Elsevier

Food Policy

Volume 33, Issue 4, August 2008, Pages 318-328
Food Policy

Smallholder market participation under transactions costs: Maize supply and fertilizer demand in Kenya

https://doi.org/10.1016/j.foodpol.2007.12.001Get rights and content

Abstract

This paper assessed the effects of transactions costs—relative to price and non-price factors—on smallholder marketed surplus and input use in Kenya. A selectivity model was used that accounts for the effects of transactions costs, assets, technology, and support services in promoting input use and generating a marketable surplus. Output supply and input demand responses to changes in transactions costs and price and non-price factors were estimated and decomposed into market entry and intensity. The results showed that while transactions costs indeed have significant negative effects on market participation, institutional innovations—such as group marketing—are also emerging to mitigate the costs of accessing markets. Output price has no effect on output market entry and only provides incentives for increased supply by sellers. On the other hand, both price and non-price factors have significant influence on adoption and intensity of input use. Overall, the findings suggest that policy options are available other than price policies to promote input use and marketed surplus. The paper concludes with implications for policy to induce greater input–output market participation among smallholders in Africa.

Introduction

The need for promoting smallholder market participation has been increasingly recognized in efforts to bring about agricultural transformation in developing countries (von Braun and Kennedy, 1994). However, subsistence agricultural producers, especially in sub-Saharan Africa (SSA), face several barriers that make it difficult for them to gain access to markets and productive assets. The most significant of these barriers are argued to be transactions costs—the observable and unobservable costs associated with arranging and carrying out a transaction (Goetz, 1992, Staal et al., 1997, Holloway et al., 2000). Although market reforms have been introduced in many countries in SSA since 1980s with a view to enhancing the efficiency of input–output markets, transactions costs in production and marketing may have actually increased rather than decreased. There is a growing concern, for example, that public programs, which traditionally provided smallholders with extension, input supply, and credit support, have collapsed in response to the reforms (Jayne and Jones, 1997). By focusing on trader entry in output markets, the reforms have also ignored the production side of the supply chain, assuming that exchanges would arise at reasonably low costs to support production and trade (Winter-Nelson and Temu, 2005).

In a pioneering study, Goetz (1992) estimated a switching regression model of market participation and amount traded—separating the decision of whether or nor to participate in markets from the decision of how much to trade. He found that fixed transactions costs significantly hindered, while better information stimulated, smallholder market participation. However, by focusing on costs that arise in marketing and by taking production as given, the few extant studies have ignored the role of production shifters in generating a marketable surplus. Indeed, fixed transactions costs involve a production threshold in market participation, where any level of production below the threshold would not make market transactions feasible. Important implications emerge for empirical analysis of the effects of transactions costs on smallholder market participation. First, apart from access to information (Goetz, 1992) and institutional innovations, such as cooperatives (Staal et al., 1997, Holloway et al., 2000), production shifters are equally important variables that can promote market participation and supply. Second, to the extent that increased production promotes output marketing, an understanding of the effects of transactions costs on input use that increase production itself should be as important as that of the effect of transactions costs on marketed supply. Estimates of supply and input demand responses to price changes are also needed that account for heterogeneous input and out market participation, where some farmers actively participate in markets, while others are simply excluded from the market economy.

The objective of this paper is thus to assess the effects of transactions costs and price and non-price factors on output marketed supply and input use among maize producers in Kenya. The next section provides an overview of maize marketing in Kenya. In the third section, we provide a conceptual framework of market participation under transactions costs. The fourth section presents the empirical methodology, whereas the fifth section gives a description of the data used in the analysis. The empirical results are presented and discussed in the sixth section, and the last section draws conclusions and implications for policy.

Section snippets

Overview of maize marketing in Kenya

Maize is the most important food crop as well as a cash crop for many Kenyan smallholders, and modern input use has always been a critical policy issue for Kenya. Maize is grown on over 1.5 million hectares, with a total average annual production of over 2.5 million metric tons. Approximately 40% of maize produced in Kenya is marketed (Friesen and Palmer, 2002). The major maize surplus areas are the Rift Valley (Nakuru, Nandi, Kericho, Uasin Gishu and Trans-Nzoia), Western, and Nyanza

Defining transactions costs

Transactions costs are the embodiment of barriers to market participation by resource-poor smallholders and have been used as a definitional characteristic of smallholders and as factors responsible for significant market failures in developing countries (de Janvry et al., 1991, Sadoulet and de Janvry, 1995). In some cases, markets do not even exist. In others, high transactions costs must be incurred in accessing markets. In yet others, there are constraints on the quantities that can be

Empirical model

Most empirical studies on output marketed supply or input demand have used the famous Heckman’s (1976) sample selection model or its variants of double hurdle and switching regression models (e.g., Goetz, 1992, Coady, 1995, Minot et al., 2000, Winter-Nelson and Temu, 2005), while some used the more restrictive tobit model to analyze output marketed supply (e.g., Holloway et al., 2000). Because fixed transactions costs are expected to affect the decision to participate in a market, but not the

Data

The data used in this study come from a sample of 802 maize producers in eight districts in western Kenya, where maize is an important food as well as cash crop. The data were collected between September and December 2005 both for 2004 and 2005 agricultural years. The sample households were selected randomly from a total of 32 villages in eight major maize growing districts in Nyanza and Western provinces. Data were collected using structured and pre-tested questionnaires, which were

Maize marketed supply

The selectivity model estimates of maize marketed supply—first-stage probit model estimates of participation and second-stage selectivity- and heteroskedasticity-corrected regression estimates of the level of participation (i.e., marketed supply), are presented in Table 2. The selectivity model fits the data reasonably well—about 77% of the participation outcomes are correctly predicted and the Wald test of the hypothesis that all regression coefficients are jointly equal to zero is highly

Conclusions and policy implications

Using survey data on maize producers in Kenya, this study assessed the effects of transactions costs, relative to price and non-price factors, on smallholder marketed surplus and input use. A selectivity model was used that accounts not only for the effects of fixed and variable transactions costs but also for the role of assets, technology, and support services in promoting input use and generating a marketable surplus. Output supply and input demand responses to changes in transactions costs

Acknowledgement

Funding for this research was provided by the African Agricultural Technology Foundation (AATF). This is gratefully acknowledged. We thank Asman Wesonga, Vianey Rweyendela, Adetunji Olanrewaju, and Bamikole Ayedun for their excellent research assistance. We are grateful to the Nyanza and Western Provincial Directors of Agriculture and their staff for their support in the design and implementation of the survey. We are grateful to three anonymous reviewers for their very useful comments and

References (18)

There are more references available in the full text version of this article.

Cited by (228)

View all citing articles on Scopus
View full text