The Rise of Social Protection, the art of Paradigm Maintenance, and a disagreement with the World Bank

November 8, 2018

     By Duncan Green     

Spent a mind-stretching day last week with a bunch of social protection experts from the LSE, IMF and assorted other bodies. Social Protection includes emergency relief, permanent mechanisms such as pensions and cash transfers, and ‘social insurance’ based on people’s personal contributions.

LSE boss Minouche Shafik set the scene really well:

‘The failure of safety nets is partly responsible for the rise of populism. Governments failed to cushion the downsides of globalization and integration and haven’t kept up with changing societies and economies. We need to rethink the welfare state.

Lots of developing countries are at their ‘Beveridge Moment’ – discussing what kind of welfare state they want to build, a holistic view to replace piecemeal progress on education, health etc. But the social contract is getting harder to build and maintain in more heterogeneous, globalized societies – Who’s in/out? Who owes whom? What about the obligations to future generations?

Some old questions also need to be revisited: Redistribution needs to be squarely back on the table; tax systems have become less and less progressive over time. Universal benefits have a real political and social logic that economists lost sight of, in their preference for means testing and efficiency. We need to revisit our worship of efficiency. The Us and Them culture thus created disengages the middle classes. And, to paraphrase Titmuss, services targeted to the poor often end up being poor quality services.’

Reach of social assistance by region

Armando Barrientos gave a useful intro to the recent trends, based on the ‘social assistance explorer’ a new (and free) online 109 country database he has helped put together (take a look – it’s amazing). Social Assistance (non-emergency Social Protection) has been rising fast in low and middle income countries. From 2000-15, the number of recipients went up nine-fold. Expansion has been uneven across regions and type of assistance (see figs).

Once we got into the detailed debate on the virtues of different kinds of protection, the arguments were so ‘inside baseball’ that they mainly went above my head/beyond my brain. At these moments, you can always revert to checking your emails, but if you can stop angsting about your inadequacies as a geek, there is usually interesting stuff to think about even in these kinds of discussions.

For example, at a more meta level, events like these can offer some fascinating into the way paradigms shift or stick in international development.

Paradigm shifts are usually unwelcome: I got very excited when the opening panel included speakers on climate change and gender, arguing that prevailing ‘technical’ approaches to social protection need to incorporate both if they are to remain relevant and useful. On Climate Change, the LSE’s Ian Gough went as far as saying ‘Endless growth is not compatible with a single planet. We have to move to post-growth’. Yup. In a seminar with the IMF.

But what happened next was sobering. No-one laughed, called for his excommunication, threw anything or shouted abuse. Instead, the tumbleweed rolled across – no-one mentioned it again, or referred back to it throughout the rest of the day. They just reverted back to the normal discussions on social protection, as the (presumably rising) waters simply swallowed up Ian’s faux pas.

Gender did a little better, not least because Jeni Klugman deliberately instrumentalized it – this was about improving the effectiveness of social protection, not overthrowing the Patriarchy (which I guess would have been the gender equivalent of calling for de-growth).

By contrast, once a paradigm is entrenched, evidence or doubt can be dismissed with the flimsiest of arguments. The World Bank’s Michal Rutkowski spoke on the implications of AI and robotics. He was a core member of the team that produced the recent World Development Report on the future of work, and he went full-on Dr Pangloss, arguing that concerns about the impact of new tech on jobs were unfounded because past periods of tech disruption were followed by the creation of new kinds of jobs and increased productivity. (i.e. the Luddites were wrong).

I’ve previously criticised the WDR on this, so thought I would raise it in Q&A

Q: On the impact of new tech on jobs, you are saying ‘the past is a good guide to the future’. But we know that is not always true. Suppose (as a thought experiment) that impact on jobs is actually serious and negative, what might a precautionary approach look like? Would you make any different recommendations for social protection in such a scenario?

A: ‘People cannot not do things. People will invent new things we cannot imagine now. I like playing chess, and thought it impossible that a computer could beat a Grand Master. I was wrong in 97 – Deep Blue beat Kasparov. I thought chess playing would die out, but numbers have been growing faster since 97.’

Okaaaay, so it’s true that people will find ways to fill their days, but will that make them happy? In the event of the impending destruction of millions of jobs (eg anyone who drives for a living, low skill manufacturing, an increasing number of routine white collar jobs), hobbies like chess or playing soccer badly won’t provide income, or the status and sense of wellbeing many people attach to having a job.

As for what this might all mean for social protection, Michal didn’t answer, but David Piachaud was pretty categorical: ‘Social Protection can’t protect against being rendered useless.’

For more on the World Bank and paradigm maintenance, see this 2014 post.

And here’s a brilliant video summary (10m views and counting) of the case for the World Bank taking this issue more seriously (ht Irene Bucelli)