Conference organisers can be congratulated for bringing together some of the most influential and up-and-coming thinkers and practitioners in social protection for last week’s International Conference on Universal Child Grants in Geneva. Top minds from international organisations and entities exchanged ideas with key national policymakers, practitioners, consultants and members of academia –all of whom were clearly united around a vision of a world in which every child has a right to, and receives, a child benefit.
Or, were they?
The targeting-universalism debate dominated the agenda
It turns out that agreeing on a flowery future by way of a vaguely defined pathway of ‘progressive universalism’ (a term coined in the 2019 World Development Report and a blog by the World Bank) is relatively easy (see our previous thought-provoking blog on this term), but when it comes to sorting out how to get there, or even where or what ‘there’ is, the consensus all but disappears. Indeed, the targeting-universalism debate loomed large during the 3-day event, and, it’s fair to say, remains heated. Just about every plenary devoted time to addressing it. One organiser’s appeal that “Surely we can all agree to… move toward reaching agreement on…” –or something to that effect—was an admirable attempt at feigned unity that, unsurprisingly, met with good-natured chuckles from an audience who knew better.
By now, the main arguments from the two camps should be familiar to most of our readers, and it makes little sense to re-hash them here. (For those who are only just beginning to follow these debates, a good starting point is here). Suffice it to say that our own Stephen Kidd was unswerving in underlining the mounting evidence that targeting is inaccurate, ineffective, costly and potentially causes more harm than good, whereas universal designs are not only more effective, but are administratively simpler, more popular and build stronger societies. You can find our previous study here, but new evidence will be shared at an event soon!
Representatives from the World Bank largely acknowledged many of these points (see here where they acknowledged exclusion errors; a presentation soon to be released where they acknowledged that it is costly; and here where they argued that “not all” those who receive targeted transfers are stigmatised –meaning, most are?).
Alas, none of that was enough to convince them that UCBs are anything other than a distant utopian dream for low- and middle-income countries, since resources are scarce everywhere, and UCBs are expensive. This is why, according to Margaret Grosh, very few countries have implemented them. (By which we presume she meant very few low- and middle-income countries, since most European countries have either universal or ‘quasi-universal’ child benefits, and it’s fair to say that they, too, were implemented under budget constraints.) The corollary is, of course, that those scarce resources should be concentrated among children and families most in need to maximise the impact of the benefits. (This assumes that transfer values are higher in targeted programmes, which is not borne out by the evidence.)
The cart before the horse?
The debate over targeting was perhaps inevitable given the conference’s theme. Throughout the event, one could not escape the feeling that, in presupposing UCG/UCBs as the solution, we had skipped over the more critical phase of defining the problem: What is the overarching objective of a child benefit, after all? Is it really to reduce poverty? If it is, then we should enter the debate over targeting (in which case, by the way, universalism still wins, as shown here and here).
But, if the objective of a child benefit is something else —for example, to equalise costs between those caring for children and those who are not, to invest in a common future, or to unite a nation, as Olli Kangas reminded us is what drove the debate in post-war Finland—then all roads would seem to lead to universally designed benefits.
Compensation of parents or caregivers for the extra cost of raising a child, in the name of the common good, is a fundamentally different policy agenda than the one that gave rise to conditional cash transfers or to their unconditional poverty targeted cousins. In fact, I would argue that this is the single biggest line of defence against the ‘limited resources beget poverty targeting’ thesis and, until we confront it, we will continue to be sucked down the targeting debate vortex.
Toward universal social protection architecture
This sort of spirited debate is always welcome, especially when it moves us toward better policy solutions, and there were certainly hints in that direction last week. One way forward is to think about how we go about building universal social protection architecture. What do we mean? A system that stops imagining that a single scheme (however it is designed) will solve the social protection coverage conundrum, stops building separate small schemes for each ‘marginalised’ or ‘vulnerable’ group, stops ignoring social insurance — and instead treats all citizens as equally deserving of robust and reliable social protection across the lifecycle, through one single social security system, building on and strengthening core schemes.
There were promising ideas put forward last week that pushed us in that direction. In particular, the idea of a progressive expansion of the age eligibility for a universally designed (tax-financed) child benefit —where eligibility starts at lower ages, for example 0-3, and then gradually rises until all children are covered, as is underway in South Africa and Thailand— was gaining popularity by the end of the conference, including from people on different sides of the targeting debate. Stephen Kidd presented an illustrative example using Kenya data, in which all children aged 0-4 years are given a benefit but not taken off until they reach their 18th birthday. As Figure 1 shows, the programme could start at a cost of 0.6 per cent of GDP and, by 2030, reach only 1.1 per cent of GDP, more or less in line with what Argentina, South Africa and Mongolia are currently investing. And, as can be seen, the impacts would be very significant. Bjorn Gelders presented microsimulations of universal child benefits for Indonesia, showing that they would have significant effects on households’ purchasing power and macro-economic poverty.
In addition, although the focus has been almost ‘universally’ on tax-financed benefits, as my own presentation argued, there is potential in many countries for building multi-tiered child benefit systems (see Figure 2) that achieve universal coverage of child benefits through a combination of social insurance and tax-financed instruments. (This is the so-called ‘coordinated mixed’ approach in the words of the conference background paper.) Such an approach has the potential to build on, and if well designed, even grow the existing social insurance system while simultaneously offering a simple and fairer alternative to poverty targeting (through benefits-testing) and guaranteeing an adequate benefit for all children (we will share more on that in the coming weeks…!). And, of course, there’s the potential use of an affluence test – as employed in South Africa – which was highlighted by Yulia Oleinik of UNICEF and Stephen Kidd in their presentation on Uzbekistan.
And finally, we can all celebrate the fact that the question of whether all children deserve social protection did not even figure into the discussion. As Alexandra Yuster from UNICEF pointed out, not investing in children, in particular young children, is not a good economic choice. On that, we all agree, which is an achievement in itself.