An analysis of cost effectiveness has to consider not only the literacy and numeracy outcomes but also the broader outcomes of the programme on people's quality of life. It has not been realistic for ACTIONAID, in such a short time, to undertake detailed cost benefit analysis which incorporates all these wider impacts on development. The section on empowerment is a guide to the kind of changes which might take place in the long term and which would have to be woven into a statistical cost benefit analysis.

In the context of Bangladesh, we have seen some impact in the economic sphere (particularly in relation to improved planning and better use of loans), in the health sphere (an increase in awareness and practical knowledge as well as concrete actions eg latrines/ tube well) and in relation to children's education (indications of more regular attendance and more support at home). Women's status has generally been improved. This impact is not just the result of the "product" of the literacy circles but the result of the "process" (participation and active discussion of local issues within the REFLECT Centres).

All the above factors need to be considered when looking at the cost of the REFLECT programme in relation to other approaches to adult literacy. If the REFLECT process produces benefits which other literacy programmes cannot then we should not simply compare cost with cost.

Costs per year for one REFLECT centre with 15 learners (in Taka)

7, 800 Facilitator's honorarium
1, 200 Facilitator's training allowance
350 Facilitator's bag/umbrella
375 Facilitator's Manual and visual cards
50 Facilitator's notebook
250 Large manila sheets (50)
640 Marker pens (20)
375 Learner's notebooks
500 Blackboard and chalk
160 Pens and pencils
100 Erasers and sharpeners
11, 800 = 786 taka per learner = approx £12 ($18) per learner.

The Unit cost of the Bhola REFLECT programme per learner can thus be said to be 786 taka (approx £12) over a year.

To put this in context, a 1992 study by CAMPE (M.H. Rahman and S.D. Khan, quoted in NFE Project Final Report, CEC/ADB 1994) estimated that costs per learner of 530 taka were normal. This estimate appears however to have been for just a six month period (the REFLECT costs for six months would be 486 taka per learner) and it is not clear whether it includes facilitators honorarium or training. Given these factors and inflation since 1992, the cost of 786 taka appears very reasonable.

It should be noted that the costs of the Bhola programme are by no means indicative of costs which all REFLECT programmes involve. The payment of facilitators is by far the largest component of the costs and this is not something automatically required by the REFLECT method. The pilot REFLECT programme in El Salvador worked with volunteer facilitators and in the next phase of REFLECT in Bangladesh at least one of the projects will work with volunteers.

Another factor to be considered is the economy of scale. The costs of producing the manual and visual cards would be much reduced if done for a larger programme and bulk purchase of other items would also reduce costs.

It should be noted that the costs calculated above are based on 15 learners per centre. The size of the circles in Bhola was based on the pre-existing shomitis (of 15 women). In future Shomitis will be larger (minimum 30 members) so this will change. In Uganda the REFLECT circles tend to have 25 or 30 learners so this should present no problem. Large circles would significantly reduce the cost per learner (eg if there were 25 learners per centre the average cost would be 472 taka or £7.50 per learner).

The average cost of the existing Bhola programme of £12 per learner per year becomes a particularly good investment when one considers the effectiveness of the REFLECT method. Compared to the World Bank estimate that on average only 25% of learners successfully complete adult literacy courses about 60% of learners successfully completed the REFLECT programme. The evaluators also suggest that there is a high likelihood that they will retain their skills (which the World Bank estimates only happens to 50% of successful learners).

There is no doubt that if the goal of a literacy programme is just to teach learners to sign their names then REFLECT is not the cheapest option. But if the goal is sustained literacy and an impact on wider development then REFLECT is a low cost approach.

Finally, it is worth making a brief comparison to children's education. The Unit cost of primary schooling in Bangladesh is $10 about a year - but with a five year cycle ($50) and only a 24% success rate (ie about $200 per successful learner). Whilst primary schooling has many other functions than teaching literacy, the same can also be said for the REFLECT process which, in comparison, comes very cheap.