Food Assistance Programs In Bangladesh

7.Concluding Observations:

68. 
    Continuing operation of such a large system of food transfers represents a big challenge for the Government of Bangladesh, particularly since the volume of international food aid received by the country has been declining in recent years. Projected total food aid to Bangladesh in 2000-01 (approximately 625,000 metric tons) has been amongst the lowest in 20 years. When resources are limited, it is all the more important to ensure that they are used to maximum effect—that they reach the people who need them most, and in a form that does the most good.
    

69. 
    The food assistance programs have, by and large, done a good job of reaching the poor. More significantly, a number of recent studies suggest that some of these programs have also had a considerable developmental impact on the lives of their participants. This is of vital importance as, large as this might be, the aggregate resource transfer under these programs represents only a small share of total per capita consumption of beneficiaries, even for those in the bottom quintile of the income distribution. The gradual shift in emphasis from relief provision to more explicit development objectives during the nineties represents one of the most important advances in the evolution of these programs, and needs to be sustained to ensure that the program resources have the greatest impact on the long-term welfare of their intended beneficiaries.
    

70. 
    However, any definitive conclusions about the “pro-poor” nature of these programs must take into account the fact that even though visible receipts of transfers are significantly pro-poor, a non-negligible share of resources allocated to these programs cannot be accounted for when using survey data to estimate total resource transfer to beneficiaries. Strengthening monitoring systems to reduce leakage from the system is likely to yield high dividends, as it would enable the Government to release additional resources to increase coverage. In addition, better monitoring systems could allow the government to compare performance across programs as well as across different regions. Program effectiveness could then be improved considerably by taking this information into account more explicitly when making its allocation decisions, for instance through rewarding good performance or penalizing misappropriation.
    

71. 
    That the needs of the poor extend well beyond food transfers alone is now increasingly being recognized in Bangladesh, both in the context of design of effective relief operations as well as in the case of more long-term development oriented programs. For instance, the European Commission is planning to look closely at the RMP program as a model for making cash transfers as it shifts away from food towards cash aid in the next 2000-02 food security cycle. There are a number of reasons why engaging on the cash vs. ink-kind transfers debate merits serious consideration: In a study of several targeted safety net programs, Ahmed et al. (1994) found that the RMP, the only cash-based transfer program, was the most cost-effective, requiring 1.2 Tk to transfer 1 Tk of benefits to a recipient household.²³ This is in part because being a cash-for-work program, it does not entail costs of handling food grains. Leakage is also believed to be very low because of the unique wage payment system. Salaries for each RMA are paid by a bank transfer every two weeks directly into a maintenance crew bank account that is operated by the elected representative of the workers. Potential for leakage is minimized since certification by officials is not required to withdraw payments.
    

72. 
    Second, in practice a number of food-based transfer programs operate in cash anyway, and have been subject to malpractice. In the case of the Food for Work program, some PICs hire labor contractors who in turn often sell the wheat or rice received so they can pay cash wages, and workers themselves sometimes end up selling part of the grain they receive to buy other goods and services (del Ninno and Dorosh, 2000). In some places, informal deductions are made from wages cover administrative costs and payments to members of the PIC for their time and effort. As a recent study of the FFE has reported (Akhter and del Ninno, 2001), the dealer-based system of food grain distribution is far from satisfactory: households participating in the program experience losses in their food grain entitlement due to dealer malpractice, as well as significant transaction costs and inconvenience in collecting their rations. Switching to a cash-based incentive system would eliminate the reliance on dealers as intermediaries. Similarly, in the case of the FFW, particularly if there are transaction costs to secondary sales of food grains by contractors and beneficiaries, there appears to be a strong case that the program should pay workers at least partly in cash rather than food.
    

73. 
    Third, if the primary objective of the FFE program is to provide an incentive to children from poor households to attend school, there appears to be little reason why an equivalent cash transfer would not provide the same inducement.²⁴ Similar arguments also hold for other programs. Cash-based transfers could also be used as incentives for destitute women in the case of the VGD, or to provide relief to VGF or GR beneficiaries. Indeed, one of the main findings of studies carried out during the 1998 flood to assess the impact of government and NGO interventions was that the needs of beneficiary households extended far beyond the need for emergency food assistance alone (e.g. del Ninno and Roy, 1999).
    

74. 
    The debate on cash vs. food transfers is nicely encapsulated by Dorosh & Haggblade (1997):
    

“The case for in-kind distribution boils down to three notions: commodities may not be physically available, as in some droughts, even for households with purchasing power; in-kind rations can most effectively improve food consumption and nutritional welfare of the poor beneficiaries; and through the resulting demand creation, this increased consumption will tend to off-set potential price depressing effects of food aid supplies. Monetization, coupled with cash transfers to beneficiaries, offers a number of compensating benefits: improved welfare for recipients, who themselves identify their most pressing needs; reduced commodity handling costs, which increase funds available for development or expanded poverty-alleviation objectives; and reduced risk of agricultural disincentives due to control over the location and timing of commodity injections.”

75. 
    First, examining more closely the extent to which food transfers to households actually improve nutrition would help shed more light on the merits of the argument cited frequently in favor of continuing food transfers. Data from the Child Nutrition Survey (CNS), which was carried out in conjunction with the 2000 HIES, could be used to analyze if, and to what extent, children in beneficiary households have better nutritional outcomes compared to those in similar non-participant households.
    

76. 
    Second, there is a need to update estimates of transfer efficiency—i.e. the cost of transferring one taka of benefits to a recipient household—across different programs. Earlier estimates of cost-effectiveness of the various programs computed by the Working Group on Targeted Food Interventions are widely cited in the literature on safety nets in Bangladesh, indicating the considerable importance and interest in the subject.²⁵ However, these estimates date back to the mid-nineties. Besides allowing a comparison of food versus cash assisted programs, cost-effectiveness calculations will also help in assessing the fiscal viability of scaling-up any of the programs currently in operation.
    

77. 
    Third, investigating further the extent to which there may be transactions costs to secondary sales of food grains by contractors and program participants would help quantify part of the efficiency gains from switching to more cash-based transfers. In this context, soliciting views from local program administrators as well as the beneficiaries themselves (both men as well as women within the households) on the relative merits of cash vs. in-kind transfers would also help bring in a critical perspective that is currently missing in the ongoing debate.