GuatTin Ng

Publication Date



This paper is an exploratory comparison of the costs of Individual Development Accounts (IDAs) relative to financial-capital programs and human-capital programs. Comparing program costs is informative but fraught with difficulties. a recurring theme is that due to a host of factors, costs vary not only between programs but also between sites in a single program. Notwithstanding the challenges involved, cost studies are important because resources are scarce and decisions must be made about resource allocation. Individual Development Accounts (IDAs) are matched savings accounts designed to assist the poor accumulate assets. Savings are matched for three purposes: to increase returns on savings, to provide an incentive to save or to save more, and to accumulate greater assets (Schreiner et al., 2001). IDA programs have a social cost because the resources used for matches or for administrative expenses could have been used for something else. As with other programs, the question is not so much how large costs are but whether the program is worth the cost. There are different approaches to this question. One approach is cost-benefit analysis, which requires estimation of both program costs and benefits accruing to participants and nonparticipants. The cost-benefit approach checks whether the present value of benefits is greater than the present value of costs. For IDAs, costs have been measured but benefits have not. Another approach to the usefulness of a program is to measure program impact, whether the benefits for or changes in program participants can be attributed to the program by comparing the changes between those who participated in a program and those who did not. The focus is on the differences between program and non-program participants. a third approach is to measure program outcomes; assessing benefits or effects for program participants only without a comparison group. Findings on savings outcome indicated that IDA participants made an average monthly net deposit of $25, and with an average match rate of 2:1 the average participant would save $75 per month and accumulate $900 per year (Schreiner et al. 2001). Outcome measurement is easier than program impact analysis but it is not possible to attribute the changes in participant behavior or attributes to program participation. Changes may have occurred even without participating in the program. Outputs, which are physical units of service, are even easier to measure than outcomes. However, programs may generate a lot of activities and not produce the intended outcomes. This paper compares costs per unit of output for IDA programs with costs per unit of output for financial-capital and human-capital development programs. IDAs are estimated to cost $1,495 per participant-year or $125 per participant- month (Schreiner, 2000). Estimates from the American Dream Demonstration (ADD) suggest lower average costs of $840 per participantyear or $70 per participant- month (Schreiner et al. 2001). Are these costs high or low? Without a baseline or benchmark the cost figures mean little. The primary purpose of this paper then is to compare cost per unit of output of IDAs with programs that are close enough in intent or operations to get a rough sense of where IDA programs stand in the plethora of capitaldevelopment programs. IDA programs are fairly new and even if benefits are unknown, knowledge of costs – and of the implications of costs—is still useful. Program costs matter because resources are scarce. Given budget constraints policy makers should have a keen interest in program costs. This paper presents cost/output figures. It does not, however, make judgments based on those figures about relative program worth. Because outputs are not exactly the same across programs, the subjective judgment of relative worth is left to the reader. IDA programs currently operate on a small scale even though they were first conceptualized as universal, permanent savings accounts. a secondary purpose of this paper is to review how costs are likely to change when pilot programs scale up. The changes that come with mass implementation matter because they may affect the relative costs and benefits of IDA programs. This paper is organized as follows. Section 1 explains IDA programs and cost measurement for IDAs. The next two sections describe programs selected for comparison with IDAs. The programs can be broadly categorized as the development of human capital (section 2) and the development of financial capital (section 3). Section 4 compares the costs of the various programs. Section 5 identifies some lessons from large-scale programs for the scale-up of IDA programs, and section 6 provides concluding remarks.

Document Type

Working Paper


Financial Inclusion


Asset Building

Original Citation

Ng, G. T. (2001). Costs of IDAs and other capital-development programs (CSD Working Paper No. 01-8). St. Louis, MO: Washington University, Center for Social Development.


American Dream Policy Demonstration (ADD)


IDA, individual development account, ADD, American Dream Demonstration