Does Geography Still Matter? Evidence on the Portfolio Turnover of Large Equity Investors and Varieties of Capitalism

Abstract

This article investigates the geography of finance through a study of the behavior of large equity investors who are key actors in capitalism. The main argument is based on their expectations in “finance-driven” capitalism: large equity investors require high returns on invested capital in a shorter time and are said to be impatient. The article focuses on their portfolio turnover in relation to geographic factors and their attachment to a specific model of capitalism. The U.S. “market-based” model is presented as a benchmark, since U.S. investors trade securities most frequently relative to other international equity investors. Our empirical findings on the proximity of investors in various models of capitalism with U.S. “impatient” investors contribute to a growing literature on the economic importance of geography in understanding global finance.

Keywords:

• geography of finance
• equity investors
• portfolio turnover
• varieties of capitalism
• global finance

Acknowledgments

This is a revised version of a paper presented at the annual international conference of the Royal Geographical Society, London, 28–30 August 2007. We thank Karel Williams, Julie Froud, Jonathan Winterton, Sarah Hall, Yannick Lung, and Jérôme Vicente for their comments, discussions, and encouragements. We also thank Henry Wai-chung Yeung for his editorial suggestions and two reviewers for their constructive comments that have improved the quality of the article. This article was made possible thanks to our participation in the European project, European Socio-Economic Models of a Knowledge-based Society, FP 6, Priority 7. This project was initiated by 60 researchers from 12 European centers in sociology, history, economics, and management, including French centers in economics (CEPREMAP, GRES, GERPISA, CERNA, INSEAD) and centers in Manchester (CRESC), Bristol (ESRU), Germany (WZB), and Italy (Università di Padova) from 2004 to 2007.

Notes

1 In 2002, currency markets transactions reached $384,400 billion versus $32,300 billion for goods and services transactions, a fact that confirms the domination of the financial sphere over the real economy (Morin 2006).

2 Financialization is a concept that describes the growing importance of financial markets on firms and household behavior and the major changes that occurred in large firms’ governance and strategies during the 1990s under the influence of institutional investors (Martin and Minns 1995; Williams 2000; Morin 2006).

3 La Porta, Lopez-De-Silanes, and Shleifer (1998, 2000) demonstrated the coexistence of different groups of financial markets explained by different legal institutions. Taking account of both civil law and common law countries, they showed that financial markets are more developed in countries where investors’ rights are better protected.

4 Data are from the Thomson Financial equity database, Thomson One Banker—Ownership (TOBO) (also known as Shareworld GEO Carson), which lists international capital flows and registers investors’ equity portfolios across international stock markets (see Table 1).

5 The portfolio turnover refers to the measure of trading activity in a fund’s portfolio. It is a percentage of the stock that is bought and sold in exchange for other stocks and is calculated by looking at the amount of new securities purchased and the amount of securities sold over the period divided by the total net asset value of the fund. It is usually reported for a 12-month period. Portfolio turnover is an indication of the time horizon of investors and allows one to identify short-term investors (with a high turnover) and long-term investors (with a low turnover). A ratio of 100 percent means that, on average, funds’ managers sell all the securities in their portfolios once a year.

6 The U.S. Securities and Exchange Commission requires every mutual fund to publish its turnover rate, an indicator of how often a fund manager trades securities. For instance, a 100-percent turnover rate means that the entire portfolio was traded over a 1-year period, and a 10-percent turnover rate indicates that it took 10 years to achieve the turnover of the entire portfolio.

7 If the investor’s turnover is high (our benchmark in the study), the variable has the value of zero; in all other cases (zero, low, and moderate), the variable has the value of 1. A multinomial analysis with the four modalities gave few robust results.

8 The choice of this method is required because we have qualitative variables (turnover rate and models of capitalism) which are binary.

9 This result can be explained by the recent development of financial markets, particularly in Spain and Portugal. The heterogeneity of countries in the Mediterranean model of capitalism is presented in Table C4.

10 The logit model cannot estimate a coefficient or a marginal effect for the Central and Oriental European model because this variable is coded only 1, since no investors belonging to this model have a high turnover. The probability of having a low turnover in this model is necessarily the highest, and we arbitrarily classified this model as the furthest from the market-based model. If the variable equity asset is significant and positive, the marginal effect remains relatively small.

11 We could not propose an estimation for New Zealand because we had only 13 observations, which is insufficient for the model. The variable equity asset is not a significant variable in this sample.

12 We kept the 23 most representative nationalities of the sample in terms of their capitalization.

13 Many authors have criticized Friedman’s (2005) thesis that “the world is flat”—among them Stiglitz (2006), who argued that national borders still matter for understanding the global economy, and Ghemawat (2007), who concluded that “the world isn’t flat.”