Causes of Government failures in resource management

Back to contents - Previous file - Next file

Everybody seems to agree today that this centralized approach has been an outright failure in the sense that natural resources have not been better managed than before. Even though a rigorous demonstration is impossible, there are some grounds (to be made more explicit below) to believe that things have actually got worse than they would have been under an alternative management regime. Just consider the case of Indian forestry. In 1980, close to 23 per cent of the Indian territory was claimed to be covered with forests by the government since this was the proportion of Indian land controlled by the Forest Department. This figure compares unfavourably with the objective of 33 per cent stated in the report of the National Commission on Agriculture as the proportion of forests required to maintain the ecological balance of the country and at the same time meet the domestic demand for forest products. If data collected through observation by satellites are to be trusted, the situation is actually much worse than that conveyed by using the official figures. According to this new estimate, indeed, forests represented only 14 per cent of Indian territory during the years 1980-2 (Desai, 1988: 192), implying that a large part of the government-controlled forests are not forests at all. Comparing satellite data for 1972-5 and for 1980-2, we reach an estimated rate of deforestation of almost 19 per cent over a period of seven years. This represents an average annual loss of 1.3 million hectares of forest, that is, eight times as much as the loss admitted by the Forest Department (ibid.)!

Increasingly, acknowledgement of failure of centralized approaches to resource management is driving governments to reconsider, amend, and even reverse their previous policies. Perhaps the most glaring illustration of such a change is the complete reversal of forest land policy which occurred in Nepal in 1978 when the new National Forestry Plan entailed devolution of responsibility for forest protection and exploitation to the panchayat level and provision for long leases of forest areas (McNicoll, 1990: 163). It was only in 1983-4, however, that transfer of forests to local control began to take place on a substantial scale (Arnold and Campbell, 1986: 426). The categories of forest transferable to local community control under the new regulation included the Panchayat Forests (entrusted to village panchayats 'for reforestation in the interest of the village community') and the Panchayat Protected Forests (entrusted 'for the purpose of protection and proper management') (ibid. 431).

In forest policy matters, India made a much more prudent turn than Nepal. The Forest Department first embarked upon a so-called social forestry programme which also proved to be disappointingly ineffective, a result of the fact that, as it is conceived, it is 'merely an extension of the state's control and a further restriction upon the use of common property resources' (Blaikie et al., 1986: 496). A more serious reorientation was obviously called for. After the Committee on Forest and Tribals in India (1982) pleaded for a broad-based approach to forest management involving the active participation of local populations (see above), the Department of Environment soon followed suit by issuing a series of 'Recommendations Regarding the Revision of the National Forest Policy' (1983) that went in exactly the same direction. After having sounded the alarm about the high rate of depletion of India's forest resources and after having deplored the government's 'false sense of complacency' about this serious predicament—'areas which have long since been stripped of cover or have been diverted to nonforest uses continue to be reported as "forest" lands in statistical returns' (ibid. 3)—the report went to propose a number of basic changes in the national forest policy, including a more community development-orientated approach involving local development bodies such as village panchayats (see section 3.8 of the report). Soon after, the Forest Department made a move towards putting this recommendation into practice by starting community forestry programmes. Unfortunately, as evidence from large social forestry projects (many of which were assisted by the World Bank) in Uttar Pradesh, Gujarat, Himachal Pradesh, Rajasthan, and West Bengal attest, community woodlots or village woodlots soon turned out to be failures. 'On account of the little interest shown by community members', the model was then modified to give considerable management authority over village woodlots to the village panchayats (Cernea, 1989: 34-5). Again, 'the slippage of community woodlots into panchayat woodlots did not remedy anything', largely because communities were not really involved and panchayats were oversized (Cernea, 1989: 35-6; Colchester, 1994: 89). The next step, known as the Joint Forest Management Policy, started by the late 1980s in the states of Orissa and West Bengal. It will be commented on further in Chapters 12 and 13.

The government of India also acknowledges the failure of watershed projects (such as the River Valley Project in the Himalayan foothills) undertaken with a view to reducing accelerated siltation of dams located on rivers and to enhancing irrigation capacity. It recognizes explicitly that its failure to involve local people in the planning of such projects, and its failure to adopt a 'multidisciplinary and integrated approach to the planning and implementation of watershed rehabilitation and development', was a major cause of poor project performance (World Bank, 1983, cited from Bromley and Cernea, 1989: 32). In neighbouring Pakistan, the situation is strikingly similar. Thus, just to quote an example, in Azad Jammu and Kashmir where the Forest Department intervened to try to stop the area's alarming deforestation and environmental deterioration, it got into open conflict with many local inhabitants. By the late 1970s, over 50,000 prosecutions for forest offences were pending in the Azad Kashmir courts, amounting to about one family in six involved in an alleged forest offence (Cernea, 1989: 11).

The reasons behind the failure of many Third World governments to manage their natural resources are numerous and complex. In the following pages, we shall be content with pinpointing the most important of them which have usually the effect of lowering the morale and reducing the incentives of local populations for managing village-level resources in an ecologically sound manner.

  1. A major difficulty with any centralized approach to resource management is a problem of information. Given the great diversity of resource types, it is difficult to establish straightforward management prescriptions that can be widely followed. No government agency can know local realities in sufficient detail to conceive of valid solutions to the highly differentiated ecological problems that arise at village level (Arnold and Campbell, 1986: 442; Dasgupta and Mäler, 1990. 24). This applies with special force to tropical areas where the number of species available (in the ocean, in the forest, etc.) is considerably larger than in temperate climatic zones. The government is clearly at a disadvantage compared to the historic users who can be expected to possess extensive knowledge of local resources and constraints (Zufferey, 1986: 86). In keeping with what we have argued in Chapter IO, it must however be added that, since the village environment has changed rapidly during recent decades, local inhabitants are likely to be imperfectly informed about ongoing processes of resource depletion and degradation. Consequently, the information gap between specialized government agencies and villagers has probably narrowed down and the latter may need external assistance to help them to better assess their resource problems and to conceive and put into effect viable solutions to them. What is required is thus a two-way process of information-sharing between rural people and the administration.
  2. Another considerable difficulty which arises in connection with resource management by government agencies is of course that of enforcement of government rules (see above, Chapter 8). Thus, for example, Arnold and Campbell have emphasized that in Nepal the nationalization of all forest lands (see above) led to an unsolvable problem of policy implementation: 'The desirable objectives of this new policy proved very difficult to achieve. Effective government supervision of thousands of patches of forest scattered through remote hill terrain, accessible only with extreme difficulty, proved impossible' (Arnold and Campbell, 1986: 430). This certainly holds true for fisheries as well. On the one hand, inland fisheries are often as highly scattered and inaccessible as patches of tropical forest while supervision measures need to be specified in great detail and tailored to the differentiated needs of each local situation. Just to give one example, it is extremely difficult to enforce mesh-size regulations when fishermen can avoid purchasing nets from central purchase points because they themselves make their nets or because they can buy smuggled nets from a neighbouring country (a common occurrence in many Third World countries with long and difficult-to-control borders). On the other hand, the monitoring of offshore fisheries is an especially arduous task given the extensive nature of the resource. This is particularly true in regard of small-scale fishermen (who form the great bulk of the marine fishing labour force in developing countries) since their fishing crafts are designed in such a way that they can be launched from, and landed on, the beaches where or near which their highly dispersed villages are usually located. As a result, the possibility for the administration to control them from harbour points is precluded.

To these considerations, one must add the constraint arising from the limited funds and staff available to enforce resource-preserving rules, a limitation typical of poor countries with many pressing priorities which may appear more important than the long-term conservation of natural resources. Thus, in Pakistan, since management of the guzara forests is only one of the myriad functions of the civil administration, there is a 'glaring absence of personnel specifically earmarked to look after these forests' and this has disastrous consequences for their future survival (Azhar, 1993: 122). Regarding fishery management, which coastal developing country can be expected to have an effective police fleet continuously patrolling the open sea to check the fishing practices of tens of thousands of small-scale fishing units or of foreign vessels?

There are at least two important consequences of this situation. First, given the absence or lack of effective supervision by state authorities, a large discrepancy has arisen between the formal decision-making system that is supposed to govern resource use and the actual pattern of interaction among resource users (Cruz, 1986: 126). More precisely, what state authorities have created in many resource domains is a de facto system of open access where, before, there often existed common property regimes regulated by local authorities or organizations. For example, one Malian official from Mopti characterized the situation in the Niger River delta 'as empty of any institutional or tenurial basis for resource management and control' (I awry, 1989a: 5). From this perspective, the 'emasculation' of local organizations, wherever they existed, appears to be a tragic outcome. Thus, with reference to woodstock management in the Sahel, the following has been observed: 'As it happened, most villages had lost their power of independent activity as the result of efforts of both the colonial and independent regimes to establish controls over major forms of organisation in rural areas. Villages (or quarters within them) had no authority to enforce sanctions against violators of locally devised use rules. In practice, few such rules appear to have been made' (Thomson et al., 1986: 399). A similar phenomenon has been noted with respect to fishing in the Niger River delta: 'Because of the limited extent of traditional village lands, the fisherman passed from a minuscule jurisdiction to a larger jurisdiction, with the resulting weakening of traditional control structures without the practical operation of a substitute control system' (Kone, 1985: 100).

In the latter instance, as in so many others, what the new rules have actually meant on the ground is the uneasy 'cohabitation of two power bodies' (traditional and modern) which may all too readily enter into conflict with each other. Thus, according to state formal rules, as we have already noted, outsiders are allowed free access to local fisheries. In practice, however, no outside fisherman, at least till recently, dared throw his net without the prior agreement of the local water master, which involved the regular making of small gifts during the period of his stay in the locality. (This is because an agreement with the water master symbolizes an agreement with the invisible powers believed to rule over the local water space.) At the same time, the ability of local water masters to impose conservation rules or restrictions, particularly on outsiders, is gradually eroded in so far as it is not legally backed by the State. Groups of outsiders equipped with modern gears are less and less ready to comply with local traditional rules (which often prohibit the use of new techniques) as the social prestige of local authorities is reduced and as they can refer to new rules more congenial to their interests (Jeay, 1984b; Kone, 1985; Fay, 1990b). We will return to this important point at a later stage of this chapter.

The problem is further exacerbated when administrative boundaries within which state management rules are supposed to apply cut across traditional social or natural ecological units. This has clearly happened in Mali where the redrawing of cercle and district boundaries throughout the delta region has obliterated the former rough correspondence between social and political units on the one hand and areas of resource use on the other (Lawry, 1989a: 5; see also Fay, 1990b: 233-4). Likewise, in Botswana, greater centralization at national and district levels has caused internal boundaries to become less important with the result that the movement of herd owners between groups and areas was greatly facilitated. This increased mobility across what had formerly been separate pasture areas as well as the erosion of the authority of local chiefs had the effect of increasingly transforming the country's pastures into an open-access resource (Peters, 1987: 187-8).

A second consequence of ineffective supervision of resource use by government agencies is the high incidence of field-officer discretion in imposing prohibitions or in levying fines (see, e.g., Blaikie et al., 1986: 490; Toulmin, 1991: 28). The situation thus resembles the external-authority game with incomplete information described in Chapter 8: to recall, when the probability that the external authority makes errors in meting out punishments is sufficiently high and this imperfect monitorability effect is not compensated by sufficiently high fines or sufficiently severe sanctions, resource users may be confronted with a prisoner's dilemma situation and therefore be induced to free ride or violate state regulations. This effect actually concurs with the above conclusion that the extreme difficulty of supervising the users' actions very closely tends to create a de facto open-access situation. Rule violations are of two major types: (1) overuse or overextraction of a common resource over and above the limits set by the State, and (2) theft of state property (Blaikie et al., 1986: 493). In Tamil Nadu, for example, the first type is mostly represented by overextraction of fuelwood and overuse of grazing land by goats while the theft of state property of timbers (such as sandalwood) clearly falls under the second type.

Concerning the latter, we are also told that illegal acts are not necessarily committed by local inhabitants—at least by poor or middle-income villagers—since most serious and blatant violations occur when 'a few private individuals, often backed by considerable capital and equipment, do mount raids on these trees' (Blaikie et a/., 1986: 494). In Senegal, to take another example, woodcutters are given permits by the forestry service that often let them enter into forest areas to cut down trees of vital importance to the survival of local communities (Bergeret and Ribot, 1990: 135-9, cited from Freudenberger and Mathieu, 1993: 22). In Guatemala, likewise, we learn that tin any encounter between the police and the bark-strippers, it was usual for the latter to bribe their we. out of trouble. This was but one aspect of the broader problem facing many Indian communities in Guatemala where the rule of law had only limited application and where the rights of Indian groups were not respected' (Utting, 1994: 241).

It may also happen that the state enforcing agency does not fulfil its duties simply because of a lack of proper incentives or poor understanding of the stakes involved. Thus, regarding a government-owned irrigation system in Sri Lanka, Harriss reports that: 'Prosecutions have to be carried out by the police, who have usually treated water offences as trivial, and who do not have the same incentives to tackle them as in other cases. Further, delays over court proceedings and the very light fines, which have been imposed on those who have been found guilty of irrigation offences, have made the legal sanctions ineffectual' (Harries, 19X4: 322).

  1. It is equally important to notice that corruption often contributes to making resource conservation still more difficult. Corruption may be either supply- or demand-induced, or both. It is supply-induced when, owing to the low level of their salaries or to the irregularity with which they are paid, state agents (like forest guards) actively seek bribes from villagers who arc caught in some rule-breaking, or get themselves involved in violations of the rules which they are supposed to enforce. For example, when the Ugandan State collapsed in chaos, forest guards' salaries were no longer paid and the former guards were directly implicated in the encroachment on the forest reserves (Bruce and Fortmann, 1989: 16).

Also, in a pioneering work, Wade has shown in great detail how, in India, the effectiveness of public service organizations, such as Irrigation, Agriculture, Forestry, and Soil Conservation Departments, is seriously impaired by well-institutionalized and predictable corruption practices. These practices are actually enmeshed in a special circuit of transactions—articulated around the mechanism of transfer of officials from less to more desirable posts—in which the bureaucracy channels illegally acquired funds upwards to higher ranks and politicians who use these funds for distributing short-term material inducements in exchange for electoral support (Wade, 1982, 1985). The effects of such a politico-administrative system of corruption may be disastrous. For instance, agents of irrigation state departments who are in charge of canal management are 'under pressure to behave almost exactly contrary to the ostensible objectives of their job': instead of reducing water uncertainty, they artificially increase it so that users pay them bribes in order to get timely deliveries of water; instead of maintaining the canals in good condition, they leave large stretches of the canal unmaintained so as to save maintenance funds for other uses (Wade, 1985: 485).

Effects appear to be still more damaging in the case of soil conservation departments, given that topsoil is India's most precious resource. Thus, these departments commonly fail to solidify the newly constructed ridges, or to provide a structure for taking water from upper and lower terraces, as a result of which rainwater cuts through the unimpacted ridges at the weakest point, thereby causing gullies to form. The ensuing soil erosion can be very high since it has been reported that the rate of soil depletion from an area treated in this way is some ten times greater than from adjacent areas without the 'conservation' programme. The costs are as follows: '(1) permanent loss of production from the land on which the programme is situated; (2) semipermanent loss of production from the alluvial fans lower down on which the sediment is deposited; and (3) increased sedimentation in lower rivers causing greater flood risk and reservoir siltation' (Wade, 1985: 485-6). According to Wade, a good part of these catastrophic effects arise from the need for officers to raise money in order to get—and then keep—decent postings. The effectiveness of India's public service organizations therefore suffers from the transfer mechanism on two counts: (1) officials do not stay long in their post if they are located in poor and remote areas and, if they do, they are obsessed with the prospect of moving; and (2) in so far as transfers can be purchased, they are incited to perform their job in ways contrary to the ostensible objectives of their departments.

On the other hand, corruption is demand-induced when the fines imposed on rule-breakers are so high compared to their current incomes that they are incited to persuade the monitoring agents who caught them to reach an (illegal) compromise acceptable to both parties; or when users are powerful enough to make these agents accept a (small) bribe rather than enforce state regulations. The first possibility is clearly encouraged by the heavy fines imposed on rulebreakers by India's Forest Department under the Indian Forest Bill (see above). It is also manifest in the common practice of Tamil Nadu's villagers to 'informally arrange an annual bribe to local forest guards to facilitate the grazing of goats, for example (by a capitation "fee" of about Rs 5 per goat-owning family)'. Artisans using bamboo follow a similar arrangement (Blaikie e' al., 1986: 494; see also Karanth, 1992: 1687, for Karnataka, where the annual payment to the forest guard varies between Rs 10-15 per sheep and goat that a farmer brings to the forest for illegal grazing). Also, when official fisheries inspectors are planted on motorized fishing boats to monitor the species of fish caught, it is usual for fishermen to offer them compensation in return for their turning a blind eye on their rule violations (personal observation of J.-Ph. Platteau). Official agents may of course find themselves in a much stronger power position—at least vis-ā-vis common villagers—than what is implied in these examples. This is particularly true in the case of tribal populations which have usually no useful connection with powerful patrons or the political elite, and are therefore easily discriminated against. In such circumstances, resource users may have no other choice than to pay amounts actually exceeding the official fines. Again with reference to Tamil Nadu, we learn that

In one village, the collection of green manure from the more productive reserve forests attracts a standardized charge of Rs 80, of which Rs 36 is an unreceipted fine to forest guards. The forest guards (and perhaps forest rangers, too) have an informal organization for dividing this rent amongst themselves and for collecting it in a variety of ways. One tribal village, wellendowed with reserve forest, has forest guards who arrive two or three times a year with a lorry, make a spot-check on fuelwood stocks of households, and confiscate and remove any timber that they believe was cut green. The value of a lorryload is estimated to be at least Rs 1,000. (Blaikie et. al., 1986: 494)

In many cases, of course, corruption seems to arise as much from demand as from supply pressures without it being possible to say whether one force or the other played a major role in getting it started. Such appears to be the case in the following example. In Mehsana district in Gujarat state (India), overexploitation of groundwater resources through electrical pumping led the government to lay down precise spacing norms for new wells and to refuse electricity connections to wells whose owners violated these norms. Yet, this regulation is systematically ignored in practice because electricity connections can easily be obtained with a fee (or a bribe) to the line man (Moench, 1992: A-11). Moreover, extensive tampering with meters was one of the reasons why the electricity boards initially moved to a flat-rate charge even though this pricing structure encourages water overextraction (ibid.). Commenting on this kind of practice, a well-known Indian expert in irrigation, Dhawan, reached the bitter conclusion that the 'eroded state of ethics' in India (a loss of the values of generalized morality preached by Gandhi and his followers?), together with an inadequate administrative set-up in the countryside and the difficulty of enforcement in the case of small landholdings, make effective implementation of state regulations of groundwater use simply impossible (quoted from Moench, 1992: A-11).

Finally, it deserves to be noted that, in some cases, corruption is actively encouraged by perverse pieces of legislation such as the 1974 Land Registration Ordinance in Pakistan, which bestowed legal recognition upon most of the unauthorized appropriation of traditional common (shamilat) lands. In the words of a former senior area official, this ordinance was a legislative disaster (that) opened the floodgates of encroachments. . . The result was that brazen-faced encroachments were made into the very heart of forest lands. Here was an opportunity for unscrupulous revenue officers to oblige friends and relatives or make hay while the sun shone at the cost of rich forests and vegetative covering of hills. (Cited from Cernea, 1989: 20)

  1. The incentive to rule violation or at least to non-co-operation on the part of resource users is almost always increased by the fact that relations between them and the state bureaucracy are usually distant and antagonistic and that, in many cases, state regulations have had the effect of setting the government against the peasant when successful resource management precisely requires the opposite circumstances (Bromley and Cernea, 1989: 17). In Mali, for example, far from being regarded as benevolent extension agents, 'forest agents are seen generally by the public as paramilitary police agents' (Lawry, 1989a: 15). Various types of reaction may be expected from resource users placed in these conditions. First, state agents may live so far away from users that the latter do not consider them a reasonable source of authorization (Thomson et al., 1986: 419). Second, and more critically, users tend to view local resources as government property rather than their own, an attitude that seriously erodes their motivation to protect them (Arnold and Campbell, 1986: 430; Azhar, 1993: 117-18), all the more so if the bureaucracy is strongly imbued with an ethic of regulation and control (Blaikie et al., 1986: 501) and if it provides them very few economic incentives (as when they are asked to work on new plantations without being given even the right to collect the produce of the trees planted). Bromley and Chapagain have well described this disincentive effect:

Such external influences are critical in the process of pitting villagers against themselves and of ultimately shifting resource stewardship away from the village. When resource responsibility is taken away from the village, so is the concern for the viability of the resource. It is the 'patron syndrome' turned on its head; villagers do not care much for things that the state gives to them, and the same thing would seem to apply to the things that the state takes away. (Bromley and Chapagain, 1984: 872; in the same vein, see Agarwal and Narain, 1989: 13, 27)