Evidence from a large-scale trial in Bangladesh shows that improving credit access in rural areas helped farmers adapt to flood risks without negative spillovers, and was profitable for microfinance institutions.
Climate change disproportionately affects developing countries
Extreme weather events are common and growing threats to homes and livelihoods, disproportionally affecting low and lower-middle-income countries. The agricultural sector, upon which 80% of the world’s poor depends for survival, is particularly vulnerable to floods, storms and drought that destroy crops, livestock and other assets (FAO 2021). The risks faced by low-income farmers are exacerbated by limited public and private resources to invest in mitigation, adaptation and response strategies.
Climate change adaptation strategies for poor farmers in vulnerable areas have often been ineffective due to obstacles preventing their adoption (Carleton et al. 2022). For example, financial institutions are reluctant to offer credit to individuals whose assets or income have been severely affected by extreme events, fearing they won't be repaid. Additionally, low-income households are often reluctant to buy insurance because they do not want to pay up-front premiums for uncertain benefits (Casaburi and Willis 2018). While some climate-resistant agricultural technologies, such as flood and drought-tolerant seeds, do exist, their adoption by low-income farmers remains low. This is because these seeds often cost more than traditional varieties and have uncertain benefits in non-shock years.
There is a small but growing body of research examining how new approaches tailored to specific risk areas can help alleviate the adverse effects of extreme climate events on low-income households. This study contributes to that work by examining how the provision of an 'Emergency Loan' following a flood can help low-income farmers more effectively adapt, improving their yields and income.
The need for climate adaptation in Bangladesh
The study takes place in Bangladesh, where nearly 80% of the country is located on floodplains and over 80% of rural households depend on agriculture for their livelihoods. In any given year, between 20 to 60% of the country can be submerged by floodwater (Brammer 1990).
The Government of Bangladesh does not have the resources to provide large-scale flood protection and relief, and existing tools and strategies to protect farmers from floods have had limited success. Moreover, at the time of this study, no financial institution in Bangladesh was willing to loan money to farmers after they had been hit by a flood.
Households in flood-prone areas tend to rely on low-risk strategies and are often reluctant to invest in potentially profitable ventures. This reluctance is due to high up-front costs during planting season when households are cash-constrained, and the uncertainty of returns. Following a flood, the vulnerability of poor households often deepens as they reduce consumption, sell assets, and take children out of school, lowering household income over time.
This study examines whether a new type of Emergency Loan can help vulnerable farm households by guaranteeing them credit after a flood. It explores whether such a loan improves their welfare by encouraging more investment in their farms before a flood and aiding in easier recovery afterward.
Measuring the impacts of the Emergency Loan on climate adaptation
The Emergency Loan was designed with BRAC, the largest microfinance organization in Bangladesh. It has over two thousand branches across the country, providing microfinance services for between twenty to sixty village organisations. BRAC loan officers visit village organizations on a regular basis to collect payments and answer questions, so households are familiar with how credit products work and tend to have high repayment rates.
The study involved 200 BRAC branches located in areas frequently exposed to flooding. Households in the sample were likely to have taken up a BRAC loan in the past year, and about half reported experiencing a flood in the past.
The 200 selected BRAC branches were randomly divided into two groups of 100 branches each. One group served as the control group and did not offer the Emergency Loan to their clients. The other group, the treatment group, offered the Emergency Loan to eligible clients. Clients were eligible if they had a credit score above a fixed threshold. This credit score was based on their past repayment behavior.
Over 150,000 clients in the treatment group were contacted one month before planting season and informed that they had been pre-approved for the loan if a flood occurred during the rest of the season. By providing the notice well before any crop decisions were made, farmers had the opportunity to invest in higher-return opportunities. BRAC loan officers emphasised to eligible borrowers that they did not have to make any upfront payments and could choose to take the loan if floods occurred. This was intended to help overcome households' aversion to making upfront payments for uncertain returns, a common constraint faced by insurance products.
Did credit access improve adaptation to climate change?
Three significant findings emerged from the study:
1. The Emergency Loan helps farmers adapt to flood risks.
Those that received the Emergency Loan increased the amount of land under cultivation by 18%, leading to a 19% increase in crop production on average. Among those farms that had access to the loan but who were unaffected by flooding, their gains were the highest: an increase in crop production of 35%. This suggest that farmers were more willing to extend their cultivation and reap the benefits due to the availability of the loan than they would otherwise.
Moreover, households that were hit by flooding, were less severely affected than those who did not have access to the Emergency Loan. Consumption levels for the former were 10% higher. The Emergency Loan mitigated the negative effects of the flood.
2. The Emergency Loan did not negatively affect households that were not eligible for the loan.
The increase in cultivated land by those eligible for the loan had no negative spillover effects on households that were not eligible. There was no reduction in available land for rent, nor did rental prices increase due to higher demand. Moreover, consumption levels in non-eligible households increased. The evidence suggests this is likely because more land under cultivation by loan-eligible households led to increased demand for agricultural labour, positively affecting those not eligible for the loan.
3. The Emergency Loan is profitable for microfinance institutions
Borrowers who accessed the Emergency Loan after a flood had higher repayment rates overall than experienced by other credit schemes. Branch profits increase, with the largest increases in profits coming from “marginal” clients who just qualified for the Emergency Loan.
This illustrates that lenders can minimise the default risk of customers affected by climate events by providing them with means to adapt in advance. Key elements of the success of the Emergency Loan were the existing relationship between the financial institution and eligible borrowers, and the loan being profitable for both parties.
Key takeaways for policy in developing countries affected by climate change
The Emergency Loan combines aspects of microcredit and insurance, addressing key limitations of both products. Specifically, the financial institution determines eligibility for the loan before the climate event occurs, overcoming the usual hesitation to lend money after the event. Moreover, eligible clients are guaranteed credit in advance without having to make an upfront payment. This encourages them to invest more in climate adaptation strategies, which they might have avoided without the guaranteed credit during a climate event.
The Emergency Loan can be scaled up and made more widely available. Scalability will depend on microfinance institutions having adequate records, including the lending history of their clients, to create a credit score that targets responsible borrowers. Not all microfinance institutions maintain such records. Additionally, branch managers need to be encouraged to adopt this new method, overcoming the traditional bias against lending to households at risk of a climate shock due to fears of loan defaults that could affect their performance. The evidence from this study should help ensure that the Emergency Loan is profitable for both the client and the lending institution.
As the severity and frequency of weather events increase with climate change, this form of credit represents a scalable and effective policy adaptation tool in low- and low-middle income countries. This study demonstrated its effectiveness in flood-prone areas, but the scheme could also be used in regions facing other severe environmental risks.
Editor’s note: Read more about research on adapting to climate change in our VoxDevLit on Climate Adaptation.
References
Brammer, H (1990), “Floods in Bangladesh: Geographical Background to the 1987 and 1988 Floods,” The Geographical Journal, 156(1): 12-22.
Carleton, T, Jina, A, Delgado, M, Greenstone, M, Houser, T, Hsiang, S, Hultgren, A, Kopp, RE, McCusker, KE, Nath, I, Rising, J, Rode, A, Seo, HK, Viaene, A, Yuan, J, and Zhang, AT (2022), “Valuing the Global Mortality Consequences of Climate Change Accounting for Adaptation Costs and Benefits,” The Quarterly Journal of Economics, 137(4): 2037-2105.
Casaburi, L, and Willis, J (2018), “Time versus State in Insurance: Experimental Evidence From Contract Farming in Kenya,” American Economic Review, 108(12): 3778-3813.
FAO (2021), “The Impact of Disasters and Crises on Agriculture and Food Security: 2021”