In a ministerial statement last week the government announced a significant change to its policy to localise Council Tax Benefit (CTB) from next April. In this observation we ask why such a significant change has been announced to a policy two years after it was first announced, less than six months before councils will have to implement it and after many have already consulted on the structure of proposed schemes.

CTB provides support to 5.9 million low-income families, more than any other means-tested benefit or tax credit in the UK. The government is proposing to localise support for council tax from 2013–14, abolishing CTB across Britain and giving grants to local authorities in England and to the Scottish and Welsh governments to design their own systems for providing support for council tax to low-income families. On top of this, the government planned to cut by 10 per cent the funding it provides for council tax support. This would save around £500 million a year.

We have analysed the effects of these proposals in some detail, concluding that localisation would create considerable complexity just as Universal Credit is being rolled out with the intention of simplifying things. It also has the potential to undermine many of the improvements to work incentives that Universal Credit is intended to deliver. For councils to save the full 10 per cent by which funding was being cut by making the system less generous, either the means test would have to be so severe that some people would be worse off after a pay rise – or else councils would have to collect some local tax from the very poorest for the first time since the poll tax. Many councils are consulting on schemes which would have these sorts of consequences.

Just last week - two years after the policy was originally announced, less than four months before local authorities have to finalise their new schemes, and only a week before the third reading of the bill in the House of Lords - new proposals have been forthcoming. In a ministerial statement a £100 million package was announced. This money – which amounts to a fifth of the total planned savings – will be available to councils whose schemes meet a particular set of criteria that the government considers “best practice”. It will, apparently, be available for one year only.

Councils will be eligible for the money if nobody currently on full CTB ends up paying more than 8.5% of their council tax liability (in practice, the costs of collecting such small amounts from very low income households who are not used to paying council tax mean that councils may well prefer to give a full rebate to such households); if the rate at which the benefit is withdrawn as income rises is no higher than 25% (compared with 20% at the moment); and if there are no “cliff edges” in the system.

Even with an extra £100 million to soften the blow, it is hard to see how most councils could design schemes that meet these criteria within the reduced funding intended for council tax support. So it looks as if the government is aiming to pay councils not only to design schemes that the government likes, but to design schemes that don’t cut support as much as councils’ funding is being cut, leaving them to make up the shortfall from elsewhere in their budgets.

It is hard to square this development with a policy whose stated aim was to devolve responsibility. And why the additional money should be appropriate in the first year of the policy and not later is unclear. But perhaps most worrying is what this says about the policy-making process. The potential downsides that the government seems to be trying to ameliorate – losses for the poorest households and weakening of work incentives – have been obvious to many observers for a long time. Yet this announcement has come very late in the process. The bill had already completed its passage through the House of Commons and scrutiny by a committee of the Lords had finished: last week’s announcement came on the eve of a key Lords debate on amendments to the bill, and just a week before the third reading in the Lords. Many councils have already been running public consultations on draft proposals (as the bill requires them to do) yet are now being incentivised to change their proposals at the last moment – perhaps only to revert to their original plans when this extra funding is withdrawn a year later.

The case for well thought through reform of the welfare system is overwhelming. The dangers of less fully considered reform - as this one appears to be - are considerable.