Managing the Resource Curse and Promoting Stabilization through the Rentier State Framework

By
Timor-Leste oil
Managing the Resource Curse and Promoting Stabilization through the Rentier State Framework - Ryan J. Connelly

Abstract

Despite much discussion of the resource curse in recent years, a consensus on a development model for resource-rich states has not emerged. Such states have tremendous economic potential but have tended to under perform when compared to resource-poor states due to a higher propensity to resort to civil conflicts. This article argues that the key to escaping the resource curse is through the application of a rentier state development mode, which creates political stability while increasing GDP per capita through (1) increased military expenditures, (2) private goods transfers to elites, (3) redistributive economic programs, and (4) security guarantees from foreign states.

Introduction

The resource curse is the common name for the observation that nations largely dependent upon natural resource have tended to underperform relative to similar countries that lack a substantial natural resource base. A particularly unsavory feature of the resource curse is that this underperformance is often due to the higher propensity for violent civil conflict in these countries.

Natural resources have different exploitation potential. The political economies of nations dependent upon diamonds, timber, and oil are drastically different. This paper will focus on the evolution of oil-rich1 states for two reasons. First, the presence of oil reserves is the most accepted as being “a significant predictor of the onset of civil war.”2 Second, 100 percent of civil war movements in oil rich states are secessionist, in stark contrast to only 68 percent when the resource is not oil.3 Although the focus will be on oil-rich states, the argument applies to the exploitation of other hydrocarbon resources, and with minor augmentation should apply to other non-renewable resources that exhibit lower exploitation potential.4

The resource curse is at odds with the rentier state framework, which argues that oil exploitation should be a source of political stability. According to the rentier state framework, autocratic regimes use oil revenues to “maintain internal peace by combining a huge security apparatus with generous distributional policies.”5 Since government revenues come from export markets, these states have no need to tax their populace. They then use distributional policies that create a largely enfranchised populace, which, combined with small winning coalitions, create very stable political systems. Thus a rentier state is able to maintain political stability in spite of, and, in fact, because of its natural resource wealth, as long as it is able to generate sufficient income.

Current development policy for post-conflict states relies on extracting policy concessions from regimes in exchange for development assistance, which tends to include a democratization agenda. This strategy does not work for resource-rich countries, which have less incentive to make policy concessions since there is little problem with capital generation. The West is therefore no longer able to effectively sanction resource-rich nations to force regime change, especially because China has made large investments in a variety of resource-rich countries in Africa and Central Asia that are “shunned as pariahs by...the rest of the world for human rights abuses.”6

In this paper, I will argue that the rentier state framework proposes a more suitable development model for post-conflict oil-rich countries than do current models. After expanding on the resource curse and the rentier state theories, I will show how a post-conflict oil-rich country exhibits the same characteristics as a rentier state in all but one area: low exploitation per capita. This lends credence to the argument for a development model that increases stability through improving exploitation per capita. Lastly, I will present the post-civil war stabilization of Angola as a successful example of rentier state development.

The Resource Curse and Civil Conflict: Greed, Grievance and Stagnation

The key issue facing resource-rich countries is the elevated threat of civil conflict.7 A large portion of the observed economic underperformance of resource-rich states as a group is due to a higher propensity to resort to civil conflicts. Encouraging political stability is a mechanism for improved economic performance.8

A brief introduction to one long-term economic issue faced by resource-dependent economies, the so-called Dutch Disease, is appropriate before framing political stabilization recommendations. Dutch Disease occurs when external demand for primary commodities leads to exchange rate appreciation, which reduces competitiveness in other sectors of the economy. This discourages diversification and enhances reliance on primary commodities, a potential source of instability since commodity prices are set by volatile international markets. Because post-conflict states tend to have very low agricultural and industrial output, they are particularly susceptible to over-reliance on primary commodity exports, and are therefore vulnerable to Dutch Disease. However, targeted policies, including stability funds and directed infrastructure investments, can mitigate these negative economic impacts.

Such long-term economic effects are a second-order concern for post-conflict states. The first concern is creating a politically stable environment by reducing the potential for civil conflicts, which involves exploring their sources. Paul Collier and Anke Hoeffler have developed a model that proposes three factors that significantly impact the probability of civil conflict: income per capita, gross domestic product (GDP) growth rates, and dependency on primary commodity exports.9 It is not debated that high income per capita and high GDP growth rates reduce potential for civil conflict: the opportunity costs for resorting to civil war are significantly higher in rich countries than in poor ones. However, a schism has emerged among academics as to why high resource dependency leads to increased potential for civil conflicts. Table 1 summarizes the three common explanations: greed, grievance, and weak institutional structures.

The greed mechanism is an economic argument positing that resource control is sought for private gains. The grievance argument insists that identity politics lie at the heart of the conflict and that “leaders simply seize on the resource issue to build support for their movement.”10 though the grievance argument does not deny the importance of resource exploitation as a capital source for rebel financing, it insists that it is merely a means to an end. Other scholars take an institutional approach, arguing that dependence on rents for state revenue leads to weak organizational structures, which increases the potential for civil conflicts.



These three mechanisms are inextricably linked. For instance, weak state institutions, which lack appropriate mechanisms to mediate conflicts between distinct identity groups, can lead to irresolvable grievances over wealth distribution. On occasion, these grievances will result in open conflict, which can lead to either a violent suppression by the State or a civil war. In the analysis of any conflict, all three of these mechanisms can provide insights.

In the case of oil-rich states, however, there is a strong argument for the greed mechanism as the driving force towards political conflict. The key difference between oil-rich states and non-resource rich states is the presence of a low-lootability resource of significant economic value. Exploitation of low-lootability resources requires attracting infrastructure capital, which requires stable control over the territory. Collier and Hoeffler argue that, in order to gain economic advantage over a resource located on their territory within a state, “secessionist political communities invent themselves when part of the population perceives secession to be economically advantageous.”12 this is especially relevant for oil-rich countries, since all prior civil wars have had a secessionist element.

By definition, a secessionist movement needs a defined territory. For a group to seek secession, this territory must be significantly richer than a separate defined territory. Otherwise, individuals in the secessionist movement would be willfully attempting to lower their standard of living. While the ideological leadership of the movement may be willing to make such sacrifices, the support for a mass movement “will come not from people who are passionately committed” but rather from people who listen to rhetoric that promises improved economic conditions.13

If the central government exerts strong control over the territory where the resources reside, a secessionist movement is unlikely. Strength in this case is a function of the magnitude of coercive forces in that region, their loyalty, and the distance from resource base to the seat of power. Conversely, a weak central government with a weak coercive apparatus, or oil fields far from the seat of government, makes secession more likely.

If fields are currently producing, a secessionist group would be able to sell oil on international markets, potentially increasing arms purchases and recruitment while also depriving the State of revenues. To prevent this scenario, a political leader would drastically overinvest in coercive institutions to far higher levels than if the country lacked the resource base. This is the intuitive logic of rentier state formation.

The Rentier State Framework

With large resource endowments and proper motivation, a central government should be able to exercise effective territorial control and maintain political stability. It is therefore difficult to account for the prevalence of civil wars in countries that have such vast capital resources to buy political support and fund coercive institutions. This is the fundamental paradox of the resource curse.

The rentier state framework has the opposite conclusion: natural resource wealth, once properly scaled up, provides stability. A rentier state expands export capacity until the bulk of state income derives from rents accrued from exports of primary commodities rather than taxation.14 revenues are then used to buy peace through “patronage, large-scale distributive policies and effective repression.”15 for individuals, low or zero taxation combined with distributive spending policies leads to an economic policy in which, though some groups may be favored with a higher proportion of revenues, no one is directly harmed. In general, public policies will “be seen as benefiting everyone.”16

An important part of the structure of a rentier state is the desire to “avoid having a clearly defined constituency.”17 the rationale for this is readily gleaned from selectorate theory.18 Political leadership has the greatest chance for survival when the minimum winning coalition is small and the selectorate large—that is, when all groups are enfranchised but few key actors are necessary to maintain political office. This is an autocratic form of government.

Autocratic regimes are particularly stable for the political leadership because the loyalty norm among the winning coalition is strong. Any member of the vast selectorate could replace any member of the small winning coalition, which means that the opportunity cost for defection from the leader is very high within the coalition. As such, “when the government is not already autocratic, incumbent leaders have an interest in making the system more autocratic...[which] is achieved by reducing the size of the winning coalition and expanding the size of the selectorate.”19

Matthias Basedau and Jann Lay attempt to resolve the contradictions between the resource curse and rentier peace arguments. They observe that “almost all studies” on the subject conducted since 1995 use primary commodity exports as a percent of GDP as the measure for resource wealth.20 this statistic only tells half of the story, however:

Dependence means that rents from resources are the most important source of income relative to other value-adding activities, while abundance or wealth refers to the absolute amount of resource rents available in per capita terms. It can be easily illustrated that these two variables may differ substantially. Nigeria and Saudi Arabia, for instance, were almost equally dependent on oil exports in 2002 —oil exports accounted for 38.9% and 38.5% of GDP, respectively. Yet, if the governments had decided to pay out the proceeds from oil exports to their citizens, Nigerians would have been given a mere US$ 140 while Saudi Arabia’s citizens would have earned US$ 2,715.21

This distinction is crucial: higher dependency ratios should only have a destabilizing impact during periods of market volatility or in the long run, which is why economic diversification is many experts’ primary recommendation. However, wealth per capita has a much more intuitive effect in politically unstable areas. In Nigeria, the central government has much less money on a per capita basis, which makes it difficult to sustain an armed forces presence large enough to maintain stability.22

The authors tested the rentier stability hypothesis with a data set that included 27 net oil-exporting countries with similar average primary commodity exports to GDP ratios. This was done in order to isolate the effects of resource dependence while analyzing the effects of production per capita. States were separated into groups based on whether annual per-capita production was above or below 10 tonnes (25 barrels per person per day).23 They then ran the data against UCDP/PRIO’s Armed Conflict Dataset24 and determined the results reproduced below:

Above the threshold of 10 tonnes of production per capita per annum, civil conflict entirely disappeared during the observation period. On the other hand, only three states with oil production below this level managed to avoid civil conflict.

Using the same data, the authors proceeded to test for causal mechanisms. They detected causal mechanisms for high spending on coercive institutions, large-scale distributional programs and the permanent presence of foreign troops: 7 out of the 11 states producing above the threshold employed all three of these policies simultaneously. These states were also found to have “more effective state institutions.”26 key findings are summarized in table 3 on the following page.

Increased cash flow allowed the central government to solidify its position through increased military expenditures and the implementation of large-scale distribution policies. The uptick in production levels also encouraged a foreign military presence. Though not part of the classic rentier state framework, a foreign military presence is a logical strategic move by Western governments looking to assure access to oil. This benefits the host country, since the mere presence of foreign troops has been shown empirically to significantly reduce the potential for civil conflict.27 Although this study failed to find a corruption mechanism, a separate study from 2009 concluded that “strategic use of public resources for off-budget and selective accommodation of private interests might reduce the risk of violent challenges to state authority in oil-rich states.”28

The downside is that rentier states tend to be autocratic (Norway is the only democracy in the sample). However, peaceful states in this sample were found to be no more repressive than the average of the data set. Another recent study, conducted over a larger time horizon, draws the even more optimistic conclusion that “increases in natural resource income are associated with increases in democracy... [particularly] among countries that had lower per capita incomes before they discovered resources.30

Most current policy recommendations for escaping the resource curse focus on economic diversification. This is the wrong conclusion drawn from a focus on the wrong variable: primary commodity dependence in GDP terms rather than production per capita. Attaining a high level of production per capita increases political stability and economic prospects, provided that additional funds are used to increase military spending, buy loyalty from key supporters, create large-scale distribution programs, and encourage large powers to take a security interest in the preservation of the regime.

The Angola Model: Developing a Rentier State

Every year that GDP per capita remains weak is another year with elevated risks for civil conflict. Every self-interested political leader wants to maintain power and avoid civil war, which is best accomplished by quickly increasing GDP per capita. The transition to a rentier state quickly increases revenues and requires fewer political concessions than other development models. Thus, the logical political economy that will develop in a post-conflict oil-rich state is a rentier state.

For the state, oil resources have no value until they are able to generate cash flow to the government—that is, until the fields are developed and oil is being sold to the open market. For this, the state needs three things: development capital, technical expertise, and interim cash flow to satisfy any expenditure needed to maintain political stability.

One model for rentier state development is the “Angola Model.” this model relies on attracting foreign investment through the sale of long-term oil leases in exchange for infrastructure development and immediate capital. A recent joint study by the UN office of the Special Adviser on Africa and the NEPAD-OECD Africa Investment Initiative described the model as follows:

Under the Angola model, the recipient country receives a loan from...[the export-Import Bank of China]; the government then awards a contract for infrastructure projects to a Chinese firm, while also giving rights for extraction of its natural resources to a Chinese company as repayment for the loan...In 2004, China extended a US$ 2 billion credit line to Angola for the development of its infrastructure, which had been destroyed during the civil war. As repayment for the loan, Angola agreed to supply China with 10,000 barrels of crude oil a day. The infrastructure projects in Angola included electricity generation and transmission, rehabilitation of power lines, rehabilitation of Luanda railway; construction of ring roads; telecomm expansion; water; and some public works projects.31

From 1975 to 2002, Angola was ravaged by civil war that was triggered by the withdrawal its former colonial power, Portugal. Cold War logic induced developed nations to view the fate of this nation in a broader security context. A number of smaller groups participated in the conflict, though it centered on the pro-communist People’s Movement for the liberation of Angola (MPLA), backed by the USSR and Cuba, and the anti-communist national union for the total Independence of Angola (UNITA), backed by Western powers and South Africa.

For the duration of the conflict, the MPLA controlled the north of the country, including Cabinda province, where most developed Angolan oil sits. Production and sales continued throughout the conflict, providing stable revenue to the MPLA and even attracting some foreign investment.32 the long history and scale of oil rents enabled the president to fund the military while sustaining “a clientele beyond the military apparatus” and thereby diversifying the winning coalition beyond the coercive institutions alone.33 Distribution of rents was targeted at the presidential entourage, high-ranking civil servants, army officers, and politicians in the form of direct cash distributions, overseas scholarships, and “subsidies on electricity, municipal water, air transport and housing...to a privileged minority within the population.”34

Meanwhile, UNITA was reliant on southern diamond mines for its capital needs. There are two types of diamonds available in Angola: alluvial and kimberlite. Alluvial diamonds require little technical expertise for extraction; production is labor intensive. Kimberlite diamonds would have required significant capital-intensive development before exploitation, which was not available, in stark contrast to the oil sector, which received foreign capital flows. Unlike in the case of oil, foreign governments did not view access to kimberlite diamonds as an important enough natural resource to encourage direct investment.

The MPLA, on the other hand, was careful to maintain foreign involvement in the oil industry at all times during the conflict, even using Cuban troops to protect the US Gulf Oil Company from nationalization.35 In the mid-1990s, this strategy finally bore fruit; in the post-Cold War environment, the West saw its strategic interest better served by protecting access to Angolan oil. The shift in coalition politics played out in a series of un Security Council resolutions that put UNITA at fault for the failure of recent peace agreements and imposed an arms embargo in 1993, followed by sanctions on diamonds in 1998. By the end of 1999, 80 percent of UNITA’s military capacity was destroyed, and the group demilitarized in February 2002.36

After the civil war, the Angolan state needed to increase revenues in order to assume control over the power vacuum left by the fall of UNITA. China provided capital in the form of loans and a credit facility in exchange for infrastructure development and oil lease contracts. As evidenced by Chart 137 on the following page, the increased oil flow stemming from Chinese infrastructure development swiftly translated into economic returns; Angola has almost doubled GDP per capita since 2002, even when taking into account the global recession and the decline in global oil prices in 2008-09.

There has been no sign of a political opening. The regime is autocratic, and MPLA has dominance over Parliament, the Constitutional Commission and state media.38 Presidential elections have been repeatedly delayed, and there are troubling reports of human rights abuses, especially in Cabinda province. Yet nationwide, basic development indicators have shown improvement. For instance, the un resident Coordinator in Angola recently stated that “poverty levels in Angola, measured in monetary income, decreased from 63 percent in 2002 to 38 percent in 2009.”39 A variety of government projects, mostly focused on increasing agricultural output, have begun in recent years. They are partly funded by the Angolan Development Bank, to which 3 percent of annual revenues and 2 percent of annual diamond revenues are funneled.40

The Angola example is a “hands-off” development model: China provides the funding and expertise needed to increase GDP per capita and leaves Angola responsible for its own internal politics. Since the civil war ended with a total victory by the MPLA, it has sought to solidify control. The rentier state framework identifies the mechanisms to do so: increased coercive capability to suppress secessionist movements, implementation of a redistribution economic program to increase the size of the selectorate, and distribution of private goods to members of the winning coalition to increase the loyalty norm. All of these are present in Angola.

The international community can play a larger role in an extension of the Angola Model. Many developing states lack access to necessary infrastructure capital and the protection of foreign security umbrellas that tend to develop with resource exploitation. One potential project model from the World Bank, the Chad-Cameroon pipeline, offered Chad infrastructure funds in order to build an expensive pipeline across Cameroon to facilitate development of Chadian oil fields. In exchange, the World Bank stipulated that certain funds be set aside for development spending. It in turn suspended the project in 2006 after the Chadian government changed its oil management laws, allowing these funds to be spent on security forces “to quell increasing unrest in the country.”41 this is a logical move from the political leader of a rentier state: to allocate resources to subdue uprisings before redistributing income to the broad selectorate. When the World Bank pulled out in 2008, the government paid off the remaining $65.7 million of outstanding loans in cash. Retraction of international capital is simply not a credible threat to a rentier state after initial oil field development is complete.42

Conclusion

Policy practitioners and development professionals lack a conceptual model for understanding political and economic transitions in countries that have now been identified as resource curse candidates. Special characteristics specific to oil-rich countries, linked to technical and capital demands in oil field development, have resulted in the empirical observation that civil conflicts have always been characterized by secessionist movement by groups living on oil-rich territories within the country. There is strong evidence that secessionist movements are mostly motivated by the potential for increased personal enrichment, although identity and organizational issues may contribute. In any case, the response of political leadership to a secessionist movement will be the same: suppression of the movement in order to preserve territorial integrity and access to capital.

The rentier state framework provides a good fit for selectorate theory’s lessons on the logical moves by political leadership in post-conflict states. For a leader seeking to maintain power, autocracy is the most stable method available, and the seemingly limitless funds available from oil exploitation provide the capital necessary to appease the winning coalition, fund the military, and increase the size and loyalty of the selectorate. But in order to pay for stabilization, resource exploitation needs to progress beyond a sufficient per capita level.

Recent studies have questioned central tenants of the resource curse thesis. Corrupt practices, such as patronage and private transfers, may not be harmful to for long-term development, as they serve to maintain stability. Similarly, it is unclear whether resource dependence actually promotes autocratic behavior. The only undisputed negative consequence of the resource curse, Dutch Disease, is manageable. Given the success of rentier states in avoiding devastating civil conflicts and escaping the resource curse, the cost-benefit ratio to development policies implementing rentier states appears significantly stronger than current development models.

Angola has thus far been a success story and thus a logical fit for the post-war political economic development in an oil-rich state. The international community can use this model when structuring future developmental programs in resource-rich economies. However, due to the time-consistency problem after initial development is complete, international lenders must be aware that any attempt to extract political concessions are limited by the nature of the development process.

Notes & References

  1. “Oil-rich” refers to natural resource endowments and not production figures. 
  2. Hanne Fjelde, “Buying Peace? Oil Wealth, Corruption and Civil War, 1985-99,” Journal of Peace Research 46, no. 2 (2009): 199-218. 
  3. Ian Bannon and Paul Collier, “Natural Resources and Conflict: What We Can Do,” in Natural Resources and Violent Conflicts: Options and Actions, ed. Ian Bannon and Paul Collier, (Washington, DC: World Bank, 2003), 1-16. 
  4. This is sometimes referred to as lootability, the ease with which a resource can be exploited. This is a function of price-to-weight or price-to-volume ratios, technical sophistication required for extraction, transportation requirements, and end-market availability. 
  5. Matthias Basedau and Jann Lay, “Resource Curse or Rentier Peace? The Ambiguous Effects of Oil Wealth and Oil Dependence on Violent Conflict,” Journal of Peace Research 46, no. 6 (2009): 757-776. 
  6. Peter S. Goodman, “CNOOC Announces $2.3B Nigeria Investment,” Washington Post, January 9, 2006, accessed December 12, 2010, http://www.washingtonpost.com/wpdyn/content/article/2006/01/09/Ar2006010901524.html
  7. Michael Ross, “The Natural Resource Curse: How Wealth Can Make You Poor,” in Natural Resources and Violent Conflicts: Options and Actions, ed. Ian Bannon and Paul Collier, (Washington, DC: World Bank, 2003), 17-42. 
  8. Paul Collier, Wars, Guns and Votes (London: Vintage, 2010). 
  9. Bannon and Collier, “Natural Resources and Conflict,” 1-16.
  10. Ibid., 1-16. 
  11. I. William Zartman. “Need, Creed, and Greed in Intrastate Conflict,” in Rethinking the Economics of War: The Intersection of Need, Creed, and Greed, ed. Cynthia J. Arnson and I. William Zartman, (Washington, DC: Woodrow Wilson Center, 2005), 256-284. 
  12. Paul Collier and Anke Hoeffler, “The Political Economy of Secession,” (Washington, DC: World Bank, June 30, 2002), http://www.worldbank.org/ogmc/pdfds/Paul-Collier.pdf.
  13. Ibid. 
  14. Giacomo Luciani, “Allocation vs. Production States: A Theoretical Framework,” in The Rentier State, ed. Hazem Beblawi and Giacomo Luciani (New York: Croom Helm, 1987), 63-82. 
  15. Basedau and Lay, “Resource Curse or Rentier Peace,” 757-776. 
  16. Luciani, “Allocation vs. Production,” 63-82. 
  17. Ibid., 63-82. 
  18. Bruce Bueno de Mesquita, Alastair Smith, Randolph Siverson, and James D. Morrow, The Logic of Political Survival (Cambridge, MA & London: MIT Press, 2003). 
  19. Ibid. 
  20. Basedau and Lay, “Resource Curse or Rentier Peace?,” 757-776. 
  21. Ibid., 757-776. 
  22. James T. Quinlivan, “The Painful Arithmetic of Stability Operations,” Rand Review, Summer 2003. 
  23. Actual revenue generated from oil exports is unavailable, explaining the choice by the authors to choose volumetric rather than notional measurements. Further study is necessary to discover how much income in PPP per capita is the identifiable threshold.
  24. This data, from the Uppsala Conflict Data Program (UCDP) at the Department of Peace and Conflict Research, Uppsala University and Centre for the Study of Civil War at the International Peace Research Institute, Oslo (PRIO) is available at http://www.prio.no/CSCW/Datasets/Armed-Conflict
  25. Basedau and Lay, “Resource Curse or Rentier Peace?,” 757-776. 
  26. Ibid., 757-776. 
  27. Fjelde, “Buying Peace?,” 199-218. 
  28. Collier, Wars, Guns, and Votes
  29. Basedau and Lay, “Resource Curse or Rentier Peace?,” 757-776. 
  30. Stephen Haber and Victor Menaldo, “Do Natural Resources Fuel Authoritarianism? A Reappraisal of the Resource Curse,” American Political Science Review (forthcoming). 
  31. “Economic Diversification in Africa: A Review of Selected Countries,” (OECD, United Nations OSAA, 2011), www.oecd.org/dataoecd/10/61/46148761.pdf
  32. Phillipe le Billon, “Angola’s Political Economy of War: the Role of Oil and Diamonds, 1975-2000,” African Affairs, 100 (2001): 55-80. 
  33. Ibid., 55-80. 
  34. Ibid., 55-80. 
  35. Ibid., 55-80. 
  36. Angola Peace Monitor 11, no. 4 (1999). 
  37. “World Bank World Development Indicators,” the World Bank, accessed online January 11, 2011, http://www.data.worldbank.org/indicator
  38. “Country Profile: Angola,” Freedom House, accessed online January 11, 2011, http://www.freedomhouse.org
  39. “Angola Poverty Levels Reduce from 68 to 38 percent,” UNDP, accessed December 12, 2010, http://www.ao.undp.org/news149.htm
  40. “Economic Diversification in Africa.” 
  41. “World Bank Announces Withdrawal from Chad-Cameroon Pipeline After Early Repayment,” Bank Information Center, accessed December 14, 2010, http://www.bicusa.org/en/Article.3892.aspx
  42. Ibid.
Ryan J. Connelly is an M.A. candidate at SAIS Bologna Center, specializing in Energy Politics and Conflict Management. He graduated from the George Washington University in 2006 with a degree in international affairs. Ryan developed an interest in the political economy of resource-dependent states while at Morgan Stanley, where he worked as a desk analyst focused on energy markets.