In “The World Bank’s New Social Protection Model: Conspirational Cash Transfers”, Sissy Teese accuses us of being part of a conspiracy to promote CCT (as opposed to unconditional transfers, UCT) and manipulating evidence for that purpose. The reality is that it is her note that follows a conspirational approach whereby she distorts and manipulates evidence to, we can only assume, pursue an ideological agenda (promote UCT). The reality is that none of us believe that CCT are necessarily superior to UCT: as is often the case with public policies, the choice of instrument is dependent on the goals being pursued and the conditions under which those policies are implemented. What we firmly believe in is the importance of basing those choices on solid evidence and, consequently, consistently seek to generate and disseminate such evidence.

In Conditional Cash Transfers: Reducing Present and Future Poverty, www.worldbank.org/ccts-report (of which two of us were the leading authors) we devote significant efforts to address the question of conditions. In Chapter 2 we review the theoretical arguments as well as the (limited) empirical evidence on the subject. In Chapter 6 we approach the question from a policy perspective and explicitly state the potential trade-offs between conditioned transfers (higher impacts on service utilization, though not necessarily on final impacts; maybe better political economy) and unconditioned transfers (likely better for the purpose of redistribution). We encourage those interested in the subject to read the book and judge by themselves the quality of our analysis. It is worth mentioning that the book (a product of the World Bank’s Research Department and not an official policy document) was thoroughly reviewed by an external panel of academic experts of the highest reputation in the field.

Sissy Teese also questions in an unwarranted way research on Ecuador by Schady and Araujo and on Malawi by Baird, Chirwa, McIntosh and Özler. Her main critique of the Ecuador paper is that the 11/05 version the abstract is explicit about the fact that the program effect on the conditioned households could be related to baseline socioeconomic differences between the conditioned and the unconditioned while in the abstract of the 6/06 Working Paper version there is no mention of this caveat. This is true. However, the caveat is not absent from the paper. Indeed, in pages 18-19 of the published Working Paper, there is a discussion of these baseline differences. One whole paragraph is devoted to discussing and describing the differences and their implications. More importantly, the second version of the paper introduces standard econometric techniques to explore further whether observable differences between conditioned and unconditioned households explain the differences in impact (they don’t). This is the version that was published in the journal Economia.

Sissy Teese also refers to a different study on the Ecuador program that shows no impacts on enrollment for either conditioned or unconditioned households. We are aware of that paper (which is quoted in the book). Moreover, one of us was a co-author of the revised version of that paper, which is also available as a World Bank Working Paper! The interpretation we had of that finding, both in the paper and in the book, is that it is due to differences in the samples used. Notably, the sample in this second paper was substantially less poor than the sample used in the Schady-Araujo paper. Given that enrollment rates among these less poor households were higher to begin with, and given that the transfer represented a much smaller fraction of total expenditures for them, it is perhaps not surprising that there was no significant impact of the BDH on these households.

The Malawi paper deals with the effects of CCT programs on the sexual behavior of adolescent girls, a topic that had not been explored before. The paper takes a policy tool that is common in many parts of the world, i.e. a CCT, and then examines its effect on an outcome previously ignored – even though there is reason to believe that these programs may have an effect on the risky behavior of their young beneficiaries. It does not aim to disentangle the channel through which this impact occurs. All the quotations from the paper that have been selected by the author of the piece in Wahenga.net are entirely consistent with the evidence the authors have: that CCT programs had significant impacts on pregnancy, marriage and onset of sexual activity. The author of the Wahenga piece, says that: "...to claim that this has anything to do with conditionality per se is highly disingenuous." However, nowhere in the paper one can find a sentence that suggests that the impacts are attributable to the condition or that unconditional cash transfers would not have had the same impact.

A working paper that actually DOES disentangle the effect of the cash from the conditionality is Working Paper 5090. It is nearly impossible for anyone to search for material on the Malawi project on the Policy Research Working Paper Series website and not come across WP 5090 by three of the same authors, which unequivocally states in its abstract (referring to schooling outcomes) that "These impacts were, on average, similar in the conditional and unconditional treatment arms". The concluding section in that same working paper states that "We cannot reject the hypothesis that, among baseline schoolgirls, the price (or substitution) effect is zero, even though we find relatively large income effects." Incidentally, a story based on this working paper appeared recently on the World Bank’s website [link].

So, it seems that by quoting only the paper that seemingly suits her purpose and selectively leaving out another (citable) paper that would essentially destroy her argument, it is Ms. Teese who is being disingenuous, and not the authors of the study. It should be remembered that equal co-authors of the two working papers on the Malawi study are three respected faculty members at George Washington University, University of Malawi, and UC San Diego, who are also being accused of being dishonest researchers.

We believe in the importance of debate and recognize that, in many cases, reasonable researchers can disagree on the interpretation of results. This, however, is not what transpires in Sissy Teese’s piece. We sincerely hope that readers of the Wahenga website genuinely interested in an evidence-based approach to social protection policies will not get distracted by such attacks and will remain focused on the growing evidence that effective social protection through cash transfers (conditional and unconditional) can indeed work in the developing world.

Ariel Fiszbein, Berk Özler and Norbert Schady