Emerging market benefits, investability and the rule of law

Emerging Markets Review 12 (2011) 47–60

Contents lists available at ScienceDirect

Emerging Markets Review j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e m r

Emerging market benefits, investability and the rule of law Bonnie G. Buchanan a,⁎, Philip C. English II

b,1

, Rachel Gordon c

a

Department of Finance, Albers School of Business, Seattle University, 901 12th Avenue, Seattle, WA. 98122-1090, United States Dept of Finance and Real Estate, Kogod School of Business, American University, 4400 Massachusetts Ave NW, Washington, D.C. 20016-8044, United States c Department of Finance, LeBow College of Business, Drexel University, Academic Bldg. 2nd Floor, 101 N. 33rd Street, Philadelphia, PA 19104, United States b

a r t i c l e

i n f o

Article history: Received 20 March 2010 Received in revised form 6 September 2010 Accepted 7 September 2010 Available online 16 September 2010 JEL classification: G15 F21 F3

a b s t r a c t We revisit the Barry, Peavy and Rodriguez (1998) paper and investigate the underlying source of emerging market performance benefits. We classify stocks according to their investability and legal origin. Emerging markets continue to represent the performance benefits they had during the Barry et al. (1998) period by providing not only return enhancement but primarily risk-reduction. More specifically, we find that an investor can achieve greater benefits by focusing on a limited set of emerging markets with a French civil law foundation and that are moderately investable stocks. © 2010 Elsevier B.V. All rights reserved.

Keywords: Investability Law and finance Emerging markets

“Capital markets in developing countries have become an important asset class. These emerging markets are commonly associated with high returns, high volatility, and diversification benefits for investors in developed markets”Barry, Peavy and RodriguezFinancial Analysts Journal, 1998. The beneficial inclusion of emerging market stocks in developed market portfolios is something generally advocated and believed today. Indeed the overwhelming capital flows channeled into these markets over the past decade are testimony to their return enhancement and diversification potential. This capital allocation to emerging markets is in part due to the academic work documenting the benefits of including emerging market ⁎ Corresponding author. Tel.: + 1 206 296 5977; fax: + 1 206 296 2486. E-mail addresses: [email protected] (B.G. Buchanan), [email protected] (P.C. English), [email protected] (R. Gordon). 1 Tel.: + 1 202 885 2745; fax: + 1 202 885 1992. 1566-0141/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.ememar.2010.09.001

48

B.G. Buchanan et al. / Emerging Markets Review 12 (2011) 47–60

assets in developed market portfolios. Beginning with the seminal works of Solnik (1974), Errunza (1977), Harvey (1991, 1995), Bekaert and Harvey (1997, 2002a,b, 2003) and continuing to Barry et al. (1998) and beyond, the performance characteristics, both relative and absolute, of emerging market indices and stocks have been widely and thoroughly investigated. The general conclusion, following Barry et al. (1998), is that investors gain higher returns and lower volatility in their overall portfolio over time by including holdings of emerging markets stocks as represented by their country-level indices. What remains in part, as they note in their concluding remarks, is to investigate the underlying source of the performance benefits. We revisit the Barry et al. (1998) paper to answer three questions: 1) As markets have become more global and integrated, has the benefit from investing in emerging market stocks changed particularly for those stocks that are investable? 2) Has the increased level of holding emerging market stocks changed the level of inclusion at which they should appear in investor portfolios2? 3) Are there characteristics logically associated with the legal traditions and families of the differing emerging markets that suggest allocation to individual emerging markets should occur along lines of legal tradition?3 Emerging market capitalization has soared in recent years, growing from less than $2 trillion in 1995 to over $12.87 trillion in 2009.4 Developing markets account for approximately 84% of the world's population in 2009 and 26.8% of the world's GNI.5 As in 1998, the emerging markets in 2008 exhibited high growth rates (7% GDP growth in emerging markets compared with 2% growth globally6) and the potential for continuing such rates into the future. We end our sample in 2006 due to data constraints and uncharacteristically volatile world markets. While investigation of the value of emerging market diversification during the financial crises may prove insightful, such an analysis would be premature until the crisis has ended and the recovery is complete. Recent work by Reinhart and Reinhart (2010) suggests that a return to “normality” could take as much as a decade. In addition, there is continued debate regarding this matter including a failure to declare the recession over by the National Bureau of Economic Research (Tseng, 2010). Given the wide-spread turmoil in global markets in 2007–2009, analysis of this period shows abnormal patterns and would not be representative of the role developing markets play. Prior to the financial crises, the volatility of world market monthly returns, as measured by the standard deviation of the monthly return to the Morgan Stanley Capital International World Index (MXWO), for the 3 years prior to the crises (2004–2006), was 0.023 or 2.3% per month while during the crises (2007–2009) it was 0.062 or 6.2% per month.7 Finally, Barry et al. (1998) cast doubt on the availability of diversification benefits during such tumultuous periods. Over the decade prior to the crises, financial markets have responded to the diversification benefits and return enhancement available in emerging markets by channeling (pouring) capital into these markets and even going so far as to create products designed to be attractive to investors interested in particular markets or sets of markets. By far, the most well-known of these are the BRIC funds which represent the high growth economies of Brazil, Russia, India and China. Accordingly, during our analysis, we also analyze

2 Barry et al. (1998) note in their concluding remarks that the optimal allocation to emerging markets changes from period to period. The answer to this question is effectively an update of their findings with regards to the proportion of the portfolio allocated to emerging markets. 3 In addition to the emerging markets, the World Bank now recognizes a classification of markets known as frontier markets. In our follow up paper, we explore the role of frontier markets in the portfolio allocation and diversification process. 4 Source: World Development Indicators (WDI). 5 Source: World Development Indicators (WDI). The sample markets we use in this paper account for 63% of the world population. 6 Source: World Development Indicators (WDI). 7 The variance estimates for these two periods are statistically significantly different at the p = 0.0001 level. For robustness, the average monthly standard deviation for the same index over the 7 non-overlapping 3 year periods within our sample is 0.041 or 4.1% per month. The difference between the average variance of monthly returns across the non-overlapping 3 year periods prior to the crises is different from the crises period monthly variance at the p = 0.0002 level.

B.G. Buchanan et al. / Emerging Markets Review 12 (2011) 47–60

49

Fig. 1. Compound value of a $1 investment in different indices (log basis).

the performance of the BRIC country indices as a separate set of emerging markets. Recent events at the Copenhagen Climate Summit suggest the BRIC countries may also align politically on certain matters to form an international voting block. This is an additional reason to analyze them independently. 1. Performance of emerging markets Following Barry et al. (1998), we use the Emerging Markets Data Base (EMDB) formerly supplied by the International Finance Corporation (IFC) and now provided by Standard & Poors (S&P).8 Our sample period starts in December 1988 and ends in February 2006. Fig. 1 shows the performance of the aggregate emerging markets composite (EFM) along with the US S&P 500, Nasdaq, T-bills and Consumer Price Index (CPI) on a log basis. As can be seen, the emerging market composite performs in a manner similar to that found in the original BPR study. Fig. 2 breaks the emerging markets composite performance into four categories: the composite representing English common law countries9; the composite representing French civil law countries10; the composite representing German civil law countries11; and the BRIC countries.12 The first three of these categories are chosen based on the seminal research of LaPorta et al. (1997, 1998) who show large differences in market capitalization and performance across differing legal traditions (common vs civil law) and families (English, French, German and Scandinavian13). Earlier contributions to the literature also include Pistor et al. (2000) who argue that both the quality and effectiveness of legal frameworks are necessary for financial development. Levine et al. (2000) find that

8

Appendix A details the construction of the performance attributes and the coverage of the data. The 8 English common law countries in the dataset are: India, Israel, Malaysia, Pakistan, South Africa, Sri Lanka, Thailand and Zimbabwe. 10 The 16 French civil law countries in the dataset are: Argentina, Brazil, Chile, Colombia, Egypt, Greece, Indonesia, Jordan, Mexico, Morocco, Peru, Philippines, Portugal, Russia, Turkey and Venezuela. 11 The 7 German common law countries in the dataset are: China, Czech Republic, Hungary, Korea, Poland, Slovakia and Taiwan. 12 Note the performance of this composite is not necessarily the same as the performance of any specific BRIC product due to differences in composition, rebalancing and fees. The performance displayed here represents a US dollar value-weighted investment in the entire index for each of the BRIC countries beginning before BRIC products were available as prepackaged investments. 13 There are no Scandinavian civil law emerging markets in our sample. 9

50

B.G. Buchanan et al. / Emerging Markets Review 12 (2011) 47–60

Fig. 2. Compound $1 investment starting in 1994.

countries with English legal origin have stronger levels of financial intermediation than countries of French legal origin and these results are also confirmed by Harper and McNulty (2008). Buchanan and English (2007) find that French civil law based emerging markets earn higher returns, have higher correlations with the world market portfolio and have higher average market capitalization growth rates and lower average market capitalization/GDP ratios than their English counterparts. More recently, Billmeier and Massa (2009) find that institutional quality has a positive and significant impact on stock market capitalization. Noting that the US is an English common law country, we reason that some or all of the diversification benefits accruing to emerging markets across the period examined by Barry et al. (1998) and prior researchers may result from low correlation between differing countries induced by a lack of common legal foundation and language. If this is indeed the case, it may be possible to achieve the same diversification and performance

Table 1 Summary performance characteristics 1988–2006. Index

Geometric mean (%)

Arithmetic mean (%)

Standard deviation (%)

French English German EFM French exBRIC English exBRIC German exBRIC BRIC EFM exBRIC Nasdaq SP500 CPI Tbills

6.021 0.550 −0.347 3.042 4.732 0.564 −0.401 4.760 2.472 0.872 0.745 0.243 −0.291

7.028 0.724 −0.077 3.489 5.708 0.769 −0.134 5.972 2.914 1.122 0.828 0.243 −0.128

17.447 5.870 7.431 10.386 17.103 6.206 7.524 17.196 10.319 7.024 4.077 0.265 5.655

Note: EFM refers to Emerging Market Composite Index and the sample is further subdivided into legal origin (English, French, German) and more specialist funds such as the BRIC (Brazil, Russia, India, China countries).

B.G. Buchanan et al. / Emerging Markets Review 12 (2011) 47–60

51

Table 2 Correlation summary 1988–2006. Correlation

French English German EFM French exBRIC English exBRIC German exBRIC BRIC EFM exBRIC Nasdaq SP500

French

English

German

EFM

French exBRIC

English exBRIC

German exBRIC

BRIC

EFM exBRIC

1.0000 0.2336 0.2851 0.9446 0.9509

1.0000 0.5824 0.4695 0.2094

1.0000 0.5938 0.2699

1.0000 0.9147

1.0000

0.2211

0.9662

0.5553

0.4513

0.2046

1.0000

0.2810

0.5726

0.9726

0.5803

0.2658

0.5512

1.0000

0.8456 0.9038

0.3057 0.4600

0.3012 0.5982

0.6796 0.9758

0.2577 0.9374

0.2398 0.4598

0.2691 0.6027

1.0000 0.3138

1.0000

0.1900 0.1585

0.4323 0.4557

0.4637 0.4214

0.3169 0.2900

0.1863 0.1690

0.4121 0.4551

0.3899 0.3847

0.2197 0.1586

0.2953 0.2905

Nasdaq

SP500

1.0000 0.8019

1.0000

benefits by focusing on a subset of emerging markets. Since expanding the asset classes an analyst or portfolio managers covers increases expenses and emerging market investment is inherently more expensive than investment in the S&P 500, reducing the set of countries under consideration while achieving the same benefits is advantageous to manager and client alike. Inspection of the performance characteristics in Table 1 immediately reinforces the conclusions of Barry et al. (1998): emerging markets have high returns and high volatility. Of particular note are the high returns and high volatility of the BRIC index and the French civil law country indices. The overall composite (labeled EFM) has a higher return and standard deviation than that found in Barry et al. (1998) (1.50% and 6.65% respectively). It is also higher than the return and standard deviation of the S&P 500 which is consistent with their findings. Table 2 provides insight into the diversification potential of these markets. The correlation between the S&P500 and English common law or German civil law indices is relatively high and relatively low for the BRIC economies and the French civil law economies. Correlation, return and standard deviation alone are not the entire story. The manner in which these combine to affect the efficient frontier determines the desirability of the differing asset classes to an investor. Table 3 provides the proportion of the minimum variance portfolio invested in each of the selected indices and the S&P 500. From this information, we can extract the entire efficient frontier resulting from including the various indices with the S&P 500 in varying proportions (shown in Fig. 3 for selected combinations14). Barry et al. (1998) find that over their entire sample period, investors can achieve the highest efficient frontier when the minimum variance portfolio has approximately 20% invested in the emerging markets composite and 80% in the S&P500. We find that, consistent with increased inclusion of emerging market assets in US portfolios, increased capital flows to emerging markets and increased globalization, the highest efficient frontier attainable with using an overall emerging markets index and the S&P 500 has only 4.42% invested in the emerging markets composite and 95.66% in the S&P500 at the minimum variance portfolio. Investigation of the alternative investment strategies using a reduced set of countries segmented by legal tradition as detailed in Table 3 reveals several interesting portfolios. First, the combination of the English law countries with the S&P500 results in a minimum variance portfolio similar to that found by BPR: 19.64%

14 The portfolios comprised of the German index or English index in combination with the S&P 500 have been omitted for clarity of presentation. They are dominated at all points by at least 2 of the presented portfolios and at most points by all of the presented portfolios.

52

B.G. Buchanan et al. / Emerging Markets Review 12 (2011) 47–60

Table 3 Proportion of the minimum variance portfolio invested in each of the selected indices and S&P500. Minimum variance portfolio representation (%)

Index

Return (%)

Standard deviation (%)

Correlation with S&P 500

4.424 1.809 19.641 8.752 1.720 16.065 10.135 4.539 1.922

EFM French English German French exBRIC English exBRIC German exBRIC EFM exBRIC BRIC

0.946 0.940 0.808 0.749 0.912 0.819 0.730 0.890 0.927

4.053 4.065 3.935 4.033 4.066 3.973 4.013 4.036 4.064

0.2900 0.1585 0.4557 0.4214 0.1690 0.4551 0.3847 0.2905 0.1586

emerging market (English common law only) and 80.36% S&P500. This is consistent with the conclusions of LaPorta et al. (1997) and, given the commonality of legal environment and, typically, language, is not surprising. Of even greater interest is minimum variance portfolio composed of French civil law investments and the S&P500. An investment of 1.81% in the French civil law countries provides a minimum variance portfolio with a return almost identical to that of investing in the complete emerging market index (0.94% vs 0.95%) but a standard deviation only marginally higher (4.065% vs 4.053%). Fig. 3 depicts the efficient frontier constructed from the emerging markets composite and the S&P500 against select efficient frontiers constructed with the S&P in combination with a BRIC investment and in the French civil law countries. As can be seen, the efficient frontier constructed of the French civil law countries dominates the other efficient frontiers over most of the range. The BRIC index provides little improvement over the EFM index and is significantly inferior to it in the lower portion of the frontier. Finally, it is particularly interesting to note the low level of inclusion of the BRIC index in the minimum variance portfolio relative to the other available investment strategies and its relatively poor performance with regards to return and standard deviation. The BRIC countries include 2 French, 1 German, 1 English legal tradition countries. The level of BRIC return was similar to the level of English common law return and below all but the German indices (both with and without BRIC inclusion). The level of standard deviation for the BRIC portfolio was the highest of all the derived indices. This could be due to a variety of factors including the extreme popularity of BRIC instruments driving up the price of BRIC investment in the latter

Fig. 3. Selected efficient frontiers.

B.G. Buchanan et al. / Emerging Markets Review 12 (2011) 47–60

53

Table 4 Summary performance characteristics segmented by investability classification and legal family. Index

Low investable

Moderately investable

Highly investable

Geometric Standard Correlation Geometric Standard Correlation Geometric Standard Correlation mean (%) deviation with S&P mean (%) deviation with S&P mean (%) deviation with S&P 500 (%) 500 (%) 500 (%) EFM 1.853 French 4.702 English 0.210 German −0.144 French exBRIC 4.161 English exBRIC −0.044 German exBRIC 0.071 EFM exBRIC 1.724 BRIC 0.718

9.479 14.661 7.301 9.635 15.012 8.436 8.470 9.571 7.630

0.3410 0.2181 0.3421 0.3523 0.1943 0.3804 0.2949 0.3122 0.4321

5.113 9.224 0.555 −0.044 6.058 0.558 −0.055 3.264 −1.613

14.361 26.627 6.802 7.907 24.753 6.829 8.041 13.096 30.851

0.2038 0.0891 0.4442 0.5212 0.1258 0.4483 0.5024 0.2686 0.0623

3.688 5.452 0.769 0.442 4.495 0.743 1.192 3.029 6.615

14.906 19.653 6.403 9.224 19.506 6.453 12.000 14.792 29.422

0.1933 0.1524 0.4232 0.4914 0.1557 0.4165 0.3977 0.1944 0.0826

period and thereby resulting in poorer returns. By contrast, one advantage to a BRIC strategy is the relatively low correlation with the performance of the S&P 500.

2. Investability We now turn our attention to a major facet of the original Barry et al. (1998) paper and an evolving feature of emerging markets, the extent to which the stocks of the various emerging markets are investable. In the literature the investability measure is often used for firm-specific dates of liberalization. Such papers can be found in Bae et al. (2004, 2006), Mitton (2006), Cũnado et al. (2006) and Boyer et al. (2006). Bae et al. (2006) state “the impact of increased openness is typically greatest for countries with poor quality of legal, disclosure and economic environment”. Jayasuriya (2005) proposes that investable markets should possess less tail risk than non-investable markets, which tend to be more influenced by adverse local factors. This is then used as an argument for the opening of investable markets to foreign investors. Mitton (2006) finds that investable companies experience higher growth, greater investment, greater profitability, greater efficiency and lower leverage. More recently, Flavin and O'Connor (2010) find that if a firm becomes investable, or cross-lists, then there tends to be an increase in equity and a reduction in debt. Standard and Poor's classifies the ability of foreign investors to access specific stocks on a scale ranging from 0 (completely inaccessible) to 1 (completely investable). The level at which a stock is classified as investable is determined by a variety of factors including market capitalization, trading volume, float, legal restrictions including foreign ownership limitations, liquidity, the mechanisms by which foreign investors can repatriate dividend income (without undue constraint), capital gains and initial investment capital.15 De Jong and De Roon (2005) and Edison and Warnock (2003) also construct capital controls at the country level. At the firm level, Bae et al. (2004) investigate how investability explains cross-sectional variations in individual firm volatility and correlation with the world market. They find that investable stocks are more volatile due to their increased world market exposure, a finding consistent with our evidence that the benefits of English common law emerging markets to a resident of the US are lower in our sample period than in that of Barry et al. (1998). Inherently, investability is not a binomial state. Bae et al. (2004) partition their stock samples into three investable categories: non-investable (investable weight= 0), partially investable (0b investable weight b=0.5), and highly investable (0.5 b investable weight b=1.0). We refine the original Barry et al. (1998) research by classifying stocks according to their investability decile rather than whether they are completely investable or not. We then group the stocks into three categories: Low Investable stocks (investable weight in

15

The Standard and Poor's Emerging Markets Equity Indices Handbook provides details on the criterion.

54

B.G. Buchanan et al. / Emerging Markets Review 12 (2011) 47–60

Table 5 Minimum variance portfolio representation invested in each of the selected indices and the S&P500 further segmented by investability. Index

EFM French English German French exBRIC English exBRIC German exBRIC EFM exBRIC BRIC

Minimum variance portfolio representation Lowinvestable (%)

Moderately investable (%)

Highly investable (%)

4.370 1.774 13.060 3.695 2.194 5.832 9.695 5.354 6.719

2.385 0.990 11.366 −4.015 0.659 10.928 −2.967 1.471 0.925

2.290 1.180 15.791 −2.762 1.154 15.682 −2.270 2.336 0.782

the lowest three deciles); Moderately Investable (investable weight in the middle four deciles); and Highly Investable (investable weight in the top three deciles). Stocks in the highest decile are more investable than stocks in lower deciles. As emerging market penetration occurs, we would expect stocks that have higher levels of investability to have increased correlation due to increased capital mobility and therefore reduced diversification benefits.16 We further extend Barry et al. (1998) by examining investability in concert with legal environment. Table 4 provides the return to the overall emerging market indices segmented by investability classification and further segmented by legal family. Table 5 provides the proportion of the minimum variance portfolio by the same. Inspection of Table 4 provides several immediate insights. First, the Low-Investable indices generally have lower return and no regular pattern of standard deviations (some lower, some higher, some similar) when compared to the Moderately Investable indices. Contrary to our expectations, they also were relatively highly correlated with the S&P 500.17 Second, the level of realized return in the Moderately Investable category is generally higher for stocks in the French civil law countries vis-à-vis the other legal families and this is not driven by the BRIC country stocks in the French index. The Moderately Investable BRIC stocks and the French index, with and without the BRIC stocks, had relatively low correlation with the S&P 500, a strong indicator of their diversification potential. Finally, the Highly Investable indices, as increased capital mobility and interest in the emerging markets would suggest had relatively high correlations with the S&P 500. Like the Moderately Investable category, this was not the case for the French indices or the BRIC index. It is also worth noting that, in most cases, the Moderately Investable indices had higher returns and similar standard deviations to the Highly Investable indices. The impact of investability on the composition of the differing possible efficient frontiers is immediately apparent in Fig. 4. When we examine the mean-variance efficiency of the differing investability classifications, the moderately investable classification dominates the other classifications and the overall EFM index. For the Highly Investable stocks, this is consistent with a coupling of increased interest in emerging markets and increased capital mobility bidding up prices thereby decreasing returns, increasing standard deviations (as evidenced in Bekaert and Harvey (1997, 2000, 2002a), and increased correlation with the S&P 500. For the Low Investable indices, this is consistent with the lack of capital mobility resulting in little bidding up of prices, etc. The stocks that offer investors the greatest possibilities for return enhancement and diversification potential are those that are not the easiest to access but are still attainable.

16 There is a body of literature that examines this aspect of investability focusing solely on financial market liberalizations caused by changes in country level economic policies (see Chari and Henry (2004), Henry (2000a,b) and Cũnado et al. (2006)). Acknowledging that liberalization allows for increased investment, it does not of necessity lead to it. We choose instead to focus on a measure which captures the full spectrum of possibility for increased investment beyond just changes in the legal environment making such investment permissible. 17 The source of this correlation in the real economy warrants greater exploration. We leave that exploration to further research.

B.G. Buchanan et al. / Emerging Markets Review 12 (2011) 47–60

55

Fig. 4. EFM index.

Fig. 5. French index.

We now combine our finding that French civil law indices offer better diversification potential with the impact of differing investability in Fig. 5. Here we see a similar pattern for the French index: the highest efficient frontier is comprised of the Moderately Investable stocks.18 When we refine this pattern by inspection of the weights comprising the minimum variance portfolios in Table 5, we note that a relatively small investment in the French composite provides as much benefit as a much higher investment in either the overall EFM composite or the BRIC composite. To the extent that transaction costs or monitoring costs are higher for emerging market investments, this suggests that investors with an S&P 500 referent portfolio can

18 It is particularly interesting to note that for the BRIC index, the Moderately Investable and Highly Investable efficient frontiers are almost coincident (not depicted but available upon request). The extreme level of interest in the BRIC countries has likely made any possible investment there exhibit similar characteristics.

56

B.G. Buchanan et al. / Emerging Markets Review 12 (2011) 47–60

achieve greater benefits by focusing on limited set of the emerging markets those with a French Civil law foundation and concentrating on the moderately investable stocks rather than those which are easily accessed.19 3. Concluding remarks Emerging markets continue to represent the performance benefits they had during the Barry et al. (1998) period by providing not only return enhancement but primarily risk-reduction. The increased focus on these markets has dramatically decreased their representation in the minimum variance portfolio when examining emerging markets in their entirety as a single asset class. When segregating emerging markets along the dominant rule of law in the varying countries, the benefits accruing to emerging market investment appears related to the low correlation between the referent investor, in our case an investor in the United States – an English common law country, and a differing legal tradition, during this period to that of French civil law. When we further inspect the impact of investability, as measured by S&P, we find that the benefits of emerging market stocks lie largely with those not as easily accessed, the moderately investable category. Combining these two features, we show that the low correlations resulting from differing legal traditions and lower investability result in an efficient frontier composed of French civil law tradition moderately investable stocks that dominates the other potential asset combinations during this time period over most of the efficient frontier. By judiciously selecting the countries, investors can achieve the benefits of investing in emerging markets with a fraction of the incremental investment and cost.

Acknowledgements We wish to thank Tom Arnold, Bruno Solnik, John Tobey, the Editor Jonathan Batten and anonymous reviewers, Valerie Cerra and participants at the Western Economic Association International 2010 Annual Conference for their helpful comments. Buchanan also acknowledges an Early Career Researchers Grant from the University of Melbourne. Appendix A. Data Used in the EMDB and Return Calculations We follow Barry et al. (1998) in our calculation of the performance characteristics of the emerging market indices and the development of the sub-indices representing the BRIC economies and the differing legal environments. One phenomenon only rarely present in their sample period plays a larger role during ours. In 11 countries during our sample period, there are approximately 90 instances where firms in several countries (Argentina, Brazil, Indonesia, Israel, Korea, Malaysia, Mexico, Pakistan, Peru, Philippines, and South Africa), engage in extremely large reverse stock splits including one with a 50,000:1 ratio (shareholders tendering 50,000 shares received one new share in return).20,21 We leave the analysis of this occurrence to future research and treat this in the manner of a diversified but informed foreign investor: when the stock reverse splits, we tender and sell the shares, effectively removing it from our portfolio. We then rebalance the portfolio to the appropriate value-weightings for the remaining stocks and do not allow the splitting stock to re-enter the portfolio during our sample period.22

19 When we combine the full spectrum of investability and legal foundation into one graph of differing efficient frontiers, the moderate investable French Civil law index dominates the other possibilities along most of the efficient frontier. The moderate investable overall EFM and the French overall frontiers are almost coincident. 20 Out of the 11 countries, Brazil and Korea have by far the greatest number of firms with large reverse stock splits. 21 There are five stocks which have two large reverse stock splits and one that has three over several years. These stocks are removed from the portfolio following the first instance of a large reverse split. 22 A number of these stocks subsequently reappear in the dataset on average 2–3 years later.

B.G. Buchanan et al. / Emerging Markets Review 12 (2011) 47–60

57

There are 31 countries, listed in Table A-1by legal origin, that are included in the emerging market composite. The data for these countries was extracted from the S&P Emerging Market Database (EMDB). From the 31 countries, approximately 2713 unique stocks spanning part or all of the 207 monthly time periods representing just over 143,000 observations for the analysis. Table A-1 Countries included in composite. Legal origin

Market

Date data available

Israel

January

1997

India Malaysia Pakistan Sri Lanka Thailand

November December February December December

1992 1988 1991 1992 1988

South Africa Zimbabwe

December July

1992 1993

Greece Portugal Russia Turkey

December December February August

1988 1988 1997 1989

Argentina Brazil Chile Colombia Mexico Peru Venezuela

December December December February December December January

1988 1988 1988 1991 1988 1992 1990

Jordan

December

1988

Philippines

December

1988

Indonesia

September

1990

Egypt Morocco

February February

1997 1997

Czech Republic Hungary Poland Slovakia

January December December February

1994 1992 1992 1997

China Taiwan Korea

December January January

1992 1991 1992

English legal origin Middle East South Asia

Africa

French legal origin Europe

Latin America

Middle East East Asia South Asia Africa

German legal origin Europe

East Asia

The overall Emerging Market Composite Index (EFM) was calculated in three separate parts. First, individual stock returns were calculated (based in local currency) per time period for each country that took into account new share issues, cash dividends, and new rights issued for each share. Next, the market-value weighted index for each country per time period was calculated and the country-level returns were multiplied by the foreign currency appreciation or depreciation. Lastly, each country received a weight based on its total market cap per time period and the entire market cap of all of the emerging market countries for that period. The weighted country returns were then aggregated to obtain the EFM Index return for each period.

58

B.G. Buchanan et al. / Emerging Markets Review 12 (2011) 47–60

Stock return calculations Share returns were calculated following the original development in Barry et al. (1998), such that:   Rt * Subt
ðPt * St Þ 1− Pret St−1 + ðRt * Subt Þ ðDt * St Þ−ðNt * Pt Þ * Local Returnit = + −1 ðPt−1 * St Þ ðPt−1 * St Þ where Pt St Rt Subt Pret Dt Nt

Price per share of the stock at time t Shares outstanding of the stock at time t Shares from new rights issues of the stock at time t Subscription price for the new rights issued of the stock at time t Prerights issue price of the stock at time t Cash dividends of the stock at time t New shares issued of the stock at time t

Country return calculations Returns for each country were calculated based on the aggregation of the market weighted returns for each stock for each time period.   Mit−1 Country Returnt = ∑Local Returnit * MCt−1 □ where Mct − 1 Total market capitalization of the country at specific time period Mit − 1 Market capitalization of the stock at specific time period Local Returnit Calculated return per share of the stock at time t The aggregate of the share returns in local currency for each country was then multiplied by the appreciation/depreciation of the currency for each country for the given time period such that:  Returnct = ð1 + Country Returnct Þ *

ERct ERct−1

 −1

where Country Returnct Calculated return per country for time period t ERct Exchange rate per country for time period t

Creation of the EFM Composite Index The EFM Composite Index is based on the total market capitalization for each country in dollars for time period t divided by the total market capitalization of all of the emerging countries in the dataset for the given time period such that:   Mct−1 EFM Returnt = ∑Country Returnct * Met−1 □ where Met − 1

Total market capitalization of all countries for time period t

B.G. Buchanan et al. / Emerging Markets Review 12 (2011) 47–60

59

Creation of legal origin and BRIC indices To create the returns for the Legal Origin Indices and for the BRIC Index, the weights for each country's return for each time period were re-adjusted following the same procedure as above. To calculate the index returns for each legal origin, the country level returns for each time period were re-weighted using the total market capitalization of each specific legal origin instead of the total market capitalization for all of the emerging market countries. The process then followed the same steps as above. The BRIC Index was created starting in December of 1992. At this point, there existed data for India, China, and Brazil. Prior to December of 1992, Brazil was the only “BRIC” country to have stock market data in the dataset and thus was included with the rest of the emerging markets. To calculate the BRIC Index, the total market capitalization of Brazil, Russia, India and China was used for each time period and each country was weighted accordingly. It is important to note that Russia did not enter the dataset, and the BRIC Index, until February of 1997. The EFM Index without the BRIC countries consisted of removing the four BRIC Number of observations

French English German Total

Low investable

Moderately investable

Highly investable

5491 16,599 35,708 57,798

6764 15,682 23,691 46,137

12,974 14,471 11,771 39,216

countries after December of 1992 and recalculating each country's weight after that point. The earliest existing BRIC product was offered by Goldman Sachs in 2001. Creation of investability indices Returns were categorized into deciles using the investable variable in the EMDB dataset. The investable variable ranges from 0 to 1, with a “1” indicating a stock that is completely investable. The deciles were grouped into three categories for the purpose of analysis. Low Investable stocks are stocks with investability rankings less than 0.3. Moderately Investable stocks encompass the middle range between 0.3 and 0.7. Highly Investable stocks have a ranking greater than 0.7. Using these new groupings, index returns were re-calculated under the various subsets (e.g. legal origins and BRIC). The numbers of observations available are as follows: References Bae, Kee-Hong, Chan, K., Ng, A., 2004. Investability and return volatility. Journal of Financial Economics 71, 239–263. Bae, Kee-Hong, Bailey, W., Mao, C., 2006. Stock market liberalization and the information environment. Journal of International Money and Finance 25, 404–428. Barry, C., Peavy, J., Rodriguez, M., 1998. Performance characteristics of emerging capital markets. Financial Analysts Journal 72–80 Jan-Feb. Bekaert, G., Harvey, C.R., 1997. Emerging equity market volatility. Journal of Financial Economics 43, 29–77. Bekaert, G., Harvey, C.R., 2000. Foreign speculators and emerging markets. Journal of Finance 55, 565–614. Bekaert, G., Harvey, C.R., 2002a. Dating the integration of world equity markets. Journal of Financial Economics 65, 203–248. Bekaert, G., Harvey, C.R., 2002b. Research in emerging market finance: looking to the future. Emerging Markets Review 3, 429–448. Bekaert, G., Harvey, C.R., 2003. Emerging markets finance. Journal of Empirical Finance 10, 3–55. Billmeier, A., Massa, I., 2009. What drives stock market development in emerging markets–institutions, remittances or natural resources. Emerging Markets Review 10, 23–35. Boyer, B., Kumagai, Yuan, K., 2006. How do crises spread? Evidence from accessible and inaccessible stock indices. Journal of Finance 61, 957–1003. Buchanan, B., English, P.C., 2007. Law, finance and emerging market returns. Emerging Markets Review 8, 181–193. Chari, A., Henry, P.B., 2004. Risk sharing and asset prices: evidence from a natural experiment. Journal of Finance 59, 1295–1324. Cũnado, J., Biscarri, J.G., de Gracia, F.P., 2006. Changes in the dynamic behavior of emerging market volatility: revisiting the effects of financial liberalization. Emerging Markets Review 7, 261–278. De Jong, F., de Roon, F., 2005. Time-varying market integration and expected returns in emerging markets. Journal of Financial Economics 78, 583–613.

60

B.G. Buchanan et al. / Emerging Markets Review 12 (2011) 47–60

Edison, H.J., Warnock, F.E., 2003. A simple measure of the intensity of capital controls. Journal of Empirical Finance 10, 81–103. Errunza, V., 1977. Gains from portfolio diversification into less developed countries securities. Journal of International Business Studies 8, 83–99. Flavin, T., O'Connor, T., 2010. The Sequencing of Stock Market Liberalization Events and Corporate Financing. Emerging Markets Review 11, 183–204. Harper, J.T., McNulty, J.E., 2008. Financial system size in transition economies: the effect of legal origin. Journal of Money, Credit and Banking 40, 1263–1280. Harvey, C.R., 1991. The world price of covariance risk. Journal of Finance 46, 111–157. Harvey, C.R., 1995. Predictable risk and returns in emerging markets. Review of Financial Studies 8, 773–816. Henry, P.B., 2000a. Do stock market liberalizations cause investment booms? Journal of Financial Economics 58, 301–334. Henry, P.B., 2000b. Stock market liberalization, economic reform and emerging market equity prices. Journal of Finance 55, 529–564. Jayasuriya, J., 2005. Stock market liberalization and volatility in the presence of favourable market characteristics and institutions. Emerging Markets Review 6, 170–191. LaPorta, R., Lopez-de-Silanes, F., Shleifer, A., Vishny, R., 1997. Legal determinants of external finance. Journal of Finance 52, 1131–1150. LaPorta, R., Lopez-de-Silanes, F., Shleifer, A., Vishny, R., 1998. Law and finance. Journal of Political Economy 106, 1113–1155. Levine, R., Loyaza, N., Beck, T., 2000. Financial intermediation and growth: causality and causes. Journal of Monetary Economics 46, 3–77. Mitton, T., 2006. Stock market liberalization and operating performance at the firm level. Journal of Financial Economics 81, 625–647. Pistor, K., Raiser, M., Gelfer, S., 2000. Law and finance in transition economies. Economics of Transition 8, 325–368. Reinhart, C.M. and Reinhart, V.R., 2010. “After the Fall”, working paper, University of Maryland and American Enterprise Institute. Solnik, B., 1974. “Why not invest internationally? Financial Analysts Journal 30, 48–54. Tseng, N., 2010. Did the recession actually end? Fortune Aug 6, 2010. http://money.cnn.com/2010/08/06/news/economy/ NBER_recession.fortune/index.htm.