Farmers with dried up wells cultivate less land with fewer profitable crops, while taking up relatively more off-farm employment opportunities
Water is becoming increasingly scarce around the world. Rainfall is becoming more erratic, and groundwater tables are rapidly declining. The populations most vulnerable are likely the hundreds of millions of small-scale farmers, who still constitute the majority of the global poor, and crucially depend on irrigation water for their livelihoods. The manner in which these farmers will cope and adapt to water scarcity and other environmental changes may have dramatic implications for global food security, social stability, and progress towards achieving the Sustainable Development Goals. Will water scarcity impede progress in alleviating poverty in developing countries?
Optimists point out the availability of technologies and irrigation practices that can enable affected farmers to manage with less water, from rainwater harvesting to efficient drip irrigation. According to this view, farmers who face water shortages will adapt by taking up such technologies. However, given the myriad factors that may impede the adoption of profitable technologies by smallholders (for a review, see Jack 2013), such adaptation is anything but assured.
Pessimists warn that as water runs out, massive waves of ‘water refugees’ whose livelihoods have collapsed, may leave rural areas and inundate cities, creating social instability. The role that the extreme drought in Syria played in exacerbating the civil war there is often cited as an example.1
Should we believe the optimists’ predictions or the pessimists’?
A few years ago, we set out to provide empirical evidence to help inform the debate. The resulting study, which provides some of the first household-level data on the impacts of water access by low income farmers, covered about 100 villages all around the Indian state of Karnataka. The state is well-known for its booming capital city, the IT hub of Bengaluru, but surrounding rural areas are reeling under the pressure of persistent drought and depleting groundwater resources.
We found that the drying up of wells was extremely widespread (over 60% of wells had dried up). To understand how farmers whose wells had dried up were affected, one could consider comparing the situation of those farmers who still possess a functioning well to those who do not. However, one would then risk attributing any differences between these two types of farmers only to the status of the well, whereas in reality, other factors could also be responsible. For example, farmers who were wealthier to begin with might have drilled more, and deeper wells, thereby maintaining access to water for longer. In this case, wealth would be driving differences in water access, rather than resulting from it.
Disentangling impacts of failing wells from other factors
How can one disentangle the impacts of the drying up of wells from other factors? It turns out that the sub-surface geology of the state of Karnataka provides an opportunity to do so. In the regional hard-rock sub-surface, water accumulates in pockets that are randomly interspersed below the surface, at depths reaching hundreds of feet down. The locations of these pockets are impossible to determine from the surface, so whether and how long a well continues to draw water, is to a large degree dependent on how many such pockets it will intersect, which is largely a matter of chance.
Consistently with this hypothesis, the data we collected show that the depth, cost or the lifetime of wells drilled by a farmer are not systematically correlated with any of the observed socio-economic characteristics of a household. Better off and worse off farmers drill similar wells with similar chances of success. What is different is the number of wells they drill. In response to failure, farmers who are wealthier to begin with are more likely to try drilling a well again.
Our analysis, however, focuses on the first borewell ever drilled by a farmer and asks: given that two households located in the same village, and which have similar initial socio-economic attributes, have drilled their first borewell in the same year, and to a similar depth, but it so happens that one household’s well has dried up while the other’s is still functional, how do the two differ in cultivation practices, employment and income at present? Since the difference between the two wells can largely be attributed to geological luck, any differences found between the two households’ economies can more reliably be attributed to the failing of the well than to other confounding factors.
Farmers whose wells had dried up cultivated less land, while taking up off-farm employment
We found that farmers whose wells had dried up had their farm income severely affected, and about 25% lower than those whose wells were still operational. They cultivated less land, fewer times a year, and with less profitable crops – especially horticultural crops that require assured control of water. The data showed no indication that farmers found ways to adapt their cultivation to increasing water stress.
On the other hand, affected households took up relatively more off-farm employment opportunities, especially in non-agricultural employment. This employment shift seemed to have largely offset the loss of farm income, but only for those farmers residing in areas in which there was a high degree of manufacturing activity by large firms. However, this came at a cost: young adolescents were also pulled out of schools to take up these employment opportunities, which is likely to have substantial long-term costs.
These conclusions bear resemblance to those of a related study carried out in Gujarat that claimed there was very little evidence of agricultural adaptation (Fishman et al. 2016). However, the main coping strategy found in that study was the migration of younger male members of the household to a big city to seek employment.
One of the unique things about these results is that they capture the medium to long-term impacts of irreversible, gradual environmental change. The impacts we document occur some five years after the wells have dried up, on average, so they incorporate whatever adaptation has occurred since then. Moreover, affected farmers would have known this was coming, given that wells were failing all around them. And yet, we do not find indications that they have made substantial anticipatory investments.
In contrast, most of the literature on the impacts of environmental change is focused on transient climatic variability, often annual. It documents a wide variety of often severe impacts (for a review, see Carleton and Hsiang 2016), but leaves open the question of whether adaptive responses to longer-term environmental change would attenuate these short-term impacts. Our study offers some new evidence on this question.
Highlights and implications
Overall, the study provides a warning about the extent of the failure of wells resulting from groundwater depletion and persistent drought, and sheds serious doubts on the notion that farmers will be able to adapt their cultivation to the changing conditions. This suggests that increasing water scarcity can have grave consequences for future food production. On the other hand, the study also shows that economic growth in non-farming sectors, such as services and manufacturing, can help employ and sustain the incomes of affected farming households. Policies that can help agriculture conserve water and also encourage more firms to set up operations in rural areas, can both help avert the threat farmers who live in water-scarce parts of the world increasingly face.
Carleton, T A and S M Hsiang (2016), "Social and economic impacts of climate", Science 353(6304): aad9837.
Fishman, R, M Jain and A Kishore (2016), “When water runs out: scarcity, adaptation and migration in Gujarat”, Working Paper, Tel Aviv University.
Jack, B K (2013), "Market inefficiencies and the adoption of agricultural technologies in developing countries".