Few common economic phenomena are as misunderstood and misrepresented as “inflation”. Unemployment represents a concrete event that manifests itself in a straight-forward manner, loss of work, application for benefits and subsequent job search. We can contrast this to inflation. Economists struggle to define what inflation is.
A “rise in the general price level” comes across as the preferred definition, but is ambiguous concept. In actual economies with their many goods and services, the “general price level” exists as a measurement concept that no one directly perceives. In addition, we have many statistical measures of inflation. I focus on the commonly used consumer price index.
If we asked the proverbial person on the street, “are you unemployed”, we are likely to receive one of three replies — no, yes or “between jobs”. Few if any adults would reply “I don’t know” or none of those”.
We could ask the same person, “was your standard of living affected by inflation last month”? There are many reasons why the person may find it difficult to answer. If the Bank of England through intention or accident kept average price increase close to its 2% target, the rise might prove insufficient for the respondent to notice. At least three characteristics of the “general price level” reinforce ambiguities of perception: people have different consumption habits; all prices increases are not inflation; and in practice price determination falls into different processes.
To take obvious examples of the first, someone who rents accommodation will be unaffected by an increase in mortgage rates, changes in the fare for the London tube will go unnoticed by a rural bus rider, and vegetarians will care little about meat prices.
More important, consumer price indices use average consumption weights. The substantial difference between average and median (mid-piont) income implies that the consumption pattern of the typical person differs substantially from the weights used by the Office of National Statistics. ONS also calculates Indices by income deciles but these are rarely used in the media. Since 2006 when the indices began, average price increases for the population in the second decile (tenth) have differed each month from those of the ninth decile by an annual equivalent of 0.3 percentage points (ONS compares deciles 2 through 9 to avoid extreme values encountered at the ends of the distribution)).
When people in the high ninth decile show no change, prices for those in the low second decile consistently show a small annual increase (calculated from ONS statistics for 2006-2019). The distribution bias provides sufficient reason to make individual perception of inflation differ by income groups over several months’ periods of time. And, of course, higher income groups may not notice small changes at all.
This ambiguity is substantially increased because (my second point), in time of low inflation quality change and new products have a calculated price impact well in excess of inflation itself. The items people purchase continuously undergo quality change as well as being joined by new products. Over twenty years ago the United States Congress commissioned a detailed investigation into the effect on inflation measurement of such changes.
That study, the Boskin Report, concluded that new products and product improvement contribute about one percentage point to consumer prices each year. While we have no equivalent UK study, the internationalisation of production and consumption suggests a similar impact. If so, when the ONS reports an annual general price increase of two percent, what we normally mean by inflation — the same thing costs more — is actually one percent or less.
Perhaps the most important problem with measuring the general price level and inflation is a third complication. The prices in our economy fall into three distinct categories: 1) goods and services who prices are determined in international markets, 2) those whose prices result from contact arrangements of varying time lengths (including public sector prices), and 3) prices determined in short term domestic markets processes (“spot market” prices).
With this complexity of price determination in mind, we can reconsider orthodox monetary policy. For example if the ONS measured rate of price increases goes to 3.5%, the Bank of England is mandated, on advice of the Monetary Policy Committee, to act to reduce the rate down toward 2%. An increase in the rate at which the Bank of England lends to private banks provides the conventional tool to archive this outcome. Finding it more expensive to access funds, banks raise their lending rates. Businesses then find their operating costs higher and reduce output. Slower private expansion reduces pressure on wages and prices, bringing inflation down.
If this logic were sound, it would mean that the slow down in price increases would concentrate in domestic markets, leaving international prices such as petroleum unchanged, as well as having little impact on prices under contracts of various lengths. The most flexible prices arise in markets with unregulated wages, such as retail and wholesale trade, where pay is also quite low.
Thus, if the logic of conventional monetary policy holds, its distributional effect should prove quite unequal, its burden carried by the lowest pad. The occasionally encountered argument that inflation disproportionately harms the poor is false; seeking to reduce inflation disproportionately harms the poor.
How did economic policy fall into this commitment to consistently inequitable monetary policy? The alleged problem, excessive price increases, defies accurate measurement. We have little evidence that the solution to this alleged problem, central bank manipulation of interest rates, would have any direct effect on it. As I argued in my previous blog, underlying this mainstream monetary policy we find the belief in automatic adjustment to full employment — market economies naturally seek full employment, and government provoked inflation is the major source of instability.
A specific model of the economy provides the bridge from the belief in self-adjusting markets to mainstream monetary policy, the infamous Quantity Theory of Money. In its simplest form that theory views the economy as generating one common output and money as created by governments. In the simple neoliberal monetary world market economies automatically find full employment; only one output is produced; governments control the money supply; and inflation results from too much government created money chasing too little output.
But market economies tend to generate unemployment not full employment. Real economies produce many goods and services with quite different process of price determination. Governments and central banks at most influence not determine money in circulation.
Inflation is not the result of too much money. That is its consequence. In a third blog I focus on that issue — what causes a broad and persistent increases in prices, when that is a problem, and what policies to manage it.
picture credit : flickr:EpicTop10.com