Stock exchange consolidation and cross-border investment:

an empirical assessment

Abstract
This paper investigates the e¤ects of stock exchange consolidation on foreign portfolio hold-
ings. Stock exchanges consolidation boosts liquidity and market size of member countries, as
expected. Since member countries’domestic holdings decline, the size e¤ect is due to a sharp
increase in foreign investment. Cross-border foreign investments among member countries in-
crease far beyond what explained by larger and deeper markets. The consolidation e¤ect is
particularly pronounced among member countries that were less foreign-oriented before the
fusion, that are smaller in size, and that feature strong familiarity.

Keywords: Stock exchange consolidation, International Portfolio Investments, familiarity

and gravity factors, …nancial regulation.
JEL Classi…cations: G11, G15, G30

1
1 Motivation and relevance

Since the beginning of the century, increased competition between stock exchanges triggered im-
portant structural changes in the securities market industry that underwent a process of gradual
consolidation. Increased cross-border capital ‡ows, on the one hand, and the evolution of infor-
mation and communication technology, on the other hand, have resulted in the transformation of
trading systems. This process gradually lead to more harmonized listing and corporate governance
standards, to more intense cross-listing activity, and, …nally, to consolidation of trading systems
and exchanges. This transformation led to an integration process of stock exchanges, …rst, within
countries, and, then, across countries and continents.
Schmiedel and Schönenberger (2006) describe the consolidation process in the securities exchange
industry of the Euro area. First, they observe the integration at the national level and between cash
and derivative markets, and then the consolidation process with the creation of the …rst pan-European
exchange, that is, Euronext.
Nielsson (2009) empirically investigates the e¤ect of the Euronext stock exchange merger on listed
…rms. The results show that gains from the stock exchange merger are concentrated among big …rms
and …rms foreign-oriented.
Hasan and Schmiedel (2004) investigate whether the adoption of network strategies by European
stock exchanges over the period 1996-2000 created additional value in the provision of trading ser-
vices. They …nd that a network strategy is associated with higher market capitalization and lower
transaction costs.
More recently, Hasan et al. (2012) investigate the e¤ect of di¤erent degrees of stock exchange
integration on exchange shareholders’value creation over the period 2000-2008. Their …ndings em-
phasize that mergers and acquisitions create more value than alliances and, among alliances, joint
ventures generate more value than nonequity alliances. Cross-border integration creates more value
than domestic, and horizontal more value than vertical. More value accruing to exchange sharehold-
ers is also associated with better shareholder protection, accounting standards, and capital market
development of the partner exchanges.

2
 Dorodnykh (2014) systematically analyzes the process of stock exchange integration over the pe-
riod 1995-2010, and investigates the determinants of stock exchange integration. She highlights that
…nancial regulation, cross-membership agreements, and openness are important drivers of mergers
and acquisitions in the stock exchange industry, while the size of the stock exchanges negatively
a¤ect the likelihood of successful mergers.
The existing literature has investigated the impact of cross-listing, foreign listing, and stock ex-
change consolidation from the perspective of …rms or exchange shareholders. The literature generally
converges on the bene…ts of stock exchange consolidation accruing to exchange shareholders and list-
ing …rms: mergers are the natural response to tough competition, and allows to substantially reduce
the costs of trading, to increase liquidity, and then to compete on a world scale. However, regulation
market authorities are also concerned that the consolidation process, by reducing the competition
through stock exchange mergers, does not result in abuse of exchange market power, so that also
investors bene…t from a reduction in transaction costs.
Our contribution departs from existing literature because we focus on the e¤ect of stock exchange
mergers on international portfolio allocation and therefore takes the stance of intermediaries and
retail investors.
This is the …rst paper, to the best of our knowledge, which analyzes the impact of cross-country
stock exchange consolidation on international stock holdings. As a unique exception, the work
by ? makes a …rst attempt to estimate the role of the Euronext creation on the international
portfolio holdings of four European investing countries (Italy, France, Spain, Sweden), in the period
2001-2004. She …nds a signi…cant role of Euronext only for individual investors, while institutional
investors and the aggregate economy appear to be non sensitive to the stock exchange consolidation
process. The limited country sample and time span, however, restrict the analysis to four European
Union countries, to the Euronext consolidation event, and to the period 2001-2004, not permitting
a rigorous and thoroughgoing investigation of the stock exchange consolidation issue.
In the time span covered by our analysis (2001-2012), several stock exchange consolidations
occurred.

3
 The …rst case of stock exchange fusion is the creation of the Euronext stock exchange, in Septem-
ber 2000, from the merger of Amsterdam, Brussels and Paris stock exchange. In September 2002,
the Euronext platform further enlarged to include the Lisbon exchange which in November 2003
successfully migrated to the common trading and clearing systems.1 In April 2007, the New York
Stock Exchange and Euronext merged to create a new transatlantic project, the NYSE Euronext
group.
In 2004, the OMX joint company was created, from the merger of the OM Group, controlling the
OM Stockholm Stock Exchange, the Helsinki exchange and the Tallin exchange. In January 2005
also the Copenhagen stock exchange joined the group, and in October 2006 the company also took
a 10 per cent stake in Oslo Holding ASA, the owner of Oslo Stock Exchange.2 In March 2008, the
Nasdaq acquisition of OMX gave birth to the NASDAQ OMX Group.
In March 2007, after a bid of the London Stock Exchange to the shareholders of the Italian Stock
Exchange, the fusion between British and Italian stock markets occurred (LSE-BI).
In September 2009 the Central and Eastern Europe Stock Exchange Group (CEESEG) was
established, comprising the Budapest, Ljubljana, Prague, and Vienna stock exchanges.
In December 2009, Peru, Colombia and Chile stock exchanges merged to create the single trading
platform Mercado Integrado Latinoamericano (MILA), the largest Latin America market in terms of
listed companies, and the second biggest stock market in terms of capitalization after the Brazilian
stock exchange.
The consolidation trend is far from concluded. Since March 2016, after the Intercontinental
Exchange -the New York Stock Exchange owner- receded from making a bid for the British London
Stock Exchange Group, the stock exchange industry has been thrilled by the perspective of an
unprecedented merger between the London Stock Exchange and the Deutsche Boerse. After the UK
vote for a Brexit, the perspective of a consolidation could appear controversial. It is the third time
1
In 2003 Euronext also acquired the London-based derivatives market LIFFE. In our paper, the standard econo-
metric speci…cation does not consider this membership, but in Table 4, column (1), we check the sensitivity of our
…ndings to a speci…cation of the NYSE-Euronext platfom including also the United Kingdom.
2
In our paper, the standard econometric speci…cations does not consider the Norwegian stock market in the OMX
group. In Table 4, column (2), however, we check the sensitivity of our …ndings to the speci…cation including also the
Oslo stock exchange.

4
since the year 2000 that Deutsche Boerse has tried to merge with the LSE, but the previous two
attempts have been rejected.3 Paradoxically, the Brexit vote could act as a fuel and further reinforce
the fusion: the merger could be the only guarantee for London to be able to trade euro-denominated
securities.4
Our …ndings can help estimate the e¤ect of such a future strategic consolidation on cross-border
investment, given the peculiarities of the stock exchanges in terms of size, geographical and cultural
distance, and openness.
The remainder of the paper is structured as follows. In Section 2, we sketch the estimable
equation. In Section 3 we describe the data. Section 4 empirically analyzes the e¤ect of stock
exchange consolidation. Section 5 concludes.

2 Estimable equation

Our paper aims to assess the role of …nancial education on foreign equity portfolios. Domestic
positions and home bias are therefore not investigated.5
The background theoretical framework rests on the return-reducing approach of Cooper and
Kaplanis (1994) and Chan et al. (2005): in equilibrium, investors are supposed to face di¤erent
information costs when investing in various …nancial markets and what matters is the investment
barrier relative to the average.
Absent any investor-speci…c factor, the "unbiased" portfolio holding of an asset depends, as in
standard portfolio choice theory, on asset characteristics (risk and return). When considering equi-
librium asset holdings without investment barriers, all investors ought to hold the same portfolio,
i.e., the value-weighted portfolio, in which each asset is weighted according to its share in world stock
3
In the last two decades, the stock exchange industry has witnessed a great deal of integration attempts that have
been blocked or failed. For a comprehensive discussion of all integration attempts in the stock exchange industry, see
Dorodnykh (2014).
4
The European Central Bank required clearing houses that handle euro-denominated securities to only be located
in the Eurozone. London disputed this in the EU court, arguing that this principle contradicts the EU’s free market,
and won. However, now the UK chose to leave the EU: the merger could be the only guarantee that euro-denominated
trading can be managed by London.
5
Domestic shares only indirectly impact our analysis, since the weight of each foreign stock index in the overall
portfolio indeed depends on the domestic share.

5
market capitalization. The same portfolio is still universally optimal in equilibrium even in the pres-
ence of investment barriers, provided that these barriers identically a¤ect all investors. Conversely,
heterogeneity in bilateral-speci…c investment barriers generates a wedge between the investor-speci…c
optimal portfolio and the value-weighted portfolio. This wedge depends, in particular, on the dis-
tance between the investment barrier of country l investing in country j and the average barrier
calculated over all countries investing in asset j.
The optimal portfolio weight in asset j by country l is wlj , while M Sj is the market share of asset
j in the world market capitalization.
wlj
We label the ratio M Sj
as "foreign bias" in asset j of a representative investor in country l. A
portfolio share wlj larger than j’s market share signals that asset j is over-weighted in country l’s
portfolio, while a ratio lower than 1 signals that country j is under-weighted.
To estimate the e¤ect of stock exchange consolidation on foreign portfolio we run standard linear
estimation techniques, as follows:

wlj P n P k P h
= + Xljn + Wlk + Zjh + "lj (1)
M Sj n=1;::;N k=1;::;K h=1;::;H

Our regression speci…cation accounts for pair-speci…c and country-speci…c factors which poten-
tially capture investment frictions.
Among pair-speci…c variables, we include n covariates, denoted by Xlj ; which are expected to
capture investment barriers. If we consider, for instance, our main variable of interest, i.e., stock
exchange consolidation (EClj ); we conjecture that sharing a common exchange induces higher invest-
ment of country l in country j, and therefore we expect a positive sign for the associated coe¢ cient.
To estimate the above parameters, we adopt a feasible Generalized Least Squares speci…cation
correcting for both heteroskedasticity and general correlation of observations across destination-
countries, with standard errors adjusted for two-way clustering at the investing-country and year
levels, as suggested for …nance panel data sets (Petersen (2009)).

6
3 Data

We consider equity portfolio investments by 40 investing countries6 in 41 destination stock markets7 ,

for the period 2001–2012. We adopt the Coordinated Portfolio Investment Survey (CPIS), released
by the IMF, a dataset which has been exploited in many recent papers (Fidora et al. (2007); Lane
and Milesi-Ferretti (2007); Sorensen et al. (2007); Giannetti and Koskinen (2010); Giofré (2013)).
This survey collects security-level data from the major custodians and large end-investors. Portfolio
investment is broken down by instrument (equity or debt) and residence of issuer, the latter pro-
viding information on the destination of portfolio investment. While the CPIS provides the most
comprehensive survey of international portfolio investment holdings, it is still subject to a num-
ber of important caveats. The most important is that the CPIS is unable to address the issue of
third-country holdings and round-tripping, very frequent in the case of …nancial o¤shore centers.8
Moreover, the survey does not report domestic positions which need to be retrieved from other
sources.
Details on dependent variable, regressors, and their respective source are reported in Appendix
A.

4 Empirical analysis

4.1 Preliminary …ndings: size, liquidity and domestic share

Table 1 reports some preliminary correlations between our main variable of interest, that is, stock
exchange consolidation (EC), and some basic indicators of national stock exchanges.
6
Argentina, Australia, Austria, Belgium, Brazil, Canada, Chile, Czeck Republic, Denmark, Estonia, Finland,
France, Germany, Greece, Hong Kong, Hungary, India, Indonesia, Israel, Italy, Japan, Malaysia,Mexico, the Nether-
lands, New Zealand, Norway, Poland, Portugal, Russia, Singapore, South Africa, South Korea, Spain, Sweden, Switzer-
land, Thailand, Turkey, the United Kingdom, the United States, and Venezuela.
7
Argentina, Australia, Austria, Belgium, Brazil, Canada, Chile, Denmark, Egypt, Finland, France, Germany,
Greece, Hong Kong, India, Indonesia, Ireland, Israel, Italy, Japan, Malaysia, Mexico, the Netherlands, New Zealand,
Norway, Pakistan, Peru, Philippines, Portugal, Singapore, South Africa, South Korea, Spain, Sweden, Switzerland,
Taiwan, Thailand, Turkey, the United Kingdom, the United States, and Venezuela. Note that there are some countries
included as investing ones but not as destination ones, and vice versa, because of relevant variables’data availability.
8
See www.imf.org/external/np/sta/pi/datarsl.htm for more details on the survey.

7
 At the head of the rows, we have three labels for di¤erent speci…cations of the EC indicator.
The …rst one –ECj - refers to destination country j, and is related to the status of country j as
a member or not of a consolidated stock exchange. The second one – ECl - refers to the investing
country l, and is related to the status of country l as a member or not of a consolidated stock
exchange. The third one - EClj - refers to the pair country lj, and is related to the status of the
pair lj as members of the same stock exchange platform.
Columns (1a) to (1b) of Table 1 report the correlation coe¢ cient between ECj and the share of
stock market capitalization of country j (M Sj ). As expected, the correlation coe¢ cient is positive
and statistically signi…cant (0.026, column (1a)), even when controlling for time …xed e¤ects (0.027,
column (1b)). In columns (2a) to (2b) of Table 2, we report the correlation coe¢ cient of ECj with
an alternative measure of market share, M S_M SCIj , based on the free-‡oat adjusted market capi-
talization.9 The coe¢ cient is still positive, statistically signi…cant, and larger in size (0.039, column
(2a)), also when accounting for time …xed e¤ects (0.041, column (2b)). Our …ndings highlight that
countries joining a common stock exchange platform enjoy an increase in stock market capitalization
up to 4%, compared to other countries.
The positive correlation between the market share and the index of stock market consolidation
can be rationalized by the higher market liquidity of national stock markets that merged in an
integrated trading platform. In columns (3a) to (4b) of Table 1, we compute the correlation between
the index ECj and two measures of liquidity of the destination country j, respectively, the relative (to
the world average) ratio of stock trades to GDP and the relative turnover ratio. As expected, both
measures, even when accounting for time …xed e¤ects (#b), display a statistically signi…cant and
large correlation (from 0.20 to 0.24) with the ECj index. Stocks traded in consolidated exchanges
show a higher liquidity than other countries: a 24% larger turnover ratio, and a 20% increase of the
ratio traded assets to GDP.
On the demand side, we are interested in understanding which investors actually hold the larger
supply of stocks associated with the fusion process. In particular, we check if the larger size is paired
9
We follow the same measure adopted by Dahlquist et al. (2003) which corrects for the percent of shares that are
not available for trading and cannot therefore be held by foreign investors.

8
with a prevalent increase in the foreign or in the domestic component. In columns (5a) to (5c)
of Table 1, we observe the behavior of investors residing in countries involved in the consolidation
process.
In column (5a) of Table 1, we report the correlation of the domestic portfolio share wll with the
index of consolidation of the investing country ECl .
The empirical evidence of overinvestment in domestic assets, often referred to as "home equity
bias", is well known in the international …nance literature, as documented by many authors (French
and Poterba (1991); Tesar and Werner (1995), among others). Interestingly, those investors resid-
ing in countries whose stock exchanges undergo a consolidation process, show a signi…cantly lower
domestic portfolio position. To net out the e¤ects due to a general decrease in “home bias” over
time, or by some peculiarities of the investing countries, in columns (5b) and (5c), we report the
correlation coe¢ cient after controlling, respectively, for time and investing-country …xed e¤ects. We
…nd that those countries involved in a process of stock exchange consolidation witness a reduction
of 11.9% in the domestic portfolio share.
Since countries belonging to a consolidated platform show larger market shares but lower domestic
holdings, then a sharp increase in inward foreign investment must have occurred.
In Table 2, we investigate hence the correlation between stock exchange consolidation and foreign
investment.
In columns (1) and (2) of Table 2, we report the correlation of the foreign share wlj with the
ECj of the destination country, and with the ECl of the investing country, respectively. The two
correlations are positive, statistically signi…cant, though economically quite small (0.003). In column
(3) of Table 2, we consider instead the bilateral index of exchange consolidation EClj , equal to 1
if investing and destination country share the same exchange platform, and to 0 otherwise. The
correlation coe¢ cient of the bilateral foreign investment wlj with the bilateral speci…c consolidation
index EClj is …ve times larger than the country-speci…c correlations of columns (1) and (2): countries
sharing a common stock exchange platform invest one in another a 1.6% larger foreign portfolio share.
When considering the three measures together (column (4) of Table 2), the above results are

9
con…rmed, also when controlling for time and investing-country speci…c …xed e¤ects (column (4a)).
In order to net out the supply side e¤ect and focus on the demand side and its composition,
we consider in columns (5) to (8a) of Table 2, the “foreign bias” measure, that is, the ratio of the
foreign portfolio share wlj to country j market share (M S_M SCIj ). This normalization allows us
to control for the size of the recipient stock market, which in turn may depend on the consolidation
process. When we correlate this measure with the ECj index, the correlation is no longer signi…cant
as it is absorbed by the larger market share (column (5)). The investing country speci…c factor ECl
remains signi…cantly associated with the foreign bias measure and becomes larger (0.192, column
(6)), and the e¤ect of the bilateral speci…c factor EClj is boosted to 1.038 (column (7)).
When considering all of the three measures together, in column (8) of Table 2, the correlation
of the foreign bias with the investing speci…c consolidation factor ECl is almost unchanged, while
the coe¢ cient of the destination-speci…c factor ECj becomes negative: this result is mainly due to
the positive e¤ect of the consolidation process on the destination-country’s market size that is the
denominator of the dependent variable, so that the latter is negatively a¤ected. The coe¢ cient
of the bilateral speci…c consolidation factor EClj only moderately shrinks to 0.899, remaining thus
economically very large. When both time and investing country …xed e¤ects are accounted for in
column (8a), the two country-speci…c factors lose statistical signi…cance and only the role of the
bilateral consolidation factor persists large and signi…cant (0.891).

4.2 Stock exchange consolidation and foreign equity holdings: the main

e¤ect.

This preliminary evidence suggests that belonging to the same trading platform justi…es the wedge
between the actual foreign portfolio share and what is predicted by the value weighted portfolio.
However, this …nding might miscapture other bilateral linkages among the countries under in-
vestigation. We need to seize the bilateral exchange consolidation e¤ect, on top of other competing
explanatory factors.
In Table 3, we report the results of a multivariate regression analysis which considers the foreign

10
bias as a dependent variable.
The literature has widely documented that the cultural and geographic proximity of the market
has an important in‡uence on investor stock holdings and trading (Brennan and Cao (1997); Kang
and Stulz (1997); Grinblatt and Keloharju (2001); Chan et al. (2005); Portes and Rey (2005)).
Column (1) of Table 3 reports the results from a regression including standard gravity variables such
as distance, common border, common language. The variable distance is measured as the great-circle
distance between the capital cities of the destination and investing countries. The common border
(language) dummy takes the value 1 if the investing and destination country share a common border
(language) and 0 otherwise. The …rst two variables, distance and common border, simply capture
the physical distance between investing and destination country. The role of the common language
dummy is intuitively interpretable, since foreign languages make collecting information more di¢ cult.
These variables play an economically and statistically signi…cant role in explaining the dependent
variable, with a particularly strong impact of the common border dummy (0.713). The coe¢ cient
of the bilateral consolidation index EClj , when standard gravity variables are accounted for, is still
large and statistically signi…cant (0.798).
In column (2) of Table 3, we include also covariates that captures cultural and institutional
linkages.
To capture cultural and/or historical ties, we control whether countries are tied by colonial
heritage. The dummy common colony variable takes the value 1 if the considered pair of countries
shares a similar colonial history. As expected, the coe¢ cient is positive and statistically signi…cant
(0.188).
We also include the common currency area (EMU), and the common legal origin index. The EMU
dummy takes the value 1 if the investing and destination countries are EMU members and 0 otherwise.
The coe¢ cient is positive and signi…cant and its e¤ect is quite large: EMU membership boosts foreign
portfolio share by 0.523 compared to non member countries. Our …ndings are consistent with the
evidence reported by Lane and Milesi-Ferretti (2007) and Balta and Delgado (2009), who …nd, as a
result of monetary integration, a notable increase in foreign investments in the Euro area by EMU

11
countries.
Finally, sharing the same legal origin might encourage cross-border investment since there is less
fear of unknown factors (Lane (2006); Guiso et al. (2009)). We include a dummy variable taking
the value 1 if the investing and destination countries share the same legal family (English, French,
German or Scandinavian) and 0 otherwise. The coe¢ cient is surprisingly negative and statistically
signi…cant in this speci…cation, though it is not the case in next richer speci…cations. Overall, the
inclusion of these pair-speci…c factors marginally reduces the impact of stock exchange consolidation,
that remains still large and signi…cant (0.646).
Institutional barriers to capital mobility can deter investment in foreign countries. In column
(3) of Table 4, we control for inward and outward capital mobility, proxied by an index measuring
the restrictions imposed by di¤erent countries on capital ‡ows, derived from the Economic Freedom
Network (Chan et al. (2005)). This index ranges from zero to 10 and measures the restrictions
countries impose on capital ‡ows, assigning a lower rating to countries with more restrictions on
foreign capital transactions. We …nd indeed that higher capital mobility of the destination country
attracts more inward investment (0.044), and higher capital mobility in the investing country pushes
foreign investment (0.130). The coe¢ cient of EClj is only marginally reduced to 0.612.
Among regulatory barriers to information acquisition by foreign investors, Barth et al. (1999)
highlight the importance of the costs faced by foreign investors in understanding other countries’
accounting principles. Bae et al. (2008) propose a measure of country-pair di¤erences in 21 accounting
rules based on an international survey of Generally Accepted Accounting Principles (GAAP), in 2001.
This measure does not attempt to assess the quality of any given set of accounting rules but the
extent to which accounting standards di¤er between two countries. Bae et al. (2008) suggest that
analysts tend to avoid following foreign …rms adopting accounting rules that are signi…cantly di¤erent
from the accounting rules used in their home country, because they incur costs to gain expertise in
understanding other countries’ GAAP. If this is the case, the "distance" in accounting standards
between two countries should decrease bilateral foreign investments. We construct the measure of
bilateral distance in GAAP and test its impact on foreign equity portfolio investment. We show

12
in column (4) of Table 3 that indeed more distant accounting principles signi…cantly deter bilateral
investment (-0.043). The coe¢ cient of the EClj index is further reinforced (0.654).
Table 1 suggests a signi…cant correlation between liquidity and stock exchange consolidation.
In column (5) of Table 3, we also include a measure of stock market liquidity, that is, the stock
turnover ratio. This covariate displays a negative coe¢ cient that can be interpreted in the light
of the positive correlation between market size and liquidity shown above: a higher stock market
liquidity is associated with a larger market share, which is the denominator of the dependent variable,
and therefore to a lower foreign bias.
Recent literature has also stressed the e¤ect of minority shareholder protection on foreign invest-
ment (Leuz et al. (2009); Giannetti and Koskinen (2010); ?). The index of shareholder rights adopted
in the paper is the "revised" antidirector rights index (Djankov et al. (2008)). This measure revises
the antidirector rights index (ADR), proposed in the seminal paper by La Porta et al. (1998), and
measures how strongly the legal system favours minority shareholders against managers or dominant
shareholders in the corporate decision making process. Consistent with the previous literature, the
e¤ect of the revised ADR on foreign investment is positive (column (6), Table 3). Interestingly,
the inclusion of both the liquidity factor and the revised ADR index has a negligible e¤ect on the
coe¢ cient of the EClj factor.10
Finally, we control for other country-speci…c factors potentially correlated with the stock exchange
consolidation covariate.
Fraudulent transactions, bribery, unenforceable contracts, legal and regulation complexity can
signi…cantly a¤ect portfolio investment (Gelos and Wei (2005); Leuz et al. (2009)). We include an
institutional variables generally related to country level governance: "control of the risk of expropri-
ation", that seizes government stance toward business. We also control for the role of the "e¢ ciency
of the judicial system" in attracting foreign investments. Overall, the introduction of these time
10
Beyond liquidity, also the index of minority shareholder rights is potentially correlated with the stock market size
(La Porta et al. (1998)). This potentially creates a problem of endogeneity because the market capitalization enters
the de…nition of the dependent variable. Given the negligible impact of the inclusion of the two covariates (turnover
ratio and revised ADR) on the coe¢ cient of the EClj variable, we will therefore not include them in the following
speci…cations, in order to prevent unnecessary sources of endogeneity.

13
invariant control factors, for both investing and destination countries, dampens the impact of EClj
to 0.395, as shown in column (7a) of Table 3.
In column (7b) of Table 3, we replace these institutional variables with two time-varying al-
ternative variables, drawn from Worldwide Governance Indicators (WGI, World Bank): "political
stability" and "control of corruption". After the inclusion of these controls, the coe¢ cient of EClj
is equal to 0.617.11
The interpretation of our …ndings is quite straightforward: beyond the liquidity and size e¤ect
bene…ted by all market participants, those investors sharing a common exchange platform can face
the same trading rules and access foreign assets issued by member countries at a cost close to the
domestic one. The standard asymmetry between domestic and foreign assets is therefore dampened,
and foreign investment is enhanced.

4.2.1 Sensitivity analysis on stock exchange consolidation

In Table 4, we check the sensitivity of our …ndings to the speci…cation of the consolidation index.12
In column (1) of Table 4, we test for an alternative de…nition of the Euronext platform including
also the UK. As mentioned above, in 2002, Euronext merged with the futures exchange LIFFE
(London International Financial Futures and Options Exchange). Since LIFFE is not strictly a stock
exchange, the lower coe¢ cient size was expected (0.324) and can be interpreted as supportive of our
thesis.
In October 2006 the OMX Group also took a 10 per cent stake in Oslo Holding ASA, the owner
of Oslo Stock Exchange. In column (2) of Table 4, we specify the OMX platform as including also
Norway from 2007 onward, and the coe¢ cient of the EClj factor remains positive and statistically
signi…cant.
Finally, we test whether our …ndings are driven by a particular stock exchange merger. In columns
(3a) to (3e) of Table 4, we display the results when NYSE-Euronext, Nasdaq-OMX, CEESEG, LSE-
11
Notice that we adopt time-invariant controls in the following speci…cations, even at the cost of losing some hundreds
observations, because their inclusion seems more challenging to the signi…cance of the EClj coe¢ cient that, in so doing,
will undergo a tougher test of signi…cance.
12
The main speci…cation we refer to is the one reported in column (7a) of Table 3.

14
BI, or MILA consolidation is, respectively, excluded from the EClj indicator. The results suggest
that a single merger does not drive our results. However, interestingly, the coe¢ cients display a
notable variability after the exclusion of one of the consolidated platforms. Columns (3a) and (3b),
for instance, suggest, respectively, that the role of NYSE-Euronext is relatively weak, while the role
of Nasdaq-OMX is relatively strong. This evidence suggests that there may exist some country
characteristics making the e¤ect of one stock exchange merger more e¤ective than another. In the
remainder of the paper we further investigate this conjecture.

4.3 Stock exchange consolidation and foreign equity holdings: interac-

tion e¤ects

The process of fusion involves a variety of countries di¤ering by several dimensions –such as size,
openness, geographical and cultural distance of member countries–, and spans a period featuring
violent crises periods and the implementation of several …nancial regulation acts. The e¤ectiveness
of the merger in terms of cross-border investment might be crucially driven by any of the above-
listed characteristics. We analyze below how the strength of the e¤ect of stock exchange consolidation
depends on the features of the countries involved in the process, and on the peculiarities of the time
period considered.

4.3.1 Size

The …rst dimension that is worth studying is the size of the economies involved in the consolidation
process. We cannot rely on the most obvious measure, that is, the stock market capitalization
because this is already accounted for in the denominator of the foreign bias ratio, which represents
the dependent variable. We therefore adopt two alternative proxies.
The …rst one, strictly related to the stock market, relies on the number of publicly listed companied
in a country, which is a meaningful indicator of the size of the stock exchange (columns (1a) to (1c)
of Table 5). The second measure is instead more generally related to the size of the economy and is
the Gross Domestic Product (GDP) (columns (2a) to (2c) of Table 5).

15
 In column (1a) of Table 5, we consider the number of publicly listed companies in the destina-
tion stock exchange. The coe¢ cient of EClj is equal to 0.403, the coe¢ cient of the (relative) size
measure is -0.010, and the coe¢ cient of the interaction term is -0.072. These coe¢ cients show that
countries with a number of publicly listed companies above the mean are relatively underweighted
by foreign investors, and, more interestingly, that the e¤ect of the bilateral consolidation index EClj
is decreasing in the relative size of the destination country. Findings are similar when considering
the size of the investing rather than the destination country (column (1b)).
In columns (2a) and (2b), we analyze the interaction between size and EClj when the size is
captured by the GDP: the results are qualitatively similar, as the larger is the size of destination and
investing countries, the less e¤ective is the attractive role played by stock exchange consolidation.
Interestingly, the e¤ect is ten times smaller in this speci…cation than when the size was captured
more speci…cally to the size of the stock market.
Beyond the size of the countries, their "distance" in size might play a role. We aim to study if
countries more similar in size bene…t more or less from joining a common platform. In columns (1c)
and (2c) of Table 5, we report the results when adopting a measure of the absolute distance in size
between investing and destination country. Our …ndings suggest that the higher the di¤erence in
size of the two member countries, the lower is the role of stock exchange consolidation in attracting
foreign investment. This last result might generally suggests that the more similar are the countries
the more e¤ective is the consolidation process in terms of foreign investment. We then investigate a
potential role for familiarity factors in driving the e¤ectiveness of the stock exchange mergers.

4.3.2 Familiarity

In Table 6, we investigate the interaction of stock exchange consolidation with familiarity factors.
In columns (1) to (3) of Table 6, the EClj index is interacted, respectively, with the geographical
distance between investing and destination country, with the common language dummy, and with
the common legal origin dummy. Results are consistent across the three measures of "distance": the
closer are the countries, the larger the e¤ect of stock exchange consolidation on foreign investment.

16
 In column (4) of Table 6, we interact the EClj index with the GAAPlj index and, consistently,
we …nd that the more distant are the accounting standard of country l and j, the less e¤ective is the
role played by the merger.
Since the ’80s, an increasing number of …rms have listed their shares on foreign stock exchanges.
A growing literature has investigated the issue, studying the determinants of the decision to list
on foreign stock exchanges and of the foreign location (Saudagaran (1988); Saudagaran and Biddle
(1995); Pagano et al. (2001)), and evaluating the …rms’ gains from foreign listing (Sarkissian and
Schill (2009); Sarkissian and Schill (2016)). An intense cross-listing activity before the merger event
can be interpreted as a signal of the closeness of the two countries. In column (5) of Table 6, we
consider the number of cross-listed …rms for each country-pair before the consolidation occurred.
We rely on the data reported in Sarkissian and Schill (2009), relative to bilateral cross-listing in
December 1998, then prior to the analyzed period. We …nd that, in line with the familiarity …ndings
bove, the larger is the number of pre-existing bilateral cross-listed assets, the stronger is the impact
of the exchange consolidation.

4.3.3 Stock exchange openness

The role of stock exchange consolidation may depend upon the level of …nancial openness of the
national stock exchanges before the merger. We consider the percentage of foreign securities traded
on the domestic market (relative to the domestic share) in 2000, prior to the period considered in our
analysis, and its interaction with our index of stock market consolidation EClj . Column (1a) and
(1b) report results when the relative percentage of foreign listed companies refers to, respectively,
the destination and the investing country. In both cases, the coe¢ cient of the interaction term is
negative and statistically signi…cant thus suggesting that the e¤ect of consolidation is lower for those
countries featuring more foreign-oriented national exchanges before the merger.
Together with the negative size e¤ect, this …nding corroborates the hypothesis of decreasing
returns of stock exchange consolidation. Since the merger allows investors to access at a lower cost
foreign securities, the advantage, and therefore the observable e¤ect, is larger the lower is the size of

17
the market and/or the initial level of foreign exposure of the national stock exchange.

4.3.4 Financial crisis

The time period we consider may be non-neutral for the analysis we perform, since it encompasses
the global …nancial crisis and the evolution in …nancial market regulation in major economies. These
events need to be accounted for, both because they might have had a direct impact on interna-
tional diversi…cation incentives, and, more importantly, because these factors can a¤ect the way
international portfolios respond to stock exchange consolidation.
During the global …nancial crisis of 2007-2008 an unprecedented large number of …nancial insti-
tutions collapsed or were bailed out by governments.
In 2010, with increasing fear of excessive sovereign debt, lenders demanded higher interest rates
from eurozone states with high debt and de…cit levels, such as Greece, Spain, Portugal, Ireland, and
subsequently, Italy. During this crisis, several of these countries had their sovereign debt downgraded
to junk status by international credit rating agencies, worsening investor fears.
Column (2a) of Table 7 reports results of an econometric speci…cation including a time dummy for
the 2007-2008 crisis and its interaction term with the EClj index. Column (2b) of Table 7, considers
instead the 2010–2012 sovereign debt crisis. In both crisis periods, we observe that the coe¢ cient of
the interaction term crisis-EClj is negative and statistically signi…cant: the crisis periods appears to
weaken the e¤ectiveness of the stock exchange consolidation.

4.3.5 Financial regulation

In the period under consideration, major changes occurred in the regulatory framework of developed
economies.
In the aftermath of a number of high-pro…le scandals, in 2002 the United States’Congress passed
the Sarbanes–Oxley Act (SOX). The Act is the most important legislation a¤ecting corporate …nan-
cial reporting enacted in the United States since the 1930s. It not only imposes additional disclosure
requirements, but more importantly, proposes substantive corporate governance mandates.

18
 In response, major economies emulated the Sarbanes Oxley Act and the related rules adopted by
US exchanges and securities regulators.
In 2002, the King Committee on Corporate Governance, issued the revision of the 1994 corporate
governance code for South Africa with requirements close to the US SOX.13
In 2003, the government of Canada’s province of Ontario passed the Budget Measures Act, known
as Bill 198, which closely duplicates the regulatory requirements contained in the Sarbanes-Oxley
Act.
In Australia, the Corporate Law Economic Reform Program Act, a modi…cation of the Corpora-
tions Act 2001 which governs corporate law, was enacted in July 2004.
Clause 49 of the Listing Agreement to the Indian stock exchange came into e¤ect from 31 Decem-
ber 2005. It has been formulated for the improvement of corporate governance in all listed companies,
by promoting corporate fairness, transparency and accountability.
In June 2006, the Financial Instruments and Exchange Act, that is the main statute codifying
securities law and regulating securities companies in Japan, was promulgated. It is often referred to
as the Japanese Sarbanes-Oxley Act.
In July 2008 the 8th EU directive 2006/43/EC of the European Parliament came into force in
member states. This Directive is generally considered as the European Sarbanes-Oxley Act.
The Goshen Committee examined the appropriate structure and format for a corporate gover-
nance code in Israel, and recommended to partially adopt, with modi…cations, Sections 302 and 404
of the Sarbanes-Oxley Act of 2002, with full application of the regulations required beginning with
the annual …nancial statements for the period ending on December 2010.
While the welfare impact of more stringent securities legislation is still under debate, these manda-
tory statutes brought about deep changes on the regulatory and supervisory framework which might
have a¤ected the e¤ect of stock exchange consolidation on foreign stock holdings.
We construct a dummy for …nancial system regulation, which is equal to 1 for the countries
13
Unlike other corporate governance codes such as Sarbanes-Oxley, the King Report code is non-legislative and is
based on principles and practices. Compliance with the King Reports is however a requirement for companies listed
on the Johannesburg Stock Exchange. Because of this peculiarity, we checked for the sensitivity of our …ndings to the
exclusion of the South African act, and our results persist.

19
involved, in the relevant years, and 0 otherwise. In column (3a) of Table 7, we check for a direct e¤ect
of this variable on foreign investment and its indirect e¤ect through the stock exchange consolidation.
While we …nd no e¤ect of the SOX legislation on average, its interaction with EClj shows a negative
coe¢ cient: stronger corporate governance acts make less relevant the common sharing of a trading
platform for foreign investment.
Another important legislation, passed in the period under consideration, might crucially a¤ect our
…ndings. The Markets in Financial Instruments Directive (MiFID), a directive that aims to integrate
the European Union’s …nancial markets and to increase the amount of cross border investment
orders, took e¤ect on November 1st, 2007. The MiFID plans to implement new measures, such
as pre- and post-trade transparency requirements and capital requirements that …rms must hold.
The new environment created by MiFID could trigger drastic changes in the architecture of capital
markets and in the organization of …nancial intermediation in Europe. A major feature of MIFID is,
in fact, to open the execution and settlement of equity transactions to a variety of operators, through
competing trading venues, in order to foster competition. Since this could potentially lead to a more
fragmented and opaque infrastructure, then best execution requirements, increased transparency and
information are demanded for the bene…t of market, and to protect investors.
Results in column (3b) of Table 7, show that countries that adopted the MiFID Directive attract
relatively more foreign investment, but no indirect e¤ect passing through stock exchange consolida-
tion can be detected.

5 Conclusions

This paper estimates the e¤ect of stock exchange consolidation on foreign investment. Stock ex-
changes mergers boost the market capitalization of member countries, as expected. Since member
countries’domestic holdings decline, the success of the merger is due to a sharp increase in foreign
investment.
Our …ndings highlight that joining a common stock exchange platform enhances cross-border

20
investments far beyond what is predicted by larger and deeper markets. The consolidation e¤ect
is particularly pronounced among member countries featuring strong familiarity, that is, among
countries that are geographically close, share a common cultural background and similar accounting
principles, and engaged an intense bilateral cross-listing activity before the fusion. The e¤ect of
the merger is stronger for smaller stock exchanges and for exchanges that were less foreign-oriented
before the consolidation. These latter …ndings might suggest the existence of decreasing returns to
stock exchange consolidation for cross-border investment: the larger and the more foreign-oriented
are the member countries’ stock exchanges, the lower is the bene…t of the consolidation. In our
setting we are not able to explore further this issue which we think may be object of investigation
in future research, within a multifacted welfare perspective encompassing equity portfolio investors,
listed …rms and exchange shareholders.

21
References
Bae, K., H. Tan, and M. Welker (2008). International GAAP di¤erences: the impact on foreign
analysts. The Accounting Review 83, 593–628.

Balta, N. and J. Delgado (2009). Home bias and market integration in the EU. CESifo Economic
Studies 55, 110–144.

Barth, M., G. Clinch, and T. Shibano (1999). International accounting harmonization and global
equity markets. Journal of Accounting and Economics 26, 201–235.

Brennan, M. and H. Cao (1997). International portfolio investment ‡ows. Journal of Finance 52,
1851–1880.

Chan, K., V. Covrig, and L. Ng (2005). What determines the domestic bias and foreign bias?
Evidence from mutual fund equity allocations worldwide. Journal of Finance 60, 1495–1534.

Cooper, I. and E. Kaplanis (1994). Home bias in equity portfolios, in‡ation hedging and international
capital market equilibrium. Review of Financial Studies 7, 45–70.

Dahlquist, M., L. Pinkowitz, R. Stulz, and R. Williamson (2003). Corporate governance and the
home bias. Journal of Financial and Quantitative Analysis 38, 87–110.

Djankov, S., R.La Porta, F. Lopez-de-Silanes, and A. Shleifer (2008). The law and economics of
self-dealing. Journal of Financial Economics 88, 430–465.

Dorodnykh, E. (Ed.) (2014). Stock Market Integration: An International Perspective. London:

Palgrave Macmillan UK.

Fidora, M., M. Fratzscher, and C. Thimann (2007). Home bias in global bond and equity markets:
The role of real exchange rate volatility. Journal of International Money and Finance 26, 631–655.

French, K. and J. Poterba (1991). International diversi…cation and international equity markets.
American Economic Review 81, 222–226.

Gelos, R. and S. Wei (2005). Transparency and international portfolio holdings. Journal of Fi-
nance 60, 2987–3020.

Giannetti, M. and Y. Koskinen (2010). Investor protection, equity returns and …nancial globalization.
Journal of Financial and Quantitative Analysis 45, 135–168.

Giofré, M. (2013). Investor protection rights and foreign investment. Journal of Comparative Eco-
nomics 41, 506–526.

Grinblatt, M. and M. Keloharju (2001). How distance, language, and culture in‡uence stockholdings
and trades. Journal of Finance 56, 1053–1073.

Guiso, L., P. Sapienza, and L. Zingales (2009). Cultural biases in economic exchange? Quarterly
Journal of Economics 124, 1095–1131.

22
Hasan, I. and H. Schmiedel (2004). Networks and equity market integration: European evidence.
International Review of Financial Analysis 13, 601–619.

Hasan, I., H. Schmiedel, and L. Song (2012). Growth strategies and value creation: What works best
for stock exchanges? The Financial Review 47, 469–499.

Kang, J. and R. Stulz (1997). Why is there a home bias? an analysis of foreign portfolio equity
ownership in japan. Journal of Financial Economics 46, 3–28.

Kaufmann, D., A. Kraay, and M. Mastruzzi (2010). The worldwide governance indicators : A
summary of methodology, data and analytical issues. World Bank Policy Research Working Paper
No. 5430 .

La Porta, R., F. Lopez-de-Silanes, A. Shleifer, and R. Vishny (1998). Law and …nance. Journal of
Political Economy 106, 1113–1154.

Lane, P. (2006). Global bond portfolios and EMU. International Journal of Central Banking 2, 1–23.

Lane, P. and G. Milesi-Ferretti (2007). The international equity holdings of euro area investors.
in The Importance of the External Dimension for the Euro Area: Trade, Capital Flows, and
International Macroeconomic Linkages (Robert Anderton and Filippo di Mauro, eds), Cambridge
University Press.

Leuz, C., K. Lins, and F. Warnock (2009). Do foreigners invest less in poorly governed …rms? Review
of Financial Studies 22, 3245–3285.

Nielsson, U. (2009). Stock exchange merger and liquidity: the case of euronext. Journal of Financial
Markets 12, 229–267.

Pagano, M., O. Randl, A. Röell, and J. Zechner (2001). What makes stock exchanges succeed?
evidence from cross-listing decisions. European Economic Review 45, 770–782.

Petersen, M. (2009). Estimating standard errors in …nance panel data sets: comparing approaches.
Review of Financial Studies 22, 435–480.

Portes, R. and H. Rey (2005). The determinants of cross-border equity ‡ows. Journal of International
Economics 65, 269–296.

Sarkissian, S. and M. Schill (2009). Are there permanent valuation gains to overseas listing? Review
of Financial Studies 22, 371–412.

Sarkissian, S. and M. Schill (2016). Cross listing waves. Journal of Financial and Quantitative
Analysis 51(1), 259–306.

Saudagaran, S. (1988). An empirical study of selected factors in‡uencing the decision to list on
foreign stock exchanges. Journal of International Business Studies 19, 101–127.

Saudagaran, S. and G. Biddle (1995). Foreign listing location: a study of mncs and stock exchanges
in eight countries. Journal of International Business Studies 26(2), 319–341.

23
Schmiedel, H. and A. Schönenberger (2006). Integration, regulation, and policy of securities market
infrastructures in the euro area. Journal of Financial Regulation and Compliance 14, 328–347.

Sorensen, B. E., Y. Wu, O. Yosha, and Y. Zhu (2007). Home bias and international risk sharing:
Twin puzzles separated at birth. Journal of International Money and Finance 26, 587–605.

Tesar, L. and I. Werner (1995). Home bias and high turnover. Journal of International Money and
Finance 14, 467–492.

24
A Data appendix
A.1 Dependent variables
Foreign stock market portfolios (wlj )
The CPIS dataset contains information on foreign holdings only and does not include domestic
positions. In order to derive the foreign portfolio positions in the overall portfolio we need to retrieve
the share of foreign assets. To accomplish this objective we need the stock market capitalization
of all country indexes, the outstanding foreign equity portfolio investments, and the corresponding
liabilities. Accordingly we can derive the ”foreign equity share”of investing-country l, F Sl 14

(F A)l
F Sl = (2)
(M CAPl + F Al F Ll )

where F A stands for "foreign equity assets", F L for "foreign equity liabilities" and M CAP for
"stock market capitalization". After obtaining the foreign share F S it is possible to recover the
share of each foreign asset in the overall portfolio (wlj ). Source: CPIS (IMF), IFS (IMF), World
Federation of Exchanges
Market share (M Sj )
Market capitalization of listed domestic companies (US$). Data are end of year values converted
to U.S. dollars using corresponding year-end foreign exchange rates.
Source: World Federation of Exchanges database.
Market share ‡oat-adjusted (M S_M SCIj )
The world ‡oat portfolio is a corrected value weighted portfolio obtained by multiplying the
market share by a fraction taking into account the fraction of closely held shares (Dahlquist et al.
(2003)). The adopted MSCI Investable Market Indexes (IMI) cover all investable large, mid and
small cap securities across the Developed, Emerging and Frontier Markets, targeting approximately
99% of each market’s free‡oat adjusted market capitalization. Source: MSCI

A.2 Main regressors

To ensure consistency with the theoretical framework, each variable X (dummy variables excluded)
enters our regression speci…cations as the ratio of X to its world average. The subscript indicates
that the corresponding variable can be referred to the country-pair lj, the destination country j
and/or to the investing country l.
Stock exchange consolidation (EC )
CEESEG: Austria, Hungary, Slovenia, Czeck Republic since 2010.
NYSE-EURONEXT: Belgium, France and the Netherlands since 2001, Portugal since 2003, and
the USA since 2007. In the speci…cation in column (1) of Table 4, it also comprises the UK exchange
since 2002 onwards (because of the fusion with the LIFFE platform).
NASDAQ-OMX: Sweden, Finland, Estonia since 2004, Denmark since 2005, and the USA since
2008. In the speci…cation in column (2) of Table 4, it also comprises the Norway since 2007 onwards
(because of the acquisition of the 10% of the ownership of the Oslo Stock Exchange).
14
Fidora et al. (2007) and Sorensen et al. (2007) follow the same procedure dealing with the CPIS dataset.

25
 MILA: Colombia, Perù, and Chile since 2010.
LSE-IT: the UK and Italy since 2007.
EClj: Dummy variable taking value 1 if the investing and the destination country share a common
exchange platform (0 otherwise).
ECl: Dummy variable taking value 1 if the investing country l is part of a consolidated platform
(0 otherwise).
ECj: Dummy variable taking value 1 if the destination country j is part of a consolidated platform
(0 otherwise).
Distance (distlj )
The distance is measured as the Great Circle distance in miles between capital cities of source
(l) and destination (j) country. The average distance from a destination country (j) is obtained as
weighted (by market share) average of the distance of investing countries. The variable included in
the regression is the ratio of the distance l j to the average distance.
Common Border. (borderlj )
Dummy variable taking value of 1 if the investing country and the destination country share a
common border (0 otherwise).
Common Language. (langlj )
Dummy variable taking value of 1 if the investing country and the destination country share a
common language (0 otherwise)
Colonial linkage (colony lj )
Dummy variable taking value of 1 if the investing country and the destination country share a
colonial linkage (0 otherwise)
EMU (EM Ulj )
Dummy variable taking value of 1 if the investing country and the destination country are mem-
bers of the European Monetary Union (0 otherwise). In our case, it coincides with a common currency
dummy since included countries do not belong to any other currency union.
Equal legal origin (equal_leg_originlj )
Dummy variable taking value 1 if the investing country and the destination country share the
same legal origin of the company law or commercial code of each country (0 otherwise). The countries
included in our sample belong to four legal families: English, French, German, Scandinavian.
Generalized Accepted Accounting Principles (GAAPlj )
Total number of GAAP (Generally Accepted Accounting Principles) di¤erences between investing
country l and destination country j. Measure based on the measure gaapdi¤2 in Bae et al. (2008).
Source: Bae et al. (2008).
International capital mobility (cap_mob )
Index (0-10) measuring the restrictions countries impose on capital ‡ows assigning a lower rating
to countries with more restrictions on foreign capital transactions. In decreasing rating order are
ranked countries where: a) domestic investments by foreigners and foreign investments by local
residents are unrestricted; b) investments are restricted in a few industries within the countries; c)
investments are permitted but regulatory restrictions slow the mobility of capital; d) either domestic
investments by foreigners or foreign investments by local residents require approval from government
authorities; e) both domestic by foreigners and foreign investments by local require government
approval. Source: Economic Freedom Network.
Turnover ratio (turnover ratioj )

26
 Turnover ratio is the total value of shares traded during the period divided by the average market
capitalization for the period. Average market capitalization is calculated as the average of the end-
of-period values for the current period and the previous period. Source: Financial Sector Indicators
(World Bank)
Traded stocks to GDP (traded stocksj =GDPj )
Stocks traded refers to the total value of shares traded during the period. Source: Financial
Sector Indicators (World Bank)
Number of listed shares (listed_shares )
Listed domestic companies are the domestically incorporated companies listed on the country’s
stock exchanges at the end of the year. This indicator does not include investment companies, mutual
funds, or other collective investment vehicles. Source: Financial Sector Indicators (World Bank)
Gross Domestic Product (GDP )
GDP at purchaser’s prices is the sum of gross value added by all resident producers in the economy
plus any product taxes and minus any subsidies not included in the value of the products. Data are
in current U.S. dollars. Source: Economy and Growth Indicators (World Bank).
Stock exchange openness (exchange_openness )
This variable is the ratio of foreign listed companies to domestic listed companies at the end of
year 2000. Source:World Federation of Exchanges.
Cross listing (cross_listinglj )
This variable captures the number of country-to-country listings of the country pair lj (Sarkissian
and Schill (2009),Panel A of Table 2).
MIFID (M iF IDj )
Dummy variable equal to 1 for the destination country j belonging to the European Union since
2008 onwards (0 otherwise).
SOX (SOXj )
Dummy variable equal to 1 for the destination country j adopted a signi…cant legislative improve-
ment in corporate …nancial reporting similar to the Sarbanes-Oxley Act in the US (0 otherwise). For
the US and South Africa: 2002; for Canada: 2003; for Australia: 2004; for Japan and India: 2006;
for the European Union: 2008; for Israel: 2011.

A.3 Other controls

Revised Antidirector Rights Index
The index amends the original LLSV (1998) index (Djankov et al. (2008)). The revised index
relies on the same basic dimensions of corporate law, but de…nes them with more precision. Both
the original and the revised anti-director rights indices summarize the protection of minority share-
holders in the corporate decision-making process, including the right to vote. The index covers the
following six areas: (1) vote by mail; (2) obstacles to the actual exercise of the right to vote (i.e., the
requirement that shares be deposited before the shareholders’meeting); (3) minority representation
on the board of directors through cumulative voting or proportional representation; (4) an oppressed
minority mechanism to seek redress in case of expropriation; (5) preemptive rights to subscribe to
new securities issued by the company; and (6) the right to call a special shareholder meeting. The
general principle behind the construction of the revised anti-director rights index is to associate bet-
ter investor protection with laws that explicitly mandate, or set as a default rule, provisions that are
favorable to minority shareholders. Methodologically, the key di¤erence between the original and

27
revised indices of anti-director rights lies in the treatment of enabling provisions. See Djankov et al.
(2008) for further details.

A.3.1 Time-invariant country governance

Expropriation risk
ICR’s assessment of the risk of "outright con…scation" or "forced nationalization". Scale from
zero to 10 with lower scores for higher risk (LLSV (1998)).
E¢ ciency of judicial system
Assessment of the "e¢ ciency and integrity of the legal environment as it a¤ects business, par-
ticularly foreign …rms" produced by Business International Corporation. Scale from zero to 10 with
lower scores for lower e¢ ciency level (LLSV (1998))..

A.3.2 Time-varying country governance

The following two time-varying variables are drawn from the Worldwide Governance Indicators (WGI,
World Bank).
Details on the underlying data sources, the aggregation method, and the interpretation of the
indicators, can be found in the WGI methodology paper (Kaufmann et al. (2010)).
The original indexes range from -2.5 to +2.5 with an average of 0. Since our variables all enter
in relative terms, we use the average as denominator and to avoid the zero in the denominator we
re-scale the range from 0 to 5 with an average of 2.5. Note that the descriptive statistics’ table
reports a mean that di¤ers from 2.5 because it reports averages across countries included in our
sample rather than global ones.
Regulatory quality
Regulatory quality captures perceptions of the ability of the government to formulate and imple-
ment sound policies and regulations that permit and promote private sector development.
Control of corruption
This index captures perceptions of the extent to which public power is exercised for private gain,
including both petty and grand forms of corruption, as well as "capture" of the state by elites and
private interests.

28
Tables
Table 1. Correlation of stock exchange consolidation with size, liquidity, and domestic
investment.
This table reports the correlation of stock exchange consolidation with size, liquidity, and domestic
investment.
***, **, and * indicate signi…cance at the 1, 5, and 10% levels, respectively.

Size, liquidity and domestic investment

market capitalization (destination) country j liquidity (destination) country j domestic share (investing) country l
MSj MS_MSCIj rel (traded stocks/GDP)j rel (turnover ratio j ) wll
(1a) (1b) (2a) (2b) (3a) (3b) (4a) (4b) (5a) (5b) (5c)

country j (destination)
stock exchange 0.026*** 0.027*** 0.039*** 0.041*** 0.222*** 0.203*** 0.219*** 0.241*** - - -
consolidation (ECj )

country l (investing)
stock exchange - - - - - - - - -0.260*** -0.248*** -0.119***
consolidation (ECl )

country l (investing)
no no no no no no no no no no yes
fixed effects
year fixed effects no yes no yes no yes no yes no yes yes

Table 2. Correlation of stock exchange consolidation with foreign investment..

This table reports the correlation of stock exchange consolidation with portfolio share (wlj ), and foreign
bias (wlj =M S_M SCIj )
***, **, and * indicate signi…cance at the 1, 5, and 10% levels, respectively.

Foreign investment
portfolio share (wlj ) foreign bias (wlj /MS_MSCIj )

(1) (2) (3) (4) (4a) (5) (6) (7) (8) (8a)
country j (destination)
stock exchange 0.003*** - - 0.002*** 0.002*** 0.016 - - -0.061*** -0.012
consolidation (ECj )

country l (investing)
stock exchange - 0.003*** - 0.002*** 0.000** - 0.192*** - 0.194*** -0.024
consolidation (ECl )

bilateral l-j stock

exchange consolidation - - 0.016*** 0.012*** 0.011*** - 1.038*** 0.899*** 0.891***
(EClj )

country l (investing)
no no no no yes no no no no yes
fixed effects
year fixed effects no no no no yes no no no no yes

29
 Table 3. Stock exchange consolidation: main …ndings.
This table reports results of a feasible GLS regression. The dependent variable is the scaled foreign
portfolio, i.e., the ratio of portfolio share to market share, (wlj /M S_M SCIj ), where the subscript lj
represents the couple investing country l -destination country j . Each regressor X (dummy variables
excluded) is expressed as the ratio of X to its world average. Two-way clustered (investing country and
time) standard errors are reported in parentheses. ***, **, and * indicate signi…cance at the 1, 5, and 10%
levels, respectively.

Exchange Consolidation: main findings

(1) (2) (3) (4) (5) (6) (7a) (7b)

exchange consolidation lj (EClj ) 0.798 *** 0.646 *** 0.612 *** 0.654 *** 0.656 *** 0.659 *** 0.395 *** 0.617 ***
( 0.037 ) ( 0.041 ) ( 0.041 ) ( 0.043 ) ( 0.043 ) ( 0.043 ) ( 0.039 ) ( 0.039 )
distlj -0.143 *** -0.135 *** -0.111 *** -0.096 *** -0.095 *** -0.095 *** -0.060 *** -0.093 ***
( 0.009 ) ( 0.009 ) ( 0.009 ) ( 0.008 ) ( 0.008 ) ( 0.008 ) ( 0.006 ) ( 0.007 )
lang lj 0.145 *** 0.113 *** 0.086 *** 0.109 *** 0.105 *** 0.099 *** 0.106 *** 0.045 ***
( 0.014 ) ( 0.015 ) ( 0.015 ) ( 0.014 ) ( 0.014 ) ( 0.014 ) ( 0.012 ) ( 0.013 )
borderlj 0.713 *** 0.586 *** 0.612 *** 0.464 *** 0.473 *** 0.471 *** 0.363 *** 0.442 ***
( 0.024 ) ( 0.024 ) ( 0.023 ) ( 0.020 ) ( 0.020 ) ( 0.020 ) ( 0.017 ) ( 0.019 )
EMUlj 0.523 *** 0.487 *** 0.483 *** 0.486 *** 0.488 *** 0.525 *** 0.451 ***
( 0.041 ) ( 0.040 ) ( 0.039 ) ( 0.039 ) ( 0.039 ) ( 0.034 ) ( 0.035 )
equal_leg_origin lj -0.025 ** -0.006 -0.009 -0.009 -0.007 0.008 0.035 ***
( 0.010 ) ( 0.010 ) ( 0.008 ) ( 0.008 ) ( 0.008 ) ( 0.007 ) ( 0.008 )
colony lj 0.188 *** 0.173 *** 0.102 *** 0.106 *** 0.098 *** 0.052 *** 0.091 ***
( 0.017 ) ( 0.017 ) ( 0.015 ) ( 0.015 ) ( 0.015 ) ( 0.012 ) ( 0.014 )
cap_mob l 0.130 *** 0.125 *** 0.125 *** 0.127 *** 0.022 ** -0.005
( 0.011 ) ( 0.011 ) ( 0.011 ) ( 0.011 ) ( 0.010 ) ( 0.011 )
cap_mob j 0.044 *** 0.041 *** 0.042 *** 0.040 *** 0.022 *** 0.012 *
( 0.006 ) ( 0.006 ) ( 0.006 ) ( 0.006 ) ( 0.005 ) ( 0.006 )
GAAPlj -0.043 *** -0.040 *** -0.042 *** -0.020 * -0.033 ***
( 0.012 ) ( 0.012 ) ( 0.012 ) ( 0.011 ) ( 0.011 )
turnover ratio j -0.028 ***
( 0.006 )
rev_ADRj 0.052 ***
( 0.012 )

country l 's and j' s controls (time invariant) no no no no no no yes no

country l 's and j' s controls (time varying) no no no no no no no yes
#obs 18409 18397 18012 16650 16599 16650 12935 16650
2
Adj-R 0.12 0.14 0.15 0.15 0.15 0.16 0.19 0.19

30
 Table 4. Stock exchange consolidation: sensitivity analysis.
This table reports results of a feasible GLS regression. The dependent variable and controls are the same
as in Table 3. The only di¤erence refers to the de…nition of our main variable of interest, the stock exchange
consolidation. In column (1) the Euronext platform also includes the United Kingdom, joining the Euronext
in 2002 through the LIFFE derivative platform. In column (2), the OMX-Nasdaq group also includes the
Norway stock market, owned at 10% by the group since 2007. In columns (3a) to (3e), one stock exchange
consolidation at a time is excluded. Two-way clustered (investing country and time) standard errors are
reported in parentheses. ***, **, and * indicate signi…cance at the 1, 5, and 10% levels, respectively.

Exchange Consolidation: sensitivity analysis

(1) (2) (3a) (3b) (3c) (3d) (3e)

Euronest-LIFFE OMX-Oslo no Euronext no OMX no CEESEG no LSE-BI no MILA

exchange consolidation lj (EClj ) 0.324 *** 0.593 *** 1.152 *** 0.118 *** 0.395 *** 0.404 *** 0.406 ***
( 0.032 ) ( 0.040 ) ( 0.074 ) ( 0.034 ) ( 0.039 ) ( 0.039 ) ( 0.039 )
dist lj -0.058 *** -0.060 *** -0.058 *** -0.062 *** -0.060 *** -0.061 *** -0.060 ***
( 0.006 ) ( 0.006 ) ( 0.006 ) ( 0.006 ) ( 0.006 ) ( 0.006 ) ( 0.006 )
dum_lang lj 0.106 *** 0.108 *** 0.109 *** 0.105 *** 0.106 *** 0.106 *** 0.107 ***
( 0.012 ) ( 0.012 ) ( 0.012 ) ( 0.012 ) ( 0.012 ) ( 0.012 ) ( 0.012 )
dum_borderlj 0.366 *** 0.347 *** 0.364 *** 0.376 *** 0.363 *** 0.363 *** 0.365 ***
( 0.017 ) ( 0.017 ) ( 0.017 ) ( 0.017 ) ( 0.017 ) ( 0.017 ) ( 0.017 )
dum_EMUlj 0.541 *** 0.504 *** 0.594 *** 0.567 *** 0.525 *** 0.523 *** 0.524 ***
( 0.033 ) ( 0.034 ) ( 0.033 ) ( 0.035 ) ( 0.034 ) ( 0.034 ) ( 0.034 )
dum_eq_leg_origin lj 0.010 0.004 0.010 0.015 ** 0.008 0.008 0.008
( 0.007 ) ( 0.007 ) ( 0.007 ) ( 0.007 ) ( 0.007 ) ( 0.007 ) ( 0.007 )
colony lj 0.049 *** 0.049 *** 0.047 *** 0.057 *** 0.052 *** 0.052 *** 0.052 ***
( 0.013 ) ( 0.012 ) ( 0.012 ) ( 0.013 ) ( 0.012 ) ( 0.012 ) ( 0.012 )
cap_mob l 0.021 ** 0.024 ** 0.024 ** 0.022 ** 0.022 ** 0.023 ** 0.022 **
( 0.010 ) ( 0.010 ) ( 0.010 ) ( 0.011 ) ( 0.010 ) ( 0.010 ) ( 0.010 )
cap_mob j 0.020 *** 0.021 *** 0.023 *** 0.023 *** 0.022 *** 0.022 *** 0.022 ***
( 0.005 ) ( 0.005 ) ( 0.005 ) ( 0.006 ) ( 0.005 ) ( 0.005 ) ( 0.005 )
GAAPlj -0.019 * -0.021 ** -0.018 * -0.016 -0.020 * -0.020 * -0.020 *
( 0.011 ) ( 0.010 ) ( 0.010 ) ( 0.011 ) ( 0.011 ) ( 0.011 ) ( 0.011 )

country l 's and j' s controls

yes yes yes yes yes yes yes
(time invariant)

#obs 12935 12935 12935 12935 12935 12935 12935

2
Adj-R 0.19 0.20 0.21 0.19 0.19 0.19 0.19

31
 Table 5. Stock exchange consolidation: interaction e¤ects. I. Market size.
This table reports results of a feasible GLS regression. The dependent variable and the regressors are
the same as in Table 3, with additional regressors and their interaction with stock exchange consolidation,
as explicitly reported in the table. Two-way clustered (investing country and time) standard errors are
reported in parentheses. ***, **, and * indicate signi…cance at the 1, 5, and 10% levels, respectively.

Exchange Consolidation: interaction effects

I. Size

(1a) (1b) (1c) (2a) (2b) (2c)

exchange consolidation lj (EClj ) 0.403 *** 0.542 *** 0.778 *** 0.464 *** 0.601 *** 0.772 ***
( 0.039 ) ( 0.038 ) ( 0.050 ) ( 0.042 ) ( 0.041 ) ( 0.049 )
listed shares j -0.010 ***
( 0.002 )
listed shares j •EClj -0.072 ***
( 0.013 )
listed shares l -0.026 ***
( 0.002 )
listed shares l •EClj -0.070 ***
( 0.013 )
|listed shares j - listed shares l | -0.114 ***
( 0.013 )
|listed shares j - listed shares l |•EClj -0.013 ***
( 0.002 )
GDPj -0.033 ***
( 0.006 )
GDPj •EClj -0.007 ***
( 0.001 )
GDPl -0.034 ***
( 0.006 )
GDPl •EClj -0.005 ***
( 0.002 )
|GDPj -GDPl | -0.048 ***
( 0.006 )
|GDPj -GDPl | •EClj -0.004 ***
( 0.001 )

gravity variables, EM U, common legal origin,

yes yes yes yes yes yes
GAAP, colonial linkage, capital mobility

country l 's and j' s controls (time invariant) yes yes yes yes yes yes

#obs 12934 12934 12934 12934 12934 12934

2
Adj-R 0.22 0.20 0.20 0.20 0.19 0.20

32
 Table 6. Stock exchange consolidation: interaction e¤ects. II. Familiarity.
This table reports results of a feasible GLS regression. The dependent variable and regressors are the
same as in Table 3, with additional regressors and their interaction with stock exchange consolidation, as
explicitly reported in the table. Two-way clustered (investing country and time) standard errors are reported
in parentheses. ***, **, and * indicate signi…cance at the 1, 5, and 10% levels, respectively.

Exchange Consolidation: interaction effects

II. Familiarity

(1) (2) (3) (4) (5)

exchange consolidation lj (EClj ) 0.846 *** 0.213 *** 0.071 0.935 *** 0.084 **
( 0.051 ) ( 0.042 ) ( 0.079 ) ( 0.136 ) ( 0.043 )
distlj -0.060 *** -0.060 *** -0.060 *** -0.060 *** -0.147 ***
( 0.006 ) ( 0.006 ) ( 0.006 ) ( 0.006 ) ( 0.010 )
(dist lj )•(EClj ) -0.730 ***
( 0.072 )
dum_lang lj 0.098 *** 0.095 *** 0.104 *** 0.106 *** 0.120 ***
( 0.012 ) ( 0.012 ) ( 0.012 ) ( 0.012 ) ( 0.017 )
(dum_lang lj )•(EClj ) 0.842 ***
( 0.069 )
dum_eq_leg_origin lj 0.007 0.011 0.006 0.008 0.052 ***
( 0.007 ) ( 0.007 ) ( 0.007 ) ( 0.007 ) ( 0.011 )
(dum_eq_leg_origin lj )•(EClj ) 0.459 ***
( 0.080 )
GAAPlj -0.024 ** -0.025 ** -0.021 ** -0.019 * -0.009
( 0.011 ) ( 0.011 ) ( 0.011 ) ( 0.011 ) ( 0.018 )
(GAAPlj )•(EClj ) -0.466 ***
( 0.119 )
cross-listing lj 0.001 **
( 0.001 )
(cross-listing lj )•(EClj ) 0.045 ***
( 0.003 )

common border, EM U, capital mobility yes yes yes yes yes

country l 's and j' s controls (time invariant) yes yes yes yes yes
#obs 12935 12935 12935 12935 8980
2
Adj-R 0.20 0.20 0.20 0.19 0.26

33
 Table 7. Stock exchange consolidation: interaction e¤ects. III. Openness, crisis,
regulation.
This table reports results of a feasible GLS regression. The dependent variable and regressors are the
same as in Table 3, with additional regressors and their interaction with stock exchange consolidation, as
explicitly reported in the table. Two-way clustered (investing country and time) standard errors are reported
in parentheses. ***, **, and * indicate signi…cance at the 1, 5, and 10% levels, respectively.

Exchange Consolidation: interaction effects

III. Openness, crisis, regulation

(1a) (1b) (2a) (2b) (3a) (3b)

exchange consolidation lj (EClj ) 0.585 *** 0.673 *** 0.430 *** 0.446 *** 0.512 *** 0.420 ***
( 0.053 ) ( 0.058 ) ( 0.043 ) ( 0.046 ) ( 0.050 ) ( 0.042 )
stock exchange openness j 0.001
( 0.001 )
(stock exchange openness j )•(EClj ) -0.051 ***
( 0.010 )
stock exchange openness l 0.013 ***
( 0.004 )
(stock exchange openness l )•(EClj ) -0.091 ***
( 0.013 )
financial crisis 0.011
( 0.016 )
(financial crisis)•(EClj ) -0.153 *
( 0.091 )
sovereign debt crisis 0.025 **
( 0.011 )
(sovereign debt crisis)•(EClj ) -0.149 *
( 0.080 )
SOXj 0.000
( 0.010 )
(SOXj )•(EClj ) -0.193 ***
( 0.071 )
MIFIDj 0.046 ***
( 0.015 )
(MIFIDj )•(EClj ) -0.020
( 0.075 )

gravity variables, EM U, common legal origin, yes yes yes yes yes yes
GAAP, colonial linkage, capital mobility
country l and j' s controls (time invariant) yes yes yes yes yes yes

#obs 11977 12935 12935 12935 12935 12935

2
Adj-R 0.19 0.19 0.19 0.20 0.19 0.20

34