe wee eee ewe wwe ewe oeoe ewe eee eee wwe wwe ewe eee owe wwe ewe wwe ewe weoe owe wee woe woe eoe eo eo ww wo oo wo oo eeoe oo eoe wo eo eoe ooe ope ope wwe wwe oee wee weoe wee wwe ewe ewe wwe wwe wwe wwe ewe owe wwe wwe weoe wwe wwe owe wwe wwe ewe wwe ewe wee ewe ewe owe eeoe ewe wee wee wee eee wwe wwe wwe wwe wwe owe oee weoe pee wee owe owe ewe wwe wee wwe wwe wee wwe ewe weoe wwe ewe wwe wee oww wow oww woe eoe wwe wee woe eeoe wo ow wo eo eo wo eo wo ooe ooe wee eoe oeo— eeowuimoeoeeeoo omens; oooww wewe ewwe eeee wewe oewe wwwe eewe oew— oewe oeee owwe wwwe oeoe oewe oewe wwwe wewe wwee ewwe oewe eewe wwee ewo wooe wew weoe www wwe oow oew eww wew oew oww www eww woe oew eeo— eoe wwe owe oee wwe wee owe owe wee wow wew eow weoe wwe oee woe oee wee wee eow www eww www www www weoe www oow eow wow www www wew oew eww eow wew www eeoe www www www www www wew www eww eww wew eow ooe weoe ewe woe eoe oow wow wew oww www eww www eww oww weoe www wew eww www eww wee wwe oww wew wee owe wwe eeoe eee eoe eoe eoe eee woe woe woe wee wee owe woe oeoe eenesoz--moeoeeeoo oeenoe oooww .ooo .>oz .ooo .ooow .o=< weoo onoo we: eesoe nose: .noe .neo eoesoo .eeeeeoeoo--.e.e eeeee 111 June for public utilities. After that, both indices declined; but, in general, the index for financial institutions was always ahead of the index for public-utilities companies. At the end of the period, the indices were 109 for financial institutions and 85 for public utilities, both in real terms. 4. Sector Analysis: Industry, Trade, Services, and Miscellaneous In Figure 4.9, the behavior of the four sectors is shown. The value-weighted indices for the four sectors are given in Table 4.8. During the bull market of 1971, the service sector reached its highest level in May at 387. The service sector outperformed all the other sectors until the middle of 1977. when the stocks included under mis- cellaneous obtained better results. It is important to note that only six companies were included in the miscellaneous sector. As shown in Figure 4.9. after June 1977, the miscellaneous sector was substantially superior to all the other sectors. Only two sectors closed the decade lower than 100: industry and trade. The trade sector had the poorest performance for the entire decade. Only dur- ing the 1971 boom was the deflated index for trade higher than 100. From the analysis of the sectors and subsectors, it is apparent that the bull market of 1971 was of tremendous importance for the Brazilian stock market. After that, the market did not recover. Thus it would seem that the stock market was not a good hedge against inflation during that decade. 112 10000 INDUSTRY ---------- zoooo TRADE .u.uuss-3°ooo SERVCIS ........ — 90000 unsoELmous INDEX moo o 1 W 1 i 1 fl I 1 1 j 1 W 1 T 1 T fi 1 1 W I2 24 36 40 0° 72 0‘ 5 m m "T”. 1970 I971 1972 I973 ’74 1.75 1976 1977 1970 I979 nm Figure 4.9.—-Va1ue-weighted indices by sectors: GPLI deflated. 113 we mm mm pm _n as as Fm ow mm mu em mkm_ em om mm mop opp epp ¢o_ mop cop o__ NNP mpp msmp mm mm mm cop mm mm mm em um Fm mm em “NmF mm mm Pm mm mop ppp mmp opp mop mop mmp mm— oxmp cup NNP mmp oep amp mop pmp cup mmp m_P ~—_ mop mumP mp_ mpp Nu mm mm mm mm mm mm mop mo. pop e~m_ Po_ mop map sup mmp om. mmp «mp mop app mo_ m_P mNmF PNN mm_ amp mep cup om_ mmp owp map New mum mom mum, _NN mmp «mp NmN www com «mm «mm mum PNF mmp mop Fump mm N“ mm as mm am No mm mm _m cm om o~m_ memuuxgpmzucm coco— pomp «opp mam. mm—_ mom m_m opm www com mpm was w¢w mxmp www mew mum mmm pom cum new new mom cmm mom oow mmmp “no Pmo mmm mmo mom mmm Foe Nae mom mom mne mme Rump one “me man new www www mum mow “we Npe mow mmc m~m_ ooe “Fe em¢ Fee ape wwe pee mmm men mmm mom mwN mum? mom PmN Pap omp mom mpm map amp «mp eww mpm pom esmp map com www emu mmm mmm mum emu www com «mp mop mNmF oow FPN mom NmN cum mew mew eww mam mom mmm mom NNmF Rpm mmm cum omm mmm mmm mum mum mmm mpm mop m~_ —Nm_ cop mm Fm mm 0w mu em ms mm mm mm _m ONmP _mcpsoz--»gpm=ucH oooo_ .umo .>oz .uuo .pgmm .m=< >_:a menu an: P_ga< saga: .nmu .cmn cowgma .msomcmppmumms vcm .mmow>gmm .mumgp .ALumsucm ”mgopumm an mmomucw kugmwmzumzpm> mghuu.m.¢ mpnop 114 mu om ow mm Nu en mm mu mu mm mm mm mum— mw om mm em um mm mm mm cop mm mm em mmmp en mm mm em em we me me om Pm me Ne mmmF me me me mm ee ee me we ee we me me oump me mm mm Fe Fe ee me me mm mm mm mm mum, mm oe mm mm oe me me me me em mm mm emmp mm mm mm mm mm mm em mm mm mm me om mmmp mm mm mm mm mm mm mm pm —n on ex om mum? mm mm em mop epp Pwp mmp omp mes om? —o~ em Pump em mm Pu pm mm mm mm mm mm mm mm mm onmp pmmmnuwumgh oooom momp emmp wmmp mppp mpop mom com mow mww nmw omm Nem mmm_ eww mew “mm mew mew mew mum men emu weo mmo —Nm mump mom wwe epe New mmm mmm mmm mom www www oow eew Rump new —NN Now mmp nm— Nap map mmp mug amp cup Pop mmmp mep mmp om— amp cup mmp ump NNP mop om em mm mmmp cop pop em mm mm mop mm mm NOF app FNF NFF eump no, mo— sop mop e—p mop mm so mop pop em em mmmp mm mm mm mm pm mm mm mop mop mm opp F—P NNmF mmp mmp Pmp mew amp mew 0mm omN omp Nmp mm_ mFP pum— oop om mm em co ma Fm mo em mm mop mm ommp pmcwsozunmumgh oooom .0mo .>oz .uuo .uamm .m=< x—ae maze an: ngae none: .30; .cme vowgma .umscwpcou--.w.e m_nmh 115 mo_ mop mm? epp mop mpp mp, mmp omp mpp cup eNF mnmp “up mmp mmp Nee eep Nee oep mmp amp mop eop mmF meme amp omp cup esp sop wmp mmp amp mop sop eop mop Rump exp Pop Rep Pup amp mm" amp amp map may qu Pmp ofimp Pup cup on_ “my mm? Pmp map mmp mm_ NMF mop wmp mum, em_ m—F mmp op— one Pmp emp “up mm? meF nmp emp eumF mm_ eo— mmp pmm amp mm_ mop mm, eom eee mmp me_ mumF mm— mm. mmp «mp mom emp m_~ mmm mmm Pom eum mum NNmF mom mmm mmm mmm mmm mmm New “mm mew mew mew mxm .Nm_ m—N oN_ mop mup mm— mm_ mm em mm mop PFF mpp oxmp pummuummup>gmm oooom mow, omm_ momp mmcp Romp emep Nee_ pomP seep __mp Kemp ommp mum. mmm— memp Rump oww? mm~_ NONF mop, NNNP oeNF NNNP mmpp eopp wxmp swap mmpp mop, no—F mmop mom omm omm Nmm pmm mum mew sump mew mun «mm “mm Nmm emu mpw om“ mmm mmm oeo one mump unm mmm oem 0mm omm eem mom emm mmm mmm mom mom mmmp new mom opm mmm mpm o_m mmm mmm com mum mum mom ego— mmm mom mwm ewe “mm eem com omm mmm wew mmm New mum— Nmm mom me Nam omm Rom mmm omm mmm omm moe mom NNmF eme omm ope Pme awe mum ome mom mem mmm mom mmm _mm_ pom mom mm— mom emp eep mop Pop mop «FF mpp up, ouop chweozunmmu_>gmm oooom .umo .>oz .uuo .pgmm .mze mean mesa he: pwgae zugmz .nmm .cme eoegwm .emacwpeou--.m.e mpnmp 116 mew cam new eem eww mew omN eew mom mmm omm ooe meme mee eme eee Nme ewe eoe ewe oee Nee mew emm emm meme mem com mom eew mum wme eme eme mme mme eee mme meme eee mme owe ome mme mme em eoe em on em mm meme mm om em eoe Noe Noe Nee ooe om on ow mm meme em mm me no mm mm mm mm mm moe mee ewe eeme «me one eme eme Nee eee wwe ewe eme wee ooe eoe meme woe woe eoe wee mee mue Nme mee mme ewe eme oee meme Nme eee mee eoe eee ome oew eew ewe moe mm em enme we we as we em mm mm mm mm woe mee em oeme emmmllmaowcm _. emum w: OOOOm mmme meme Nowm omom mmmm momm meom NeNm Nomm comm eemm mmoe meme mcoe emoe mace emoe mmee memm emmm eemm emem ooem mmem ewww meme Nomm emme mmee wmme emme come mom mmw eww wa wwe mew meme eoe mew omm emm ooe mmm ooe mNe 0mm mom owm emu meme com mow eem wem nem com mmm mmm eww Nme mew mmm meme emu oew eee woe eom new mme wme oew omm oew new eeme 0mm emu eww mew www com mmm mmm mmm eom oee one meme mee wee mee wwe emm mme mom emm mom ome mme eom meme new wow wow mum New mmm omm com emu mme mee eee emme om mm mm om mm eoe em ooe ooe Nee mme mm oeme Pmcwsoz--m=omcm__mumwz cocoa .umo .>oz .uuo .pamm .mze mesa wcze an: eegae cage: .30; .cme noegma .emacppcou--.m.e m_nmh 117 5. Market Rates of Return From the deflated value-weighted index, tables expressing average rates of return were constructed. Table 4.9 was constructed assuming 2l0 investment strategies. The semi-annual average rates of return for each strategy are shown. The table can be used as follows: If an investor bought a market portfolio on the first day of January l970 and held his portfolio until June 1970, his loss would have been 20 percent for the six-month period. On the other hand, if he held his portfolio for six months more, his gain would have been l7.90 percent per semester. Finally, if he held his portfolio until the end of the decade, his average loss would have been 9.86 percent per semes- ter. All the other semi-annual investment strategies are included in this table. 0f the 2l0 investment strategies, there were 138 negative rates of return in real terms or 65.7l percent of the cases. The best investment strategy was to buy in June l970 and to sell one year later, giving a 96.53 percent real gain per semester. After June l97l, of l57 investment strategies, l23 showed real negative rates of return. If viewed in this way, the stock market cannot be considered a hedge against inflation for the period of this study. 0n the other hand, large losses were found for those who invested at the l97l peak. For an investor who bought in the market in June l97l, the loss until the end of the decade was 7.38 percent every six months in real terms. Again, the bull market of 1971 had a negative effect on the Brazilian stock market. Investors lost confidence and changed 118 oe.m- oo.o- me.ee- w~.o- m~.m- em.m- ec.wu ce.o- we.m- ee.n- we.e- oc.m- ea.mu m~.m- m~.mu ew.o- mm.eu ce.~- o~.o ow.ou meme .uoa mo.oe- ow.~e- Ne.m- em.e- ~c.m- mo.eu me.mu em.o- we.mu ne.o- oe.~- om.ma ow.~n wo.eu om.0u o~.e- oe.~u we.o me.o- meme vase mm.ee- ow.m- ee.~- mm.n- en.e- em.eu mm.m- we.~- ee.c om.e- ew.~o ee.~- ~m.e- ae.o- co.e- oo.e- om.e oe.c meae .uoo ~e.~- mw.e om.c mm.mu ee.~- mm.eu m~.c- em.~ oo.o mm.eu ow.o- ec.mu ae.mu we.wa ~m.o- ~o.~ Ne.e weae acne me.~e mo.~ om.o- ew.~- ne.e- ee.o e~.m en.o Ne.e- o~.o- ee.nu ee.m- we.cn em.o ca.~ mn.e eeoe .umo om.e- mm.ee- om.e- oo.m- e~.~- ee.e mm.e- oc.nu oc.~- -.m- o~.eu ec.m- we.e- cm.~ eo.c eeae «cza ae.e~- ee.e- No.o- ww.o- me.m mN.Ou an.~u mm.eu mm.e- m~.~- em.mu o~.eu oe.n m~.e oeo— .uwo no.» e~.~- no.5 ne.ee mm.e me.e mm.e ce.~- mm.mu em.o- oe.o ee.m -.m meme vase mo.~e- en.o mm.ee mm.m a~.ou ee.o -.e- Ne.e- me.mu co.c me.m we.~ meme .uma oe.w~ mm.e~ cm.o mm.~ om.m om.~- me.o- eo.o- me.e eo.e m~.e meme ocze e~.e~ mm.o me.ea em.en ee.o- e~.eeu mm.~e- ~m.eu oa.e mo.~u eew. .00: m_.m_- e~.cn en.m- em.ee- ec.eeu m~.ee- n~.eu ce.~ ee.9u eeme acne we.ee- ~o.en ow.me- oe.men oe.ee- m~.~n mo.c ee.~ mnae .uwo ~c.o oe.meu oo.~e- ew.ee- m~.o em.m mo.m meme maze o~.m~- mm.-- om.e~- oe.en ce.oe em.e weoe .00: oe.m~- oo.-a om.oe ew.m~ oe.ne Neme acne me.meu mo.em am.oe ao.m~ eeme .uwo om.-e mm.em mo.me eeme «can me.me oa.ee came .uoo oo.o~- ceoe mesa mwwm www. meae meme meoe eeo. weoe ceme wee. meme weo. eeoe Meme neo— Neoe New. eeoe eeoe ceme cmme oe\e\e a own case owo wcae owe ucae owo maze owo ocae .005 maze .umo mesa .um: 9:33 .umo 0:26 .csaumg we mean; mo~go>o enacca-esomu-umuceewu ~4aw "uoxcae euuou 0;» L05 xwuce amazoemzumaea>uu.m.e venue 119 investments to papers that offered fixed income, thus avoiding risky investments. It will take time for confidence in the stock market to be restored. Conclusions The conclusions from this chapter are simple and may be stated as follows: From the comparison between the value-weighted index and the equally weighted index, it is apparent that small com- panies did better than larger ones. This conclusion is supported by the fact that a few large companies represented a very high percentage of the volume traded at the Sao Paulo Stock Exchange. The BOVESPA Index, despite its construction bias, seems to be a reasonable representation of the market. When compared with the value-weighted index, which is considered theoretically more sound, the difference was small. Common stocks did better than preferred as was expected, except during the bull market of l97l. This means that during the boom market, less risk was associated with more return. This is evidence of irrationality during this period. The bull market of l97l had a tremendous effect on the Brazilian stock market. The market has never recovered. Finally, the analysis of rates of return for 2l0 investment strategies during the lO-year period showed that investment in the stock market was not a good hedge against inflation. CHAPTER V CAPITAL-ASSET-PRICING MODEL: EMPIRICAL TESTS In financial literature, the stock market in the United States has been extensively examined. However, due primarily to the absence of machine-readable data bases, there are not many studies about stock markets in other countries. In this chapter, validity tests of the capital-asset-pricing model (CAPM), developed in the United States, are reproduced. These tests are discussed as they were developed by Black, Jensen, and Scholes and by Fama and MacBeth. The Risk-Free Rate of Interest in Real Terms To reproduce the test developed by Black, Jensen, and Scholes, as it is performed in real terms, it was necessary to know the monthly real return of the risk-free security. It was assumed that the indexed national treasury bonds (ORTN) were the risk-free security. Unfortunately, as shown in Appendix A, the ORTN indexation did not follow strictly the variation in the purchasing power of the cur- rency. Thus it was necessary to compare the nominal return from these bonds with the GPLI to obtain the real risk-free return. The follow- ing procedure was used. l. The ORTN pay 6 percent interest annually on a nominal value that is adjusted monthly. Thus, a monthly interest rate of 0.4868 percent was assumed. 120 121 2. An index was created as of January l, 1972, having 1000 as a base. At this time, the ORTN price was Cr$6l.52. Therefore, the value of the index at the end of January l972 was expressed as: 62. 26 1000x 61.52 x l,004868=1016.9552 where: Cr$62.26 = the ORTN value in February l972 The index at the end of February 1972 was expressed as: 63. 09 1016.9552 x 62. 26 x 1. 004868= 1035.5290 where: Cr$63.09 = The ORTN value in March l972 This procedure was followed, and 96 monthly index values were obtained for the period January l, 1972, to January l, l980. 3. The index obtained as described above was then inflated according to the variation in the GPLI. Thus the inflated index for January 1972 was expressed as: 37l4_ 1016.9552 302 =12506. 5285 where: 1016.9552 = The index in January l972 3714 = GPLI in January 1980 302 = GPLI in January 1972 The inflated index for February l972 was expressed as: 122 3714 1035.5290 +- 308 = 12486.8659 where: 1035.5290 = The index for February l972 4. Finally, the real return was obtained by a comparison between two consecutive real-index values. Thus, for January 1972, the ORTN real return was expressed as: 12486.8659 - l2506.5285 _ l2506.5285 - '0-001572 The ORTN real monthly returns are shown in Table 5.l. The CAPM Test Based on the Procedure of Black, Jensen, and Scholes The capital-asset-pricing model in real terms was expressed as in Equation 63. E(Rj) = E(RM) Bj where: E(Rj) = Expected real-excess return of security j E(R”) 83' Expected real-excess return of the market Beta measure of systematic risk The following generating function was used to perform the test: The coefficients bj were estimated by Equation 72 for a two-year period, from January l970 to December l97l, for all securities listed on the 55o Paulo Stock Exchange at the beginning of January 123 wemmeo.o- aeomwo.on momecc.ou moeeoo.ou moommo.on mmewmo.Ou oeeeec.On mnemoo.o oemaeo.o oeoeoo.o ememmo.on owoooo.o- meme meomoo.o- omcmec.o oeomoc.c emeooc.o wwmmoo.o e—emoc.o emeeoc.o omwooo.o- newmoo.o Nmmooo.o moeuoo.cu mmemoo.cu meme emmooo.o- ommmoo.o mmeooo.o- someoo.o- mmaooo.o oeomeo.c Reecec.o mmwmeo.o owmeoo.o mmeeoo.o- mmmmeo.Ou someoo.ou meme mmmmec.o- Nmomoo.o oeeoeo.o mmmoeo.o moomoo.c emeeoo.ou memooc.o- eececo.o oemcoo.c- eommoo.o- wwemoo.o- commeo.o- meme mmeeoo.o- cemmme.o cmmwwe.cu semeoo.o oweeoo.o emmcoo.o- cmecoo.o emeeoo.o eeoeoo.o wmeeoo.o emeeoo.o meeeoo.oa meme eOmmoo.ou mnemoo.o- memeoc.o Nmmueo.o eeoemo.o ooooeo.o memomo.o mmomeo.o mmewoo.cu oeomwo.o- oncomo.ou oowooo.ou eeme cemeeo.c- oeo—eo.o eeemoc.c mmmmco.o- meoeoo.o emmmoo.o omemoo.o mmomoo.o ommcoo.o eoemoo.o mmemoo.o Nowmoo.o meme mmowoo.o- oeeoeo.o moomoc.c Nmemco.o mmoooo.o- Nocmoo.o- meeooc.o cemmeo.o commeo.o eommoo.o cemmoc.c Nemeoo.o- Neae .umo .>oz .uuo .ugwm .oae xese meze em: eeeae seem: .nom .eow noeema llllilu'llizll. IIII‘.HI.MII:1II 11.”1I.'Il. :93... 1.--. N1'.1.lalvo|¢l11nvlll .- l0.|lll-l.u.t,n..i1a1l‘ljv . .l 0 I 1,111. .eeaawe eeme--umwgouce mo mums mmeeuxmemu-.e.m wenme 124 1972 for which 24 monthly returns were available. These securities were ranked on the basis of estimates bj and were assigned to l0 portfolios. The 10 percent of securities with the largest bj were assigned to the first portfolio, and so on. The return in each of the next l2 months (January 1972-December 1972) for each of the 10 portfolios was calculated. This entire process was repeated for securities listed at the beginning of January 1973 to estimate new coefficients hi to be used for ranking and assignment to the 10 port- folios. This process was repeated for each time period from January 1970 through December 1979. This procedure yielded 96 monthly returns on l0 portfolios. The data used in the tests were collected as described in Chapter II. All returns were in real terms computed as described before. The Time-Series Test of the Model Using Equation 72 and the 96 monthly returns, it was possible to calculate the least-square estimates of the parameters a5 and bj for each of the l0 portfolios. In Table 5.2 the following are pre- sented: the bj and “j estimates, the student t values of “j and bj, the coefficient of correlation between R. J tion of residuals, the average monthly excess returns, and the stan— and RM, the standard devia- dard deviation of the monthly excess returns. Note that the intercepts a were consistently positive but that the significance tests expressed by t values were greater than 2 in only five cases. The estimate-risk coefficients ranged from 0.67l0 to 0.l94l with t(b) always greater than 2 except in Portfolio 10, in which t(b) 125 eewo.o memo.o ammo. ecoo.o memo.o Nmmo.o eoeo.o memo.o emwo.o Neee.o b oweo.o omoo.o ooeo.o mmoo.o emoo.o wwoo.o eoeo.o: omeo.o meeo.o momo.o .m ammo.o oemo.o memo.o memo.o memo.o memo.o ommo.o emmo.o mm~o.o ommo.o Aemvo omme.o mmmm.o emee.o mmoe.o mmem.o meem.o mmem.o momm.o eeem.o mmom.o Azm.mva eeNN.N momm.e moom.N mmme.e meme.e eomm.m wemm.o memo.m mee.e mmam.m Aavp Nmom.e eewe.m meom.e mmem.e ewmm.m Name.m ommm.o omoo.m Nmem.m memo.m Aavu omeo.o meoo.o mmeo.o eaoo.o moeo.o mmeo.o emoo.o moeo.o omeo.o memo.o a eeme.o mmm~.o eemm.o mmom.o wmem.o emme.o mmee.o some.o Noem.o oeeo.o n oe m m m m m e m N e Empe emnszz oeeoeueoa .moepmepmwmuumpmmp mmeemm-meee--.~.m menme 126 was l.9032. Most surprising was the fact that no portfolio had a risk coefficient greater than one. In analyzing the reasons for this behavior, it would seem that the real return on the risk-free security and the computation of the real-market return were the causes of this fact because of the following: l. The indexation of the ORTN did not follow the variation of the currency-purchasing power, especially after 1975. (The nega- tive real returns for the risk-free security are shown in Table 5.l.) In the space representing security-excess return and market excess return, the points corresponding to the second half of the decade were depressed resulting in a reduction in the beta estimate. 2. The real-market return was computed according to the value-weighted index. The portfolios were formed on an equally weighted basis. As discussed in Chapter TV, the equally weighted index and the value-weighted index presented substantially different results. The equally weighted index was always ahead of the value- weighted index. This fact tended to increase the real returns of the portfolios and to reduce the market return, introducing a bias downward in the beta estimates. The correlation coefficients between the portfolio returns and the market returns were substantially lower than the coefficients obtained by Black, Jensen, and Scholes for the American market. The standard error of residuals was very small. The intercept « appeared to be nonstationary. 127 Cross-Sectional Tests of the Model The cross-sectional procedure provided an opportunity to analyze the linearity between risk and return without any specifi- cation being made about the intercept. The two-factor formulation of the capital-asset-pricing model was expressed as follows: ~. = ” - . + ~ . E(RJ) E(Rz) (1 BJ) E(RMMBJ (73) where: E(Rz) = Expected real return in the zero-beta portfolio The following generating function was used in the cross- sectional test: ~. = . + . . + ~. RJ 03 bJYJ eJt (74) The traditional form of the capital-asset-pricing model implied that aj = 0 and Yj = RM. The two-factor model required only that the linearity of Equation 73 be valid and that the intercept aj could be nonzero. The cross-sectional parameters of Equation 74 were estimated through use of the portfolio mean-excess returns over the period and of the risk coefficients obtained from the time-series procedure described in the previous section. The results obtained from the cross-sectional test are pre- sented in Figure 5.l and Table 5.3. The visual representation shown in Figure Sul does not imply that there was a linear relationship between risk and return. This implication is confirmed by the data in Table 5.3. The coefficients mi and yj had very low t values. The 128 AVERAGE EXCESS ”ONTHLY RETURNS 0.0 32 0.0 30 0.028 0.028 0.024 0.022 0.020 0.01 8 0.0 16 ODLQ 0.012 0.01.0 0.008 0.006 0.004 0.002 0.000 'QOOZ J j T T 1 T I 1 I | j OJ 0.2 0.3 0.4 0.5 0.6 0,7 0.8 0.9 no 1 svsreumc msx Figure 5.l.--Relationship between risk and return: cross-sectional test. 129 correlation coefficient and the F statistics were also very low. Because of the bias introduced in this test, it was not possible to conclude that the relationship between risk and return was not linear. The Fama and MacBeth procedure, discussed below, seemed to be more appropriate. Table 5.3.--Relationship between risk and return: cross-sectional test. .=oc.+ .b+ . R J Y.1 ea 3 Item Value «j 0.0044 vj 0.0171 t(“j) 0.6109 t(Yj) 0.9263 R 0.3112 R2 0.0968 C(Ej) 0.0085 F 0.8581 The CAPM Test: Based on the Procedure of Fama and MacBeth According to Fama and MacBeth, Equation 47 had three testable implications. First, the relationship between the expected return on a security and its risk in any efficient portfolio was linear. Second, 8 was a complete measure of risk, and third, in a market of risk-averse 130 investors, higher risk should have been associated with higher expected return: that is, E(Rm) - E(Rz) > O. The Methodology of This Procedure From data for the first two-year period (l970-7l), 20 port- folios were formed on the basis of ranked bi for individual securi- ties. The 5 percent of the securities with the largest bi were assigned to the first portfolio, and so on. From data for the follow- ing two-year period (l972-73), bi was recomputed, and these bi were averaged across securities within portfolios to obtain 20 initial portfolio 0 for the risk-return test. Thus, equal weights were pt applied to individual securities. The subscript t was added in bpt to indicate that, for each month t of the following year (l974), the b were recomputed as single averages of individual securities, thus pt allowing for delisting of securities. The month-by-month returns on the 20 portfolios, with equal weighting of securities, were also computed for 1974. For each month t of l974, the least-square method was used to compute: Rpt = Ylt + YZt bp,t-l + Y3t b§,t-l + Y4t5(ei p,t-l + ej (75) Rpt = Ylt + th bp,t-1 + Y3t b§,t-l + ej (76) Rpt = Y1t + th bp.t-1 T Y4t°(ei);,t-1 * 95 (77) Rpt = Y1t + Y2t bp,t-1 T ej (78) The variable b? was included to test linearity. The first testable implication mentioned above was that E(T3t) = 0. The term 131 involving 0(ei), represented some measure of risk of security i that was not related to 81' The second testable implication was that E(y4t) = 0. The disturbance term ad was assumed to be independent of all other variables. The explanatory variable bp,t-l was the average bi for securi- ties in each portfolio; b§,t-l was the average of the squared values of these bi; and C(61),t_] was the average of 0(9i) for securities in Portfolio p. The 0(9i) were sample standard deviations of market- model residuals for individual securities. The results from these equations were the time series of month-by-month values of y]t, th, Y3t’ and y4t. Results for other periods were obtained in the same manner for yearly intervals. Thus, there were six different portfolio- formation periods. These portfolio-formation periods, the initial- estimation periods, and the testing periods are shown in Table 5.4. All tests were predictive in the sense that the explanatory variables were computed from a period prior to the month of the returns. After this time, to facilitate the tests, the subscripts p and t were dropped. The Results of the Fama- MacBeth Procedure The relationships between return and risk are shown in Table 5.5. The relationships between return, risk, and average squared risk are shown in Table 5.6. The relationships between return, risk, and average standard deviation are shown in Table 5.7. Finally, the relationships between return, risk, average squared risk, and average standard deviation are shown in Table 5.8. The 132 meme meme neae meme meme eeme voegma aceumme weueeme meumeme oeumeme menemme emumnme muumeme coegma coepmseummuemeuece mesmeme meueeme enumeme meumeme Neueeme eeuo~me coeema coepesgoeuoeeoeueom o m e m N e .muoeema meeummp eew .coeueeepmm .coepmeeom oeeoepeomuu.e.m menme 133 owee.o oeee.o omoo.o mmeo.o come.o oemm.o mmeo.o Nome.o wwee.o eoeo.o eooo.o mooo.o meme emeo.o omeo.o mmoo.o oeoo.o ome~.o emem.o mmoe.o mooo.o mmoo.o mmeo.o oooo.o meeo.o meme oooo.o oeeo.o NmNN.o emmo.o eeee.o mooe.o mmmo.o eomo.o ommo.o Nooo.o mmoo.o eweo.o meme mmmo.o mneo.o eNNe.o some.o omoo.o Nmme.o emeo.o emmo.o nowm.o wmmo.o mono.o emoe.o meme eemo.o mmm0.o omoo.o eeoo.o oomm.o mmo~.o mmmo.o eomo.o memo.o memo.o meee.o ~mmm.o meme mmee.o moeo.o mmoo.o eomo.o mmoo.o meeo.o meeo.o Neeo.o memo.o meee.o omeo.o ... eeme mm oewo.o meme.o- emmo.o mono.o emeo.ou mwmo.o meeo.o- Nmmo.o- mmoe.o- mmmo.o mooo.o- emoo.o meme mmmo.o mmmo.o- memo.o wmeo.o- mmom.ou Nmee.o ewoo.o- mmoo.o mmmo.o mmeo.o- wooo.o memo.o- meme emeo.o emmo.o- memo.o emmo.ou memo.o mmeo.o memo.o mmmo.o- memo.o- mmoo.o mmeo.ou eeNo.o meme mo~o.o mmeo.o mowo.ou emee.o- mmmo.o- mmmo.ou eeeo.o mmmo.o meme.o memo.o wmmo.o mmmo.o meme emwo.o mmmo.o- mmuo.o- meeo.o- emme.ou meem.o emme.o emoo.o- oeeo.o emmo.o- mmee.o meee.o meme memo.o memo.o eeeo.o- Neeo.o- mmeo.o emoo.o mmoo.o mmeo.o- mmo¢.ou emmo.o memo.o- ... eeme N» meeo.o- eoeo.o mowo.o mooo.o smoo.o emoo.o- Nmmo.o- ooee.o meme.o Demo.o ommo.on wmoo.ou meme mmeo.o- mmeo.o momo.o- omeo.o mooo.o eeeo.o omoo.o- mmoo.o mmeo.o- emeo.o mmoe.o mmeo.o meme memo.o mooo.o memo.o “Neo.o memo.o eemo.o omoo.o mumo.o. emeo.o emoo.o emoo.o oooo.o Neme mooo.o ommo.o mmoo.o mmeo.o- memo.o mmeo.o mmoo.o ammo.o emmo.o- meeo.o- meoo.o mooo.o- meme oeoo.o mmeo.o ammo.o ammo.o emoe.o- emmo.o meme.o emeo.o meoo.o- emoo.o- ooeo.o- mono.ou meme memo.o- mceo.o ammo.o mmmo.o- ammo.o- meeo.o meeo.o mmmo.ou mmwo.o- memo.o- omoo.o ... eeme ; .umo .>oz .puo .pgmm .mae xeze mean he: eeeae cuemz .nmm .cme coeewm 5+3: I... ...m .xmee eee assume emmzuma necmcoeememe 0;» e0 eeoume emcee-»n-;peos use--.m.m m_nme 134 omwo.o Nuke.ou eoeo.o mmoo.o emeo.o- memo.o mmeo.o- memo.o- mmoe.o- mome.o mooo.o mmoo.o meme eemo.o eemo.o- eeeo.o um~o.¢- mmmm.o- mmmm.o mmeo.o- memo.o- mmeo.o emmo.o- owoo.o- eemo.o meme memo.o- mmmo.o- mmmo.o mmeo.o- eeeo.o mooe.o mmmo.o oemo.o- memo.o- eooe.o eoeo.o oemo.o meme owoo.o emee.o mmem.ou oeoe.o- emmo.o- owee.ou emmo.o memo.o omeo.o woee.ou mmmo.o ommo.o- meme emo~.o mome.ou woeo.o- mmme.o- Nome.o- emme.o mmme.o memo.o- ooeo.o memo.o moe~.o mmoe.o meme mmmo.o moeo.o- moeo.o- mmeo.o- Nmmo.o omoo.o mmeo.o ommo.o- memo.o- eeeo.o mmmo.o- ... eeme N» emoo.o- oeme.o omeo.ou oemo.o eewo.o mmeo.o- ammo.o- oeee.o Nem~.c wmoe.o mmeo.o- eeoo.o- meme emoo.o- eemo.o Nomo.o- mmoo.o- Neeo.o eemo.o weeo.o- memo.o- ommo.ou mmmo.o emmo.o mmmo.o meme ooeo.o mmmc.o ~o~o.o eeeo.o ommo.o memo.o eeoo.o memo.o memo.o mmeo.o emeo.o Neoo.o- meme meeo.o ooeo.o aomo.o ~moo.o- memo.o oomo.o eeoo.o- emeo.o meeo.o- omoo.ou wmoo.o- mmeo.o meme eemo.ou emoo.o weeo.o eeeo.o Neeo.o eemo.o- ammo.o meme.o emoo.o meeo.o- emeo.ou eowo.o- meme emeo.o- emmo.o mmmo.o Nmmo.o- eoeo.ou meeo.o mmoo.o memo.ou memo.o- ooeo.o- Nmoo.o ... eeme e» .umo .>oz .puo .uamm .mze xeze mcze me: eeeae gage: .nmu .eme eoeemm em + Nam» + em» + e» u we .xmee emgmzem mmmem>e eee .xmee .cezpme cmmzuma aezmcoeueem; ms» e0 eeomme gaeosuxangpcoe mceuu.o.m menee 135 emee.o emem.o eeee.o memo.o ooo~.o meom.o ewmo.o oeee.o wmme.o memo.o eoeo.o Neoo.o meme eeoo.o memo.o mmmo.o omeo.o omom.o meme.o eeae.o eoMe.o eemo.o «wwo.o mmoo.o emme.o meme Nmme.o mmmo.o oeNN.o mmeo.o owee.o mome.o emmo.o meme.o mmmo.o ooee.o owee.o mmeo.o Neme emmo.o emoe.o Neme.o mome.o omoo.o eemw.o owee.o ammo.o «wom.o emmm.o emmo.o meme.o meme mmee.o moee.o emoo.o omeo.o eoem.o emmm.o mme¢.o mmNe.o eeom.o eemo.o owee.o mmem.o meme ooe~.o mmme.o mmoo.o mmmo.o mmwo.o memo.o memo.o emmo.o oeoe.o Nme~.o Nee~.o ... eeme Nm eweo.o mmmo.o- memo.o omeo.o mmeo.o- mmoo.o mmoo.o om~o.ou ammo.o- memo.o- omoo.o meoo.o- meme oemo.o- memo.o- memo.o moeo.o memo.o emmo.o- oemo.o ammo.o memo.o mmeo.on eemo.o Nmmo.o- meme once.o Nmeo.o- owoo.o mmoo.o oeoo.ou memo.o- emoo.o emmo.ou eeoo.o- emeo.o- memo.o- mmoo.o- meme mmeo.o meeo.o- emeo.o emeo.o oooo.ou emmo.o memo.ou mmeo.o- emmo.o eeoo.o emeo.o- emee.o meme «mee.o- omeo.o eeeo.o mooe.o emmo.ou mmee.o eNoo.o emeo.o oemo.ou mmmo.o- mmmo.o- mmeo.o meme eomo.o emmo.o Noeo.o- meoo.o- ooeo.o oooo.o mmwo.o- mwoo.o- meeo.o- cemo.o e~mo.o ... eeme m» .umo .>oz .puo .uamm .mze xeze meze he: eeeqe gage: .amm .eee eoeema .emzcepcouuu.m.m menme 136 ewao.o ~eo~.o- emmo.o ommo.o emeo.ou memo.o omeo.ou emmo.o- meme.o- meeo.o Nooo.o eomo.o- meme memo.o “eeo.ou Nmeo.o ammo.o- ome~.o- emee.o «eeo.o- mmwo.ou weoo.o oeeo.ou mmoo.o emuo.o- meme eeeo.o mmmo.ou eemo.o omeo.o- emeo.o- oeoe.o meeo.o- omeo.o ww~o.o- mm~o.ou emeo.o memo.o “eoe emeo.o- eoeo.o o~oo.o omoo.o emoo.o moeo.ou memo.o mooo.o- emoo.o- eeee.o- mmoo.o mmee.o- meme Nmeo.o ammo.o- mooo.o ooeo.o emwe.o- mmem.o meme.o ommo.on mmeo.o mmeo.ou emme.o oeee.o meme mmoe.o mweo.o ommo.o- mmeo.ou emeo.o meoo.ou omeo.ou emeo.o- ooeo.ou memo.o oeoo.o ... eeme N» mmoe.o- mmme.o emoo.o- emee.o- emmo.o Nmoe.o- mmoo.o- owmo.o mem~.o omoo.o- eeoo.o- some.o- meme memo.o- eemo.o memo.o wmeo.oa enme.o omeo.o eoeo.o- emmo.ou meoo.o- mwoo.o- meme.o mmmo.o meme «eeo.o eeeo.o eweo.ou mmmo.o- memo.ou oeoe.o meeo.o- ammo.o mm~o.ou wmeo.ou emeo.o memo.o Nem— emeo.o- eoeo.o omoo.o omoo.o emoo.o Noeo.o- ~eeo.o wooo.on ewoo.ou Neme.o- m~oo.o mmee.o- meme eemo.o- eeeo.o- emee.o wmme.o eweo.o- memo.o ooe~.o- come.o meee.o mmmo.o owee.o eowo.o- meme eewo.o- mewo.ou eemo.o memo.o- moeo.o- memo.o womo.o mmeo.o- eeeo.o- meeo.ou eoeo.o- ... eeme ; .umo .>oz .umo .uzmm .az< aeze meze he: eeeae gage: .amm .eme eoeema mm + eemvo e» + am» + e» u em .coepme>me eemecmum momem>e eee .xmee .cezpmg :mmzumn aegmeoeumeme mew mo eeomme gpcosuxauzpeos mceun.e.m meame 137 omme.o emom.o Nmeo.o oem~.o Nume.o mmwm.o memo.o ewee.o mmee.o memo.o eeeo.o meme.o meme memo.o emmo.o eemo.o memo.o Noem.o emem.o mmoe.o memo.o eooo.o eeeo.o omoo.o meeo.o meme mmoo.o eemo.o Nmem.o ommm.o eemm.o eeme.o eoeo.o emmo.o mmme.o memo.o moeo.o emmo.o enme mooo.o eemo.o emme.o meme.o mmoo.o emm~.o mama.o Neeo.o emm~.o mmmm.o Nomo.o Neme.o meme mmoo.o eeeo.o wmmo.o emme.o emmm.o eme~.o omme.o memo.o mwoe.o mmeo.o moom.o emmm.o meme emmm.o oeeo.o meoo.o «Noe.o ammo.o oooe.o oeeo.o meeo.o memo.o emee.o owee.o ... enm— mm meem.o moem.o- eeme.o emoe.e Nmoo.o- momN.o memo.o ~mm~.o eeom.o- memm.o meme.o emoe.o meme emee.o w-e.o mmmm.o- Nemm.o meem.o- emoo.o- omeo.o mNNe.o memo.o Nowe.o emoe.o- mmme.o meme Neeo.o- meme.o mem~.o eeom.o owew.o emue.ou mmee.o Nmee.o- oeom.o eemm.c emme.o- wmm~.ou meme mmee.o meem.o emoo.o meme.o- ooee.o oemm.o emee.o- emwe.o mome.o- emeo.e memo.o- emew.o meme mome.o ewo—.o Neoo.e- mmmm.o- NNom.o eme~.e- memo.m meee.o- omem.o- mmem.o- meee.o- mmeo.o meme enme.o mewm.o memo.o- Nume.o- omeo.o Noom.ou meeo.o- eeeo.o- emmo.o- eewo.o omme.o ... eeme e» .umo .>oz .uuo .pamm .mze xeze meze he: eeeme euemz .nmm .cme eoeemm .emacwpeou--.~.m azame 138 eeeo.o oamo.ou memo.o eNmo.on mNeo.o- emoo.o meoo.o o~mo.o- eemo.o- meeo.ou owoo.o mono.ou meme emmo.ou ammo.o- meeo.o mmmo.o mNmo.o mNoo.ou ammo.o ommo.o mmNo.o oemo.ou mmNo.o emeo.o- meme meee.o ooNo.ou eeoo.o ooeo.o- emmo.o- mmeo.ou mNoo.o- mmNo.ou ewNo.o- eoNe.ou moeo.ou oeoo.o meme eeNo.o mmoo.ou moo0.o eoNo.o Neoo.o- emeo.o eeeo.ou mNNo.o- wNeo.o omco.o omeo.ou mNeo.o meme eeee.o- mNeo.o emee.o emeN.o mmee.ou meem.o momN.on memo.o mNmo.on emeo.o- emNo.on mmmo.o meme meNo.o meoe.o mmoo.o- ammo.o emmo.o mmNc.ou meeo.o- mmeo.ou Nmmo.ou memo.o meoo.o ... eeme m» emmo.o emee.ou meeo.o meoe.o mNeo.ou ommo.o- meeo.ou Nomo.o- wwee.o- mmeo.o mmoo.o mwNo.ou meme omeo.° emmo.ou oeeo.o mmmo.ou momN.ou eeee.o MNeo.ou memo.ou eoeo.o Nemo.o- emoo.o wmmo.o meme emmo.o- mONo.ou mmmo.o Nmmo.o- eoeo.o memo.o emNo.o NeNo.ou NONc.ou meNe.o emeo.o ammo.o meme eeeo.o mmNe.o mOON.ou mNme.ou eoNo.ou eeee.o- memo.o memo.o emmo.o owee.o- NemN.o emoo.ou meme moNN.o mmme.ou mNee.ou eooN.ou emmo.ou eeNo.o- emom.o mmoo.ou mmeo.o NNmo.o oeme.o emeo.o meme meeo.o mmeo.ou oeNo.o- emee.o- Neoo.o emoo.o eNeo.o eoNo.o- NmNo.ou mmeo.o mooo.ou ... eeme N» Neoe.o- mNce.o emoo.o- Nmee.o- Newo.o eoee.o- MNeo.o- eemo.o mweN.o mNNo.o mooo.on mome.o- meme mNee.o- mmoo.o- ecmo.o eNmo.o- eNeN.o eeoo.o Nooo.o ooeo.ou wemo.on emoo.o- Nome.o oeNo.o meme eo~o.o eomo.o Neeo.o1 owmo.ou emeo.ou mNmo.o wweo.o- memo.o eeeo.o- emee.ou eeeo.o momo.o meme NNNo.o Neme.ou omme.o Nooo.o mmoo.ou ONNo.o emNo.o moeo.ou ammo.o memo.ou mNeo.ou ero.o- meme meoe.o- eoNo.o eoeN.o NoeN.o mmme.ou emmN.o mocm.o- emeN.o wmwo.o eoeo.o oeoe.o mNmo.o- meme Neoo.o- Nemo.o eemo.o Nmoo.o mNeo.ou Neeo.o moeo.o- memo.o- emoe.o- memo.ou Nmmo.o- ... eeme S .mmo .>oz .puo .unmm .mze xeze mcze em: eeea< goes: .nmm .cee eoeemm mm + Aemvb e> + am» + nN> + e> u nm N .coepme>me eemeeepm mmmem>e eee .xmee ememzcm mueem>m .JMeg .cgzpme cmmzumn aegmcoeumeme mgu mo egomme cucosuznucecoe mcenu.m.m meaee 139 NeON.o omem.o come.o mooN.o eOON.o waN.o meeo.o came.o meoN.o eeee.o weNo.o moee.o meme eeme.o wmo0.o ewee.o oeeo.o mmwm.o omem.o eeoN.o woee.o mmeo.o meme.o moeo.o emme.o meme owee.o emeo.o emeN.o moeN.o mmee.o eoee.o eeeo.o oeNe.o wmwe.o eoem.o oewe.o emmo.o meme mmmo.o meeN.o ommm.o meme.o mmoo.o emmN.o meme.o eoee.o ewNe.o meem.o omeN.o mmeN.o meme mwme.o meee.o momo.c mmmm.o mee.o mNNm.o eeme.o meme.o NMON.o memo.o Nmom.o momN.o meme eemN.o meme.o meoo.o mmme.o meee.o oemo.o cowo.o memo.o mmee.o omeN.o owee.o ... eeme Nm emmN.o mmmo.on mmmo.on meeN.e mNoo.ou omoN.o ewmo.o emmm.o mooN.Ou ewoe.o mmoe.o mmmw.o meme moee.o mmme.o mmem.ou NOON.o emmm.eu omeo.o oeee.Ou Neee.o emwe.o mmem.o mmwN.ou mewm.o meme ONee.ou emeN.o weeN.o Nmmo.o emem.o memm.on meme.o moeo.ou meem.o memo.e mmme.ou mNmN.on meme ommo.o: mmNo.e mee.Ou ewem.ou meNN.o meme.o emeN.On mome.o Nwmm.ou came.o mmme.on memo.o meme omwe.o memo.o: NeNe.en mmNo.en mooN.e eeem.Nn mmme.N eme.on Nmme.ou omeN.ou NNem.eu meNe.ou meme ommo.o wmme.ou eeeo.on eoeN.Ou NmNe.on meNo.ou memo.o meeo.o oeee.o mmee.ou mmme.o ... emme e» .omo .>oz .uuo .uamm .mze zeze meze an: eeeae seem: .nmm .eme eoeemm .emzcwpcou--.m.m 6_nme 140 month-by-month values of R2 , which were the coefficients of deter- mination, are also given in these tables. The major tests are summarized in Table 5.9. The following statistics are given in the table: 73, the average of the month-by- month least-squares values of vj; 0(yj), the sample standard devia- tion of the monthly th;‘62 and C(02), the mean and standard deviation of month-by-month coefficients of determination. Also shown is the first-order autocorrelation of the various yj labeled 6(y5). Finally, the t values are presented. The results presented in panels C.and D were consistent with the hypothesis E(i4) = 0. The t values for the mean values of Y4 were small. Therefore, the hypothesis that beta is a complete measure of risk cannot be rejected. The results presented in panels 8 and D did not reject the proposition that the relationship between return and systematic risk was linear. In panel B, the t value for y3 is 0.6l762 and in panel D, 0.26lOl. Thus, the results were consistent with the hypothesis E(§3) = 0. The autocorrelations of Y3 and Y4 were always low, so there was no information in the time series about past values. This result was consistent with a market that is effi- cient. The average coefficient of determination 52 was small in all panels. The small values t(yé) reflected the variability of 72' Fama and MacBeth found large values for t(Yz) only for the entire period of their study, 34 years. Thus, in the present study, which included data for only 10 years, no statistically significant relationship between risk and return was found. 0n the other hand, mocme.o Nemwe.o weewo.o eoemN.o mmwwo.o emomeguweQON.oleNcwm.0gmemmo.oloeNeN.o- Nmeee.o NNmeo.o emcee.o NMNee.o wommc.o mMNoo.o Ne—oo.o meooo.o o em +136: + n~>+ aw» + _> n we N moeoo.c eNmNe.o omNem.e ... mmwmm.o omNemgucoeme.OL ... femwe.olmommc.o: enmem.o ... mmmwo.o emmmc.o omewc.o ... nmmoo.o eomco.o u em + 2363., z~>+ I ... .z 141 moeee.o mmwme.o ... Nee—o.o womMN.o memem.N ... CNmae.o.ememe.o.m¢o0m.o ... meeeo.o me¢a0.o ommmo.c ... ommoo.o meNoo.o wmoeo.o m hm + Nam» + 3N» + e» u em mnemo.o emeeo.o ... ... Nemme.o scone.N ... ... mNewo.0+ meemm.o ... ... wmmwo.o oeNmo.o ... ... woeoo.c Nemeo.o < H0+DN>+~>D¢ e38 N. 2...: 2605.6. E: 3: 1:... .3“. 2% 2:... 3... 38 2:. f m N. : :5 Trilfillll p 11' I 1 III. aux . “hid? M rurflvl.l. N. n F ”huthwfilfluhurluptph. n .1 trA'IIEVFUHT E “HULL .Elflvlvhl...‘ In? I. I H h 1| Eli . .meos 83253-03 mfi yo mumme--.m.m menme 142 as shown in all panels, Y2 was consistently positive. This is what was expected in a market of risk-averse investors. CHAPTER VI THE RELATIONSHIP BETWEEN INFLATION AND STOCK PRICES In this research, the relationship between stock returns and inflation was studied by linear regression. having as a dependent variable the monthly real return in the value-weighted index and as independent variables the anticipated and the unanticipated inflation. The anticipated inflation was measured by the inflation two months previous to the time in question. Alt = “’2 OH (79) pt-3 where: pt = Inflation index at time t The unanticipated inflation was measured by the real rate of infla- tion less the anticipated rate of inflation. UIt = (RIt - Alt) (80) This model was expressed as: Rmt = a + bAIt + cUIt + E (81) t Anticipated Inflation As mentioned before, the anticipated inflation was calcu- lated in a nontraditional way. It was not possible to use the 143 144 one-month return on the nonindexed treasury bill as a measure of inflation as Fama1 did because of the enormous discrepancy between this rate and real inflation. For example, in June, July, and August 1980, inflation was over 6 percent a month in Brazil, and 30-day treasury bills were sold at a discount of approximately 3 percent with no indication that there would be any spectacular decrease in inflation. The only reason for this behavior was the lack of other investment opportunities or of government intervention to lower interest rates. A lag of two months was necessary for computing anticipated inflation because the inflation indices for each month are published at the end of the following month. Therefore, inflation indices may be acted upon only after a lag of two months. Results The month-by-month variations in the GPLI are presented in Table 6.1. The relationships among monthly real return in the value- weighted index, anticipated inflation, and unanticipated inflation are shown in Table 6.2. The relationship between monthly real return and total inflation is also shown in Table 6.2. As can be seen, no statistically significant relationship was found between return and inflation. The t values of the coefficients b and c were always very small. The hypothesis cannot be rejected that there was no relationship between inflation and real return from 1Eugene F. Fama, "Short Term Interest Rates as Predictors of Inflation,“ American Economic Review 65 (June 1975): 269-82. l45 ONmo.o mmeo.o emmo.o mNmo.o eeeo.o memo.o mmeo.o eemo.o omNo.o eemo.o ommo.o memo.o meme memo.o emeo.o meNo.o mmNo.c omNo.o moNo.o mwNo.o oemo.o eNmo.o mmmo.o mNmo.o eemo.o meme emNo.o eeNo.o NmNo.o meNo.o omeo.o eNeo.o oeNo.o mmeo.o ammo.o moeo.o eoeo.o mNmo.o meme memo.o mmNo.o oweo.o meNo.o ammo.o moeo.o memo.o eoNo.o emmo.o eemo.o memo.o eoeo.o meme memo.o NONo.o NNNo.o oeNo.o meNo.o meNo.o oeNo.o eNNo.o moNo.o oneo.c eoeo.o oNNo.o meme mNNo.o oMNo.o mmeo.o mmeo.o emeo.o NNeo.o mNeo.o mmeo.o Nemo.o mNmo.o mmeo.o oeNo.o eeme NNNo.o eNeo.o moeo.o emeo.o ooeo.o eoeo.o mmeo.o oceo.o oeeo.o oeeo.o Neeo.o meeo.o meme meeo.o mmoo.o mmoo.o mmoo.o eNeo.o mmeo.o eNeo.o emoo.o mmoo.o mNeo.o omeo.o mmeo.o Neme moeo.o Noeo.o moeo.o eoeo.o eeeo.o Noeo.o eeeo.o eNNo.o wmeo.o mmeo.o emNo.o oeoo.o eeme eeeo.o emoo.o eeoo.o ooeo.o NeNo.o eeNo.o eeeo.o mNNo.o emeo.o meoo.o emeo.o Neeo.o oeme .0mo .>oz .uuo .uamm .mze aeze meze an: eeea< emeez .nmm .eee eoeemm .ezmw ms» ce m:o_zm_em>--.P.o 6.88e 146 risky assets. Brazil has an indexed economy, and in an indexed economy the effect of inflation is minimized. All long-term loans are indexed as are most of the long-term receivables. Table 6.2.--Relationship between inflation and stock prices. RMt = a + bAIt + CUIt + ei Period a b c t(b) t(c) 70-73 -0.0162 1.9527 1.3035 0.4543 0.4553 74-79 0.0389 -1.3487 -O.8197 -1.7873 -1.1390 70-79 0.0262 -0.9581 -O.5017 -1.2657 -0.6052 RMt=°c+bI+ei Period « b t(b) 70-73 -0.0061 1.2470 0.4538 74-79 0.0305 -1.0574 -1.7157 70-79 0.0219 -O.7642 -1.1935 Unanticipated inflation could influence the results of a company only to the extent that the company is a net debtor or creditor on a short-term basis. The expectation of inflation in the short run is generally accurate, and so this influence seems to be minimal. This analysis is valid to the extent that inflation affects in the same way cost and value of sales. In the aggregate, it seems to be reasonable to suppose that sales prices and cost will increase in the same way as inflation. This being so, it would be reasonable to anticipate no 147 significant relationship between inflation and real returns in an indexed economy. CHAPTER VII SUMMARY AND CONCLUSIONS Summary The structure of the Brazilian capital market was presented in Chapter I. As discussed, the 1964 political revolution altered the economy. Prior to l964, the economy had grown very quickly. How- ever, despite this fact, lack of adequate structure in the capital market had provoked severe distortions. The usury law had forbidden interest on savings over l2 percent per annum. With high levels of inflation, the real interest paid to savers was negative, providing a severe disincentive to investing funds in fixed-income paper. The inflationary climate, together with a government inclined to the left, created an unbearable political situation. In l964 the revolu- tion brought into being a new government that proposed to reorganize the country, reinstate economic growth, and reduce inflation. The most important changes in the capital market occurred under the first government after the l964 revolution. Law 4357 of June l6, l964, created the indexed national treasury bonds. Law 4595 of December 31, l964, the Bank Reform Law, introduced a complete refor- mulation of the national financial system. This law defined the role of the national financial system, specifying its components as follows: Conselho Monetario Nacional (CMN) (National Monetary Coun- cil), Banco Central da Republica do Brasil (BACEN) (Brasil Central l48 149 Bank), Banco do Brasil (Bank of Brazil), Banco Nacional de Desenvol- vimento Economico (BNDE) (National Economic Development Bank). Other private and public financial institutions were also established. Law 4728 of July l6, l965, assigned to the National Monetary Council the function of regulating the capital market and, to the Central Bank, its control. This law created new agents, organized the capital- market distribution system, and determined the institutions that could act in this market. Law 4380 of August l964 created and regulated the Sistema Financeiro da Habitacfio (Housing Finance System). In 1976, the ComissEo de Valores Mobiliarios (CVM) (Stock Exchange Com- mission) undertook most of the regulatory functions of the stock market. In the first years after l964, inflation was reduced at the cost of a serious slowdown in economic growth. In l967, the second government of the revolution tried to stimulate the economy, having inflation reduction as only a secondary target. During the period l967-73, the gross national product grew at an average annual rate of 10.5 percent. Inflationary rates were gradually contained, decreas- ing from 28.2 percent in l967 to 15.4 percent in l973. However, the strong external pressures to increase prices caused a return to the growth in inflation. By l975, inflation was already 28.7 percent, reaching ll0 percent in 1980. After 1973, economic growth continued but at a slower rate. The Brazilian experience after l964 showed that high infla- tion rates may be associated with an orderly development of the capital market. The main reason for this seemed to be the indexation 150 of the Brazilian economy. The origin of indexation was Law 4357 of l964, which created the indexed national treasury bonds and applied the indexation principle to overdue fiscal debts and companies' balance sheets. Law 4380 of l964 extended the concept of indexation to loans provided by the national housing system. Afterwards, this principle was extended to several other sectors, including savings, loans, debts, rents, insurance, interest, wages, the real estate sector, and the exchange rate. In the Brazilian economy now, the state controls a high pro- portion of financial savings, taking in the funds through compulsory savings or through assets yielding interest and subject to monetary adjustment. Law 157 of 1967 instituted the fiscal mutual funds that allow individual taxpayers to deduct from 12 to 24 percent of their tax bills (depending on their gross incomes) and to invest this money in the purchase of fiscal-fund quotas. The fiscal funds, by reason of the volume and constancy of the resources that they operate, are among the most important investors in the market. However, the con- centration of a large volume of stocks with relatively few institu- tions can reduce the efficiency of the market. Chapter I also showed the objectives of this research, which were as follows: l. To build a machine-readable data base of price information for securities traded at the $50 Paulo Stock Exchange in the past decade; 151 2. To analyze the rates of return for common and preferred stocks traded at the S30 Paulo Stock Exchange for the period covered by the study; 3. To reproduce the tests developed by Black, Jensen, and Scholes and by Fama and MacBeth with new data from the $50 Paulo Stock Exchange and to compare the findings; 4. To determine the relationship between inflation and returns from risky assets in an indexed economy. The methodology for collecting the data was presented in Chapter II. The period covered by this study was from January l970 to December l979. Therefore, there were data collected on each stock and monthly returns for listed securities since January l970. The total universe of securities traded on the S50 Paulo Stock Exchange was searched. To compute the monthly security returns, it was assumed that the ideal investor received dividends and reinvested them in the same securities, buying shares quoted ex dividend. It was also assumed that, in the case of subscription rights, the ideal investor subscribed all shares corresponding to his rights and immediately sold the exact number of stocks quoted ex subscription rights neces- sary to recover the amount spent. Adjustments were also made in the case of stock dividends, stock splits, and all possible combinations of these events. Through the use of two deflators, two sets of real returns for each stock were obtained. Commissions and income taxes were not considered. 152 Information about 395 stocks was collected from 203 companies. The stocks included in the data file represented about 90 percent of the volume traded on the $50 Paulo Stock Exchange. Two magnetic tapes are now available. The first gives basic information with no further computations. The second tape offers as a final product the monthly returns of the securities in real and nominal terms. A review of the literature was presented in Chapter III. The choice of the proper weighting system and the methods of computing the value-weighted index and the equally weighted index in nominal and real terms were discussed in this chapter. Then, a brief review of the theory of the capital-asset-pricing model in its single-factor and two-factor formulations was presented. The tests realized by Douglas; Lintner; Miller and Scholes; Black, Jensen, and Scholes; and Fama and MacBeth to test the validity of the capital-asset-pricing model were discussed in some detail. Finally, studies about the rela- tionship between inflation and return from risky assets were mentioned. The performance of the market was presented in Chapter IV. The value-weighted index and the equally weighted index in real and nominal terms were presented for the entire period of this study. The behavior of the BOVESPA index (the official index of the $50 Paulo Stock Exchange) was compared with the behavior of the value- weighted index. The performance of preferred and common stocks was shown through the computation of the value-weighted index in real and nominal terms. Finally, the behavior of several sectors and sub- sectors was presented. 153 In Chapter V, the tests realized by Black, Jensen, and Scholes and by Fama and MacBeth were reproduced with the new set of data. For the procedures of Black, Jensen, and Scholes, the coefficients bj were estimated for a two-year period, from January 1970 to December l97l, for all securities listed on the $50 Paulo Stock Exchange at the beginning of January l972. These securities were ranked on the basis of estimates bj and were assigned to l0 portfolios. The l0 percent of the securities with the largest bj were assigned to the first portfolio and so on. The returns in each of the next 12 months for each of the l0 portfolios were calculated (from January 1972 to December l972). This entire process was repeated for securities listed at the beginning of January l973 to estimate new coefficients hi to be used for ranking and assignment of securities to the l0 port- folios. This process was repeated from January 1970 through December l979. This procedure gave 96 monthly returns on l0 portfolios. For the Fama and MacBeth procedure, using the first two years of data (l970-8l), 20 portfolios were formed on the basis of ranked bi for individual securities. The 5 percent of securities with the largest bi were assigned to the first portfolio and so on. Data from the following two-year period (l972-73) were then used to recompute bi’ and these bi were averaged across securities within portfolios to obtain 20 initial portfolio bpt for the risk-return test. Thus, equal weights were applied to individual securities. The month-by- month returns on the two portfolios, with equal weighting of securi- ties, were also computed for l974. For each month t of l974, the least-square method was used to compute: 154 _ . 2 ” Rpt ‘ Y1t + Y2t bp,t-1 + Y3tbp,t-1 + Y4to(ei)p,t-l + 90 2 pt = Ylt + th bp,t-1 + Y3t b pt ' Ylt + Y2t p,t-l + Y4t°(ei)p,t-l + ej b + e pt ‘ Y1t + th p,t-l j. The explanatory variable bp,t-l represented the average bi for securities in each portfolio; b§,t-l was the average of the squared values of these bi; 0(ei)’t-l was the average of 0(ei) for securities in portfolio p. The 0(ei) were sample standard deviations of market-model residuals for individual securities. The results from these equations were the time series of month-by-month values of Ylt’ YZt’ Y3t’ and Y4t‘ To get results for other periods, the above steps were repeated for each year. The relationship between inflation and risky assets was analyzed in Chapter VI by a linear regression having as dependent variable the monthly real return in the value-weighted index and as independent variables the anticipated and the unanticipated inflation. The anticipated inflation was measured by the inflation two months previous to the time being studied. The unanticipated inflation was measured by the real rate of inflation less the anticipated rate of inflation. 155 Conclusions From the comparison between the value-weighted index and the equally weighted index, the conclusion was reached that small com- panies did better than larger ones. This conclusion was supported by the fact that a few large companies represent a very high percentage of the volume traded at the S50 Paulo Stock Exchange. The BOVESPA index, despite its construction bias, seemed to be a reasonable representation of the market. When compared with the value-weighted index, which was considered theoretically more sound, the difference was small. Common stocks did better than preferred, as was expected, except during the bull market of l97l. This meant that during the boom market, less risk was associated with more return. This was evidence of irrationality during this period. The bull market of l97l had a tremendous effect on the Brazilian stock market. Since then, the market has not recovered. Finally, the analysis of rates of return for 2l0 investment strate- gies during the l0-year period showed that investment in the stock market was not a good hedge against inflation. The results of the test procedure of Black, Jensen, and Scholes did not show a linear relationship between risk and return. However, because of bias introduced in this test, it was not possible to conclude that the relationship between risk and return was not linear. Nor was it possible, from the results of the procedure of Fama and MacBeth, to reject the hypothesis that the relationship between risk and return was linear. Neither could the hypothesis be 156 rejected that beta was a complete measure of risk. Despite the fact that the coefficients of beta were greater than zero in all four cases, they were not statistically significant. Therefore, no statistically observable positive relationship between expected real return and risk was found. The results of this test were similar to results of tests by Fama and MacBeth for similar periods of time. Fama and MacBeth found a statistically significant relationship between risk and return only for the entire period of their study--34 years. In addition, no statistically significant relationship was found between real return and anticipated inflation, unanticipated inflation, or total inflation as could be expected for an indexed economy. APPENDICES 157 APPENDIX A METHOD OF COMPUTING THE ORTN VALUE 158 APPENDIX A METHOD OF COMPUTING THE ORTN VALUE Law 4357 of July l6, l964, which created the indexed national treasury bonds (ORTN), made the following stipulations: that the bonds have a nominal value of Cr$l0.00 and that they be indexed every three months based on variations in national currency purchasing power. Therefore, initially, the indexed bonds were corrected only every three months; that is, the ORTN first-month value was maintained during the entire trimester. In September l965, indexed bonds with monthly indexation were created, gradually replacing the former ones, which were withdrawn in l973. Interest, which presently varies from 6 to 8 percent accord- ing to the ORTN's term, is paid every six months or annually, through the Banco do Brasil. Changes in the Formula of the Calculation a. January l965 to September l969 The first method used to estimate the nominal value readjust- ment of the ORTN represented a moving trimestral average, with a semester time gap and a base period of April, May, and June l964. The wholesale-price index was used as the adjustment base. The initial calculation formula was the following: 159 160 (It-4 + It-S + It-6) x l/3 Vt = 10°00 x ”4.64”5.64”6.64) “’3 where: Vt = ORTN value in month t 10.00 = ORTN initial value It-i = Wholesale-price index estimated for month t-i 14.64 = Wholesale-price index for April 1964 15.64 = Wholesale-price index for May l964 16.64 = Wholesale-price index for June l964 From the above formula, the following simplifications can be performed: 14.64 + 15.64 + 16.64 = 2517 + 2568 + 2678 = 7763 (I + I + I ) _ t-4 t-5 t-6 _ v = (It-4 + I,c_5 + It_6) x 0.001288 b. October l969 to November l972 The Fundacfio Getfilio Vargas, which estimated the wholesale- price index, divided the index into the wholesale-price index of total availability and the wholesale-price index of internal availa- bility. The latter was used to estimate ORTN variations with the period July, August, and September 1969, as the base. t-5 t-6 + I + I I Vt = vt-1 x I + I + I t-6 t-7 t-4 t-5 161 vt = (1t_4 + It_5 + It_6) x 0.073524 c. December l972 to March 1974 Beginning in this period, the index was estimated in a weight- ing system of two parts: the first represented the actual indexation estimated according to the above formula, using a weighted value of l/2; the second, also with a weighted value of l/2, reproduced the inflationary residual foreseen by the government at l2 percent per annum. The nonpublished formula could be expressed as: I t-l x I + I + I t-6 t-7 t-5 t-6 t-4 t—5 + 0.5 V - 1.00949 Vt = 0.5 V + I + I t-1 d. April l974 to July 1975 In this period there was a return to the former system; that is, the weighting system that introduced the desired inflation in the index was withdrawn. I t = vt-l x I + I + I t-6 t-7 t-5 t-6 t-4 t-5 V +1 +1 e. August l975 to June l976 In August l975, the government included a new alteration in indexation. The concept of accident was included to eliminate from the index price increases provoked by accidental events, such as climatic catastrophes or sudden alterations in imported-merchandise prices. 162 f. After July l976 New alterations in the form of the calculation were performed in July 1976. Government authorities, hoping that inflation would initiate a period of decreasing rates, introduced two measures to reduce the impact that fed inflation. l. The indexation index time gap was reduced to two months. 2. Once again, a weighting system was introduced in the index to reflect both the inflation that actually occurred (80 percent) and the inflation target of 15 percent per annum. The new formla was expressed as: I I + I + I t-3 t-4 t-2 t-3 t‘4 + 0.2 (1.011715) t-5 V = V x 0.8 t t-1 + I + I where (l.0l7l5) is the adjustment of an inflation rate of l.l7l7 percent per month, that is, 15 percent per annum. SOURCE: ORTN Indexation, Calculation Forms, Special Study, Conjuntura Econ6mica 28 (March l978): 92-95; and Luna, "Indexation and Capital Market: The Brazilian Experience." APPENDIX 8 METHOD OF COMPUTING THE GENERAL PRICE-LEVEL INDEX 163 APPENDIX 8 METHOD OF COMPUTING THE GENERAL PRICE-LEVEL INDEX The general price-level index is the weighted average of three indices that are as follows: 11.9.09. New. Wholesale-price index Consumer-price index 000001 Civil-construction-price index On the other hand, the three basic indices mentioned above are calculated as follows: a. Wholesale-price index The wholesale-price index is based on the following weighting system: m l. Consumer goods 55.8450 Durable 4.5956 Nondurable 51.2494 2. Production goods 44.1550 Raw materials l8.0954 Construction materials 7.7299 Machinery and vehicles 5.9434 Others 12.3863 TOTAL 100.0000 164 165 b. Consumer-price index The consumer-price index is based on the following weighting system: we Domestic food 37.5428 Nondomestic food 4.0179 Clothing 5.4440 Housing 14.4632 Housing goods 10.8366 Hygienics and medical care 4.2158 Personal services 13.7784 Public services 9.7013 TOTAL 100.0000 c. Civil-construction price index The civil-construction price index is based on the following weighting system: Weight Materials 58.9753 Labor 41.0247 TOTAL 100.0000 APPENDIX C DIVISIONS AND CODES OF THE SECTORS 166 APPENDIX C DIVISIONS AND CODES OF THE SECTORS l0.000--Industry 11.000--C0nstruction 12.000--Mining 13.000--Manufacturing Industry 13.010--Food, Tobacco, and Beverages 13.020--Cement 13.030--E1ectronics 13.040--Wood and Furniture 13.050--Transportation Materials l3.060--Mechanics l3.070--Metalurgic 13.071--General Metalurgic Products 13.072--Forging 13.080--Paper and Graphic 13.090--Chemistry l3.09l--Fertilizer 13.092--0i1 13.093--P1astics 13.094--General Chemical Products 13.100--Texti1es and Clothing 13.110--Other Manufacturing Industries 167 168 20.000--Trade 21.000--Domestic Equipment 22.000--Oi1 Dealers 23.000--Automotive, Machines, and Equipment 30.000--Services 31.000--Financial Institutions 31.0lO--Commercia1 Banks 31.011--Government Banks 31.012--Private Banks 31.020--Investment Banks 31.030--Credit, Financing, and Investment Companies 32.000--Public Utilities 32.0lO--Communication 32.020--Electrical Energy 32.030--Transportation 90.000--Misce11aneous APPENDIX D TYPES OF STOCKS 169 APPENDIX D TYPES OF STOCKS Tape Code Market Code Type 01 PP Nonregistered preferred stock 02 OF Nonregistered common stock 03 PN Registered preferred stock 04 ON Registered common stock 05 PPA Nonregistered preferred stock Type A 06 PPB Nonregistered preferred stock Type B 07 PPC Nonregistered preferred stock Type C 08 OE Registered common stock with no par value 09 PE Registered preferred stock with no par value 170 APPENDIX E COMPANIES INCLUDED IN THE TAPE 171 APPENDIX E COMPANIES INCLUDED IN THE TAPE Tape Sector . Code Classifi- Name in the Tape Full Name of the Company cation 001 13072 ACESITA Companhia de Acos Especiais Itabira--Acesita 002 13080 AGGS AGGS--Inddstria Graficas S.A. 003 13100 ARTEX Artex S.A.--F5brica de Artefatos Texteis 004 13030 ARNO Arno S.A. 005 13072 ACOS VILARES Acos Vilares S.A. 006 13010 ANTARTICA Companhia Antértica Paulista Indfistria Brasileira BC 007 13093 ATMA Atma Paulista S.A. Indfistria e Comé rci o 008 13010 ANTARTICA NORDESTE IndGstria e Bebidas Antértica do Nordeste S.A. 009 13072 ANHANGUERA Acos Anhanguera 010 13010 ANDERSON CLAYTON Anderson Clayton S.A. Indfistria e Comércio 011 13110 ARTHUR LANGE Arthur Lange S.A.--Ind65tria e Comércio 012 13010 ACUCAR uurfio Companhia Uniao dos Refinadores de Acucar 013 90000 AUDI Audi S.A. AdministracSo e Participacio 014 13072 APARECIDA Indfistria Metalfirgica Nossa Senhora Aparecida 015 31012 BRADESCO Banco Brasileiro de Descontos 172 S.A. 173 Tape Sector Code Classifi- Name in the Tape Full Name of the Company cation 016 31020 BRADESCO INVESTI- Banco Bradesco de Investimento MENTO S.A. 017 31011 BRASIL Banco do Brasil S.A. 018 31012 AUXILIAR Banco Auxiliar S.A. 019 31011 BANESPA Banco do Estado 06 S30 Paulo S.A. 020 31020 COMIND BANCO DE Comind Banco de Investimento S.A. INVESTIMENTO 021 31012 NOROESTE DO ESTADO Banco Noroeste do Estado de 550 Paulo S.A. 022 31012 ITAUBANCO Banco Itau S.A. 023 31030 REAL DE INVESTIMENTO Banco Real de Investimento S.A. 024 31012 NACIONAL Banco Nacional S.A. 025 31012 AMERICA 00 SUL Banco América do Sul S.A. 026 31012 MERCANTIL DE 5A0 Banco Mercantil de S50 Paulo PAULO S.A. 027 31012 UNIBANCO Unibanco UniSo de Bancos Bra- sileiros S.A. 028 31030 BMG FINANCEIRA BMG Financeira 029 31012 REAL Banco Real S.A. 030 31012 ECONOMICO Banco Econ6mico S.A. 031 31012 COMERCIO E INDUSTRIA Banco Comércio e Indfistria de S.P. S30 Paulo S.A. 032 31012 FRANCES BRASILEIRO Banco Francés e Brasileiro S.A. 033 31011 NORDESTE BRASIL Banco Nordeste do Brasil S.A. 034 31020 ITAfi INVESTIMENTOS Banco Itafi de Investimentos S.A. 035 31012 BRASIL Banco Comercial Brasil 174 Tape Sector Code Classifi- Name in the Tape Full Name of the Company cation 036 90000 BRASMOTOR Brasmotor S.A. 037 13071 BUNDY TUBING Bundy Tubing S.A. Indfistria e Comércio 038 13060 BARDELLA Bardella S.A. Inddstrias Mecanicas 039 13110 BAUMER Baumer S.A. 040 13040 BERGAMO Bérgamo Companhia Industrial 041 13050 BARLEM Barlem S.A. Empreendimentos Industriais 042 13071 BRASIMET Brasimet Comércio e Industria S.A. 043 13010 BIC MONARK Bicicletas Monark S.A. 045 13091 BENZENEX Benzenex Abudos e Inseticidas 046 13020 BRASILIT Brasilit S.A. 047 13100 BRASILEIRA DE ROUPAS Companhia Brasileira de Roupas 048 13010 CACIQUE Companhia Cacique de Café Solfivel 049 11000 CONSTRUTORA ADOLPHO Construtora Adolpho Lindenberg LINDENBERG S.A. 050 13072 CIMETAL Cimetal Siderfirgica S.A. 051 13100 CREMER Cremer S.A. Produtos Texteis Cirfirgicos 052 13020 CIMENTO GAUCHO Companhia de Cimento Portland Gaficho 053 21000 CASA MASSON Casa Masson S.A. Comércio e Indfistria 054 13030 CONSUL Consul S.A. 175 Tape Sector Code Classifi- Name in the Tape Full Name of the Company cation 055 23000 COMERCIAL BORDA DO Companhia Comercial da Borda do CAMPO Campo 056 11000 CONSTRUTORA BETER Construtora Beter S.A. 057 21000 CASA ANGLO Casa Anglo Brasileira S.A. Modas e Confeccfies. Bazar 058 13010 CICA Companhia Industrial de Conser- vas Alimenticias Cica 059 11000 CONSURSAN Consursan Engenharia e Comércio S.A. 060 13060 CBV;INDUSTRIAS CBV--Indfistrias Mecfinicas S.A. MECANICAS 061 13050 COBRASMA Cobrasma S.A. 062 32020 CESP Cesp--Companhia Energética de 550 Paulo 063 13020 CIMENTO ITAU Companhia de Cimento Portland Itafi 064 13010 CAFE BRASILIA Café Solfivel Brasilia 065 21000 CASA JOSE SILVA Casa José Silva Confecqaes S.A. 066 13020 CIMENTO CAUE Cimento Caue S.A. 067 13071 CIMAF Companhia Industrial e Mercantil de Artefatos de Ferro 068 13010 CITROBRASIL Citrobrasil S.A. 069 11000 CONCRETEX Concretex S.A. 070 32020 CEMIG Centrais Elétricas de Minas Gerais S.A. 071 11000 MENDES JUNIOR Construtora Mendes JGnior S.A. 072 13020 CIDAMAR Cidamar S.A. IndGstria e Comércio 073 13010 CERVEJARIA POLAR Cervejaria Polar S.A. 176 Tape Sector code CIBSSIfi' Name 1" the Tape Full Name of the Company cation 074 13010 CONFRIO Companhia Nacional de Frigori- ficos Confrio 075 13091 COPAS Companhia Paulista de Fertili- zantes Copas 076 32030 DOCAS DE SANTOS Companhia Docas de Santos 077 13010 DREHER Dreher S.A. Vinhos e Champanhas 078 32030 DOCAS DE IMBITUBA Companhia Docas de Imbituba 079 11000 DIAMETRO Di3metro Empreendimentos S.A. EMPREENDIMENTOS 080 13050 DF VASCONCELOS D.F. Vasconcelos S.A. Otica e Mecanica de alta precisao 081 13040 DURATEX Duratex S.A. 082 22000 DISTRIBUIDORA Distribuidora de Produtos de IPIRANGA Petroleo Ipiranga S.A. 084 13030 ERICSSON Ericsson do Brasil Comércio e Industria S.A. 085 13110 ESTRELA Manufatura de Brinquedos Estrela S.A. 086 13071 ELUMA Eluma S.A. Indastria e Comércio 087 03010 EMILIO ROMANI Emflio Romani S.A. 088 13040 EUCATEX Eucatex S.A. IndGstria e Comércio 089 11000 ECEL Escrit6rio de Construcfio e Engenharia Ecel S.A. 090 11000 ECISA Ecisa Engenharia, Comércio e Indfistria S.A. 091 13020 ETERNIT Eternit S.A. 092 13080 EDITORA DE GUIAS Edit6ra de Guias LTB S.A. LTB 177 Tape Sector Code Classifi- Name in the Tape Full Name of the Company cation 094 11000 ENBASA Enbasa Engenharia Comercial S.A. 095 20000 EMBRAVA Empresa Brasileira de Varejo S.A. 096 13094 ELE KEIROZ Produtos Quimicos Ele Keiroz S.A. 098 31030 FINANCIADORA Financiadora Bradesco S.A. Crédito BRADESCO Financiamento e Investimento 099 13072 FERRO LIGAS Companhia Paulista de Ferro Ligas 100 13091 FERTISUL Fertisul S.A. 101 13010 FRIGOBRAS Frigobras(Companhia Brasileira de Frigor1ficos 102 13050 FORD BRASIL Ford Brasil S.A. 103 11000 FIQUET Fiquet S.A. 104 13071 FUNDICAO TUPY Fundicfio Tupy S.A. 105 13071 FERRO BRASILEIRO Companhia de Ferro Brasileiro 106 13091 FERTIPLAN Fertiplan S.A. Adubos e Inseti- cidas 107 13050 FNV Fabrica Nacional de VagBes S.A. 108 90000 FERRAGENS E Ferragens e Laminacio Brasil LAMINACAO BRASIL S.A. 109 13050 GEMMER BRASIL TRW Gemmer do Brasil S.A. 110 13100 GUARARAPES Confeccfies Guararapes S.A. 111 13010 GERMANI Germani Companhia Paranaense de Alimentos 112 13093 GOYANA Goyana S.A. Inddstria Brasileira de Materiais Plasticos 113 11000 HINDI Hindi Companhia Brasileira de Habitacoes 114 11000 HELENO FONSECA Heleno Fonseca Construtécnica S.A. 178 Tape Sector Code Classifi- Name in the Tape Full Name of the Company cation 115 13071 HERCULES Hercules S.A. Fabrica de Talheres 116 11000 H.C. CORDEIRO H. C. Cordeiro 117 13091 IAP Iap S.A. Indfistria de Fertili- zantes 118 13010 IGUACU CAFE Companhia Iguacu de Café Solfivel 119 13060 INDUSTRIAS VILLARES Indfistrias Villares S.A. 120 13100 INDUSTRIA HERING Indfistria Textil Companhia "Hering" 121 13060 INDUSTRIA ROMI Indfistria Romi S.A. 122 13071 ENGEMIX Engemix S.A. Estudos e Projetos de Engenharia 123 13030 IBESA Pereira Lopes Ibesa Indfistria e Comércio S.A. 124 13093 KERALUX Keralux S.A. Revestimentos Cerfimica 125 13010 KIBON Kibon S.A. Indfistrias Alimen- ticias 126 13093 KELSONS Kelson's Indfistria e Comércio S.A. 127 32020 LIGHT Light Servicos de Eletricidade S.A. 128 21000 LOJAS AMERICANAS Lojas Americanas S.A. 129 21000 LOBRAS Lojas Brasileiras S.A. 130 13010 LACTA Industria de Chocolates Lacta 131 13050 LONAFLEX Lonaflex S.A. Guarnicfies para freios 132 21000 LOJAS RENNER Lojas Renner S.A. 179 Tape Sector Code Classifi- Name in the Tape Full Name of the Company cation 133 32020 LUZ FORCA SANTA CRUZ Companhia Luz e Forca Santa Cruz 134 13080 LISA Livros Irradiantes S.A. 135 13050 MARCOPOLO Marcopolo S.A. Carrocerias e Onibus 136 13010 MOINHO LAPA Moinho da Lapa S.A. 137 13110 MOINHO SANTISTA S.A. Moinhos Santista Industrias Gerais 138 13071 METALURGICA GERDAU Metalurgica Gerdau S.A. 139 13040 MANASA Manasa Madeireira Nacional S.A. 140 13071 MADEF Madef S.A. Industria e Comércio 141 12000 MAGNESITA Magnesita S.A. 142 13091 MANAH Manah S.A. 143 13060 MAQUINAS PIRATININGA Méquinas Piratininga S.A. 144 13080 MELHORAMENTOS 5A0 Companhia Melhoramentos sao PAULO Paulo Indfistria de Papel 145 21000 MESBLA Mesbla S.A. 146 13050 METAL LEVE Metal Leve S.A. Indfistria e Comércio 147 13071 MANGELS INDUSTRIAL Mangels Industrial S.A. 148 13060 MECANICA PESADA Mecanica Pesada S.A. 149 13071 METALURGICA ABRAMO Metalfirgica Abramo Eberle S.A. EBERLE 150 13071 MICHELETTO Indfistrias Micheletto S.A. 151 13071 METALURGICA LA FONTE Metalfirgica La Fonte S.A. 152 13050 NAKATA Nakata S.A. Indfistria e Comércio 180 Tape Sector Code Classifi- Name in the Tape Full Name of the Company cation 153 13060 NORDON METALURGICA Nordon Indfistrias Metaldrgicas S.A. 154 13094 OXIGENIO 00 BRASIL Oxigénio do Brasil S.A. 155 13094 ORNIEX Orniex S.A. 156 22000 PETROLEO IPIRANGA Companhia Brasileira de Petréleo Ipiranga 157 13030 PIRELLI Pirelli S.A. Companhia Industrial Brasileira 158 13092 PETROBRAS Petr61e0 Brasileiro S.A. Petrobrés 159 22000 PARAGAS Companhia de 055 do Paré Paragés 160 11000 PARANAPANEMA Paranapanema S.A. Mineracio, Indfistria e Comércio 162 13010 PERDIGAO PerdigEo S.A. Comércio e IndGstria 163 13071 PREMESA Premesa S.A. Inddstri a eComércio 164 13093 PIRAMIDES BRASILIA Pir3mides Brasilia S.A. Inddstria e Comércio 165 32020 PAULISTA DE FORCA Companhia Paulista de Forca e Luz E LUZ 166 21000 PROSDOCIMO Prosd6cimo S.A. Importacfio e Comercio 167 13093 PLASTICOS MONSANTO Companhia Brasileira de Plasticos Monsanto 168 11000 PBK EMPREENDIMENTOS PBK Empreendimentos Imobiliérios IMOBILIARIOS S.A. 169 13094 PHEBO Perfumarias Phebo S.A. 170 13010 REALCAFE Realcafé Soleel do Brasil S.A. 181 Tape Sector Code Classifi- Name in the Tape Full Name of the Company cation 171 13030 REFRIPAR Refrigeracfio Parana S.A. 172 13030 SEMP Semp Toshiba S.A. 173 13010 SOUZA CRUZ Companhia Souza Cruz Indfistria e Comercio 174 11000 SERVIX Servix Engenharia S.A. 175 13072 MANNESMAN Mannesman S.A. 176 23000 SOPAVE Sopavg S.A. Sociedade Paulista de Ve1culos 177 13072 SIDERURGICA COFERRAZ Siderfirgica Coferraz S.A. 178 13030 SHARP Sharp S.A. Equipamentos Ele- tr6nicos 179 23000 SAVENA Savena S.A. de Veiculosfiacionais, Comercio e Representacao 180 13020 SANO Sano S.A. IndGstria e Comércio 182 13010 SCHLOSSER Companhia Industrial Schlosser S.A. 183 13072 SIDERURGICA NACIONAL Companhia Siderfirgica Nacional 184 90000 SAMCIL Samcil S.A. Servico de Assis- téncia Médica ao Comércio e Indfistria 185 13040 SUDESTE Sudeste S.A. Indfistria eComércio 186 13091 SOLORRICO Solorrico S.A. Indfistria e Comercio 187 13010 SADIA CONCORDIA Sadia’Concérdia S.A. Indfistria e Comercio 188 13072 SIDERURGICA Siderfirgica Aconorte S.A. ACONORTE 189 13072 SIDERURGICA GUAIRA Siderfirgica Guaira S.A. 182 Tape Sector Code Classifi- Name in the Tape Full Name of the Company cation 190 13010 SADIA Afo0LA Sadia AvIcola S.A. 191 13071 SIFCO DO BRASIL Sifco do Brasil S.A. Inddstrias Metalfirgicas 192 13072 SANTA OLIMPIA Usina Santa Olimpia Industria de Ferro e Aco S.A. 193 32010 TELESP Telecomunicacfies de 550 Paulo S.A.--Telesp 194 13100 TEKA Tecelagem Kuenrich S.A. 195 23000 TRANSPARANA Transparané S.A. 196 90000 TECNOSOLO Tecnosolo Engenharia e Tecno- logia de Solos e Materiais S.A. 197 32010 TELERJ TelecomunicacBes do Rio de Janeiro S.A.--Te1erj 198 21000 TECHNOS RELOGIOS Technos Re16gios S.A. 199 21000 ULTRALAR Ultralar S.A. Aparelhos e Servicos 200 90000 UNIPAR Unipar Uniio de Industrias Petroqu1micas S.A. 201 13110 VIGORELLI Vigorelli do Brasil S.A. Comércio e Industria 202 12000 VALE 00 R10 DECE Companhia Vale do Rio Doce 203 32030 VARIG Varig S.A. ViacSo Aérea Rio- Grandense 204 13110 VIDRARIA SANTA MARINA Companhia Vidraria Santa Marina 205 13100 VULCABRAS Vulcabrés S.A. Indfistria e Comércio 206 11000 VEPLAN Veplan--Resid€ncia, Empreendi- mentos e Construcaes S.A. 183 Ta e Sector Coge Classifi- Name in the Tape Full Name of the Company cation 207 13094 WHITE MARTINS S.A. White Martins 208 13060 ZANINI Zanini S.A. Equipamentos Pesados APPENDIX F INDEXED NATIONAL TREASURY BONDS (ORTN) AND GENERAL PRICE-LEVEL INDEX VALUES 184 APPENDIX F INDEXED NATIONAL TREASURY BONDS (ORTN) AND GENERAL PRICE-LEVEL INDEX VALUES . General Price-Level Date Period ORTN Value Index Value 1/1/70 1 42.35 211 1/2/70 2 43.30 214 1/3/70 3 44.17 218 1/4/70 4 44.67 219 1/5/70 5 45.08 222 1/6/70 6 45.50 227 1/7/70 7 46.20 231 1/8/70 8 46.61 236 1/9/70 9 47.05 241 1/10/70 10 47.61 245 1/11/70 11 48.51 246 1/12/70 12 49.54 248 1/1/71 13 50.51 252 1/2/71 14 51.44 256 1/3/71 15 52.12 262 1/4/71 16 52.64 266 1/5/71 17 53.25 271 1/6/71 18 54.01 277 1/7/71 19 55.08 281 1/8/71 20 56.18 284 1/9/71 21 57.36 288 1/10/71 22 58.61 291 1/11/71 23 59.79 294 1/12/71 24 60.85 297 1/1/72 25 61.52 302 1/2/72 26 62.26 308 1/3/72 27 63.09 312 1/4/72 28 63.81 316 1/5/72 29 64.66 319 1/6/72 30 65.75 322 1/7/72 31 66.93 326 1/8/72 32 67.89 331 1/9/72 33 68.46 335 1/10/72 34 68.95 338 1/11/72 35 69.61 341 1/12/72 36 70.07 343 1/1/73 37 70.87 349 1/2/73 38 71.57 353 1/3/73 39 72.32 358 186 General Price-Level Date Per1od ORTN Value Index Value 1/4/73 40 73.19 363 1/5/73 41 74.03 367 1/6/73 42 74.97 371 1/7/73 43 75.60 374 1/8/73 44 76.48 378 1/9/73 45 77.12 382 1/10/73 46 77.87 388 1/11/73 47 78.40 392 1/12/73 48 79.07 397 1/1/74 49 80.62 408 1/2/74 50 81.47 419 1/3/74 51 82.69 438 1/4/74 52 83.73 461 1/5/74 53 85.10 477 1/6/74 54 86.91 486 1/7/74 55 89.80 492 1/8/74 56 93.75 498 1/9/74 57 98.22 507 1/10/74 58 101.90 514 1/11/74 59 104.10 522 1/12/74 60 105.41 534 1/1/75 61 106.76 546 1/2/75 62 108.38 558 1/3/75 63 110.18 567 1/4/75 64 112.25 577 1/5/75 65 114.49 589 1/6/75 66 117.13 602 1/7/75 67 119.27 615 1/8/75 68 121.31 632 1/9/75 69 123.20 647 1/10/75 70 125.70 661 1/11/75 71 128.43 676 1/12/75 72 113.93 690 1/1/76 73 113.34 712 1/2/76 74 135.90 741 1/3/76 75 138.94 769 1/4/76 76 142.24 797 1/5/76 77 145.83 825 1/6/76 78 150.17 847 1/7/76 79 154.60 879 1/8/76 80 158.55 915 1/9/76 81 162.97 946 1/10/76 82 168.33 969 1/11/76 83 174.40 987 1/12/76 84 179.68 1010 1/1/77 85 183.65 1047 187 . General Price-Level Date Per1od ORTN Value Index Value 1/2/77 86 186.83 1081 1/3/77 87 190.51 1125 1/4/77 88 194.83 1171 1/5/77 89 200.45 1213 1/6/77 90 206.90 1237 1/7/77 91 213.80 1263 1/8/77 92 219.51 1279 1/9/77 93 224.01 1302 1/10/77 94 227.15 1337 1/11/77 95 230.30 1372 1/12/77 96 233.74 1401 1/1/78 97 238.32 1438 1/2/78 98 243.35 1487 1/3/78 99 248.99 1536 1/4/78 100 255.41 1588 1/5/78 101 262.87 1639 1/6/78 102 270.88 1698 1/7/78 103 279.04 1746 1/8/78 104 287.58 1793 1/9/78 105 295.57 1839 1/10/78 106 303.29 1891 1/11/78 107 310.49 1943 1/12/78 108 318.44 1973 1/1/79 109 326.82 2045 1/2/79 110 334.20 2122 1/3/79 111 ' 341.97 2245 1/4/79 112 350.51 2329 1/5/79 113 363.64 2384 1/6/79 114 377.54 2466 1/7/79 115 390.10 2574 1/8/79 116 400.71 2723 1/9/79 117 412.24 2933 1/10/79 118 428.80 3087 1/11/79 119 448.47 3258 1/12/79 120 468.71 3497 1/1/80 121 487.83 3714 APPENDIX G THE STOCK MARKET IN 1971 188 APPENDIX G THE STOCK MARKET IN 1971 The 1969 economic growth and the process of inflation control intensified the stock market reaction, which had already started in 1968. The BOVESPA Index (initiated in the early part of 1968 based at 100) exceeded 200 at the end of 1968, 400 in late June 1969, and oscillated between 500 and 600 for the period August-December 1969. Following a period of relative stability, an expansion process was initiated in July 1970 that maintained itself to the middle of 1971. The market growth attracted investors belonging to all levels, through direct stock-exchange operations or through mutual and fiscal funds. The mutual-funds patrimony showed a significant evolution: 243 percent in 1969, 57 percent in 1970, and 150 percent in 1971, reaching the amount of Cr$3.167 million at the end of 1971. Fiscal funds also represented a high volume of resources. Their assets by November 1969 reached the sum of Cr$260 million and by December 1970 Cr$411 million. At this date, there were 896,000 investors. The demand for stocks grew faster than the corresponding supply. The number of companies quoted in 1970 and early 1971 was still relatively small, provoking a fast price increase and stimu- lating speculation. Volume traded reached extremely high values. The two largest stock markets, those in Rio de Janeiro and $50 Paulo, negotiated, in 189 190 all, 1.2 billion dollars in 1970 and 4.9 billion in 1971; during May 1971 alone, the total reached 767 million dollars. This excess demand, which existed up to the middle of 1971, prompted a reversion, and in 1972 all stock-market indicators showed the market readjust- ment. SOURCE: Adroaldo Moura da Silva, Francisco Vidal Luna, and Helio Nogueira da Cruz, Inflacfio e Mercado de Capitais a Experi- éncia Brasileira (Inflation and the Capital Market: The Brazilian Experience). APPENDIX H PARTICIPATION OF LARGE COMPANIES IN THE TOTAL VOLUME TRADED 191 .1; APPENDIX H PARTICIPATION OF LARGE COMPANIES IN THE TOTAL VOLUME TRADED Participation (%) Company 1974 1975 1976 Petrobras 21.41 32.41 24.13 Banco do Brasil 12.18 21.06 22.15 Belgo Mineira 13.80 6.97 5.24 Vale do Rio Doce 7.46 7.29 4.22 TOTAL 54.85 67.73 55.74 SOURCE: Anuarios da Bolsa de Valores de 550 Paulo. 192 BIBLIOGRAPHY 193 BIBLIOGRAPHY Abrahamson, Allen A., and Emery, John T. "Risk and Price Distribu- tions." Journal of Financial and Quantitative Analysis 9 (November 1974): 847. Altman, Edward I.; Jacquillat, Bertrand; and Levasseur, Michel. "Comparative Analysis of Risk Measures: France and the United States.“ Journal of Finance 29 (May 1975): 1495-1511. Arditti, Fred 0. "Skewness and Investors' Decisions: A Reply." Journal of Financial and Quantitative Analysis 10 (March 1975): 73-76. Beckerman, Paul. "The Trouble With Index-Linking: Notes on the Recent Brazilian Experience." 1978. (Mimeographed.) Belkaqui, Ahmed. "Canadian Evidence of Heteroscedasticity in the Market Model." Journal of Finance 32 (September 1977): 1320-24. Biger, Nahum. "The Assessment of Inflation and Portfolio Selection." Journal of Finance 30 (May 1975): 451-67. "Portfolio Selection and Purchasing Power Risk-—Recent Canadian Experience." Journal of Financial and Quantitative Analysis 11 (June 1976): 251-67. . "Real Returns, Portfolio Decisions and the Capital Asset Pricing Model." Ph.D. dissertation, York University, 1974. Abstracted in "Abstracts of Doctoral Dissertations." Journal of Finance 30 (June 1975): 911. Black, F. "Capital Market E uilibrium With Restricted Borrowing." Journal of Business 45 ?July 1972): 444—54. Black, F.; Jensen, M. C.; and Scholes, M. "The Capital Asset Pricing Model: Some Empirical Results." In Studies in the Theory of Capital Markets. pp. 79-121. Edited by Michael Jensen. New Ybrk: TPraeger, l972. Blume, M. E. “On the Assessment of Risk." Journal of Finance 26 (March 1971): 1-10. . "Portfolio Theory: A Step Toward Its Practical Applications. Journal of Business 43 (April 1970): 152-73. 194 195 Branch, Ben. "Common Stock Performance and Inflation: An International Comparison." Journal of Business 47 (January 1974): 48-52. Brennan, M. J. “The Optimal Number of Securities in a Risky Asset Portfolio When There Are Fixed Costs of Transacting: Theory and Some Empirical Results." Journal of Financial and Quantitative Analysis 10 (September 1975): 483-96. Cagan, Phillip. “Stock Values and Inflation." National Bureau of Economic Research, September 1973, p. 102. Chen, A. H., and Boness, A. J. "Effects of Uncertain Inflation on the Investment and Financing Decisions of a Firm." Journal of Finance 30 (May 1975): 469-83. Chen, Elaine. "Capital Asset Prices Under Uncertain Inflation." University of California, Berkeley, 1974. (Typewritten.) Choate, 0. March, and Archer, Stephen H. "Irving Fisher, Inflation, and the Nominal Rate of Interest." Journal of Financial and Quantitative Analysis 10 (November 1975): 675-85. Comissao Nacional de Bolsas de Valores. "Introducfio ao Mercado de Acfies." 1979. Cornell, W. Bradford. "Which Inflation Rate Affects Interest Rates?" Business Economics 12 (May 1977): 22-25. CRSP Master File. Chicago: Center for Research in Security Prices, Graduate School of Business, University of Chicago, March 1978. Douglas, G. W. "Risk in the Equity Markets: An Empirical Appraisal of Market Efficiency." Yale Economic Essays 9 (Spring 1969): 3-45. Durand, D. "The Cost of Capital, Corporation Finance and the Theory of Investment: Comment." American Economic Review 49 (September 1959): 639-55. . "Costs of Debt and Equity Funds for Business: Trends and Problems of Measurement." Reprinted in The Management of Cor- porate Capital, pp. 91-116. Edited by Ezra Solomon. New York: Free Press, 1959. Fama, E. F. "The Behavior of Stock Market Prices." Journal of Business 37 (January 1965): 34-105. . "Efficient Capital Markets: A Review of Theory and Empiri- cal Work." Journal of Finance 25 (May 1970): 383-417. 196 . "Risk Return and Equilibrium: Some Clarifying Comments." Journal of Finance 23 (March 1968): 29-40. . "Short Term Interest Rates as Predictors of Inflation." American Economic Review 65 (June 1975): 269-82. Fama, E. F., and MacBeth, J. 0. "Risk, Return and Equilibrium: Empirical Tests." Journal of Political Economy_81 (May-June 1973): 607-36. Fama, E. F., and Miller, M. H. The Theory of Finance. New York: Holt, Rinehart and Winston, Inc., 1972. Findlay, Ronald, and Wellisz, Stanislaw. "Project Evaluation, Shadow Prices and Trade Policy." Journal of Political Economy 84 (1967): 543—52. Fisher, Irving. Appreciation and Interest. New York: Macmillan, 1896; reprint ed., New York: Augustus M. Delley, l965. The Theory of Interest. New York: Macmillan, 1930. Fisher, Lawrence, and Lorie, James H. A Half Century of Returns on Stocks and Bonds. Chicago: University of Chicago, Graduate School of Business, 1977. . "Rates of Return on Investments in Common Stocks: The Year by Year Records, 1926-65." Journal of Business 41 (July 1968): 291-316. Fogler, H. Russel, and Radcliffe, Robert C. "A Note on Measurement of Skewness." Journal of Financial and Quantitative Analysis 9 (June 1974): 485-87. Foster, G. "Asset Pricing Models: Further Tests." Journal of Finan- cial and Quantitative Analysis 13 (March 1978): 39-53. Francis, J. C. Investment Analysis and Management. 2nd ed. New York: McGraw-Hill, Inc., 1976. . "Skewness and Investors' Decisions." Journal of Financial andeuantitative Analysis 10 (March 1975): 163-72. Francis, J. C., and Archer, S. H. Portfolio Analysis. Englewood Cliffs, N.J.: Prentice-Hall, 1971. Friedman, M., and Savage, L. J. "The Utility Analysis of Choices Involving Risk." Journal of Political Economy 4 (August 1948): 279-304. 197 Friend, F., and Blume, M. "Measurement of Portfolio Performance Under ggceyga1nty." American Economic Review 60 (September 1969): Friend, Irwin, Lands-Kroner, Yoram; and Losg, Etienne. "The Demand for Risky Assets Under Uncertain Inflation." Journal of Finance 31 (December 1976): 1287-97. Friend, Irwin; Westerfield, Randolph; and Granito, Michael. "New Evidence on the Capital Asset Pricing Model." Journal of Finance 33 (June 1978): 903-20. Gibson, W. E. "Interest Rates and Inflationary Expectations: New Evidence." American Economic Review 62 (December 1972): 854-65. Gonzales-Gaviria, Nestor. "Inflation and Capital Asset Prices: Theory and Tests." Ph.D. dissertation, Stanford University, 1973. Gordon, Myron J., and Halpern, Paul J. "Bond Share Yield Spreads Under Uncertain Inflation." American Economic Review 66 (Septem- ber 1976): 559-65. Hagerman, Robert L., and Kim, E. Han. "Capital Asset Pricing With Price Level Changes." Journal of Financial and Quantitative Analysis 11 (September 1976): 381-91. Hamada, R. S. "The Effect of the Firm's Capital Structure on the Systematic Risk of Common Stocks." Journal of Finance 27 (May 1972): 435-52. Haugen, R. A., and Pappas, J. L. "Equilibrium in the Pricing of Capital Assets, Risk-Bearing Debt Instruments and the Question of Optimal Capital Structure." Journal of Financial and Quan- titative Analysis 6 (June 1971): 943-54. Hendershott, Patric H., and Van Horne, James C. "Expected Inflation Implied by Capital Market Rates." Journal of Finance 28 (May 1973): 301-14. Hess, Alan C. "The Riskless Rate of Interest and the Market Price of Risk." Quarterly Journal of Economics 89 (August 1975): 444-55. "The Riskless Rate of Interest and the Market Price of Risk: A Correction." Quarterly Journal of Economics 92 (November 1978): 689-91. Hess, Patrick J., and Bicksler, James L. "Capital Asset Prices Versus Time Series Models as Predictors of Inflation." Journal of Finan- cial Economics 2 (December 1975): 341-60. 198 Hirshleifer, J. "Efficient Allocation of Capital in an Uncertain World." American Economic Review 54 (May 1964): 77-85. _________ "Investment Decision Under Uncertainty: Choice-Theoretic Approaches." Quarterlleournal of Economics 79 (November 1965): 509-36. Hogan, William W., and Waren, James M. "Toward the Development of an Equilibrium Capital-Market Model Based on Semivariance." Journal of Financial and Quantitative Analysis 9 (January 1974): 13-23. Horne, J. C. V. Financial Management and Policy. 4th ed. Englewood Cliffs, N.J.: Prentice-Hall, 1977. . Function and Analysis of Capital Market Rates. Englewood Cliffs, N.J.: Prentice-Hall, 1970. Ibbotson, Roger G., and Sinquefield, Rex A. "Stocks, Bonds, Bills, and Inflation: Year by Year Historical Returns 1926-1974." Journal of Business 49 (January 1978): 11-43. Jaffe, Jeffrey F., and Mandelker, Gershon. "The 'Fisher Effect' for Risky Assets: An Empirical Investigation." Journal of Finance 31 (May 1976): 447-58. Jahankhani, Ali. "E-V and E-S Capital Asset Pricing Models: Some Empirical Tests." Journal of Financial and Quantitative Analysis 11 (November 1976): 513-28. Jensen, M. C. "Capital Markets: Theory and Evidence." Bell Journal of Economics and Management Science 3 (October 1972): 357-98. Joehnk, Michael D., and Nielsen, James F. "The Effects of Conglomerate Merger Activity on Systematic Risk." Journal of Financial and Quantitative Analysis 9 (March 1974): 215-25. Keran, Michael W. "Expectations, Money and the Stock Market." Review: Rederal Reserve Bank of St. Louis. January 1971. Kraus, Alan, and Litzenberger, Robert H. "Skewness Preference and the Valuation of Risk Assets." Journal of Finance 31 (September 1976): 1085-99. Latané, H. A.; Tuttle, D. L.; and Young, W. E. "How to Choose a Market Index." Financial Analyst Journal, September-October 1971, pp. 75-85. Lau, C. Sheila; Quay, Stuart R.; and Ramsey, Carl M. "The Tokyo Stock Exchange and the Capital Asset Pricing Model." Journal of Finance 29 (May 1974): 507-14. 199 Lee, Cheng F. "Functional Form, Skewness Effect, and the Risk-Return Relationship." Journal of Financial andyguantitative Analysis 12 (March 1977): 55-72. "On the Relationship Between the Systematic Risk and the Investment Horizon." Journal of Financial and Quantitative Analysis 11 (December 1976): 803-15. Lessard, Donald R. "International Portfolio Diversification: A Multivariate Analysis for a Group of Latin American Countries." Journal of Finance 28 (June 1973): 619-33. Levy, H., and Sarnat, M. "Diversification, Portfolio Analysis and Uneasy Care for Conglomerate Mergers." Journal of Finance 25 (September 1970): 795-802. Lintner, .1. "Inflation and Common Stock Prices in a Cyclical Context." National Bureau of Economic Research, 53rd Annual Report, 1973. "Inflation and Security Returns." Journal of Finance 30 " (May 1975): 259-80. _________ "The Market Price of Risk, Size of the Market and Investor's Risk Aversion." Review of Economics and Statistics 52 (February 1970): 87-99. "Security Prices and Risk: The Theory and a Comparative Analysis of A.T.T. and Leading Industrials." Paper presented at the Conference on the Economics of Regulated Public Utili- ties, Chicago, 24 June 1965. _________ "Security Prices, Risk and Maximal Gains From Diversifi- cation." Journal of Finance 20 (December 1965): 587-615. Lloyd, William P. "International Portfolio Diversification of Real Assets: An Inquiry." Journal of Business Research 3 (April 1975): 113-20. Luna, Francisco Vidal. "Mercado de Capital e Economia Brasileira" [Capita1~Market and the Brazilian Economy]. Paper presented at the Sao Paulo Stock Exchange, 1979. . "E1 Papel de la Bolsa en la Capitalization de la Empresa Privada" [The Effect of the Stock Exchange on Capitalization of Private Enterprise: The Brazilian Case]. Paper presented by the $50 Paulo Stock Exchange at the Seventh General Assembly of the Ibero-American Federation of Stock Exchanges, Santiago, Chile, 1980. 200 Luna, Francisco Vidal, and Nogueira Neto, Thomaz de Aquino. "Correcfio Monetfiria e Mercado de Capitais" [Indexation and Capital Market]. Paper presented by the $50 Paulo Stock Exchange to the Fifth General Assembly of the Ibero-American Federation of Stock Exchanges, Buenos Aires, Argentina, October 1978. Luna, Francisco Vicao,; Silva, Adroaldo Moura; and Cruz, Helio Nogueira. “Inflacio e Mercado de Capitais: A Experiéncia Brasileira" [Inflation and the Capital Market: The Brazilian Experience]. Paper presented at the Fourth General Assembly of the Ibero-American Federation of Stock Exchanges, Barcelona, Spain, 1977. Manaster, Steven. "Real and Nominal Efficient Sets." Journal of Finance 34 (March 1979): 93-102. Mandelker, Gershon. "Risk and Return: The Case of Merging Firms." Journal of Financial Economics 1 (December 1974): 303-35. Markowitz. H. "Portfolio Selection." Journal of Finance 7 (March 1952): 77-91. Martin, John 0., and Klemkosky, Robert C. ”Evidence of Heterosce- dasticity in the Market Model." Journal of Business 48 (January 1975): 81-86. Miller, M., and Modigliani, F. "Dividend Policy, Growth and the Valuation of Shares." Journal of Business 34 (October 1961): 411-33. Miller, M., and Scholes, M. "Rates of Return in Relation to Risk: A Re-Examination of Some Recent Findings." In Studies in the Theory of Capital Markets, pp. 47-77. Edited by Michael Jensen. New York: Praeger, 1972. Modigliani, F., and Miller, M. "Corporate Income Taxes and the Cost of Capital: A Correction." American Economic Review 53 (June 1963): 433-43. . "The Cost of Capital, Corporation Finance and the Theory of Investment." American Economic Review 48 (June 1958): 261-97. _________ "The Cost of Capital, Corporation Finance, and the Theory of Investment: Reply." American Economic Review 49 (September 1959): 655-69. Morgan, 1. G. "Prediction of Return With the Minimum Variance Zero-Beta Portfolio." Journal of Financial Economics 2 (December 1975): 361-76. 201 Mossin, J. "Equilibrium in a Capital Asset Market." Econometrica 34 (May 1976): 768-83. Nelson, Charles. "Inflation and Rates of Return on Common Stocks." Journal of Finance 31 (May 1976): 471-87. Officer, R. R. “Seasonality in Australian Capital Markets." Journal of Financial Economics 2 (March 1975): 29-51. Panton, Don 8.; Lessig, V. Parker; and Joy, 0. Maurice. "Comovement of International Equity Markets: A Taxonomic Approach." Journal of Financial and Quantitative Analysis 11 (September 1976 : 415-32. Pettit, Richardson R., and Westerfield, Randolph. "Using the Capital Asset Pricing Model and the Market Model to Predict Security Returns." Journal of Financial and Quantitative Analysis 9 (September 1974): 579-605. Pogue, Gerald A., and Solnik, Bruno H. "The Market Model Applied to European Common Stocks: Some Empirical Results." Journal of Financial and Quantitative Analysis 9 (December 1974): 917-44. Poole, William. "Indexing and Capital Markets." American Economic Review 66 (May 1976): ZOO-204. Pradhan, Kamalakar Vinayak. "Capital Asset Pricing Models Under Uncertain Inflation: Theory, Tests, and Applications." Ph.D. dissertation, University of Toronto, 1978. Puggina, W. A. “Analysis of Rates of Return and Risk for Common and Preferred Stocks--The Brazilian Experience." Ph.D. dissertation, Michigan State University, 1974. Reilly, Frank K.; Smith, Ralph E.; and Johnson, Glenn L. "A Correc- tion and Update Regarding Individual Common Stocks as Inflation Hedges." Journal of Financial and Quantitative Analysis 10 (December 1975): 871-80. Robichek, Alexandre A., and Cohn, Richard A. "The Economic Determi- nants of Systematic Risk." Journal of Finance 29 (May 1979): 439-47. Robinson, R. 1., and Wrightsman, 0. Financial Markets. New York: McGraw-Hill, Inc., 1974. R011, R. "Assets, Money and Commodity Price Inflation Under Uncer- tainty." Journal of Money, Credit and Bankigg, November 1973, pp. 903-23. 202 . "A Critique of the Asset Pricing Theory's Tests." Journal of Financial Economics 3 (May 1977): 129-76. Ross, Stephen A. "The Current Status of the Capital Asset Pricing Model (CAPM)." Journal of Finance 33 (June 1978): 885-901. Rozeff, Michael S. "The Association Between Firm Risk and Wealth Transfers Due to Inflation." Journal of Financial and Quanti- tative Analysis 12 (June 1977): 151-63. Rubinstein, M. E. "The Fundamental Theorem of Parameter-Preference Security Valuation." Journal of Financial and Quantitative Analysis 8 (January 1973): 61-69. . "A Mean-Variance Synthesis of Corporate Financial Theory." Journal of Finance 28 (March 1973): 167-81. Scholes, Myron, and Williams, Joseph. "Estimating Betas From Non- synchronous Data." Journal of Financial Economics 5 (December 1977): 302-27. Sharma, J. L., and Kennedy, Robert E., "A Comparative Analysis of Stock Price Behavior on the Bombay, London, and New York Stock Exchanges." Journal of Financial and Quantitative Analysis 12 (September 1977): 391-413. Sharpe, W. F. "Capital Asset Prices: A Theory of Market Equilibrium Under Conditions of Risk.“ Journal of Finance 19 (September 1964): 425-42. Siegel, Jeremy J., and Warner, Jerold B. "Indexation, the Risk-Free Asset and Capital Market Equilibrium." Journal of Finance 32 (September 1977): 1101-7. Silber, William L. "Thinness in Capital Markets: The Case of the Tel Aviv Stock Exchange." Journal of Financial andyQuantitative Analysis 10 (March 1975): 129-42. Soldofsky, Roberto M. "Securities as a Hedge Against Inflation 1910-1969." Journal of Business Research 3 (April 1975): 165-73. Solnik, Bruno H. "Inflation and Optimal Portfolio Choices." Journal of Financial and Quantitative Analysis 13 (December 1978): 913-25. "An International Market Model of Security Price Behavior." Journal of Financial and Quentitative Analysis 9 (September 1974): 537-54. 203 . "Testing International Asset Pricing: Some Pessimistic Views." Journal of Finance 32 (May 1977): 503-12. Tobim, J. "Liquidity Preference as Behavior Towards Risks." Review of Economic Studies 25 (February 1958): 65-85. Williams, John Burr. Theory of Investment Value. Cambridge: Harvard University Press, 1938.