IFS got their inflation calculations wrong by a factor of 10

Posted on

I have received an email from the IFS in response to some of the points I made on this blog last week regarding their suggestion that VAT be charged on all goods and services at standard rate, so increasing VAT on food, heat, power, children's clothes and much more besides.

In particular I have been advised that I:

identif[ied] an error which the authors have now corrected, for which thanks. The paper did understate the effect of the example VAT reform on retail prices - the decimal point was in the wrong place and the one-off increase in the RPI should be 3.5% and not 0.35% (roughly one-fifth of the RPI basket rises by just under 17.5%). This remains a one-off increase in the price level, not a permanent change in the rate of inflation.

I am grateful for the IFSs' acknowledgement of this point. It was readily apparent from the information I calculated and published that the inflationary impact of the proposal they made could not have been of the order they indicated.

But I do not think this the end of the issue. I do not accept that this adjustment is a "one-off increase in the price level, not a permanent change in the rate of inflation". I also think that an inflationary impact of this order of magnitude makes another explicit assumption of the paper, which is that "it would not be difficult to compensate most poor losers from the move to a uniform VAT".

To make these observations firstly ignores several important issues concerning inflation. Second they ignore the nature and impact of the compensation package the IFS proposes in its paper. Finally (for now) they ignore the likely behavioural impact of the change on other economic policy. All three flaws could be forgiven if the assumed inflation impact were 0.35%, as originally stated. When it is 3.5%, as all the published documentation now confirms, the errors compound to suggest that the policy proposal is fundamentally flawed. This needs explanation.

First, inflation is never "one-off". Inflation of one period affects the base level at which future inflation compounds. A one off increase does therefore have permanent consequences for price levels, and if the rate of compounding thereafter varies for different income groups and commodities a single one off change can have permanent differential effects.

Second, this proposal would not have a one off inflationary effect unless these conditions held true a) all losers from the change were completely compensated b) the compensation was exactly coincident in timing with the price increase resulting from the VAT increase c) nobody took opportunity of the confusion in pricing that the imposition of VAT would have on the food and other markets to generally reprice, improve margins or product reposition d) all losers perceived the compensation as being due in respect of the imposition of VAT and nothing else e) the take up of compensation was perfect f) the price elasticity of demand of all the goods affected by the change in pricing was constant.

Actually, there are more conditions than that but they are sufficient for now. Since either the IFS does not recommend that these situations be crated, or it would be impossible for them to be created it is absolutely certain that a perception of general inflation requiring compensation through wage rate increases would be created by this change. This is inevitable when the IFS only recommend compensation in full for those in the lowest three deciles of income earners (who earn £16,000 a year of less on average). I think it fair to say those in deciles 4 to 9, earning between £16,000 and about £40,000 a year will seek some compensation through wage inflation in this case, and so the argument presented is simply wrong. Put simply, this is not a static case scenario as the IFS represent. It will be a dynamic change with ripple effects precisely because they do not propose full compensation for all effected.

So, third, let's consider just one of the possible behavioural consequences of a 3.5% hike in the inflation rate leading to wage inflation, as is inevitable. The Monetary Policy Committee of the Bank of England has a statutory duty to keep inflation at the rate of 2% or lower, and is given the sole instrument of changing interest rates to achieve this goal. It is given no power to consider any other economic issues, or to make recommendation for use of any other measure to tackle inflation. This is of course economically insane, but that's not the issue here: this is the scenario we face and which we should assume will persist. Right now inflation is approaching 4% and as a result although it is very obvious that almost all other woes in the economy could be relieved by significant reductions in interest rates the MPC is refusing to make that cut. It would rather face the destruction of banks, massive increase in evictions and all the resulting social trauma that follows, plus very real risk of recession than deviate from its required path to tackle inflation.

So let's assume it will continue to do that in the face of this 3.5% increase, which will be on top of current rates. Inflation would exceed 7%. So too, no doubt, would interest rates, I suggest. Economic meltdown would result; there would be mass evictions from houses as mortgage defaults sky-rocketed, bank bad debts would cripple the economy, recession would become depression, unemployment would spiral, and all this to "finance [ £11 billion of] further desirable tax reductions" as the IFS puts it.

It's an interesting economic scenario of the tax tail wagging way out of control. It would be inconsequential if an inability to handle a decimal point and the lack of understanding of macro-economic management by such an influential think tank (for let's be clear: the IFS is no more than that) did not have such potential serious consequences. But in this case the consequences are enormous, and so damaging that I think my comments last week were very entirely justified.

But I gather from information supplied that despite that the IFS are concerned about the accuracy of some of my other observations, even though they couldn't spot a tenfold error in their reported inflation rate. I'll turn to those issues in other blogs, soon.