The World Bank,contradictory lending,and forests: A cross-national analysis of organized hypocrisy

Abstract

Spanish

French

This article draws on the theory of organized hypocrisy to test the hypothesis that World Bank lending in different sectors has contradictory impacts on forests. The authors use ordinary least squares regression to analyze newly available satellite imagery data on forest loss from 2001 to 2014 for a sample of 89 low and middle income nations. The analysis finds support for the theory of organized hypocrisy. The results indicate that World Bank structural adjustment lending and investment lending in the agriculture and forestry sectors are related to more forest loss but World Bank investment lending in the environmental sector is related to less forest loss. The article concludes with a discussion of the theoretical, methodological, and policy implications.

Keywords

Cross-national environment forest loss organized hypocrisy World Bank

Introduction

Since it was established in 1944 after delegates from 45 nations negotiated a monetary agreement following the Second World War (Babb, 2008), the World Bank (WB) has undergone many changes in its activities. Originally created in order to finance the rebuilding of Europe, as its initial task came closer to being completed by the late 1950s, the bank began to turn more toward making investment loans for capital intensive projects (Peet, 2003). The purpose of such investments, including funding projects in agriculture and forestry sectors (McMichael, 2004), was to support the provision of basic needs in poor nations by promoting economic development. By the late 1970s, the bank also began making structural adjustment loans in order to help borrowing nations resolve balance of payment issues precipitated by high oil prices (Peet, 2003).

Because these investment and structural adjustment loans were predicated on stimulating economic growth via natural resource extraction, growing exports, and cutting government spending (Downey, 2015), the WB’s lending has been criticized for creating forest loss (Bryant and Bailey, 1997). For reasons we discuss below, the bank came under intense pressure from United States lawmakers, non-governmental organizations, and indigenous groups for the forest loss caused by its agriculture and forestry investments. The WB responded by implementing a variety of reforms that deal with environmental issues. These included an organizational restructuring, hiring more environmental staff, writing environmental directives, and providing loans for environmental protection and conservation (Rich, 1994).

Despite investments in conservation projects and funding environmental protection, the World Bank also continued to make investments in extractive sectors and imposed structural adjustment conditionalities that emphasized privatization of environmental resources and focused on export-led growth. In an attempt to explain why the WB puts contradictory policies into place, we build on the work done by Weaver (2008), who, in Hypocrisy Trap: The World Bank and the Poverty of Reform, draws on the theory of organized hypocrisy to explain how the bank responds to external pressures that threaten its finances and legitimacy. Weaver develops her argument through an examination of the WB’s implementation of anti-corruption reforms (Weaver, 2008). Her insights serve as the starting point for our study, but we extend the argument in a novel way. We contend that organized hypocrisy also manifests itself at the WB through its lending across purposes in different sectors (i.e., extractive versus environment), which may produce contradictory impacts on forests.

However, there is no cross-national research to our knowledge that considers this insight. We are surprised because Weaver’s book was published almost a decade ago and provides the theory to inform cross-national work. Methodologically, it is also possible to obtain data on different types of WB lending (i.e., investment or structural adjustment) by sector (e.g., agriculture, forestry, and environment) to test hypotheses drawn from this theory. Thus, we undertake this task here. We now turn to a discussion of the theory of organized hypocrisy and the application of it to the WB. We then relate this theory to the WB’s lending with an emphasis on why investment and adjustment lending in different sectors may have contradictory effects on forests. To test the theory, we develop a model of forest loss and discuss the variables, analytic methods, and empirical findings. We conclude by reviewing the theoretical and policy implications, as well as highlight some methodological challenges along with limitations and directions for future research.

A theory of organized hypocrisy and the World Bank

Brunsson (1989) was the first to theorize the concept of organized hypocrisy. Brunsson and Olsen (1993) went on to apply it to organizational reform. The theory draws on resource dependency and sociological institutionalism literatures. The key idea taken from these two fields, applied to the World Bank, is that as an organization it depends upon its external environment for financial support and conferred legitimacy (Weaver, 2008). While the conditions under which either factor brings about change is up for debate, an organization must appear responsive to these demands in order to survive (Weaver, 2008).

However, problems can emerge for an organization when the pressures exerted on it are constituted by ‘inconsistent expectations’ (Weaver, 2008). Weaver explains that, ‘If the nature of the dependency is such that organizations must placate multiple masters to attain needed material resources and conferred legitimacy, neither acquiescence nor defiance is a viable option’ (2008: 27). Rather, as Oliver (1999: 154) contends, the most likely response involves an organization implementing ‘rules, guidelines, and procedures’ that allow it to incorporate the inconsistent expectations into its operations and, thereby, ‘exhibit conformity with external demands.’ It is in instances like this that organized hypocrisy emerges (Brunsson, 1989). In its most blatant form, hypocrisy involves reforms never being implemented and which serve as ‘window dressing’ in response to the external pressures (Oliver, 1999). This type of response, where an organization will say one thing and do another, is the most often studied – see Oliver’s (1999) examination of a retail company responding to pressures by stockholders and consumers. However, Weaver (2008) contends that another response involves an organization implementing reforms while continuing with other aspects of its operations (Weaver, 2008). This hypocrisy takes the form of incorporating contradictory mandates and policies into an organizational structure (Wade, 2002).

Most research on organized hypocrisy focuses on corporations and governments – see Brunsson and Olsen (1993), Oliver (1999), or Barnett and Coleman (2005). However, Weaver (2008) extends the theory to the WB because it, too, must act in a way ‘that reflects not only the interests of those that provide critical material resources but also prevailing international ideals and norms in the broader global development regime’ (2008: 30). In this regard, because of the conflicting demands it experiences, the WB puts contradictory policies into place.

On the one hand, the WB experiences pressure from the United States Treasury and private capital markets, which it is largely dependent upon for capital to operate (Weaver, 2008). Thus, in response to these demands, it implements policies that correspond with a ‘finance ministry agenda’ and favors export-led growth, privatization, and trade liberalization (Wade, 2002). These policies serve the interests of the United States Treasury and private capital markets by opening up the economies to trade and financial flows from donor nations – even if it is at the expense of the natural environment (Weaver, 2008). For our purposes, such policies include investment and structural adjustment lending in agriculture and forestry.

On the other hand, the WB also must defer to issues being raised by United Nations agencies, non-governmental organizations, indigenous groups, and lawmakers of donor governments that appropriate funding to it (Weaver, 2008). These groups are pushing the WB to adopt a ‘civil society agenda’ (Wade, 2002). The policies associated with this agenda include investment lending in areas such as environment, education, health, gender, and the provision of basic needs (Wade, 2002). In this study, we use investment loans in the environmental sector as indicative of the civil society agenda.

Weaver concludes that the WB, ‘faced with the necessity of appearing responsive to both sets of demands, reacts by embracing both sets of agendas in its broad policy paradigms, leaving the inconsistencies and contradictions to be worked out in its daily operations’ (2008: 32). These ‘inconsistencies’ and ‘contradictions’ in the WB’s lending may lead to different impacts on forests. Before we evaluate this possibility, we review in turn the history and details of the WB’s finance ministry agenda (investment and structural adjustment lending in extractive sectors) as well as its civil society agenda (investments in the environmental sector).

The finance ministry agenda and forest loss

The World Bank’s investment lending for agriculture and forestry

The underlying logic of the WB’s investment lending in the agricultural and forestry sectors is predicated on stimulating economic growth by supporting projects that allow nations to boost exports (Rich, 1994). Following the rebuilding of Europe, the bank sought to diversify and expand its lending portfolio by making ‘soft’ loans to nations (McMichael, 2004). These loans had lower interest rates, longer repayment schedules, and longer grace periods than the WB’s early offerings (McMichael, 2004). They also moved beyond supporting infrastructure to include other sectors like agriculture, forestry, and mining (Peet, 2003).

The WB created a demand for these loans via a campaign of institution building and technical training (Rich, 1994). The WB encouraged governments to create development parastatals or semi-autonomous government agencies (e.g., development boards or electricity generating companies) charged with the task of promoting economic growth. In turn, the WB and these newly created agencies engaged in a mutually rewarding pattern. The parastatals borrowed from the WB in order to expand their mandate, which, in turn, created a continuous demand for the WB’s investment loans (Bryant and Bailey, 1997).

The WB’s investment loans for agriculture and forestry typically cost anywhere from $50 million to over $500 million spread across a 5- or 10-year period (WB, 2005). There are several characteristics that the loans share in common. First, investment loans focus on export-led growth. In the agricultural sector, the exports include cattle, coffee, cotton, soybeans, wheat, and corn (Rich, 1994). In the forestry sector, the exports include logs, fuelwood, pulp, paper, and plywood (Bryant and Bailey, 1997). Second, investment loans fund processing facilities and ‘better market linkages’ to facilitate exports (Rich, 1994). This includes lumber mills, paper mills, and slaughterhouses but also roads that connect such facilities to ports so exports can be sent abroad (WB, 2005). Third, investment loans often entail governments working with the private sector (McMichael, 2004). In some instances, governments may allow companies to purchase land or lease it to them often at below market rates (Tockman, 2001). In other instances, governments may enter into an operational agreement for a corporation to run a facility being financed by the WB (Rich, 1994).

We now turn to the question of how the WB’s investment lending for agriculture or forestry increases forest loss. We suggest some possible mechanisms by drawing on various examples of WB programs. In the late 1980s, Brazil received a $443 million loan from the WB. The funds were used to build roads along with 39 agricultural settlements in the Amazon. In these settlements, newly arriving migrants would grow cocoa and coffee for export on single-family farms or land leased to them by corporations (Bryant and Bailey, 1997). The government would provide credit to buy the seeds, fertilizers, and pesticides along with titles to land (Rich, 1994). The WB also encouraged cattle ranching in areas around the settlements. The project had a direct impact, with forests being cleared to expand farming and cattle ranching. However, there were secondary impacts. The Brazilian government was not able to provide the agricultural inputs needed to grow cocoa and coffee to the 500,000 migrants (Bryant and Bailey, 1997). As a result, more and more forests were cleared often by ‘slash and burn’ techniques by the migrants to grow subsistence crops because the soil could not sustain cocoa or coffee beyond one season without inputs (Rich, 1994).

The WB’s investments in palm have also created forest loss. For instance, Indonesia received several WB investment loans to increase exports of palm oil during the 1990s. The loans supported government efforts to convert forests to fields. It also created forest loss with trees being removed to build processing facilities and roads necessary to export the palm oil (Bryant and Bailey, 1997). The government in addition leased the newly established plantations and processing facilities to foreign companies, who expanded production into surrounding forests including national parks (Rich, 1994).

The WB’s investment loans in the forestry sector have also been linked to increased forest loss. The most obvious way it plays out is when loans support logging by financing roads, equipment, and processing facilities necessary to bring valuable wood to market but result in large swaths of forest being cleared. A report by the WB’s Independent Evaluation Group (2013) documents this process at work in Cambodia, Cameroon, Democratic Republic of Congo, Liberia, and Ghana. However, its investment lending in the forestry sector is not limited to logging. It also includes projects that seek to expand tree plantations, which leads to the clearing of natural forests (Rich, 1994). For instance, Thailand received a $60 million dollar loan during the late 1980s where farmers were encouraged to clear forests to make room for rubber plantations (Rich, 1994). A similar process has played out with natural forests being cleared to make room for eucalyptus plantations across Sub-Saharan Africa and Asia (Hurst, 1990).

The World Bank’s structural adjustment lending for agriculture and forestry

As noted above, the WB also provides structural adjustment loans in the agricultural and forestry sectors, in response to the ‘debt crisis’ of the 1980s, which was highlighted by the inability of many nations to repay their foreign debts (Magdoff, 1986). The WB responded to the crisis by providing indebted nations with new loans. These loans do not directly fund projects. Rather, they seek to resolve balance of payment issues by requiring nations to adopt macro-economic policy reforms in order to receive the money (McMichael, 2004). They typically range in value from $50 million to $400 million with a loan duration between two and five years (Peet, 2003). The reform conditionalities include devaluing the local currency, cutting government spending, liberalizing trade, and privatizing government assets (Rich, 1994). The underlying purpose of such reforms involves generating currency for debt repayment by boosting exports while cutting spending (Bryant and Bailey, 1997).

While this ‘earn more’ and ‘spend less’ model may facilitate debt repayment, structural adjustment also creates negative repercussions for forests. First, the WB requires nations to increase export earnings in order to make debt payments. The most common way to achieve this goal is a currency devaluation, which creates a demand for a country’s exports. In general, poor nations meet the increased demand by expanding the extraction of forestry and agricultural goods for export (Rich, 1994).

Second, WB structural adjustment often attempts to increase exports by having governments liberalize trade (Barbosa, 2001). This tends to involve removing barriers to foreign investment by providing corporations with a variety of regulatory concessions and financial incentives (Clapp, 1998). The regulatory concessions may include exemptions on logging harvest quotas, permission to export raw logs, the ability to log protected species, and logging in protected areas (Hurst, 1990). The most notable financial incentives entail ‘tax holidays’ like exemptions on export and import duties (London and Ross, 1995). The WB may also promote liberalizing trade by encouraging governments to sell off or lease public land (Downey, 2015). Accordingly, governments may be required to implement land reforms that make it easier for foreign corporations to obtain titles to land or limit the ability of a government to reclaim public land, even if obtained by illegal or semi-illegal means (Tockman, 2001). Taken together, liberalizing trade may stimulate foreign investment, but it also results in forest loss because it makes extractive investments much more profitable and easier to undertake (Jorgenson and Kuykendall, 2008).

Third, WB structural adjustment may require deep cuts in government spending to correct for budgetary imbalances (Tockman, 2001). The nature of the cuts varies from country to country, but a common theme has been the slashing of budgets and staffs from environmental departments (Bryant and Bailey, 1997). The cuts hamper enforcement of forestry regulations, demarcation of protected areas, and monitoring of illegal logging (Rich, 1994).

The preceding discussion leads us to our first main hypothesis. We argue that WB investment and structural adjustment lending for agriculture and forestry, which are representative of a finance ministry agenda, may increase forest loss. However, the WB also experienced pressure from non-governmental organizations, indigenous groups, and lawmakers in donor nations in response to the forest loss that structural adjustment lending was causing. We now turn to a discussion of the campaign against the WB and how it came to integrate a civil society agenda into its investment lending.

Pressuring the World Bank on forest loss

The experience of increasing forest loss and environmental degradation, such as what took place in Brazil (described in the example above), led to growing criticism of the WB. The case of Brazil is instructive as it may have been the catalyst for a wave of broader criticism of the bank. Initially, international non-governmental organizations and domestic non-governmental organizations in Brazil organized a campaign directly against the WB. This included sending letters to WB executives that highlighted anthropological research detailing the forced relocation of indigenous groups and corruption in titling of land. The letter was signed by leaders of international non-governmental organizations, the head of the Brazilian Bar Association, and the presidents of the American and Brazilian anthropological association. Nevertheless, the WB ignored the efforts.

In 1984, non-governmental organizations took a different approach, which involved pressuring key members in the United States Congress in an effort to stop the WB’s recapitalization (Wade, 2002). The first attempt at this involved flying Jose Lutzenburger, Brazil’s leading agronomist, to the United States to testify before a House sub-committee on agricultural research. In his testimony, Lutzenburger detailed how the WB project had led to extensive forest loss and declining agricultural productivity in Brazil. The testimony was carried on the evening news in Brazil.

This initial hearing caught the attention of Senator Robert Kasten of Wisconsin, who was the Chair of the Senate Appropriations Sub-Committee on Foreign Affairs that is responsible for approving the WB’s budget (Rich, 1994). The non-governmental organizations presented correspondence from WB officials about the project along with evidence from academics and environmentalists about the forest loss being created by the WB in Brazil (Rich, 1994). The correspondence served as evidence to Senator Kasten about the lack of accountability at the WB. Consequently, Senator Kasten sent a letter to WB President Barber Clausen, questioning its response to date and indicating that it may complicate securing the United States financial support (Rich, 1994). Kasten sent an even blunter letter to Secretary of the Treasury Donald Regan. Kasten wrote, ‘The World Bank’s response … is an insult. It was outrageous, reinforcing feelings by many that … the World Bank is arrogant and unwilling to receive constructive criticism’ (Rich, 1994: 124). He went on to ask Regan to deal with the matter personally and reiterated that the WB’s attitude ‘obviously complicates our problems in securing appropriations in the Senate’ (Rich, 1994: 124).

Within 24 hours of sending the letters, Kasten received phone calls from the United States Executive Director at the WB, James Burnham, asking for a meeting. The WB was clearly troubled by the thought that Kasten, as chair of the key sub-committee that appropriated financing to the WB, was in a position to withhold one-fifth of its annual budget (Rich, 1994). Kasten demanded a meeting between the non-governmental organizations and the WB’s entire senior management including President Clausen (Rich, 1994). In the meeting, Ernest Stern, Senior Vice President of Operations, launched into a discussion about the importance of the environment but that the WB’s main priority was ensuring the stability of the international financial system following the debt crisis (Rich, 1994). Tired that nothing new was being said, Kasten turned to Clausen and stated that the WB would not receive its appropriation unless changes were made to address the issues raised by the non-governmental organizations about the project in Brazil (Weaver, 2008). With eight months passing since Kasten’s initial letter had been sent, President Clausen announced that the WB was suspending remaining loan disbursements totaling over $250 million ‘pending the preparation and carrying out of emergency environmental protection measures by the Brazilian government’ (Rich, 1994: 126). This marked the first time that the WB halted a loan disbursement for environmental concerns as a result of pressure brought about by non-governmental organizations and lawmakers from a donor nation (Rich, 1994).

Within the next year, Clausen left the WB and Barber Conable was appointed to lead it. In response to this campaign, he delivered a speech in which he conceded that ‘the WB has been part of the problem in the past’ in terms of forest loss, and admitted that it ‘had misread the human, institutional, and physical realities of the frontier’ in regard to investment projects (Rich, 1994: 146). Conable then went on to implement specific environmental reforms that would be taken, which incorporated a civil society agenda. We now turn to an extended discussion of how the WB implements its civil society agenda.

The civil society agenda and forest loss

Investment lending in the environmental sector

The reforms undertaken by the WB began with it increasing environmental staff and locating them in a newly created Environment Department (Bryant and Bailey, 1997). During the late 1980s, the WB had only five staff members working on environmental issues (Clapp and Dauvergne, 2005), which had increased in number to 60 environmental staff members by 1990. By the year 2000, this had grown to just under 1000 staff (WB, 2001).

The staff worked to create issue papers with the purpose of reviewing and addressing how environmental problems were affecting borrowing nations while also evaluating the WB’s role (Rich, 1994). The staff drafted operational policies for how the WB would address issues of biodiversity loss, conservation, mining, dam building, forced resettlement, and forestry in its lending (WB, 1990). From the beginning, it collaborated with non-governmental organizations on the guidelines. For instance, the first major publication in this regard was Conserving the World’s Biological Diversity, which offered guidelines for preserving natural habitats to minimize species loss (WB, 1990). It followed with guidelines on curbing forest loss via land tenure, agroforestry, and community participation in project design, implementation, and monitoring (WB, 1989).

The most important publication was the Operational Directive on Environmental Assessment (see Rich, 1994). This directive ensures that development projects were environmentally sound and adverse impacts were recognized and addressed early in a project’s design (WB, 1990). Otherwise, projects would be abandoned (Clapp and Dauvergne, 2005). This report also provided a mechanism that allowed affected local groups and non-governmental organizations to be heard and their concerns addressed (WB, 1990). The publication served to provide WB personnel with guidelines that seek to mitigate the adverse environmental impacts of its lending. Most importantly, it described policies for standalone environmental investment (WB, 1990).

The number of projects that dealt with the environment was negligible prior to 1985, but by 1990, lending for projects for the environment accounted for a third of the projects annually. This translated into environmental lending totaling $1.6 billion in 1990, which was about 7% of the WB’s total portfolio (Piddington, 1992). From 1990 to 2000, investment lending in the environmental sector averaged $1 billion annually (AidData, 2016).

The WB identified five priority areas to fund. These include destruction of natural habitats, land degradation, fresh water depletion, industrial pollution, and the ‘global commons’ (WB, 1990). However, the ‘cross-cutting’ focus of lending in the environmental sector was forest conservation (WB, 1990). The WB’s earliest efforts involved the demarcation of forest borders as its first step in conservation. For instance, Guinea received a loan for $40 million to support the demarcation of 150,000 hectares of humid forest and 160,000 hectares of dry forest along with the monitoring of the demarcated areas for illegal logging (WB, 1990). It is important to note that the loan doubled the government’s expenditure for conservation, which was reinvested in monitoring (WB, 1990).

The WB’s approach in such instances relies on armed patrol guards and physical barriers to protect a forest. However, this reliance on ‘guns, fences, and fines’ to promote conservation has been criticized. First, a nation may receive funds from the WB for conservation but, in reality, the loan may serve as a means for a country to exert control over a remote region and the people living there (Peluso, 1993). The creation of a national park or nature reserve not only entails the marking of the boundaries but also the appointment of a ‘whole army of park rangers and guards to ensure that excluded actors do not interfere with park management’ (Bryant and Bailey 1997: 65). The ultimate goal may be to open the region to development, which could exacerbate forest loss (Bryant and Bailey, 1997).

Second, ‘strict’ conservation practices tend to remove local populations from protected areas and declare any extractive activities as illegal (Kubo and Supriyanto, 2010). This was the case with the WB’s support of debt-for-nature swaps in Madagascar. The transaction paid off a portion of the country’s debt but required complete protection of forests. Bryant and Bailey (1997: 142) note, ‘grassroots actors are denied access as part of the terms of the swap and resort, thereafter, to illegal forest extraction or new land uses like grazing.’ Toward this end, forest loss increased along with conflict with government officials due to the enclosure (Bryant and Bailey, 1997).

While ‘environment-first’ rather than ‘people-first’ projects remain part of its environmental lending, the bank began investing in conservation efforts that sought to reduce pressure on forests by providing alternative sources of income to local populations (Bryant and Bailey, 1997). These ‘mixed’ conservation projects follow recommendations from the study ‘People and Parks: Linking Protected Areas with Local Communities’ carried out by the WB, World Wildlife Fund, and United States Agency for International Development. It recommended that attention be given to implementing institutional and incentive structures that encourage local people to work with governmental agencies and non-governmental organizations to put forestry and agricultural management plans into place (WB, 1990).

For instance, people living in rural areas of Nepal are dependent on forests for fuelwood, fodder, food, building materials, medicinal plants, and fertilizers (WB, 2001). Thus, improving the management of forest resources was imperative for the welfare of a large portion of the country’s population. The WB aimed to establish a ‘community-based’ forestry management system in order to protect and expand forest resources with a $100 million environmental loan. The program involved turning over responsibility for forestry to over 9000 forest user groups, representing nearly 40% of rural households in Nepal (WB, 2001). The WB loan helped usufruct rights for more than 400,000 hectares of forest to be transferred in perpetuity to local communities (WB, 2001). The project resulted in forest regeneration and created substantial income for participants through higher production yields of non-timber forest products along with fuelwood and fodder (WB, 2001). There is a key difference between the WB’s investment loans across extractive and environmental sectors. The former emphasizes promoting exports via corporate investment in rural areas, while the latter focuses on protecting forests and setting production levels for domestic consumption in defined areas.

The WB maintained that its environmental projects would only be successful if it was able to work with all stakeholders to strengthen legal and institutional deficiencies within a nation (WB, 1990). Many projects sought to improve the legislative frameworks and financial incentive systems in the forestry sector while also strengthening the institutional capacity of governments to implement and monitor conservation laws (Clapp and Dauvergne, 2005). For instance, the WB invested $117 million in Brazil to reduce forest loss in the Amazon. The project involved creating protected areas, hiring personnel for monitoring, improving training of personnel, and purchasing equipment to carry out monitoring (Clapp and Dauvergne, 2005). It also entailed establishing a national network for environmental information including centers to interpret remote sensing images used to monitor illegal logging in the region (WB, 1990). The project emphasized annual training of staff at the centers and in the field in defining property rights over forests (WB, 1990).

However, concerns about the WB’s people-first lending remain. At a basic level of concern, it takes substantial amounts of effort to write up project loan assessments and their potential environmental impacts (Rich, 1994). However, little time and money is set aside for the process beyond WB staff flying to a country and holding brief talks with involved parties (Goldman, 2005). There is also little funding to carry out research to complete this task (Goldman, 2005). Toward this end, meaningful participation in the design, implementation, and evaluation is limited and may translate into a project’s failure. This problem is enhanced by career considerations of its employees, who are rewarded for increasing the WB’s lending portfolio and delaying loans (Goldman, 2005).

A far more serious concern comes from Goldman (2005), who fears that the WB is putting ‘green’ neoliberal policies into place. Goldman (2005) describes how this process played out in Laos. The WB carried out scientific studies with non-governmental organizations and scientists from the United States and Europe to determine the causes of forest loss in Laos. The government then used the data to draft new forestry laws that defined the ecological zones and what activities could be carried out in certain zones and not others. This included allowing the construction of large hydroelectric dams with the areas around them completely protected, thereby out of bounds to people who may want to protest their construction (Goldman, 2005).

We may also be seeing this process at work in our discussion of the WB’s investment in Nepal. This project defines property rights over forests managed by local communities. However, it also allowed the government to designate forests that could be used for other purposes like logging. Despite the designation, forest protection is predicated on establishing ownership by a community, company, or government and is centered upon extracting resources (Bryant and Bailey, 1997).

This discussion leads us to our second hypothesis. We argue that WB investment lending in the environmental sector, which is representative of a civil society agenda, should correspond with less forest loss, even in spite of the potential limitations we have discussed. Before testing our hypotheses, we turn to a discussion of the variables and methods.

Data

Dependent variable

Forest loss

The cross-national research on forest loss tends to use data made available in the United Nation’s Food and Agriculture Organization’s Global Forest Resources Assessment (e.g., Shandra et al., 2011). Their reliability has been called into question because they are gathered using collection methods that vary by nation (Grainger, 2008). In some places, forestry statistics are more reliable because they are based on remote sensing surveys (Food and Agriculture Organization, 2015). In other places, estimates are less reliable because they are based on expert opinions or extrapolated from an outdated forest inventory (Grainger, 2008).

Therefore, we use newly available data on forest loss derived from high resolution satellite imagery (30 × 30 meters) in order to eliminate this potential source of error. The data may be obtained online from the World Resources Institute’s (2016) Global Forest Watch web page. There is another advantage of these data in addition to the comparability across nations. We can identify forest loss across a range of tree canopy cover density levels or the estimated percentage of a pixel that is covered by a tree canopy. The different tree canopy cover density levels can be used to distinguish loss of certain ‘types’ of forests. Miles et al. (2006) argue that data derived from a 75% tree cover canopy density level indicate loss of ‘tropical’ or ‘wet’ forests. Rudel et al. (2016) indicate that data derived from a 50% tree canopy cover density level correspond with loss of ‘temperate’ or ‘dry’ forests.

We follow Rudel’s (2013) procedure for calculating the dependent variable of rate of forest loss. We calculate the loss of wet forests in the following way. First, we obtain the number of hectares of forest cleared from 2001 to 2014. Second, we collect the total amount of forest area in hectares for each county in 2000. The data are derived using the 75% canopy cover level. Third, we divide the number of hectares cleared in a country from 2001 to 2014 by the country’s total forest size in hectares for 2000. This yields the rate of wet forest loss over this period of time. We calculate the rate of dry forest loss in a similar way. However, we have to subtract the hectares of forest area cleared under a 75% tree cover canopy density threshold from the hectares of forest area under a 50% tree cover canopy density measure prior to doing so. This procedure yields the total hectares of dry forest in a country, because estimates based on the 50% tree cover canopy density include the areas that would also be classified as wet forests using the 75% tree cover canopy density level. We then calculate the rate of change over this time. Please see Table 1 for descriptive statistics and bivariate correlation matrix of all data.

Table 1.

Descriptive statistics and bivariate correlation matrix (N = 89).

	Mean	SD	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)	(12)	(13)	(14)	(15)
(1) Wet forest loss, 2001–2014	.117	.162	1.000
(2) Dry forest loss, 2001–2014	.096	.189	.395	1.000
(3) WB investment lending in environmental sector	39.201	235.704	−.044	−.022	1.000
(4) WB investment lending in agriculture and forestry sectors	5.909	7.128	−.094	.280	−.012	1.000
(5) WB structural adjustment lending in agriculture and forestry sectors	.687	1.114	.146	.232	−.053	.133	1.000
(6) Multilateral debt service, 2000	3.546	.879	.027	.123	.108	.139	.236	1.000
(7) Non-governmental organizations, 2000	4.821	.100	−.101	−.069	.077	−.064	−.101	−.016	1.000
(8) Gross domestic product, 2000	6.679	1.044	−.253	−.206	−.041	−.072	−.272	−.318	.305	1.000
(9) Economic growth rate, 1990–2000	3.714	3.853	−.161	−.204	.027	−.028	−.243	−.206	−.068	−.003	1.000
(10) Agricultural land area, 2000	3.675	.737	.164	.012	.100	−.026	−.297	−.136	−.301	−.129	.310	1.000
(11) Corruption, 2000	.387	.678	.189	.084	.008	.093	.046	.073	−.228	−.451	−.002	.130	1.000
(12) Democracy, 2000	11.924	8.645	−.181	−.208	−.137	.057	.082	−.055	.014	.410	−.096	−.119	−.304	1.000
(13) Total population growth rate, 1990–2000	.200	.134	.382	.300	−.033	−.069	.115	.129	−.036	−.316	−.172	−.179	.170	−.375	1.000
(14) Rural population growth rate, 1990–2000	.106	.131	.372	.203	−.026	−.076	.217	.105	.007	−585	−.124	−.092	.272	−.332	.757	1.000
(15) Urban population growth rate, 1990–2000	.289	.210	.202	.270	.094	−.116	.078	.222	−.052	−.330	−.037	−.090	−.038	−.486	−.699	.358	1.000

Main independent variables

WB investment lending in environmental sector

Our first independent variable measures the total amount of general environmental protection lending a nation receives between 1990 and 2000. We divide this amount by a country’s total population size in 2000 to standardize it. We collect the data from 1990 to 2000 for a couple of reasons. First, while the WB began implementing its environmental reforms in 1988, its first environmental loans were not disbursed until 1990 (WB, 1990). Second, WB loans are often implemented over many years in a nation. If we used data for one year, we may underestimate the effects of WB lending. This is especially true for the structural adjustment variable, whose conditions stay in place even after a loan is fully disbursed – see above. Third, we end the collection of data beyond 2000 to avoid simultaneity bias. This variable was collected online from the WB’s (2017) Projects and Operations Database. All other data can be found in the WB’s (2016) World Development Indicators database online unless otherwise indicated. The project database allows users to search loans by type, sector, and approval date. We begin by identifying nations that received loans categorized with the ‘general environmental protection’ classification. We then identify loans and record amounts only if a project involves forest protection, conservation, or management. We code with this level of detail by using project descriptions and documents for each loan. It allows us to exclude environmental loans that may not affect forests including air pollution or wastewater treatment technology. From the theory of organized hypocrisy, we hypothesize that higher amounts of lending are associated with less forest loss.

WB investment lending in agriculture and forestry sectors

The second independent variable is the total amount of investment lending in the agriculture and forestry sectors received by a nation from 1990 to 2000 divided by its population size in 2000. The data may be obtained from the WB’s (2017) Project and Operations Database. As discussed above, we hypothesize that higher amounts of WB investment lending in these sectors, which is associated with a finance ministry agenda, are associated with higher rates of forest loss.

WB structural adjustment lending in agriculture and forestry sectors

We include a variable that measures the total amount of money that a nation receives as part of structural adjustment loans in the agriculture and forestry sectors from 1990 to 2000. We divide this value by a country’s population size in order to standardize it. This variable is logged. Again, as discussed above, we expect higher amounts of structural adjustment lending, which is also representative of a finance ministry agenda, should be related to increased forest loss.

Other independent variables

Debt service

We include debt service or the sum of principal and interest payments in foreign currency, goods, or services on long-term public and publicly guaranteed private debt or debt with a maturity of two years or longer as a percentage of exports of goods and services. The data may be obtained from the WB (2016). We expect that higher levels of debt service are associated with higher rates of forest loss because indebted nations tend to increase exports of natural resources to earn cash for repayments (Marquart-Pyatt, 2004).

Non-governmental organizations

We include a measure of international non-governmental organizations (NGOs) working on ‘environmental’ and ‘animal rights’ issues in a nation for the year 2000 standardized by total population. The data are collected by Smith and Wiest (2005) from the Yearbook of International Associations. Schofer and Hironaka (2005) find that higher levels of non-governmental organizations are associated with lower rates of deforestation. This may be the case because non-governmental organizations finance local conservation projects, support social movement activity, and shape the language of environmental laws.

Gross domestic product

We employ a measure of gross domestic product per capita for 2000. We log this variable to correct for its skewed distribution. Burns et al. (2003) find that higher levels of economic development are associated with lower rates of deforestation and argue that this finding is a result of wealthier nations ‘externalizing’ their environmental costs by importing natural resources. It may also be that higher levels of wealth are accompanied by increased concern for the environment and put pressure on governments to pass environmental protection laws (Jorgenson, 2006).

Economic growth

We include the average annual economic growth rate from 1990 to 2000. The cross-national research that examines how economic growth affects forest loss yields contradictory findings. On the one hand, economic growth is associated with higher rates of deforestation (Jorgenson, 2006). This is because there are large amounts of capital available for investment in activities (i.e., infrastructure) that accelerate forest loss during periods of economic expansion (Rudel, 1989). On the other hand, an absence of economic growth may increase forest loss. The lack of growth offers rural populations little incentive to migrate to urban centers for work and, as a result, rather expand agriculture and forestry to make a living (Ehrhardt-Martinez et al., 2002).

Agricultural land area

We include the percentage of land within a nation that is being used for agriculture or land that is permanently under crops or pasture (WB, 2016). This variable measures the size of the agricultural sector. We expect that higher amounts of agricultural land correspond with more forest loss because forests are converted to pastures or fields (Austin, 2010).

Corruption

This variable may be found online from the Varieties of Democracy web page (Coppedge et al., 2016). We carry out a principal components factor analysis of countries’ executive, legislative, judicial, and public sectors corruption scales to create a weighted index. In the executive, legislative, and public sectors, corruption refers to its members or agents granting favors in exchange for bribes, kickbacks, embezzlement, and misappropriation of public funds for personal or family use (Coppedge et al., 2016). In the judicial sector, corruption involves judges and their agents receiving undocumented extra payments or bribes from an individual or company to speed up, delay, or obtain a favorable decision (Dahlberg et al., 2016). We hypothesize that higher levels of corruption should be related to forest loss. This is because corruption across sectors diverts government funds away from investment in conservation (Koyuncu and Yilmaz, 2009). It also allows illegal logging and other violations of a country’s forestry laws to go unenforced and continue with little fear of punishment (Koyuncu and Yilmaz, 2009).

Democracy

We use Vanhanen’s (2014) measure of democracy,¹ which is calculated by taking the average of his political competition index with his political participation index. According to Vanhanen (2014), political competition measures the percentage of votes gained by smaller parties in parliamentary and presidential elections. The political participation variable measures the percentage of the population that voted in parliamentary and presidential elections (Vanhanen, 2014). According to Li and Reuveny (2006), higher levels of democracy are associated with lower rates of deforestation because democratic nations have more political activism than repressive nations. This is a result of democratic nations guaranteeing freedoms of speech, press, and assembly in addition to electoral accountability (Marquart-Pyatt, 2004).

Total population growth

We include the average annual percentage change in total population growth from 1990 to 2000. Rudel (1989) suggests that ‘geometric’ growth in population outstrips ‘arithmetic’ growth in the means of subsistence, leading to ‘carrying capacity’ problems (e.g., forest loss). We expect that higher rates of population growth correspond with more forest loss.

Rural and urban population growth

Jorgenson and Burns (2007) find that higher rates of rural population growth are associated with increased deforestation while higher rates of urban population growth are associated with lower rates of deforestation. They argue that expanding urban centers create economic opportunities other than agriculture, which leads to increased rural to urban migration (Jorgenson and Burns, 2007). This must be accompanied by the importing of food, which reduces pressure on forests (Rudel, 2013). However, the effect of urban population growth may increase forest loss. This occurs because higher incomes among an emerging middle class in cities drive up the price of agricultural and forestry products (DeFries et al., 2010). The rate of forest loss increases as producers respond to meet consumer demand. Given these potentially differential effects on forest loss, we also examine the impact of the average annual percentage change in rural and urban populations from 1990 to 2000. We log these variables because of skew.

Analysis plan

We use ordinary least squares regression to analyze the data.² This model is often employed to analyze forest loss (e.g., Austin, 2010).³ It is denoted by the following formula:

y_{i} = a + b_{1} X_{1} + b_{2} X_{2} \dots + b_{k} X_{k} + e_{i}

where,

y_i = dependent variable for each country,

a = the constant,

b₁ to b_k = unstandardized coefficients for each independent variables,

X_k = independent variables for each country, and

e_i ₌ error term for each county.

We must ensure that we are not violating regression assumptions if the results are to be valid and reliable. First, we calculate mean and highest variance inflation factor scores for each model. We report the values in Table 2. There does not appear to be any potential problems with multicollinearity because mean and highest variance inflation factor scores do not exceed a value of 2.5 (York at al., 2003).

Table 2.

Ordinary least squares regression estimates of World Bank lending on forest loss, 2001–2014.

	Equation (2.1)	Equation (2.2)	Equation (2.3)	Equation (2.4)
	Wet forest loss	Wet forest loss	Dry forest loss	Dry forest loss
Independent variables
World Bank investment lending	−.001*	−.001*	−.001*	−.001*
in environmental sector, 1990–2000	−.052	−.059	−.035	−.055
	(.001)	(.001)	(.011)	(.001)
World Bank investment lending	−.001	−.001	.007*	.008*
in agriculture and forestry sectors, 1990–2000	−.069	−.052	.288	.303
in agriculture and forestry sectors, 1990–2000	(.002)	(.002)	(.004)	(.004)
World Bank structural adjustment	.026*	.023*	.032*	.031*
lending in agriculture and forestry sectors, 1990–2000	.195	.177	.189	.185
lending in agriculture and forestry sectors, 1990–2000	(.002)	(.013)	(.017)	(.017)
Multilateral debt service ratio, 2000	−.009	−.005	−.002	−.007
	−.058	−.035	−.011	−.033
	(.016)	(.018)	(.013)	(.013)
Non-governmental organizations, 2000	.492	−.275	.080	.073
	.049	−.002	.006	.005
	(.414)	(.476)	(.436)	(.519)
Gross domestic product, 2000	−.009	.010	−.001	.008
	−.066	−.073	−.002	.048
	(.015)	(.019)	(.017)	(.017)
Economic growth rate, 1990–2000	−.007	−.008	−.008	−.009
	−.205	−.220	−.176	−.198
	(.005)	(.005)	(.005)	(.006)
Agricultural land area, 2000	.079***	.073***	.049*	.047*
	.335	.311	.167	.158
	(.022)	(.022)	(.023)	(.023)
Corruption, 2000	.019	.016	−.018	.001
	.049	.075	−.066	.001
	(.019)	(.022)	(.020)	(.021)
Democracy, 2000	.001	−.001	−.003	−.002
	.011	−.011	−.170	−.129
	(.119)	(.002)	(.002)	(.145)
Total population growth rate, 1990–2000	.375***		.346**
Total population growth rate, 1990–2000	.331		.245
	(.119)		(.140)
Rural population growth rate, 1990–2000		.328*		.114
Rural population growth rate, 1990–2000		.285		.079
		(.152)		(.145)
Urban population growth rate, 1990–2000		.083		.211*
Urban population growth rate, 1990–2000		.117		.235
		(.087)		(.099)
Constant	−.142	−.247	−.126	−.180
	(.167)	(.187)	(.177)	(.178)
R²	.267	.247	.266	.263
Sample size	88	88	89	89
Highest variance inflation factor score	1.91	2.59	1.92	2.60
Mean variance inflation factor score	1.33	1.50	1.34	1.51
Breusch–Pagan test	35.90***	34.27***	93.43***	94.46***

Notes: * indicates p < .05, ** indicates p < .01, and *** indicates p < .001 for a one-tailed test.

The first number is the unstandardized coefficient, the second number is the standardized coefficient, and the third number in parentheses is the robust standard error.

We remove Sudan from the wet forest loss models because it is an outlier.

Second, we examine scatterplots of each independent variable against the dependent variables to determine if there are any problems with linearity (Allison, 1999). We transform variables when appropriate with the natural log (Tabachnick and Fidell, 2013). We note any transformation.

Third, we calculate standardized residuals to determine if outliers are a problem. There are no nations with standardized residuals greater than an absolute value of 2.5 (Tabachnick and Fidell, 2013). We also examined Cook’s distance statistics to detect influential cases. The results indicate Sudan is an outlier in the wet forest loss models.

Fourth, we calculate Breusch–Pagan statistics for each model to determine if heteroskedasticity is problematic. The null hypothesis for this chi-square test is that the error variances are homoskedastic or equally distributed. The coefficients for these chi-square statistics reach a level of statistical significance in every model, indicating potential problems with heteroskedasticity (Tabachnick and Fidell, 2013). We present robust standard errors to help deal with this issue, which is common in cross-national research.

Findings

In Table 2, we present the regression estimates of forest loss.⁴ The first number is the unstandardized coefficient, the second number is the standardized coefficient, and the third number in parentheses is the robust standard error. We report one-tailed hypothesis tests because of the directional nature of the hypotheses.

In every equation,⁵ we include WB investment lending in the environmental sector, WB investment lending in forestry and agriculture, WB structural adjustment in forestry and agriculture, debt service, non-governmental organizations, gross domestic product, economic growth,⁶ agricultural land area, corruption, and democracy. In odd-numbered equations, we examine the impact of total population growth, and in even-numbered equations, we examine the impact of rural and urban population growth. In equations (2.1) and (2.2), we examine change in wet forests. In equations (2.3) and (2.4), we examine change in dry forests.

Let us begin with the WB variables.⁷ In equation (2.1) through (2.4), we find that WB environmental lending, representative of a civil society agenda, is associated with decreased forest loss. The coefficients for this variable are negative and significant in all the models. We also find that other aspects of WB lending have different impacts on forests. In every equation, we find that WB structural adjustment lending in extractive sectors, representative of a finance ministry agenda, is associated with increased forest loss. The coefficients are positive and significant. The coefficients for WB investment lending in the forestry and agricultural sectors, another indicator of the finance ministry agenda, are positive and significant in two of four equations. Taken together, the results support the theory of organized hypocrisy. The WB is providing loans to borrowing nations that have contradictory effects on the natural environment, with its investment lending in the environmental sector related to less forest loss while its structural adjustment and investment lending in extractive sectors is related to more forest loss.

We find other factors are associated with forest loss. First, we find that higher levels of agricultural land area correspond with increased forest loss. The coefficients are positive and significant. Second, we find that total population growth is related to increased forest loss. The coefficients for this variable are positive and significant in equations (2.1) and (2.3). Third, we find that the remaining demographic measures have different effects on wet and dry forests. The coefficient for the rural population growth is positive and significant in the wet forest loss model while the coefficient for the urban population growth is positive and significant for the dry forest loss model.

There are non-significant findings that should be discussed. First, we find that a number of economic factors are not related to forest loss. These include debt service, gross domestic product, and economic growth. Second, political characteristics of nations, including democracy, non-governmental organizations, and corruption, do not reach a level of significance.

Discussion and conclusion

We began by noting that the existing cross-national research has mainly drawn on political-economic thinking to theorize how the WB affects forests. For instance, Shandra et al. (2011) find that WB investment and structural adjustment in extractive sectors correspond with increased forest loss. However, this approach ignores how other forms of lending, including in the environmental sector, may impact forests. Insights from organizational sociology suggest that such lending arises due to conflicting pressures on the bank. In this research, we thus seek to test the potential differential effects that may arise due to the WB’s contradictory lending policies. By applying insights from Weaver (2008), we draw on the theory of organized hypocrisy to explain why the WB pursues both a finance ministry and civil society agenda simultaneously in its lending. We find support for this more nuanced theory: investment and structural adjustment lending for agriculture and forestry is associated with increased forest loss, while lending for the environment corresponds with decreased forest loss.

We offer the following theoretical implication. As illustrated in our findings, cross-national research needs to move beyond considering only insights from political-economic theory, which are primarily concerned with the consequences of the WB’s finance ministry agenda and contend that lending only adversely affects the environment. This leaves us with an incomplete understanding of the WB. Sociologists should also consider reforms that the WB puts into place in response to external pressures that threaten its finances and legitimacy. These reforms often represent a civil society agenda, including the WB’s environmental lending. By using the theory of organized hypocrisy to synthesize the reasons behind these two agendas, we are able to assess both the finance ministry and civil society agendas in the same model. This permits us to arrive at a more comprehensive understanding of how the WB impacts the natural environment than would be possible by only considering one of the explanations alone.

There is a methodological implication that follows as well. Our results here correspond with the approach put forth by Jorgenson (2008), who ‘decomposes’ foreign investment into primary, secondary, and tertiary sectors, and examines the impact on the natural environment of investments in each sector. We agree with this author’s approach to use more refined measures in cross-national research, as we do here. Utilizing sector specific data allows researchers to develop a more nuanced understanding of forest loss and to offer more specific policy recommendations.

We also note that our findings have potential implications for policy decisions. They indicate that WB lending in the environmental sector is associated with less forest loss. This contradicts critiques that suggest this form of lending is not effective for the several reasons that we described above. Thus, our first suggestion is for the WB to expand its investment lending in the environmental sector. Second, there should be an attempt by the WB to reduce corruption at all levels but, realistically, this would start with its own projects. These suggestions may best be accomplished by the WB reinforcing its commitment to partnering with local people and non-governmental organizations in planning and design of projects along with using an independent auditor for project evaluation.

However, these policy recommendations can be critiqued as ‘reformist’ because they do not address the fundamental causes of forest loss (i.e., structural adjustment and investment lending in extractive sectors) (Bryant and Bailey, 1997). If this critique is correct, then it supports Weaver’s conclusion that ‘lending for environmental projects does not by itself indicate a significant shift in the WB’s thinking and action’ (2008: 24). This appears to be the case here.

While environmental lending is related to less forest loss, its effect size is smaller than that of investment and adjustment lending in the agriculture and forestry sector. Given the smaller ‘positive’ effect of environmental lending as well as its smaller share of lending compared to investment and structural adjustment, at a minimum the WB can be criticized for ‘greenwashing’ (Bryant and Bailey, 1997). The WB appears, in the eyes of the international community, to be a leader in conservation largely stemming from its support for environmental projects (Karliner, 1997). However, this image allows the WB to divert attention from its investments in extractive sectors and to profit off repayment for loans in both extractive and environmental sectors (Bryant and Bailey, 1997).

However, one concern remains particularly salient. We agree with Goldman (2005) that the WB’s investment lending in the environmental sector has the consequence of further putting into place ‘green neoliberal’ policies. When reviewing the characteristics common to investment loans in the environmental sector, in short these loans often require governments to rewrite laws related to property rights and natural resource management (Goldman, 2001). The legal changes often allow other forms of development to proceed with little or no recourse for affected communities (e.g., mining, dam building, logging) (Goldman, 2005). There also tends to be a restructuring of government ministries to ensure that the revised laws are monitored, enforced, and prosecuted (Goldman, 2001). In the end, projects based on extractive industry become enshrined in a government’s laws under the guise of conservation and remain the cornerstone of development like they have been for decades.

This discussion suggests that to prevent forest loss, the WB should abandon lending in extractive sectors like forestry and agriculture along with environmental lending that allows it to occur in certain areas. We describe above how political pressure in the United States Senate during the late 1980s forced the WB to undertake environmental reforms. It may well be that the only way to bring about any meaningful changes would entail non-government organizations, social movements, and concerned citizens to again focus their efforts on legislators in rich nations to withhold funding from the WB if it continues to pursue policies that cause environmental degradation.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Notes

Author biographies

Jamie M Sommer is a PhD candidate in Sociology at the State University of New York at Stony Brook. Her research interests include environmental sociology, global political-economy, and development.

John M Shandra is a faculty member at the State University of New York at Stony Brook. His research interests include environmental sociology, globalization, and statistics.

Michael Restivo is a faculty member in Sociology at the State University of New York at Geneseo. His research interests include global disparities related to health, technology, and the environment.

References

AidData (2016) Analytic dashboard. AidData: Open data for international development. Available at: AidData.org

Allison

(1999) Multiple Regression: A Primer. Thousand Oaks, CA: Pine Forge Press.

Austin

(2010) Soybean exports and deforestation from a world-systems perspective: A cross-national investigation of comparative disadvantage. The Sociological Quarterly 51(3): 511–536.

Babb

(2008) Behind the Development Banks: Washington Politics, World Poverty, and the Wealth of Nations. Chicago: University of Chicago Press.

Barbosa

(2001) The Brazilian Amazon Rainforest: Global Ecopolitics, Development, and Democracy. Lanham, MD: University Press of America.

Barnett

Coleman

(2005) Designing police: Interpol and the study of change in international organizations. International Studies Quarterly 49(4): 593–620.

Brunsson

(1989) The Organization of Hypocrisy: Talk, Decisions and Actions in Organizations. Chichester: John Wiley.

Brunsson

Olsen

(1993) The Reforming Organization. London: Routledge.

Bryant

Bailey

(1997) Third World Political Ecology, 1st edn. New York: Routledge.

10.

Burns

Kick

Davis

(2003) Theorizing and rethinking linkages between the natural environment and the modern world-system: Deforestation in the late 20th century. Journal of World-Systems Research 9(2): 357–392.

11.

Clapp

(1998) The privatization of global environmental governance: ISO 14000 and the developing world. Global Governance 4: 295–316.

12.

Clapp

Dauvergne

(2005) Paths to a Green World: The Political Economy of the Global Environment, 2nd edn. Cambridge, MA: The MIT Press.

13.

Coppedge

Gerring

Lindberg

et al . (2016) V-Dem Dataset v6.2. Varieties of Democracy (V-Dem) Project. V-Dem Institute, Gothenburg, Sweden: University of Gothenburg and Notre Dame, IN: Helen Kellogg Institute for International Studies, University of Notre Dame.

14.

Dahlberg

Holmberg

Rothstein

B et al

et al . (2016) The quality of government basic dataset (version January 2016), University of Gothenburg, The Quality of Government Institute. Available at: www.qog.pol.gu.se

15.

DeFries

Rudel

Uriarte

Hansen

(2010) Deforestation driven by urban population growth and agricultural trade in the twenty-first century. Nature Geoscience 3(3): 178–181.

16.

Downey

(2015) Inequality, Democracy, and Environment. New York: New York University Press.

17.

Ehrhardt-Martinez

Crenshaw

Jenkins

(2002) Deforestation and the environmental Kuznets curve: A cross-national investigation of intervening mechanisms. Social Science Quarterly 83(1): 226–243.

18.

Food and Agriculture Organization (2015) Global Forest Resources Assessment 2015: Food and Agriculture Organization Forestry Paper. Rome: United Nations.

19.

Freedom House (2015) Freedom in the World. New York: Freedom House.

20.

Goldman

(2001) The birth of a discipline producing authoritative green knowledge, WB-style. Ethnography 2(2): 191–217.

21.

Goldman

(2005) Imperial Nature: The WB and Struggles for Social Justice in the Age of Globalization. New Haven, CT: Yale University Press.

22.

Grainger

(2008) Difficulties in tracking the long-term global trend in tropical forest area. Proceedings of the National Academy of Sciences 105(2): 818–823.

23.

Hurst

(1990) Rainforest Politics Ecological Destruction in South-East Asia. Atlantic Highlands, NJ: Zed Books.

24.

Jorgenson

(2006) Unequal ecological exchange and environmental degradation: A theoretical proposition and cross-national study of deforestation, 1990–2000. Rural Sociology 71(4): 685–712.

25.

Jorgenson

(2008) Structural integration and the trees: An analysis of deforestation in less-developed countries, 1990–2005. Sociological Quarterly 49(3): 503–527.

26.

Jorgenson

Burns

(2007) Effects of rural and urban population dynamics and national development on deforestation in less-developed countries, 1990–2000. Sociological Inquiry 77(3): 460–482.

27.

Jorgenson

Kuykendall

(2008) Globalization, foreign investment dependence and agriculture production: Pesticide and fertilizer use in less-developed countries, 1990–2000. Social Forces 87(1): 529–560.

28.

Karliner

(1997) The Corporate Planet: Ecology and Politics in the Age of Globalization. Berkeley: University of California Press.

29.

Koyuncu

Yilmaz

(2009) The impact of corruption on deforestation: A cross-country evidence. The Journal of Developing Areas 42(2): 213–222.

30.

Kubo

Supriyanto

(2010) From fence-and-fine to participatory conservation: Mechanisms of transformation in conservation governance at the Gunung Halimun-Salak National Park, Indonesia. Biodiversity and Conservation 19(6): 1785–1803.

31.

Reuveny

(2006) Democracy and environmental degradation. International Studies Quarterly 50(4): 935–956.

32.

London

Ross

RJS

(1995) The political sociology of foreign direct investment. International Journal of Comparative Sociology 36(3): 198–218.

33.

McMichael

(2004) Development and Social Change: A Global Perspective. Thousand Oaks, CA: Pine Forge Press.

34.

Magdoff

(1986) Third world debt: Past and present. Monthly Review 37: 1–10.

35.

Marquart-Pyatt

(2004) A cross-national investigation of deforestation, debt, state fiscal capacity, and the environmental Kuznets curve. International Journal of Sociology 34(1): 33–51.

36.

Miles

Newton

DeFries

et al . (2006) A global overview of the conservation status of tropical dry forests. Journal of Biogeography 33(3): 491–505.

37.

Oliver

(1999) Strategic responses to institutional pressures. Academy of American Management Review 16: 145–179.

38.

Peet

(2003) Unholy Trinity: The IMF, WB and WTO, 2nd edn. New York: Zed Books.

39.

Peluso

(1993) Coercing conservation? The politics of state resource control. Global Environmental Change 3(2): 199–217.

40.

Piddington

(1992) The role of the WB. In: Hurrell

Kingsbury

(eds) The Institutional Politics of the Environment: Actors, Interests and Institutions. Oxford: Clarendon Press, pp. 212–227.

41.

Rich

(1994) Mortgaging the Earth: The WB, Environmental Impoverishment, and the Crisis of Development. Washington, DC: Island Press.

42.

Rudel

(1989) Population, development, and tropical deforestation: A cross-national study. Rural Sociology 54(3): 327–338.

43.

Rudel

(2013) The national determinants of deforestation in sub-Saharan Africa. Philosophical Transactions of the Royal Society B 368(1625).

44.

Rudel

Sloan

Chazdon

Grau

(2016) The drivers of tree cover expansion: Global, temperate, and tropical zone analyses. Land Use Policy 58: 502–513.

45.

Schofer

Hironaka

(2005) The effects of world society on environmental protection outcomes. Social Forces 84(1): 25–47.

46.

Shandra

Shircliff

London

(2011) The International Monetary Fund, WB, and structural adjustment: A cross-national analysis of forest loss. Social Science Research 40(1): 210–225.

47.

Smith

Wiest

(2005) The uneven geography of global civil society: National and global influences on transnational association. Social Forces 84(2): 621–652.

48.

Tabachnick

Fidell

(2013) Using Multivariate Statistics, 6th edn. Boston: Pearson.

49.

Tockman

(2001) The IMF-Funding Deforestation: How International Monetary Fund Loans and Policies are Responsible for Global Loss. Washington, DC: American Lands Alliance.

50.

Vanhanen

(2014) Global Inequality as a Consequence of Human Diversity. London: Ulster Institute for Social Research.

51.

Wade

(2002) US hegemony and the WB: The fight over people and ideas. Review of International Political Economy 9(2): 215–243.

52.

Weaver

(2008) Hypocrisy Trap: The WB and the Poverty of Reform. Princeton, NJ: Princeton University Press.

53.

WB (1989) World Development Report 1989: Financial Systems and Development. Washington, DC: World Bank.

54.

WB (1990) Conserving the World’s Biological Diversity. Washington, DC: World Bank.

55.

WB (2001) Linking Poverty Reduction and Environmental Management. Washington, DC: World Bank.

56.

WB (2013) Independent Evaluation Group. Washington, DC: World Bank.

57.

WB (2005) WB Debt Servicing Handbook. Loan Services Group. Washington, DC: World Bank.

58.

WB (2016) World Development Indicators. Washington, DC: World Bank.

59.

WB (2017) Projects and Operations Database. Washington, DC: World Bank. Available at: http://projects.worldbank.org

60.

World Resources Institute (2016) World Resources Institute: Global Forest Loss 2016. Washington, DC: World Resources Institute.

61.

York

Rosa

Dietz

(2003) Footprints on the earth: The environmental consequences of modernity. American Sociological Review 68(2): 279–300.