Stocks, Brady Bonds, and Currencies: Investment Opportunities in Latin America Paul D. Kaplan is vice president and chief economist at Ibbotson Associates in Chicago, which he joined in 1988. He was previously on the economics faculty of Northwestern University, teaching international finance and statistics. Mr. Kaplan holds a B.A. in mathematics, economics, and computer science from New York University and an M.A. and Ph.D. in economics from Northwestern University. The advantages of investing globally are well known. In general, it is always better to have more assets to invest in than fewer. The principles set forth by Harry Markowitz some 44 years ago are as apt today as when they were first introduced. In today’ s global economy, investors have the largest set of investment vehicles in history. Cross-border investing has become a generally accepted practice among U.S. investors, both institutional and individual.1 While most of the focus has been in the markets of other developed countries, the opportunities to invest in emerging markets have greatly increased in the last few years. Why Emerging Markets? There are at least three reasons why investors should consider emerging markets. Emerging markets investing can be considered a means of diversification, an opportunity to exploit inefficiencies in the global capital markets, and a unique historical investment opportunity. Let us consider each of these in more detail. A Means of Diversification Stocks and bonds issues in emerging countries represent part of the world portfolio. To achieve the highest possible return for the level of risk taken, even in a perfectly integrated efficient world, all investors should consider all risky assets in forming their portfolios.2 The logic of the CAPM leads to the conclusion that every investor should hold every asset in proportion to its market capitalization. Although emerging markets represent a small portion of the total world market, every investor should have at least a small portion of their risky portfolio in these markets to achieve complete diversification.

1

From 1988 to 1993, U.S. tax-exempt assets invested overseas grew from $62 billion to $262 billion From 1984 to 1993, international mutual fund investments grew from $1.2 billion to $71 billion. See Meyer [1996]. While cross-border investing has become a recent practice in the U.S., it has a long history in continental Europe. See Ibbotson, Siegel, and Kaplan [1991] for a discussion. 2 This conclusion is not specific to a mean-variance model. See Harlow [1991] and Gruaer and Hakansson [1987] for examples of global asset allocation models that are not based on the mean-variance model.

An Opportunity to Exploit Market Inefficiencies The theory of efficient markets presumes a set a circumstances that do not yet exist in the world markets. The capital markets of developed countries are not yet completely integrated, let alone those of the entire world. In a world in which markets are segmented, correlations between markets are low. These low correlations present diversification opportunities to investors who can invest across borders. As we shall see later, correlations between U.S. and Latin American markets have been low. As emerging nations develop, their capital markets should become more integrated with the rest of the world.3 The low correlations available to U.S. investors now may not be available in the future. A Unique Historical Opportunity Countries which are beginning to emerge as modern capitalist societies may be at unique historical crossroads, much like the United States at the end of the nineteenth century or Japan after World War II. Any investor who participated in those markets at those times would have been richly rewarded. Latin America countries may be in a similar situation today. These countries are rich in resources and have great potential for economic development. As these nations move towards capitalistic forms of economic organization, new opportunities for investors are emerging. Unfortunately, just because a country is poised for development does not guarantee that it will succeed or that foreign investors will be able to keep their claims. To an investor in 1900, Russia, Germany, and Japan would have looked like good opportunities. New Opportunities to Invest in Latin America Until recently, the opportunities for foreigners to invest in Latin American capital markets were limited. Several developments have greatly increased these opportunities. These include the growth in the publicly traded capital in these countries, the liberalization of laws on foreign investors, the creation of Brady bonds, and the introduction of derivative contracts on equity indices and Brady bonds. The first two developments increase the opportunity for direct investment in Latin American stocks. The creation of Brady bonds makes it possible for foreign investors to take on sovereign risk and receive compensation for it. Because Brady bonds are denominated in U.S. dollars, there is no local currency risk. Finally the introduction of derivatives makes it possible to participate in Latin American markets without direct investments in the underlying securities. In the remainder of this paper we present data on Latin American equity indices, Brady bonds, and currencies. In presenting this data, we take the perspective of an unhedged U.S. investor. Our purpose is to begin to quantify investment opportunities in Latin America and stimulate further study.

3

Bekaert and Harvey [1995] model the time-variation in the level of integration between emerging and developed markets. Their results suggest that world capital markets are less integrated than is commonly perceived. Page 2

Equity Markets To measure the past performance of Latin American equity markets, we used three types of indices: Morgan Stanley Capital International (MSCI) total return indices, International Finance Corporation Investable (IFCI) total return indices and, where available, local market capital appreciation indices.4 Exhibit 1 shows U.S. dollar annual returns for these indices for Argentina, Brazil, Chile, Colombia, Mexico, and Venezuela to the extent that data are available.5 Details of how these indices are constructed are in the Appendix. In comparison to data on developed markets, the time series presented in Exhibit 1 are quite short (the longest series dates back only to 1988). However, given the changing nature of these markets, the period for which data are available may closely correspond to the period for which the data are relevant. It is simply too early to draw conclusions about the long-run performance of these markets. Exhibit 2 shows the growth of one U.S. dollar invested in the four markets for which the IFCI data are available, beginning in 1989, along with the S&P 500 for comparison. Over this sevenyear period, a U.S. investor would have greatly been rewarded for taking on the risk of these Latin American markets. Exhibit 3 shows the risk/return relationship for all of the indices by plotting compound rates of return against annualized monthly standard deviations.6 This exhibit confirms that the high returns of the Latin American equity markets came at high levels of risk. The value of investing in Latin American markets to a U.S. investor is not just in the potential for higher returns, but also in the opportunity to diversify a portfolio. Exhibit 4 shows monthly correlation coefficients for the Latin American markets with the S&P 500 using the IFCI data.7 With the highest correlation being less than 0.3, the diversification opportunities appear to be substantial. Brady Bonds The introduction of Brady bonds in 1990 created new opportunities in global fixed income investing. By investing in Latin American Brady bonds, a U.S. investor takes on the risk of Latin American sovereigns without taking on currency risk. As with other risky securities, investor willingness to take on risk is due to the potential rewards in the form of higher expected return and diversification opportunities. 4

Dividends are not accounted for in the capital appreciation indices. Capital appreciation indices were used because total return indices were not available. 5 Derivative users might be especially interested in the results for the local market index for Mexico, the IPC, since options and futures on the index are going to be launched in the near future. 6 Monthly standard deviations were annualized by treating an annual return as a compound return of twelve independent monthly returns. See Ibbotson Associates [1996], p. 106 for details. When monthly returns are highly volatile, annual returns are very unevenly dispersed above and below their average. The usual rule of thumb that two-thirds of returns fall within one standard deviation of the average cannot be applied. For example, the 264 percent annual standard deviation for IFCI Argentina does not mean that returns are likely to fall below 100 percent (the complete elimination of value). Rather it is a reflection of the fact that monthly returns varied between -67 and 148 percent over the period. 7 The results are similar using the MSCI data. Page 3

The introduction of Brady bonds was the direct outcome of the implementation of the Brady Plan to reduce the sovereign debt in emerging countries. During the 1980s, debt negotiations focused on new private lending and debt restructuring. Seeing that this approach was not solving the debt crisis, the Bush administration, through Secretary of the Treasury Nicholas Brady, put forth a new market-oriented plan for debt reduction. Under the Brady plan, debt was reduced in recognition of the fact that the market value of the commercial bank loans was less than face value. The loans were restructured as publicly traded U.S. dollar-denominated bonds. In many cases, the principal was collateralized with U.S. Treasury securities. We use J.P. Morgan Emerging Market Bonds (EMBI) indices to represent Brady bonds for Argentina, Brazil, Mexico, and Venezuela. These capitalization-weighted total return indices represent the entire Brady float of the issuing countries in addition to other dollar-denominated sovereign restructured bonds. The time series for Brazil, Mexico, and Venezuela begins in December 1990. The time series for Argentina begins in April 1993. For indices that predate the Brady agreements, the index is comprised of non-Brady dollar-denominated sovereign restructured bonds. For a complete methodology of these indices, refer to the Appendix. Exhibit 5 shows U.S. dollar annual returns for the Brady bond indices for Argentina, Brazil, Mexico, and Venezuela. Exhibit 6 shows the growth of one U.S. dollar invested in the four Brady bond indices, along with the Lehman Brothers Government/Corporate Bond index and the First Boston High-Yield bond index for comparison. While returns on Brady bonds were high compared to the broad U.S. bond market, similar returns could have been achieved with corporate high-yield bonds. Exhibit 7 shows the risk/return relationship for all of the indices by plotting compound rates of return against annualized monthly standard deviations. This graph shows that the high returns on Brady bonds came at considerable risk. The risk on Brady bonds, at least during the short period depicted, was far greater than the risk of corporate high-yield bonds that produced similar returns. Exhibit 8 shows monthly correlation coefficients of the Brady bond indices with the Lehman Brothers Government/Corporate index, the CS First Boston High-Yield Bond index, and the S&P 500 index. It also shows the correlations of the high-yield bond index with the government/corporate index and the S&P 500. All of the correlations fall within the of range of 0.29 to 0.51. Hence, Brady bonds as well as high-yield corporate bonds can play a role in a diversified portfolio. The correlations with corporate high-yield bonds are uniformly higher than the correlations with the broad U.S. market.

Page 4

Currencies Anyone investing in foreign markets faces the problem of currency risk. High returns realized in local currency terms can be eaten away by a depreciating currency. Exhibit 9 shows year-by-year rates of appreciation against the U.S. dollar for the currencies of Argentina, Brazil, Chile, Colombia, Mexico, and Venezuela.8 All of these currencies have depreciated sharply against the U.S. dollar. (Strings of zeros are due to periods of fixed exchange rates). Exhibit 10 depicts the depreciation of Latin American currencies against the U.S. dollar. It shows what has happened to the value of one dollar’ s worth of each Latin American currency since 1986. In spite of these depreciating currencies, as Exhibit 2 shows, U.S. investors would have done very well by investing in Latin American equity markets. In these markets, equities have proven to be a very effective currency hedge. Conclusion The world’ s capital markets in general are becoming more open to all investors. In particular, opportunities for investing in Latin American markets have greatly increased. This is not only due to the growth and liberalization of these markets, but also to the introduction of new investment vehicles. Brady bonds allow U.S. investors to take on sovereign risk without currency risk. New futures and options contracts on the Chicago Mercantile Exchange and other worldwide exchanges allow investor to participate in Latin American equity and Brady bond markets without going through the costly process of buying and selling the underlying securities. To the extent that they are available, we have collected and presented data on Latin American markets. While the time period studied is too short to draw any long-term conclusions, we make the following observations: • Latin American equity markets have had very high returns, but with very high risk. • The high risk of Latin American equities can be partially mitigated in a well-diversified portfolio that includes U.S. equities. • The high returns that would have been realized by U.S. investors are in spite of the high rates of currency depreciation. • Brady bonds have had returns similar to that of high-yield U.S. corporate bonds, but with more risk. However, because the correlations with U.S. bonds and stocks have been 0.51 or less, they may play a role in diversified portfolios.

8

See the Appendix for a description of the data source. Page 5

Clearly more research is needed on the risks and correlations of Latin American markets. For example, Erb, Harvey, Viskanta [1995] have found that risks and correlations of country equity markets can differ in up and down markets. As the investment process becomes more globalized, findings like these need to be explored in more and more detail. Finally we note that collecting and presenting data are always valuable. In the context of studying the data on developed markets, Siegel and Kaplan [1990] point out that: Data are always valuable, and time series of capital market data become statistically more valuable as the period covered grows. There is no precise interval called “long enough,” but as a data study ages it eventually does become long enough. One might as well start collecting data on any market now. This certainly applies even more to emerging markets where large amounts of new money are flowing with little historical data as a guide.

Page 6

Exhibit 1:

1988 1989 1990 1991 1992 1993 1994 1995

Annual Returns on Equity Indices in % MSCI

Argentina IFCI

83.85 108.46 -28.25 370.41 -38.73 58.60 -24.12 12.75

149.86 -37.21 403.39 -25.74 77.65 -24.13 13.10

Colombia MSCI 1988 1989 1990 1991 1992 1993 1994 1995

29.54 3.11 -25.50

Merval*

MSCI

Brazil IFCI

13.42

129.85 81.08 -68.38 161.80 9.56 72.97 63.43 -19.30

65.38 -76.18 303.68 0.49 88.69 63.94 -18.37

IFCI

MSCI

Mexico IFCI

IPC*

49.86 40.38 6.78 -25.00

60.56 92.79 50.36 123.45 31.74 46.86 -42.72 -22.88

129.80 27.20 113.88 16.72 51.91 -39.30 -24.34

89.81 71.23 36.63 116.03 22.54 48.45 -42.63 -24.92

Bovespa*

105.58 56.90 -13.65

Venezuela MSCI

-10.13 -35.21 -25.69

Chile MSCI 44.68 54.52 49.54 105.34 22.30 37.24 40.63 -3.24

IFCI

56.94 -50.68 17.54 -15.62 -29.55

* Returns on local market indices do not include dividends. Note:

All returns are stated in US Dollars

Source: Morgan Stanley Capital International, International Finance Corporation, and the Chicago Mercantile Exchange

Page 7

IFCI 49.61 51.06 110.39 18.42 37.93 41.20 -1.32

Exhibit 2:

Growth of $1 Invested in IFCI Indices and S&P 500 (January 1989 to December 1995)

$16

IFCI Chile IFCI Argentina IFCI Mexico

Wealth in US Dollars

$12

IFCI Brazil S&P 500

$10.82 (40.53%) $8.94 (36.74%)

$8

$5.09 (26.17%) $4

$4.04 (22.06%) $2.74 (15.49%)

$0 1988

Note:

1989

1990

1991

1992

1993

1994

1995

Figures in parentheses indicate compound annual rates of return. All returns are in stated in US Dollars.

Source: International Finance Corporation and Standard & Poor’ s

Page 8

Exhibit 3:

Risk vs. Return of Equity Indices (January 1989 to December 1995*)

60%

Argentina IFCI

Brazil

MSCI IFCI

40%

Bovespa

Compound Annual Return

IFCI MSCI

IFCI 20%

IPC

Chile

MSCI IFCI

Colombia

IFCI

Mexico MSCI

S&P 500

Venezuela

MSCI

USA

0% 0%

-20%

50% Merval

100%

150%

200%

250%

300%

MSCI

-40%

Standard Deviation

*

with the exception of MSCI Colombia and MSCI Venezuela (Jan 1993-Dec 1995); IFCI Colombia (Mar 1991-Dec 1995); IFCI Venezuela (Feb 1990-Dec 1995); Bovespa (Nov 1992-Dec 1995); and Merval (Apr 1994-Dec 1995)

Note:

All returns are stated in US Dollars

Source: Morgan Stanley Capital International, International Finance Corporation, Standard & Poor’ s, and the Chicago Mercantile Exchange

Page 9

Exhibit 4:

Correlation of IFCI Indices with S&P 500 (January 1989 to December 1995*)

0.40

0.30

0.29 0.23

0.22

0.20

0.10

0.07

0.05

0.00

Mexico

Chile

Brazil

Colombia

Argentina

Venezuela

-0.10

-0.15 -0.20

*

with the exception of Colombia (Mar 1991-Dec 1995) and Venezuela (Feb 1990-Dec 1995)

Source: International Finance Corporation and Standard & Poor’ s

Page 10

Exhibit 5:

Annual Returns of Brady Bonds in % Argentina

1991 1992 1993 1994 1995 Note:

-27.24 34.30

Brazil 36.26 -1.56 60.64 4.83 20.49

Mexico Venezuela 38.95 39.78 13.61 -9.65 31.00 42.58 -23.14 -30.10 25.86 41.05

All returns are stated in US Dollars

Source: JP Morgan

Page 11

Exhibit 6:

Growth of $1 Invested in Brady Bonds and US Bonds (December 1990 to December 1995)

$3.00 Brazil US High Yld Corp Bonds

$2.72 (22.17%)

Mexico $2.50

Venezuela LB Gvt/Corp

Wealth in US Dollars

$2.32 (18.31%)

$2.00 (14.88%)

$2.00

$1.78 (12.17%) $1.60 (9.80%) $1.50

$1.00 1990

Note:

1991

1992

1993

1994

1995

Figures in parentheses indicate compound annual rates of return. All returns are stated in US Dollars.

Source: JP Morgan, CS First Boston, and Lehman Brothers

Page 12

Exhibit 7:

Risk vs. Return of Brady Bonds and US Bonds (January 1991 to December 1995*)

24%

Brazil US High Yield Bonds

Compound Annual Return

20%

Mexico

16%

12%

Argentina

Lehman Brothers Govt/Corp

Venezuela

8%

4%

0% 0%

5%

10%

15%

20%

Standard Deviation

*

with the exception of Argentina (May 1993-Dec 1995)

Note:

All returns are stated in US Dollars

Source: JP Morgan, CS First Boston, and Lehman Brothers

Page 13

25%

30%

Exhibit 8:

Correlations of Brady Bonds and US High-Yield Bonds with US Bonds and Stocks (January 1991 to December 1995*)

0.60

0.49

0.50

LB Govt/Corp

0.51

US Hi-Yld Corp Bnd

S&P 500

0.45 0.39

0.37

0.35 0.32

0.31

0.31

0.32

0.31

0.30

0.29

0.30 0.31

0.15

0.00

Argentina

*

Brazil

Mexico

Venezuela

US Hi-Yld Bonds

with the exception of Argentina (May 1993-Dec 1995)

Source: JP Morgan, CS First Boston, Standard & Poor’ s, and Lehman Brothers

Page 14

Exhibit 9:

Year-by-Year Rates of Appreciation Against US Dollar Argentina

1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 Note:

0.00 -50.00 0.00 -71.43 -85.71 -78.97 -86.96 -77.68 -36.32 -66.48 -71.95 -99.02 -73.82 -47.56 0.05 0.00 -1.00 0.07

Brazil

Chile

Colombia

-66.67 -89.29 -91.30 -94.50 -84.52 -91.14 -96.29 -86.47 -12.55

-51.21 -37.70 -17.64 -12.95 0.00 0.00 -46.89 -16.11 -31.75 -30.25 -10.19 -14.03 -3.67 -16.87 -8.99 -10.36 -1.62 -10.86 3.63 -1.34

-9.25 -4.32 -7.41 -6.82 -13.59 -13.80 -15.96 -20.82 -22.06 -33.86 -21.37 -16.95 -21.49 -22.60 -17.03 -9.04 -7.08 -9.31 -17.95 -16.01

All returns are stated in US Dollars

Page 15

Mexico 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 -37.34 -12.27 0.09 -0.35 -1.98 -11.32 -72.81 -32.97 -25.25 -48.19 -59.75 -58.02 -5.18 -13.54 -8.97 -4.56 -0.87 0.35 -37.15 -35.80

Venezuela

-0.17 0.00 0.00 0.00 0.00 0.00 0.00 -0.17 -42.67 0.00 -48.28 0.00 0.00 -66.34 -13.39 -18.19 -22.21 -25.18 -38.51 -49.04

Exhibit 10:

Growth of $1 Invested in Latin American Currency (December 1986 to December 1995)

$1.00 Chile Colombia Mexico $0.80

Venezuela Argentina

Wealth in US Dollars

Brazil $0.60 $0.50 (-7.34%) $0.40

$0.22 (-15.43%)

$0.20

$0.12 (-20.99%)

$0.00 1986

Note:

1987

1988

1989

1990

1991

1992

1993

1994

$0.04 (-29.41%) $0.00 (-63.13%) $0.00 (-87.04%) 1995

Figures in parentheses indicate compound annual rates of return. All returns are stated in US Dollars.

Page 16

Appendix All calculations are performed on an annualized monthly data basis. All data are represented in US dollars. International Finance Corporation Investable (IFCI) Equity Indices The IFC Investable indices represent the portion of each country’ s equity market that is available to foreign investors. The data are a subset of the IFC Global index produced by the International Finance Corporation (IFC), the private sector arm of the World Bank. Data availability is as follows: Argentina Brazil Chile Colombia Mexico Venezuela

January 1989 to present January 1989 to present January 1989 to present February 1991 to present January 1989 to present February 1990 to present

For each stock, IFC maintains a time series with a 'foreign availability factor' that shows the share of total market capitalization that foreigners can purchase. This factor is multiplied by the stock's market capitalization to generate its investable market capitalization, which is used in calculations of the investable indices. The IFCI indices are weighted by investable market capitalization, measuring the change in components' value, adjusted for changes in capitalization and for addition or deletion of stocks to the indices. The market capitalization in the Total Returns Indices is adjusted for payment of cash dividends and stock dividends implicit in rights issues. Rights issues at market prices have no impact on total returns, while rights issues at prices above market decrease total returns. Stocks for the IFC Investable indices are selected from constituents of the IFC Global index based on their availability to foreign institutional investors; an investable market capitalization of at least US$25m; and meeting a liquidity screen for trading at least US$10m over the previous year. An IFC Global country index represents approximately 65-75 percent of the local market capitalization. Where multiple classes of stocks are common, IFC may include in the Index certain classes of stocks which are not actively traded, but whose presence is necessary to give a balanced view of the capitalization of companies which have other classes of stock actively traded. The IFC uses exchange rates from WM/Reuters daily spot rate service. Until October 1994, IFC was using exchange rates from The Wall Street Journal, or if not available there, the Financial Times (London). The exchange rates are collected for each Friday and for the last day of each month.

Page 17

Morgan Stanley Capital International (MSCI) Equity Indices MSCI Indices are designed to reflect the performance of the entire range of stocks available to investors in each local market. Stocks are chosen for the indices by the following criteria: 1) The MSCI Indices aim for 60% coverage of the total market capitalization for each market. 2) The companies included in the indices replicate the industry composition of each global market. 3) The chosen list of stocks includes a representative sampling of large, medium, and small capitalization companies from each local market, taking into account the stocks’ liquidity. 4) Stocks of non-domiciled companies and investment funds are not eligible for country indices. 5) Companies with restricted float due to dominant shareholders or cross ownership are avoided. Each stock in the local index is weighted by market capitalization. Dividend reinvestment methodology is as follows: • In the period between the ex date and the date of dividend reinvestment, a dividend receivable is a component of the index return. • Dividends are deemed received on the payment date. • To determine the payment date, a fixed time lag is assumed to exist between the ex date and the payment date. This time lag varies by country, and is determined in accordance with general practice within that market. Argentina and Brazil have a time lag of one month. Chile, Colombia, Mexico, and Venezuela have no time lags. • Reinvestment of dividends occurs at the end of the month in which the payment date falls. Data availability is as follows: Argentina Brazil Chile Colombia Mexico Venezuela

January 1988 to present January 1988 to present January 1988 to present January 1993 to present January 1988 to present January 1993 to present

Page 18

Local Market Equity Indices Local market indices were derived from daily price quotes provided by the Chicago Mercantile Exchange. Dividends are not included in the returns. Data availability is as follows: Merval (Argentina) Bovespa (Brazil) IPC (Mexico)

April 1994 to present November 1992 to present January 1987 to present

S&P 500 Total Return This index is a readily available, carefully constructed, market value weighted benchmark of common stock performance. Currently, the S&P Composite Index includes 500 of the largest stocks (in terms of stock market value) in the United States. From 1977 to the present, the common stock total return has been provided by the American National Bank and Trust Company of Chicago, based on monthly income numbers provided by Wilshire Associates, Santa Monica, California. Dividends (measured as of the ex-dividend date) are accumulated over the month and invested on the last trading day of the month in the S&P 500 Index at the day's closing level. Brady Bonds Brady bonds are represented by the J.P. Morgan Emerging Markets Bonds Index (EMBI) for the individual countries. These indices are total returns expressed in US dollars. The data availability for each country is as follows: EMBI Argentina EMBI Brazil EMBI Mexico EMBI Venezuela

April 1993 to present December 1990 to present December 1990 to present December 1990 to present

The EMBI indices track the traded market for dollar-denominated sovereign restructured bonds, which includes all Brady bonds and Brady-style bonds. The indices are marketcapitalization weighted and fully invested at all times. Individual bond prices are calculated based on daily changes in bid prices and exact coupon accrual and payment conventions. Index returns are calculated by weighting daily bond returns in proportion to market capitalization. The indices are always fully invested because cash payments are reinvested in the indices on the date paid. Rebalancing is done on the last business day of the month, and is only required when issues are added to or dropped from the indices, or when a government repurchases its bonds and cancels a portion of an outstanding issue. Sovereign restructured bonds which have not traded since issue, whose prices are seldom quoted, or which are not dollar-denominated issues are excluded from the indices.

Page 19

Lehman Brothers Government/Corporate Bond Index This data is available from 1973 to the present and is provided by Lehman Brothers Inc. This index is an aggregate of the Lehman Brothers (LB) Government Bond Index and the Lehman Brothers Corporate Bond Index. All returns are market-value-weighted inclusive of accrued interest. The LB Government Bond Index includes all public obligations of the U.S. Treasury, excluding flower bonds and foreign-targeted issues, all publicly issued debt of U.S. government agencies and quasi-federal corporations and corporate debt guaranteed by the U.S. government. All issues have at least one year to maturity and an outstanding par value of at least $100 million. Price, coupon and total return are reported on a month-end to month-end basis. All returns are market value weighted inclusive of accrued interest. The LB Corporate Bond Index includes all publicly issued, fixed-rate, nonconvertible investment grade domestic corporate debt and Yankee bonds. All issues have at least one year to maturity and an outstanding par value of at least $100 million. Price, coupon and total return are reported on a month-end to month-end basis. All returns are marketvalue-weighted inclusive of accrued interest. High-Yield Corporate Bond Portfolio The High-Yield Corporate Bond Portfolio is represented by the First Boston Corporation High-Yield Index (TM). Bonds included are weighted by rating and including ratings of BB/BBB & below. The First Boston High-Yield Index is designed to mirror the investable universe of the highyield public debt market. As a result, the index includes: new issues, fallen angels, private issues and defaulted issues. The index excludes securities with the following criteria: called, retired, exchanged, or upgraded issues. Currency Returns Currency returns are calculated using exchange rates as quoted by Bankers Trust Co., NY at 3 p.m. in The Wall Street Journal, or if not available there, the Financial Times (London). They represent the returns to a US investor.

Page 20

The financial support of the Chicago Mercantile Exchange is greatly appreciated. Matthew T. Moran and Michael Gorham of the Chicago Mercantile Exchange and Mark Riepe and Lori Lucas of Ibbotson Associates provided helpful comments. Hoa Quach made substantial contributions to the research. References Bekaert, Geert and Campbell R. Harvey, “Time-Varying World Market Integration,” Journal of Finance, June 1995. Erb, Claude B., Campbell R. Harvey, and Tadas E. Viskanta, “Do World Markets Still Serve as a Hedge?”Journal of Investing, Fall 1995. Gruaer, Robert R. and Nils H. Hakansson, “Gains from International Diversification: 1968-85 Returns on Portfolios of Stocks and Bonds,”Journal of Finance, July 1987. Harlow, W.V., “Asset Allocation in a Downside-Risk Framework,”Financial Analysts Journal, September-October 1991. Ibbotson Associates, Stock, Bonds, Bills, and Inflation 1996 Yearbook, Chicago. Ibbotson, Roger G., Laurence B. Siegel, and Paul D. Kaplan, “World Equities: The Past and The Future,”Global Portfolios: Quantitative Strategies for Maximum Performance, Robert Z. Aliber and Brian R. Bruce, eds., Irwin, 1991. Meyer, Barbara J., “International Investments”, Pension Investment Handbook, Mark W. Riepe and Scott L. Lummer, eds., Panel Publishers, 1996. Siegel, Laurence B. and Paul D. Kaplan, “Stocks, Bonds, Bills, and Inflation Around the World,”Managing Institutional Assets, Frank J. Fabozzi, ed., Harper & Row, 1990.

Page 21