Our faulty data collection

By Zubeida Mustafa
Source: Dawn

Get your facts first, and then you can distort them as much as you please. — Mark Twain I gather, young man, that you wish to be a Member of Parliament. The first lesson that you must learn is, when I call for statistics about the rate of infant mortality, what I want is proof that fewer babies died when I was prime minister than when anyone else was prime minister. That is a political statistic.

— Winston Churchill

Errors using inadequate data are much less than those using no data at all. — Charles Babbage

Anyone who cannot cope with mathematics is not fully human. At best he is a tolerable subhuman, who has learned to wear shoes, bathe, and not make messes in the house. — Robert Heinlein

If you read these quotations carefully you will know where we are going wrong in economics — all the boastful claims made about Pakistan’s GDP growth and the decline of poverty and so on. First, Mark Twain, the American satirist, conveys deep wisdom when he asks you to get the facts correct before distorting them.

At least our policymakers should know the correct position themselves before they try to fool the public otherwise they will get fooled themselves which can be dangerous. Their data may well be based on what Churchill calls a political statistic. Wouldn’t it be better to dispense with the number crunching altogether? No, says Babbage, the 19th century mathematician, who advises us to use statistics to support our argument even if inadequate.

But that is not easy because many products of our education system can’t differentiate between a billion and a million, and worse still between a million and a lakh. Irrespective of what traits Heinlein, the science fiction writer, finds in us, the fact is that most of us are not able to cope with mathematics.

Seen against this backdrop, one wonders how the government collects the data as some of the figures stated hardly seem to support people’s observations. Doesn’t wrong data distort the reality and impair the effectiveness of policies?

To begin with, culturally we don’t have a strong penchant for accuracy. It is common for us to be deliberately vague in our speech and writings when it comes to numbers. Crowds are estimated at random as “lakh, do lakh”. Even government publications give different figures for the same head in different chapters. That is because different sources are used. The Pakistan Economic Survey 2005-06 conveniently switches from one source to another and leaves the reader totally confounded.

How do we go wrong when we have so many organisations engaged in the task of collecting data, computing and compiling it? There is the Federal Bureau of Statistics, the provincial bureaus of statistics and agencies in various ministries concerned with collecting data in their specific field. The accuracy of the information collected and the figures released depend on a number of factors, and a mistake in any of them can distort results.

Take the case of the inflation rate which is invariably disputed every year it is announced. Based on the price index, it is determined by the items that are covered, the weightage given to each, the areas surveyed and, of course, the base period.

To determine inflation the prices of 374 items, that constitute the basket of goods/services, are collected on a monthly basis from 35 cities (71 markets) and computed on the basis of weightage determined by the family budget. Nearly 42,000 families are surveyed and they are supposedly drawn from four income groups and live in rural and urban areas. The family survey is held in the base year — in 2000-01 in the present case. A five-year period is considered to be ideal. But in Pakistan the period has been stretched to a decade since 1980. This logically detracts from the accuracy of the survey because ten years is a long period in which changes creep into a family’s lifestyle affecting its pattern of spending.

A look at the weightage given to various items calculated on the basis of family budgets six years ago appears quite unrealistic. The item “food and beverage” is highest on the list with a weightage of 40 per cent that is believable. But an expenditure of two per cent on medicine, 3.4 per cent on education and 7.3 per cent on transport appear to be too low. And these are the areas in which prices have risen the most ever since the government left it to the private sector to fill in the vacuum.

The family surveys are conducted ostensibly on scientific lines. Then where do the distortions come in? There is plenty of room for error which may not always be deliberate. Thus the rural areas are not surveyed for the prices because they are said not to have an organised market. The retail prices are taken to calculate the consumer price index and it is common knowledge that the fluctuations in the retail prices are more pronounced.

But the biggest distorting factor is the unwillingness of the people to participate in surveys. Be they individuals or organisations accuracy and authenticity are not their strong points. People, even educated ones, are not used to record keeping. How many of us record our daily household expenditure in detail as our mothers and grandmothers did? Those of us who go and do our shopping at the supermarket and pay by credit card or cash never note down each item individually.

Along with food items we also buy other household goods such as tissue paper, detergent and sometimes cosmetics and medicine which are lumped together in one bill. All we can tell the surveyor when he comes are broad figures which may be totally out. As for the poor who are illiterate and have a daily income, they would not know themselves how they spend their money.

Furthermore, people are most reluctant to disclose any information of a financial nature such is the lack of confidence in the government and its functionaries. No amount of reassurance that their confidentiality will be respected can persuade them to cooperate with the data collector. People are convinced that whatever information they provide to a stranger would be used against them. The business, trade and industry sectors are not transparent in their financial dealings and account keeping. They either refuse to give information or suppress part of it.

Some government departments maintain their own records of institutions that operate under them. These are based on paperwork. For instance, the education department will give you a list of schools that were entered in the registers when these institutions were set up. But they no longer exist. There is no record of that and they constitute the ghosts schools that have closed down long ago without the parent body even being aware of it.

That would explain why the inflation figure released by the government generally appears to be so ridiculously low. In this context, poverty also would not have fallen as much as the government would like us to believe. So all one can do is to keep one’s fingers crossed and make broad estimates. In this guessing game it would be better to err on the side of caution, rather than boast of a rosy picture that robs the government of its credibility.