The World’s Rural Poor Lack Access to Credit, Is Technology a Solution?
This post was written by Utsav Manjeer, PhD student in Economics at Northwestern University, in collaboration with Christopher Udry (Northwestern University) and Monica Lambon-Quayefio (University of Ghana). Their research on digital credit for farmers in Ghana is supported by a grant from CEGA’s Digital Credit Observatory.
The Realities of Agricultural Credit and the Role of Technology
Timely and adequate access to credit is key to the success of small-scale farmers, and the overall growth of the agricultural economies of developing countries. The FAO estimates that 3 billion of the developing world’s rural poor population lives in 475 million farm households, with plots of land averaging less than 2 hectares. Inadequate infrastructure and a shortage of formal banking institutions in rural areas leads to a persistent and systemic shortage of much-needed credit to rural, agricultural households. A lack of credit has the potential to reduce investments in agriculture, and as a result, reduce future income and investments in health care, human capital, and the broad quality of life. Technology, in the form of digital credit, may offer a path to overcome these physical barriers since it is instant, automated, and remote.
What is the role of technology in this space?
Prior to the digital credit boom, obtaining loans in rural, unbanked areas entailed subscribing to prohibitively expensive loans in the informal market or several costly visits to distant banks. Difficulties with formal authorities, a lack of credit histories, the perception that farmers have high default rates, and other factors could exacerbate the cost of these bank visits.
In Ghana, an innovative digital platform called Paytime, developed by the researchers’ study partner, Farmerline Ltd, utilizes a credit scoring algorithm that farmers can access on their smartphones, or with a local agent (more on that later). With this platform, loan applications are completed digitally and a loan decision is made within days. The farmer no longer needs to travel repeatedly to brick-and-mortar banks or engage in complex interactions with bank officials.
The Impacts of Digital Loans on Farmers
Our team is studying digital credit for farmers in the Ashanti region in central Ghana. We seek to understand whether providing farmers credit through Paytime prior to the planting season can impact farmer investment decisions and influence financial behavior, yields, and profits. Funded by CEGA’s Digital Credit Observatory (DCO) and supported in the field by Innovations for Poverty Action (IPA), we randomly provide loans, in the form of agricultural inputs, to farmers who may also access several other of Farmerline’s services.
While Paytime’s digital loans can technically be transferred through mobile money, we directly provide loans to farmers in the form of agricultural inputs such as seeds and fertilizer. This allows us to answer our first research question: what is the impact of providing digital input loans to farmers on their investment decisions, and subsequent harvests, yields, and profits?
By randomly assigning credit access to farmers from within the group of farmers already deemed eligible for a loan, we ensure that we are comparing farmers who actually get a loan with other farmers capable of getting a loan. Therefore all farmers in the study are likely to be similar in all other aspects, except that they are not given a loan. Ineligible farmers are likely to be very different eligible farmers, and this might affect how they utilize the loan and their agricultural outcomes.
The Role of Agents
Lending to poor rural farmers is perceived as extremely risky by formal lending institutions; the institutions lack both the knowledge of farmers’ potential returns on investments and the capacity to actually monitor the farmers and recover the loans made. Our setting offers a chance to explore one potential mechanism to deal with this information asymmetry — loan agents. These agents are typically locals living in close physical proximity of the loan applicants.
We hypothesize that agents may be able to bridge the gap between the lending institution and the farmer, since their geographical proximity allows for knowledge of idiosyncratic ground realities, an understanding of the farmer’s quality, and the monitoring and influencing of farmer’s choices, thereby reducing the risk of default. This information may serve as an important complement to other forms of digital credit that rely solely on digital data streams including call detail records and social media activity.
To test our hypothesis, we assign loan agents to two distinct types of loans made to farmers. If agents recover “special incentive loans,” they will receive an extra bonus, over and above the regular payment they receive for the successful recovery of loans. By observing whether farmers assigned to special incentive loans behave differently compared to farmers assigned to the same agent but without a high-powered incentive from the agent’s perspective, we aim to shed light on agent influence over farmer choices. This exercise helps answer our second research question: can high-powered incentives for agents influence farmers’ choices, and reduce the risk of default?
Our study aims to investigate two distinct policy challenges. With our first research question, we aim to provide insight into the benefits of digital credit that small-scale farmers can accrue in terms of investment decisions, yields, and profits. With our second research question, we can comment on the sustainability of such a business model by incentivizing local loan recovery agents to reduce farmers’ default rates.
We hope that this research helps answer whether and how digital credit can be extended in agriculture across developing countries.
