Cross Listing And Liquidity

Cross listings and Liquidity Lígia Catarina Marques Febra ([email protected]) Escola Superior de Tecnologia e Gestão do Instituto Politécnico de Leiria Campus 2, Morro do Lena, Alto Vieiro, Apartado 4163, 2411-901 Leiria, Portugal

This version: May 2007

Abstract

My research examines the effects of international cross listings on liquidity. We first measure the impact of international listings in terms of the volume of the cross- listed stocks, for the home and international markets, and evaluate which factors drive the fragmentation of the order flow. We also look upon any spillover effects in terms of the changes in liquidity of the non-cross listed firms and of the aggregate home market. Second, we reanalyse the impact of cross listings in returns. In particular we evaluate whether changes in liquidity account for the observed changes in returns.

JEL Classification: G12, G15 Keywords : international capital markets, international cross- listings, stock returns, liquidity, location. ______________________________________________________________________

1. Introduction

It is well established (Foerster and Karolyi (1993); Karolyi (1998); Jayakumar (2002)) that there is an increase in trading volume for cross- listed stocks. Further, most studies report that the increase in trading volume observed for the international market is accompanied with increased trading in the home market (Foerster and Karolyi (1993); Korczak and Bohl (2003)) but there is some contradictory evidence regarding this latter fact (Levine Schmukler (2003)). The purpose of our analysis is twofold. First we reanalyse whether the observed changes in liquidity are significant for an extensive worldwide sample of cross listings. We evaluate the impact in terms of home and international trading volume and propose a model to evaluate the role of a set of previously identified factors in explaining the fragmentation of the order flow. Given that we cover a large cross-section of countries and listing locations we also measure the impact for the aggregate liquidity across markets in order to evaluate any related spillover effects. Second, we investigate whether the well-established effect on returns of cross- listed shares is associated with the changes in liquidity that follow the cross listing. Our analysis is motivated by the work of Amihud and Mendelson (1986) that show that liquidity is priced risk factor in the return generating process. International cross listing is one, among several corporate policies, that may impact the liquidity of shares (see, for example Alexander, Eun and Janakiramanan (1988)). Our main contribution is to evaluate the role of liquidity in the observed effects of cross listings using an extensive data set that covers a large number of countries and listing locations, and to shed light on the important factors that drive the cross-sectional variation in the observed effects. The structure of this proposal is the following. We first present a brief literature review and previous findings for the two issues analysed. Second, for each essay, we present the testable hypotheses (and expected results) and the proposed methodology.

2. Related literature and previous findings

2.1. Cross listing: trading volume and fragmentation of the order flow

2

Mittoo (1992) and Bancel and Mittoo (2001) report, on the basis a survey done with Canadian and European firms, that managers perceive that international cross listings increase the total trading volume of the share of a firm. In fact several studies have looked upon the effects of cross listing on trading volume. Karolyi (1998) and the references therein conclude that there is overwhelming evidence that the total volume of trading increases following an international cross listing. In many cases the evidence shows that trading volume in the home market also increases. Most of these results are corroborated in later studies using different samples (Jayakumar (2002); Korczak and Bohl (2003)). Yet there are contradictory results on the impact of international cross listings in terms of the fragmentation of the order flow. While Levine and Schmukler (2003) find a reduction in the trading volume of cross listed shares in the home market, Halling, Pagano, Randl and Zechner (2004) report that the increase in trading volume that occurs in the international market immediately after the international cross listing is followed by a decline later on. Several other studies claim that the effects on trading volume and on fragmentation of the order flow vary across stocks, markets, and listing locations (Mittoo (1992); Domowitz, Glen and Madhavan (1998); Hargis (1997); Doidge (2001); Baruch, Karolyi and Lemmon (2003); Halling et al. (2004)). For example, Baruch et al. (2003) show that the distribution of trading volume is related with the correlation between the return of the cross listed stock and other international traded stocks. The same authors also refer that the effects also depend upon the country of origin of the cross listed stock, in particular, whether it originates from an emerging or developed market. Doidge (2004) shows that the impact in trading volume is a function of the changes in ownership that occur after the listing. Halling et al. (2004) show that the ratio between home and total trading is a function of the characteristics of the cross listed firm such as industry affiliation, foreign sales, etc.; and of the characteristics of the home and international exchanges such as differences in investor protection and in information disclosure.

One related question analysed in several studies is whether the cross listing of a particular stock has any impact on the development of the home stock market. Hargis (2000) analyses whether the international listing of Latin American stocks on US stock markets hinders the development of markets by deviating order flows from the original markets. He reports that in general there seem to be benefits associated with cross 3

listing, even if these vary accross markets. Yet other studies find cont radictory evidence and report a negative impact for the home market liquidity (Claessens, Kingebiel and Schmukler (2002); Moel (2001); Jaykumar (2002)). Karolyi (2004) finds a significant positive relation between the number of cross listings and a subsequent increase in the aggregate liquidity of the originating home market. This result seems to be driven by an increase in the liquidity of cross listed shares with no spillover effects for the other (non-cross listed). Thus the ratio of the turnover of non cross listed in total home turnover necessarily decreases. Fernandes (2005) analyses the impact of the first ADR on the liquidity of non cross- listed home stocks and finds a positive effect. The few other studies that address the impact of cross listing on the non cross listing home shares focus on the impact on returns (Melvin and Valero-Tonone (2004); Bradford, Martin and Whyte (2002)) and claim that the observed effects (positive or negative) are either information or competition-driven.

2.2. Cross listing: returns and liquidity

The large and extensive literature on cross listings consistently reports a positive impact on the valuation of cross listed stocks (Karolyi (1998); Miller (1999); Serra (1999)). This result reflects that there investors required lower returns as a consequence of lower risk. The reduction in risk could be driven, for example, by an increase of international market integration, greater investors’ protection and awareness, better accounting standards and risk sharing (Alexander et al. (1987, 1988); Miller (1999), Foerster and Karolyi (1999).

Some studies report that the return impact of the cross listing varies across markets, listing locations and firms (Karolyi, 1998). Some authors (Foerster and Karolyi (1999)) refer that this reduction of risk could also be associated with greater liquidity. The seminal study of Amihud and Mendelson (1986, 1988, 1989) indicates that investors require higher returns to hold stocks with lower liquidity to compensate them for the higher trans action costs and that lower trading costs induce greater participation and better risk sharing. The same authors show that average returns are associated with bid-ask spreads (used as a proxy measure of liquidity) even after controlling for other systematic risk measures. Datar, Naik and Radcliffe (1998) confirm that liquidity is a 4

priced risk factor using turnover as an alternative measure of liquidity. However, Rouwenhorst (1999), examining a sample of emerging markets’ firms, does not find a significant relation between return and turnover. As mentioned in the previous section cross listings are in most cases associated with increases in trading volume. Yet, there are very few studies that evaluate whether the cross listing impacts liquidity priced risks and whether the observed effect on returns is explained in fact by a reduction in liquidity risk. A study by Kadlec and McConnell (1994) analyses the impact on return of bid-ask spread for firms that upgrade from NAS DAQ to NYSE and finds that the firms that experience a reduction in bid-ask spreads also show an increase in prices, but the significance of the relation is sensible to the way the changes in spread are measured (absolute or relative terms). Baker, Khan and Edelman (1994) analyse the behaviour of the returns of dual listed stocks and suggest that part of the specific risk could reflect liquidity risk. Mittoo (2003) examines the particular case of Canadian cross listed stocks and finds that impact in returns is positively correlated with changes in liquidity after controlling firm, industry and market specific factors. More recent studies indicate that an individual stock liquidity risk measure such as the stock turnover or the bid ask spread is not the relevant liquidity risk measure because part of liquidity risk can be diversified away and the relevant risk measure has to be computed relative to aggregate market liquidity (Huberman and Halka (2001)). The results of Amihud (2002) suggest that the premium of stock risk includes a compensation for the weak market liquidity. Pástor and Stambaugh (2003) propose a four factor model that incorporates a systematic liquidity risk factor based on top of the three factors model proposed by Fama and French (1993). Acharya and Pedersen (2003) provide a theoretical model that accommodates empirical specification. The model predicts that both individual (total) and systematic liquidity are priced. Several studies have tested the role of systematic liquidity risk in explaining expected returns and found a significant premium for illiquidity (see, for example, Jun, Marathe and Shawky (2003), for the particular case of emerging markets). Thus, none of the existing studies that have tried to relate the liquidity effects of cross listed stocks with the observed price impact, has evaluated separately the role of changes in systematic and total liquidity risk measures in returns.

5

3. Essays

3.1 Essay I: Cross-listing: trading volume and fragmentation of the order flow

3.1.1 Testable hypotheses

We first analyse how international cross listings impact trading volume in global and for the home market, and examine the role of a series of previously identified factors in explaining the variation of order flow fragmentation across listings.

Managers suggest that the success of the decision to cross list should be measured by its global trading volume impact (Mittoo(1992)). Our first test is thus to measure if the global trading volume of the cross listed firm increases after the international listing. Second, we evaluate if the increase (decrease) of global trading volume results solely from effects through trades of shares now listed in the international market or/and whether it results also because of an increase in the transactions in the home market. 1

To test the effects on trading volume, we compare total trades before and after the international cross listing. We control the changes in trading in a particular market that are not associated with the cross listing event by standardizing the global volume (home and international trades) by the total trading volume of the home market.

To explain the cross sectional variation of the observed changes in trading, we use firm, and market – level factors.

Firm characteristics

The literature has identified a set of characteristics influencing the distribution of the trading volume of cross-listed shares.

1

One could think that ceteris paribus, we would only observe a shift in the trades of cross listed shares. For example, in the case of ADRs, the American investors would now trade their shares on an American Stock Exchange.

6

Mittoo (1992) and Halling et al. (2004) argue that the foreign sales of a firm increase the trades of cross- listed shares in the international markets. This stylised fact may reflect that foreign investors prefer, and trade more, assets of firms that have sales in the foreign country. This is related with Merton’s (1987) recognition hypothesis. Related with that, Baruch et al. (2003) suggest that investors prefer more stocks that are more correlated with foreign stocks in the cross- listed exchange. This is against the prediction of the diversification theory that would dictate that investors would be more willing to buy stocks that would result in better hedging. Previous empirical studies show in fact that investors hold and trade a subset of stocks belonging to the same category. Kang and Stulz (1997) for example, refer that foreign investors elect large firms. Yet, this may merely reflect that it is cheaper to trade these stocks or that these are stocks for which they do no t require an additional risk premium to compensate for imperfect information. The same authors refer that given information asymmetries, foreign investors prefer stocks less risky and with low debt. Finally, according to Halling et al. (2004) the greater the firm-specific information conveyed by the international market relative to the home market, the larger the trading volume in the international market.

Market Characteristics

Hargis and Ra manlal (1998) show that the fragmentation of order flows is more severe for the stocks of firms that originate from emerging markets. Faff, Hodgson and Saudagaran (2002) refer that differences in disclosure requirements between the international and home market also influence the fragmentation of order flow. Home markets that have higher disclosure requirements are more prone to retain a larger proportion of trades. Levine e Schumukler (2005) suggest that the differences of size and quality across markets produce different “gravitational pulls” on trading volume of cross listed stocks. Yet, traditional gravitational models that are used to analyse international trade flows predict that trading volume would be a direct function of both the size of the home and international markets. Halling et al. (2004) show as well that the fragmentation of the order flows may also be associated with the size of the home and international stock markets. Market size may be associated with stock market development. Yet, this measure could also be correlated with the factors mentioned 7

previously such as disclosure or whether a stock originates from emerging or mature markets.

Time zone differences also impact the fragmentation of order flow. The results of the study of Pulatkonak and Sofianos (1999) indicate that the time zone effect is an important country specific factor in determining the distribution of trading volume. The most relevant information regarding a firm is disseminated locally. If trading sessions overlap, informed investors may now trade in the international market. Further, overlapping trading hours enhance arbitrage trades.

Our second goal is to evaluate the impact of cross listing on the aggrega te home market liquidity and whether the impact varies across markets. In particular we test for significant differences between emerging and mature markets. Related to that, we evaluate whether there are any significant effects on the liquidity of the non-cross- listed firms in the home market.

3.1.2 Empirical model and tests

3.1.2.1 Impact in trading volume

To analyse the impact of the cross-listing on global and home trading, we compare the trading volume before and after cross listing. First, we use an event study procedure. We use daily data for 3, 6 and 12 month-windows. We evaluate changes in absolute measures of trading volume (number of stocks or turnover) and relative to the trading in aggregate market. Second we analyse whether these changes are persistent. To access that we compare the breakdown of local and international trading over time up to 3 years after the listing date.

3.1.2.2 Cross sectional analysis of fragmentation in order flow

Considering the findings of Levine and Schumukler (2005), Halling et al.(2004) and Pulatkonak and Sofianos (1999) we evaluate the distribution of the trading volume using an expanded and adapted gravitational model. Welded stock market development arguments imply that trading volume would be concentrated on the (larger) international 8

markets. Traditional gravitational models predict that both the size of home and international markets positively impact trading in both locations and negatively impact the distance (measured by the time zone).

We estimate the following model 1 year and 3 years after the cross-listing:

[( )(

)]

ln (OFFi ) = β 0 + β1 ln MC I MC H + β 2TZ i + β 3PFSi + β 4CORi + β 5SIZEi + β 6SDi + β7 LEVi + β 8MI i + β9M i + β10DIS i +

n −1

∑

(1)

β10+ j I j ,i + εi

j =1

where: OFFi - is the proportion of international trading in global trading of firm i. OFFi is given by

TVi I H TVi + TVi I

, where, TV iI is the trading volume of firm i in the international

market and TV iH is the trading volume of firm i in the home market; MC I – represents the size of the international stock market measured by the aggregate market capitalization; MC H – represents the size of the home stock market measured by the aggregate market capitalization; TV H ; TV I – is the aggregate trading volume of the home market and international market, respectively. PFSi – is the percentage of foreign sales to total sales of firm i as of year-end just before listing date; CORi – is the correlation between the returns of the cross listed stock i and the international market; SIZEi – is given by the ratio of the absolute market capitalization of the cross listed firm i and relative to market capitalization of the home market; SDi – is the standard deviation of the returns of the cross listed firm i for the year previous to listing; LEVi – is given by the ratio between long term debt and market capitalization, of firm i, as of year-end just before listing date; Ij – is a dummy variable for industry affiliation of firm i;

9

MIi - measures the sensitivity of the cross listed stock i, the information of the international market relatively to the information of the home market. We use a proxy suggested by Baruch, Karolyi and Lemmon (2005). (See below) Mi – is a dummy variable, that takes the value of 1 if the cross- listed firm i originates from an emerging market and 0 otherwise; DISi - index of the disclosure level of the home country of cross listed firm i; TZi – is a dummy variable that is equal to 1 if there is a common trading time between the home market of cross- listed firm i and the international market, and 0 otherwise; εi

- is the i.i.d error.

To proxy the degree of the information that is conveyed by the international market for a particular firm i, Baruch et al. (2005) propose using a statistic F that compares the explanatory power of two models. The information content is given by the increase in R2 that results from using a model that includes on top of the returns of the home market index, the returns of the international market index. These models are estimated using home return data lagged one year.

To evaluate the impact of the cross listing on the development of the home market, we use the liquidity measure proposed by Bekaert, Campbell, Harvey and Lundblad (2005). This measure replaces the return observations that are equal to zero by a transformed measure that aims capturing at true price impact. We use daily data of returns.

Finally, to test the effect of the international cross listing on the liquidity of the remaining firms listed solely in the home market, we evaluate the changes in daily trading volume of these firms before and after the cross listings. We use the same tests proposed above to evaluate changes in trading for the cross listed stocks.

Table 1 below summarize the expected results for each of the parameters of interest in equation (1) above.

Table 1. Fragmentation of the trading volume for cross listed stocks (equation (1)): expected results 10

Independent variables  ln  MC I  MC H   TZi

Expected sign in the Ln(OFFi)

 

PFSi CORi SIZEi SDi LEVi MIi Mi DISi

+/- ? + + + + + + -

3.2 Essay II: Cross listing: returns and liquidity

3.2.1. Testable hypotheses

The first essay looks upon the magnitude and significance of the impact of cross listings in trading volume and sheds light on the determinants of different changes across stocks. The second essay measures cross listing valuation effects and examines whether these are risk or non-risk related, and if the observed impact in volume changes brought by cross- listing can account for those effects. We first evaluate the effects of cross listings on the returns of cross listed stocks. We then proceed and test if the effects observed in returns and risk of cross listed stocks are associated with liquidity changes. We explicitly evaluate whether the impact results from changes in total or systematic liquidity.

3.2.2 Empirical models and tests

3.2.2.1 Impact of the international cross listing on return

To test the first hypothesis we follow the methodology used in previous studies. We measure abnormal returns around the listing date. Abnormal returns are measured by the differences between observed and expected returns. Expected returns are given by market model, for which parameters are estimated with weekly data for the period two

11

years before the listing date. Observed returns are total returns (adjusted for dividends and changes in capital). We test the null hypothesis that average abnormal returns are statistically different from zero at the event date. We also compute cumulative abnormal returns for several event windows around the event date. These windows are defined in order to accommodate market reaction to the cross listing announcement and/or delayed reaction2 . We compute daily abnormal returns up to one year.

To test if there are structural changes in the risk measures of the stock associated with home market and international market the cross listing, we first use the following model suggested by Foerster and Karolyi (1999):

PRE eH eI POST eH POST Rite = αiPRE + βiH Rmt + β iIPRE Rmt + α iLIST DitLIST + αiPOST DitPOST + β iH Rmt Dit +

( 2)

eI POST β iIPOSTRmt Dit + ε it

where: Rite

- is the weekly excess return of firm i, at moment t;

eH Rmt

,

ReI mt

- are the excess return for home (H) and international markets (I),

respectively, at moment t; DitLIST -

is a dummy variable, that equals one for the listing date and zero otherwise;

DitPOST -

is a dummy variable, that equals one for the period after listing date and zero

otherwise; αi

- are drift parameters that measure average abnormal returns associated respectively

with the period before 3 and after listing date as well as the listing date itself; PRE, LIST and POST - denote the periods before, at, and after the listing date, respectively; ε it

- is the i.i.d error.

2

Several studies focus announcement dates to evaluate market reaction to cross listing and find a significant market reaction. We use listing dates because we cover a larger number of listing for most of which there is no reliable announcement information. 3

Please note that αiPRE captures the abnormal return before listing and could be associated with

announcement effects.

12

We estimate this model regression based on pooled weekly returns for a period for two years around listing date (one year before and one year after listing date).

We then proceed to evaluate our centre hypothesis that is if the changes observed in the returns generating process are associated to liquidity changes. We modify the model developed by Foerster and Karolyi (1999) by including two additional variables:

PRE eH eI PRE POST eH POST Rite = α iPRE + βiH Rmt + β iIPRE Rmt + β iL LIQit + αiLIST DitLIST + αiPOST DitPOST + βiH Rmt Dit

(3)

eI POST + βiIPOST Rmt Dit + βiLPOST LIQit DitPOST + ε it

where: LIQit

- is a measure of liquidity of stock i, at moment t;

PRE β POST βiL , iL

- measure whether the returns are associated to total liquidity in periods

before and after the cross listing, respectively; The remaining variables are defined as above. We use several measures of liquidity.

Several authors (see Miller (1999) for example) suggest that the impact of cross listing is a function of the degree of the market segmentation between home market and international market. To test if the type of market (developed market versus emerging market) influences return of the cross listed stocks, we estimate the previous models using two sub-samples.

To further address the role of liquidity in the return generating process and the impact of cross listing relevant risk measures, we extend the model of Pástor and Stambaugh (2003). This model is different from Fama and French (1993) in that it includes a liquidity systematic factor on top of the three traditional risk factors. As before, we run a pooled regression of weekly returns for a period of two years around the listing date.

Model of Pástor and Stambaugh (2003):

13

e +β Rite = αi + βim Rmt iSMB SMBt + βiHMLHMLt + βiIMV IMVt + ε it

( 4)

Model to estimate (adapted):

PRE e PRE PRE PRE Rite = α iPRE + β im Rmt + β iSMB SMBt + β iHML HML t + β iIML IML t + α iLIST DitLIST

(5)

POST e POST POST + α iPOST DitPOST + β im Rmt D itPOST + β iSMB SMBt DitPOST + β iHML HML t D itPOST POST + β iIMV IMV t D itPOST + ε it

where: Rite e Rmt

- represents the excess return of the stock i, at moment t; - represents the excess return of the market, at moment t;

SMB, HML, IML – stand for Small Minus Big, High Minus Low and Iliquid Minus Liquid differential portfolio returns. DitLIST

DitPOST

- is a dummy variable, that equals one for the listing date and zero otherwise; - is a dummy variable, that equals one for the period after listing date and zero

otherwise; PRE, LIST and POST - denote the periods before, at, and after the listing date, respectively; ε it

- is the i.i.d error.

(Work in progress)

14

4. References Acharya, V.V. and Pedersen, L.H. (2003). Asset pricing with liquidity risk. Working Paper London Business School. Alexander, G.J., Eun, C.S. and Janakiramanan (1987). Asset pricing and dual listing on foreign capital markets:a note. The Journal of Finance, 42, 151-158. Alexander, G.J., Eun, C.S. and Janakiramanan, S. (1988). International listings and stock returns: some empirical evidence. Journal of Financial and Quantitative Analysis, 23, 135-151. Amihud, Y. (2002). Illiquidity and stock returns: cross-section and time-series effects. Journal of Financial Markets, 5, 31-56. Amihud, Y. and Mendelson, H. (1986). Asset pricing and the bid-ask spread. Journal of Financial Economics, 17, 223-249. Amihud, Y. and Mendelson, H. (1988). Liquidity and asset prices: financial management implications. Financial Management, Spring, 5-15. Amihud, Y. and Mendelson, H. (1989). The effects of beta, bid-ask spread, residual risk, and size on stock returns. The Journal of Finance, 44, 479-486. Baker, H.K., Khan, W.A. and Edelman, R.B. (1994). The post-dual listing anomaly. Journal of Economics and Business, 46, 287-297. Bancel, F. and Mittoo, U.R. (2001). European Managerial perceptions of the net benefits of foreign stock listings. European Financial Management, 7, 213-236. Baruch, S., Karolyi, A. and Lemmon, M.L. (2003). Multi-Market trading and liquidity: theory and evidence. EFA 2004 Maastricht Meetings paper, 4642. Baruch, S., Karolyi, G.A. and Lemmon, M.L. (2005). Multi market trading and liquidity. Working paper, University of Utah. Bekaert, G., Harvey, C. and Lundblad, C. (2005). Liquidity and expected returns: lessons from emerging markets. Working paper . Bradford, B.M., Martin, A.D. and Whyte, A.M. (2002). Competitive and information effects of cross-border stock listings. The Journal of Financial Research, 25, 399-413. Claessens, S., Klingebiel, D. and Schmukler, S. (2002). Explaining the migration of stocks from exchanges in emerging economies to international centres. Working paper, N.º 2002/94, UNU World Institute for Development Economies Research. Datar, V.T., Naik, N.Y. and Radcliffe, R. (1998). Liquidity and stock returns: an alternative test. Journal of Financial Markets, 1, 203-219.

15

Doidge, C. (2001). Do changes in law matter for corporate ownership and control? Evidence for emerging markets firms that lists in the U.S.. Mimeo Ohio State University. Doidge, C. (2004). U.S. cross- listings and the private benefits of control: evidence from dual-class firms. Journal of Financial Economics, 72, 519-553. Domowitz, I., Glen, J. and Madhavan, A. (1998). International cross- listing and order flow migration: evidence from an emerging market. The Journal of Finance, 53, 20012027. Faff, R.W., Hodgson, A. and Saudagaran, S. (2002). International cross listings towards more liquid markets: the impact on domestic firms. Journal of Multinational Financial Management, 12, 365-390. Fama, E.F. and French, K.R. (1993). Common risk factors in the returns on stocks and bonds. Journal of Financial Economics, 33, 3-56. Fernandes, N. (2005). Market liberalization at the firm level: spillovers from ADRs and implications for local markets. EFA, 2003, Annual conference paper, N.º83; EFMA, 2003, Helsinki meetings . Foerster, S. and Karolyi, G.A. (1993). International listings of stocks: the case of Canada and the U.S.. Journal of International Business Studies, 24, 763-784. Foerster, S. and Karolyi, G.A. (1998). Multimarket trading and liquidity: a transaction data analysis of Canada - US interlistings. Journal of International Financial Markets, Institutions and Money, 8, 393-412. Foerster, S.R. and Karolyi, G.A. (1999). The effects of market segmentation and investor recognition on assets prices: evidence from foreign stocks listing in the United States. The Journal of Finance, 54, 981-1013. Halling, M., Pagano, M., Randl, O. and Zechner Josef (2004). Where is the market? Evidence from cross- listings. CEPR Discussion Paper No. 4987, Available at SSRN: http://ssrn.com/abstract=771993. Hargis, K. (1997). When does multimarket trading improve the quality of the primary market? Evidence from international. Goldman, Sachs and Co. Working Paper. Hargis, K. (2000). International cross- listing and stock market development in emerging economies. International Review of Economics and Finance, 9, 101-122. Hargis, K. and Ramanlal, P. (1998). When does internationalization enhance the development of domestic stock markets? Journal of Financial Intermediation, 7, 263292. Huberman, G. and Halka, D. (2001). Systematic liquidity. The Journal of Financial Research, 24, 161-178.

16

Jayakumar, V. (2002). Impact of international cross- listing on local exchanges: evidence from Chile. International Review of Finance, 3:3/4, 189-211. Jun, S.-G., Marathe, A. and Shawky, H.A. (2003). Liquidity and stock returns in emerging equity markets. Emerging Markets Review, 4, 1-24. Kadlec, G.B. and McConnell, J.J. (1994). The effect of market segmentation and illiquidity on asset prices: evidence from exchange listings. The Journal of Finance, 49, 611-636. Kang, J.-K. and Stulz, R.M. (1997). Why is there a home bias? An analysis of foreign portfolio equity ownership in Japan. Journal of Financial Economics, 46, 3-28. Karolyi, G.A. (1998). Why do companies list shares abroad?: A survey of the evidence and its managerial implications. Institutions&Intrument, 7, New York, University Salomon Center . Karolyi, G.A. (2004). The role of ADRs in the development of emerging equity markets. Review of Economics and Statistics, 86, 670-690. Korczak, P. and Bohl, M. (2003). Return performance and liquidity of cross- listed central European stocks. Working paper available at SSRN: www.ssrn.com.

Levine, R. and Schmukler, S. (2003). Migration, spillovers, and trade diversion: the impact of internationalization on stock market liquidity. Policy Research Working paper series 3046, The World Bank. Levine, R. and Schmukler, S. (2005). Internationalization and stock market liquidity. NBER working paper 11894, December. Melvin, M. and Valero-Tonone, M. (2004). The dark side of international cross- listing: effects on rival firms at home. Working Paper, Carey Scholl of Business, Arizona State University, May. Miller, D.P. (1999). The market reaction to international cross- listings: evidence from depositary receipts. Journal of Financial Economics, 51, 103-123. Mittoo, U.R. (1992). Managerial perceptions of the net benefits of foreign listing: Canadian evidence. Journal of International Financial Management and Accounting, 4:1, 40-62. Mittoo, U.R. (2003). Globalization and the value of US listing: Revisiting Canadian evidence. Journal of Banking and Finance, 27, 1629-1661. Moel, A. (2001). The role of american depositary receipts in the development of emerging markets. Economia, 2 (1), 209-273. Pástor, L. and Stambaugh, R.F. (2003). Liquidity risk and expected stock returns. Journal of Political Economy, 111, 642-685.

17

Pulatkonak, M. and Sofianos, G. (1999). The distribution of global trading in NYSElisted non-U.S. stocks. NYSE Working paper 99-03. Rouwenhorst, K.G. (1999). Local return factors and turnover in emerging stock markets. The Journal of Finance, 54, 1439-1464. Rouwenhorst, K.G. (1999). Local return factors and turnover in emerging stock markets. The Journal of Finance, 54, 1439-1464. Serra, A.P. (1999). Dual- listings on international exchanges: the case of emerging markets' stocks. European Financial Management, 5, 165-202.

18