Letter to the Editor and a Reply
Dear Editor,
(i) How is the Consumer Price Index for Industrial Workers (CPI-IW) calculated? Similarly, what is the Wholesale Price Index for "all commodities". It is said to be 100 in 1981-82. We need to know about inflation -- how it actually affects people.
(ii) What is the `primary market' -- it is commonly mentioned in relation to the share market?
-- Harbag Singh,
Faridkot.
Thank you for your letter. In regard to your questions:
Different measures of inflation ought to be used depending on one’s purpose. Actually, the Government deliberately creates some confusion on this score.
All price indices use a particular year as a "base year". That means that rises or falls in prices are measured with reference to the price in that year. For example, the base year used for the Wholesale Price Index is 1981-82. Wholesale prices of all products in that year are treated as "100". If, in 1981-82, the wholesale price of gur was Rs 2 a kg, and rose by 50 paise the following year, it would mean that the wholesale price index for gur would rise to 125 in 1982-83. (The corresponding method is used for prices for years before 1981-82, too; the index numbers would probably be less than 100 in the previous year, and so on for each previous year, since inflation is a permanent feature of "free" economies.) Different base years are used for different price indices, so one should not be confused by the figures in one index being higher than the figures in another index, for the same year (you can see how an index with the base year 1960-61 would have figures far higher today than one with the base year 1981-82); what matters is the growth rates.
The WPI: The Wholesale Price Index (WPI) covers all commodities: ie, primary goods, power/fuel, and manufactured goods. Some primary goods (agricultural, plantation, and mineral output) are bought by consumers (eg cereals), but many are not (eg. minerals). Of the manufactured goods, intermediates (such as steel and cement) are only bought by producers, to be used in the course of production. And then there are consumer goods. But, for all of these, the WPI takes the wholesale prices. In all, the WPI tracks the prices of 447 commodities.
Each of these is given a different weightage in the index, according to its importance. Imagine that the price of washing machines has remained at the same level for two years running (let us say the index of washing machine prices has remained at 260), but the index of edible oil prices has risen from 230 to 290. Naturally, since edible oil plays a much larger role in the economy than washing machines, when calculating the overall price index (on the basis of the indices of increase in each individual commodity), the weightage given to the index for edible oil prices will be greater. Hence the stability in washing machine prices will not have much effect on the index, whereas the rise in edible oil prices will.
The WPI is not intended to capture the effect of price rise on the consumer. So one should not be surprised when one finds that the weightage to foodgrains in the WPI is less than 8 (out of 100), and indeed to all food items only about 17; whereas, of course, food makes up more than half the expenditure of the average consumer. Onions have a weight of only 0.16 per cent, and potatoes 0.47 per cent. For that reason, one needs a separate consumer price index.
The CPI: There are in fact several Consumer Price Indices. Each tracks the retail prices of goods and services for a different group of people, because the consumption patterns of different groups differ. For Industrial Workers (CPI-IW), a basket of 260 commodities is tracked; for Urban Non-Manual Employees (CPI-UNME), 180 commodities; for Agricultural Labourers (CPI-AL), just 60 commodities. Here again, each commodity is given a different weightage, which differs from group index to group index. For example, the CPI-AL would give a greater weightage to foodgrains than the CPI-UNME, since a greater proportion of the agricultural labourer’s expenditure would go toward foodgrains, and he would be unlikely to buy the sort of items the office-goer would buy.
These baskets and the weightages to each item have been determined on the basis of some surveys of consumption patterns. (It is reported that these surveys are very outdated, so they do not give a proper picture of consumption patterns — and therefore the CPI data do not give a proper picture of price rise.)
Further, the information about prices is supposed to be collected according to the places from which each section would purchase. Information also differs, naturally, from centre to centre around the country; the all-India figures declared are merely averages.
The WPI is useful in certain contexts. For example, if one were working out the costs of setting up a factory over the course of several years, and further wished to calculate the costs of production and returns over several years, one would obviously not use the CPI. First, the basket of items in the CPI does not include machinery, chemicals, and so on; secondly, the price of electricity in the CPI would be the consumer tariffs, not the industrial tariffs; and so on. Here, obviously, one would turn to the WPI. (There is, further, a capital goods price index, which tracks specifically the prices of capital goods; this one uses in order to calculate the real growth in capital expenditure. And there is also a GDP deflator — an index of prices of every good and commodity included in the Gross Domestic Product, which is pretty close to the WPI. We use the GDP deflator to calculate the real GDP growth.)
Normally, however, when we refer to the rate of inflation in the economy, we mean the rise in retail prices, which is what directly affects people. In such a case, the correct measure would be the CPI for various sections. The consumer price index, in fact, is the measure normally used abroad when referring to inflation.
However, the Indian government deliberately uses the WPI, because the figures for inflation in the WPI are on the average much lower than those in the CPI indices. There could be two reasons for this difference in rates between the WPI and CPI: first, prices of the items in the CPI basket might have risen more sharply than items excluded from it — this would mean that prices of mass consumption goods have risen more sharply than inputs for production; secondly, the retail prices of commodities might have grown more sharply than the wholesale prices, indicating that middlemen have taken a bigger share.
A source of confusion in understanding inflation data is the Government’s use of point-to-point figures for inflation. That is, if bananas were Rs 10 a dozen on July 1 last year and Rs 12 a dozen on July 1 this year, the price rise is given as 20 per cent. However, it might be that bananas were Rs 14 for most of this year, and have only fallen to Rs 12 on July 1; so the rate of inflation for the whole year is not really captured by the point-to-point method. Or it might be that the price last July 1 might have been unusually high for that year (it might have been just Rs 8 in June and August 1997). In that case the real price rise over the past year would not be captured by showing the increase of July 1, 1998, over July 1, 1997. A better measure would employ averages (eg. averages of three months over the previous year’s corresponding three months; or averages of point-to-point increases for all the months of the preceding year; etc.) The use of point-to-point figures is one of the reasons why the inflation rate suddenly "falls" without any prices falling: the reason may be that there was steep price rise just preceding the corresponding date of the earlier year, on top of which the latest date’s prices have not grown much. At the very least, the Government should calculate inflation rates by both methods, and give readers the choice.
One more source of confusion is that the figures of inflation which we normally read in the press are only provisional; they are later revised, though the revised figures do not get blared in the headlines, and instead lie buried in some documents. The revised figures for inflation are generally higher than the provisional ones. (In fact, the Government’s provisional calculations of inflation follow a questionable method: they take the provisional price index figure of today over the revised index figure of a year ago. As it turns out, if one instead takes the provisional index figure of today over the provisional index figure of a year ago, the inflation rate calculated therefrom is pretty close to the final, revised one.)
Now let us look at the news items of July 20, 1998, which are based on the official hand-outs. They tell us that "inflation touched a 71-week high of 7.59 per cent for the week ended July 4." But these are provisional figures, which are usually lower: the same item also tells us that the final figures are now available for the week ended May 9. We find that the final WPI inflation figure for that week is 6.6 per cent, whereas the provisional figure was only 6.1 per cent.
More importantly, the figure for inflation in the CPI-IW is now available for the month of May. This turns out to be 10.5 per cent — compared to a WPI rate of 6.4 per cent in the corresponding period. Note that the gap between the WPI and the CPI is more than four percentage points; and in fact, throughout 1996, 1997, and the first quarter of 1998, the gap has been of this order. (For example, the WPI point-to-point inflation for 1996-97 was 6.9 per cent, but CPI-IW was 10 per cent; CPI-UNME was 10.2 per cent; and CPI-AL was 10.5 per cent.) So we would be safe in assuming that when the CPI-IW inflation figures for July 1998 are finally calculated, they will show an inflation rate of around 11.5 to 12 percent. (This letter was written in July 1998. As it turns out, our figure was a gross underestimate. In September 1998 it was revealed that CPI-IW inflation figures for July 1998 were 14.8 per cent! — Editor, Aspects)
In fact, the great talk last year of inflation having been brought under control, or down to five per cent or even four per cent, was just a grand hoax of using point-to-point WPI figures. Inflation in the CPI-IW (average of the months of each year) was as follows since the beginning of the IMF reforms:
1991-92 — 13.5%
1992-93 — 9.6
1993-94 — 7.5
1994-95 — 8.1
1995-96 — 12.2
1996-97 — 9.3
1997-98 — over 9
and now running riot!
Thus the essential point is: If you want a figure for the current inflation rate, don’t look for the WPI figures, but look for the latest CPI-IW figures available; then look for the gap, if any, between CPI-IW and WPI for that date; and make a guess, on that basis, of the current CPI-IW rate.
Even the Consumer Price Index may not reflect the real rate of inflation, as experienced by common people, for several reasons. First, the basket of commodities that is chosen for the purpose of measuring price rise may not reflect actual consumption patterns — either the commodities may be different, or the weightage given to each commodity may be different. Secondly, the reporting of prices from various centres may be understated.
A third reason is one that is particularly relevant now, when there is rapid price rise in essential commodities: During such a period, despite the higher prices, people cannot greatly reduce their consumption of such commodities (the term used is that "demand is inelastic in relation to price changes"); consequently, in their total basket of consumption, common people cut back more on other commodities. In that case, even if the consumption basket for CPI were initially correct, it would get rapidly outdated during such a period. Let us assume that food items have a weight of 60 out of 100 in a consumer price index; when prices of food rise rapidly, food items might actually consume 70 per cent of the expenditure of that section of consumers. But the inflation rate would still be calculated as if they only made up 60 per cent, and hence the official rate would understate the real consumer price inflation. This is one reason why people feel the official rates have little to do with the real rate they experience, and the impact of it on their lives. (Another reason is, of course, the fact that most common people are already having difficulty making ends meet. In such circumstances even a `moderate’ price rise can push them over the edge into dire poverty.)
The IMF, World Bank and government have for some time been propounding the theory that price rise results from too much demand in the economy. Their answer therefore is to suppress demand by raising interest rates, whereupon companies will have to reduce their borrowing and therefore their purchases and investments, and will also have to squeeze the amount they spend on wages. Once companies cut their spending, and people cut their spending, demand will go down and prices will accordingly fall.
How absurd this theory is becomes clear when you look at the current runaway price rise: there is a recession in industry, demand in the rural areas too is depressed because of poor agricultural performance, and yet prices are soaring. The real causes of price rise are inherent to the structure of our economy: a backward agriculture dominated by parasitical forces, where periodic shortfalls in production are bound to occur; a trade sector in the grip of big hoarder-speculators with access to a vast pool of black money, who can capture the supply and manipulate prices upward; continuing dependence on imports of foreign goods and inflows of foreign capital, both of which render the economy vulnerable to price increases by foreign interests and the rupee vulnerable to devaluation; in many industrial commodities, a handful of big industries dominating the market, and hence easily colluding to fix prices; and so on. The governments of various hues occupying the seat of office at the Centre and in the states cannot control price rise because that would mean acting against those very sections that have brought them to power.
*
You also asked about the meaning of "primary" and "secondary" markets. When a company issues shares, it does so for a specific amount. Investors can apply for these shares. Depending on how many people apply, and for how much, the would-be investor may or may not get allocated some shares. Sometimes, if an issue is "oversubscribed", the company may decide to increase the size of the issue to accommodate the additional investors. All this activity is called the "primary market".
Later on, the shareholder can sell his/her shares to any other person. The shares of the major companies are bought and sold on the stockmarket every day. All this activity is called the "secondary market".
The same phrases are also used in the case of debt instruments such as "debentures" issued by companies or "Government securities" issued by the Reserve Bank. The initial issue is called the primary market, and the subsequent sale and re-sale is called the secondary market. These markets are not physically limited to a specific spot (the sale might take place between two persons talking on the telephone); the term is used to describe a type of activity.