- Email: [email protected]
Forest logging and institutional thresholds in developing south-east Asian economies: A conceptual model
Forest Policy and Economics 9 (2007) 1079 – 1089 www.elsevier.com/locate/forpol
Forest logging and institutional thresholds in developing south-east Asian economies: A conceptual model Ryan R.J. McAllister a,⁎, Alex Smajgl b , John Asafu-Adjaye c b
a CSIRO Sustainable Ecosystems, 306 Carmody Road, St. Lucia, QLD, 4067 Australia CSIRO Sustainable Ecosystems, Davies Laboratory, Private Mail Bag, Aitkenvale QLD, 4814 Australia c School of Economics, The University of Queensland, St Lucia, QLD, 4072 Australia
Received 3 February 2006; received in revised form 25 August 2006; accepted 27 October 2006
Abstract Many developing south-east Asian governments are not capturing full rent from domestic forest logging operations. Such rent losses are commonly related to institutional failures, where informal institutions tend to dominate the control of forestry activity in spite of weakly enforced regulations. Our model is an attempt to add a new dimension to thinking about deforestation. We present a simple conceptual model, based on individual decisions rather than social or forest planning, which includes the human dynamics of participation in informal activity and the relatively slower ecological dynamics of changes in forest resources. We demonstrate how incumbent informal logging operations can be persistent, and that any spending aimed at replacing the informal institutions can only be successful if it pushes institutional settings past some threshold. © 2006 Elsevier B.V. All rights reserved. Keywords: Formal and informal sectors; Institutional arrangements; Deforestation; Illegal logging; South-east Asia
1. Introduction In developing south-east Asian social-environmental systems, welfare is generally highly dependent on forest resources. In money-poor, resource-rich countries like Myanmar, Cambodia and Laos, largely rural based populations depend on tangible non-timber forest products, as well as less tangible flow-ons like water quality. One common dilemma faced in these countries ⁎ Corresponding author. Tel.: +61 7 3214 2359; fax: +61 7 3214 2308. E-mail address: [email protected] (R.R.J. McAllister). 1389-9341/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.forpol.2006.10.004
is that while their rural populations are dependent on intact forest ecosystems, national development aspirations often depend on logging, and hence on reducing forest resources. What constitutes sustainable forestry in these countries is not the topic of this paper. Rather the topic is that of institutional change, and in particular the balance of power between formal and informal institutions. Formal government institutions in many developing south-east Asian countries are not strong. Semi-autonomous logging operations controlled by police forces and the military tend to control many areas of their
1080
R.R.J. McAllister et al. / Forest Policy and Economics 9 (2007) 1079–1089
economies. Of particular concern for governments in these countries is that logging operations and their related cash flows are often controlled by such informal institutions. This current informal institutional setting violates general principles for robustness because, among other things, it lacks clearly defined rules on resource usage, mechanisms for resolving conflict and adequate monitoring (Ostrom, 1990). Accordingly, the deforestation literature commonly recommends government policies aimed at strengthening institutions, mainly legal (IUCN, 2001). However, a circular argument emerges: weak institutions do not control the cash flows often required to dislodge incumbent informal institutions, and therefore formal institutions find it difficult to establish their influence. Thus, institutional settings tend to be persistent, for better or worse (Dietzl et al., 2003). Some economists would argue that the informal institutions lower transaction costs and therefore are an important and positive force in their respective systems (Williamson, 1985; Williamson, 1998; Mankiw, 2000). Certainly some institutional setting is a prerequisite to large-scale logging operations, and with weak formal governance, the informal institutions at least facilitate and regulate trade (Coolidge and Rose-Ackerman, 1997). After all, the informal logging institutions generate rent from forest resources just as the formal sector would. Establishing formal sector dominance of logging activity would certainly benefit not only the formal sector itself, but also the entire economy (Tanzi, 1997; McAllister and Bulmer, 2002). Relative to those formal, informal logging systems tend to (1) fail to distribute profits fairly, (2) fail to include the negative environmental externalities of logging into log prices, (3) fail to consider long-term sustainability of the industry, and (4) fail to consider broader social issues in the economy in general. In summary, informal logging institutions consider forests as open-access resources and exploit them accordingly (Clarke et al., 1993). Instituting formal governance therefore remains an important policy arena for resource-rich, money-poor Asian countries. This paper examines the balance of power between formal and informal logging institutions in south-east Asian countries. We do not assert that our model represents a particular country, nor the complete part of the problem. Rather we use a model to make a point about forest institutions. Using phase diagrams derived
from a simple formalization of individual decisionmaking processes, we first show how informal institutions come to dominate governance. Second, we show that given dominance, informal institutions are resistant to change. We structure our paper by first presenting some background and related literature. Methods then results follow, and we conclude with a discussion of policy implications, and a summary. 2. Informal logging institutions in south-east Asia An institutional framework defines behavioral regularities (Ostrom et al., 1994) through human-devised constraints, or rules, that structure human interactions (North, 1993b). These rules can be formal, like statute law, common law and regulations, informal, like conventions, norms of behavior or codes of conduct, or they can incorporate both characteristics (North, 1993a). In south-east Asia a large degree of logging activity is facilitated through informal conventions. Such activity is prohibited by the formal institutions. Technically, this makes such activity illegal, but in this paper we refer to this activity as informal logging. Other literature does define this type of logging as “illegal logging” (for example, Clarke et al., 1993; Dudley et al., 1995; McAllister and Bulmer, 2002; Blundell and Gullison, 2003). People are governed by a dominating institution, whether this institution is a formal government or otherwise (Klooster, 2000). Therefore, here we choose the term informal because from the perspective of the residents in many less-developed south-east Asian countries, actions chosen are based on the institutional setting in which they exist, irrespective of formalities. Cultures in more developed countries tend to do the same, except that in these cultures a greater proportion of institutions are formal (Schneider and Enste, 2000). We do not discount the rent-losses to small-land holder and communities, (stemming from issues ranging from poorly defined properties rights or lack of knowledge, Shanley and Rodrigues Gaia, 2002; Engel and Palmer, 2006 to violence) associated with such logging many developing countries, but in proposing new ideas about logging institutions we choose less confrontational language. Even if informal activity contradicts formal rules the normative position is not clear as to whether the formal rule can be seen as a dbadT rule (Feige, 1997; Leitzel, 1997). More important than the issue of legality, is the
R.R.J. McAllister et al. / Forest Policy and Economics 9 (2007) 1079–1089
concept of what constitutes moral behavior (Ostrom, 2005). Social approval of individual behavior and its feedback effects such as the presence of approval motives may lead to permanent negative effects on rule compliance (Fehr and Falk, 2002). There are cultural reasons for different perceptions of informal activity (Osborne, 1997) (though not all differences are cultural, e.g. Wei, 1999). For example, there is some empirical evidence to show that Thai bribes have to be much higher than in most other countries before it is seen as illicit, and that a certain level of payment is considered to be normal (Pasuk and Piriyarangsan, 1994). In light of cultural differences, the reference to legality becomes rather an abstract term (Casson and Obidzinski, 2002). (Klooster, 2000) Given that we are concerned with an examination of the institutional settings and the decisions of individuals rather than planners, the informal and formal distinction is more appropriate. Logging operations require some form of institutional framework. As international demand for timber developed, informal institutions facilitated logging operations where formal governance was ill-equipped to do so. In many cases, military forces were the strongest, most capable institution, and as such assumed control. This legacy is evident still. For instance, the Lao military is self-funded by logging operations (Stuart-Fox, 1997; Lang, 2001). In Cambodia the Khumer Rouge funded their guerrilla operations by logging activities (Hajari, 1999). When the Khumer Rouge disintegrated, the Cambodian military assumed control of logging operations. In Myanmar, the military not only controls logging operations, but effectively all governance, so the formal governance that we refer to in this analysis does not strictly exist. In Vietnam, the military is less involved than in Laos and Cambodia, but they still control logging operations informally. Formal legislation exists on paper, but regional semiautonomous logging operations dominate the control of logging (Lang, 2001). In Vietnam the military has at least some role, logging from within Lao borders as Lao debt repayment for services rendered in the 1955–1975 Second Indochinese War (Vietnamese War). As formal governance has developed in less-developed Asian countries, forest-property rights have become legally owned by governments. These formal institutions generally establish that governments control forest resources in a manner which should provide
1081
sustainable benefits, and perhaps more importantly, in a way that satisfies international agencies and donors. However, the broad institution of governance is not strong in many of these countries (Barrett et al., 2001) and dislodging the incumbent informal-logging institutions, which are dominated by a semi-autonomous military, has not proven easy. In some cases it appears as though government attempts to dislodge the informal logging institutions have not been concerted. But it is also possible that no concerted efforts have been made merely because governments are aware of the enormity of the task. Policies aimed at dislodging informal-logging institutions generally assume that bringing the informal institutions under strict government control is unlikely — ruling members of those institutions have too much to lose by surrendering control to the government. Rather, the establishment of a loyal military is suggested, whose role would be to police forest property rights — which in essence means wrestling control from the semi-autonomous military. Clearly this means a country may have two military forces. Having two competing militaries does not always equate to civil war. However, each “military” is best viewed as a logging institution which seeks dominance by numbers. Each institution distributes payments, presumably funded by profits from logging, in order to attract participants into their sector in the hope of gaining some critical mass (i.e. the minimum number of participants to allow an institution to be persistent without substantial external investment). In many developing south-east Asian countries, the informal institutions have the required critical mass. Here we use an analytical model to first demonstrate how such informal institutions can dominate the control of logging systems, and second how government spending on enforcing property rights may have limited lasting effect if the spending levels are not sufficient to push the system past some institutional threshold (see Walker et al., 2006 for a discussion about thresholds in socio-ecological systems). 3. Methodology Previous research into participation in the informal sector has concentrated on how taxes (Ihrig and Moe, 2004), civil service wages (Van Rijckeghem and Weder, 2001) and enforcement (Clarke et al., 1993; Ihrig and
1082
R.R.J. McAllister et al. / Forest Policy and Economics 9 (2007) 1079–1089
Moe, 2004) impact on participation in an informal sector. In this paper we explore how institutions can compete for control over logging operations. Rather than viewing deforestation from a social or forest planning perspective concerned directly with extraction rates, we turn our attentions to individuals who make decisions only on which sector to participate in. Formal governments have the backing of international donor agencies (such as the IUCN, World Bank, etc.) and a formal government dictates the country's legislative framework. However, in practice, people in these countries frequently operate based on their established traditional informal institutions. Rather than changing the existing informal institution, a more promising policy prospect is to create a new formal institution to replace the existing informal institutions. The new institution's role is twofold. The first is to enforce the legislations of the government, while the second is to switch the balance of power away from the informal institution. In many developing south-east Asian countries, informal institutions are dominant while the formal government logging institution is weak. The alliances of the power-brokers in each institution are constant. Each institution controls its labor force, but for the people in the labor force, there is little distinction between either institution aside from the pay and conditions. The penalties for being in the informal sector basically amount to a loss of income when the enforcement activity of the formal sector is effective. As a result, people participate in whichever sector they expect to earn more. Simple individual decisions have some interesting collective properties. Most important here is that what individuals expect to earn is also dependent on how many people are already participating in the respective institutions. This is because given a level of funding, the number of participants directly determines how many people share the returns. Expected returns from participating in the informal institution are also dependent on the chance of losing any returns because of enforcement activity, which directly depends on how many people are participating in the formal sector. This implies a threshold effect, where the incumbent institution is difficult to dislodge and in practice the informal institution may become dominant. It is unlikely that an individual will switch to the formal sector when informal returns are higher and there is little chance of
enforcement activity. If the formal sector dominates control, then there is little chance that participants would switch to the informal sector because with the formal sector dominating, there is a high risk associated with informal returns (i.e., higher levels of monitoring and enforcement activity). We use phase diagrams, typical of control and game theory models, which we derive below using a simple conceptual model. The phase diagrams are described using two ‘motion equations’. One defines human dynamics: how the level of participation in the informal sector relative to the formal sector changes over time, da1 / dt, where a1 is the number of people participating in the informal sector. The second defines the role of forest resources: how the effectiveness of enforcing formal government property rights changes over time as a result of the remaining stock of natural forest resources, dK / dt, where K refers to the level of activity in enforcing the property rights of the formal sector. 3.1. Formal model In our model we have individual people who make individual, myopic choices which indirectly affect some common stock which then feeds back into individuals' decision making processes (for a similar approach see Clarke et al., 1993; Angelsen, 1999).We are interested in the balance of power between an informal and formal logging institution. This balance is commonly contested between a semiautonomous military (informal sector), and a government which may seek to enforce its forest property rights through the sector it actually controls (i.e., formal sector). In the model, there are people who make a choice between allocating their effort to either the formal or informal sector. In each instance of time individuals ask themselves “where will I expect to make the most money?” Returns are assumed to depend on several interrelated factors: (1) total expected returns; (2) total number of people in each sector; and (3) additional risk. 1. The total expected returns to be shared between participants in either institution. In systems where informal institutions dominate control, public sector employees are frequently not paid and are instead expected to receive income by extracting their share of covert payments. Accordingly, in our
R.R.J. McAllister et al. / Forest Policy and Economics 9 (2007) 1079–1089
model informal sector workers expect a share of returns from illegal logging, π1, based on an allocation of total covert payments. In the formal sector, participants expect a share of the available (or budgeted) wages allocation, π2. We assume that expected returns are greater in the informal sector, because there may be a higher level of logging and formal taxes are avoided (For some anecdotal evidence of this in south-east Asia, see Walker, 1999). Because initially, at least, we are interested in the short-term threshold between institutions, we treat expected returns as static. Note that the total profits from either the informal or formal sectors are greater than the participants' expected returns, which are the amounts allocated by institutions to pay its “blue-collar” members. Total expected returns are assumed constant and exogenous, which is a simplification. 2. The total number of people participating in each sector, which we define as a1 and a2 for the informal and formal sectors respectively, and a as the total in both. Given total expected returns, having fewer participants in a sector increases expected returns per person because we assume (above) that the total returns for a given sector are constant and shared evenly between sector participants. 3. The risk associated with returns generated from participating in the informal sector 1 / ω(a1,K). If there are no participants in the formal sector (which enforces forest property rights) then there is no risk that formal property rights will be enforced. However, as the number of participants in the formal sector increases, there is an increasing chance that the formal sector will reduce informal sector returns by monitoring formal property rights (below we extend our treatment of this risk to account for forest resource dynamics). We assume that when property rights breaches are successfully policed, logs are confiscated but no further tangible penalties are enforced. The term K refers to the level of enforcement activity. Participants in the formal sector can monitor informal activity but enforcement requires physical capital and bureaucratic support. We use the function ω(a1,K) to estimate the expected percentage of returns from participating in the informal sector to be retained. This
1083
function is approximated by a distribution of the functional form, xða1 ; KÞ ¼ ðK / þ 1Þða1 =aÞ/ =ðK / þ ða1 =aÞ/ Þ: ð1Þ This assumption is a key driver of the model. Where K N 0, ϕ N 0 and 0 ≤ a1 / a ≤ 1, the function ω(a1,K) increases asymptotically as a1 increases at a rate determined by K and the shape parameter ϕ, and equals zero when a1 = 0 and one when a1 = a. For each individual person in the model, their decisions can be expressed as a myopic choice between two expected returns, E½Informal Sector ¼ xða1 ; KÞd p1 =a1 ; and E½Formal Sector ¼ p2 =a2
ð2Þ
Because we assume the individuals have no institutional preference, we simply assume that in each time period they associate themselves with the institution which is expected to provide the greatest return. The relative number of participants in the formal and informal sectors will remain stable when there is no incentive for individuals to switch between sectors. This will occur if E[Informal Sector] equals E[Formal Sector]. If E[Informal Sector] N E[Formal Sector], an individual is likely to switch to the informal sector. If E [Formal Sector] N E[Informal Sector], an individual is likely to switch to the formal sector. There are no penalties for switching between sectors. This is realistic in the economies we are analyzing. Eq. (1) states that if there are no participants in the formal sector (a1 = a), then there is no risk associated with returns. As the number of participants in the formal sector increases, there is an increasing risk. 3.2. The fast dynamics: informal sector participation The motion equation for da1 / dt = 0 is derived as the space where E[Informal Sector] = E[Formal Sector] (Eq. (2)). This defines space where there is no incentive for an individual to switch between sectors, xða1 ; KÞd p1 p2 ¼ : ð3Þ a1 a2 We define the line da1 / dt = 0 in terms of the level of enforcement activity by the formal sector K, as a da1 =dt ¼ 0 when
1084
R.R.J. McAllister et al. / Forest Policy and Economics 9 (2007) 1079–1089
function of a1 the level of participation in the informal sector. Expanding Eq. (3) to include Eq. (1) gives, ðK / þ 1Þd ða1 =aÞ/ p1 p2 ¼ : ðK / þ ða1 =aÞ/ Þ a1 a2
ð4Þ
Eq. (4) relates to individuals but we need to aggregate in order to analyze the problem from a systemwide perspective. From this point onward we redefine a1. We set a equal to one and bound a1 between 0 and 1 to represent the proportional participation rates. We then assume that each agent experiences (and perceives) the same levels of all other parameters in the model. Solving Eq. (4) for K gives our informal sector participation (i.e. human) dynamics curve, " K¼
a/1 ðp1 −a1 p1 −a1 p2 Þ
a/þ1 p1 1
þ
a1 p2 −a/1 p1
#1
/
:
ð5Þ
3.3. The slow dynamics: forest resources The individual-myopic decisions occur over a relatively short-time period, probably every logging season (poor roads often mean that logging is constrained to a dry season). In the systems we are interested in, we know there are many additional variables and interactions. These may be dynamic, non-linear and/or path dependent. They are also likely to be slower variables; that is they affect the state of the system gradually. People and governments learn new strategies, institutions and environmental systems adapt, and their level of “resilience” may be strengthened or weakened. Even with the aim of presenting a simple model, it is important to provide a link between the human driven institutional setting and the natural environment which underpins it. We therefore need to know, over longer time scales, how the system behaves in relation to the state, or remaining stock, of the forest resources. We have defined our phase space in terms of K and a1, so we develop the institutional link to resources through the role of forest resource stock on the expected returns from participating in the informal institution. We redefine K here as an endogenous variable: as above we still interpret K as the level of enforcement activity, but now we introduce the notion that it is not only affected by the level of spending on enforcement activity, but also by how much of the forest resource remains. We
define x as the remaining stock of forest resource at a point of time, c as the maximum capacity of the forest resource (both in area units), and k as a parameter which accounts for the exogenous expenditure on enforcement activity. K is now given by, x ð6Þ K ¼ k 1− : c To complement our phase-diagram analysis, space must be defined where there is no system pressure for K to change (i.e. dK / dt = 0), which is variable in time only due to the influence of x. The forest resource stock x will not change over time when the level of regeneration f(x) is equal to the rate of extraction h. Logging is therefore actually a function of a2 as well as a1 given both formal and informal sectors log, but a2 = a − a1 so it is convenient to use h(a1). Our forest dynamics are, dx=dt ¼ f ðxÞ−hða1 Þ:
ð7Þ
We present the curve f(x) as a traditional logistic growth curve (Conrad and Clark, 1987) and define h (a1) as (1 + a1 / a) · H, where H is some exogenous fixed extraction rate defined by dual informal and formal sector infrastructure constraints. Our stylized extraction rate assumes that constraints at the national level limit logging, but that the increased participation in the informal sector results in greater pressure on forest resource stocks. Our model becomes, dx=dt ¼ hf ðxÞ−hða x1 Þi h a1 i ¼ x 1− a − 1 þ dH : c a
ð8Þ
Eq. (8) is equated to zero (to define the space where dx / dt = 0) and solved for a1. Eq. (6) is solved for x and then substituted into Eq. (8) which defines the space where forest-resource stocks are stable dK / dt = 0, a1 ¼
cðk−KÞKa−hk 2 : hk 2
ð9Þ
4. Results The demarcation line da1 / dt = 0 (Eq. (2)) provides a conceptual framework for analyzing thresholds of institutional change in logging operations in
R.R.J. McAllister et al. / Forest Policy and Economics 9 (2007) 1079–1089
developing south-east Asian countries (Fig. 1). The shaded areas in Fig. 1 show space where participants' expected returns are greater in the informal sector compared to the formal sector. Anywhere in the shaded area, there is an incentive for participants to switch to the informal institution. In all areas not shaded, expected returns from the formal sector are greater than the corresponding returns from the informal sector. In the non-shaded area, participants have an incentive to switch to the formal sector. Treating K as exogenous (an assumption we relax later), Fig. 1 demonstrates how an incumbent informal sector can remain the dominant institution. When K is at the level K˜, the system contains three equilibrium points ‘1’, ‘2’ and ‘3’. The stability of each point has policy implications. Point ‘1’ is not a good outcome from the government's position. At point ‘1’ not only is the informal institution entrenched, but the equilibrium is stable. Perturbation either increasing or decreasing the number of participants in the informal sector invokes pressure to return to ‘1’ where ã individuals participate in the informal sector. Point ‘2’ is better from a government perspective but still dangerously unstable. Perturbations decreasing participation in the informal sector causes a complete exodus from the informal sector and a collapse of the informal institution. However, perturbations increasing participation in the informal sector cause the reverse effect. When informal returns
Fig. 1. Phase diagram showing the incentives of an individual to participate in either the informal or formal sector. Shaded areas show space where the expected returns from participating in the informal sector are higher than returns from participating in the formal sector.
1085
exceed formal returns, participation in the informal sector increases until point ‘1’ is reached and the dominance of the formal institution is lost. Even though point ‘3’ does not correspond with the line da1 / dt = 0, it too is an equilibrium point. At point ‘3’ there is pressure to switch from the informal to formal sector, but there are no participants in the informal sector so a boundary condition is met. Point ‘3’ is stable; any perturbation in participation invokes pressure to leave the informal sector. At point ‘3’ the formal institution is completely dominant. What Fig. 1 therefore shows is that when K = K˜, even with some spending on enforcing property rights, unless that spending is sufficient to change institutional settings, the informal logging institutions will remain dominant. Only when the level of enforcement activity is increased past the right most point on the curve da1 / dt = 0 (i.e. right of K =Kˆ ) can institutional change be brought about under such a regime (with a1 b ã). This assumes a fixed ratio between the expected returns from participation in either sector, changes in which would change the shape of the curve da1 / dt = 0. Often logging in countries where informal institutions exist is unsustainable and forest stocks have important institutional, dynamical feedbacks. We analyze these dynamics, which occur over longer time scales than do individual decisions, using the forest dynamics curve dK / dt = 0 (Fig. 2). In space above the curve dK / dt = 0 forest logging rates are at unsustainable levels. As a result, the area enforced per expenditure decreases, adding to the potential uncertainty used in the decision making of the informal logging institution participants. Below the curve dK / dt = 0 forest logging rates are at a sustainable level, so forest-resource stocks increase and accordingly K decreases. Forest dynamics (dK / dt) occur over a longer temporal scale than the dynamics of participation preferences (da1 / dt). They impact on the possible long-term outcomes for the system. In Fig. 2A points ‘1’, ‘2’ and ‘3’ are possible equilibriums, though point ‘2’ is unstable. In Fig. 2B the only long-term equilibrium is point ‘1’. The level of forest stock is linearly and negatively related to K (see Eq. (6)), and it indicates that forest dynamics determine possible longterm outcomes. This highlights an important aspect of analyzing the stability of forest based economies. What we define as stable depends on how we define
1086
R.R.J. McAllister et al. / Forest Policy and Economics 9 (2007) 1079–1089
Fig. 2. Phase diagrams for two alternative forest dynamics curves (dK / dt = 0). (A) shows a forest dynamics curve with relatively fast regeneration of forest resources and (B) shows relatively slower regeneration.
our time scale because our short-term analysis (Fig. 1) produces different equilibrium points to those obtained when a longer-time scale is considered (Fig. 2). 5. Policy implications and discussion Our model indicates that low levels of spending on property rights may be sufficient for an incumbent formal logging sector to maintain its dominance. In many developing countries, however, it is the informal institutions which are incumbent. The same low level of spending that is capable of maintaining formal institutional dominance is insufficient to push the system over some threshold because the informal institutions remain resilient to change. We argue this point with the aid of our model, but for this we ignore the long-term
dynamics of the forest resource and consider only the rapid dynamics of our myopic agents regarding whether they allocate their efforts to the informal formal logging institutions (Fig. 3). Given a system where the informal sector is dominant (point ‘1’ on Fig. 3A–C), any policy set aimed at bringing logging revenues under government control must involve a spending commitment sufficient in level and length to change the fundamentals of the system. Spending that results in either an insufficient level (Fig. 3C) or insufficient longevity (Fig. 3B) in enforcement will produce no long-term benefits. The level of spending required to bring about institutional change (Fig. 3A) may be so high as to outweigh the short-term benefits of forest property rights protection. However, the long-term benefits of controlling logging are likely to be financially rewarding. The problem for many countries is that they are financially very poor. Even those governments which could commit enough financially to maintain control of forest property rights, may not be able to spend sufficiently to wrestle control if they do not already have it. This unfortunately is particularly relevant for countries most reliant on their forest resource because without any control, they lack forest related cash-flows and typically have few alternate avenues for generating cash outside of international assistance. Cash constrained governments where informal institutions control logging are faced with a dilemma. They can spend what limited funds they may have in the hope of gaining some control over the forest resource, or acknowledge that they cannot spend sufficiently to dislodge the informal system and therefore spend nothing. The latter seems to be more common. For example, in Laos international pressure led to what was termed as the ‘New System of Resource Management’ (Government of Laos, 1992). Property rights enforcement, including the Forest Police, was legislated as part of this policy. Laos, however, is a very money-poor country. The military controls logging and accordingly the Lao government losses rent from the forest resources. Without cash, however, little has been done to change institutional settings. The Forest Police was never established and anecdotal reports indicate that public servants involved in the forest industry are frequently not paid wages. People therefore have the choice of how they allocate their labor. Given the choice between the metaphorical informal and formal logging
R.R.J. McAllister et al. / Forest Policy and Economics 9 (2007) 1079–1089
Fig. 3. Short-term policy responses: a pulse in enforcement activity aimed at bringing institutional change must be sufficient to push the system past some threshold. (A) shows policies sufficient in level and longevity to bring about institutional change. (B–C) show polices insufficient in longevity and level respectively.
1087
sector, however, the choice for Lao is quite simple (metaphorical because our model is stylized and the actual choices are more complex). The formal sector cannot afford to pay wages, so expected returns are very low. With low or non-existent returns in the formal sector, more people choose the informal sector. Since the informal sector logs unsustainably, greater returns are available to fund its workers, and there is a feedback effect; in the absence of enforcement by the formal sector, informal sector returns are not seen as risky, further entrenching the informal institution. This type of situation is described by Holling (2001) as a “maladaptive cycle”. Three characteristics seem to categorize maladaptive cycles in forest-based economies: (1) they are money poor; (2) they have a relative abundance of tropical timber; and (3) their logging industries are controlled by the informal sector. We have shown that to break out of this cycle one of two things must happen. A poor outcome is for the informal logging sector to disintegrate once the forest resource diminishes below some threshold. A better outcome is for a formal sector to establish control over logging, which we have shown requires a pulse of resources to push the institutional arrangements past some threshold (point ‘2’ in Fig. 3A). After the institutional threshold is crossed, however, a relatively smaller financial commitment is required to maintain formal sector dominance. Developing Asian economies may not have the resources required to achieve the required institutional change. If multiple informal institutions operate in clearly defined geographical regions, then one policy option is to establish the formal sector region by region, but even this option may not be financially realistic. Also many of the conservation benefits which are associated with dislodging informal logging institutions in developing south-east Asian countries may extend to richer western countries (Barrett et al., 2001). The implication is that the international community has a continued role to play in supporting institutional change in developing southeast Asian countries, and while community-based programs may have a role to play in forest conservation, any action must be supported by institutional change at the top level. The policy implications above have conveniently ignored forest dynamics, which impact on institutions less rapidly than does the participation choices of
1088
R.R.J. McAllister et al. / Forest Policy and Economics 9 (2007) 1079–1089
individuals discussed above. We included them in our model, however, as a way of reflecting on their importance. Governments have some control over forest dynamics through their role in conservation. Gaining institutional dominance of the logging sector is a critical short-term goal for governments, but in the long-term natural resource management will play an equally important role in determining how sustainable and profitable logging will be. 6. Summary Many developing Asian countries face a dilemma. Forest resources provide economic benefits to the country. For example, non-timber forest products are important for rural-based populations, and are also a major source of foreign currency. Forest property rights are legally owned by governments. The formal government institutions are supposed to control forestry in a manner that provides sustainable benefits, and perhaps more importantly in a way that satisfies international agencies and donors. However, the broad institution of formal governance in the forestry sector is not strong in many of these countries, and in some cases logging operations are controlled by a semiautonomous military. Government hopes for replacing the informal institution center around developing a loyal military which it can use to enforce its formal property rights. In this study, we have presented two competing logging sectors. We did not model a particular country, rather we presented a highly stylized sub-problem which we used to demonstrate a point about institutions. In our model, each sector seeks dominance by numbers and distributes profits from logging in order to attract workers in the hope of gaining some critical mass. In many developing south-east Asian countries, the informal institutions have achieved the required critical mass. Here we used an analytical model to first demonstrate how informal institutions can dominate the control of logging systems, and second, to show how government spending on enforcing property rights may have limited effect if the spending levels are not such as to push the system past some institutional threshold. In this regard, international agencies have a critical role to play in the sustainable management of forest resources in the region by assisting governments to
build effective institutional structures and providing funding for monitoring and enforcement activities. Acknowledgements We are grateful to Chris Stokes, Peter Roebeling, Dan Walker, Joseph Clark, Amy Thams, Rodney Beard and four anonymous reviewers for comments on various drafts on this paper. All remaining errors remain our responsibility. The first author is funded by the CSIRO's Complex Systems Science program as a post-doctoral fellow. References Angelsen, A., 1999. Agricultural expansion and deforestation: modelling the impact of population, market forces and property rights. Journal of Development Economics 58, 185–218. Barrett, C.B., Brandon, K., Gibson, C., Gjertsen, H., 2001. Conserving tropical biodiversity amid weak institutions. BioScience 51, 497–502. Blundell, A.G., Gullison, R.E., 2003. Poor regulatory capacity limits the ability of science to influence the management of mahogany. Forest Policy and Economics 5, 395–405. Casson, A., Obidzinski, K., 2002. From new order to regional autonomy: shifting dynamics of “illegal” logging in Kalimantan, Indonesia. World Development 30, 2133–2151. Clarke, H.R., Reed, W.J., Shrestha, R.M., 1993. Optimal enforcement of property rights on developing country forests subject to illegal logging. Resource and Energy Economics 15, 271–293. Conrad, J.M., Clark, C.W., 1987. Natural Resource Economics: Notes and Problems. Cambridge University Press, Cambridge. Coolidge, J., Rose-Ackerman, S., 1997. High-level rent-seeking and corruption in African regimes: theory and cases. Policy Research Working Papers, vol. 1780. World Bank. Dietzl, T., Ostrom, E., Stern, P.C., 2003. The struggle to govern the commons. Science 302, 1907–1912. Dudley, N., Jeanrenuad, J.P., Sullivan, F., 1995. Bad Harvest? The Timber Trade and the Degradation of the World's Forests. Earthscan, London. Engel, S., Palmer, C., 2006. Who owns the right? The determinants of community benefits from logging in Indonesia. Forest Policy and Economics 8, 434–446. Fehr, E., Falk, A., 2002. Psychological foundations of incentives. European Economic Review 46, 687–724. Feige, E.L., 1997. Underground activities and institutional change: productive, protective, and predatorial behavior in transition economies. In: Nelson, J.M., Tilly, C., Walker, L. (Eds.), Transforming Post-Communist Political Economies. National Academy Press, Washington DC, p. 19. Government of Laos, 1992. New System of Resource ManagementPolicy and Institutional Framework. GOL, Vientiane. Hajari, N., 1999. Wooden Solders. Time International, p. 153.
R.R.J. McAllister et al. / Forest Policy and Economics 9 (2007) 1079–1089 Holling, C.S., 2001. Understanding the complexity of economic, ecological, and social systems. Ecosystems 4, 390–405. Ihrig, J., Moe, K.S., 2004. Lurking in the shadows: the informal sector and government policy. Journal of Development Economics 73, 541–557. IUCN, 2001. Regulation of the Trade in Timber and Non-Timber Forest Products in the Lower Mekong Basin Countries. The World Conservation Union, Bangkok. Klooster, D., 2000. Institutional choice, community, and struggle: a case study of forest co-management in Mexico. World Development 28, 1–20. Lang, C., 2001. Deforestation in Vietnam, Laos and Cambodia. In: Vajpeyi, D. (Ed.), Deforestation, Environment, and Sustainable Development. Praeger, Westport, p. 112. Leitzel, J., 1997. Rule evasion in transitional Russia. In: Nelson, J.M., Tilly, C., Walker, L. (Eds.), Transforming Post-Communist Political Economies. National Academy Press, Washington DC, p. 118. Mankiw, N.G., 2000. Principles of Economics. Harcourt College Publishers, Fort Worth. McAllister, R.R.J., Bulmer, M., 2002. A game of corruption in a stylised south-east Asian timber industry: a genetic algorithm approach. In: Kozan, E., Ohuchi, A. (Eds.), Operations Research/ Management Science at Work. Kluwer, Boston, p. 93. North, D.C., 1993a. Institutions, transaction costs and productivity in the long run. Economic History 9309004, Economics Working Paper Archive at WUSTL, St Louis. North, D.C., 1993b. The new institutional economics and development. Economic History 9309002, Economics Working Paper Archive at WUSTL, St Louis. Osborne, D., 1997. Corruption as a counter-culture: attributes to bribery in local and global society. In: Rider, B.A.K. (Ed.), Corruption: the Enemy Within. Kluwer International Law, The Hague. Ostrom, E., 1990. Governing the Commons: the Evolution of Institutions for Collective Action. Cambridge University Press, Cambridge.
1089
Ostrom, E., 2005. Doing institutional analysis: digging deeper than markets and hierarchies. In: Ménard, C., Shirley, M.M. (Eds.), Handbook of New Institutional Economics. Springer, New York. Ostrom, E., Gardner, R., Walker, J., 1994. Rules, Games, and Common-Pool Resources. The University of Michigan Press, Ann Arbor. Pasuk, P., Piriyarangsan, S., 1994. Corruption and Democracy in Thailand. Silkworm Books, Chiang Mai. Schneider, F., Enste, D.H., 2000. Shadow economies: size, causes, and consequences. Journal of Economic Literature 38, 77–114. Shanley, P., Rodrigues Gaia, G., 2002. Equitable ecology: collaborative learning for local benefit in Amazonia. Agricultural Systems 73, 83–97. Stuart-Fox, M., 1997. A History of Laos. Cambridge University Press, Cambridge. Tanzi, V., 1997. Corruption: arm's length relationships and markets. In: Fiorentini, G., Peltzman, S. (Eds.), The Economics of Organised Crime. Cambridge University Press, Cambridge, p. 161. Van Rijckeghem, C., Weder, B., 2001. Bureaucratic corruption and the rate of temptation: do wages in the civil service affect corruption, and by how much? Journal of Development Economics 65, 307–331. Walker, A., 1999. Legend of the Golden Boat: Regulation, Trade and Traders in the Borderlands of Laos, Thailand, China and Burma. University of Hawai'i Press, Honolulu. Walker, B., Gunderson, L., Kinzig, A., Folke, C., Carpenter, S., Schultz, L., 2006. A handful of heuristics and some propositions for understanding resilience in social-ecological systems. Ecology and Society 11, 12. Wei, S.-J., 1999. Corruption in economic development: beneficial grease, minor annoyance, or major obstacle? Working Paper 2048. The World Bank, Washington DC. Williamson, O.E., 1985. The Economic Institutions of Capitalism. Free Press, New York. Williamson, O.E., 1998. Transaction cost economics: how it works, where it is headed. De Economist 146, 23–58.
Forest logging and institutional thresholds in developing south-east Asian economies: A conceptual model Ryan R.J. McAllister a,⁎, Alex Smajgl b , John Asafu-Adjaye c b
a CSIRO Sustainable Ecosystems, 306 Carmody Road, St. Lucia, QLD, 4067 Australia CSIRO Sustainable Ecosystems, Davies Laboratory, Private Mail Bag, Aitkenvale QLD, 4814 Australia c School of Economics, The University of Queensland, St Lucia, QLD, 4072 Australia
Received 3 February 2006; received in revised form 25 August 2006; accepted 27 October 2006
Abstract Many developing south-east Asian governments are not capturing full rent from domestic forest logging operations. Such rent losses are commonly related to institutional failures, where informal institutions tend to dominate the control of forestry activity in spite of weakly enforced regulations. Our model is an attempt to add a new dimension to thinking about deforestation. We present a simple conceptual model, based on individual decisions rather than social or forest planning, which includes the human dynamics of participation in informal activity and the relatively slower ecological dynamics of changes in forest resources. We demonstrate how incumbent informal logging operations can be persistent, and that any spending aimed at replacing the informal institutions can only be successful if it pushes institutional settings past some threshold. © 2006 Elsevier B.V. All rights reserved. Keywords: Formal and informal sectors; Institutional arrangements; Deforestation; Illegal logging; South-east Asia
1. Introduction In developing south-east Asian social-environmental systems, welfare is generally highly dependent on forest resources. In money-poor, resource-rich countries like Myanmar, Cambodia and Laos, largely rural based populations depend on tangible non-timber forest products, as well as less tangible flow-ons like water quality. One common dilemma faced in these countries ⁎ Corresponding author. Tel.: +61 7 3214 2359; fax: +61 7 3214 2308. E-mail address: [email protected] (R.R.J. McAllister). 1389-9341/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.forpol.2006.10.004
is that while their rural populations are dependent on intact forest ecosystems, national development aspirations often depend on logging, and hence on reducing forest resources. What constitutes sustainable forestry in these countries is not the topic of this paper. Rather the topic is that of institutional change, and in particular the balance of power between formal and informal institutions. Formal government institutions in many developing south-east Asian countries are not strong. Semi-autonomous logging operations controlled by police forces and the military tend to control many areas of their
1080
R.R.J. McAllister et al. / Forest Policy and Economics 9 (2007) 1079–1089
economies. Of particular concern for governments in these countries is that logging operations and their related cash flows are often controlled by such informal institutions. This current informal institutional setting violates general principles for robustness because, among other things, it lacks clearly defined rules on resource usage, mechanisms for resolving conflict and adequate monitoring (Ostrom, 1990). Accordingly, the deforestation literature commonly recommends government policies aimed at strengthening institutions, mainly legal (IUCN, 2001). However, a circular argument emerges: weak institutions do not control the cash flows often required to dislodge incumbent informal institutions, and therefore formal institutions find it difficult to establish their influence. Thus, institutional settings tend to be persistent, for better or worse (Dietzl et al., 2003). Some economists would argue that the informal institutions lower transaction costs and therefore are an important and positive force in their respective systems (Williamson, 1985; Williamson, 1998; Mankiw, 2000). Certainly some institutional setting is a prerequisite to large-scale logging operations, and with weak formal governance, the informal institutions at least facilitate and regulate trade (Coolidge and Rose-Ackerman, 1997). After all, the informal logging institutions generate rent from forest resources just as the formal sector would. Establishing formal sector dominance of logging activity would certainly benefit not only the formal sector itself, but also the entire economy (Tanzi, 1997; McAllister and Bulmer, 2002). Relative to those formal, informal logging systems tend to (1) fail to distribute profits fairly, (2) fail to include the negative environmental externalities of logging into log prices, (3) fail to consider long-term sustainability of the industry, and (4) fail to consider broader social issues in the economy in general. In summary, informal logging institutions consider forests as open-access resources and exploit them accordingly (Clarke et al., 1993). Instituting formal governance therefore remains an important policy arena for resource-rich, money-poor Asian countries. This paper examines the balance of power between formal and informal logging institutions in south-east Asian countries. We do not assert that our model represents a particular country, nor the complete part of the problem. Rather we use a model to make a point about forest institutions. Using phase diagrams derived
from a simple formalization of individual decisionmaking processes, we first show how informal institutions come to dominate governance. Second, we show that given dominance, informal institutions are resistant to change. We structure our paper by first presenting some background and related literature. Methods then results follow, and we conclude with a discussion of policy implications, and a summary. 2. Informal logging institutions in south-east Asia An institutional framework defines behavioral regularities (Ostrom et al., 1994) through human-devised constraints, or rules, that structure human interactions (North, 1993b). These rules can be formal, like statute law, common law and regulations, informal, like conventions, norms of behavior or codes of conduct, or they can incorporate both characteristics (North, 1993a). In south-east Asia a large degree of logging activity is facilitated through informal conventions. Such activity is prohibited by the formal institutions. Technically, this makes such activity illegal, but in this paper we refer to this activity as informal logging. Other literature does define this type of logging as “illegal logging” (for example, Clarke et al., 1993; Dudley et al., 1995; McAllister and Bulmer, 2002; Blundell and Gullison, 2003). People are governed by a dominating institution, whether this institution is a formal government or otherwise (Klooster, 2000). Therefore, here we choose the term informal because from the perspective of the residents in many less-developed south-east Asian countries, actions chosen are based on the institutional setting in which they exist, irrespective of formalities. Cultures in more developed countries tend to do the same, except that in these cultures a greater proportion of institutions are formal (Schneider and Enste, 2000). We do not discount the rent-losses to small-land holder and communities, (stemming from issues ranging from poorly defined properties rights or lack of knowledge, Shanley and Rodrigues Gaia, 2002; Engel and Palmer, 2006 to violence) associated with such logging many developing countries, but in proposing new ideas about logging institutions we choose less confrontational language. Even if informal activity contradicts formal rules the normative position is not clear as to whether the formal rule can be seen as a dbadT rule (Feige, 1997; Leitzel, 1997). More important than the issue of legality, is the
R.R.J. McAllister et al. / Forest Policy and Economics 9 (2007) 1079–1089
concept of what constitutes moral behavior (Ostrom, 2005). Social approval of individual behavior and its feedback effects such as the presence of approval motives may lead to permanent negative effects on rule compliance (Fehr and Falk, 2002). There are cultural reasons for different perceptions of informal activity (Osborne, 1997) (though not all differences are cultural, e.g. Wei, 1999). For example, there is some empirical evidence to show that Thai bribes have to be much higher than in most other countries before it is seen as illicit, and that a certain level of payment is considered to be normal (Pasuk and Piriyarangsan, 1994). In light of cultural differences, the reference to legality becomes rather an abstract term (Casson and Obidzinski, 2002). (Klooster, 2000) Given that we are concerned with an examination of the institutional settings and the decisions of individuals rather than planners, the informal and formal distinction is more appropriate. Logging operations require some form of institutional framework. As international demand for timber developed, informal institutions facilitated logging operations where formal governance was ill-equipped to do so. In many cases, military forces were the strongest, most capable institution, and as such assumed control. This legacy is evident still. For instance, the Lao military is self-funded by logging operations (Stuart-Fox, 1997; Lang, 2001). In Cambodia the Khumer Rouge funded their guerrilla operations by logging activities (Hajari, 1999). When the Khumer Rouge disintegrated, the Cambodian military assumed control of logging operations. In Myanmar, the military not only controls logging operations, but effectively all governance, so the formal governance that we refer to in this analysis does not strictly exist. In Vietnam, the military is less involved than in Laos and Cambodia, but they still control logging operations informally. Formal legislation exists on paper, but regional semiautonomous logging operations dominate the control of logging (Lang, 2001). In Vietnam the military has at least some role, logging from within Lao borders as Lao debt repayment for services rendered in the 1955–1975 Second Indochinese War (Vietnamese War). As formal governance has developed in less-developed Asian countries, forest-property rights have become legally owned by governments. These formal institutions generally establish that governments control forest resources in a manner which should provide
1081
sustainable benefits, and perhaps more importantly, in a way that satisfies international agencies and donors. However, the broad institution of governance is not strong in many of these countries (Barrett et al., 2001) and dislodging the incumbent informal-logging institutions, which are dominated by a semi-autonomous military, has not proven easy. In some cases it appears as though government attempts to dislodge the informal logging institutions have not been concerted. But it is also possible that no concerted efforts have been made merely because governments are aware of the enormity of the task. Policies aimed at dislodging informal-logging institutions generally assume that bringing the informal institutions under strict government control is unlikely — ruling members of those institutions have too much to lose by surrendering control to the government. Rather, the establishment of a loyal military is suggested, whose role would be to police forest property rights — which in essence means wrestling control from the semi-autonomous military. Clearly this means a country may have two military forces. Having two competing militaries does not always equate to civil war. However, each “military” is best viewed as a logging institution which seeks dominance by numbers. Each institution distributes payments, presumably funded by profits from logging, in order to attract participants into their sector in the hope of gaining some critical mass (i.e. the minimum number of participants to allow an institution to be persistent without substantial external investment). In many developing south-east Asian countries, the informal institutions have the required critical mass. Here we use an analytical model to first demonstrate how such informal institutions can dominate the control of logging systems, and second how government spending on enforcing property rights may have limited lasting effect if the spending levels are not sufficient to push the system past some institutional threshold (see Walker et al., 2006 for a discussion about thresholds in socio-ecological systems). 3. Methodology Previous research into participation in the informal sector has concentrated on how taxes (Ihrig and Moe, 2004), civil service wages (Van Rijckeghem and Weder, 2001) and enforcement (Clarke et al., 1993; Ihrig and
1082
R.R.J. McAllister et al. / Forest Policy and Economics 9 (2007) 1079–1089
Moe, 2004) impact on participation in an informal sector. In this paper we explore how institutions can compete for control over logging operations. Rather than viewing deforestation from a social or forest planning perspective concerned directly with extraction rates, we turn our attentions to individuals who make decisions only on which sector to participate in. Formal governments have the backing of international donor agencies (such as the IUCN, World Bank, etc.) and a formal government dictates the country's legislative framework. However, in practice, people in these countries frequently operate based on their established traditional informal institutions. Rather than changing the existing informal institution, a more promising policy prospect is to create a new formal institution to replace the existing informal institutions. The new institution's role is twofold. The first is to enforce the legislations of the government, while the second is to switch the balance of power away from the informal institution. In many developing south-east Asian countries, informal institutions are dominant while the formal government logging institution is weak. The alliances of the power-brokers in each institution are constant. Each institution controls its labor force, but for the people in the labor force, there is little distinction between either institution aside from the pay and conditions. The penalties for being in the informal sector basically amount to a loss of income when the enforcement activity of the formal sector is effective. As a result, people participate in whichever sector they expect to earn more. Simple individual decisions have some interesting collective properties. Most important here is that what individuals expect to earn is also dependent on how many people are already participating in the respective institutions. This is because given a level of funding, the number of participants directly determines how many people share the returns. Expected returns from participating in the informal institution are also dependent on the chance of losing any returns because of enforcement activity, which directly depends on how many people are participating in the formal sector. This implies a threshold effect, where the incumbent institution is difficult to dislodge and in practice the informal institution may become dominant. It is unlikely that an individual will switch to the formal sector when informal returns are higher and there is little chance of
enforcement activity. If the formal sector dominates control, then there is little chance that participants would switch to the informal sector because with the formal sector dominating, there is a high risk associated with informal returns (i.e., higher levels of monitoring and enforcement activity). We use phase diagrams, typical of control and game theory models, which we derive below using a simple conceptual model. The phase diagrams are described using two ‘motion equations’. One defines human dynamics: how the level of participation in the informal sector relative to the formal sector changes over time, da1 / dt, where a1 is the number of people participating in the informal sector. The second defines the role of forest resources: how the effectiveness of enforcing formal government property rights changes over time as a result of the remaining stock of natural forest resources, dK / dt, where K refers to the level of activity in enforcing the property rights of the formal sector. 3.1. Formal model In our model we have individual people who make individual, myopic choices which indirectly affect some common stock which then feeds back into individuals' decision making processes (for a similar approach see Clarke et al., 1993; Angelsen, 1999).We are interested in the balance of power between an informal and formal logging institution. This balance is commonly contested between a semiautonomous military (informal sector), and a government which may seek to enforce its forest property rights through the sector it actually controls (i.e., formal sector). In the model, there are people who make a choice between allocating their effort to either the formal or informal sector. In each instance of time individuals ask themselves “where will I expect to make the most money?” Returns are assumed to depend on several interrelated factors: (1) total expected returns; (2) total number of people in each sector; and (3) additional risk. 1. The total expected returns to be shared between participants in either institution. In systems where informal institutions dominate control, public sector employees are frequently not paid and are instead expected to receive income by extracting their share of covert payments. Accordingly, in our
R.R.J. McAllister et al. / Forest Policy and Economics 9 (2007) 1079–1089
model informal sector workers expect a share of returns from illegal logging, π1, based on an allocation of total covert payments. In the formal sector, participants expect a share of the available (or budgeted) wages allocation, π2. We assume that expected returns are greater in the informal sector, because there may be a higher level of logging and formal taxes are avoided (For some anecdotal evidence of this in south-east Asia, see Walker, 1999). Because initially, at least, we are interested in the short-term threshold between institutions, we treat expected returns as static. Note that the total profits from either the informal or formal sectors are greater than the participants' expected returns, which are the amounts allocated by institutions to pay its “blue-collar” members. Total expected returns are assumed constant and exogenous, which is a simplification. 2. The total number of people participating in each sector, which we define as a1 and a2 for the informal and formal sectors respectively, and a as the total in both. Given total expected returns, having fewer participants in a sector increases expected returns per person because we assume (above) that the total returns for a given sector are constant and shared evenly between sector participants. 3. The risk associated with returns generated from participating in the informal sector 1 / ω(a1,K). If there are no participants in the formal sector (which enforces forest property rights) then there is no risk that formal property rights will be enforced. However, as the number of participants in the formal sector increases, there is an increasing chance that the formal sector will reduce informal sector returns by monitoring formal property rights (below we extend our treatment of this risk to account for forest resource dynamics). We assume that when property rights breaches are successfully policed, logs are confiscated but no further tangible penalties are enforced. The term K refers to the level of enforcement activity. Participants in the formal sector can monitor informal activity but enforcement requires physical capital and bureaucratic support. We use the function ω(a1,K) to estimate the expected percentage of returns from participating in the informal sector to be retained. This
1083
function is approximated by a distribution of the functional form, xða1 ; KÞ ¼ ðK / þ 1Þða1 =aÞ/ =ðK / þ ða1 =aÞ/ Þ: ð1Þ This assumption is a key driver of the model. Where K N 0, ϕ N 0 and 0 ≤ a1 / a ≤ 1, the function ω(a1,K) increases asymptotically as a1 increases at a rate determined by K and the shape parameter ϕ, and equals zero when a1 = 0 and one when a1 = a. For each individual person in the model, their decisions can be expressed as a myopic choice between two expected returns, E½Informal Sector ¼ xða1 ; KÞd p1 =a1 ; and E½Formal Sector ¼ p2 =a2
ð2Þ
Because we assume the individuals have no institutional preference, we simply assume that in each time period they associate themselves with the institution which is expected to provide the greatest return. The relative number of participants in the formal and informal sectors will remain stable when there is no incentive for individuals to switch between sectors. This will occur if E[Informal Sector] equals E[Formal Sector]. If E[Informal Sector] N E[Formal Sector], an individual is likely to switch to the informal sector. If E [Formal Sector] N E[Informal Sector], an individual is likely to switch to the formal sector. There are no penalties for switching between sectors. This is realistic in the economies we are analyzing. Eq. (1) states that if there are no participants in the formal sector (a1 = a), then there is no risk associated with returns. As the number of participants in the formal sector increases, there is an increasing risk. 3.2. The fast dynamics: informal sector participation The motion equation for da1 / dt = 0 is derived as the space where E[Informal Sector] = E[Formal Sector] (Eq. (2)). This defines space where there is no incentive for an individual to switch between sectors, xða1 ; KÞd p1 p2 ¼ : ð3Þ a1 a2 We define the line da1 / dt = 0 in terms of the level of enforcement activity by the formal sector K, as a da1 =dt ¼ 0 when
1084
R.R.J. McAllister et al. / Forest Policy and Economics 9 (2007) 1079–1089
function of a1 the level of participation in the informal sector. Expanding Eq. (3) to include Eq. (1) gives, ðK / þ 1Þd ða1 =aÞ/ p1 p2 ¼ : ðK / þ ða1 =aÞ/ Þ a1 a2
ð4Þ
Eq. (4) relates to individuals but we need to aggregate in order to analyze the problem from a systemwide perspective. From this point onward we redefine a1. We set a equal to one and bound a1 between 0 and 1 to represent the proportional participation rates. We then assume that each agent experiences (and perceives) the same levels of all other parameters in the model. Solving Eq. (4) for K gives our informal sector participation (i.e. human) dynamics curve, " K¼
a/1 ðp1 −a1 p1 −a1 p2 Þ
a/þ1 p1 1
þ
a1 p2 −a/1 p1
#1
/
:
ð5Þ
3.3. The slow dynamics: forest resources The individual-myopic decisions occur over a relatively short-time period, probably every logging season (poor roads often mean that logging is constrained to a dry season). In the systems we are interested in, we know there are many additional variables and interactions. These may be dynamic, non-linear and/or path dependent. They are also likely to be slower variables; that is they affect the state of the system gradually. People and governments learn new strategies, institutions and environmental systems adapt, and their level of “resilience” may be strengthened or weakened. Even with the aim of presenting a simple model, it is important to provide a link between the human driven institutional setting and the natural environment which underpins it. We therefore need to know, over longer time scales, how the system behaves in relation to the state, or remaining stock, of the forest resources. We have defined our phase space in terms of K and a1, so we develop the institutional link to resources through the role of forest resource stock on the expected returns from participating in the informal institution. We redefine K here as an endogenous variable: as above we still interpret K as the level of enforcement activity, but now we introduce the notion that it is not only affected by the level of spending on enforcement activity, but also by how much of the forest resource remains. We
define x as the remaining stock of forest resource at a point of time, c as the maximum capacity of the forest resource (both in area units), and k as a parameter which accounts for the exogenous expenditure on enforcement activity. K is now given by, x ð6Þ K ¼ k 1− : c To complement our phase-diagram analysis, space must be defined where there is no system pressure for K to change (i.e. dK / dt = 0), which is variable in time only due to the influence of x. The forest resource stock x will not change over time when the level of regeneration f(x) is equal to the rate of extraction h. Logging is therefore actually a function of a2 as well as a1 given both formal and informal sectors log, but a2 = a − a1 so it is convenient to use h(a1). Our forest dynamics are, dx=dt ¼ f ðxÞ−hða1 Þ:
ð7Þ
We present the curve f(x) as a traditional logistic growth curve (Conrad and Clark, 1987) and define h (a1) as (1 + a1 / a) · H, where H is some exogenous fixed extraction rate defined by dual informal and formal sector infrastructure constraints. Our stylized extraction rate assumes that constraints at the national level limit logging, but that the increased participation in the informal sector results in greater pressure on forest resource stocks. Our model becomes, dx=dt ¼ hf ðxÞ−hða x1 Þi h a1 i ¼ x 1− a − 1 þ dH : c a
ð8Þ
Eq. (8) is equated to zero (to define the space where dx / dt = 0) and solved for a1. Eq. (6) is solved for x and then substituted into Eq. (8) which defines the space where forest-resource stocks are stable dK / dt = 0, a1 ¼
cðk−KÞKa−hk 2 : hk 2
ð9Þ
4. Results The demarcation line da1 / dt = 0 (Eq. (2)) provides a conceptual framework for analyzing thresholds of institutional change in logging operations in
R.R.J. McAllister et al. / Forest Policy and Economics 9 (2007) 1079–1089
developing south-east Asian countries (Fig. 1). The shaded areas in Fig. 1 show space where participants' expected returns are greater in the informal sector compared to the formal sector. Anywhere in the shaded area, there is an incentive for participants to switch to the informal institution. In all areas not shaded, expected returns from the formal sector are greater than the corresponding returns from the informal sector. In the non-shaded area, participants have an incentive to switch to the formal sector. Treating K as exogenous (an assumption we relax later), Fig. 1 demonstrates how an incumbent informal sector can remain the dominant institution. When K is at the level K˜, the system contains three equilibrium points ‘1’, ‘2’ and ‘3’. The stability of each point has policy implications. Point ‘1’ is not a good outcome from the government's position. At point ‘1’ not only is the informal institution entrenched, but the equilibrium is stable. Perturbation either increasing or decreasing the number of participants in the informal sector invokes pressure to return to ‘1’ where ã individuals participate in the informal sector. Point ‘2’ is better from a government perspective but still dangerously unstable. Perturbations decreasing participation in the informal sector causes a complete exodus from the informal sector and a collapse of the informal institution. However, perturbations increasing participation in the informal sector cause the reverse effect. When informal returns
Fig. 1. Phase diagram showing the incentives of an individual to participate in either the informal or formal sector. Shaded areas show space where the expected returns from participating in the informal sector are higher than returns from participating in the formal sector.
1085
exceed formal returns, participation in the informal sector increases until point ‘1’ is reached and the dominance of the formal institution is lost. Even though point ‘3’ does not correspond with the line da1 / dt = 0, it too is an equilibrium point. At point ‘3’ there is pressure to switch from the informal to formal sector, but there are no participants in the informal sector so a boundary condition is met. Point ‘3’ is stable; any perturbation in participation invokes pressure to leave the informal sector. At point ‘3’ the formal institution is completely dominant. What Fig. 1 therefore shows is that when K = K˜, even with some spending on enforcing property rights, unless that spending is sufficient to change institutional settings, the informal logging institutions will remain dominant. Only when the level of enforcement activity is increased past the right most point on the curve da1 / dt = 0 (i.e. right of K =Kˆ ) can institutional change be brought about under such a regime (with a1 b ã). This assumes a fixed ratio between the expected returns from participation in either sector, changes in which would change the shape of the curve da1 / dt = 0. Often logging in countries where informal institutions exist is unsustainable and forest stocks have important institutional, dynamical feedbacks. We analyze these dynamics, which occur over longer time scales than do individual decisions, using the forest dynamics curve dK / dt = 0 (Fig. 2). In space above the curve dK / dt = 0 forest logging rates are at unsustainable levels. As a result, the area enforced per expenditure decreases, adding to the potential uncertainty used in the decision making of the informal logging institution participants. Below the curve dK / dt = 0 forest logging rates are at a sustainable level, so forest-resource stocks increase and accordingly K decreases. Forest dynamics (dK / dt) occur over a longer temporal scale than the dynamics of participation preferences (da1 / dt). They impact on the possible long-term outcomes for the system. In Fig. 2A points ‘1’, ‘2’ and ‘3’ are possible equilibriums, though point ‘2’ is unstable. In Fig. 2B the only long-term equilibrium is point ‘1’. The level of forest stock is linearly and negatively related to K (see Eq. (6)), and it indicates that forest dynamics determine possible longterm outcomes. This highlights an important aspect of analyzing the stability of forest based economies. What we define as stable depends on how we define
1086
R.R.J. McAllister et al. / Forest Policy and Economics 9 (2007) 1079–1089
Fig. 2. Phase diagrams for two alternative forest dynamics curves (dK / dt = 0). (A) shows a forest dynamics curve with relatively fast regeneration of forest resources and (B) shows relatively slower regeneration.
our time scale because our short-term analysis (Fig. 1) produces different equilibrium points to those obtained when a longer-time scale is considered (Fig. 2). 5. Policy implications and discussion Our model indicates that low levels of spending on property rights may be sufficient for an incumbent formal logging sector to maintain its dominance. In many developing countries, however, it is the informal institutions which are incumbent. The same low level of spending that is capable of maintaining formal institutional dominance is insufficient to push the system over some threshold because the informal institutions remain resilient to change. We argue this point with the aid of our model, but for this we ignore the long-term
dynamics of the forest resource and consider only the rapid dynamics of our myopic agents regarding whether they allocate their efforts to the informal formal logging institutions (Fig. 3). Given a system where the informal sector is dominant (point ‘1’ on Fig. 3A–C), any policy set aimed at bringing logging revenues under government control must involve a spending commitment sufficient in level and length to change the fundamentals of the system. Spending that results in either an insufficient level (Fig. 3C) or insufficient longevity (Fig. 3B) in enforcement will produce no long-term benefits. The level of spending required to bring about institutional change (Fig. 3A) may be so high as to outweigh the short-term benefits of forest property rights protection. However, the long-term benefits of controlling logging are likely to be financially rewarding. The problem for many countries is that they are financially very poor. Even those governments which could commit enough financially to maintain control of forest property rights, may not be able to spend sufficiently to wrestle control if they do not already have it. This unfortunately is particularly relevant for countries most reliant on their forest resource because without any control, they lack forest related cash-flows and typically have few alternate avenues for generating cash outside of international assistance. Cash constrained governments where informal institutions control logging are faced with a dilemma. They can spend what limited funds they may have in the hope of gaining some control over the forest resource, or acknowledge that they cannot spend sufficiently to dislodge the informal system and therefore spend nothing. The latter seems to be more common. For example, in Laos international pressure led to what was termed as the ‘New System of Resource Management’ (Government of Laos, 1992). Property rights enforcement, including the Forest Police, was legislated as part of this policy. Laos, however, is a very money-poor country. The military controls logging and accordingly the Lao government losses rent from the forest resources. Without cash, however, little has been done to change institutional settings. The Forest Police was never established and anecdotal reports indicate that public servants involved in the forest industry are frequently not paid wages. People therefore have the choice of how they allocate their labor. Given the choice between the metaphorical informal and formal logging
R.R.J. McAllister et al. / Forest Policy and Economics 9 (2007) 1079–1089
Fig. 3. Short-term policy responses: a pulse in enforcement activity aimed at bringing institutional change must be sufficient to push the system past some threshold. (A) shows policies sufficient in level and longevity to bring about institutional change. (B–C) show polices insufficient in longevity and level respectively.
1087
sector, however, the choice for Lao is quite simple (metaphorical because our model is stylized and the actual choices are more complex). The formal sector cannot afford to pay wages, so expected returns are very low. With low or non-existent returns in the formal sector, more people choose the informal sector. Since the informal sector logs unsustainably, greater returns are available to fund its workers, and there is a feedback effect; in the absence of enforcement by the formal sector, informal sector returns are not seen as risky, further entrenching the informal institution. This type of situation is described by Holling (2001) as a “maladaptive cycle”. Three characteristics seem to categorize maladaptive cycles in forest-based economies: (1) they are money poor; (2) they have a relative abundance of tropical timber; and (3) their logging industries are controlled by the informal sector. We have shown that to break out of this cycle one of two things must happen. A poor outcome is for the informal logging sector to disintegrate once the forest resource diminishes below some threshold. A better outcome is for a formal sector to establish control over logging, which we have shown requires a pulse of resources to push the institutional arrangements past some threshold (point ‘2’ in Fig. 3A). After the institutional threshold is crossed, however, a relatively smaller financial commitment is required to maintain formal sector dominance. Developing Asian economies may not have the resources required to achieve the required institutional change. If multiple informal institutions operate in clearly defined geographical regions, then one policy option is to establish the formal sector region by region, but even this option may not be financially realistic. Also many of the conservation benefits which are associated with dislodging informal logging institutions in developing south-east Asian countries may extend to richer western countries (Barrett et al., 2001). The implication is that the international community has a continued role to play in supporting institutional change in developing southeast Asian countries, and while community-based programs may have a role to play in forest conservation, any action must be supported by institutional change at the top level. The policy implications above have conveniently ignored forest dynamics, which impact on institutions less rapidly than does the participation choices of
1088
R.R.J. McAllister et al. / Forest Policy and Economics 9 (2007) 1079–1089
individuals discussed above. We included them in our model, however, as a way of reflecting on their importance. Governments have some control over forest dynamics through their role in conservation. Gaining institutional dominance of the logging sector is a critical short-term goal for governments, but in the long-term natural resource management will play an equally important role in determining how sustainable and profitable logging will be. 6. Summary Many developing Asian countries face a dilemma. Forest resources provide economic benefits to the country. For example, non-timber forest products are important for rural-based populations, and are also a major source of foreign currency. Forest property rights are legally owned by governments. The formal government institutions are supposed to control forestry in a manner that provides sustainable benefits, and perhaps more importantly in a way that satisfies international agencies and donors. However, the broad institution of formal governance in the forestry sector is not strong in many of these countries, and in some cases logging operations are controlled by a semiautonomous military. Government hopes for replacing the informal institution center around developing a loyal military which it can use to enforce its formal property rights. In this study, we have presented two competing logging sectors. We did not model a particular country, rather we presented a highly stylized sub-problem which we used to demonstrate a point about institutions. In our model, each sector seeks dominance by numbers and distributes profits from logging in order to attract workers in the hope of gaining some critical mass. In many developing south-east Asian countries, the informal institutions have achieved the required critical mass. Here we used an analytical model to first demonstrate how informal institutions can dominate the control of logging systems, and second, to show how government spending on enforcing property rights may have limited effect if the spending levels are not such as to push the system past some institutional threshold. In this regard, international agencies have a critical role to play in the sustainable management of forest resources in the region by assisting governments to
build effective institutional structures and providing funding for monitoring and enforcement activities. Acknowledgements We are grateful to Chris Stokes, Peter Roebeling, Dan Walker, Joseph Clark, Amy Thams, Rodney Beard and four anonymous reviewers for comments on various drafts on this paper. All remaining errors remain our responsibility. The first author is funded by the CSIRO's Complex Systems Science program as a post-doctoral fellow. References Angelsen, A., 1999. Agricultural expansion and deforestation: modelling the impact of population, market forces and property rights. Journal of Development Economics 58, 185–218. Barrett, C.B., Brandon, K., Gibson, C., Gjertsen, H., 2001. Conserving tropical biodiversity amid weak institutions. BioScience 51, 497–502. Blundell, A.G., Gullison, R.E., 2003. Poor regulatory capacity limits the ability of science to influence the management of mahogany. Forest Policy and Economics 5, 395–405. Casson, A., Obidzinski, K., 2002. From new order to regional autonomy: shifting dynamics of “illegal” logging in Kalimantan, Indonesia. World Development 30, 2133–2151. Clarke, H.R., Reed, W.J., Shrestha, R.M., 1993. Optimal enforcement of property rights on developing country forests subject to illegal logging. Resource and Energy Economics 15, 271–293. Conrad, J.M., Clark, C.W., 1987. Natural Resource Economics: Notes and Problems. Cambridge University Press, Cambridge. Coolidge, J., Rose-Ackerman, S., 1997. High-level rent-seeking and corruption in African regimes: theory and cases. Policy Research Working Papers, vol. 1780. World Bank. Dietzl, T., Ostrom, E., Stern, P.C., 2003. The struggle to govern the commons. Science 302, 1907–1912. Dudley, N., Jeanrenuad, J.P., Sullivan, F., 1995. Bad Harvest? The Timber Trade and the Degradation of the World's Forests. Earthscan, London. Engel, S., Palmer, C., 2006. Who owns the right? The determinants of community benefits from logging in Indonesia. Forest Policy and Economics 8, 434–446. Fehr, E., Falk, A., 2002. Psychological foundations of incentives. European Economic Review 46, 687–724. Feige, E.L., 1997. Underground activities and institutional change: productive, protective, and predatorial behavior in transition economies. In: Nelson, J.M., Tilly, C., Walker, L. (Eds.), Transforming Post-Communist Political Economies. National Academy Press, Washington DC, p. 19. Government of Laos, 1992. New System of Resource ManagementPolicy and Institutional Framework. GOL, Vientiane. Hajari, N., 1999. Wooden Solders. Time International, p. 153.
R.R.J. McAllister et al. / Forest Policy and Economics 9 (2007) 1079–1089 Holling, C.S., 2001. Understanding the complexity of economic, ecological, and social systems. Ecosystems 4, 390–405. Ihrig, J., Moe, K.S., 2004. Lurking in the shadows: the informal sector and government policy. Journal of Development Economics 73, 541–557. IUCN, 2001. Regulation of the Trade in Timber and Non-Timber Forest Products in the Lower Mekong Basin Countries. The World Conservation Union, Bangkok. Klooster, D., 2000. Institutional choice, community, and struggle: a case study of forest co-management in Mexico. World Development 28, 1–20. Lang, C., 2001. Deforestation in Vietnam, Laos and Cambodia. In: Vajpeyi, D. (Ed.), Deforestation, Environment, and Sustainable Development. Praeger, Westport, p. 112. Leitzel, J., 1997. Rule evasion in transitional Russia. In: Nelson, J.M., Tilly, C., Walker, L. (Eds.), Transforming Post-Communist Political Economies. National Academy Press, Washington DC, p. 118. Mankiw, N.G., 2000. Principles of Economics. Harcourt College Publishers, Fort Worth. McAllister, R.R.J., Bulmer, M., 2002. A game of corruption in a stylised south-east Asian timber industry: a genetic algorithm approach. In: Kozan, E., Ohuchi, A. (Eds.), Operations Research/ Management Science at Work. Kluwer, Boston, p. 93. North, D.C., 1993a. Institutions, transaction costs and productivity in the long run. Economic History 9309004, Economics Working Paper Archive at WUSTL, St Louis. North, D.C., 1993b. The new institutional economics and development. Economic History 9309002, Economics Working Paper Archive at WUSTL, St Louis. Osborne, D., 1997. Corruption as a counter-culture: attributes to bribery in local and global society. In: Rider, B.A.K. (Ed.), Corruption: the Enemy Within. Kluwer International Law, The Hague. Ostrom, E., 1990. Governing the Commons: the Evolution of Institutions for Collective Action. Cambridge University Press, Cambridge.
1089
Ostrom, E., 2005. Doing institutional analysis: digging deeper than markets and hierarchies. In: Ménard, C., Shirley, M.M. (Eds.), Handbook of New Institutional Economics. Springer, New York. Ostrom, E., Gardner, R., Walker, J., 1994. Rules, Games, and Common-Pool Resources. The University of Michigan Press, Ann Arbor. Pasuk, P., Piriyarangsan, S., 1994. Corruption and Democracy in Thailand. Silkworm Books, Chiang Mai. Schneider, F., Enste, D.H., 2000. Shadow economies: size, causes, and consequences. Journal of Economic Literature 38, 77–114. Shanley, P., Rodrigues Gaia, G., 2002. Equitable ecology: collaborative learning for local benefit in Amazonia. Agricultural Systems 73, 83–97. Stuart-Fox, M., 1997. A History of Laos. Cambridge University Press, Cambridge. Tanzi, V., 1997. Corruption: arm's length relationships and markets. In: Fiorentini, G., Peltzman, S. (Eds.), The Economics of Organised Crime. Cambridge University Press, Cambridge, p. 161. Van Rijckeghem, C., Weder, B., 2001. Bureaucratic corruption and the rate of temptation: do wages in the civil service affect corruption, and by how much? Journal of Development Economics 65, 307–331. Walker, A., 1999. Legend of the Golden Boat: Regulation, Trade and Traders in the Borderlands of Laos, Thailand, China and Burma. University of Hawai'i Press, Honolulu. Walker, B., Gunderson, L., Kinzig, A., Folke, C., Carpenter, S., Schultz, L., 2006. A handful of heuristics and some propositions for understanding resilience in social-ecological systems. Ecology and Society 11, 12. Wei, S.-J., 1999. Corruption in economic development: beneficial grease, minor annoyance, or major obstacle? Working Paper 2048. The World Bank, Washington DC. Williamson, O.E., 1985. The Economic Institutions of Capitalism. Free Press, New York. Williamson, O.E., 1998. Transaction cost economics: how it works, where it is headed. De Economist 146, 23–58.