Thesis: Analysis of the Efficiency and the Future of the Foreign Cross-Listing

IV.3.b. Current Tendency: the "Fading Listing"

According to a study of the Banque de France made at the end of December 2007, 38.5% of the companies composing the French index CAC 40⁵⁰ was owned by nonresidents investors. However, the trading activity analysis of the same crop of companies highlights the predominant share of 96.8%⁵¹ held by Euronext Paris, clearly demonstrating that the very great majority of investors are used to buying and to selling their shares on the company's main trading market.

Furthermore, a more detailed look at the evolution of volume distribution in cases of efficient foreign cross-listings reinforces this notion of investors' preference for the most liquid trading place.

The following exhibits #38 illustrate this tendency applied to bi-national companies, materialized by steadily diminishing volumes on the secondary listing places during the years following the merger (see most representative cases of ABB, AstraZeneca, InBev, SABMiller and Anglo American).

By this way, after a certain period of time the most liquid place overrides the other ones, leading to the marginalisation of the secondary foreign cross-listings. Hence, an efficient foreign cross-listing in year T, may become no longer efficient in year T+8 (e.g. see cases of Aventis, SAS, AngloGold Ashanti and Altadis). In this research, this new notion will be called the "Fading Listing". After a merger operation, the fading listing seems to be verified whatever the company's origin country and the secondary stock exchange. The average CAGR⁵² fading listing calculated thanks to the examples presented in the exhibits #38 reaches -19.01%, suggesting that after a merger operation the share in the total volumes held by the secondary listing places losses on average -19.01% per year.

The case of the Anglo-South African SABMiller is probably the most interesting case (see figures in exhibit #38). In 1999 the group SAB (South African Breweries, and later renamed SABMiller after the purchase of the U.S. based company Miller Brewing in 2002) initiated a listing on the L.S.E. At this time, the share in volumes traded in London accounted for 32%, the rest being traded in Johannesburg. Nine years later, the influence is reversed since London henceforth accounts for 73% of the total volumes. During these years, the overall greater liquidity of the British stock exchange had played the role of catalyst by attracting

⁵⁰ Are considered 36 companies belonging to the CAC 40, i.e. except those which have their head offices outside France like EADS, ArcelorMittal, Dexia and STMicroelectronics

51

6-months average, source: ThomsonReuters Datastream

⁵² Compound Annual Growth Rate (CAGR), for more details see Appendix 1

more and more investors. This example confirms the preference of investors for the most liquid places.

However, it is important to have a closer look at the foreign cross-listings in the United States. Indeed the exhibit #37, clearly suggests that a foreign cross-listing in the United States is generally not impacted by the fading listing phenomenon. As we may notice, companies resulting from merger operations with an American company naturally have a significant share of their trading activity performed in New York; share which is even used to gaining in weight over the years (GlaxoSmithKline 18% in 2008 vs. 12% in 2000), Alcatel-Lucent (32% in 2008 vs. 24% in January 2007). But such levels may also concern companies not resulting of a merger operation such as the German companies SAP (24% in 2008 vs. 10% in 2000) and Infineon AG (17% in 2008 vs. 5.8% in 2004), the British Diageo (22% in 2008 vs. 8% in 2000). Therefore, after a foreign cross-listing in the United States, we may state that the American listing gains in importance in the company's total trading activity.

#37: Evolutions of the Share in Total Volumes of Foreign Cross-Listings
in New York (Period 2000-2008)

	British Companies	French Companies	German Companies		Japanese Companies	New York Examples
	+44%	+37%	+373%
	32% Alcatel-Lucent	24% SAP	25% Nidec		% max of volumes traded in New York Company

Source: ThomsonReuters Datastream

AstraZeneca

As of after the merger, i.e. 01/2000 As of 11/2008

TeliaSonera

As of after the merger, i.e. 01/2003 As of 11/2008

Source: ThomsonReuters Datastream

66

#38: Shares of Volume Evolutions

ABB

Aventis

As of 01/2000 As of 11/2008

As of after the merger, i.e. 01/1999 As of 1/2004

Nyse Euronext

Euronext

Brussels
0.04%

Amsterdam
0.03%

Brussels
0.1%

As of before the merger w ith Nyse,
i.e. 5/2007

As of after the creation of Euronext,
i.e. 08/2001

As of after the merger, i.e. 05/2007 As of 11/2008

Amsterdam
-89%

Paris
99%

Paris
8%

9%

Paris

91% Amsterdam

Amsterdam
1%

New York
92%

Paris

-88%

New York

99% Paris

1%

Stockholm

Frankfurt
-88%

-34%

Paris
92%

Zurich
79%

Frankfurt
1%

Zurich
68%

Stockholm
32%

Stockholm
21%

Frankfurt

8% Paris

99%

Unibail-Rodamco

As of after the merger, i.e. 11/2007 As of 11/2008

EADS

As of after the merger, i.e. 01/2001 As of 11/2008

Amsterdam
-31%

Frankfurt
12%

Paris
87%

Paris
90%

Paris
76%

Spain
12%

Amsterdam
13%

Paris

91% Amsterdam

9%

Frankfurt -25%
Madrid -92%

Frankfurt

9% Spain

1%

London
62%

Stockholm
38%

Stockholm

-37%

London
76%

Stockholm
24%

Stockholm
76%

Helsinki
24%

Helsinki

-54%

Stockholm
89%

Helsinki
11%

Fortis

As of after the merger, i.e. 01/1992 As of 11/2008

Amsterdam
90%

Brussels
10%

Amsterdam
-13%

Amsterdam
78%

Brussels
22%

As of 01/2002 As of 11/2008

Copenhagen
26%

Stockholm

41%

Oslo

33%

Oslo -91%

Oslo

Stockholm 3%
68%

Copenhagen
29%

Anglo American AngloGold Ashanti

As of after the merger, i.e. 01/1 999 As of 11/2008

As of after the merger, i.e. 06/2004 As of 11/2008

Altadis

As of after the merger, i.e. 05/2000 As of 2/2008

SABMiller

As of after the IPO on L.S.E, i.e. 12/1999 As of 11/2008

Royal Dutch Shell

InBev

Paris

0.89%

Paris

0.59%

Nasdaq
4%

Johannesburg
-38%

Sydney -75%
Paris -34%

London
68%

Sydney
2%

London
54%

Johannesburg
42%

Johannesburg
26%

Johannesburg

91% Sydney

8%

Johannesburg
97%

Nasdaq
6%

Paris -94%

London
73%

London
32%

Madrid

88% Paris

12%

Madrid

99% Paris

0.7%

Johannesburg
68%

Johannesburg
-60%

Johannesburg
27%

As of after the merger, i.e. 04/2004 As of 11/2008

As of after the merger, i.e. 07/2005 As of 11/2008

Brazil
48%

Brazil -48%

Brazil
25%

London

38%

London -13%

London
33%

New York
8%

New York
11%

Belgium
75%

Belgium
52%

Amsterdam
51%

Amsterdam
59%

Source: ThomsonReuters Datastream

#39: Evolutions of the Share in Total Volumes of Foreign Cross-Listings in
Financial Places Specialized in the Basic Resources and Mining Sectors
(Period 2006-2008)

As Secondary Stock Share of Volume % of Positive

Exchange Variation Variations *

										TSX & TSX Venture L.S.E
		+12.9% +42.2%				38% 33%			ASX
		+45.2%				57%

Source: ThomsonReuters Datastream

* Percentage of foreign cross-listing cases presenting an increase of the weight of TSX, L.S.E or ASX in the total volumes

A first glance at the exhibit #39 would let suggest that foreign cross-listings in specialized financial places are efficient thanks to positive variations of the share in total volumes during the period 2006-2008. The listing fading seems not to be verified for these cases of foreign cross-listings, with shares in volume variation reaching +12.9% in Toronto, +42.2% in London and +45.2% in Sydney.

However, this presumption has to be mitigated since the tendency is not as clear as the one showed in the cases of foreign cross-listings in the United States. Indeed, the findings are not totally convincing, since the percentage of cases with positive variations is relatively mixed for all three listing places, but above all a majority of negative variations on the Toronto and the London stock exchanges. Only the cases presenting negative shares of volumes variation suffer from the fading listing phenomenon.