Mapping the Bonanza: Geographies of Mining Investment in an Era of Neoliberal Reform

Abstract

Geographic analyses of how national policies of economic liberalization influence global patterns of economic activity often draw their conclusions from studies of the paradigmatic sectors of manufacturing and, to a lesser extent, services. There is, by contrast, relatively little work examining how neoliberal policy reforms in the developing world may be driving changes in the geography of primary sector (i.e., extractive) activities at the global scale. This article presents and analyzes new data on direct investment in the international mining industry. It reports methods and results from a research project to systematically map and evaluate changes in the commodity mix and geographical spread of mining-related investment in the world economy since 1990. It confirms and quantifies what was hitherto anecdotal evidence of a geographic shift in investment during the 1990s away from mature targets toward a small number of “rising stars” in the developing world, following the adoption in many countries of neoliberal economic policies from the mid-1980s onward. However, the findings challenge conventional interpretations of this shift as an investment “bonanza” in the periphery and highlight how recent investment trends are highly specific in geographical scope, concentrated within a few commodities, and how the allocation of investment between established and emerging targets is variable over both time and space.

Key Words:

• investment
• mining
• globalization
• economic liberalization

Notes

1 Not all components of the primary sector have been overlooked. There is now a rich geographical literature on the restructuring of the global agro-food complex (see, e.g., 1 b25 b20 Arce and Marsden 1993; McMichael 1994; Goodman and Watts 1997 ).

2 FDI refers to equity investments made by a firm resident in one national economy in enterprises located in another national economy, where a purpose of investment is to gain an effective voice in the management of the enterprise ( 37 UNCTAD 1994 ).

3 Lest the inclusion of Mongolia be considered hyperbole, it is worth noting that Mongolia has become—in mining circles at least—“the Chile of Asia” by adopting in 1997 what is widely regarded as a “progressive” investment regime (from the point of view of inwardly investing mining firms). This provides a “market-based mechanism for access to, and maintenance of, mineral rights similar to that found in the leading Latin American jurisdictions” ( 28 Naito, Remy, and Williams 2001 , 14).

4 These were InfoMine ( http://www.infomine.com ), Raw Materials Group ( http://www.rmg.se ), Engineering and Mining Journal Annual Surveys, and Metals Economics Group's Mine Search ( http://www.metalseconomics.com ).

5 The MineSearch database classifies projects into seven categories of project status, ranging from “Raw Prospect” to “Production.” Data analyzed here were drawn from the top three categories—feasibility, preproduction, and production—because it is in these stages that significant capital expenditure occurs.

6 The decision to expense costs across a number of years, rather than allocate them to a single year, recognizes that large capital sums are rarely expended in a single year but are distributed throughout the development and commissioning phase of a mine project. A large greenfield copper mine, for example, may have capital costs in the region of $500 million to $1 billion and take between three and four years to develop into a working mine (the narrative histories in MineSearch enable a determination, for each project, of the time period between capital first being expended and project commissioning). In nearly all cases, capital flow was distributed normally; for a four-year project, the largest capital flows were assumed to be in years 2 and 3, with relatively smaller flows in years 1 and 4. Discussion with analysts familiar with the industry validated this decision to distribute costs.

7 Many of these downstream phases have their own geography. For example, the geography of alumina smelting (the most energy-intensive phase of aluminum production) is shaped by the availability of low-cost energy supplies.

8 The category Europe here also includes the former Soviet Union.

9 Of the two metals, copper dominates in Indonesia (70%) and Peru (68%), but is subsumed to gold in Papua New Guinea (42% versus 56%).

10 The profile for Australia (not shown) is an exception. It demonstrates a response much more like that of Chile and Peru, with investment flows tripling throughout the 1990s, followed by a slump in investment in the post-1999 period. In part this reflects the broad commodity mix of mining investment in Australia, but it is also reflective of policy initiatives adopted by the Australian government to attract overseas investment into the mining sector. In this regard, Australia has behaved less like the United States and Canada and much more like an emerging market country in seeking investment in its natural resource sectors.

11 It is no surprise to find that the geography of investment activity for specific mineral commodities is unevenly distributed at the global scale. Only a small proportion of countries around the world have significant reserves of copper or nickel, for example, so the number of potential targets for investment is limited by basic geology. It is considerably more interesting, however, to find that the geography of investment for specific mineral commodities is unevenly distributed even among those countries with significant reserves of those commodities. Gini coefficients—which calculate the gap between a hypothetical equal distribution of investment and the actual distribution and are expressed as an index between 0 and 1, where 0 indicates perfect equality and 1 indicates perfect inequality—provide one way of measuring this inequality. When calculated for all 207 countries worldwide, the Gini coefficient for copper is 0.96, reflecting investment's high degree of concentration in just a small fraction of these countries, most of which have no prospect of receiving investment. More meaningful measures are Gini coefficients calculated using only those 30 countries that actually received investment in copper (0.72) or using only those countries that rank among the top 10 in terms of copper reserves and represent 75% of global copper reserves (0.66).

12 These apparently “physical” criteria are also social products since they derive from earlier exploration activity and processes of knowledge accumulation.

13 It is also possible to use investment-reserve quotients for different time periods to identify how policies of liberalization have enabled individual countries to shift their relative position “up” or “down” over time. Space considerations prevent demonstration of this here.

14 Membership of the Organization for Economic Cooperation and Development (OECD) was used as a proxy for “developed.” Membership of the OECD evolved during the 1990s to include Mexico (1994), the Czech Republic (1995), Korea (1996), Hungary (1996), and Poland (1996)—countries that throw into sharp relief the limitations of binary classifications such as “developed” and “developing.” For a complete listing of OECD countries, see the OECD web-site at http://www.oecd.org .

*A Junior Faculty Development Grant from the University of Oklahoma supported initial stages of this research. An NSF Research Experience for Undergraduates (REU) Supplement Grant supported additional research assistance by April Rankin. I thank Julie Parker for her work developing and mapping the pilot dataset, Gerardo Castillo and Todd Fagin for their assistance extracting data from MineSearch and producing maps, and Joe Stoll for his help with graphics. Russell Carter, Graham Davis, Magnus Ericsson, Rob Krueger, and Gregory Theyel provided insightful discussion and/or helpful comments on a much earlier version of this article. Special thanks to Bakes Mitchell at the Metals Economics Group for his interest in this project and for continuing discussions on the geographies of mining investment. I would also like to thank three anonymous referees for their comments. The usual disclaimers apply.