The Institute for Fiscal Studies, based out of London, is a research institute that aims to foster the development of effective fiscal policy, principally through regular examination of information concerning national economics. In early 2015, the institute published a review of the Consumer Price Index, which measures changes in the price level of a household basket of goods, and even acts as the official measure of inflation in the United Kingdom. With inflation frequently making front page news, often generating divisive reactions from the public, one would expect the figures to be wholly accurate. But are they? And if not, are there ways we can improve on our measures of inflation?
Oh, I forgot to, erm, carry the one
Though long a staple of an economy’s national figures, the Consumer Price Index has come under scrutiny recently for its various limitations. According to the Board of Governors of the Federal Reserve System, there are five major sources of error that might lead to inaccuracies in the Consumer Price Index: upper and lower level substitution bias, new outlet bias, weighting bias, and quality change and new items bias. In a paper published by the board, the index’s inability to account for the change in quality of existing items, and the introduction of new items into the market, were noted as the most severe biases. Because of them, inflation measures fail to take into account the fact that consumers can exchange inexpensive goods for expensive ones, thereby overstating the cost of living.
Further inaccuracies are said to arise from a weighting bias, that is, the fact that ‘consumer expenditure’-based weights are often inaccurate due to a lack of reliable information. The limitation here is that information gathered from consumer expenditure surveys may lead to imprecise results, due to consumers needing to recollect expenditures made in the past, because of their inability to accurately estimate the imputed value of their homes, and because of their potential reluctance to record private expenditures like alcohol and tobacco. Moreover, new outlet bias is also a major source of error, arising as a consequence of outlet stores being rotated in and out of the Consumer Price Index sample, and thus leading to the index failing to capture consumers’ changes in consumption habits in response to discounted goods.
Overall, it is estimated that these five errors alone contribute a total index error of 0.62%, at a confidence interval of 0.10% to 1.20%, a lower figure than the Advisory Commissions’ previous estimates and thereby evidence suggesting that our inflation measures might not be completely accurate.
Yes, but how does it affect me?
The Consumer Price Index plays a prominent role in everyday life, affecting entire governments down to the baker across the road. On a government level, the index plays an important role in a number of measures – central banks, for instance, use inflation to set official interest rates. As interest rates have a direct bearing on investment and consumption, inaccuracies in the level of prices could lead to an economic bubble, as in the case of the ‘Japanese asset price bubble’ of the late 1980’s. Additionally, an error in the level of prices can spur on speculation in financial markets and affect bond prices, thereby influencing a nation’s economy.
But how is granddad affected?
Measurement errors can impact an individual’s benefits, as is the case with the unemployed and the elderly. Employment contracts and Social Security benefits, for instance, are adjusted annually with respect to the Consumer Price Index, and so an overstating of inflation could negatively influence a person’s income and, subsequently, standard of living. The same is true for food stamp beneficiaries and military retirees, children eating school lunches, and individuals signed with private firms who use the Consumer Price Index to adjust alimony and child support payments.
Progress, progress, progress
The good news is that the limitations of the Consumer Price Index can be eliminated, to an extent. For one thing, the information used in formulating and presenting the index can be improved. As of today, the Bureau of Labor Statistics rounds the percentage change of the index to one decimal place before publicly releasing it, which can signal speculation in financial markets. This short-term error can be corrected, however, by converting original data, before it has been rebased, into current values.
The sample size and accuracy of consumer expenditure surveys can also be improved upon, with more accurate results possible through both the use of personal consumer expenditure surveys, as well as an increase of the sample size to include not only more households, but also individual reports by each household member. Furthermore, the assumption of a constant basket of goods leads to substitution bias, the effect of which can be lessened through the use of a geometric mean, as addressed by the Fisher index. Lastly, policymakers ought to collect computerised information from retail stores of sold goods as to readily account for both changes in prices, as well as changes in expenditures, due to the fact that a more accurate cost of living index ought to “take account of changes in both prices and quantities”.
As opposed to improving upon the index, however, there exist also arguments for replacing the index entirely. In the review carried out by the Institute for Fiscal Studies, for instance, it is argued that the Consumer Price Index be replaced with the CPIH, an index that incorporates a measure of an occupiers’ housing costs. As such costs are said to account for 10% of total household expenditure, the fact that such costs are excluded from the original index can lead to significant imprecisions on a macroeconomic scale. Nevertheless, replacing entire indices might be unrealistic in the short term, with the CPIH, for instance, having its designation as a national statistic suspended due to “concerns about the way it is calculated”.
Maybe we should just ditch it?
The effects of the Consumer Price Index are far reaching, influencing every facet of an economy from top to bottom. But the question of whether to continue using the index as a measure of inflation remains up for debate. Although there have been numerous changes proposed as to improve the accuracy of the index, equally good arguments stand that the index should be replaced completely. Regardless of where one stands, what can be agreed upon is that the real life implications of such errors are severe, affecting entire populations and entire economies, and, no matter the means to go about spurring it, with respect to inflation measures, change is in order.