68255 Latin America and the Caribbean Region LCSSD Occasional Paper Series on Food Prices Improving Agricultural Productivity and Market Efficiency in Latin America and The Caribbean: How ICTs can Make a Difference? Aparajita Goyal and Carolina González-Velosa March 2012 Thanks to Augusto de la Torre, John Nash, Grahame Dixie, Willem Janssen, Nabil Chaherli, Marie-Helene Collion, Eija Pehu for many useful comments and discussions. This paper was prepared as a background paper for the World Bank regional study on Assessing Agricultural Export Performance in LAC. The views expressed in this paper are those of the authors and do not necessarily reflect those of the World Bank, its Board of Directors, or the countries they represent. Contact Address: World Bank, 1818 H Street, NW, Washington, DC 20433, Email: agoyal3@worldbank.org; Inter-American Development Bank, 1300 New York Avenue, N.W., Washington, D.C. 20577, Email: cagonzalez@iadb.org 2 work has been partly financed Market Efficiency in for Environmentally and Socially The Improving Agricultural Productivity andby the Trust Fund Latin America and The Caribbean Sustainable Development (TFESSD) LATIN AMERICA AND THE CARIBBEAN REGION LCSSD Food Papers Series IMPROVING AGRICULTURAL PRODUCTIVITY and MARKET EFFICIENCY in LATIN AMERICA and THE CARIBBEAN: HOW ICTS can MAKE a DIFFERENCE? Aparajita Goyal and Carolina González-Velosa March 2012 How ICTs can Make a Difference? 1 TABLE of CONTENTS I. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 II. The Agricultural Sector in Latin America and the Caribbean . . . . . . . . . . . . . . . . . . . . . . . . 4 III. Potential Benefits of ICTs in the Agricultural Sector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 IV. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 2 Improving Agricultural Productivity and Market Efficiency in Latin America and The Caribbean I. INTRODUCTION In the last decade, mobile phone penetration in Latin America and the Caribbean (LAC) grew at an impres- agricultural markets and increase their bargaining power. Moreover, ICT applications can help farmers meet the sive pace. Between 1998 and 2008, the average number escalating demands in distribution and quality control in of mobile phone subscriptions increased from 3.4 to 86.3 modern supply chains. ICTs can also play a role in the al- per 100 people (IDB 2011). Mobile phones in the rest of leviation of credit and insurance constraints in agricultural the developing world also grew at staggering rates. While economies by reducing the costs of rural service provision in 1999 only 10 percent of the African population had and of information. Finally, ICTs can influence the adop- mobile phone coverage, by 2008 over 65 percent of the tion of productive technologies in agriculture by reducing people in Africa had access to mobile phone service. the costs of knowledge dissemination systems such as Other more costly information and communication tech- extension services. nologies (ICTs) have also had a fast penetration. For in- However, even though ICTs are promising, limits to stance, in the 1998-2008 period, subscriptions to internet in their potential benefits can be expected. On the one the LAC region increased from 4.7 to 27.3 per 100 people hand, the adequate provision of ICTs, such as internet and the number of personal computers doubled in the services, often demands a minimum infrastructure and eight-year span between 1998 and 2006 (IDB 2011). The education level. For example, a main operational difficul- adoption of ICTs has not been limited to urban, wealthier ty in the provision of mobile phone services in Nigeria has households. For example, mobile phone ownership been the lack of reliable electricity supply, which is an es- among rural households in the Dominican Republic and sential requirement for the introduction of wireless services Guyana is as high as 50-60 percent (Jensen 2010). (Barret and Slavova 2011). In other cases, the low level of The rapid adoption of ICTs in LAC has generated human capital can be an even more relevant constraint. considerable optimism regarding its consequences on For instance, illiteracy is an impediment to the introduc- the economic development of the region. In the case of tion of ICT-related technologies such as text messaging agriculture, ICTs are especially promising. To the extent and internet (IDB 2011). that rural regions are sparsely populated and often have This paper attempts to present an overview of the ag- poor infrastructure and dispersed markets, the introduc- ricultural sector in LAC, discuss its distinctive features, and tion of technologies that reduce the cost of communica- the potential role of ICTs in improving agricultural produc- tion may lead to important transformations. tivity and market efficiency in this region. The discussion The literature that tries to assess the economic conse- in this paper will refer to the evidence provided by studies quences of ICTs on rural areas is still very recent. This paper that evaluate the impact of ITCs interventions. While the summarizes findings from very recent economic research emphasis will be put on the studies that evaluate interven- and complementary anecdotal evidence. The results, tions in the LAC region, there will also be references to thus far, are very promising and suggest that, by reduc- studies in other developing economies whenever these ing disparities and gaps in information, ICTs can help are pertinent to the LAC context. farmers find and exploit the opportunities offered in the How ICTs can Make a Difference? 3 II. THE AGRICULTURAL SECTOR in LATIN AMERICA and THE CARIBBEAN As is characteristic of developing economies, there has of agricultural growth on the reduction of rural poverty been a long-run steady decline in the relative importance has been a distinctive feature of Latin America’s agricul- of the agricultural sector in LAC countries overall. The con- tural economy (World Bank 2008a). Over the years, the tribution of agriculture to GDP fell steadily from 17 percent incidence of rural poverty has been surprisingly resilient, in 1960 to 6 percent in 2008. Moreover, while the share and has doubled the rate of urban poverty (Ravallion of employment in agriculture amounted to 54 percent et al. 2007). The rate of rural poor has remained above of the workforce in 1950, only 16 percent of workers were 50 percent since the 1970s in countries like Guatemala employed in agriculture in 2008 (World Bank 2011; Igle- and Honduras. Moreover, the number of rural poor has sias 1992). Despite these decreasing trends, agriculture increased in most countries, with the exception of Brazil, remains a sector of great importance for many countries Chile and Mexico (de Janvry and Sadoulet 2002). in the region. Indeed, aggregate figures at the regional In addition to a persistent rural poverty, the agri- level mask important differences across countries. While in cultural economies in LAC are characterized by a dual 2008, agriculture contributed at least 30 percent of total nature in which small traditional subsistence farms coexist employment in Honduras, Guatemala, Bolivia and Para- with large corporate landholdings. In fact, the LAC region guay, less than 1 percent of the workers in Argentina were has the world´s most unequal land distribution and those employed in agriculture (World Bank 2011). The relative who have access to land often use it inefficiently (UN importance of agriculture also varies within countries. For 2008). For example, in 2001, 90 percent of the total arable instance, the Mexican states of Zacatecas and Sinaloa land in Latin America was in large farms that accounted have the character of primarily being agricultural-based for 26 percent of the total number of farms, and land in economies (World Bank 2008a). these farms was often under-used or idle. The 50 percent Agricultural growth rates in the LAC region have been smallest farms, which accounted for 2 percent of the much slower than the rest of the developing world. In land, were subsistence farms in which land was generally the regions of East Asia, South Asia and Middle East and overused (de Janvry et al. 2001). Such inequality in land North Africa, the annual growth of agricultural GDP in ownership has not given rise to an active tenancy market, 1980-2004 exceeded 3 percent, while growth in Sub- on the contrary to what would be expected. Despite hav- Saharan Africa averaged almost 3 percent. In contrast, ing by far the world´s most concentrated land patterns, annual growth rates in Latin America’s agricultural sec- Latin America´s land rental markets are surprisingly thin tor did not reach 2 percent. However, due to a decline (de Janvry et al. 2002). in the agricultural population, the per capita annual To support the more disadvantaged smallholders, growth of agricultural GDP averaged 2.8 percent (World many countries in the region have implemented social as- Bank 2008b). Unfortunately, this growth in per capita sistance programs that often take the form of cash trans- agricultural income did not translate into benefits for fers targeted at the rural poor. These cash transfers are much of the rural population. In fact, the limited impact often conditional on the beneficiaries taking certain long- 4 Improving Agricultural Productivity and Market Efficiency in Latin America and The Caribbean run beneficial measures, such as enrolling their children LAC region, including small holders, can increase their at school or receiving vaccinations. In countries like Brazil, productivity and take advantage of the new opportuni- these cash transfers seem to have had a greater impact ties brought by the expansion of agricultural markets, on rural poverty reduction than agricultural growth. Thus, several structural issues should be tackled. These can be in spite of a booming agriculture, Brazil’s recent decline in classified into three broad categories: i) agricultural mar- rural poverty seems to be mostly driven by social assis- keting and supply chains; ii) agricultural insurance and tance and non-farm employment, rather than increased credit; and iii) adoption of productive technologies. agricultural earnings (World Bank 2008a). Some countries are turning to an alternative ap- AGRICULTURAL MARKETING AND proach in which poverty reduction can be diminished by SUPPLY CHAINS the increase of agricultural incomes instead of the provi- The commercialization of agricultural products has sion social assistance (World Bank 2008a). This requires suffered important transformations in recent decades, measures to foster agricultural productivity in a way that posing big challenges for farmers in the LAC region. First, enhances farm earnings among the poorest. Unfortunate- there are challenges at the initial stage of marketing, ly, the performance of agricultural productivity in many when farmers are required to identify their potential buy- countries has been unsatisfactory. With the exception of ers. There are also big challenges in distribution and qual- Costa Rica, countries in Central America and the Carib- ity control, to the extent that the food markets are being bean experienced very low productivity growth in the last transformed by the procurement practices of the new decade. Instead, the largest countries in the region, such supply chains. Finally, an effort should be made in terms as Argentina, Brazil, Chile, Colombia, Mexico and Venezu- of information and transaction costs in order to increase ela, made considerable improvements (IDB 2010). efficiency in the agricultural markets as well as the gains At an aggregate level there has been a satisfactory obtained by the producers. A more detailed description growth in LAC’s agricultural productivity since the 1960. of these issues follows. The LAC region has outperformed all other regions in the world except for high income countries. These produc- Finding the right buyers tivity gains have been entirely driven by technological In the first stage of marketing, farmers are required to improvements -- such as the adoption of more productive collect short-term information on the quality and quan- technologies and modern crop varieties-- as opposed tity currently demanded as well as long-term information to a more efficient use of the existing resources (Ludena on future market trends. Moreover, as food distribution 2010). Further gains in agricultural productivity are not systems become more integrated and globalized, farm- only important for the creation of better-paid agricultural ers not only require information on domestic consumers, jobs in the rural economies. They are also important if the but also on international markets. The collection of such countries in the region want to respond to the worldwide information can be quite costly in rural areas of LAC. If increases in food demand without affecting domestic no in-site sources of information are available, farmers food security. It has been estimated that worldwide food may need to travel personally to collect information. This production must increase by 70 percent in order to satisfy can involve significant costs due to the long distances in the global food demand in 2050. With the rising scarcity sparsely populated rural regions. Poor conditions of road and degradation of arable land, fresh water reserves infrastructure will worsen the situation. In this respect, the and biodiversity, the majority of the required increases in LAC region has a particularly poor record, lagging behind production should come from productivity gains rather non-LAC middle income countries in terms of road infra- than an expansion in the use of natural resources (FAO structure. A third of the population in the region has poor 2009). Thus, to meet a rapidly increasing global food road access, meaning they don’t live within 2 km of an demand, farmers in the region should make an effort to all-season passable road. In countries like Nicaragua, ac- increase their productivity. Transformations in the produc- cess to transport is particularly limited, reaching just over tive process may also be required if farmers want to take one-fourth of the population (Calderón and Servén 2010). advantage of the opening of new markets for high-value Some evidence on the consequences of these high in- primary and processed products (World Bank 2008a). formation costs has been collected for Colombia where, In order to find ways in which agricultural producers in the due to lack of information, many agricultural products are How ICTs can Make a Difference? 5 not produced or are inefficiently commercialized (Cama- México, where the US retail chain Wal-Mart and several cho and Conover 2011). national supermarkets have expressed interest in directly procuring chili habanero from small-scale growers. In spite Delivering the product on time of this, small-scale producers rarely distribute their output In the second stage of marketing, in which agricultural directly to these retailers because they lack the infrastruc- products are delivered to the consumers through supply ture and technology needed to comply with strict quality chains, the role of transportation and logistics is deci- control standards. Another example is provided by Cava- sive. Transformations in the procurement practices in the tassi et al (2009), who describe how Ecuadorian small po- agricultural industry are posing important challenges to tato growers have been unable to meet the volume and farmers in terms of supply chain logistics, quality assur- quality requirements of the multinational food processor ance and process management. Indeed, in recent years, Frito-Lay. the food industry has been subject to significant changes due to the industrial deregulation in many developing na- Getting the best price tions, the lowering of trade barriers in many industrialized If farmers are to increase their profits from agricultural countries and the incorporation of private market agents. marketing ventures, it is necessary to reduce informa- In this new industrial environment, private entrepreneurs tion and transaction costs that lead to inefficiencies and lead expensive supply chains that link consumers to agri- weaken their bargaining position. To take full advantage cultural producers. Moreover, supermarkets and the food of the opportunities offered by markets, agricultural pro- processing and food service industries play an increasing- ducers should have timely and accurate information on ly important role (World Bank 2008b, Beaumont et al 2011). the prices paid by potential buyers, the costs of alterna- By the early 2000s, in many Latin American countries retail tive distribution channels and the prices and outside op- food sales in supermarkets exceeded 50 percent of total tions of input suppliers. retail sales (World Bank 2008b). The demand for food Unfortunately, due to high information costs, many services is also growing rapidly, as “eating out� becomes farmers make their production and sales decisions in the increasingly popular in Latin American countries like Brazil, absence of sufficient information. Some evidence of this where spending on food services accounts for 22 percent issue has been gathered for the department1 of Boyacá, of food budgets (World Bank 2008b). in central Colombia. Camacho and Conover (2011) docu- If farmers want to participate in these new markets, ment how, in this region, 26 percent of the farmers don’t their products must meet escalating safety, quality and know the price of their product if it is purchased at the distributional requirements. Indeed, agri-food systems in farm, 43 percent don’t know the price of their product LAC are increasingly pervaded by food safety and quality at the municipal market and 63 percent don’t know the standards from the private sector which, although not price of their product in Bogotá. legally binding in the regulatory sense, may be de facto Such lack of information has an adverse impact in mandatory for farmers (IDB 2011). Adequate logistics are terms of efficiency, to the extent that optimal arbitrage also essential to overcome common problems, such as requires farmers to have full information on prices. Limited the inability to communicate timely orders to producers, information can also have distributional consequences the incapacity to fully trace the production cycle for cer- (Jensen 2010). As mentioned earlier, farmers rarely sell tification purposes and delays in the process of collection, directly to consumers; instead, there is usually a supply delivery and payment (Beaumont et al 2011). Meeting the chain composed of transportation agents, wholesalers, requirements of these value chains poses serious chal- retailers and other intermediaries. These intermediaries lenges to farmers in Latin America, especially to small can gain pricing power if the producers have limited holders who are often unable to cater to demanding information on alternative trading opportunities. In this supermarket standards (World Bank 2008a). Brehm et al way, information asymmetries lower the profits received (2007) illustrate this point with a case study in Yucatán, by producers from agricultural sales. 1 Departments are the largest subnational administrative units in Colombia, analogous to US states. 6 Improving Agricultural Productivity and Market Efficiency in Latin America and The Caribbean INSUFFICIENT AVAILABILITY OF INSURANCE different countries in the LAC region. They report that only AND CREDIT 8.4 percent of the surveyed rural households had savings Agricultural producers in LAC are exposed to numerous in a formal institution and only 3.4 percent had accessed sources of risk. Fluctuations in weather and in commodity credit from a formal source. prices translate into large shocks to agricultural income. Such constraints in the formal credit market have an Pests and disease outbreaks as well as risks related to especially high impact on the farmers’ ability to en- health shocks can have major impacts on yields. To cope gage in production ventures. Due to its seasonal nature, with these risks and uncertainty, agricultural households agricultural production often requires the payment of could resort to formal insurance arrangements that can upfront costs in anticipation of future returns. To surmount cover natural, biological and health hazards. However, the constraints faced in the formal credit market, many access to these products in the LAC region is still very farmers resort to informal sources of loans. However, in- limited. In Argentina, Brazil, Chile, Colombia, Costa Rica, formal credits are often more expensive and can also be Dominican Republic, Ecuador, Panamá, Paraguay and insufficient. In the absence of credit rating mechanisms Venezuela, at most 4 percent of the cultivated area is and proper collateral, informal lenders often charge high covered with a formal insurance product. México is an interest rates to offset the risk of no repayment. These exception to this low insurance coverage, with a share of higher interest rates may perversely attract only those insured farmland of 15 percent. The lack of coverage in farmers who have no intention of repaying, driving the the LAC region contrasts with access to formal insurance rates even higher and reducing the access to credit for in the US, where 75 percent of the farmland under cultiva- small farmers. Tejerina and Westley (2007) document the tion is insured (Werner 2005). insufficient supply of informal credit in 12 countries in the In the absence of formal insurance arrangements, LAC region by reporting that, even when both formal and farmers often resort to informal insurance networks com- informal sources are considered, only 14.1 percent of rural posed by family or community members. They can also households have access to credit. adopt alternative mechanisms to cope with risk. One alternative is taking steps to reduce their exposure to risk LIMITED TECHNOLOGICAL ADOPTION shocks before they occur. This is often achieved by adopt- As was mentioned previously, the adoption of produc- ing income smoothing strategies, such as engaging in tive technologies explains the steady improvement in conservative income activities or in crop diversification. To agricultural productivity in the LAC region since the 1960s. the extent that these production choices probably do not There are, however, important cross-country differentials in coincide with those taken on profit maximization grounds productivity trends. This reflects the fact that in many local only, they will typically yield lower returns. agricultural economies in the region there are important Households may also respond to risk shocks ex post, constraints to technological adoption (IDB 2010). Techno- by depleting nonfinancial assets, adjusting their supply of logical adoption may be limited by credit and insurance labor or forgoing investment decisions. Thus, the ability of constraints, which, as discussed previously, are prevalent in farmers to mitigate risk is affected by their access to finan- rural areas in LAC. Insurance and credit imperfections will cial services, which can be used to smooth consumption constrain the adoption of technologies that require large across different states of nature. Unfortunately, access upfront investment costs or whose returns are uncertain. to credit markets in rural areas in the LAC region remains Limits to adopting new technologies may also arise fairly limited in spite of the increase in banking competi- from informational inefficiencies. Farmers may simply not tion and the expansion of financial products that followed know of a technology that is beneficial or have no infor- the wave of financial liberalization in the 1990s. While the mation on how to use it. One of the more commonly used financial markets in urban areas benefitted from such policies in LAC to diffuse the information on technological reforms, financial services in rural areas remain underde- adoption is the provision of publicly-funded agricultural veloped and non-competitive (IDB 2010). Thus, formal extension services (IDB 2010). However, despite decades sources of credit in rural areas are scarce. Some evidence of investment in public extension programs, evidence of of this is presented by Tejerina and Westley (2007), who their impact in the LAC region is fairly limited (González et review more than 400 household surveys that span 12 al 2009). How ICTs can Make a Difference? 7 In fact, rigorous impact evaluations of agricultural overs that accrue to the wider community. For example, extension programs in other developing countries are also there is evidence that in developing countries, practices scarce. The existing evidence suggests that the effective- that control pests or reduce erosion will be adopted at ness of extension services varies across settings and that a lower level than what is optimal from the point of view the effects can be very weak. While this can be attributed of the community. (Jack 2011) To overcome these exter- to the methodological difficulties involved in performing a nalities, arrangements that align the individual farmers´ rigorous evaluation, the weak results can also be attrib- incentives can be developed. However, these may be uted to the quality of the agricultural extension systems ineffective if the actions that generate the externalities themselves. In many developing countries agricultural are costly to observe. For instance, Costa Rica´s program extension systems are barely functioning. This is partly due of payments to reduce deforestation has had little impact to the problems of scale involved in the provision of exten- on the actual deforestation rates because the majority of sion services in small-farm agricultural economies in which payments go to farmers that are unlikely to deforest in the farmers live in geographically dispersed areas. In this absence of the program (Pfaff et al 2008). environment, the provision of extension systems becomes Finally, the adoption of agricultural technologies will so costly that it may be financially unsustainable. Also, the also be constrained by insecure land rights. Investing weak performance incentives of field agents are often a in technologies with long-run returns will not be attrac- barrier to efficient extension. Monitoring the performance tive if farmers are uncertain about their property rights in of field extension agents can be quite costly given that the future (Jack, 2011). This is certainly an issue in several they often work in geographically disperse regions. countries in LAC, where land conflicts, expropriation and Agricultural technologies also remain at low levels of de facto ownership are common.2 adoption if they generate positive externalities or spill- 2 In fact, some scholars have argued that insecure property rights have hindered the development of a land rental market in several countries in Latin America such as Brazil and the Dominican Republic. (See, for instance, Alston and Mueller, 2010; Macours et al. 2004) 8 Improving Agricultural Productivity and Market Efficiency in Latin America and The Caribbean III. POTENTIAL BENEFITS of ICTS in the AGRICULTURAL SECTOR The rapid spread of ICTs in developing countries offers an Aker (2008) illustrates this point with a comparison of the opportunity to tackle some of the problems afflicting ag- per-search costs of searching for price information for ricultural economies in the LAC region. The most obvious different types of search mechanisms in Niger. While the impact that these technologies can have is the reduction approximate per-search costs of making personal visits in the information and communication costs in agricul- are, $USD 0.8, the cost of using mobile phones is $USD 0.2. tural economies. This can be quite beneficial, given that Mobile phones therefore allow people to obtain infor- rural regions are sparsely populated and often have mation on a more frequent basis and take a more active dispersed, poorly connected markets. Indeed, by reduc- role in the process of searching for information. To the ing information costs, ICTs can: i) increase the efficiency extent that farm producers can be better informed, they and producer surplus in agricultural markets; ii) strengthen may be more able to make efficient production and sales the provision of rural services; iii) facilitate the adoption of decisions. agricultural technologies. If the introduction of ICTs allows farmers to operate more efficiently and take advantage of the existing op- INCREASING AGRICULTURAL MARKET ACCESS portunities for arbitrage, there should be a reduction in AND GAINS WITH ICTS price dispersion across markets and time. An emerging Research on the impact of ICTs on agriculture is fairly body of research has provided direct evidence of this ef- recent. Recent literature provides some new evidence fect in developing countries. In an influential paper, Jen- indicating that ICTs can increase the farmers´ access to sen (2007) shows that the introduction of mobile phones markets and production profits mainly by reducing the in the state of Kerala, India, leads to a sizeable decline in costs of acquiring relevant market information (e.g., po- the price variability of fresh fish across markets and over tential buyers, prices of inputs, distribution channels) and time and to the complete elimination of waste. The results by supporting the logistics and quality control required to of Goyal (2010) point to the same direction. She finds access more demanding markets. evidence of a reduction in the spatial dispersion of the price of soybeans in the state of Madhya Pradesh in India, Lowering the cost of information where the introduction of internet kiosks allowed farmers Consider, for instance, the case of mobile phones, which to bypass intermediaries and obtain daily information of can significantly reduce the cost of acquiring market in- wholesale prices. Evidence of a similar effect of ICTs in formation in remote, rural areas. Once the initial fixed cost the LAC region is provided by Camacho and Conover of installment is paid for, the variable costs associated (2010). They implement an experiment that randomizes with the gathering of information via mobile phones are the price and weather information provided to farmers significantly lower than the equivalent costs of travelling. through text messages (SMS technology) in the depart- How ICTs can Make a Difference? 9 ment of Boyacá, Colombia. They find that, relative to the Fafchamps (2010) also suggest that there are intervening control group, farmers who received the text message factors that may hinder the possible impact of ITCs on the had a smaller dispersion in the expected crop price and market efficiency and the producer welfare. First, even a significant reduction in crop loss. in the presence of greater information, farmers may be The gains in efficiency from the introduction of ITCs unable to arbitrage in response to the additional infor- can translate into an increase in the welfare of agricultur- mation if the high transportation costs limit the access to al producers. ICTs may also increase the farmers’ surplus alternative input and output markets. This, as discussed if a greater access to information increases their market previously, is an important limitation in the LAC region. power. Goyal (2010) provides suggestive evidence of Also, even after obtaining additional information and these welfare effects in Madhya Pradesh, India, where the learning of better market opportunities, farmers may introduction of the internet kiosks allowed for the bypass- continue to trade with the marketing agents with whom ing of intermediaries and the creation of a direct market- they have repeated interactions. This will be the case if ing channel. As a result, there was not only a reduction there are interlinked transactions in which, for instance, in price dispersion but also an increase in the average traders extend credit to farmers in return for the exclusive soybean price received by farmers and an increase in rights to purchase their output (Jensen, 2010). Coon et the area of soy under cultivation suggesting the existence al. (2010) provide some illustrations of this kind of relation- of net welfare gains for farmers. Jensen (2007) also finds ships in Central America. For example, they describe how positive welfare effects from ITC´s, to the extent that the NicFoods, a Nicaraguan company dedicated to process introduction of mobile phones increased the profits of and export tubers, has acted as a guarantor for plan- fishermen in Kerala. tain farmers to access loans at a local bank. A second The evidence of increases in welfare effects for farm example is the case of Parmalat, a multinational dairy producers in the LAC region is still very limited. Beuer- corporation that provides credit to dairy farmers in Nica- mann (2010) examines the impact of an intervention ragua when they are unable to pay the upfront operation that provided at least one public (satellite) payphone costs of milk refrigerators. to rural villages in rural Peru that previously had neither landlines nor cell phones. Results show a sizable increase Supporting logistics and quality-control in agricultural profitability which the author attributes to ICT applications can also help farmers meet the escalat- an increase in the farmers’ bargaining power. Chong ing demands in terms of distribution and quality control et al (2009) also study the impact of the introduction of posed by modern supply chains. An interesting example payphones in rural Peru and find a sizable positive impact for the LAC region is the TRAZ.AR program in Argentina. on agricultural income. This program provided small and medium cattle farm- However, the empirical literature has not always found ers with internet-connected software that allowed them a welfare enhancing effect of ICTs on farmers. Aker and to track each animal from the time of its delivery until Fafchamps (2010) find that while mobile phone coverage the meat is distributed and thus follow the evolution of reduces producer price dispersion, it does not increase the stock and the sanitary situation of the animals. Since producer prices in Niger. Suggestive evidence indicates tracking is an essential quality requirement in many inter- that the limited welfare-enhancing effects could be driv- national value chains, the use of TRAZ.AR strengthened en by insufficient mobile phone coverage in remote areas. the competitiveness of cattle farmers in the global meat In the LAC region, Camacho and Conover (2010) present market (IDB 2011). Moreover, through the use of this pro- evidence in the same direction. They find that sending text gram farmers improved reproduction selection, reduced messages with price and weather information to farmers in animal stress and improved sanitation. Galiani and Jait- Boyacá, Colombia, has no significant impact on agri- man (2010) find that, relative to a control group with simi- cultural prices, revenues or household expenditures. The lar characteristics, farmers in the TRAZ.AR program were authors argue that the lack of effects could be explained less affected by a severe drought experienced during the by the short duration of the intervention. period and were able to sell at better prices. While the absence of ITC on farmers’ welfare could ICT applications that provide similar services for farm- be associated to the characteristics of the program, the ers in the developing world are becoming more common. results of Camacho and Conover (2010) and Aker and Choudhary and Sen (2011) describe several software ap- 10 Improving Agricultural Productivity and Market Efficiency in Latin America and The Caribbean plications that have helped overcome supply chain prob- inclusion since only 4 million people own a bank account lems in countries like Kenya, India and Bangladesh. These in Kenya (IDB 2011). However, even though this technol- systems typically support the communication of orders ogy has provided “banking for the unbanked�, it should to farmers, the coordination of storage and distribution be noted that most of the transactions occur within urban logistics, the tracking of the produce along the supply areas. In Colombia’s coffee sector m-banking services chain for quality control purposes and the process of have also been introduced. In a very recent pilot proj- payment. While some of these software applications are ect, coffee growers have been given access to mobile sophisticated and costly, there is also a supply of cheaper phones with which they can perform financial transac- systems that are affordable for small holders since they tions including the purchase of inputs (IDB 2011). rely on lower-cost ICT devices such as, for example, mo- ICTs have served a similar function by reducing the bile phones or PDAs. cost of providing public services to the rural poor. An ex- Choudhary and Sen (2011) note how many of the ICT ample is the electronic implementation of the conditional applications for supply chain management are provided cash transfer programs in Colombia and Mexico (Familias by the private sector. To the extent that private compa- en Acción and Oportunidades, respectively). The subsi- nies support the operations with a viable business model, dies are being transferred through electronic payments these interventions are likely to be sustainable. In this re- to individual and collective bank accounts (IDB 2011). The gard, an interesting example is the ICT intervention led by Bhoomi project in the state of Karntaka, India, is also an ITC Limited, a large buyer of soybeans in India. In order to interesting example. In this project, several million land bypass intermediaries and lower the transaction costs, ITC records were computerized and made publicly avail- Limited set up internet kiosks in rural villages that enabled able through a network of rural land-record kiosks. To the farmers to access daily information on wholesale prices of extent that public officials were bypassed, this process soybeans in local markets as well as the price offered by reduced the opportunities for corruption and eased farm- ITC. As discussed by Goyal (2010), this intervention was fi- ers’ access to documentation needed for land transac- nancially sustainable, for it was profitable both for farmers tions and loans. (McNamara 2009). and for ITC. In contrast, similar efforts undertaken by NGOs ICTs can also have an impact in the exposure to risk or governments to remove intermediaries and provide in- of agricultural households. As discussed earlier, agricul- formation to farmers have not had the same success due tural economies are inherently risky and, in the absence to their lack of sustainability. of formal insurance mechanisms, farmers often resort to members of their social network for informal insurance. STRENGTHENING FINANCIAL SERVICES AND To the extent that ICTs improve communications among MITIGATING RISK WITH ICTS the social network, they can increase the effectiveness ICTs can facilitate the provision of services in agricul- of these informal insurance arrangements. For example, tural economies by reducing the costs of reaching and mobile phones can increase the speed of information servicing scarcely populated rural areas. In the case flows within the network, allowing them to respond more of banking, a very successful innovation has been the rapidly to shocks. However, the evidence on the effect of introduction of mobile financial applications (a.k.a “m- mobile phones on social networks is limited, but the topic banking�). These systems facilitate the operation of a has been extensively studied in the field of sociology. variety of financial transactions via mobile phones, such ITCs may also facilitate the transmission of information on as the payment of bills or the transfer of money among potential shocks. This was the goal of the intervention in bank accounts. Boyaca, Colombia, in which farmers were provided with M-money technologies have been adopted through- price and weather related information via text messages out the developing world, in Asia, Africa and LAC region. (Camacho and Conover 2010). An improved system of Particular attention has been given to the Kenyan m- alerts natural disasters are also a good example of the banking service, M-Pesa, which has had a dramatically role of ITCs in reducing risks. In response to the 2010 earth- rapid penetration, attracting one million subscribers in quake, the Chilean government initiated an SMS earth- the first ten months after its introduction. There are cur- quake alert system program that, by 2012, should have rently 9.5 million subscribers (23% of the population) which incorporated all mobile phones in the country (IDB 2011). implies a remarkable achievement in terms of financial Another example of how the adoption of ICTs can affect How ICTs can Make a Difference? 11 risk exposure is presented by Muto and Yamano (2009), From a different standpoint, ICTs can also improve the who study the impact of a large expansion in mobile effectiveness of agricultural extension. In particular, they phone coverage in Uganda. After the expansion of the can be used to improve the accountability of exten- coverage, there was an increase in the sales of banana sion services by facilitating the collection of agricultural but not of maize. The authors argue that the greater information. For example, instant messaging systems can impact on banana is due to the fact that, as opposed be used to collect information on technological adop- to maize, banana is a perishable product whose price tion and use of inputs on a more frequent basis than the depends heavily on freshness at the time of sale. The new regular agricultural surveys. Mobile phones may also be flow of information made available by the mobile phones used to verify the extension agents visits (Aker, 2008). allowed the farmers to reduce the risks of producing ba- ICTs can also be beneficial to the adoption of ag- nanas by facilitating a timely coordination in production ricultural technologies by strengthening social ties and and transportation that avoided spoilage. the diffusion of private information on technologies. The adoption of productive technologies can be sped up if HOW ICTS CAN INCREASE THE ADOPTION OF there is increased communication between farmers and AGRICULTURAL TECHNOLOGIES other technological adopters. The economics literature One of the main policies put forth by many countries provides interesting evidence on the role of peer effects, to promote the adoption of agricultural technologies is knowledge spillovers and learning externalities on the agricultural extension. Extension services were conceived adoption of agricultural technologies. In an influential of and developed in response to the limited access that paper, Foster and Rosenzweig (1995) provide evidence of farmers often have information on farming technologies. learning spillovers in farming technologies in rural India. To the extent that ITCs reduce the costs of communica- They find that farmers with neighbors that have adopted tion, they can ease the provision of effective extension new technologies (i.e., high-yielding seed varieties) de- services. vote more land to the new technologies and have more “Traditional� ICTs such as television and regular radio profitable farms. Conley and Udry (2010) exploit a rich broadcasts have been used for long to support the ser- set of data on the communication patterns of pineapple vice of agricultural extension. Governments around the farmers in Ghana and provide evidence of social learn- world are now incorporating new versions of extension ing. Specifically, they show that farmers align their level services that are supported on more “modern� ICTs, such of fertilizer input with the amount used by farmers in their as voice-based information systems, SMS and e-learning. information neighborhood who were successful in the The voice-based information systems primarily consist of previous season. call-in centers and hotlines that provide information on Finally, by facilitating the degree of cooperation farming methods and market access. SMS-based exten- among economic agents, ITCs may also influence the sion services essentially use text-messages via mobile adoption of technologies with spillovers and externali- phones to disseminate information. E-learning programs ties. Evidence of this type of effect is quite scarce. There usually consist of internet kiosks or centers that allow is, however, anecdotal evidence from the Huaral Valley farmers to access agricultural information on the inter- in Perú indicating that the installation of telecommuni- net. While the introduction of these new technologies cation information centers improved the distribution of can prove beneficial, there are challenges to support- water from irrigation sources and helped the communities ing the diffusion of agricultural extension via ICTs. On the coordinate its use in times of water scarcity (IDB 2011). one hand, the effectiveness of the technology is highly dependent on the type of information provided. For example, while the information on crop prices can be easily transmitted by text messages, these may prove inadequate to disseminate more nuanced, complex infor- mation about agricultural practices. Also, some of these technologies require users to have some literacy level and technological knowledge. 12 Improving Agricultural Productivity and Market Efficiency in Latin America and The Caribbean IV. CONCLUSIONS The rapid dissemination of ICTs in rural areas in LAC has empower smallholders. ICTs can also foster productivity by been received with a lot of optimism, as these tech- facilitating the dissemination of technological knowledge nologies are thought to be potentially effective tools of and expand the access to financial and public services agricultural development. Moreover, ICTs can become among the rural population by making service provision sustainable development interventions to the extent that more affordable. they can be delivered as viable businesses for the private Nonetheless, to the extent that the effective provision sector. However, rigorous analyses of the impacts of ICTs of ICTs has certain minimum requirements in terms of hu- on agriculture are still very scarce and lag behind the man and physical capital, many agricultural economies rapid penetration of these technologies. This paper sum- will be unable to reap the full benefits of these technolo- marizes recent findings from some of the few academic gies. Given the speed of innovation in the ICT industry and studies addressing this topic. The analysis is complement- the diversity and complexity of its applications, there is an ed with anecdotal evidence and findings from case stud- urgent need for more analyses on their potential benefits ies. Overall, the available evidence indicates that ICTs and pitfalls. More rigorous evidence is necessary to guide can play a major role in promoting agricultural productiv- development practitioners and policy makers on how to ity and rural development in LAC. By closing information harness the opportunities that ICTs can bring to the agri- gaps and reducing transaction costs, ICTs can improve culture sector in Latin America. the opportunities of farmers in agricultural markets and How ICTs can Make a Difference? 13 REFERENCES • Alston, Lee and Bernard Mueller (2010) “Property Rights, Land Conflict and Tenancy in Brazil� NBER Working Paper No. 15771 • Aker, Jenny (2008) “Does Digital Divide or Provide? The Impact of Mobile Phones on Grain Markets in Niger� BREAD Working Paper No. 177 • Aker, Jenny and Marcel Fafchamps (2010) “How does mobile phone-coverage affect farm- gate prices? Evidence from West Africa�. Mimeo. Medford: Tufts University • Barrett, Michael and Mira Slavova (2011) “Making ICT Infrastructure, Appliances, and Services More Accessible and Affordable in Rural Areas� In: “ICT for Agriculture e-Sourcebook� Agricul- tural and Rural Development Department & infodev Washington DC: World Bank • Beaumont Dixie, Grahame and Nithya Jayaraman (2011) “Strenghtening Agricultural Market- ing with ICT� In: “ICT for Agriculture e-Sourcebook� The Agricultural and Rural Development Deparment & infodev World Bank • Beuermann, Diether (2010) “Information and Communication Technologies, Agricultural Pro- ductivity and Child Labor in Rural Peru�. Mimeo. College Park: University of Maryland • Camacho, Adriana and Emily Conover. (2011) “The Impact of Receiving Price and Climate Information in the Agricultural Sector� IDB Working Paper Series No. IDB-WP-220. Washington DC: IDB • Cavatassi, Romina, Mario González, Paul Winters, Jorge Andrade-Piedra, Graham Thiele and Patricio Espinosa (2009) “Linking small holders to the new agricultural economy: an evaluation of the Plataformas program in Ecuador.� Working Papers 09-06. Agriculture and Development Economics Division of the Food and Agriculture Organization of the United Nations. Rome: FAO • Brehm, Kevin, Amanda Enrico, Cheray Kiendl, Emily Morgan,Alexandra Teachout and Ka- tie Vasquez (2007) “Globalization of food retailing and transformation of supply networks: consequences for small-scale agricultural producers in Southeastern Mexico� Journal of Latin American Geography, Vol 6. No. 2 • Calderón, César and Luis Servén (2010) “Infrastructure in Latin America� Policy Research Work- ing Paper No. 5317. Washington D.C: World Bank • Chong, Alberto, Virgilio Galdo y Máximo Torero (2009) “Access to Telephone Services and Household Income in Poor Rural Areas Using a Quasi-natural Experiment for Peru.� Economica, London School of Economics. Vol. 76, No. 304, pp. 623-648 • Choudhary, Vikas and Soham Sen (2011) “ICT Applications for Smallholder Inclusion in Agri- business Supply Chains� In: “ICT for Agriculture e-Sourcebook� The Agricultural and Rural Development Deparment & infodev Washington DC: World Bank • Conley, Timothy G. and Christopher R. Udry (2010) “Learning about a New Technology: Pine- apple in Ghana.� American Economic Review. Vol 100, No. 1, pp. 35-69. • Coon, Jeremy, Anita Campion and Mark Wenner (2010) “Financing agriculture value chains in Central America� Technical Notes. No. IDB –TN-146. Washington DC: IDB • de Janvry, Alain and Elisabeth Sadoulet. (2002) “Land Reforms in Latin America: Ten Lessons toward a Contemporary Agenda�. Document prepared for the World Bank’s Latin American Land Policy Workshop, Pachuca, Mexico, June 14th 2002. http://are.berkeley.edu/~sadoulet/ papers/Land_Reform_in_LA_10_lesson.pdf 14 Improving Agricultural Productivity and Market Efficiency in Latin America and The Caribbean • de Janvry, Alain, Elisabeth Sadoulet and Karen Macours (2002) “El acceso a tierras a través del arrendamiento�. In: “El acceso a la tierra en la agenda de desarrollo rural�, Serie de in- formes técnicos del Departamento de Desarrollo Sostenible, Washington DC: IDB • de Janvry, A., G. Gordillo, J. P. Platteau and E. Sadoulet (2001) “Access to Land, Rural Poverty, and Public Action� Oxford: Oxford University Press. • Dirven, Martine (2004) “Alcanzando las metas del milenio: una mirada hacia la pobreza rural y agrícola�. Serie Desarrollo Productivo. Red de desarrollo agrícola. Santiago de Chile: CEPAL • FAO (2009) “How to feed the world in 2050� Briefings of the High-Level Expert Forum. Rome: FAO • Foster, Andrew and Mark Rosenzweig (1995) “Learning by Doing and Learning from Others: Human Capital and Technical Change in Agriculture.� Journal of Political Economy. Vol 103, No. 6, pp. 1176-1209. • Galiani, Sebastián and Laura Jaitman (2010). “Traceability applied to SMEs in Argentina: A Case Study�. Mimeo. Washington DC: IDB. • González, Verónica, Pablo Ibarrarán, Alessandro Maffioli, and Sandra Roz (2009) “The Impact of Technological Adoption on Agricultural Productivity: The Case of the Dominican Republic�. Office of Evaluation and Oversight. Washington DC: IDB • Goyal, Aparajita (2010) “Information, direct access to farmers, and rural market performance in Central India� American Economic Journal: Applied Economics Vol. 2, No. 3, pp. 22-45 • IDB (2010) “Assessing the Effectiveness of Agricultural Interventions� Development Effective- ness Overview. IDB • IDB (2011) “Development Connections: Unveiling the Impact of New Information Technolo- gies� Washington DC: IDB • Jack, Kelsey (2011) “Constraints on the adoption of agricultural technologies in developing countries.� White paper, Agricultural Technology Adoption Initiative, Boston: J-PAL (MIT) and Berkeley: CEGA (UC Berkeley) • Jensen, Robert (2007) “The Digital Provide: Information (Technology), Market Performance and Welfare in South Indian Fisheries Sector�. Quarterly Journal of Economics Vol. 24., No.3, pp: 1057-942 • Jensen, Robert (2010) “Information, Efficiency and Agricultural Markets�. Agricultural Econom- ics, forthcoming. Available at: http://ageconsearch.umn.edu/bitstream/53206/2/ICTs_Agri- culture_Jensen.pdf • Iglesias, Enrique V. (1992) “Agricultural Development in Latin America and the Role of Re- search and Technological Change: An Overview�. Sir John Crawford Memorial Lecture. Wash- ington DC: CGIAR • Ludena, Carlos E. (2010). “Agricultural productivity growth, efficiency change and techni- cal progress in Latin America and the Caribbean.� IDB Working Paper Series No. IDB-WP-186. Washington DC: IDB • Macours, Karen, Alan de Janvry and Elisabeth Sadoulet (2004) “Insecurity of Property Rights and Matching in the Tenancy Market� Cudare Working Papers. Berkley: UC-Berkeley • Mendoza, Rene, and Johan Bastiaensen (2003) “Fair Trade and the Coffee Crisis in the Nica- raguan Segovias.� Small Enterprise Development, Vol. 14, No. 2 How ICTs can Make a Difference? 15 • McNamara, Kerry (2009) “Improving Agricultural Productivity and Markets: The Role of Infor- mation and Communication Technologies� Notes. Agriculture and Rural Development. Issue 47. Washington DC: World Bank • Muto, Megumi and Takashi Yamano (2009) “The Impact of Mobile Phone Coverage on Market Participation: Panel Data Evidence from Uganda� World Development, Vol. 37. No. 12 • Pfaff, Alexander, Andres Robalino and G. Arturo Sanchez-Azofeifa. (2008) “Payments for Environmental Services: Empirical Analysis for Costa Rica�. Stanford Institute of Public Policy. Working Paper Series. SAN08-05. Durham: Duke University. • Ravallion, Martin, Shaohua Chen and Prem Sangraul (2007) “New Evidence on the Urban- ization of Global Poverty� Policy Research Working Paper. Development Research Group. Washington DC: WB • Tejerina, Luis and Glenn Westley (2007) “Financial Services for the Poor. Household Survey Sources and Gaps in Borrowing and Saving� Technical Paper Series. Sustainable Development Department. Washington DC: IDB • United Nations (2008) “Trends in Sustainable Development�. Economic and Social Affairs. New York: UN • Wenner, Mark (2005) “Agricultural Insurance Revisited: New Developments and Perspectives in Latin America and the Caribbean� Rural Development Unit- RUR-05-02. Washington DC: IDB • World Bank (2008a) “The Agenda for Latin America and the Caribbean�. Agriculture for Devel- opment Policy Brief. Washington DC: WB • World Bank (2008b) “World Bank Development Report: Agriculture for Development� Wash- ington DC: WB • World Bank (2011) “World Development Indicators�. Washington DC: WB 16 Improving Agricultural Productivity and Market Efficiency in Latin America and The Caribbean About the series: The LCSSD Occasional Paper Series is a publication of the Sustainable Development Depart- ment (LCSSD) in the World Bank’s Latin America and the Caribbean Region. The papers in this series are the result of economic and technical research conducted by members of the LCSSD community. The series addresses issues that are relevant to the region’s environmental and social sustainability; water, urban, energy and transport sector development; agriculture, forestry and rural development; as well as cross-cutting topics related to sustainable develop- ment such as climate change; logistics; crime and violence; and spatial economics. While all papers in this series are peer reviewed and cleared by the LCSSD Economics Unit on behalf of the Director of LCSSD, the findings, interpretations, and conclusions expressed in this paper, as in all publications of the LCSSD Occasional Paper Series, are entirely those of the authors and should not be attributed in any manner to the World Bank, to its affiliated organizations or to members of its Board of Executive Directors or the countries they repre- sent. The World Bank does not garantee the accuracy of the data included in this paper and accepts no responsibility for any consequences of their use. For environmental sustainability reasons as well as ease of distribution, the primary vehicle for the dissemination of this series is electronic, with papers available on the following web- site: www.worldbank.org/lcssdpapersseries