Peter Saunders has challenged me to provide a more detailed response to his paper ‘Do we still need the welfare state?’ The full paper is here (PDF). The core of the argument is summarised reasonably well by this para from his comment.

My argument is essentially about ‘churning’ (i.e. the fact that most of us pay in taxes for most of what we get back in welfare payments and services), not poverty. I explicitly recognise that some provision still needs to be made for those who cannot look after themselves (I estimate this at no more than 5% of the working-age population – back in the sixties it was only 3% and we are a lot wealthier now than we were then, so it really shouldn’t be much more than 5%). In my AIFS paper in February this year, for example, I note: “There will always be some people who for one reason or another are incapable of supporting themselves on a long-term basis. But their numbers are not so large as to require a welfare system on anything like the current scale to support them.”
My key point is that, unlike 100 years ago when the *mass* of the population could not earn enough to provide its own basic necessities (which is why a *mass* welfare system evolved to fill the gap), today this is patently no longer the case. …The image that most of us have of the welfare state is of a system that takes from those who can afford to pay and reallocates to those who are poor. The reality, however, is that as welfare spending has grown, so it has had to be funded increasingly by taking money from all sections of society, not just the wealthy or high income earners. This means many people now pay with one hand and receive the money back with the other, and that most beneficiaries end up paying for most or all of what they receive. Indeed, the main reason many welfare recipients appear to need government assistance is that the government has taken so much of their income away in taxation, thereby pushing them into dependency.

The central problem with this argument is that Saunders confuses arguments about transfer payments with arguments about publicly provided services, lumping both together under the term ‘welfare’ and assuming that the purpose of these payments is to alleviate poverty.

As far as transfer payments (age and other pensions) are concerned, I agree with Saunders’ basic point that we should not tax ourselves heavily to fund universal benefits, effectively taking money out of one pocket and putting it back into the other. But of course, we don’t. As far as I’m aware, there are no cash welfare payments in Australia that are not subject to means tests, usually tight ones. Saunders may perhaps favour even tighter means tests, but this gets us straight into a huge and messy debate about effective marginal tax rates.

So even though the majority of Saunders’ discussion is about poverty-alleviation and cash transfers, the real issue, as far as Australia is concerned, is whether governments should reduce spending on services like health and education. There are strong arguments for public funding of these services which have little to do with income redistribution. Saunders concedes the crucial point early on his paper. While he’s opposed to governments actually providing health and education (a ‘socialised’ mode of consumption) and welcomes shifts towards funding of private schools and hospitals ‘a privatised mode of consumption’, he concedes that this does not imply big reductions in spending:

The overall level of government social expenditure might not fall much as a result of this transition, but what does change very visibly is the form in which services are produced and consumed. (emphasis in original)

If, with Saunders, we rule out substantial cuts in expenditure on health and education and we are already imposing tight means tests on cash payments, the only options remaining are to reduce the value of pensions and benefits and to tighten eligibility. These have been the approaches adopted in the US, and they have contributed to the outcomes noted in the previous post.