Wednesday, 12 October 2016

Tax-driven migration

Both the BBC and the Western Mail gave a lot of prominence a week or two ago to a report produced by economists at Cardiff University suggesting that reducing the higher rate of tax in Wales might actually lead to an increase rather than a reduction in government revenues, as a result of people moving from one side of the border to the other to chase lower tax rates.  The headlines were clear and explicit: “… a Welsh tax cut would end up boosting revenues”, according to the Western Mail, and “Higher income tax rate cut in Wales 'would raise money'”, according to the BBC.
The BBC headline reflected the detail of the story which followed it, which presented the report in definitive terms.  However, the detail in the Western Mail report was, in fairness, rather less conclusive, being full of qualifying words such as ‘may’, ‘might’, ‘if’, and ‘possible’.  Looking at the report itself, on this occasion the Western Mail’s reporting (after the first few sentences at least) seemed to me to be a much more accurate reflection of the content.
Lack of certainty is never a deterrent to politicians of course, especially those who don’t bother to read the source material, and the Tories’ Andrew RT Davies was quick to pounce on the report as justification for a policy of reducing both the higher rate and the basic rate of tax.  It’s a statement which owes more to his own ideological commitment to low levels of direct taxation than to the conclusions of the report.
Having seen the level of uncertainty indicated by the Western Mail’s report, I was interested in seeing just how the report’s authors had calculated the migratory impact of a change in tax level.  The idea that a difference in taxation rates between two neighbouring jurisdictions could, other things being equal, lead some people to take a decision as to which side of the border they should live, is one which has been made before, and in terms of theoretical economics, assuming that all individuals ultimately take rational decisions based solely on their own economic self interest, it sounds like common sense.  And if it’s going to apply anywhere, it’s likely to apply along a border such as that between Wales and England where cross-border commuting is (comparatively) easy for a large number of people, and where there are few other obvious differences.
The mechanism by which it might work is clear as well; the report itself uses an example to show how much a higher rate taxpayer might save from a given reduction in the rate of tax over a period, and compares that with the costs of moving. 
So the theory is clear enough; but how many real people, as opposed to theoretical economic concepts, are actually likely to sit down and do that calculation before deciding where to live?  And how many will base that decision solely on considering one variable, i.e. the rate of income tax?
The study itself accepts that there is simply no data which can directly tell us how many people will migrate based on tax rates, so there’s some analysis in the study based on establishing a correlation between the levels of council tax and the rate of migration between council areas, and then adjusting that to account for the differences between a property tax and an income tax.  It’s clever academic stuff, but it is ultimately based on an assumption that we can in some way measure the migratory impact of council tax variations.
And that’s my problem - whilst there certainly is a degree of correlation between in-Wales migration between council areas and the level of council tax, I couldn’t see that the model actually established a clear causal relationship between the two, i.e. evidence that people are actually moving from one council area to another specifically in order to benefit from lower council tax rates.  It’s a very interesting academic study of the likely outcomes of particular tax changes if people do indeed choose to migrate in response, but its conclusions aren’t – and cannot be – as conclusive as the media reports have suggested.  Not for nothing do the report’s authors themselves caveat a lot of their statements with words like if, might, and may.
Personally, I’m inclined to accept that there are indeed some people who will decide where to live on the basis of such calculations, particularly if the gap is a large one.  Isn’t economic difference, after all, one of the drivers of mass migration?  But I tend to believe that in the more limited case of migration across the Wales-England border, the number is much lower than a purely economic model would suggest, not least because (a) the differences are likely to be quite small and (b) most people will be considering a range of other, softer factors as well.  On top of that, a change to an individual tax rate rarely happens in isolation. 
But one thing of which I am convinced is that we should not allow bold headlines, or politicians with an ideological commitment to tax-cutting, to bounce us into adjusting our tax regime to suit the few who are so narrowly motivated by differences in marginal rates of taxation.  The assumption that significant numbers of people (or smaller numbers with significant enough incomes) will simply move elsewhere in response to tax differentials is far from being proven in practice as opposed to theory.


Gav said...

Probably just a variation of the old "Laffer curve" argument that reducing tax rates increases government revenue. However often these arguments are knocked down they always reappear in some form or other - the technical term for this is "whack-a-mole".

Anonymous said...

We currently have different tax rates along the border. There is currently about a £400 difference between Powys and Shropshire for H band homes. I can't see a rush from western Shropshire to eastern Powys to get the tax break.