A tale of two ‘low’ inflation episodes

http://www.thehindubusinessline.com/2007/11/12/stories/2007111251910100.htm
This article is so vital I am reproducing it in full.

Harish Damodaran

New Delhi, Nov. 11 The annual wholesale price index (WPI)-based inflation rate falling below the three per cent mark for the first time in over five years may please mandarins and analysts, who would see it as a vindication of proactive supply-side measures in conjunction with sensible monetary and fiscal policies.

The 2.97 per cent year-on-year WPI increase for the latest recorded week ended October 27 is the lowest-ever since the 2.86 per cent level of July 20, 2002. The current headline inflation rate has, thus, touched a 275-week low. While that may seem impressive for the habitual number-cruncher or policymaker, it is unlikely to, however, cut ice with the common man.

The reason for the aam aadmi’s scepticism partially lies in the WPI-based inflation not being of much consequence to him, against the more relevant rates derived from the consumer price index (CPI). And here, the numbers reveal a stark contrast. The WPI-based annual inflation during September averaged 3.39 per cent, but the corresponding CPI rates were far higher: 7.89 per cent for agricultural and rural labourers (AL), 6.4 per cent for industrial workers (IW) and 5.74 per cent for urban non-manual employees (UNME).

But isn’t this gap between WPI and CPI-based inflation rates a normal phenomenon? Well, not really so, if one looks back at the previous low inflation period of mid-2002.

In July 2002, the average WPI-based inflation stood at 2.79 per cent, while being 2.27 per cent for CPI-AL, 3.89 per cent for CPI-IW and 3.84 per cent for CPI-UNME. In other words, unlike the present ‘low’ inflation episode, there was very little divergence between WPI and CPI-linked rates.

Then and now

The divergence is, in turn, reflective of the nature of inflation (or lack of it) then and now. The recent inflationary episode has been largely food articles-driven. While the annual WPI increase for the week ended October 27 may have been only 2.97 per cent, hidden within this overall number are corresponding rates of 11.68 per cent for edible oils, 12.92 per cent for tomatoes, 8.27 per cent for milk, 6.10 per cent for rice and, well, 108.76 per cent for onions!

The huge price jump in these essential commodities is better captured in the CPIs, where the combined weight of food is 40 per cent for UNME and 46 per cent for IW. On the other hand, food articles (edible oils and sugar included) have a weight of just 22 per cent in the WPI, which is why the apparently low headline inflation is yet to be ‘felt’ by the consumer.

The difference between mid-2002 and now is that the former period coincided with burgeoning public grain inventories and downward pressure on domestic as well as global commodity prices. As a result, not only was inflation officially low, but the consumer, too, shared this perception.

http://www.incometaxindia.gov.in/ItInformation/CostInflation.asp

1981-82	100
1982-83	109
1983-84	116
1984-85	125
1985-86	133
1986-87	140
1987-88	150
1988-89	161
1989-90	172
1990-91	182
1991-92	199
1992-93	223
1993-94	244
1994-95	259
1995-96	281
1996-97	305
1997-98	331
1998-99	351
1999-2000	389
2000-2001	406
2001-2002	426
2002-2003	447
2003-2004	463
2004-2005	480

http://www.india-briefing.com/article/2007/08/24/notified-cost-inflation-index-financial-year-2007-08.html
“27 2007-08 551” is notified as inflation index for current year.

The economist
http://www.economist.com/world/asia/displaystory.cfm?story_id=8708934

et the government's most pressing need, with GDP growth of 9.2% predicted this financial year (ending in March), is to prevent India's economy from boiling over. Prices for manufactured goods, which have a weighting of 64% in the WPI, have risen by 6.6% over the past year. Wages are soaring; property and financial-asset prices are bubbling; lending to companies and households has increased by 30% in the past 12 months. The RBI will probably have to tighten monetary policy further, in an effort to cool the overheating economy. To sustain India's current growth, longer-term measures, such as cutting subsidies and building infrastructure, are needed. But that takes too long to win the votes of those whose most pressing concern is the price of onions.
http://www.thehindubusinessline.com/2007/02/08/stories/2007020800500800.htm
Are inflation figures the least trusted official statistics?

R. VAIDYANATHAN

Inflation is the talk of the town, yet the government has no reliable data as the price indices either do not reflect the modern milieu or are available with too much lag to be of any real use. Citing the creation of a personal inflation calculator by Britain, R. VAIDYANATHAN suggests that India could consider outsourcing key statistical activity to generate reliable and timely numbers.

The various channels that are desperate for `breaking news' can think of conducting an interesting competition: Identifying the least trusted government statistics. Is it the number of people dead in a riot or the people below poverty line or the inflation figures? My guess is that inflation figures would be the runaway winner.

The Government's Economic Advisory Council, at its recent meeting, expressed concern about inflationary pressures in the economy but is confident that they can be contained and the growth rate maintained. But, without malice, one would like to ask the Government about the numbers over which it is concerned. Indeed, more than the rate of inflation, the unreliable and non-credible numbers are a major embarrassment to the Government and that should be a matter of concern. At the national level there are four consumer price index numbers. These are:

Consumer Price Index (CPI) for industrial workers (IW),

CPI for Agricultural Labourers (AL),

CPI for Rural Labourers (RL) and

CPI for Urban Non-Manual Employees (UNME).

While the first three are compiled by the Labour Bureau in the Ministry of Labour, the last one is released by the Central Statistical Organisation in the Ministry of Statistics and Programme Implementation.

The CPI numbers are considered only partial indices as each caters to a specific segment of the population with different base years. The State-level CPI numbers lack uniformity with the oldest base year being 1939 (for a segment of Bihar). Fewer than 20 States compile and construct an index. Only a few States have base year after 1981-82.

The major concerns and discussions stem from the Wholesale Price Index (WPI) compiled by the Office of the Economic Advisor (OEA) in the Ministry of Industry on a weekly basis based on the price quotations collected by official and non-official agencies.

The main issues pertain to the coverage and the index not reflecting the travails of the common man. For instance, the annual inflation (in WPI) increased to 6.11 per cent on January 20, higher than the previous week's annual rise of 5.95 per cent. The increase is primarily on account of higher prices of manufactured products and food articles — foodgrains (9.46 per cent year-on-year), pulses (21.39 per cent Y-o-Y), fruits (14.50 per cent Y-o-Y), and oilseeds (21.27 per cent Y-o-Y).

Thus, the prices of a plate of two idlis has gone up from Rs 4 to Rs 8 in the last year. That is, for a student or a worker living on his own, the price of just one item of consumption has gone up by 100 per cent. Check the grocery bills, you will see that urad dal now sells at Rs 75 a kg from Rs 40 last year. Easily perceptible is the significant gap between wholesale and retail and critical are the regional variations — for instance, wheat impacts people in the North more than in the South, and so on with different type of dals and oils.

Then, there is the question of weights given to the WPI components. For instance, food items have around 22 per cent weightage in the WPI but it is well-known that they constitute more than 60 per cent of the consumption basket of the vast majority of people. Hence, food inflation hurts hard the vulnerable segments but this is not fully reflected in the WPI, giving the government a cover.

Under-pricing

Another issue is the suppressing of price increase by government decisions or by subsidies, as in the case of kerosene, LPG, or electricity. Under-pricing facilitates short-term suppression but exacerbates the inflationary expectations. In the case of sugar, the government controls it through the PDS (public distribution system) prices and freesale sugar by monthly dispatch orders. Hence, there can be a temporary freeze on the prices of some items to minimise logical rise due to demand-supply mismatch.

The non-inclusion of a number of services in the WPI, especially when the sector constitutes nearly 60 per cent of the economy, distorts the real inflation picture. For instance, doctor's fees have gone up by more than 100 per cent in the last three years as also the cost of education and rentals. Yet, most of these are not reflected in the WPI. The Consumer Price Index for Industrial Workers covers those working in factories, mines, plantations, railways, public transport undertakings, electricity generation and distribution units, and ports. Given the significant changes in the employment scene, especially the larger role of the services sector, including information technology and IT-enabled services, it is only appropriate to completely redefine "industry" rather that hang on to the mid-19th century definition.

Inflation calculators

The UK statistical agency — Office for National Statistics (ONS) — has initiated a novel scheme by which British consumers will be able to work out their own inflation rate. The agency has put up on its web site a personal inflation calculator after insinuations became rife that the official index under-estimates inflation.

The tool will help individual households get an idea of their spending patterns and compare it with the average and the effect it has on current inflation. Interestingly, an ONS survey two years ago found that the inflation figures were among the least trusted official statistics, ranked fourth out of six key measures. The ONS's Chief Statistician argues that "inflation figures are based on the average spending patterns. But every household is different and the way the inflation affects a family depends on how they spend their money."

Will our government statistical agency be enthusiastic in attempting to create such a calculator? It may not be feasible given our state of affairs in collecting and disseminating data. For example, the scheme on Consumer Price Index for Industrial Workers was cleared by the Expenditure Finance Committee in 1996 but the work on it could start only in 1999 due to various administrative reasons (see report of the National Statistical Commission, page 332).

One possibility is to consider outsourcing this important statistical activity to private agencies such as CMIE, with the Government Statistical Commission monitoring the accuracy of the data collection, the sample frame and the interpretation.

The entire process of collection, analysis and dissemination can be left to private agencies. This would result in a more reliable and timely inflation indicator. This should be the first step in gradually outsourcing most of the data series and downsizing the government statistical set-up, particularly at the State-level, which is mostly ineffective.

Else, such important numbers as inflation will not be reliable and the wide divergence between government figures and reality can become politically explosive. The next time a minister wants to know about inflation, the suggestion is just visit the nearest Mahila Mandal on an afternoon. The politician will not only understand real inflation but also learn the method to compute it. That is called practical education.

(The author is Professor of Finance and Control, Indian Institute of Management-Bangalore, and can be contacted at vaidya@iimb.ernet.in. The views are personal and do not reflect that of his organisation.)