POVERTY THE WORLD BANK REDUCTION AND ECONOMIC MANAGEMENT NETWORK (PREM) Economic Premise MAY 2012 • Number 82 JUN 010 • Numbe 18 69118 Reducing Distortions in International Commodity Markets Bernard Hoekman and Will Martin World commodity markets—and particularly the markets for agricultural commodities—remain highly distorted despite the wave of liberalization that has swept world trade since the 1980s. Commodity markets are distorted on both the export and the import sides, with serious implications for world prices and their volatility. Very few of the price distortions found in commodity markets can be justified on the grounds of dealing with market failures. Rather, most policies that affect com- modity prices are designed to transfer resources to favored groups by raising or lowering prices. Policies may target the level and/or the volatility of prices, and the pursuit of one type of policy objective may have unintended consequences in generat- ing further distortions. Moreover, some commodity markets are characterized by imperfect competition. Where monopolies or oligopolies in trade arise, either because of government regulation or through other barriers to entry, distortions may arise that call for application of antitrust laws and other forms of pro-competitive policy action. The negative spillovers for other countries that are generated by many developing countries over the last two decades, and in- national policies or the exercise of market power imply there is creasing competition (demand) for natural resources have significant potential to realize global welfare gains through in- changed the political economy landscape. The recent upsurge ternational cooperation and negotiation of disciplines on the in the use of export barriers suggests that the general mercantil- use of specific policies. The main multilateral institution that ist reluctance to restrict trade cannot be taken for granted when offers a framework for countries to pursue such cooperation is world food prices rise, or when there is a significant increase in the World Trade Organization (WTO). The WTO embodies a global demand for scarce natural resources (Frankel 2010). In set of rules that apply to the agricultural trade policies of mem- this situation, many governments may place a higher weight on ber states, but these address only a subset of the relevant poli- the welfare of consumers and downstream industries than on cies. Compared to the historical focus of the membership on the welfare of upstream producers (farmers, miners, and so negotiating agreements to improve access to markets for export- forth) when deciding on whether and how to use trade policies. ers, rules applying to import policies are more extensive and The use of export restrictions may increase as emerging mar- better developed than rules pertaining to export policies. In kets continue to experience high rates of economic growth, principle, the WTO’s framework of rules and disciplines can be which generates greater demand for food products and raw ma- applied to both types of policies—what is needed is for mem- terials. These recent developments suggest that a more compre- bers to be willing do so. hensive approach to negotiating rules of the game for policies Rising real incomes, the emergence of a multipolar world affecting commodity markets could potentially benefit a much economy resulting from the sustained economic growth in wider set of stakeholders than those that have so far driven 1 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK www.worldbank.org/economicpremise WTO negotiations. However, implementing a broader ap- help individual countries reduce domestic price volatility, they proach would require going beyond traditional trade policies to result in a serious collective-action problem. As more coun- encompass policies that affect investment in natural resource tries stabilize their domestic prices, world prices become more production/extraction and market competition. volatile, and the policy is completely ineffective if all countries use it to the same degree. Martin and Anderson (2012) found Agricultural and Commodity Market that almost half of the increase in the world rice price in Distortions 2005–8 could be explained by countries’ attempts to insulate Many agricultural commodities tend to be heavily protected in themselves from the primary shocks causing the world rice industrial countries for reasons such as: (i) food is typically a price rise. While some countries were relatively successful in small share of consumers’ expenditures in these countries; (ii) insulating themselves against the increases in world prices, do- the number of farmers tends to be small, making it relatively mestic prices in low-income countries in Africa rose almost as easy for them to coordinate to apply political pressure; and (iii) much as world prices did, suggesting that price volatility may farmers sell virtually all of their output and use substantial have been greater as a result of the insulating policies than it amounts of purchased inputs—increasing the leverage of output would have been otherwise. prices on their net returns. Conversely, historically, agricultural Governments also have a long history of intervening in products in developing countries have tended to be taxed be- markets for other natural resources, both renewable and nonre- cause: (i) food expenditures are frequently a large share of the newable. Imperfect competition, market power, and high levels income of most people; (ii) the number of farmers is large, mak- of concentration characterize some commodities markets and ing it hard for them to organize; (iii) urban consumers are a rela- may result in price distortions. Political uncertainty and risk tively small group, able to organize against rising food prices; that preclude efficient investment and generate inefficient and (iv) farmers are mainly subsistence oriented—selling only forms of trade can also generate important costs (Collier and part of their output and using few intermediate inputs so that Venables 2010; WTO 2010). In the case of natural resources, there is little leverage between food prices and their net returns. producers have focused much more on export restrictions de- A recent comprehensive multicountry study shows that signed either to lower domestic prices or to raise export prices— agricultural distortions in industrial countries remain large, including through the creation of producer cartels. Export re- but have declined from the high levels in the mid-1980s (Anderson 2009). In developing countries, the average Figure 1. Average Nominal Rates of Assistance to Agriculture, 1955–2004 rate of taxation on agriculture declined sharply, as shown 70 in figure 1, and has switched to positive assistance on av- 60 50 erage. The changes in these rates of assistance reflect a 40 sea-change in the pattern of agricultural distortions in percent 30 developing countries, perhaps related to the high rates of 20 economic growth in developing countries in the latter 10 period of the sample, and the sharp shift away from de- 0 -10 9 4 9 4 9 4 9 4 9 –4 pendence on exports of commodities toward greater reli- –5 –6 –6 –7 –7 –8 –8 –9 –9 00 55 60 65 70 75 80 85 90 95 -20 20 ance on exports of manufactures. In developing coun- 19 19 19 19 19 19 19 19 -30 19 tries’ agricultural sectors, there is a sharp difference HIC and ECA developing countries HIC and ECA, including decoupled payments between the rates of protection provided to import-com- peting agriculture and those provided to export-oriented Source: Five-year averages from Anderson (2009) and www.worldbank.org/agdistortions. Note: ECA = Europe and Central Asia; HIC = high-income countries. agriculture (figure 2). The almost complete elimination of taxation on agriculture has sharply reduced the cost of Figure 2. Nominal Rates of Assistance to Developing Countries’ Agriculture distortions on export-oriented agriculture in developing 50 countries. The rise in protection to import-competing ag- 30 riculture is likely to be particularly costly to the poor, who spend a large share of their income on food. 10 percent For a number of staple foods, governments intervene -10 4 9 59 4 19 9 4 9 4 9 –4 to reduce the volatility of domestic prices relative to –7 –7 –6 –6 –8 –8 –9 –9 00 – 70 75 55 60 65 80 85 90 95 20 world prices. Such measures have a long history and were 19 19 19 19 19 19 19 19 - 30 also used in high-income countries to stabilize domestic prices prior to the Uruguay Round multilateral trade ne- -50 gotiations. Insulating policies are heavily used for key sta- import competing exportables total ples such as rice and wheat. While these measures may Source: Five-year averages from Anderson (2009) and www.worldbank.org/agdistortions. 2 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK www.worldbank.org/economicpremise strictions are predominantly used by developing countries enhancing for the countries concerned, may lower welfare for (WTO 2010). Export taxes on natural resources account for the world as a whole by generating negative spillover effects for about one-third of all export taxes imposed—some 11 percent trading partners and inducing affected governments to imple- of world trade in natural resources is covered by such taxes, ment measures in turn to offset such effects, in the process gen- with timber, iron, copper, pearls, and gemstones among the erating further distortions and inefficiencies. most frequently affected.1 Countries may also use other instru- Rule-Making Implications and Priorities ments that have effects similar to those of export restrictions on international prices, such as efforts to agree on joint production Many of the policies that affect commodity markets are subject limitations (quotas), as in the case of the Organization for Pe- to multilateral rules under the WTO. But a number of the ma- troleum Exporting Countries (OPEC). In practice, most ef- jor policy instruments that are used by governments are not forts by producers to form a commodity cartel and improve subject to effective disciplines, either because countries have their terms of trade by restricting supply have not been success- not been willing to make commitments—for example, limits on ful (OPEC is an exception). the level of export taxes—or because there are no agreed on Finally, state ownership and control of entities that pro- rules. In practice, domestic policy measures may be used that duce or trade natural resources may affect the operation of the have an equivalent effect to trade policies that are subject to dis- markets concerned and distort international competition as a ciplines—for example, nondiscriminatory excise taxes on a good result of de facto subsidization—soft budget constraints, privi- for which there is little national production or output quotas leged access to credit, and so forth—or market power effects. by exporting countries. The multilateral trading system has fo- State-owned enterprises (SOEs) are particularly prevalent in cused primarily on lowering import barriers and making them the energy and mining sectors—often ranking among the largest less variable. nonfinancial multinational enterprises.2 The policy and negotiating agenda on import protection is Protection rates for imports of nonagricultural natural re- well understood and is already squarely on the table in the sources are generally low because of pressure from user indus- WTO. Making progress in further disciplining the scope to use tries. However, importing countries confronting exporters of import barriers is important—the estimated welfare gains from resources with market power and thus the ability to affect lowering applied levels of protection and bringing down tariff world prices may seek to use import tariffs or, more frequently, bindings are significant. But extending the effort to agree on nondiscriminatory excise taxes to lower demand and shift rents disciplines on export restrictions is equally important, not least away from supplying nations. In addition, tariff escalation is because the greater freedom of countries to use export restric- often used to increase the effective rate of assistance to domes- tions is likely to have a direct bearing on the willingness of many tic processing industries. importing countries to accept greater disciplines on their im- Commodity value chains are often characterized by im- port policies. perfect competitive market structures. Domestic processors, WTO disciplines are much less comprehensive on export for example, may have a degree of market power, as may suppli- measures and policies that reduce output or restrict supply. Ar- ers of inputs such as seeds, fertilizers, or chemicals. Economic ticle XI of the General Agreement on Tariffs and Trade (GATT) effects similar to those that are generated by trade policies can prohibits the use of quantitative restrictions, whether on im- arise from the (non)application of antitrust law. One example ports or on exports, but permits temporary quantitative trade is an exemption under competition law for behavior by nation- restrictions to prevent critical shortages of food or other goods. al firms on export markets—such as export cartels—that would It leaves open the possibility of a country imposing production otherwise be illegal, as long as the actions do not have negative quotas or using excise taxes that primarily affect imports. The effects on consumers in the home market. Another issue that is use of export taxes is also unconstrained unless a country has attracting more attention in policy circles is concentration (a made specific commitments in this regard—the counterpart of dominant position) at specific segments of global value chains. import tariff ceiling bindings—which most WTO members The concern is that monopoly power of input providers and/or have not. Article XVII GATT requires state-trading enterprises monopsony power on the part of buyers (large trading compa- to abide by the WTO’s nondiscrimination rules (national treat- nies or retailers) may lower domestic farm gate prices and/or ment and most favored nation), but does not impose other con- result in retail prices that are higher than they would be if the straints on the behavior of SOEs (Hoekman and Kosticki relevant markets were characterized by greater competition. 2009). In sum, a wide mix of policy instruments may influence There has been a recent upsurge in the use of export barri- the level and volatility of domestic prices for commodities. Na- ers in response to rising world prices of food staples and strong tional policies may affect international prices, either indirectly demand for scarce natural resources (Datt, Hoekman, and Ma- or directly, especially if the countries concerned are large sup- louche 2011). Export barriers are often used for political econ- pliers or importers. The pursuit of such policies, even if welfare omy reasons—as, for example, to drive down the price of a com- 3 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK www.worldbank.org/economicpremise modity used as an input by a more politically powerful Imperfect competition, market power, and high levels of “preferred� sector. If a country can affect its terms of trade, ex- concentration characterize some commodities markets and port restraints can make economic sense for that country. An may result in price distortions. At the national level, competi- important difference between political economy and terms-of- tion policy and/or regulation is the appropriate instrument to trade motivated trade interventions is that in the case of the address market behavior that may result in distorted pricing. former, the probability that intervention raises national welfare From a global rule-making perspective, the question is what can is lower.3 Whatever the motivation, the interventions are likely international cooperation do to address the cross-border nega- to generate negative spillovers for other countries, providing an tive spillovers that are created by the behavior of firms located incentive to cooperate and negotiate reciprocally binding disci- in a foreign country (or, in the case of multinationals, that are plines on the use of specific policies. subject to multiple jurisdictions). “Competition issues� may Fears of inadequate access to supplies by resource-scarce arise in the operation of both food and nonfood commodity countries and of inappropriate exploitation in resource-rich re- markets, but are more likely in the case of natural resources be- gions have significant potential to generate trade conflicts and cause production and/or exports of commodities concerned create negative spillovers for the world as a whole (WTO 2010). often involve a relatively small number of large firms (many of Responses by importers to actions by exporters to restrict sup- which may also have strong links to the state). Market power ply—whether the government does so directly through a tax or and oligopoly have a number of implications, including possi- other type of policy, or allows firms based in its jurisdiction to ble foreclosure of markets for more efficient foreign producers. exploit their market power in foreign markets (through an ex- Competition policy was one of the three so-called Singa- port cartel, for example)—may result in some of the rents being pore issues suggested for negotiation at the 1996 WTO Minis- shifted from exporters to importers, but the net result for terial Meeting that eventually were taken off the table at the world welfare is negative. 2003 Cancun ministerial. One reason was that the focus of Conceptually, the agenda for cooperation in this area is discussions was not primarily on the negative spillovers associ- rather straightforward and revolves around agreeing on a ban ated with competition law enforcement (or nonenforcement). on export quotas and on binding commitments on export taxes Instead, most of the deliberations revolved around the benefits and equivalent disciplines for export cartels. Given that coun- of national competition policy per se—something that can and tries that have the ability to set prices have an incentive to do so, has been autonomously implemented by countries. Arguably, affected trading partners will have to be willing to engage in any effort to negotiate rules of a competition policy nature quid pro quo negotiations and offer concessions to the coun- must address situations that involve private sector behavior tries that currently benefit from being able to impose export that gives rise to cross-border negative externalities—antitrust restrictions. In principle, this is of course exactly what the exemptions for export cartels and international cartels being WTO is set up to do. The challenge is to design a negotiating obvious examples. A number of major antitrust cases in recent agenda out of which win-win deals can be constructed. Despite years against global cartels connected with the food industry the difficulty in concluding the Doha Round negotiations— have illustrated the importance of active enforcement and in- which in part is arguably a result of an agenda that does not of- ternational cooperation between competition authorities.4 fer enough in the way of potential win-win deals—the history of In many of the areas that can potentially give rise to compe- negotiations under the GATT indicates that this is a challenge tition concerns, there is significant uncertainty/ambiguity re- that members should be able to meet. garding whether a practice, level of market concentration, pre- As has been argued by Collier and Venables (2010), multi- vailing market structure, and so forth should be of concern, that lateral efforts to agree on policy disciplines that will reduce dis- is, whether they can affect price levels or generate excessive vola- tortions of international commodity markets need to span not tility. The same is true for the operation of SOEs. This suggests a only export and import trade policy and domestic substitutes first priority is to compile much better data and to undertake a for trade policy instruments, but also policies that determine concerted effort to identify negative cross-border spillovers and the extent to which the most efficient firms are permitted ac- analyze whether these should be accepted (as in the case of coop- cess to resources. Uncertainty regarding the conditions of com- eration between countries for nonrenewable natural resources petition for contracts and leases and the future operating and such as oil). Thus, greater transparency and analysis should be tax regime that will affect extraction activities and rates of re- part of any forward-looking work program on commodity mar- turn on investment generate distortions that may result in re- kets and natural resources in the WTO. ductions in supply, excess costs and inefficiencies, and lower Conclusion revenues for host governments. Rule-making efforts need to include issues that affect access to resources through foreign Extending the multilateral trading regime to cover the main direct investment (FDI) and contracting transparency (govern- policy-induced distortions of global commodity markets is a ment procurement). complex endeavor. However, the case for a concerted focus on 4 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK www.worldbank.org/economicpremise the set of policies that distort agricultural and natural resource 3. Although from a global point of view, improvements in the markets is strong. Distorted price levels and excess price volatil- terms of trade are purely transfers, and removal of these barri- ity are detrimental to both producers and consumers. Agree- ers generally increases world income by reducing the efficiency ment to discipline the use of policies that generate distortions costs of these distortions. would benefit all countries. To broker such an agreement, the 4. See the contributions in Evenett and Jenny (2012). Regarding focus must be comprehensive and address all of the major poli- export cartels, a distinction should be made between cartels that cy areas that affect the operation of international commodity involve states and cartels of private firms. The former may be an markets, not just a subset. efficient mechanism if the product concerned is a nonrenewable There has been much analysis and discussion of why the natural resource (Collier and Venables 2010). In any event, it is Doha Round negotiations have proven so difficult to con- unlikely that any effort to declare such arrangements illegal will clude successfully. One argument is that the negotiations be successful because for many of the producing countries the (and thus prospective agreements) were not structured to natural resource represents a major source of national wealth. make them relevant enough to international business and References the concerns of citizens. In the case of commodities, the Doha negotiations have focused primarily on agricultural Anderson, K. 2009. Distortions to Agricultural Incentives: A Global products and only a subset of the relevant policies—subsidies Perspective. Washington DC: World Bank. and import protection. This narrow focus significantly re- Anderson, K., and W. Martin, eds. 2009. Distortions to Agricultural Incentives in Asia. Washington DC: World Bank. duces the potential global welfare payoffs from any agree- Collier, P., and T. Venables. 2010. “International Rules for Trade in ment and does not do enough to address the concerns of Natural Resources.� Journal of Globalization and Development countries that rely heavily on imports of food and natural 1(8). http://www.bepress.com/jgd/. resources. Given the social importance of food prices and ac- Datt, M., B. Hoekman, and M. Malouche. 2011. “Taking Stock of cess to food and the impact of increasing demand and com- Trade Protectionism since 2008.� Economic Premise 72, World petition for other natural resources, renewable and nonre- Bank, Washington, DC. newable, a concerted multilateral effort to agree to disci- Evenett, S., and F. Jenny, eds. 2012. Trade, Competition, and the Pric- plines on the use of the major policies that distort global ing of Commodities. London: CEPR. commodity markets should be a priority. The international Frankel, J. 2010. “Food Security: Export Controls Are Not the Cure for Grain Price Volatility But the Cause.� Blog post. http://con- community is now beginning to focus on the issues involved tent ksg harvard.edu/blog/jeff_frankels_weblog/2010/08/23/ in addressing these challenges (G-20 2012), but rigorous ef- food-security-export-controls-are-not-the-cure-for-grain-price- fort and follow through will be needed to implement coop- volatility-but-the-cause/ erative solutions for these problems. G-20. 2012. “Mexico’s Presidency of the G-20.� Discussion Paper. http://www.g20.utoronto.ca/2012/2012-loscabos-disc-en.pdf Acknowledgment Hoekman, B., and M. Kostecki. 2009. The Political Economy of the This note draws on Hoekman and Martin (2012). World Trading System. Oxford: Oxford University Press. Hoekman, B., and W. Martin. 2012. “Reducing Distortions in About the Authors International Commodity Markets: An Agenda for Multilateral Cooperation.� In Trade, Competition, and the Pricing of Commodi- Bernard Hoekman is Director of the International Trade Depart- ties, ed. S. Evenett and F. Jenny. London: CEPR. ment at the World Bank. Will Martin is Research Manager, Agri- Martin, W., and K. Anderson. 2012. “Export Restrictions and Price culture and Rural Development, Development Research Group in Insulation during Commodity Price Booms.� American Journal the World Bank. of Agricultural Economics 94(2): 422–27. Sauvant, K., and J. Strauss. 2012. “State-controlled Entities Control Notes Nearly US$ 2 Trillion in Foreign Assets.� Columbia FDI Per- spectives No. 64, April. http://academiccommons.columbia. 1. Export restrictions on natural resource products accounted edu/catalog/ac:146184 for one-third of the 7,328 notified export restrictions in the WTO (World Trade Organization). 2008. “Revised Draft Modali- time period covered by WTO (2010). ties for Agriculture.� Committee on Agriculture, Special Ses- 2. Forty-nine of the 200 largest nonfinancial multinationals are sion, TN/AG/W/4/Rev. 12, May 19, Geneva. SOEs, defined as firms in which the government stake is at least ———. 2010. World Trade Report 2010: Trade in Natural Resources. 10 percent (Sauvant and Strauss 2012). Geneva. The Economic Premise note series is intended to summarize good practices and key policy findings on topics related to economic policy. They are produced by the Poverty Reduction and Economic Management (PREM) Network Vice-Presidency of the World Bank. The views expressed here are those of the authors and do not necessarily reflect those of the World Bank. The notes are available at: www.worldbank.org/economicpremise. 5 POVERTY REDUCTION AND ECONOMIC MANAGEMENT (PREM) NETWORK www.worldbank.org/economicpremise