Previous Page Table of Contents Next Page



'Before I built a wall I'd ask to know what I was walling in or walling out ...'

Robert Frost. 1874-1963

In this and in Chapters 4 and 5 the several structural elements of a farm-household system noted in Section 1.4 are more closely examined.


3.1.1 Importance of boundary specification

In analysing any farm-household system or any other agricultural system, an obvious first step is to define the scope of such a system, i.e., its boundary as relevant to the purpose of analysis (FAO 1990). Sometimes this will present no problem, particularly if, as here, the focus of analysis is on farm production and its management. In this case, unless they relate significantly to production management, interfaces of a purely social, religious or political nature between the farm-household and its environment can be ignored (Dillon 1992). However, as the following examples illustrate, even if the focus of analysis is production and its management, the specification of the farm-system boundary (and likewise of the farm-household system boundary) may be quite complicated. Boundary definition may also vary with the purpose of analysis. Thus the relevant boundary for annual enterprise planning may simply correspond to the boundary of the physical farm entity. In contrast, for long-term development planning, the boundary may need to include off-farm income-generating activities and the interface with suppliers of long-term credit.

Modern Western and Asian farms in a highly commercial environment (e.g., West Texas cotton farms and Malaysian rubber smallholdings and estates) have one or a limited number of farm production enterprises and clear sets of trading relationships between the farm and its input suppliers on the one hand and its output markets on the other. Relationships between a given farm and other farms will often be minimal, even non-existent. From a production management perspective, the boundary of this type of farm system will encompass the farm in its physical extent and its input suppliers and market outlets, but will typically exclude other farms.

In a second situation, as exemplified by many isolated near-subsistence farm-household systems of the Himalayan valleys, the boundary of the farm system also corresponds closely with the farm's physical boundary but in this case the system excludes the outside world.

More common is a third situation. It comprises the great bulk of small traditional farms of Types 1, 2 and 3 (i.e., subsistence, semi-subsistence and specialized independent, respectively) which are moderately or heavily dependent on each other for supply of inputs - exchange or hired labour, oxpower, village transport etc. - and often also for the disposal of produce, e.g., by barter or trading among friends and neighbouring families. In this situation the real boundary of a farm-household system, from an operational management viewpoint, might encompass all the farms of a hamlet or even of several mutually dependent villages.

This third situation in some of its variations can be quite complex. For example, the farms of the villages around Karanganyar in the Solo Valley of Java commonly consist of some area of household garden devoted to fruit trees and bamboo plus a larger area of fields (Prabowo and McConnell 1993). They are nominally 'owned' (in the Javanese sense, or held in trust in perpetuity) by the operating families who in fact do exercise full control over the garden or 'pekarangan' parts of their farms. In three years out of four the field area of a farm will usually be under paddy, grown either by the farmer himself or herself or more typically as a quasi-cooperative/mutual-help undertaking ('gotong royong') by the farmer and his or her neighbours. Thus in this first phase the boundary around each farm system encompasses its household garden and some area of paddy shared with other farm households. But in the fourth year the farm paddy fields, together with one fourth of all other village fields (which lie within the command area of a publicly owned sugar mill) are pooled, put under sugarcane, and the farmer becomes both a labourer (on his or her own land) for the mill and a landlord (receiving rent from the mill). In short, the boundary of this farm system is fluid over time: it always encompasses the pekarangan (household garden) lands; it sometimes encompasses the rice field and other farm households which participate in the rice phase; and it sometimes exists only as one part of the boundary of a much larger sugar-estate system. The situation is depicted diagrammatically in the righthand side of Figure 3.1. The lefthand side of Figure 3.1 refers to an isolated subsistence farm system. The central part of Figure 3.1 depicts a commercial system in which the alternatives are to define the system as consisting only of the farm (implying boundary B') or as also including input suppliers and output markets (implying boundary B). The contrast in complexity between these three examples of farm-system boundaries is obvious.

3.1.1 Importance of boundary specification

At farm-household level, one obvious essential step in planning farm development programs and investment projects is to accurately identify the relevant boundary of the subject system; yet it is a step sometimes not taken. In the 1970s a project in North East Africa was aimed at village economic development through introduction of modem rice-growing technology to Dinka tribespeople in the hope of supplementing and if possible replacing their traditional millet (as well as supplying rice to the urban population). Project planning overlooked the fact that, while millet was indeed important, the real boundary of the traditional farming system encompassed also keeping cattle and catching mudfish. Thus, while project planning management was directed at zealously maintaining field polder embankments to grow rice, Dinka management was directed even more zealously at breaking them down to get at the fish. (Planning that was less technically oriented would have aimed for the integration of all three components in a locally acceptable system.) This and similar mistakes in other farm projects also provide a caution that structuring or restructuring of a system must always start with an understanding of the farm-household system as discussed in Section 3.2.

At the levels of subordinate subsystems, the boundaries of processes, activities and enterprises, although set approximately by technical factors, are determined also by the purpose of the analysis or by operational convenience (as discussed in Chapter 4).

FIGURE 3.1 - Boundaries of Three Contrasting Farm Systems


3.2.1 Farm household as resource manager
3.2.2 Farm household as system beneficiary

'I grant indeed that fields and flocks have charms
for him that grazes or for him that farms;
But when amid such pleasing scenes I trace
the poor laborious natives of the place ...'

George Crabbe, 1754-1832

The household component of a farm-household system is a somewhat flexible concept. It can consist only of the farm's nucleus family but more often includes extended-family members. It also commonly includes some number of more or less permanent domestic and farm workers and miscellaneous dependants. Most farm development projects are structured on the assumption that households are headed by males, but this is often not so. In a typical Javanese village anywhere from 10 to 25 percent of households might consist only of women and children. Where poverty exists, here and elsewhere, it will generally be concentrated on such households.

3.2.1 Farm household as resource manager

The two roles of the household as resource manager and as system beneficiary were noted in Section 1.3.1. The household's role as resource manager is to provide purpose, direction, objectives and management to the whole-farm system and its subsystems (Shaner, Philipp and Schmehl 1982). On farms of Type 5 (large commercial family farms) and especially Type 6 (estates) the place of the household might be taken by professional management. However, on small farms and within limits set by the physical environment and available resources, the planning objectives are set by the household's broad social-value structure and local tradition (though the latter is increasingly being corroded by commercialisation and the influence of modem communications technology). The farm economist is usually not well equipped to question these goals or the farm-planning objectives which result from them. His or her role is to attempt to optimize the farm component of the system in terms of whatever mix of socio-cultural, religious, traditional and material goals is relevant to the household. He or she might also evaluate and convey the material consequences of alternative choices and management strategies within the farm component.

Decision making within the household

It will often be necessary to enquire more deeply into who does what and who decides what within the household if sound farm planning is to be possible (in Field A) or sound farm development programs are to formulated (Fields C and D). Real power and responsibility for the farm component might not always lie with the apparent or nominal household head. Thus in the Wadi Hadramant of Yemen, as throughout most of Africa, although women might play an otherwise subservient role - or seem to - they are in fact the effective farmers, doing everything except the heaviest chores from planning the production system and executing most of it to preparing and allocating its output. Yet farm 'development' projects here and elsewhere - with few more than token exceptions and invariably conceived, planned, evaluated and managed by men - do not acknowledge this fact and remain directed at the mirage of some unspecified but assumed patriarchal 'farmer'. Without at least an insight into local culture, the conceptual models of how a system works which an analyst may have can be unproductive if not dangerous things.

Gender analysis

As noted above, sound farm planning often necessitates an understanding of how resources, responsibilities, tasks and benefits are distributed between the men, women and children of the farm household. This is the subject of gender analysis as discussed and illustrated by, e.g., Feldstein and Jiggins (1994), Feldstein, Flora and Poats (1990), Gondowarsito, van Stanten and Bottema (1995), Quisumbing et al. (1995) and van Herpen and Ashby (1991). The need for gender analysis, particularly in relation to the role of women in the farm-household system, is well illustrated by the following response by a small farmer when interviewed in a farm survey (Gabriel 1995, p. 68):

Q. 'Does your wife work?'

A. 'No, she stays at home.'

Q. 'I see. How does she spend her day?'

A. 'Well, she gets up at four in the morning, makes the fire and cooks breakfast. Then she goes to the river and washes clothes. After that she goes to town to get corn ground and buy what we need. Then she cooks the midday meal.'

Q. 'You come home at midday?'

A. 'No, no. She brings the meal to me in the fields - about three km from home.'

Q. 'And after that?'

A. 'Well, she takes care of the hens and pigs, and of course she looks after the children all day. Then she prepares supper so it is ready when I come home.'

Q. 'Does she go to bed after supper?'

A. 'No, I do. She has things to do around the house until about nine o'clock.'

Q. 'But you say your wife does not work?'

A. 'Of course she doesn't work. I told you, she stays at home.'

Tables 3.1 and 3.2 provide an illustration of some aspects of gender analysis (labour and task allocation - but not the distribution of benefits) relative to a sample of small-farm households in Northern Mindanao, Philippines.

TABLE 3.1 - Gender Analysis of Labour Use in Cassava Production in Northern Mindanao, Philippinesa

a Based on a sample survey of 75 small farms (Rola 1995)

TABLE 3.2 - Gender Analysis of Participation of Farm-household Members in Household and Other Activities in Northern Mindanao, Philippinesa

a Based on a sample survey of 75 small farms (Rola 1995)

3.2.2 Farm household as system beneficiary

On small farms the primary beneficiaries are usually the members of the household itself. However, external beneficiaries are also often important. In Bhutan it is common for farm family members long resident far from the farm and not dependent on its output to still retain important rights. They exercise these by returning at harvest time and taking their benefits in the form of pork and grain. In the villages of Central Java up to 30 percent of households might be landless and subsist as farm labourers or - especially important in the case of households consisting of only poor women and children - by harvesting their neighbours' paddy in exchange for retaining one seventh or one tenth of the crop (the 'bawon' system). This latter can be the only significant source of income for otherwise practically destitute people. They are perhaps proportionately more dependent on the farms of their more fortunate neighbours than are the farmers themselves, although they are from a formal viewpoint external to these systems. This is one common reason why some forms of Western-style farm 'development' do not occur. Javanese villagers have a strong sense of mutual economic and social responsibility. 'Advanced' technology which would prepare land more quickly (tractors) or give higher grain recovery rates (mechanical harvesters, grain dryers) or other technology whose first effect would be to increase the comparative wealth of the few would be frowned on because of its adverse effect upon the many. Thus external beneficiaries can impose constraints on the household, in its first role as system manager and internal resource allocator, by exerting a collective moral influence on how crops are to be produced and even which crops are to be grown.

Other obvious groups of external beneficiaries consist of landlords, village traders, local governments as taxing authorities etc. and, at further remove, national governments deriving their foreign exchange from the export of farm produce. These can also influence how a farm-household functions in its first role as system manager, e.g., through tenancy agreements, provision of credit, price policy etc.


3.3.1 Farm resources from an accounting view
3.3.2 Farm resources from an operational view
3.3.3 Operational resource categories
3.3.4 Other relevant resource properties
3.3.5 Resource acquisition and generation
3.3.6 Relationships between resources, capital and costs

The supply of resources in a whole-farm system can be examined from either an accounting or, more fruitfully, an operational perspective.

3.3.1 Farm resources from an accounting view

From an accounting viewpoint with its ex post or backward-looking emphasis, farm resources fall into two broad categories (Makeham and Malcolm 1986):

· Fixed resources provide services over a number of years or at least over a period longer than the production cycle of short-term (seasonal, annual) crop or livestock enterprises. Common examples are land, machinery, an irrigation system. These services may be used either by individual enterprises or to maintain the farm as a whole. In the very long run, of course, few resources are truly fixed in supply. Even land and climate in their productive dimension can be created (as in the controlled-environment greenhouses of Saudi Arabia, UAE and Qatar).

· Short-term or variable/operational resources are those that are usually entirely used up in the annual production cycle, e.g., a supply of seed or fertilizer.

The essential difference between fixed and short-term resources is that the former provide a stream or flow of services over time while the latter consist primarily of quickly exhaustible material things or time-bound institutional sanctions (as noted in Section 3.3.3).

From an accounting viewpoint, both fixed and short-term resources have two other relevant dimensions. In their economic dimension they become respectively fixed/long-term capital and operating/short-term capital; and in their financial dimension these generate, respectively, fixed costs and operating/variable/direct costs, viz.:

The distinction between these two cost categories of fixed and variable is discussed further in Sections 4.3.8 and 9.

Discussion so far has concerned resources when used at whole-farm level, i.e., at Order Level 10. At lower Order Levels of farm systems (processes, activities and enterprises), the above classification of resources relative to capital and costs is parallel to that for the whole-farm: e.g., the resources assigned to a paddy crop can be broken down and become fixed and operating paddy-crop capital, generating fixed and operating/variable/direct paddy costs.

Variable/direct costs are discussed in Section 4.3 in the context of enterprises and activities, and fixed costs in Section 5.3 relative to the whole-farm service matrix.

3.3.2 Farm resources from an operational view

From an operational viewpoint the picture of farm resources is somewhat different. Here emphasis is on the ex ante potential or planned use of resources rather than the results of their past use. In this forward-looking context it is more fruitful to regard farm resources from a systems viewpoint. Distinction between fixed and short-term resources is now less relevant. As shown in Figure 3.2, more relevant are the flows of both short-term or operational resources and of services from fixed resources to the farm's subsystems.

FIGURE 3.2 - Direction of Resource Flows within a Whole-farm System

Central to these resource flows is the farm's resource pool (a system of Order Level 8, discussed from an operational perspective in Section 3.3.3 below). Short-term or operational resources flow from the pool primarily to lower Order Level production subsystems (processes, activities and enterprises) to generate the farm's intermediate and final outputs. Some also flow directly to the whole-farm service matrix to maintain the structure (repair of fences, buildings etc.) or permit the functioning of this subsystem (e.g., payment of land, water and road taxes).

The services of long-term or fixed resources flow both to maintain the service matrix and to the resource pool from where they are assigned to the various production subsystems.

Any initial stock of both fixed and short-term resources must sooner or later be replenished from output of the productive enterprises/activities, either as materials purchased with income from the activities or as resources/intermediate products generated by these activities, e.g., oxen produced as a by-product of the dairy activity. These resource return flows are indicated in Figure 3.2 as flowing both to the resource pool and to maintain the stock of fixed resources or capital in the whole-farm service matrix.

3.3.3 Operational resource categories

From a planning and operational viewpoint, farm resources fall into five categories. Discussion from an operational viewpoint is focused on how specific resources might constrain or limit farm production. The five resource categories are:

(1) Material long-term: This category consists of material things which yield their services over relatively long time periods. They were referred to above as fixed resources or fixed capital. Land is typically the most important such item and will usually provide its services indefinitely (but see system sustainability, Section 6.2.7.) On the other hand, land is relatively less important on many capital- and labour-intensive specialist Type 3 farms such as those producing orchids, poultry and pigs. Other examples of resources in this category are irrigation systems and farm sheds generating their services over 20 to 30 years or an ox pair providing draught power over five or six years.

(2) Material short-term: This category, exemplified by such items as seed and other seasonal inputs, was also discussed above. In a commercial environment where these items are purchased, the production constraint is generally set by the amount of money available to buy them, not by the supply of these items in themselves.

(3) Financial: This category consists of cash, debts receivable, and access to credit from formal (banks, cooperatives) and informal (shops, traders, relatives) sources.

(4) Institutional: This category consists essentially of rights of access to materials, markets and services. In its financial dimension, this category takes the form of land and road taxes, water-use license fees, payments for production-quota rights (as sometimes prevail for sugarcane, milk, tobacco etc.). They are termed 'institutional' because they consist of relationships between the farm family on the one hand and institutions/agencies/persons on the other. Note that where they are transferable and have financial value, these rights are assets as well as production resources.

(5) Labour: This consists of family labour available for general farm work or which might be available only for specific tasks. For example, specific-purpose labour might consist of the very old and young family members who can do only light work such as tending livestock; or a family member who prefers and is especially skilled in tapping toddy palms etc. (Management ability is an important attribute of family labour, sufficiently so as to sometimes warrant attempts at separate evaluation of its productivity, but it is usually not possible to measure management as an ex ante input, only in terms of what it actually achieves as discussed in Sections 7.2.3 and 5).

Resource inventory for planning

For planning (ex ante) purposes, specification of the farm resource pool simply amounts to making a list of the availability of those resource items which might limit production in the planning period, e.g., for the coming year the resource pool may include the following resources:


Material long-term:


0.25 ha irrigated lowland

0.60 ha eroded upland


1 pair


3 units


1 unit


1 house, 1 shed (100 m2)


Material short-term:


6 bags

cow feed

3 tonnes




1 500 Rs

expected from crops

3 500 Rs



milk quota

20 litres/day, sold in town

water license

0.25 ha



work-age adults

490 labour days


180 days (cattle only).

Listing of the initial resources will depend partly on the resources actually available but also on the uses to which they are likely to be put, i.e., the types of enterprises/activities which are likely to be operated. The procedure, therefore, is somewhat circular and subjective but in practice it poses no serious problems. The aim is to identify all the main farm resources, particularly those which are likely to be production-limiting and relatively costly, rather than to compile an exhaustive list of all resources which are present or which could conceivably act as constraints.

In almost all planning situations, the analyst will be able to form a general idea of the likely production possibilities and therefore the types or categories of resources needed to exploit such possibilities. If, as is often the case on many small farms, there exists two or three times the amount of labour likely to be needed to execute the production plan, one would not devote much attention to this resource. But, on the other hand, if the production system is likely, e.g., to include strawberries, and this requires the deft fingers of children, then one would measure this particular category of available labour with some precision.

3.3.4 Other relevant resource properties

Resource quality

For planning purposes it will often be necessary to quantify some resources according to the specific uses to which they can be put. These uses may have a quality as well as a quantity dimension. The example of child labour was noted above. If a farmer has 1.5 hectares of land and an asset statement is being prepared (below), it would be enough to describe this as simply 'Land ... 1.5 ha', but for planning purposes it would be necessary to list this in the resources pool as, e.g.:


(necessary for paddy)




(suited to maize)




(inter-cropping possible)



Similarly, if total crop-storage capacity is six tonnes, but only three tonnes of this is secure and rodent-proof, this fact will be specified in the resource pool statement. Likewise, while 'Pasture' might be an adequate description of a land parcel if all cattle/donkeys/buffalo/sheep are to run together as a combined herd, it would not be adequate if the producing dairy cows are to be given preferential treatment. The pasture might then be quantified as:

good fresh pasture

(for cows only)



rough grazing

(all other stock)



Resource-use time dimension

From the viewpoint of resource use in relation to time, resources fall into two groups: those which provide a flow of services over time, and those which consist of a consumable store or stock of materials or other farm resources. Land, family labour, oxpower, tractors and machinery, crop storage space, fences and livestock housing are examples of service-generating resources which provide their services as a flow over time. A store of seed or agricultural chemicals in a shed, a pond of irrigation water, a contract to produce sugarcane, or a shed of animal feed are examples of store-type resources, since once they have been consumed in the production process they cease to exist.

Flow resources have a time as well as a quantity dimension. Because agricultural production does not occur instantaneously but over some time period of months or years, the resources allocated to such production must be provided at discrete points in time or during specific time periods. Thus a one-hectare paddy crop to be grown from March to June might require the following resources: one hectare of irrigated land, 90 days of family labour, 21 days of ox work etc. This would be sufficient information if the purpose is to compare the economics (input costs versus output value) of paddy against other crops; but for planning purposes a more satisfactory statement concerning paddy-enterprise resource needs would be as follows:





land (ha)





labour (days)





oxen (days)





where 12 types of inputs are identified rather than just three, each having a quantity-time dimension. In short, in planning where the resource inputs of an enterprise have to be time-scheduled, e.g., by months, then the farm resource pool also has to be specified in these same quantity-time units. Labour-days or ox-days in March are, respectively, different resources from labour-days or ox-days in June.

However, the more closely that the resource needs of a production activity are specified (as occurring in June, early June, first week of June etc.), the greater the degree of detail needed in quantifying the resource pool. Obviously this could result in the specification of an unmanageably large number of resource items. Some limits must be imposed on the procedure of adding a time-dimension to flow-type resources; but, on the other hand, a resource pool constructed without noting when resources are to be available would be of little use in planning. The problem is largely removed by the fact that farm resource-using activities fall into four groups according to their relative needs for precision in time-specification, viz.:

· Highly time-specific: These are farm production activities which must be operated according to some tight time schedule or calendar of operations. Examples are found in the production of vegetables/fruit/poultry/fish to be marketed on festival days such as Id or Chinese New Year. They must be ready on that day, not three days earlier or they would spoil, nor one week later or there would be no demand. All production operations leading up to final sale will be according to a close timetable. Thus, for the activities producing these commodities, the time dimension of inputs might be specified as closely as within a given week, perhaps even on a given day, rather than within, say, a looser one-month timeframe. In planning this type of enterprise it follows that the resource pool must also be specified in similar small quantity-time units: e.g., so many labour days or so much transport in the first/second/third week of August etc. Note also that production on a continuous-flow basis throughout the year (as with orchids, pigs, coconuts etc. on some specialist farms of Types 3, 4, 5 and 6) implies ensuring adequate resource availability on a continuous basis throughout the year. Paradoxically, such production is thus highly time-specific on a continuing basis.

· Moderately time-specific: Typical of these activities is the planting/production of a crop under seasonal irrigation conditions. The appropriate time-unit for operation scheduling will usually be as broad as one month. In South India, the monsoon will probably arrive in late May and provide enough rain for paddy fields to be ploughed during June. The irrigation channels will start flowing in July to permit planting. July, August and September are the growing months, followed by harvesting some time in September. In this case, crop input requirements, and farm resource specification, need be on only a monthly basis (e.g., June: oxpower and labour; July: planting labour; September: harvest labour etc.).

· Low time-specific: Activities in this category include some rain-fed crop production where the agricultural year typically consists of only two periods, the wet and dry seasons. If the farming system (in terms of the number of production activities) is not very complex, it may be sufficient to define farm resources only on the basis of season. The timing of operations will be quite flexible and if a wet-season crop is not to be followed by a dry-season crop it will not matter much, within limits, how long some mature crops are left in the field (e.g., for cassava this might be many months).

· Not time-specific: A few agricultural enterprises are not managed with respect to time or the seasons. A small two-cow dairy herd kept for household milk supply is one example. Commonly such stock are run on the common lands along with the other livestock of the village. They receive no special attention whether they are in milk or not, and inputs are the same throughout the year. The resource supply for this type of activity need not be specified as having a time dimension.

Family labour resources

Obviously family labour is a very important resource on small farms. In some areas of high population pressure, small farms might support nucleus and extended families of up to ten or more persons, all of whom except the very young and very old can supply some labour. But the actual available supply is often difficult to measure because family labour has quantity, quality, time and often custom dimensions. Difficulty in measurement arises from the fact that the different family age/sex population classes often generate different amounts of labour service (e.g., as differentially provided by young men, women, older children and grandparents), and that some farm operations/tasks are labour-type specific while others can use any class of labour. The custom dimension complicates things further by insisting that certain tasks which could well be done by men (or women) must in fact be done by women (or men) (van Herpen and Ashby 1991).

Assuming that such activity-specific available labour has been identified, the procedure for quantifying the general family labour resource is to standardize all remaining labour according to some common quality unit, frequently adult male equivalents (AME). In the following example, an extended family of 18 persons shown in column (2) is converted to a general farm labour force of 9.1 'adult male equivalents' in column (4). When dealing with these typically heterogeneous populations, this or a similar standardization procedure is necessary. But the limitations should be noted.

Col (1)

Col (2)

Col (3)

Col (4)

Age Group in Years

Family Members

Conversion Factor

Work Force in Adult Male Equivalents





0 to 8






9 to 15






16 to 55






56 and over









First, the conversion coefficients of column (3) are necessarily subjective and thus somewhat arbitrary. Even if correct for one society, they might not be for another. In Nepal a man might be considered old at 50; but in parts of Iran he might be thought still capable of hard work at 70. Second, the male to female comparison suggested in column (3) is a generalization to which there will be many exceptions. There are many field tasks which women can or do in fact do better than men (e.g., planting seed, plucking tea, weeding, harvesting paddy). In these tasks the 1:0.8 comparison ratio of male to female labour is invalid (it possibly holds for heavy tasks). Finally, for some jobs children might in fact be equivalent in labour supply to only half an adult male, but for such jobs as herding livestock they are fully equivalent because one child can do as much as one adult.

Before leaving this topic, one further possible problem should be noted. The above procedure will result in a farm labour supply that is (it can be assumed) arithmetically valid but in some cultures this might still not be a measure of family labour actually available. Many families set a high value on what in developed countries would be called 'leisure' but which in South Asia would be regarded by even poor families as necessary participation in social and religious festivals, community affairs and village politics. The labour supply standardization procedure might result, for example, in an apparent labour supply of, say, 750 AME labour days; but when this is adjusted for non-farm socio-cultural demands on the family's time, the actual supply might be only half this. The point could be of obvious practical importance in the planning of new farms (on irrigation or settlement projects etc.) where such planning should obviously be based on actual rather than apparent family labour resources.

Resources versus assets

Quantifying a farm resource pool is roughly akin to constructing an asset statement for the farm family, but there are differences. An asset statement is a listing of all property owned and its value. Value might be based on the (imputed) productivity value of the assets (Section 7.2), but is more commonly based on their current market value. An asset statement might be prepared periodically as an instrument for measuring a family's economic progress over time (in terms of its changing 'net worth'), or occasionally as evidence of security in applying for a bank loan etc. A family's asset statement includes both farm and non-farm property and other items of financial value.

3.3.5 Resource acquisition and generation

As indicated in Section 2.2.2, the main structural characteristic of small farms of Types 1 and 2 is their orientation to internal generation of most of their needed resources. Farms of the other types must also generate their resources, at least in the long run, but do so through the medium of cash sales and external or off-farm purchase. The various ways by which small farms can obtain their resources are discussed in Sections 4.4.1 and 9.3.1.

3.3.6 Relationships between resources, capital and costs

Perhaps the best way to conclude this discussion of resources is to summarize the relationships between resources, capital and costs, as shown in Figure 3.3. This shows (a) how the five categories of resources discussed above become, in their economic dimension, either fixed or operating capital; (b) the various types of costs which are generated by the use of capital; and (c) how each resource/capital category (1) to (5) of Section 3.3.3 contributes to final total farm costs. Figure 3.3 serves as a useful introduction to discussion of the whole-farm service matrix (Section 5.3) and the evaluation of past whole-farm system ex ante operational planning (Chapter 9). While Figure 3.3 refers to a whole-farm system, the cost structure of an enterprise or activity subsystem would be analogous.

In Figure 3.3, starting with the first resource category of long-term fixed material resources in column 1, row (ii) shows this as becoming farm fixed capital. In row (iii) this type of capital gives rise to fixed capital costs. Then, going to row (iv), this group of fixed costs forms one component of total farm fixed costs (the other component of farm fixed costs being those fixed costs which might arise from institutional resources as discussed below).

Returning to column 1 and continuing down to row (v), this indicates the purposes for which total farm fixed costs are incurred: replacement of capital resources as these wear out or become obsolete, routine maintenance of these capital items, and (for some fixed capital or institutional items only) costs of operating them so long as these costs cannot be allocated to any specific farm activity or enterprise - if they could be so allocated, they would be included as a variable cost in column 3.

Continuing, row (vi) is a more precise description of each of the cost purposes of row (v): 'replacement' is achieved by setting up a depreciation fund (depreciation costs), 'maintenance' by incurring costs for repairs or periodic servicing of capital items, while 'operating' remains as 'operating'.

FIGURE 3.3 - Relationships among Categories of Resources, Capital and Costs in a Whole-farm System Context

Column 2 refers to institutional resources. In row (ii) these do not become either fixed or working capital (although one could argue that such institutional resources as a licence to produce and sell milk to a town market is a sort of capital). In row (iii) the use of institutional resources could generate a type of farm fixed costs known as general charges (see Section 5.3.1), e.g., the payment of a land or road or house tax which purchases the right to use or occupy these resources (as distinct from the physical resources themselves). In row (iv) such general charges or institutional fixed costs, if any, would combine with capital fixed costs from column 1 to give total farm fixed costs. In addition, in row (iii) some of the institutional resources might also incur variable costs (e.g., payment of an output-based tobacco production tax), in which case they would combine with the variable costs of column 3 and (if relevant) column 4 to give total farm variable costs in row (iv) or total direct production costs for the system in row (v).

Considering resource category columns 3 and 4 in Figure 3.3, these show short-term material and financial resources as operating or working capital in row (ii), which generates variable costs in row (iii); these in turn are totalled in row (iv) as total variable costs of the system. The specific purpose of these could be shown in rows (v) and (vi). Note that, as indicated in the diagram, while financial resources are used to pay for the fixed and direct farm costs, they may also incur variable costs through such items as bank fees and interest charges.

Family labour is shown as a resource in column 5 of Figure 3.3. In contrast to hired labour, it is usually not costed. Instead, family labour is taken as receiving income (for its services of labour and management) as a beneficiary of the farm system. However, as outlined in Section 7.2.5 and illustrated in Tables 7.8 and 7.9, family labour is costed on an opportunity cost basis in the calculation of such total productivity measures as total factor productivity and return on capital or equity.

Row (vi) of Figure 3.3 shows how all fixed and variable costs arising from the employment of the various categories of resources become total farm costs.

The specific types/categories of resources used in and generated by individual farm enterprises and activities are discussed in Sections 4.1 and 9.3.1.


Dillon, J.L. (1992). The Farm as a Purposeful System, Miscellaneous Publication No. 10, Department of Agricultural Economics and Business Management, University of New England, Armidale.

FAO (1990). Guidelines for the Conduct of a Training Course in Farming Systems Development, FAO Farm Systems Management Series No. 1, Food and Agriculture Organization of the United Nations, Rome.

Feldstein, H.S. and J. Jiggins (eds) (1994). Tools for the Field: Methodologies Handbook for Gender Analysis in Agriculture, Kumarian Press, West Hartford.

Feldstein, H.S., C.B. Flora and S.V. Poats (1990). The Gender Variable in Agricultural Research, IDRC, Ottawa.

Gabriel, T. (1995). The Human Factor in Rural Development, Belhaven Press, London.

Gondowarsito, R., C. van Stanten and T. Bottema (1995). 'The Role of Women in Upland Agriculture: Gender Issues Raised by Case Studies in Indonesia, Sri Lanka and the Philippines', Palawija News (CGPRT Centre Newsletter) 12(2): 3-15.

Makeham, J.P. and L.R. Malcolm (1986). The Economics of Tropical Farm Management, Cambridge University Press.

Prabowo, D. and D.J. McConnell (1993). Changes and Development in Solo Valley Farming Systems, Indonesia, FAO Farm Systems Management Series No. 4, Food and Agriculture Organization of the United Nations, Rome.

Quisumbing, A., L.R. Brown, H.S. Feldstein and L. Haddad (1995). Women: The Key to Food Security, Food Policy Report, IFPRI, Washington, D.C.

Rola, M.F.M. (1995). 'Gender Roles and Attitudes in Upland Farming Systems in the Philippines', Palawija News (CGPRT Centre Newsletter) 12(4): 1-12.

Shaner, W.W., P.F. Philipp and W.R. Schmehl (1982). Farming Systems Research and Development: Guidelines for Developing Countries, Westview Press, Boulder.

van Herpen, D. and J.A. Ashby (eds) (1991). Gender Analysis in Agricultural Research, Publication No. 204, CIAT, Cali.

Previous Page Top of Page Next Page