MALFUNCTIONING OF AFRICAN RURAL FACTOR MARKETS: THEORY AND A KENYAN EXAMPLE
| Published date | 01 May 1983 |
| Date | 01 May 1983 |
| Author | Paul Collier |
| DOI | http://doi.org/10.1111/j.1468-0084.1983.mp45002001.x |
OXFORD BULLETIN
of
ECONOMICS and STATISTICS
Volume 45 May 1983 No. 2
MALFUNCTIONING OF AFRICAN RURAL FACTOR
MARKETS: THEORY AND A KENYAN EXAMPLE
Paul Collier
I. INTRODUCTION
This paper is a study of the failure of rural factor markets. An instance
of this failure is the paradoxical finding that among Kenyan small-
holdings there is an inverse relationship between farm size and the use
of hired labour per hectare. Transactions in the hired labour market
actually amplify large differences in labour input per hectare between
farms of different size, and thus run precisely counter to the predic-
tions of conventional analysis. Explanations for such paradoxical
transactions in rural factor markets are developed at two levels. First,
existing contractual forms of transactions are shown to be subject to
inefficiencies which would indeed generate observed behaviour. More
fundamentally, Kenyan factor market transactions are shown to be
constrained from adopting modes of contract common in rural areas of
many Asian developing countries. The consequences of these con-
straints for output, equity and poverty are then analysed. Whilst our
evidence relates only to Kenya it is possible that the theoretical analysis
may have a more general application in rural Africa.
In rural Kenya, as in most developing countries, there are substantial
differences in both absolute and relative factor endowments among
smallholder households. Such differences arise partly because house-
holds are at different stages in the life-cycle and thus have different
labour entitlements (see Sen, 1975), and partly because of underlying
entitlement differentiation resulting from demographic, ecological or
economic processes. Thus, cross-section descriptions of household
entitlements reveal widely different proportions of land, labour, fixed
capital, material inputs and finance. If all households have access to
the same production functions, then these different proportions
generate differences in marginal products prior to market transactions.
In turn these differences in marginal products create the opportunity for
mutually profitable transactions in either factor or commodity markets.
141
142 BULLETIN
If market transactions were unconstrained, costless, certain and
enforceable, then the particular form of the transaction would be
undetermined. It would make no difference whether landowners hired
labour or workers rented land, whether capitalists rented land and hired
labour, or acted as money lenders. A further alternative would be for
some degree of specialization in commodity production which could
remove the need for any factor market transactions. In a world of such
markets, poverty would be related directly to a lack of entitlements.
The 'exchange-entitlement mapping' (the ability to convert given
entitlements into meeting wants, see Sen, 1981) would be the same for
all households.
In practice, rural factor and commodity markets are not character-
ized by costless, certain and enforceable possibilities for exchange.
Labour transactions may involve an uncertain quantity of 'effort' on
the part of the worker and may involve some skill formation. Land
transactions may involve uncertainty over changes in land quality due
to cultivation practices, and in the possibility of re-possession at the
end of the contract, both viewed from the perspective of the landlord,
and an uncertain land fertility as viewed by the tenant. Credit trans-
actions may involve uncertainty concerning repayment. In addition to
these contract-specific risks and uncertainties which arise from the
conflicting interests of the parties to a contract, both parties operate
in an environment of objective uncertainty concerning climate and crop
prices and so may wish to enter into insurance or hedging transactions.
Because exchanges in rural markets potentially involve these
elements of risk, quality uncertainty, and enforcement difficulties, the
contractual form needed to overcome these obstacles to efficient trans-
actions can be complex. A contract confined to a single transaction in a
single market to be undertaken now at a specified price and quantity -
the ideal-typical 'spot' contract - is often inappropriate. Sharecropping
is an instance of a single contract blending transactions in the labour,
land, credit and insurance markets. Recent theoretical work has estab-
lished that sharecropping is an economically rational contract superior
to other forms of contract in environments in which there are multiple
risks or limited information about labour quality. Empirical work on
Asia has established that sharecropping is generally efficient in the sense
that output per acre is not lower than on owner-operated farms.' Many
labour transactions in rural Asia are now understood to be only part of
multi-transaction contracts which include credit transactions (see
Bardhan (1980) and Bardhan and Rudra (1980)). Such packaging of
transactions into a single contract can be seen as attempts to overcome
the obstacles which make 'spot' contracts unattractive.
In some circumstances there may be no feasible contract which
enables efficient transactions to take place in a particular market.
'On both the theory and Asian evidence see Biswanger and Rosenzweig (1981).
MALFUNCTIONING OF AFRICAN RURAL FACTOR MARKETS 143
However, this need not upset the direct connection between poverty
and entitlements. This connection appears robust because it is only
broken if all of the possible mechanisms for preserving a common
exchange entitlement mapping fail simultaneously. Initial differences in
land/labour ratios between households will give rise to differences in
exchange-entitlement mappings only if neither specialization in com-
modity production nor efficient exchange in land or labour markets is
feasible. In the event of such a global failure of markets, poverty is not
determined purely by the same process which determines entitlements.
The complex of market failures results in both a generalized loss in
efficiency and in a different distribution of income compared with
the case of efficient exchange. To understand poverty it becomes
necessary to understand the constraints which confine feasible
contracts to inefficient exchanges in all markets.
Factor and commodity markets in rural Kenya are unusual for the
absence of contractual complexity in transactions. In contrast to rural
Asia there is no sharecropping nor any linking of labour and credit
transactions into single contracts. It will be argued that this is not
because Kenyan circumstances are such that risk, quality and enforce-
ment problems do not render 'spot' contracts unattractive. Rather, it
will be suggested, there are in Kenya additional constraints which
preclude certain modes of contract. The roots of these constraints are
a non-capitalist conception of property rights over land, combined with
widespread absenteeism among land owners. In this environment land
owners cannot monitor labour input, making labour hiring inefficient,
but land hiring is precluded by the limited nature of the rights of land
holders. Because of these contractual constraints upon transactions no
market exchange mechanism remains by which a common exchange
entitlement mapping is generated. The consequence of this market
failure is that rural poverty becomes a compound of low entitlements
and an unfavourable exchange entitlement mapping.
In order to substantiate this thesis we need first to show that
efficient transactions in Kenyan factor and commodity markets are
contractually infeasible. This is undertaken in sections II (the labour
market), III (the land market), IV (the commodity market) and V (the
credit market). Section VI brings together the analysis of the causes of
these market failures and assesses their consequences. Section VII
places the analysis in a dynamic setting and investigates how con-
tractual modes have changed and are likely to change in the future.
II. LABOUR TRANSACTIONS
In sub-section (a) data are compiled on the malfunctioning of the
Kenyan rural labour market. Readers prepared to take stylized facts
To continue reading
Request your trial