This page contains an overview of some of the research themes I am involved in, citing both published work and on-going research. You can click on the buttons below to take you directly to a thematic area.

Drinking Water Quality

In collaboration with Sheila Olmstead (Akram and Olmstead 2011), I sought to understand the willingness of poor urban households in Lahore (Pakistan) to pay for improvements to their water access and quality (as a service from the local government). This was my first foray into social science field work and involved designing the research from end to end – finding a suitable location and local partner, designing the survey and administering randomized WTP questions. The study used a contingent valuation survey to estimate the demand for drinking water quality among poor urban households in Lahore Pakistan. We found that households in Lahore are willing to pay amounts comparable to the monthly cost of in-home water treatment and about three to four times the average monthly water bill for piped water supply that is clean and drinkable directly from the tap.


My work with Sheila Olmstead showed that the poor are willing to pay for improved water access and quality. However, municipal governments in the developing world are unlikely to provide drinkable piped-water to the poorest anytime soon, even if they are willing to pay for it. In the meantime, poor households must arrange to clean their own water. Despite the availability of low-cost water decontaminating technologies like chlorine tablets, uptake is puzzling low in the developing world. In a randomized controlled trial in Karachi, Pakistan, Robert Mendelsohn and I test the hypothesis that a simple, visual recordkeeping of diarrheal incidence among their children helps households better understand the impact of chlorinated drinking water on the health of their children thus increasing their likelihood of tablet use. The intervention, which we called Info-Tool, was simple to implement requiring only paper-and-pencil and a short visit by our field staff. We found that this treatment doubled usage, even a year after treatment. We also found that children in the treated group weighed more than those in the control.*

My intent now is to refine Info-Tool and also to scale this intervention up. In this regard, I am in initial talks with BRAC, IPA and the World Bank, all of whom have expressed interest in replicating and scaling up my Info-Tool in Sub-Saharan Africa.

* We are finalizing our manuscript and readying it for publication. In the meantime, please see a blog that describes this on the Defeat Diarrhea website and I can email you a copy of an older version of the manuscript.



Parts of Northern Bangladesh experience an annually occurring lean season between September and November that is linked to the agrarian cycle when jobs off-farm are scarce – so the landless poor are not able to find work, therefore suffer deprivation. Simultaneously, nearby urban areas have relatively more employment opportunities. Yet those affected by this seasonal scarcity do not take advantage of these nearby employment opportunities. Gharad Bryan, Shyamal Chowdhury and Mushfiq Mobarak (2014) address this seeming puzzle and demonstrate that providing a small travel grant dramatically increases temporary migration, with a large associated increase in welfare, suggesting that the risk of migration is reduced by providing the grant.

While Bryan et al’s (2014) research design evaluated the direct effects of migration on beneficiary households, it did not answer questions about spillover effects. Shyamal Chowdhury, Mushfiq Mobarak and I expanded on that design in several ways during the 2014 lean season to study general equilibrium effects on the rural labor market. We offered travel subsidies to 5,792 potential seasonal migrants, randomly varying the proportion of landless agricultural workers across 133 villages induced to move, to generate labor supply shocks of different magnitudes in different villages. With this variation, we were able to document general equilibrium changes in the village labor market. We found that a larger number of simultaneous migration offers in the village increases the likelihood that each individual takes up the offer, which suggests some benefits of coordinated travel. Migration offers lead to large increases in income earned at the destination, and also income earned at the origin. The increase in home income is due to increases in both the agricultural wage rate for rural workers and in available work hours. For every 10% increase in emigration, male agricultural wage rate increases by 2.8%.

My co-authors and I are finalizing our research manuscript and Evidence Action is engaged in scaling up the program to about 300,000 households in Bangladesh.

Media coverage:

No Lean Season: Encouraging Seasonal Migration to Address Income Insecurity. Policy Brief, Yale School of Management.

No Lean Season: Reducing Seasonal Poverty. Policy Brief, the International Growth Center.


Pakistan has a complex irrigation canal network and a well-established management institution for this infrastructure but the principle for water allocation is rigid and may not be optimal. My work in this area started with the idea of thinking through the physical feasibility of a water market in Pakistan’s irrigation network (Akram, 2013). In my paper, I argue that a water market is physically feasible in the existing reality of Pakistan’s Indus Basin Irrigation System at the lower levels of the canal system. I detail the extent of modification to water structures that would be needed to enable trading based on structure type and the scale of the water-trading region, along with a first glance at the relative costs of those modifications.


Having explored the feasibility of a water market in the Pakistani irrigation context, I next turned to demonstrate and quantify the extent of inefficiency of water allocation in it, in collaboration with Robert Mendelsohn (Akram and Mendelsohn, 2017). Irrigation systems power agriculture in large parts of the developing world, yet there have been no studies that assess whether these systems allocate water efficiently. This is largely a data issue because agricultural production surveys in the developing world are not able to capture volumetric use of water by farmers – traditionally, proxy measures such as total irrigated area or number of surface irrigations applied are employed. We provide, for the first time, an assessment of canal irrigation water allocation based on actual in-season volumetric measurement of water use by farmers and provide simple tests of allocative efficiency. We use “traditional” measures of water use and find that these misleadingly indicate that allocation is efficient. Next, we use the more precise in-season volumetric measures of water use by farmers to demonstrate allocative inefficiency. Our analysis builds a strong evidence-based case for a water market, which would provide substantial welfare gains.


Media coverage: 

- Making Irrigation Work for Farmers. Panel discussion at the Consortium for Development Policy Research.

- How much are farmers losing from inefficient irrigation? Developing Pakistan Blog.

Irrigation Water and Agriculture


Air Pollution Monitoring Technology

A final area of interest for me is air pollution monitoring technology. As a freshly minted undergraduate in computer science back in 2004, I designed, acquired funding for and developed a low-cost air pollution monitoring network called the Volunteer Internet-based Environment Watch or VIEW (Ikram and Akram 2007). VIEW was deployed in the city of Lahore, Pakistan. I designed and wrote the funding grant for the second iteration of the air pollution monitoring network - VIEW2 (Ikram et al 2012). With my collaborator, Jahangir Ikram, I am actively looking for funding a third iteration of this air pollution monitoring network. My goal, now that I am a trained economist, is to generate data on air pollution in a developing country context that can be used by social scientists. This would be something truly unique, as air pollution data at a high spatial and temporal granularity does not exist for most developing countries, especially those in South Asia.