A/HRC/38/33 social protection from the standpoint of fiscal policy”, 97 as illustrated by the cases of subsidy reform and social spending floors. Subsidy reform 43. One third of IMF arrangements approved during the period 2006–2015 contained structural conditionality pertaining to energy price subsidies.98 Reforming such subsidies makes perfect economic sense: they depress growth, are regressive, discourage investment, crowd-out social spending, diminish competitiveness, incentivize smuggling, negatively impact the balance of payments, encourage overconsumption, and contribute to pollution that exacerbates climate change and negatively impacts individual health. 99 In practice, however, reform efforts must take account of political economy considerations, as shown by recent experience in many countries, including the Islamic Republic of Iran 100 and Tunisia.101 Acknowledging such tensions, IMF has generally advised that subsidy reform should be accompanied by the reallocation of part of the savings to “well-targeted”, “welldesigned” and “pro-poor” social protection programmes.102 44. However, experience demonstrates that these seemingly neat and rational prescriptions beg more questions than they answer. First, history shows that social protection systems have gained traction where they were not politically divisive. As one author observes about Western Europe, “although the greatest immediate advantage was felt by the poor, the real long-term beneficiaries were the professional and commercial middle-class”.103 Similarly, it has been noted that, while “welfare States are sometimes seen as an achievement of the industrial working classes … they mostly came about through broad coalitions representing farmers, middle classes and professionals as well. Furthermore, the most extensive, most durable welfare States are generally those that serve the interests of the broadest and most powerful coalitions.” 104 The rapid elimination of regressive but universal price subsidies to be replaced by limited social protection programmes for the poorest will garner very little political support from the key political players, thus undermining the political (rather than fiscal) sustainability of such reforms. Unsurprisingly, ILO recorded 322 “social protection reforms” worldwide in 2017, only 5 per cent of which had been implemented to address the needs of the poor or extreme poor.105 45. Second, while price subsidies are often deeply rooted in a country’s political and institutional structure, the “pro-poor” social protection programmes that replace them (such as cash transfer programmes) are often entirely or largely new, and thus easily subject to rollback, leaving many of the poor, and non-poor, uncovered.106 While IMF, in its role as emergency lender, may have significant leverage over a Government to undertake subsidy reform in times of crisis, there is a tension between the short-term focus of IMF and the significant time it takes to invest in and build up a proper social protection system, a problem that is unlikely to be solved only by devoting more resources to addressing social protection issues in Article IV surveillance. 97 98 99 100 101 102 103 104 105 106 14 IMF, The IMF and Social Protection, p. 11. Ibid, p. 67. See Clements et al, Energy Subsidy Reform: Lessons and Implications (Washington, D.C., 2013), ch. 3. See Development Pathways, “Role of IMF-backed elimination of universal social protection in protests”, available at www.developmentpathways.co.uk/resources/role-imf-backed-eliminationuniversal-social-protection-protests/. U. Laessing and T. Amara “Tunisia seen softening economic reforms to avoid unrest”, Reuters (4 December 2017). Clements et al, Energy Subsidy Reform, pp. 20–21; IMF, The IMF and Social Protection, p. 25; and IMF, “Social Safeguards”, p. 22. Judt, Postwar, p. 76. David Garland, The Welfare State: A Very Short Introduction (2016), p. 41. ILO, Social Protection Measures throughout the World: January to December 2017 (2018). See reports of the Special Rapporteur on his mission to Saudi Arabia (A/HRC/35/26/Add.3, pp. 7–10) and on his mission to Mauritania (A/HRC/35/26/Add.1, pp. 14–16).