In the Argentine economy, the correct measurement of inflation is of great relevance since it is linked to the evaluation of the success or failure of an economic policy that seeks to reduce it. Currently, the country has experienced a substantial reduction in the inflation rate (although still high for international standards) while the need arises by the official body in charge of measuring the consumer price index (CPI) to put on the table the effectiveness of the index to capture the true behavior of prices.
To continue with the analysis, it is important to highlight the conceptual difference between the official price indices surveyed by statistics agencies with respect to the cost of living. The first of these relieves the evolution of the prices of a fixed basket of goods consumed by a reference population, while the last is a theoretical concept that attempts to measure the variation of the cost of maintaining the same standard of living, considering the capacity of consumers to adapt to price changes. The proposed difference results in the appearance of biases that will be treated later.
Under the described plot, it is worth bringing on stage the debate that occurred in the mid -1990s in the US on the measurement of the CPI. Precisely, in 1995 the US Senate appointed a commission called “Advisory Commission for the study of the Consumer Price Index” whose main task was to study possible biases in the calculation of the CPI. The conclusion of the task resulted in a report colloquially as “Boskin Report”, since Michael Boskin was the president of the Advisory Commission, in which it was concluded that the CPI overestimated inflation in approximately 1.1 percentage points annually in 1996 (it is also highlighted that in previous decades the calculation has exaggerated even more the measurement).
The relevance of the Boskin report was transcendental for its time. The inflation calculated by the “Bureau of Labor Statistics” was used to index the annual increases in social security payments, among other transfers programs. The implication of the overestimation of inflation is that the federal budget had increased more than well as the projections of budget deficits.
With regard to the technical issue that the report reviews, it is worth highlighting the main biases present in the measurement of the CPI:
Assets substitution bias: occurs in front of a change in relative prices, where the fixed basket of goods that IPC usually considers fails to capture the replacement of goods that become relatively more expensive for more cheaper ones. For example, if apples increase by 10%, it does not imply that the consumer is 10% worse since it can replace some apples with pears whose price has not risen. It should be noted that statistics agencies for the elaboration of the CPI usually consider a fixed basket of goods, which arises from household expense surveys carried out every certain number of years.
Sale substitution bias or discount stores: In this case the substitution occurs at the trade level. Since the CPI is relieved in a fixed sample of places of purchase, changes in consumer behavior would not be contemplated which accesses other types of stores that offer discounts.
Quality change bias: This bias occurs when you improve the quality of a product is not contemplated in prices properly. An example that illustrates the respective point is that of lighting foci that last more and consume less energy, giving the case where the consumer is willing to pay more for an improved version of the product (technological improvements would not be considered in the price index).
BIEWORK PRODUCT: This bias is typical of an index where the participation of goods and services in the consumer basket remain constant over time (technically called Laspeyres Index). It occurs when new products that turn out to be relevant today are not included in the consumption basket. A typical example of this phenomenon was the lack of incorporation of cell telephony in the measurement of the CPI.
Recently in Argentina it was announced that INDEC will update the CPI with a basket that will include new consumption habits[1]. Among the measures are the inclusions of cell phone and streaming services (Netflix), to name a few. This modification is a step towards the construction of a relevant index for consumers and policy makers. It is important not to lose sight of the macro-fiscal implications that this measurement has, as reported by the Boskin report, while allowing consumers to make better decisions based on having more appropriate statistics.
*Norberto Mangiacone is an economist at CEFIP-IEE- (FCE UNLP)
[1] The current CPI is calculated based on the National Household Expenditure Survey (Engo) 2004-2005, with a 2016 base index.
By Norberto Mangiacone
