24 February 2006
Everyone agrees that the council tax is regressive – but there's less consensus on the solution. The interim Lyons report plumped for reforming the benefit system and not the bands. But both are needed, argue Peter Kenway, Ines Newman and Nick Cuff
One of the main points to come out of Sir Michael Lyons' interim report on local government finance was the regressive nature of council tax. The remedy he proposes for this problem is to reform and fully integrate the council tax benefit system. This is an important and welcome conclusion and one the government should take on board.
Less pleasing is the report's conclusion that adding additional bands to council tax, beyond the current eight bands (A-H), 'would appear not to have a significant impact on fairness in terms of the link to ability to pay as measured by income'. The importance of this assertion cannot be overstated because it appears to rule out reforms to the council tax itself, such as adding extra bands or changing the 'multipliers' for some of the bands.
In our response to the interim report, we have shown that this conclusion is based on incomplete analysis – and to that extent is therefore wrong.
The inquiry uses the yardstick of 'ability to pay' to assess the fairness of the tax and the merits of various proposals for reform. It then uses the ratio of council tax paid as a percentage of income as the measure of 'ability to pay'.
The final report will need to consider in more depth whether this is the right basis on which to judge this tax. Council tax is, in part, a property tax and if a person with a higher income chooses to live in a smaller home, it is quite right that their council tax should take a smaller percentage of their income.
Furthermore, a household's ability to pay will depend not only on their income but also on their other living costs. A couple with two children might have a higher income than the pensioner next door, but they are also going to have more demands on their income.
But our main disagreement with the interim report lies in the way in which the households have been grouped together to assess overall ability to pay. The interim report does this by dividing households into ten equal-sized groups on the basis of their income. It then shows a graph of council tax as a percentage of income for each of the ten groups (see figure 1). This demonstrates that on average, council tax represents a higher percentage of income for households in lower income deciles than it does for households in higher deciles.
Click here for council tax graphs (Figure 1 and Figure 2) (this will open up a new browser window)
The interim report then shows that adding more bands to the council tax system does not really change the graph. It explains this 'negative' result as reflecting 'the fact that people living in houses within a specific band tend to have widely different income levels'. But that is only a small part of the explanation. Two other factors need to be taken into account, which alter the conclusions about what should be done.
First, the report considers only limited changes, with none at all to the multipliers for bands B through F and only a very small one for band G. The only significant changes are to half of band A, the bottom band, which gets a reduction of one-sixth, and to band H, the top band, which goes up by a third. If more radical changes were to be made, reducing the tax paid for bands A to C and increasing it for bands E to H, then, according to our analysis, that would indeed make a big difference to the line in figure 1.
Second, the glaring unfairness in the tax structure is confined to the top two bands, G and H. This can be seen in figure 2, which measures fairness in the same way as figure 1 but reports the results according to the band rather than income decile. This diagram shows that the tax on properties in band G, and especially band H, takes a much lower percentage of household income than it does for properties on average. The obvious policy conclusion is to increase the tax in these top two bands relative to all the others.
There are a number of reasons why it is important to look at figure 2 as well as figure 1. The single most important one is that bands G and H contain less than 5% of all properties. Even if the match between income and council tax band were perfect, this unfairness would hardly show up in a graph like figure 1 which lumps the top-earning 10% of households together.
Figure 2 is the basis for our long-held view that while the heart of the council tax system is basically sound, the ends of the system are not. Looking at ability to pay by band can help with explaining the changes. For example, a statement to the effect that: 'We are increasing the tax on properties worth more than £1m because, at the moment, the households living in them pay much less relative to their income than average' can be understood by the public. Both what is being done and why are made clear.
If the government is going to sell property revaluation to the public, a problem that has merely been postponed, then additional bands will be vital. One of the charts in the interim report shows the number of properties moving up and down one or more bands under the National Update revaluation option. In Yorkshire and the Humber, 19% of properties are shown as due to move down a band. However, none of these properties will be among the 45% of properties in band A, since they are already at the floor. Consequently, the benefits of revaluation in the region will be focused on the properties of above-average value. Likewise, in London, the pain of revaluation will hit 35% of households who drift upwards, while the lucky 2% already in band H will escape scot-free.
Council tax revaluation will be hard enough politically without these glaring iniquities. The remedy for both is more bands at the top and the bottom. Revaluation, in other words, will require council tax reform.
Peter Kenway is director of the New Policy Institute. Ines Newman and Nick Cuff are, respectively, head of policy and campaigns officer at the Local Government Information Unit
PFfeb2006
