Centre for the Study of African Economies
Department of Economics
.
University of Oxford
.
Manor Road Building
.
Oxford OX1 3UQ
T: +44 (0)1865 271084
.
F: +44 (0)1865 281447
.
E: csae.enquiries@economics.ox.ac.uk
.
W: www.csae.ox.ac.uk
CSAEWorkingPaperWPS/2013Ͳ09
Mobile Phone Coverage and Producer Markets:
Evidence from West Africa
Jenny C. Aker
and Marcel Fafchamps
May 2013
Abstract. Expansion in mobile phone coverage has improved access to information throughout the
developing world, particularly within sub-Saharan Africa. The existing evidence suggests that
information technology has improved market efficiency and reduced consumer prices for certain
commodities. There are fewer studies assessing the impact of the technology on producers. Using
market-level data we estimate the impact of mobile phone coverage on producer prices in Niger. We find
that mobile phone coverage reduced the spatial dispersion of producer prices by 6 percent for a semi-
perishable commodity, cowpea. These effects are strongest for remote markets and lowest at harvest time.
Mobile telephony, however, has no effect on price dispersion for millet and sorghum, two storable crops.
There is also no impact on the average producer price, but mobile phone coverage is associated with a
reduction in the intra-annual price risk, primarily for cowpeas. These findings are confirmed by data from
a farmer-level survey: we find that farmers owning mobile phones obtain more price information but do
not engage more in spatial arbitrage and hence do not receive higher prices – except for peanuts. The
additional evidence presented here helps understand how mobile phone coverage affects agricultural
market efficiency in developing countries. It suggests that the impact differs across agents – depending on
whether they use the information for arbitrage or not – and across crops – depending on whether inter-
temporal arbitrage is possible or not. (JEL O1, O3, Q13)
Key words: Africa, Information, Information Technology, Market Performance, Search Costs, Niger.
*Jenny C. Aker: Department of Economics and the Fletcher School, Tufts University and Center for Global Development, 160
Packard Avenue, Medford, MA 02155. Email: Jenny.Aker@tufts.edu. Marcel Fafchamps: Department of Economics and Center
for the Study of African Economies, University of Oxford, mfafchamps@economics.ox.ac.uk. This research was partially
funded by Rocca Dissertation Fellowship, Catholic Relief Services, CARE, World Vision, the Ford Foundation and UC-
Berkeley’s CIDER. We are grateful for comments from three anonymous referees, Reena Badiana and participants at the Center
for the Study of African Economies (CSAE) and American Economic Association (AEA) conferences. Daniel Hammer provided
excellent research assistance. All errors are our own.
2
1. Introduction
Price information plays an important role in arbitrage behavior and market efficiency. Access to
such information has improved significantly in developed countries, especially with the introduction on
online databases (Autor 2001, Anderson and Magruder 2012). In sub-Saharan Africa, limited
infrastructure has historically made obtaining information costly. Over the past decade, however, the
spread of mobile telephony (Aker and Mbiti 2010) has enabled consumers and producers to send and
receive information more quickly and cheaply. This paper examines how this has affected the spatial
integration of agricultural markets in Niger.
In Niger farmers typically sell to traders, either by transporting their output to a local market or
by waiting for traders to come and purchase it from their home village. Traders then sell the goods
throughout the country. Between 2001 and 2008, mobile phone networks were introduced in Niger, with
over 44 percent of the population having access to mobile phone service by 2008 (GSMA 2010). Even if
service remained limited in the villages where most farmers live, by 2008 90 percent of local agricultural
markets in our sample had mobile phone coverage. For farmers and traders, mobile phones reduced the
cost of obtaining market information by 35-50 percent (Aker 2010).
We estimate the impact of mobile phone coverage on the spatial dispersion of producer prices and
on producer price levels. We develop a simple model explaining how the circulation of information
among traders is likely to improve spatial arbitrage. The model shows how, in the context of Niger, better
circulation of information about local market conditions reduces the spatial dispersion of producer prices
because it allows traders to reorient their purchases from high to low price markets. As local prices adjust,
some farmers are hurt while others benefit – and hence the average producer price need not be affected.
The model also predicts the spatial arbitrage effect to be strongest when intertemporal arbitrage is absent,
as would be the case for perishable commodities.
Mobile telephony may affect agricultural prices in ways other than improving spatial arbitrage.
Lower information costs may, with sufficient competition between traders, lead to a reduction in their
gross margins, i.e., to a reduction in the average difference between producer and consumer price. Better
3
informed farmers may also reduce the monopsony power of traders, putting added downward pressure on
trader margins. Because intertemporal arbitrage can smooth producer price fluctuations even in the
absence of spatial arbitrage, we expect little or no effect of mobile telephony on gross margins for
storable commodities, but a possible effect for perishables.
Reduced gross margins can translate into either higher producer prices or lower consumer prices.
A straightforward application of tax incidence theory predicts that if supply is much more price elastic
than demand, a reduction in the wedge between producer and consumer price results in a lower consumer
price but leaves the producer price unchanged, and vice versa. Applied to our setting, this means that a
reduction in trader gross margins may either lower consumer prices or increase producer prices, or both,
depending on the relative elasticities of demand and supply.
We test these predictions using detailed information about producer prices in agricultural markets
over time. To estimate the impact of mobile phone coverage on producer price dispersion, we use a
difference-in-differences estimation strategy. We control for market-pair and monthly fixed effects and
for variables that could be simultaneously correlated with mobile phone coverage and producer price
dispersion. We also check for pre-treatment balance on observables and formally test the validity of the
parallel trends assumptions.
Results suggest that mobile phone coverage reduced the spatial dispersion of producer prices by 6
percent for cowpeas, a semi-perishable commodity. We do not observe the same effect for millet or
sorghum. However, we find heterogeneous effects, with larger effects for more isolated markets, during
certain periods of the year, and for surplus markets. We do not find that mobile phones are associated
with a reduction in gross margins for cowpea, millet or sorghum.
Next we investigate whether markets in which mobile telephony was introduced experienced an
increase in producer prices relative to markets without mobile phone access. We find no evidence that
mobile phone coverage translated into higher producer prices for either millet or cowpea, but reduced
intra-annual price risk for cowpea by 6 percent. To explore this question further, we use a farmer survey
to measure the impact of mobile phone ownership – rather than coverage – on prices received. Overall
4
results are consistent with those from the market level data – except for peanuts where we find mobile
phone ownership to be correlated with a positive and statistically significant increase in the price received
by farmers. Unfortunately we do not have peanut price data at the market level.
We also use the farmer-level data to assess the impact on farmers’ marketing behavior. Aker
(2008) found prior evidence that improved access to information changes traders’ search and arbitrage
behavior. The analysis presented here indicates that mobile phone ownership is associated with better
access to market information but does not significantly change farmers’ marketing and arbitrage behavior
and is not reflected in a higher price received. Using an experimental methodology, Fafchamps and
Minten (2012) report similar findings for India.
This paper makes two contributions. First, the results contribute to a substantial economic
literature showing that information is crucial for the effective functioning of markets, both from a
theoretical (Stigler 1961, Reinganum 1979, Stahl 1989) and empirical perspective (Autor 2001, Brown
and Goolsbee 2002, Jensen 2007, Aker 2010, Goyal 2010). Recently some of this evidence has been
extended to producer markets for commodities that are grown by a large percentage of farmers in the
developing world (Goyal 2010, Fafchamps and Minten 2012, Muto and Yamano 2009). Yet few of those
studies are able to assess the interaction between producer and consumer markets as we do in this study.
Second, much of the existing literature concentrates on the impact of information technology on a
single good (Jensen 2007, Aker 2010, Goyal 2010). To our knowledge, only Muto and Yamano (2009)
assess the relative impact of information technology on perishable and non-perishable commodities. Our
study is not only able to look at price levels and market participation for multiple goods, but also at
producer price dispersion over a long time period.
1
The rest of this paper proceeds as follows. Section 2 provides an overview of the context and
research design. Section 3 presents the theoretical framework. In Section 4 we present our data and
Section 5 discusses the empirical strategy. Section 6 provides the main empirical results. Section 7
concludes.
1
In addition, having pre-treatment price data makes it possible to test the validity of the parallel trends assumption.
5
2. Context
2.1. Agricultural Markets in Niger
With a per capita Gross Domestic Product (GDP) of US$330 and an estimated 61 percent of the
population living in extreme poverty, Niger is one of the lowest-ranked countries on the United Nations
Human Development Index (United Nations Development Program 2010). Agriculture employs more
than 80 percent of the total population and contributes approximately 41 percent to GDP (World Bank
2010). The majority of the population consists of subsistence farmers who depend upon rain-fed
agriculture and livestock as their main source of food and income. The main grains cultivated are millet
and sorghum. Cowpea is the primary cash crop and is primarily exported to Nigeria. Peanuts and sesame
are grown as cash crops in some areas.
Millet and cowpea are produced by smallholders in the Sahelian and Sudano-Sahelian agro-
climatic zones of Niger. Although both commodities can in principle be stored for several years, over 90
percent of farmers and traders in our study area do not engage in inter-annual storage. Unlike millet and
sorghum, cowpea is highly susceptible to pests during storage and is considered semi-perishable.
A variety of market intermediaries are involved in taking agricultural commodities from
producers to rural and urban consumers. Small-scale farmers typically sell their agricultural products to
traders in their village or in a nearby market. Traders then sell to wholesalers in local markets, who in
turn sell to buyers (wholesalers, retailers or consumers) in regional markets. As there is only one rainfed
harvest per year, traders and farmers engage in intra-annual storage but seldom store for more than a year
(Aker 2008).
Trade in agricultural commodities takes place through a system of weekly markets. Across all of
Niger, the average distance between any pair of markets is 350 km. Farmers are located on average 7.5
km away from the closest weekly market. While an agricultural market information system has existed in
