When to Purchase Food Aid Locally?

by Will Lynch

Download in pdf

The United States is the world's largest provider of food aid. This aid can be used only to purchase U.S. commodities, an often-criticized policy known as "tied" aid. The alternative to the purchase of U.S. commodities is local purchase in the state or region where the food is needed.

It's easy to assume that food aid purchased near the site of the emergency is the optimal response. Proponents argue that the food will be available quicker and will cost much less than imported food aid shipped from the United States. There is an element of truth to this, which is what makes local purchase an attractive idea.

Under limited circumstances, adequate local surplus stocks of appropriate foods can be delivered to victims of emergencies in a timely manner. A few examples from India demonstrate how local purchase can help.

Following the January 25, 2001 earthquake in Gujarat, Catholic Relief Services (CRS) responded by releasing food aid from an ongoing feeding program to meet emergency needs. During the first 48 hours, before other relief was available, CRS Title II partners cooked meals for survivors using USAID-donated bulgur wheat and oil.

The acute food shortage during the period immediately following the earthquake was the result of destroyed infrastructure and disrupted local trade flows. CRS and its partner organizations were able to address this situation by becoming key links in the logistics chain for relief supplies. Within a few days local food supplies were procured, and the acute food shortage in the earthquake-affected areas was redressed with surplus stocks shipped into Gujarat from neighboring states such as Punjab

During the floods of 2003, CRS' partners in West Bengal diverted food aid from a development program to roadside cooking stations. As the emergency wore on, CRS obtained funding to purchase high protein commercially produced biscuits from Calcutta. The baker delivered four 10-ton truckloads of biscuits to people marooned on elevated roads following floods. The biscuits were generally considered a children's food and were overwhelmingly consumed by the most vulnerable populations: women and children.

In these instances, local purchase was used to speed the response. An existing development program relying on U.S.-purchased food aid provided an immediate source of food to address the needs of those worst affected by the crisis. The lesson here is that integrating food aid development programs with emergency response capacity is an effective means of providing timely, locally led emergency response. In CRS programs, local partners train their staff to borrow from development programs to meet emergency needs.

Moreover, these examples demonstrate that quick onset emergencies have limited outside food-aid requirements. It is the long-term emergencies such as protracted civil unrest, locust invasions and drought that require large aid flows, which are likely to prove difficult to procure locally or regionally on short notice.

The delays associated with such purchases need not adversely affect the health and nutrition of people trapped in crisis situations. In such long-term scenarios, local governments, donors and relief agencies know well in advance what the food aid needs will be. Because of this advance knowledge, it should be possible to purchase and ship the food regionally or internationally with adequate lead-time to avoid both food insecurity and market disruption.

However, the reality of such slow-onset emergencies is that far too often a lack of attention to the warning signs means relief agencies are forced to deal with chronic emergencies that were allowed to become acute. In these situations, and as a result of the short planning horizon, the amount of food necessary cannot be procured solely from local resources without disrupting local markets.

It is simple economics that the local purchase of thousands of tons of commodities for emergency food aid will drive up the local price.  Higher prices will force people who were not food insecure to either cut their consumption due to the price increase or become recipients of food aid themselves. Just as food aid needs to be metered into economies, local purchases have to be carefully metered out of the economy.

 In the Sahel in 2005, local commodity traders anticipated that relief agencies would need to buy locally available cereals to meet emergency feeding needs in Niger. From May through August, traders bid up the price of cereals in the warehouses. Relief agencies limited local purchases to keep the price from rising further than it already had on speculation. This speculation had the double-edged effect of raising prices for the urban consumer and reducing local purchases by some relief agencies, thereby limiting the commodities immediately available for the truly food insecure. This very real, very significant impact tempers the attractive notion that local purchase is always the best response to food shortages.

There are other potential problems that would detract from the suitability of local food aid purchases. First of all, it is simplistic to equate commodity costs with program costs. Getting local purchase right costs money. In a local purchase program costs will include laboratory fees (necessary to ensure the safety and suitability of the food) and added expenditures associated with conducting transparent tendering, as well as commissions to traders, rents, office equipment and supplies, salaries of administrative staff, in-house commodity traders, distribution teams, travel, transport, distribution, program monitoring, evaluation, audits and reporting.

These are many of the same costs that critics say make tied food aid expensive relative to local purchase. However, these costs are inescapable when carrying out accountable and safe programming. Unless there is an ongoing pipeline of purchases from dependable sources, such as with a development program, the process needs to be recreated on an ad hoc basis—at great expense—each time a program of emergency purchase is undertaken. 

In 1995 CRS purchased several hundred tons of local beans to fill a food aid pipeline break in Burundi. During the initial camp distributions, substantial amounts of rocks were found mixed in with the beans. Subsequently a large sample of bags had to be opened and checked to ensure that the bags were free of rocks and foreign matter. This was an expensive and time-consuming process. Had this been a long-term relationship with suppliers, the supplier would probably not have attempted to profiteer. As it was, in the wake of widespread civil unrest and large scale purchases by relief agencies, some sellers did whatever they could to maximize profits. 

In Sierra Leone a regional purchase of three containers of beans from Mali was delayed in transshipment and arrived three months after the donor-provided rice and oil were distributed. Following 9/11, CRS employed an army of quality checkers in Pakistan to inspect truckloads of relief food because we could not rely on the integrity of suppliers. This cost time and money—a hidden premium on the purchase price.

All of the problems described here can be overcome, but it is overly simplistic to suggest that all food aid should be purchased through local sources in the affected country or region. A carefully executed local purchase can be an effective development tool and a wise use of resources. Poorly done, local purchase programs can destroy markets, waste money and do more harm than good.

Will Lynch has worked for more than 20 years in international relief and development in Europe, Africa and Asia. His food aid programming experience includes Lesotho, Ghana, Sierra Leone, Rwanda, Liberia, Albania, Guatemala, Kosovo, India, Pakistan and Indonesia.

This piece was originally published in Bread for the World Institute's 2006 Hunger Report, Frontline Issues in Nutrition Assistance. Find out more about the publication or order your copy from our online store.