Why is the corporate welfare state so flawed – and how could it be made better?
I want to suggest that corporate welfare could learn a lot from some of the best reforms being introduced to welfare for people.
The first is reciprocity. It’s become a firm principle in welfare that payments from the state should, where possible, involve reciprocity – something for something. So entitlements come with conditions attached (like the requirement to look for work) and funding for organisations is tied to performance – with payment by results contracts (like the Work Programme) or Social Impact Bonds for adoption or ex-offenders.
The second is experimentalism. Here the principle – less firm but nevertheless gaining ground - is that programmes and policies should be tested first before they are taken to scale. So we see the proliferation of randomised control trials and pilots, in everything from schools policy to poverty alleviation.
These principles are often ignored (as with the over-rapid implementation of the Work Programme and Universal Credit which skipped over the crucial phase of experiment and improvement). But together these two principles of reciprocity and experimentalism have the potential to make social policy a lot more rigorous.
Contrast this with policy for business. Corporate welfare looks remarkably unreformed by comparison with human welfare. Indeed almost opposite principles sometimes seem to apply. Large sums of money are distributed without any strings attached, or any link to performance – and often following consultation with the very organisations that end up receiving the money.
Meanwhile much business support is evidence free – it’s rarely tested, almost never uses control groups, and when serious evaluations are done they often struggle to find any evidence of impact.
This has long been true of sector-specific support – for example for aerospace, or the car industry. It’s become even more true of recent changes to policy. The patent box and R&D tax credits are particularly stark examples, strongly lobbied for by some firms, generously funded by the tax payer but with very little sign that they achieved any demonstrable changes in behaviour.
In other countries there are glaring examples of something for nothing funding. A recent one was the funding of Verizon in US to provide fibre broadband to homes. Billions of dollars of public money went out - but there was nothing to show for it in return. We know that some industries are highly likely to become hopelessly dependent. Nuclear power sometimes looks like the most extreme case in point. Having promised cheap or free electricity for half a century it funds armies of lobbyists to sustain a flow of public guarantees and subsidies, the most recent of which is the long-term guarantee of a price double the market price.
The wrong response to this would be to conclude that government should stand off. Instead what we need is a mix. On the one hand we need strong engagement and collaboration between government and business. Particularly where an industry is growing fast it needs good access to government to sort out regulatory barriers, to deal with planning, R&D or supply chains.
But the aim should not be one-sided welfare, but rather better deals in which both sides win. Groups of firms or sectors should be encouraged to engage with government and propose deals – where governments sorts out the things only it can fix, like basic research, infrastructure, regulations and skills – and business gets on with the job of developing better goods and services. The crucial point is that these deals need to be structured around performance – with funding released in response to demonstrable evidence of growth. Welfare needs, in other words, to become conditional, while policy becomes experimental.
These two principles would together imply a revolution in the way industry departments around the world work. But it would be a rather healthy revolution, and would reflect the best of how markets work rather than the worst.