Policy Research Institute - PRI Bangladesh

The Policy Research Institute of Bangladesh (PRI) is a private, nonprofit, nonpartisan research organization dedicated to promoting a greater understanding of the Bangladesh economy, its key policy challenges, domestically, and in a rapidly integrating global marketplace.

Inflation and statistical paradox

Published: Wednesday, May 11, 2011

Economic Analysis
Inflation and statistical paradox

Ahsan H. Mansur

Bangladesh Bureau of Statistics (BBS) has recently published the consumer price index (CPI) inflation data for March 2011.

Recent figures show that the upward trend in inflation is continuing and the point-to-point inflation rate has reached 10.86 per cent fueled by hefty increases in food prices, with food inflation touching almost 14 per cent. This inflation rate is the highest since July 2008. Due to persistently high point to point inflation, inflation calculated on a 12 month moving average basis remained above 8.0 per cent for the last two quarters.

Point to Point Inflation in March-2011

General   10.49

Food        13.87

Non-food   4.32

Of which

Gross rental, fuel & lighting (national)  3.51

Gross rental, fuel & lighting (urban)     1.37

Gross rental, fuel & lighting (rural)       4.16

A sharp increase in international commodity prices, a depreciation of the taka exchange rate against all major currencies, and strong domestic demand fueled by monetary expansion of around 22 per cent, has led to this high level of inflation rate (well above the initial budgetary estimation of 6.5 per cent) is not surprising at all. But what surprises many observers including myself is the benign nature of non-food inflation which is reported to be only 4.32 per cent. This low level of non-food inflation is a bit puzzling when we go for renting an apartment or house or purchase non-food goods or services. All rickshaw pullers have certainly increased their fares by 20-25 per cent in the last 12 months. Minimum fares which used to be Tk 6-8 is now at least Tk 10. My observations are certainly anecdotal, but I will bet that most citizens of Dhaka will agree that a rickshaw puller is now asking for Tk 25 instead of Tk 20, a shirt is now costing Tk 900 instead of Tk 800 or a household appliance is priced 10 per cent more now than that of one year back.

This counter intuitive development forced us to look further at the BBS's inflation data, which unfortunately raised more questions than answers. Here, I will mention some of the findings.

Normally, one would expect that food and non-food inflation rates will move together in the same direction with a short lag. When food price increases, cost of services and wages are adjusted upwards quickly, resulting in increased cost of production of other commodities. But in the case of Bangladesh, on-food inflation generally tends to go down when food inflation goes up, especially during the high inflationary periods. It tends to defy normal expectation based on economic rationality.

When we look at the components of the non-food basket used in constructing BBS's inflation index, we find that gross rent, fuel and lighting together constituting about 40 per cent of the non-food basket and 16 per cent of the total CPI basket, is the single-most important component in the non-food price index. In all likelihood, anomalies associated with the non-food index has its origin in this component since the current index level for this component is only at 155 for urban areas in comparison with 100 in the base year of 1995/96. The number implies that in the urban areas the price of this basket has increased by only 55 per cent over the last 15 years. During the 12-month period through March 2011, the increase was less than 1.4 per cent.

Anyone who is renting an apartment and who is looking for or shifting to another apartment after getting a rent increase or eviction notice from the homeowners will laugh at the number. It seems that being the largest component of non-food expenditure, the brunt of adjustment is artificially being imposed on gross rental when inflation rates are calculated.

It is also counterintuitive to observe that the price index of the non-food basket comprising gross rent, fuel and lighting increased at a much higher pace in the rural areas reaching the level of 217. The price increase was 117 per cent in the rural areas over the last 15 years, compared to only 55 per cent in urban areas. Certainly this is a paradoxical outcome considering the rapid urbanisation, pressures on urban housing and the surge in urban real estate prices in recent years.

The consumer price index, measuring the rate of inflation, is the most politically sensitive headline item. Thus, it is no wonder that the inflation rate is a very crucial indicator of macroeconomic management. Key economic policies such as monetary policy and fiscal policy are designed and revised taking into account the targeted inflation rate. Any policy based on the inflation rate which is different from the reality, will fail to bring out the desired outcome. At the micro level, decisions are adjusted based on perception about the current inflation and expectation about the future inflation rate. It is thus imperative upon BBS to ensure integrity of the index.

The issues raised in this write up are not simply my own, similar doubts have been cast by a wide range of professional economists and development partners and practitioners during various policy dialogues and other contexts. The current government has given greater importance to BBS, as it should be, by elevating the position of the head of the organization to a full Secretary level. We sincerely hope BBS will look into the issues raised in this commentary and in other contexts about the quality of the CPI data in a comprehensive manner and take necessary measures to restore confidence in the integrity and quality of the statistics published by bureau. (The writer is Executive Director of Policy Research Institute (PRI) of Bangladesh. He can be reached at e-mail:  amansur@pri-bd.org)

Speech