Monga and seasonal migration

Mushfiq Mobarak of the Yale School of Management, USA, takes on the issue of seasonal hunger with an eye to the role of migration in mitigation. Throughout the world, approximately 805 million people are reported to be food insecure. Of them, three-fourths (600 million) are rural. Among these 600 million, just half are said to suffer from seasonal hunger. Seasonal hunger is often found to afflict agrarian economies between planting and harvest seasons - called lean season in the literature - when rural families, particularly the landless agricultural wage labour, experience hunger. During this period, food stock falls and demand for agricultural labour decreases, thus, adversely affecting income and meals for a two- to three- month period.
The lean season, known as morakartik or monga in northern part of Bangladesh, is linked to the agrarian cycle. During the pre-harvest lean season, between September and November, farmers wait for the crop to grow. Meantime, they face misfortunes in many ways such as scarcity of jobs in the villages, falling wages, rising prices, etc. It has been estimated that close to half of Rangpur's about 16 million inhabitants live below the poverty line, and are subjected to the sufferings of the lean season. Interestingly, the nearby urban and peri-urban areas do not face the same seasonal downturns. This contrast can be construed as a seasonal labour misallocation, and a window of opportunity for rural workers to find seasonal work in urban or semi-urban centres.
However, the researchers reckon, despite this apparent spatial arbitrage opportunity, the areas prone to seasonal poverty display a low degree of domestic remittances - a feeble 5.0 per cent compared to the rest of the country. Work by Gharad Bryan, Shyamal Chowdhury and Mushfiq Mobarak suggests that "people living close to the margin of subsistence are unwilling to take on the risk of sending a member away for work. During the lean season, for the very poor a small chance that the cost of migration fails to generate income could push household below the subsistence level of income. Thus, the persistent risk of the lean season among the landless poor could stem from a poverty cycle, in which the extreme poor fail to take advantage of profitable income opportunities because they are so poor."
However, seasonal temporary migration can address low food consumption and income reduction during the lean season in a cost-effective way. Rigorous evidence suggests that a programme offering small travel grants to low-income agricultural workers enables them to migrate during the lean season to nearby areas with higher wages and better work opportunities and consequently mitigate the adverse effects of the seasonal downturn at home.
Evidence in favour of this intervention is supported by a series of randomised evaluations conducted in Bangladesh between 2008 and 2014 that have demonstrated persistent impact on the take-up of seasonal migration for employment and welfare gains for families at risk of famine. These studies provided small travel grants ($8.50 USD) covering the cost of a round-trip bus ticket and a couple of days of food. The evidence is quite compelling that the provision of the subsidy substantially increased migration rates. In the original study, grants increased migration rates by 22 percentage points (a 61 per cent increase compared to households in non-incentivised villages). Moreover, this effect extends beyond the intervention period. Households targeted by the intervention were 9.0 percentage points (25 per cent) more likely to migrate in the subsequent lean season in 2009, and 7.0 percentage points (22 per cent) more likely to migrate two-and-a-half years later, in 2011-without receiving an additional travel grant. Many of the grant recipients travel back to work for the same employer.
The evidence also shows that seasonal migration in response to receiving a travel grant generated large welfare gains for migrant families. This includes an increase in caloric intake of 758 calories per person per day in 2008 and 435 in 2009; and an increase in consumption and expenditure of Tk 355 (37 per cent) per person per month in 2008 and Tk 260 (23 percent) in 2009. All of these effects are statistically significant. The $8.50 investment therefore generated very large returns, suggesting that this type of programme offers a cost-effective intervention in areas that face a lean season and disparities in available wages.
The most recent study (2014) confirmed these results and also showed important spillover effects. Migration rates increased in both the low-intensity and high-intensity villages. Consistent with results of the 2008 study, about one-third of the households in control villages sent a seasonal migrant (34.2 per cent). Households offered a grant in the low-intensity group were 26.4 percentage points more likely to migrate than a household in a village where no grant offers were made. Each household in the high-intensity group had a 42.6 percentage point higher propensity to migrate than households that were not offered an incentive. This is evidence of an important spillover - the take-up rate of the migration offer is significantly higher when a larger number of people are simultaneously planning to travel. This positive spillover even extends to those not directly receiving migration offers. Households that did not receive an offer in the high-intensity village had a 12.2 percentage point greater propensity to migrate than households in control villages. High spillover rates underscore existing high demand for the intervention, as long as the risk is mitigated by friends and family travelling simultaneously. Additionally, these spillovers have implications for cost-effectiveness of the intervention at scale: offering travel grants to greater numbers induces higher take-up rates.
Subsequent migration (without any incentive provision) shows the long-term impact of the intervention. The migration rate in 2009 was 9.0 percentage points higher in villages where incentives were provided a year before, even though the incentive offers were not repeated in 2009. The data indicate that those who had more successful migration experiences in 2008 chose to re-migrate a year later using their own funds. Even two and a half years later in 2011, during a less extreme lean season, without any further incentive, the migration rate was 7.0 percentage points higher in the villages randomly assigned to receive cash or credit in 2008.
The writer is a Professor of Economics at
 Jahangirnagar University.
[email protected]