«DIREC TIONS IN DE VELOPMENT Human Development Public Disclosure Authorized The Cash Dividend The Rise of Cash Transfer Programs in Sub-Saharan Africa ...»
Biometric Systems and Smart Cards Biometric systems and smart cards hold promise for Sub-Saharan African CT programs. Biometric systems using fingerprinting can improve registration and identification of beneficiaries in some Sub-Saharan African CT programs. They also add security and checks against fraud in registration and payment processes.
Technology is also being tested to improve distribution of payments in remote, hard-to-access environments or in situations where the beneficiaries’ location may change over time. This flexibility is important for a program such as Kenya’s Hunger Safety Net Programme, where nomadic beneficiaries use smart cards and biometric identification to retrieve benefits from locations of their choice (HSNP n.d.).
Mobile Phones Mobile phone technology has multiple uses in CT programs. The fast-expanding coverage of cell phone systems throughout much of the African continent may allow mobile phone technology to be used in CT programs. Nigeria’s Kano CCT for Girls’ Education will test how mobile phones can be used in a limited-capacity environment to distribute transfers and send messages to beneficiaries (Gerelle 2009). Field workers also use mobile phones to request a class register and record girls’ school attendance for monitoring purposes. When mobile networks are unavailable, data are stored on the phone until reception is available and the information can be transmitted (Mobenzi Researcher 2011a).
Mobile phones may also be used to conduct research. For instance, mobile researcher technology allows smart phones to be used to conduct household surveys. The technology is able to walk enumerators through complex skip and repeat patterns one question at a time, immediately noting data inconsistencies and reducing data entry errors. Collected data can be sent in real time to a central information system, and communication between central offices and enumerators allows problems to be addressed as they occur. No data entry personnel are needed, and survey data can be exported and analyzed immediately.
The controls in the system improve data integrity, because there is a clear record of who has dealt with and altered data. Phones can be used to take pictures of survey respondents and record GPS (global positioning system) locations
Box 3.4 (continued) of households to discourage ghost respondents. This technology allows for a 40 percent reduction in required supervision staff members. Costs of the technology include the phones (less than US$200), data transfer costs, and training and payment of enumerators and supervisors (Mobenzi Researcher 2011b).
Mobile researcher technology has already been used in multiple countries in Sub-Saharan Africa.
Web-Based Management Information Systems Web-based management information systems also may prove useful. This technology lets officials access a single system from decentralized locations, thus allowing for more up-to-date recordkeeping and facilitating communication and the transmission of data among program offices. In this way, central offices can stay better apprised of field-level implementation.
The Need for Appropriate Infrastructure, Training, and Communication Key issues affecting the willingness to adopt new technologies are ease of use and comfort with the modality. Adoption of new technologies depends on welldesigned training that addresses all obstacles to proper use by the specific beneficiary group. When users refuse to adopt, a wise approach is to look into what those who introduced the technology did (or failed to do) in communicating it.
Program officials using sophisticated technology need to have reliable access to technical assistance and support for problem solving when inevitable glitches and problems arise. Access to support is especially important when officials have little experience with the technology and are working in a low-capacity environment. The need for support was noted for Ethiopia’s PSNP (World Bank 2010a).
Without the necessary technical assistance, technology can become more of a hindrance than an enabling force.
In considering innovative approaches, program managers must also keep in mind that a technology that is appropriate in some settings may prove inappropriate in others. A small CT in Malawi known as Dowa Emergency Cash Transfers encountered many difficulties while trying to use a distribution mechanism very similar to the one being used in Kenya’s HSNP. The difficulties stemmed from poor coverage of the appropriate cell phone network (Global System for Mobile Communications, or GSM) and from the lack of compatible sites to retrieve cash. Rather than improving accessibility, the technology actually made it harder for beneficiaries to retrieve transfers (Langhan, Mackay, and Kilfoil 2008).
112 The Cash Dividend 10 percent of reviewed programs. Other programs, such as Mali’s Bourse Maman, provide regular transfers through only a portion of the year (in Bourse Maman’s case, the school year). Still others, such as Eritrea’s RBF, award transfers only after households have fulfilled certain requirements. (RBF transfers are given to women after they complete a certain number of medical checkups.) Figure 3.8 shows how often the reviewed CTs are distributed.
Several factors must be weighed when deciding the frequency of transfer distribution, including the cost and timing of distribution and the ability of households to incorporate transfers of a given frequency into their income stream. In particular, when a program intends to help rural agricultural households, CTs should be distributed at appropriate times in the production cycle (Sabates-Wheeler, Devereux, and Guenther 2009). For instance, cash should not arrive at a time when labor demands on household members discourage them from retrieving transfers, and it should arrive in time, or frequently enough, to allow households to use transfers to purchase time-sensitive agricultural inputs.
Just as appropriately timing the transfer schedule is important, transfers must be distributed as planned. Some Sub-Saharan African programs Figure 3.8 Transfer Frequency
Source: Authors’ representation.
Note: Sample size is 61. Data are limited to programs that provide more than one transfer and for which specific information on frequency of transfers was available. Some programs give benefits monthly, but the benefits are not given for the entire year (for example, they are given during the school year only). In that case, they are classified as monthly, because they are given every month during the time when transfers are distributed.
Design and Implementation of Cash Transfers in Sub-Saharan Africa 113 mentioned that transfer delivery was unreliable. This unreliability can undercut the stability that households need to incorporate the additional cash as a regular part of their income. An evaluation of Ethiopia’s PSNP found that program impacts were significantly dampened in households that received either low or irregular transfers (Gilligan and others 2009a).
Although timeliness of distributions has gradually improved, only a little over one-quarter of PSNP households said that they were able to plan for their transfers. Improving this indicator has been a key goal of the program. It has required significant capacity building, monitoring, and continuous adjustments and improvements to help improve on-time transfer delivery (World Bank 2010a).
Transfer Recipients Many CT programs (and especially CCT programs) throughout the world have provided benefits primarily to female beneficiaries; however, the reviewed CT programs in Sub-Saharan Africa showed less preference for distributing cash exclusively to women. Almost half of CCT programs and 9 in 10 UCTs distributed cash to either male or female beneficiaries (see figure 3.9). Some of this distinction is due to the individual nature of many transfer programs, such as social pensions, in the region. Other transfers do not specifically direct cash to women, although in practice,
Source: Authors’ representation.
Note: Figure shows the intended beneficiary (payee) of cash transfers. Sample size is 15 for CCTs and 38 for UCTs.
Results are limited to those programs that clearly explain which household member should receive cash.
114 The Cash Dividend women are the primary recipient in most households. Despite this inclination, slightly over 20 percent of all programs indicated that transfers should be given to women whenever possible. This tendency is markedly greater in CCTs than in UCTs.
Burkina Faso’s Pilot CCT-CT is studying which household member should be the transfer recipient. Some transfers in the experimental evaluation are being given to male adults, and others are being given to female adults in the household. The results of the evaluation will inform this feature in other CTs in Sub-Saharan Africa. Programs in Morocco and the Republic of Yemen, not covered in this review, are also testing how program outcomes change depending on whether mothers or fathers receive transfers. Results from these programs may be especially helpful in African countries that share relatively similar cultures.
More recently, CT programs have begun distributing some benefits to adolescent beneficiaries, a practice that has been used in educationfocused CCT programs in Bangladesh and Colombia (Fiszbein and Schady 2009), among others. This program feature is being tested experimentally in Malawi’s Zomba CT to assess any variations in impact based on the proportion of transfers given to parent versus adolescent beneficiaries. After two years of program implementation, evaluation results showed that increasing the share of the transfer given to adolescent girls did not affect the schooling, marriage, or fertility outcomes examined (Baird, McIntosh, and Özler 2010).
Areas for Further Analysis in Cash Payment Systems Sub-Saharan African countries are currently investigating several issues important to the design of CTs, including differences in outcome based on which household member receives the cash. An important factor to keep in mind is that the intrahousehold dynamics in Sub-Saharan Africa are different from those in other regions. The dynamics may vary greatly even within countries, so decisions about the transfer recipient need careful attention and evaluation. Both quantitative and qualitative methods should be used to gather information about what arrangements are most conducive to the desired program outcomes. Care should be taken to understand how receipt of the CT, especially by women, affects intrahousehold relationships. Transferring cash to women rather than to men may improve household welfare and give women greater household bargaining power. Conversely, it could also be counteracted in the household. These questions are currently being tested within Sub-Saharan Africa, and additional analysis in this area will be helpful.
Design and Implementation of Cash Transfers in Sub-Saharan Africa 115 Many programs in Sub-Saharan Africa are also experimenting with novel means of delivering cash to beneficiaries, and these experiences will be helpful to others seeking to implement programs, particularly in remote rural settings. Although the optimal frequency of cash distribution merits further analysis, much of this decision rests on pragmatic questions of human resource availability and program capacity to process payments and deliver cash. Therefore, case studies and other information related to improving on-time delivery of transfers may be more useful.
Conditional versus Unconditional Cash Transfers Conditions have the potential to enhance transfer-related outcomes, but context matters. The decision to condition transfers depends on multiple factors, and conditions may or may not be cost-effective in various countries in Sub-Saharan Africa. Even if conditions are important for program impacts in countries outside the region, this result may not translate similarly to Sub-Saharan Africa. For example, returns to certain education levels may be lower in certain Sub-Saharan African countries than they are in other parts of the world, which could affect how beneficiaries use CTs with and without education-related conditions attached. The true value of applying conditions in CTs within the region needs to be, and currently is being, tested. If the benefit of enforcing conditions outweighs the costs, a more appropriate approach for most programs is to focus on conditional CTs, unless they have another reason for not imposing conditions.
Most studies have been inconclusive on the exact effect of conditions.