CPI cannot be credibly discredited unless there is an alternative index

Dear Editor,
In his letter published in the local press on September 8, 2024, with the caption “Given questionable credibility of inflation data, it is important to know the weight of each of the nine groups in consumer basket”, Dr. Ramesh Gampat continues to question the credibility of the Consumer Price Index (CPI).
It is important to establish at the outset that, unlike GHK Lall, Dr. Gampat is not a “wishy washy” economist. He is technically sound and knows his onions, he has far more pedigree than I possess, and he commands respect in his field. As such, I do not wish to get into a debate with Dr. Gampat, but would just like to make two comments: one is a clarification and the other is a substantive commentary on the subject.
First, Dr. Gampat was defensive of GHK Lall, whom I had responded to on the subject of the CPI. I argued that GHK committed an error. Incidentally, at the time of writing, GHK published another letter in the Stabroek News edition of September 8, 2024, where he reinforced that error, that Dr. Gampat sees no issue with. To this end, GHK’s letter suggests that there is no separate index to measure changes in food prices. Nonetheless, this is not the case.
In my original letter it was illustrated that the six-months inflation (Jan-June, 2024) was 1.6% and the twelve-months inflation (June 2023-June 2024), was 4% (admittedly, the language used in my original letter did not make this point clearly). Further, I explained that the CPI is made up of several sub-categories, and if one wants to look at food prices separately, the sub-category for food prices showed that the six-month inflation was 3% and the twelve-month inflation for this category was 8%. In both instances, the food price sub-category was higher than the other sub-categories and the aggregate CPI (all items).
Additionally, I had also explained why the other sub-categories did not record any significant changes, which was due to direct and indirect policy interventions by the Government. Notably, Dr. Gampat did not dispute or refute this aspect of my argument.
The other substantive commentary I wish to make is that, for someone of Dr. Gampat’s calibre, in his quest to discredit the CPI or question the credibility of the CPI, he could only credibly do so on the basis of an alternate CPI that he has constructed.
Therefore, is there an alternate CPI developed by Dr. Gampat? Does he collect, update, and maintain a time series of data from the local market? If so, I humbly and respectfully request of him to please publish his alternate CPI, assuming that this is the basis upon which he is questioning the credibility of the current CPI developed, updated, and maintained by the Guyana Bureau of Statistics.
The main point that I was trying to explain for the average reader, however, is that the CPI is an index. It is not a simple arithmetic comparison of price changes for commodities as Dr. Gampat has correctly highlighted. And this is precisely what the likes of GHK and others have been doing, a simple arithmetic comparison, with just a few items. There is no established statistical methodology employed as with the CPI.
Accordingly, the methodology by which the CPI is calculated is much more complex than a simple arithmetic comparison, which can be summarised as follows (for simplicity):
(i) Prices for different types of goods (that make up the basket of goods used in the CPI) are collected each month. The prices collected are reviewed to ensure that they refer to the same quality and quantity of items observed in the previous month.
(ii) The average price for each item is calculated. This means that prices for the items are collected from various sources.
(iii) A price relative is computed by taking the ratio of the current month’s average price and the base month’s average price.
(iv) Since the CPI for each item is calculated using equally weighted geometric mean, the weight of the items in the base period is required.
(v) Once the item indexes are calculated the higher-level indexes for categories are compiled. The index for the subgroup is then calculated by dividing the total weighted price relative by the total weight of the subgroup then multiplying the result by 100.
(vi) Price indices are computed for each subgroup in all categories in the same manner. These index numbers are then linked together to obtain an aggregate index for each of the sections. This movement or average change in price is identified as the “All Items CPI” and provides a measure of the rate of inflation/deflation.
Henceforth, it is reiterated that in order for Dr. Gampat or anyone else to credibly discredit the current CPI, they can only do so on the basis of an alternate CPI that they would have developed.

Yours sincerely,
Joel Bhagwandin