What if the Malian Cotton Parastatal is Privatized?
A Substantive Performance Analysis
Department of Agricultural Economics, Michigan State University
This study provides a substantive framework to understanding the privatization debate in the Malian cotton industry. A Situation-Structure-Performance (SSP) impact and change analysis is used to identify the main sources of interdependences, the relevant structures required to sort out these interdependences, and their corresponding performance. The paper argues, among others, that when farmers have alternative buyers, repayment of credit given by any one buyer is compromised. People’s beliefs and motivations appeared to be the main forces that control dynamics in the industry. The industry presents strong patterns of path dependence, irrespective of what is chosen between privatization and state coordination.
------------------------------------------------------------------------------------------------------------------
The focus on market and state as separate spheres of the society, and the urge for normative policy prescriptions, resulted in the neglect of the “legal-economic nexus” (Samuels, 1989a, 1989b) in the cotton policy debate. The paper reaches beyond this traditional market-state sorting out of societal conflicts, and explores a third way of looking at the interdependences within the cotton sub-sector in Mali. The interdependences considered here include physical cotton output at farm level, the provision of credit and the uncertainties inherent to it, the supply of marketing services at various stages of cotton production process, the distribution of potential surplus obtainable from production and marketing activities. The paper is an attempt to answer the key substantive performance question: Who gets what in the cotton industry? The aim here is to identify whose interests count or whose interests are made to count under different structural arrangements in the industry.
The paper heavily draws on the Situation-Structure-Performance paradigm (SSP hereafter), which allows for a rigorous treatment of the interdependences among different interest groups. The SSP theory (see Schmid, 1987; 2001 for details) offers a typology of situations, a list of structural arrangements and hypothesized performances, a practical dynamic grid through which most interdependence can be filtered and understood. The SSP theory allows for both impact and change analysis but the emphasis is placed in this paper on impact analysis. Four different situations are considered in the analysis, namely (i) incompatible use goods (IUG) such as seed cotton, cottonseed, cotton lint, and cotton oil; (ii) high information cost goods (HIC) such as credit-worthiness of cotton growers (e.g. information management systems); (iii) goods with economy of scale such as inputs produced or bought in bulk, ginning or processing facilities, transportation services, etc; and (iv) high exclusion cost goods (HEC) such as different social services that the state provides in cotton growing zones, or the collective goods produced by village associations.
The remaining sections of the paper expand each of the four situations above, using the SSP impact framework. A tentative change analysis is performed on village associations (VAs), looking at their evolution and sophistication over time in a changing policy environment. The paper also draws on empirical evidences, mainly from Sub-Sahara Africa, to formulate testable hypotheses that appeared plausible from the performance column in an appended SSP flow chart. The paper ends with some concluding remarks mainly based on the hypotheses.
Finished or intermediary products such as seed cotton, cottonseed, cotton lint, and cotton oil are all incompatible use goods (IUGs): the use by one party excludes the other party from using the good. There are many other IUGs in the cotton production system, including lands, farm implements, seeds and other productive inputs. The intrinsic feature of these IUGs will not change with privatization. IUGs are not the main focus in this paper, as the structural arrangements needed to sort out IUG interdependences are relatively well known. These structures, as identified in the SSP theory (Table 1), are factor ownership and degree of competition in the market. Property rights to IUGs are the least controversial in this case, though land reform is an issue in Africa (Zimbabwe for example). However, people seem generally to be at ease with the idea that the holder of property rights also holds rights to sell IUGs.
Table 1: Independences in the Malian Cotton Industries: Incompatibility Use Goods
Situation
|
Structure |
Performance |
|
IUGs include seed cotton, cottonseed, cotton oil, cotton lint, land, implements and other productive inputs |
1. Degree of competition: monopsony or competition 2. Factor ownership |
1. Known. Farmers, traders, and parastatal etc, all receive their market value. Market sorts out independences 2. People may buy or sell these goods based on their own valuation and the prices offered by market |
In the case of the Malian cotton, the market is not confined to national boundaries, since most of the cotton produced is exported in the form of lint. This is a common feature in the entire Francophone Africa where cotton is marketed through the marketing division (COPRACO) of the French Textile Development Company (CFDT). This issue will be discussed more in detail in the next sections. Market performance under a state monopsony or under a privatization monopsony regime depends not only on prices but also on policy objectives and people’s valuations of IUGs. The state-owned company may choose to forgo higher prices in the international market and to sell cotton to a domestic textile company for many valid reasons. Similarly, a domestic textile company may prefer consuming locally produced cotton instead of using cheaper sources of regional supply. All is not just a matter of prices. Passion for a national identity, pride to consume locally, solidarity towards domestic industries, knowledge that opportunities are being created and used locally, and so on, are all attributes of a national choice set. Some of these attributes are likely to be valued differently if the cotton sub-sector is privatized. The change in emotional attributes will also shift the opportunity set available to different parties in the industry. For example, Malian farmers showed their dissatisfaction in the system by boycotting cotton production during the 1999-2000 campaign.
The cotton market itself presents many dynamics, reflecting the rapid change in consumers’ tastes and preferences. For example, the use of toxic pesticides in the region, as reported by Myers (2000), results in specific characteristics of the cotton produced. Environmentalism is rising in importing countries, and may force the production in exporting countries to observe a set of environmental rules. Should this occur, we would observe a change in the input mix, and also in the quality of products. It appears that market forces can ultimately change the intrinsic characteristics of a number of IUG used in cotton industry, and this irrespective to the degree of privatization in the industry. Obviously, the problem is slightly more complicated if we consider other elements such as advertising, which are used to alter consumer preferences. The key idea lies in the potentials of the market rules to sort out interdependences, and to feed back to the production process, which in turn adjusts to changes in market demands, following a cyclical and cumulative causation process (Arthur, 1990; Skott, 1994). Of course there are other issues such as high information cost, which we discuss in the next section with respect to the provision agricultural credit.
Cost and availability of credit is central to cotton production. Credit is needed to purchase inputs (seeds, fertilizer and pesticides), pay temporary workers (planting, weeding, picking), invest in machinery and for subsistence until harvest. It is fair to say that without any form of credit, there will be no cotton production in rural Mali. Poulton (1998) provides an interesting example from Ghana where competing cotton companies must “campaign” for farmers and the only way to get them “vote” for them is to offer deals on different packages of credits (tilling their food crop plots in addition to cotton plot).
The HIC situation in credit provision in Mali is due to the uncertainty regarding the credit worthiness of borrowers. Farms activities are largely dependent on unreliable weather conditions and a few credit institutions are willing to take the risk to lend to agricultural enterprises. There are also risks of idiosyncratic behavior of individual farmers such as work effort and farmers’ ability to repay. Besides, the dispersed spatial organization of rural communities, the lack of suitable national ID system and the absence of effective courts for contract enforcement make it difficult to track individual borrowers, thus increasing risks for defaults. Three possible structures are considered in Table 2: (i) privatized credit markets, whose functioning depends on a working collateral system; (ii) group lending as a way to use the dynamics of peer monitoring in the group; and (iii) buyer monopoly where hierarchy and functional interlinking play key role in minimizing risks for defaults.
The performance under the first structure is already stated above: evidence from Sub-Saharan Africa supports the idea that cotton production activities are likely to be relegated to lower priority in the absence of a credit system. The second structure is based on the working formula of Grameen banks (Stiglitz, 1990). Farmers are expected to self-select into groups based on trust and mutual monitoring processes. An interesting question is to know the optimal size, if any, of these groups. The idea of peer monitoring may also be viewed with respect to status grant institutions in small groups (in the Schelling (1978) sense). These institutions are built over generations and are based on internalized beliefs of group cohesiveness and responsibility towards the community. Peer monitoring is therefore a direct use of trust, or in a broader sense the use of social capital (see e.g., Fukuyama, 1993; Staatz, 1998). With regards to the third structure, one can think about a scenario where production activities are totally reorganized. Hierarchical control of credit provision may be achieved through tenant-landlord contracts (Ray, 1998). The sharecropping arrangements proposed in Ray (op. cit.) and common throughout the Asian agriculture are not likely to work in the cotton production, because farmers do not use directly the seed cotton produced. A formula that may work is the interlinking of the credit and output market.
Table 2: Independences in the Malian Cotton Industries: High Information Cost Goods
Situation
|
Structure |
Performance |
|
Credit in kind or cash from cotton company to farmers. The situation involves uncertainty, which leads to HIC about repayment capacity versus opportunism of borrowers |
1. Privatized credit markets—with collaterals; credit provided by specialized firms; No effective courts. 2. Group lending (peer monitoring) 3. Buyer monopoly: more functions linked by hierarchy—interlinked activities; tie-in sales |
1. Failure: poor farmers do not receive credit-No cotton production and no subsidiary advantage for the economy 2. Credits provided, cotton produced, other advantages also provided—prior experiences matter
3. Credit may be provided; monopolist collect risk rent from producers through interlinking credit and output markets |
While the first structure (market allocation of credit) is likely to fail under both state-control and private control, the two other structures can potentially control for risk and guarantee effective credit provision, even if interlinking may lead to other high exclusion cost and monopoly issues. These issues are discussed in section D. We now look at the question of economies of scale, which is the third source of interdependence considered in this study.
There are economies of scale in marketing services such as input supply, output collection and transportation, as well as in ginning and oil producing. All these activities use physical equipment, and most conflicting interdependences arise from the use or non-use of these physical goods. Ginneries and oil crushers need very specialized equipments, implying that their use may lead to some sort of asset specificity issues in the cotton industry (Williamson[2], 1979; 1985). Like in the previous case, three institutional structures are considered here: (i) keep the status quo, that is CDFT and national research institutes intervene in cotton seed research, CMDT coordinates all production activities, COPRACO control marketing and HUIPROMA produce oil; (ii) Liberalize some periphery functions such as input supply and transportation; and (iii) fully privatize CMDT and fully liberalize all periphery functions (Table 3).
Table 3: Independences in the Malian Cotton Industries: Economies of Scale and Uncertainty
Situation
|
Structure |
Performance |
|
· Marketing services such as input supply, output collection, transportation · Ginning, oil from cottonseed (byproduct from ginneries) The dominant characteristics are economy of scale and uncertainty (e.g., asset specificity), but there are also some IUG such as the cotton oil. |
1. Status quo: CMDT coordinate all production activities, COPRACO controls marketing and HUIPROMA produces oil 2. Some periphery functions such as input supply, transportation are liberalized 3. Full privatization of CMDT and full liberalization of all functions in wholly and competing firms |
1. Transfer pricing (to local oil factory)—Pan-territory pricing will continue; access to remote areas and possible political redistribution of resources; current structural rigidities maintained or adjusted; benefits of economies of scale effective but must trade prices of good and service against variety and quality of services 2. Economies of scale not fully “harvested”; limited cooperation among firms; proliferation of services and inputs and quality differentiation 3. Competition between different autonomous companies; networks of economies of scale; increased risk default and collusion among firms; possible increased asset specificity |
These three structures may appear to reflect not only the structural organization of the market. In fact, they also embody other important structural issues, which can theoretically be inferred from the existence of economies of scale. A look at predicted performance of these three structures will further illustrate this point. If the status quo prevails, we will continue to see transfer pricing favors to the local oil factory, because when the stages are integrated in one management there may be sweetheart deals wherein the transfer price may differ from arm length market price[3]. Pursell and Diop (1998) describes this practice as inefficient, since the cottonseed transferred to HUIPROMA have higher market value elsewhere.
This raises again the questions: what are the objectives of government policies, who are the beneficiaries of transfer pricing and who are the losers? Unless the questions are answered in the light of the specific context of the Malian legal-economic nexus, it will be too quick to conclude that such a practice is inefficient.
The persistence of the status quo also implies that the pan-territorial pricing scheme will continue. The pan-territorial pricing scheme, which primarily allows growers in remote areas to benefit from the cotton incentive package, cannot be presumptively labeled as inefficient. The concept of efficiency in policy making can only be relative, if not subjective. As previously argued, it depends on whose interests count and whose interests are made to count.
Status quo will also mean that the structural rigidities in the cotton subsector will be maintained. Also, the benefits of economies of scale will be effectively “harvested”. However, state coordination of the cotton industry will necessarily require a tradeoff between prices of goods and services on the one hand, and the variety and quality of these goods and services on the other hand. For example, seed is provided by CFDT in collaboration with the national agronomic institute. Farmers have no control over the variety of cotton they grow, which is chosen by researchers. The same issue applies to all other inputs and to different services provided to cotton growers. Liberalization, whether full or peripheral, will introduce diversity but it will increase unit cost in the absence of scale economies.
With regards to prices, liberalization is synonymous to the end of administrative pricing. While this could eventually lead to a more efficient pricing, as the termination of administered prices (and the emergence of competitive or oligopolistic pricing) will offer a diverse set of prices corresponding to a wide range of qualities supplied and demanded, it could also give rise to increased opportunism both on the part of farmers and emerging cotton firms. Inferring once again from the Ghanaian example reported in Poulton (op. cit.), one might conjecture that ending the administered prices will create an additional demand for institutions capable of monitoring farmers’ and firms’ loyalty. One could indeed fear that, unless credit and output markets are effectively interlinked[4], some farmers would fail to honor their debts towards credit providers by selling their products to other cotton companies offering slightly higher prices. Similarly, cotton companies that did not commit themselves in the risk-prone credit market will be willing to participate in the less-risky output market, providing foreseeable disincentives for credit provision. Thus, credit market would fail, as no company will take the chance to pre-commit its resources to productions that may be snatched by other companies.
According to Poulton (op. cit.) Ghanaian firms resorted to collusive pricing in order to control for farmers opportunism, but were unable to sustain the collusion, also because of opportunism. This is classical situation of a prisoner’s dilemma where firms fail to achieve a collectively desired outcome as each of them tries to take an opportunistically higher share of the cotton output market. Accepting the claim that cotton may not be produced in the absence of credit, it possible that the persistence of the farmers/firms’ opportunistic behaviors will dry up credit sources over time, and ultimately result in the elimination of cotton from the cropping system following liberalization. We show this using a normal form representation of the prisoner’s dilemma in Table 4.
Table 4: An Illustration of a Prisoner’s Dilemma between Cotton Companies Interacting in the Credit and Output Market
|
Firms’ behaviors |
Firm 2 is opportunistic in at
least one market |
Firms 2 cooperates in both
markets |
|
Firm 1 is opportunistic in at
least one market |
Firms are likely not to participate in the riskier, less-attractive credit market, but may be willing to participate in the output market. Credit is not likely to be provided. Consequently, cotton is unlikely to be produced. Even if both firms do participate in the credit market, at least one of them will be opportunistic in the output market (deviating from the collusive prices). Successive retaliations through price increases may not be sustained, or may not lead to the best outcome, which is the collusion price. A possible result is the lack of credit provision and the elimination of cotton production over time. |
If Firms 2 cooperates in the two markets, Firm 1 will defect from the credit market while trying to compete (loyally or disloyally) with Firm 2 in the output market. Being opportunistic is a dominant strategy for Firm 1, knowing that Firm 2 will cooperate. |
|
Firms 1 cooperates in both markets |
If Firms 1 cooperates in the two markets, Firm 2 will defect from the credit market while trying to compete (loyally or disloyally) with Firm 1 in the output market. |
Collectively desired outcome: both firms cooperate in both credit and output market. Farmers grow cotton and this equilibrium is sustainable overtime. However, this equilibrium will be difficultly achieved, as it is dominated by firms’ opportunistic strategies. |
While successful collusion between firms may be necessary to ensure the continuation of cotton production activities, collusion does not warrant, in itself, inefficiency. Cotton firms may collude, not to shield themselves against farmers’ opportunism, but rather to “unduly enhance” prices in their favor. Such behavior is illegal under many legal systems, and thus is usually not an option in many countries. As mentioned earlier, cotton firms were nevertheless able to collude in Ghana. OTAL (2001) reports that the collusive prices were perceived as very low by cotton growers, who threatened in year 2001 to abandon cotton production in the absence of price increases. Perhaps, firms would not have to resort to collusion in order to monitor the pervasive risk inherent to farmers’ high propensity to default, were the society endowed with enough social capital or other status grant institutions that encourage self-monitoring.
Also important to the success of the cotton business is the employees’ loyalty. Financial embezzlement in the top management of the Malian cotton parastatal (CMDT) has been one of the underlining cause of the Malian cotton crisis that resulted in a historical farmers strike and a consequent halving of cotton production in Mali between 1999 and 2000. In Ghana, Poulton (op. cit) reports intra-company opportunistic behaviors whereby employees use companies’ equipment in personal side-businesses, such as “moonlight ploughing” (p.73) in non-cotton farms. Such opportunism can arguably be controlled more effectively in private firms than in a state-managed industry. However, status grant institutions can also control for such opportunism or facilitate it.
A final important issue relates to asset specificity. Specific assets may exit the economy at a loss, as the ownership of large-scale ginneries is disarticulated. Perhaps this is an area where functional and periphery liberalization would provide an interesting result, especially for the ginneries and the oil factories. Private owners would then make “hostage” deals with seed providers and share the benefits accruing from the economies of scale. Such deals will involve negotiations and certainly coalitions, and these are obviously not free. But once the ginning services are provided, some other companies can easily benefit. One can inevitably ask who will pay the fixed costs and who will pay the marginal cost of provision of the asset specific services. The issue might arise if farmers paid for ginning and there were different prices to different farmers. Will companies instead rely on general assets? Not likely, because they will not do it by hand! However, if effective contract enforcement is lacking and the community is endowed with poor status grant resources, agents in the industry may be forced to manage uncertainties by using general assets or simply exiting the business.
The redistribution of surplus from the cotton industry, if any, will depend to a large extent on regulations. If the current structure persists, the use of CMDT profit for infrastructures and other social goods presents a typical situation of high exclusion cost (HEC). In fact, a part of CMDT profit is used to provide social infrastructure such as education, health center and rural roads. Some may be used to finance political campaigns and excessive employment in CMDT (Table 5).
Once HEC goods are
provided, their benefits accrue to everyone, irrespective of whether they grow
cotton or not and how much they grow. Proceeds from the industry are also
reinvested in the sector in the form of research. Each cotton grower benefits
equally from the research efforts, but all the farmers do not contribute
equally to the research fund (farmers have different farm size and the
contribution is a function of patronage). Also, Tefft (1999) rightly points out
that cotton earnings may be used as strategic investments in other sectors. But
one may wonder whether it is appropriate that some people work so that others
benefit. The question is an interesting one but the answer goes beyond the
objective of this paper. All depend on how stakeholders feel about the use of
their money, their actions in order to press for some specific uses, or their
indifference vis-à-vis the use of the proceeds, provided they are paid what
they perceive as a fair price for their respective efforts. Formal and
informal rules, habits, preferences, experiences, etc., each of these elements
will influence the final re-distributive outcome.
Table 5: Independences in the Malian Cotton Industries: High Exclusion Cost Goods
Situation
|
Structure |
Performance |
|
Profit from MDT and provision of social services: education, health facilities, infrastructures—HEC Village associations |
1. CMDT (Profit used for social services) 2. Privatization (Tax used for social services and to support SYCOV) 3. SYCOV relies on voluntary dues |
1. Spillover to regional non-cotton growers through different investments in the region; reinvestment into cotton sub-sector; financing government budget or strategic investment into other sectors, all are possible performance 2. Performance depends on regulations, and motivations of stakeholders; unwilling riders
3. Free riders |
There is also a high exclusion cost involved in the goods provided by village associations. This discussion is deferred to the next section, which also looks at the institutional change analysis on SYCOV or South Mali Cotton Producers’ Union.
E. CHANGE ANALYSIS: SOME ILLUSTRATIONS FROM SYCOV
This section illustrates the process of change in the Malian cotton industry over the past 70 years. This historical account is based on several publications by APM (2001). In French colonial Africa, cotton production began in 1930s. At this time, farmers were forced[5] to grow cotton, which gave them cash payment needed to pay personal taxes to the colonial administration. The colonial rules did not provide room for exit, nor voice[6]. In fact the command word under the colonial authority was “execution, not complaint”, and dissidents were subject to harsh punishments, which serve as efficient short-run reinforcers in the Skinnerian sense (Nye, 1992). We will see later that these reinforcers will gradually erode as the colonial power wanes, thus opening up avenues for both voice and exit.
CFDT took over the organization of cotton production after World War II, diffusing technologies and securing quality-based prices to producers. Loyal farmers then received price incentives for their cotton, but the opportunities for voicing or exiting were still limited under authoritarian rule. CFDT ruled the cotton sector as a monopoly for the first 15 years following the independence (1960) before CMDT took the command of some activities. However, CFDT (through COPRACO) has kept the control of cotton marketing in Mali. Despite all these changes, cotton activities have maintained their cozy integrated features[7], and this suggests these changes were highly path-dependent in the sense of North (1990). Farmers are given an input package to produce cotton, while cotton is bought at an administratively decided price. This path dependence is likely to be observed if CMDT is privatized, as is actually the case in Ghana (see Poulton, op.cit).
Since the CMDT’s control of the sector, farmers have benefited from several social services, but these advantages began to be withdrawn in the late 1980s as a result of structural adjustment programs (SAPs) under foreign pressure. The SAPs have forced CMDT to refocus its activities exclusively on the cotton sub-sector. They have also altered the terms of provision of inputs and credits, and these new changes encouraged farmers to organize and collectively fight for their interests. These events coincide with the so-called “Eastern Wind”, a democratization wave emanating from the former socialist block of Eastern Europe. The Malian dictatorship was toppled, which provided the Malian farmers with a new sense of political and economic freedom. For example, the French law of 1901 on the right to associate became widely used to support the formation of several farmers groups[8]. These farmers’ associations succeeded to negotiate two successive contracts with the state and CMDT, first over the period 1990-1994, and then for the period 1994-1998. The terms of the contracts improved to the advantage of the groups over time.
Unfavorable prices in the international cotton market combined with financial embezzlement at senior management level in CMDT, however, led in the late 1990’s to a worsening of farmers’ conditions, which ultimately resulted in the so-called Malian cotton crisis with a historical farmers’ strike and a cut-down of cotton production by half between 1999 and 2000.
One of element that increased tremendously farmers’ bargaining position was the creation of South Mali Cotton Producers’ Union (SYCOV) in 1992. SYCOV was involved in many other stages in the sub-sector: managing inputs and fixing their prices, evaluation of tenders, setting production quotas, planning collecting activities, controlling buying and weighing operations. SYCOV has access to CMDT financial information and “participates” in the management of the stabilization fund. All these were possible because of all the changes that took place in the production environment. Farmers were able to voice their concern, maintain their cohesiveness (through unified leadership) and achieve a collectively desired outcome[9]. However, these collective goods present the characteristics of HEC. Many farmers may want to free ride, but SYCOV needs resources to be operational. These resources are currently obtained through administrative taxes, automatically retained from the profit rebate received by farmers. This has inescapably created unwilling riders. Under privatization and liberalization, the union may continue to negotiate with private traders, but with no tax and rebate, voluntary contributions to sustain the activities of the union may not be sufficient to finance the union. These services may therefore not be provided. This may discourage some farmers who may choose to exit the market.
There are many sources of interdependence in the cotton industry in Mali. This paper has investigated how state-owned firms and private firms may sort out these interdependences under different structural rules. The paper is based on a substantive performance approach using a SSP framework. Hypothesized outcomes of the SSP analysis as well as evidence from the West African region suggest that the debate between proponents and opponents of privatization is not a clear-cut one. In the light of the potential substantive performance of the industry, the arguments evoked to support or oppose privatization appear to be partial. Privatization may still result in collusive prices and opportunism may prevent the development of an integrated industry, which is critical in yielding substantial economies of scale.
The cotton sector presents recurrent characteristics, which suggest that structural changes are likely to be path-dependent. Motivations and beliefs appear to be strong forces that are currently controlling and can potentially control dynamics in the cotton industry, irrespective of which state (status quo or privatization) occurs. Farmers overcame the free rider temptation long enough to organize, but not sufficient to voluntarily pay dues to support their union. Farmers have made a long march over several decades in order to gain more control of the industry. This power may slip away from them under privatization!
APM (Agricultures paysannes, sociétés et mondialisation). (2001). “La Longue Marche des Paysans du Mali Sud”, Online: www.fph.ch/French.wlproj/web%20apm/sycov2.html, November 20, 2001.
Arthur, B. (1990). “Positive Feedbacks in the Economy”, Scientific American, February 1990: 92-99
Bates, R. H. (1981). “Markets and States in Tropical Africa”, Berkeley, University of California Press.
Fukuyama, F. (1995). “Trust : The Social Virtues and the Creation of Prosperity”, New York, Free Press.
Hirschman, A. (1970). “Voice, Exit and Loyalty”. Cambridge, Havard University Press.
Myers, D. (2000). “Cotton Tales”. New Internationalist 323, May 2000. Online: www.oneworld.org/ni/issue323/tales.htm, December 2, 2001.
North, D. (1990). “Institutions, Institutional Change and Economic Performance”, Cambridge, University Press.
Nye, R.D. (1992). “The Legacy of B.F. Skinner: Concepts and Perspectives, Controversies and Misunderstandings”. Wadsash: Brooks/Cole Publishing Co.
OTAL. 2001. “Ghanaian Farmers Threaten to Abandon Cotton Production”, Cotton News, OT Africa Line, Online: www.otal.com/ccotton.htm, November 25, 2001
Poulton, C. (1998). “Cotton production and marketing in Northern Ghana: The Dynamics of Competition in a System of Interlocked Transactions”. In A. Dorward, J. Kydd and C. Poulton (eds)-Smallholder Cash Crop Production Under Market Liberalization: A New Institutional Economics Perspective. Wye College, University of London, UK
Ray, D. 1998. “Development Economics”, Princeton, Princeton University Press.
Slantchev, B. L. “Markets and States in Tropical Africa: A Review”. Online: http://www.gotterdammerung.org/books/reviews/bates-markets-africa.html, September 6, 2002
Pursell, G. and M. Diop. (1998). “Cotton Policies in Francophone Africa”, Preliminary Draft, Development Research Group, The World Bank.
Samuels, W. J. (1989a). “The Legal-Economic Nexus”. George Washington Law Review 57: 1556-78.
Samuels, W. J. (1989b). “Some Fundamentals of the Economic Role of Government”. In W.J. Samuels (ed.), “Fundamentals of the Economic Role of Government”, pp. 167-72, New York, Greenwood Press.
Schelling, T.C. (1978). Micromotives and Macrobehavior”, New York, Norton.
Schmid, A. A. (1987). Property, Power and Public Choice: An Inquiry into Law and Economics”, New York, Praeger.
Schmid, A. A. (2001). “Institutional and Behavioral Economics”, East Lansing, Walbridge Press.
Skott, P. (1994). “Cumulative Causation”, In Elgar Companion to Institutional and Evolutionary Economics, G. Hodgson (ed), pp.119-22
Seck, T.A. 1999. Le Coton en Afrique Subsaharienne Lettre de la Cade (Coordination pour l’Afrique de Demain), No. 26, February 1999. Online: http://perso.wanadoo.fr/cade/Lettres/lettre26/lettre26.html, November 30, 2001
Staatz, J.M. (1998). The Application of Social Capital in Understanding Economic Development”. MSU Agricultural Economics Staff Paper no. 98-17.
Stiglitz, J.E. (1990). “Peer Monitoring and Credit Markets”. The World Bank Economic Review 4 (3): 351-66
Williamson, O. E. 1979. “Transaction Cost Economics: The Governance of Contractual
Relations”, Journal of Law and Economics 22: 233-161
Williamson, O. E. (1985). “The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting”, New York, Free Press.
Endnotes
[1] A personal experience (1998) in the villages of Turiani and Gairo in the Morogoro region in Tanzania.
[2] It is useful to point out that Williamson works on firm-specific asset issues, which are slightly different from the types of asset specificities that characterize the one firm Malian cotton industry (from ginning onward). Privatization will, however, result in the multiplication of ginning and marketing firms, ultimately leading to a Williamsonian type of asset specificities.
[3] Other examples may be firms supplying their own inputs through vertical integration, or firms that are owned by the firm’s management, such Enron in US.
[4] The same cotton company provides credit to farmers and by output from them.
[5] The State may force the allocation
of resources to a particular sector in order to encourage a particular pattern
of industrial development, or to respond to the need of special, and often
powerful, interest groups. Bates (1981) discusses, in the Eastern African
context, forms of State coercion, and how this may result in rural demobilization.
Slantchev (2001) provides a short review of Bates’ work.
[6] Hirschman (1970) uses the concepts of “exit”, “voice” and “loyalty” to describe interactions between firms and their customers under different situations. These words are used much more loosely in this section. They have no direct link with Hirschman’s terminolgy.
[7] Old habits persisted even if firms were nominally separated. If you don’t pay farmers very much, there is plenty of profit to share among friends.
[8] Other local regulations may also have been used, but I have not looked into those in this paper.
[9] This may have been possible because of the emotional response that sustained the initial activities of the Union. In theory, such response gradually dies out with time; it needs to be replaced by other sources of motivations that bring out and sustain the effectiveness of the initial collective achievements.