Don't Believe the Inflation Figures

Inflation figures are said to be a measure of the price level. However what is the price level? This is normally calculated using a weighted basket of goods and calculates the change in prices over a time period, normally a year. However there are numerous problems with this approach. Firstly it fails to take into account changes in quality of goods- when deriving the price level it calculates the change in price of a good over the year however no-one can actually objectively determine that the good bought in the second year was identical to that of the first. Since quality, in this context, can only really be defined as “those properties to which the buyers and would be buyers pay-heed” and since these properties are in constant flux, determined by individuals’ value judgements, one is essentially adding apples and oranges to create this statistical construct.

Secondly, since various goods play-varying importance in one’s existence, coefficients are needed to show the relative importance of various commodities which is of course are arbitrary since importance to each actor is subjective. The conventional method for this is to calculate how much one spends on a good to see its relative importance- so in low-income households the main expense would be mortgage, or rent, payments. However if the price of a good rose, such as butter, this would change the pattern of consumer spending since they would substitute it for other goods, such as margarine, changing the relative importance of each commodity based on their expenditure upon it. So any change in a good will change the individual’s expenditure pattern which is neglected by the computation of the price level.

Another problem with this weighting is that different people buy different things and thus have different “price levels” which the government statistics do not take account of; Murray Rothbard always complained that the prices of books kept increasing but it made no indentation on the price level. Astonishingly mortgage repayments are omitted from the CPI even though they constitute a large percentage of household expenditure. Further there are various statistical techniques to compute these averages: arithmetic mean, geometric mean, harmonic averages, median and such like. However there is no objective way of deciding which is a superior technique and thus any decision must again be arbitrary. Finally, as Mises points out, a “judicious housewife knows more about price changes as far as it affect her household than the statistical averages can” and she is no less scientific than the statisticians.

So how should we measure inflation? The older, and far more cogent, definition of inflation is the increase in the money supply. Suppose one had two goods in the economy A and B which both had prices of £5 each. Now if good A was to rise in price to £6 the price good B would have to fall to £4, assuming a constant money supply, which would yield an average price of £5. The only way prices in general can rise is if the money supply increases: suppose the money supply increased by £2 then the average price in the economy would rise by 20% to £6. Thus the prices rises are only a consequence of the increase of the money supply. It must though be stated at this point that prices in the real world do not increase, necessarily, at the same rate as the growth in the money supply since new money enters via a step process with different people receiving the new money at different times.

So instead of looking at the Price Level we should observe the M3 (a monetary aggregate) growth to see the true inflation rate and its wealth reducing consequences.