A/HRC/38/33 children, which has since been raised again to 80 per cent.89 That structural benchmark in the IMF loan has reportedly been strictly enforced. 90 Similarly, in January 2018, IMF announced that Kyrgyzstan would “amend the law on universal child allowances to reintroduce targeting”.91 That policy change is said to have been adopted following the IMF review of the country’s Extended Credit Facility.92 40. Nonetheless, recent research on targeting in sub-Saharan Africa presents a compelling challenge to the logic of targeting, assuming the objective is to reduce poverty, rather than to promote neoliberal orthodoxy. 93 The authors of that research explore key factors affecting the design and implementation of anti-poverty policies. The first is information, for which they demonstrate that the proxy means tests — consistently promoted for that purpose by IMF and others — yield highly unsatisfactory results in practice. They conclude that the test “misses many of the poorest households in almost all countries: on average, 80 per cent of poor households are counted as non-poor by the test, and 40 per cent of non-poor households are counted as poor”. The second factor is incentive effects, where they argue that targeted policies encourage non-poor people to change their behaviour in order to qualify for benefits intended for the poor. And the third factor is political economy, which, as discussed below, suggests that policies targeted only at the poorest will not receive sufficient political support from the elites and middle classes who dominate political decision-making. The conclusion to be drawn is that anti-poverty programmes with broader coverage, rather than finely targeted ones, are much more likely to reduce poverty. 41. A recent analysis published in the IMF Fiscal Monitor goes tentatively in the same direction by acknowledging that targeting requires adequate administrative capacity and that, where such capacity is lacking, there is often “undercoverage of the poor and leakage of benefits to the rich”.94 As a result, the analysis cautiously explores the option of universal basic income schemes as a “second-best” alternative to targeted social protection. Consistent with that approach, a recent IMF Article IV report on Mozambique encouraged the continuation of “efforts to improve coverage and generosity” of the non-contributory social protection system and noted the Government’s shift from a targeted, charity focus to a rights-based universal benefit approach. 95 By IMF standards, such an approach seems wholly anomalous. Social protection practice 42. The idea of social protection as “a certain minimum of justice, or fairness”,96 is in fact relevant to all of the core work of IMF. One Executive Director explained it well when he told the Special Rapporteur that the work of IMF in any of its five main pillars — monetary, fiscal, financial, balance of payments or structural policies — could either promote or disrupt social protection. Nevertheless, IMF “has traditionally approached 89 90 91 92 93 94 95 96 IMF, “Mongolia: First and Second Reviews under the Extended Fund Facility” (December, 2017), p. 10. See S. Kidd, “Mongolia and Kyrgyzstan lose out in their struggle with the IMF over the targeting of child benefits”, Development Pathways, 21 February 2018. Available at www.developmentpathways.co.uk/resources/mongolia-kyrgyzsg-child-benefits/. For a response by the IMF, see www.developmentpathways.co.uk/resources/mongolia-kyrgyzsg-childbenefits/#comment-1358. See IMF, “IMF Staff Concludes Visit to the Kyrgyz Republic”, press release (31 January 2018), available at www.imf.org/en/News/Articles/2018/01/31/pr1834-imf-staff-concludes-visit-to-thekyrgyz-republic. Kidd, “Mongolia and Kyrgyzstan lose out”. The research is summarized in C. Brown, M. Ravallion and D. Van de Walle, “Reaching Poor People”, in Finance & Development (December 2017), pp. 48–51. IMF, “Tackling Inequality”, p. 16. IMF, “Republic of Mozambique”, p. 12. T. Judt, Postwar: A History of Europe since 1945 (Penguin Books, 2005), p. 76. 13