Previous - Next
Part one : Smallholders and their characteristics
Introduction
Chapter 1 : What is a smallholder?
Chapter 2 : The role of smallholders in economic
development
Since the mid-1970s, as a consequence of the "debt crisis" and other developments, new economic policies have been proposed to many developing countries by a number of international organizations such as the International Monetary Fund (IMF), the World Bank, the Inter- American Bank, and other institutions, under the general heading of "structural adjustment programmes". These cover a wide spectrum of proposals, from tight monetary policy to tax system reforms and rate of exchange adjustments. The general idea behind these proposals is that the debt crisis is fundamentally a consequence of inadequate economic structures, that is, a bad arrangement of all the parts of the economic system. Therefore, the solution to the debt crisis depends on structural adjustment. They are not so new: most of them have been implemented long ago, in many developed countries. During the 1950s, few western European countries were exempt from the kind of recommendations we are dealing with here. These recommendations, however, are now being made to developing countries. This is an important point since, in this context, historical investigations can be helpful in predicting the consequences of present policies, in so far as historical and present situations are comparable.
Another characteristic of these policies is that they are painful. This is not a case of pure masochism on the part of their promoters. On the contrary; their intended long-term effects are highly beneficial, with steadier economic growth and higher incomes for every section of the population. But, in the short term, changing the rules of the economic game makes economic agents revise their expectations and strategies, thus introducing the risk of error and failure. Apart from expressing scepticism about their final success (a point which will not be examined here), this is probably the most serious objection that can be made about structural policies.
But there are other causes for concern. It is quite possible that, in spite of their long-term beneficial effects, structural policies may at least temporarily impose severe losses to certain people. Such a possibility must be taken seriously, especially if these people are the poorest. In this case, it would probably be necessary to provide some form of compensation, or to mitigate the contemplated policies to avoid their most detrimental effects. Together with the urban poor and the landless rural workers, smallholders may be permanent or temporary losers in the structural policy game. This is an important reason for taking an interest in them.
Even so, although they are poor, smallholders play an essential role in the economic life of most developing countries. The success or failure of structural adjustment policies will thus depend upon how much they are resented by smallholders, whose production is individually negligible but who, taken as a whole, represent a significant share of the GNP.
The subject of this paper is, therefore, to examine the possible consequences of structural policies on smallholders, both in the long and in the short term, in terms of equity as well as productive efficiency.
But, before going further, the notion of smallholder must be defined.
Chapter 1 : What is a smallholder?
The difficulties of a statistical definition
Common features of smallholders
The difficulties of a statistical definition
In agricultural matters, especially when poor and traditional farmers are concerned, economic ideas are usually severely biased by the European and North American situation. The farmer is a well-identified decision-maker, most of the time head of a family presenting a certain degree of "stability". The farm itself is well defined by a set of clearly identified plots, which either belong to the farmer (who, in this case, possesses the corresponding titles) or are hired from a landlord, who can produce proof of his or her rights. Land is supplemented by machinery, cattle, buildings and trees, each of which has a clearly identified owner. In this case, defining a smallholder is mainly a matter of threshold: more precisely, a farm size index can be defined and below a certain size a farmer is called a smallholder.
Certain problems are involved in these two operations. The size of a farm may be defined by the cultivated area (however, a vegetable grower with 5 ha is obviously larger, from an economic point of view, than a cereal grower with 10 ha), or by gross income (however, the size will depend upon the price system), or by a weighed sum of outputs (with the same problem as above, the size then depending upon the weights, which play exactly the same role as prices and are just as arbitrary). The threshold is even more difficult to define: changing the minimum size required for being considered a farm in the European Community from I ha to 0.5 ha doubles the number of farms. This is because the size distribution of the plots of land belonging to, or being cultivated by, the same person has a very long lower tail (Figure 1). A slight change in any threshold in this region of the distribution may considerably change the number of cases belonging to one category or another. The same is true whatever measure is chosen for farm size. This makes any definition of a smallholder rather arbitrary in these countries: one would have preferred a "double-peaked" distribution, such as the one shown in Figure 2. Such distributions are not uncommon in many developing countries, with the lower peak corresponding to peasant or traditional farmers, and the upper peak corresponding to colonial or modem or cooperative farming (other words are used, but these are adequate for our present purpose). In this case, defining smallholders is relatively easy, because fixing the threshold at a particular level in the valley between the two peaks is of limited consequence for the classification of most of the farms in the population.
FIGURE 1 : Typical distribution of farm area in developed countries.
Unfortunately, in many other developing countries, the problem is even more difficult, because the real problem is not so much the choice of a threshold as the definition of a farm. In these countries, land-use decisions are made at different levels, and by different individuals, so that the notion of "farmer" is not clearly identified. As an example, let us consider the Senegalese carré. The carré is a local community headed by a chief, usually the elder of a large family. All carré members live together in the same village, each household (husband, wife and young children) having its own house. The carré, as a whole, collectively cultivates cash crops, such as groundnuts or cotton. In this respect, the carré is a collective farm, the farmer being its head, just as the farmer of a cooperative farm is the president of the board of the cooperative.
FIGURE 2 : Double-peaked distribution of farm areas in many devoloping countries.
Apart from collective fields, the chief allocates plots of land to each household. Originally the plots allocated were intended for food crops. But it is now quite common to see ambitious young men growing cash crops on their individual plots, because cash crops give them more freedom, or are more profitable. At the same time, there is a division of labour between the sexes in the household. The wife is responsible for food, including growing crops such as paddy or cassava, necessary for feeding the family, whereas the husband is concerned exclusively with cash crops. Thus the household plot is again divided into two parts, one for food crops under the responsibility of the wife, and one for cash crops operated by the husband. In such a case, should husband and wife be described as two separate farmers? This is a question that may be raised.
An additional complication is that many meals from the carré kitchen are taken in common, men on one side and women on the other. The kitchen itself is supplied with foodcrops taken from collective fields. The working time allocation between individual and collective field is a matter of permanent negotiation between the various members of the community, with presumably an enormous amount of time devoted to these discussions.
Whatever the reasons for this situation, it does not facilitate the definition of the smallholder. The same kind of difficulty with other farms, for example, semi-collective farming, such as the eijidos in Mexico; the colonos in Peru; many forms of sharecropping arrangements; and even the kolkhoz in the USSR, which is a typical large farm, but which fosters links with the farm workers' individual plots in such a way that it is really difficult to define the latter either as an intrinsic part of the big kolkhoz, or as separate, small part-time holdings).
Another aspect is the existence of herder and nomads who do not own land in the usual sense. They usually claim right of passage or of pasture, although these claims are often denied by sedentary farmers whose land they cross. The size of their holdings cannot, therefore, be measured in terms of acreage. An alternative measurement is the number of head of cattle they possess, although this is often a somewhat elastic yardstick. Herders are disappearing in many countries, but still play an important role in Sahelian Africa, especially with respect to meat exports.
A difficulty arises from the fact that farming is not necessarily the only occupation of a smallholder. In the case of pan-time farmers, the judgement of the economic situation of a smallholder may be considerably modified by the consideration of other jobs he or she may have. In many Latin American countries, for example, it is fashionable for a doctor or a lawyer to operate a farm, usually not too far from the large town where they live. As a rule, this farm is relatively small (because the time spent on the farm is limited), and "modern" (again because this kind of "farmer", unlike real poor farmers, can obtain credit from banks and other sources). In this context, the naive analyst of a farm survey could reach the preposterous conclusion that small farms are among the richest and most modern!
The notion of a smallholder must not be considered in a purely static framework. If a particular man or woman is described as a smallholder at a certain date, he or she may not have been so for long, and the situation may not last until his or her death. In fact, it is a well-known feature of social statistics that the permanency of the number of cases falling into a given category may hide an important turnover, with many individuals departing, and being replaced by newcomers, between two surveys. This may be the case for smallholders. In the best situations, some of them will become mediumholders, or they may get a job in town. At worst, they may become landless workers or unemployed. Conversely, landless workers may become smallholders. Even industrial workers (although this is not very likely) may return to the land and also become smallholders. In any ease, the outcome of this process of state transition may be just as important as, if not more than, what happens to the state of the smallholder, taken in isolation.
A statistical definition of a smallholder is, therefore, very difficult. This accounts for the scarcity of statistical data on performance of smallholders and is also an obstacle to the knowledge of their specificities and potentialities, as well as the main cause of the difficulty in assessing the impact of structural adjustment programmes on them. Nevertheless, smallholders do exist and can be defined at least notionally, even if such a definition is not of much use to a statistical office.
They constitute mainly the rural poor, because the notion of smallholders refers to poverty, although poverty is essentially relative, i.e. a full-time farmer with 10 ha of grain crops in Africa is rather rich, whereas the same situation would classify him/her automatically as poor in any western European country. However, the smallholder category excludes the rural poor who are so poor they would never be able to operate a farm of their own. This is because of the specific role of smallholders, in comparison with "simple" landless workers and other rural poor, in shaping economic activity in their countries. If the impact of structural adjustment programmes on smallholders was only a matter of equity, there might be no reason for taking an interest in this segment of the population. But smallholders play a important role in economic development, a role that is probably much more important than is apparent from their relative number or from their total income. Neglecting their welfare could actually jeopardize the success of economic policies that would otherwise have every chance of succeeding. Here lies the basic justification for the present paper. We shall try to explain the reasons in the next chapter. For the moment, we shall define the common features which, despite their diversities, make smallholders a useful category for economic analysis.
Common features of smallholders
Despite the incredible variety of socio-economic status and organization of poor agricultural producers, they have many common features, especially with respect to their production techniques.
Specificity of production techniques
It is almost tautological to say that a poor entrepreneur has virtually no capital and thus the basic system of production is mainly a combination of land and labour. This is the original of "traditional agriculture" and the source of many justified comments about it.
First, it is not "backward" (with its pejorative connotation of irrationality). On the contrary, plain elementary economic theory teaches us that, with very small capital, the optimal techniques of the income-maximizing entrepreneur should actually be labour- and land-intensive. This has a number of consequences.
FIGURE 3 : Total output of labour functions at various levels of capital availability
Low productivity of labour
Labour productivity is low. Again, this is a consequence of the most elementary economic theory (see Figure 3).
As a consequence, since most of their incomes are derived from the rewards of labour, incomes are low, and this is both the cause and the consequence of poverty.
Land productivity depends on demography
Land productivity is also low, as shown by low yields, and for the same reason: far from being a proof of backwardness, the low yields of smallholders are a consequence of their lack of capital, and a mark of economic rationality. Two contrasted situations can nevertheless occur with respect to yields: in some cases, land is abundant at the regional or national level. This is the case in many parts of Africa or in the Amazonian region in South America. The productivity of land can then be extremely low, the limiting factor being the working time lost in cultivating remote fields. On the contrary, in overpopulated regions, such as the "groundnut basin" in Senegal, northeastern Brazil and many other regions, land is scarce. Yields are comparatively higher than in underpopulated areas. But the marginal productivity of labour is lower and people, in general, are poorer, especially if the land rent is appropriated by landlords not belonging to the class of smallholders.
Of course, there are intermediary situations between these two extremes, because land is never homogeneous. Scarcity usually begins with some types of land being defined by their pedological characteristics (such as the famous "grey soils" in Africa), or by distance to road or village, or by other geographical features. In this case, some soils are intensively cultivated, whereas others remain apparently idle (they are often used for extensive activities such as hunting, wild fruit harvest or even pasture). In any case, demography is the main determinant of these types of soil occupation, because land scarcity arises when there is population growth over a certain limit (otherwise, with free land available, newcomers simply occupy virgin land).
High productivity of capital
Capital productivity, by the same token, is extremely high: again, it is well known in economics that a very scarce factor can have a very high return. Simple budget calculations show the high profitability of simple tools such as spades, or bicycle wheels, or even donkeys or buffaloes. High capital productivity is also indirectly shown by the large interest rates charged by traditional moneylenders, which may reach several hundred percent. Without passing a moral judgement on usury, it must be recognized that such rates could not exist if the marginal productivity of capital were not large enough.
In such a case, economic theory assumes that firms only have to borrow to increase the quantity of capital they use, thus decreasing the marginal productivity of capital at a level commensurate with that in other activities. But, here, economic theory is wrong, because it neglects two points that are very important and, to a large extent, specific to smallholders.
First, the elements of capital at stake are very small: for example, a plough, or a fertilizer dose per hectare; the largest indivisible piece of equipment is a pair of oxen. Yet, oxen are often hired for few individuals can afford them. Nowadays, which bank would open a file for a credit of $50? Thus, the relative efficiency and the comparative advantage of the traditional moneylender begin to appear. He or she can make a loan of $50 without opening a file, simply because they know their client personally.
Second, lender availability is not the only obstacle to borrowing. Unless the peasant is dishonest (and this is rare), he or she knows the loan must be reimbursed. But nobody is ever sure of anything, especially in the case of a smallholder, who is subject to a variety of risks, both climatic and economic. Thus, peasant behaviour in borrowing matters is extremely conservative and highly "risk-averse". In fact, borrowing is a risk in itself! Only rich people can afford it. This is why so few poor people in developing countries borrow (and also, why is it so difficult for them to find a lender).
The blocking scheme just described must not be considered as the only obstacle to smallholders' economic development. If it were possible to remove it, other obstacles such as land scarcity or market saturation would appear. But it is an important factor in the explanation of the permanency of the smallholder's situation, whatever form it takes. It accounts for the combination of the low productivity of land and labour and the high capital productivity that are the fundamental characteristics of smallholders. The consequences of this situation will be examined from a macro-economic point of view.
Role of risk in economic decisions
One of the lessons learned from the above is the importance of risk as a deterrent to investment. But this is not the only effect of risk on smallholders. Risk plays a particular role in the decisions of smallholders simply because they are poor and, thus, considerably risk-averse. Whatever the technical solution chosen to express this risk aversion in models, it can be said that risk imposes diversified patterns of production, and costly practices to avoid it as far as possible. Also, the existence of risk and of risk aversion implies that average prices and yields are not the only elements entering into utility computations of rational producers. Their variability is just as important, as is illustrated by the following simple example. Let us assume that a farmer has the choice between two crops, with the same prices, the same yields and the same inputs. Of course, the decision-maker will be indifferent between the two. At the same time, the slighest change in prices or yields will induce him or her to shift the totality of their resources from one production to another. If all producers act in the same manner, the markets of these crops will be characterized by a devastating instability.
Let us assume now that the yields are random, and independently distributed with, again, the same mean and variance. Nothing is changed with a risk-neutral decision-maker. But only a slight degree of risk aversion will suffice to induce any risk-averse farmer to choose unhesitatingly a 5050 output mix, which minimizes income variance. Even better, in such a situation, if the price of crop A decreases slightly, farmers will increase their production of B. but never cease to devote a part of their production capacity to A. Thus, risk not only clearly defines an optimum where it apparently does not exist, but it may stabilize markets and stand as the major determinant of price supply elasticity.
We shall now assume that the government decides to stabilize crop A. Since farmers are compensated for any loss due to random changes in the price or in the yield of A, with the unitary income at the previous average level being guaranteed, crop A is riskless. Then, every risk-averse farmer shifts to A, so that the production of B is nil. If the government wants to have the least quantity of B harvested, the return for A should be fixed at a level significantly below the average unitary income of B. particularly as farmers are more risk-averse.
This simple theoretical example has more applications than one might imagine at first. Consider the case of export crops and food crops (that is, crops consumed within the country) in most African countries. As a rule, export crops have a price guaranteed by stabilization boards, or other agencies, whereas food crops are sold (when they are sold) on unorganized markets, with tremendous seasonal and interannual fluctuations. Thus, mutatis mutandis, export crops resemble crop A, and food crops crop B in the preceding example. At present, prices give a definite advantage to export crops, and this has been the object of complaints from international organizations, and of recommendations to increase the prices of food crops, to make them competitive with export crops. The above analysis suggests that it would suffice to stabilize food crops at their previous average level to induce a very significant shift of resource allocation from cash to food crops. This illustrates the importance of risk considerations in agricultural policy. Other examples are given later.
Supply responsiveness
The question of smallholder sensitivity to price variations has been the subject of controversy. Some economists deny any price responsiveness on the part of smallholders, either because they are "irrational" (a very strong assumption) or because their rationality is different from that of the capitalist entrepreneur. In this respect, 20 years ago, Sicco Mansholt, a member of the European Community's Commission in charge of agricultural questions, prepared a plan for the restructuring of European agriculture. The backbone of the "Mansholt plan" was to increase farm size to make farms more responsive to prices. European farms nowadays are, on average, considerably larger than was intended by Mansholt, and apparently not much more responsive to prices than before. Of course, smallholders in developing countries cannot be compared with European farmers; nevertheless, this story shows the need for intellectual modesty.
The smallholders response to price is never zero, but is determined by complicated patterns that jeopardize any attempt to measure it statistically. This is because a price variation in one crop (say, a price increase) has two separate effects:
These considerations interfere with the risk effect which has been described above, in two different ways:
Admittedly, such results are complex and represent a challenge for econometric analysis (see Mundlak, 1988). It is not surprising, in such a context, that the latter gives back contradictory results. What is surprising is that, despite the difficulty of taking account of the preceding considerations, the results nevertheless generally suggest a relatively large and positive direct supply elasticity for many developing countries and for many commodities. A possible reason for this result might be found in the fact that fixed factors for smallholders in developing countries are not numerous, thus giving decisive weight to direct profitability in the decision-making process. This is contrary to what happens with farmers in developed countries. This point deserves funkier research.
Having thus shown both the diversity and the common characteristics of smallholders, before examining the likely impact of structural adjustments we have yet to demonstrate their economic role. This is the subject of next chapter.
Chapter 2 : The role of smallholders in economic development
They are a huge reservoir of labour
They constitute almost the sole source of
domestic food production
They are often export crop producers and a
source of foreign exchange
They constitute a potential market for non-food
goods
Their disappearance should be among the targets
of economic development
They are a huge reservoir of labour
Because of low labour productivity, smallholders represent a huge reservoir of labour which, like any reservoir, is probably capable of dampening the effects of sudden changes in labour demand.
The problem of labour transfers from the "traditional" to the "modern" sectors is recurrent in literature pertaining to development from the early works of Nurkse or Sir Arthur Lewis (1954) to the modern "computable general equilibrium models''.
In the early stages of economic analysis applied to development, it was thought that labour from the "traditional" sector (that is, essentially, labour coming from agriculture) was available in unlimited quantities, for virtually any wage. Of course, this was an over simplified view of reality and was later reassessed. It is now widely acknowledged that smallholders and landless workers do act according to economic considerations when leaving farming to seek an urban job, or a non-agricultural rural job. They compare their marginal productivity as smallholders or hired workers in traditional farming with expected wages in other sectors. The computation is greatly complicated by the considerable uncertainty on both sides of the balance sheet. As we have seen, the productivity of smallholders and rural workers is extremely variable, depending upon many circumstances. And if urban wages are more predictable, in general, than agricultural incomes, the probability of finding a job is always small, and the risk of staying idle in an unknown town in very precarious conditions is never negligible.
In such situations, mean values of agricultural incomes and urban wages do not play a large role in the decision. Their variability, as measured, for instance, by variances or other dispersion indicators, is actually much more important. The migration flow between rural and urban zones may exist in both directions, and is regulated by forces resembling market forces (although perhaps more complicated) in the sense that kinship and other sociological factors play a large role in determining the subjective degree of uncertainty attached to migration. These non-measurable factors are, thus, the main determinants of this special "market". This is also true for changes from rural agricultural to rural non-agricultural occupations, such as local transport entrepreneurs (for example, the "taxi brousse", whose existence is so important in most regions of West Africa) or craftsmen.
But if migration and occupational changes can take place in both directions, this means that the "labour reservoir" constituted by the smallholder and rural poor may play an important regulatory role in the economy. Changes in the "wage differential" between urban and rural labour may change the pattern of migration, with sometimes dramatic consequences.
Indeed, over the last 20 years urban growth has become a real economic problem for most developing countries. The accumulation of huge concentrations of men in large towns is politically dangerous and economically costly. It is one of the causes of the present "debt crisis", because at least part of the excessive borrowing at the origin of the crisis was explicitly aimed at lodging, feeding, sanitizing and educating these urban concentrations. The final outcome, in these respects, is far from satisfactory. Efforts to put public goods at the disposal of the urban poor, however, increase the implicit urban wage and the probability of migrants finding better conditions in town than in their village, thus creating a vicious circle: each effort to relieve the predicament of the urban poor increases their number by attracting new migrants from rural areas.
Although it is probably impossible to reverse the movement (that is, to create a massive migration flow from suburban bidonvilles or favellas to country villages), it is probably possible to reduce significantly the flow of migration from rural to urban zones. Here, the situation of smallholders plays an important role, because migrants are either smallholders themselves (this is the case of people fleeing after failing to reimburse debts, or people in despair at their low income and hoping for better job opportunities in town), or, more frequently, people likely to be employed by smallholders, such as relatives (sons, cousins) or landless workers who could get temporary or permanent jobs in smallholdings, if there were a chance of their being paid.
Thus, smallholders represent one of the key determinants of the equilibrium between rural and urban labour forces. Improving their situation, even slightly, could significantly contribute to solving urban problems of congestion and unemployment, although, of course, it would be necessary in each specific case to compare the effects of one dollar spent helping smallholders with spending it directly on improving the situation of the urban poor.
In this respect, an important and somewhat ironical macro-economic effect must be pointed out. Improving the situation of smallholders requires the use of more capital-intensive techniques (animal draught ploughing instead of hand-tilling, for instance) which will decrease the optimal ratio of workers per hectare. Especially in regions where land is limited, this will require the emigration to town or to other rural occupations of at least a fraction of former land workers. This fact must be considered when planning rural development.
They constitute almost the sole source of domestic food production
Except in very specific situations (such as Argentina), large farms or agricultural plants seldom contribute to supplying domestic food markets. This is because large farms are more or less compelled to put into operation techniques similar to those in use in developed countries (for example, the extensive use of machinery, especially for harvesting as well as for ploughing). Otherwise, the necessity of monitoring the large number of workers necessary to compensate for machines would require a numerous and costly staff, together with coercive measures which were possible on the "plantations" when slavery existed but which, fortunately, are now unthinkable.
As a consequence, large firms in developing countries do not have a large comparative advantage over the producer of the same (or similar) product in developed countries. The fact that the wages of the (necessary) workers are lower is counterbalanced by natural conditions which are often worse in tropical than in temperate countries (this is the reason for the specific nature of Argentina - a country located mainly in a temperate climate, with definite natural comparative advantages). Therefore, the comparative advantage of large farms in tropical countries cannot be materialized other than by crops which cannot be profitably grown in the conditions of temperate countries, such as coffee, cocoa and rubber.
However, smallholders are obliged to make use of radically different labour- and land-intensive techniques. They accept a very low reward for their labour and, in so far as they are their own employers, they monitor themselves very efficiently. In such a situation, they may be perfectly competitive with large-scale and foreign producers. This was at least true until recently, when agricultural surpluses in developed countries led Western governments to dump large quantities of staple food a/international prices which bear no relation to production costs. Even so, despite the incredibly low price of grain on international markets, African maize and sorghum growers or Latin American pea growers are still competitive and continue to produce large quantities of European and American grain substitutes. Admittedly, this competitiveness is now being challenged. It is increasingly difficult to sell traditional food in African city markets. In the countryside, however, as well as for their own food needs (which, because of their large number, represent a significant share of the nation's food intake), the competitiveness of traditional producers remains largely unaffected. The cost of transporting imported food, and of dividing large quantities into small batches, rapidly offsets its artificial price advantage. At the same time, pseudo-traditional techniques are now emerging: these techniques (sometimes called "soft technologies") do not belong to the set of traditional techniques sensu stricto, but they are shaped to be put into operation by smallholders with a little more capital and a little less labour than the "true" traditional 8 techniques. In this way, smallholders may be placed in the position of adapting themselves to a new environment, hopefully where there is a little less poverty. This is the economic interpretation of "draught animal cultivation" and other similar systems. Whatever the validity of this interpretation, such improvements are likely to reinforce the competitivity of smallholders against large-scale farming, and even against open dumping.
They are often export crop producers and a source of foreign exchange
Apart from producing staple foods, smallholders also often produce export crops. It is a source of permanent astonishment to many general economists that small-scale producers can remain competitive against large firms. Although there are few explanations for this, scale economies are virtually absent from agriculture. Whenever this is the case, large colonial firms have no comparative advantages in producing export crops. They may actually have severe comparative disadvantages, given the workers' monitoring difficulties referred to above. This explains the failure not only of many colonial or neo-colonial plantations but also of a number of state or pseudo-cooperative farms, which can survive only by massively substituting capital for labour, against all economic rationality, in countries where labour is abundant and capital is scarce.
Conversely, the same considerations explain the success of contract fanning, where a large firm makes a contract with smallholders for the production of a certain crop, grown according to certain rules. Here it is the smallholder who monitors, whereas the technical specifications which are imposed by the large firm (and enforced by the direct execution of sensible and standardized tasks, such as tillage, as well as by the direct supply of adequate seeds) guarantee the achievement of certain standards of quality. The large firm remains in charge of operations for which economies of scale do exist, such as transport and marketing.
Everything that has been said already about what determines the supply to markets of food crop products is also valid for "cash crops" (except, perhaps, that as a rule cash crops do have guaranteed prices and, therefore, are less risky to be marketed than food crops)."
Much has been said about the prices of cash crops which, owing to their high levels, are blamed for discouraging food crop production. Ironically, other authors criticize taxes, which disconnect the internal prices of export crops from their international level, and advocate farm gate price increases. These two positions are not only contradictory, but also irrelevant. We have already seen that not only the average level of prices but also their variability are among the determinants of farmers' decisions on their crop mix. Therefore, it is possible to change the pattern of crop production without changing the average price levels. These issues will be discussed later.
They constitute a potential market for non-food goods
Smallholders, in general, do not at present represent a very lucrative market. They buy a few plastic articles (such as basins), some carbonated beverages, a few radio sets and a few pieces of cloth. For most of the other consumption goods they are self-sufficient. But this is, hopefully, temporary and a result of their poverty. Their income demand elasticity for simple articles is considerable.
Furthermore, among the reasons why smallholders often seem to be reluctant to increase their food production in excess of their foreseeable needs is their conviction that they would not be able to exchange possible food surpluses for the non-food goods they require (Berthelemy and Morrisson, 1987). At present, many reports from Madagascar, the United Republic of Tanzania and other similar countries tend to show that increasing the availability of non-food commodities in rural areas is likely to increase agricultural production. Otherwise, peasants restrict their production to their immediate needs.
Such production deadlocks frequently arise when socialist governments stop the activities of traditional traders (often the same people as the traditional moneylenders) on the grounds of their exaggerated profits, and without replacing them by institutions that provide the same services. Just as for traditional moneylenders, providing services for smallholders is extremely costly because, as for credit, very small quantities must be spread overlarge areas. As soon as smallholders become richer, they buy in larger quantities. They may even acquire their own means of transport. Thus, quick cumulative effects can be expected of even limited rural development. But the opposite is true also. A slight lowering of smallholders' income, due to various causes, can very well result in the disappearance of traditional shopkeepers, and to the collapse of rural economic life.
Their disappearance should be among the targets of economic development
Of course, such a statement may seem very odd, after what has been said about their potential contribution to development. Nevertheless, it must be taken seriously, if one recalls that most of the virtues of smallholders (such as high productivity of capital) come from the fact that they are poor. Since removing poverty is clearly one of the ultimate targets of economic development, it follows that smallholders should disappear during the process. This is actually what has happened in developed countries since the end of the eighteenth century (a movement which gained momentum in the 1880s, and even more in the 1950s. See Bairoch, 1964).
At the same time, this statement has many far-reaching implications for economic policy. It means that, in most countries, the total area under cultivation, instead of being divided into plots of between 0.5 and 5 ha, will have to be shared between farms of 100 to 150 ha. This also means that a large number of people will be expelled from farming and will have to find other occupations, either as wage workers, or in the trade sector or in administration. In Western countries, these changes have often been very painful. Would it be possible for currently developing countries to obtain similar results at a smaller cost? This is a point to be taken into account when devising agricultural policy.
Before this final outcome, agricultural policies (or their absence) are likely progressively to reveal formerly unnoticed constraints resulting from the fact that a previously free resource suddenly becomes scarce as a consequence of depletion. An example is firewood in Africa, which was in ample supply 20 years ago, and which now poses a serious problem (Webley, 1986). More broadly, Boserup (1965) contends that increasing the number of people leads naturally to an increase in farming intensity. This long-term dynamic must be kept in mind.
We have not mentioned the role of smallholders as "money traps", which is a controversial subject. In any case, the points just presented are sufficiently important for some care to be taken to avoid "hurting" smallholders by structural adjustment programmes, aside from considerations of equity.
