Spending forms a pattern
The Household Expenditure Survey (HES) is conducted once every five years in order to collect up-to-date information on the expenditure patterns of households DAVID and his family were invited by the Census and Statistics Department to take part in the 1994-95 Household Expenditure Survey (HES).
He knew that his uncle, Tom, took part in the same survey a few weeks ago. David met Tom and they started discussing the survey.
David: Tom, I have received a letter from the Census and Statistics Department inviting my family to participate in the 1994-95 HES. It is mentioned that the information collected is important for the compilation of the Consumer Price Index (CPI).
Tom: My family took part in the survey a few weeks ago. I have also got a copy of the report of the last HES.
David: Oh, really? Do you know why the HES is conducted? Tom: As I know, the HES collected detailed information on the expenditure of different categories of households. Such information is required for compiling the CPI, and this round of the HES will produce the latest expenditure information which will be used to keep the CPI reliable and up-to-date.
David: I am not sure whether I understand this completely. Please explain.
Tom: For compiling the CPI, two types of information are required. First, data on the price movements of goods and service items which households purchase.
Second, a weighting system which represents the relative importance of the goods and service items. Information to derive the weighting system is obtained from the HES, while information on price movements is collected from a continuous pricing survey.
As consumers spend more on some items and less on others, similar price movements of different items may have a different impact on the overall price change. Therefore, weighting of each item, which represents the importance of the item in the total household expenditure, is essential information for the compilation of the CPI.
In Hong Kong, the HES is conducted once every five years in order to collect up-to-date information on the expenditure patterns of households, and hence the weighting system used for compiling the CPI.
The 1994-95 HES is being conducted in the period from October last year to September this year.
David: Once every five years! Is it adequate? Tom: Expenditure level of households generally increases over time in response to factors such as changes in income levels and prices. However, only the household expenditure pattern, but not the expenditure level, is used in the compilation of the CPI.
Hence, changes in expenditure level should not affect the reliability of the CPI as long as the expenditure pattern does not undergo substantial changes.
David: Can you explain a little bit more on what is meant by 'expenditure pattern'? Tom: The household expenditure pattern is the relative expenditure by the household among various items. This is measured by the percentage of the total expenditure spent on each item of goods and services. The expenditure pattern is actually the weighting system I have explained earlier.
It may be noted that the expenditure pattern only changes gradually. However, their cumulated changes over a period of several years, whether substantial or not, should be taken into account in order to ensure that the CPI is reliable and up-to-date.
Therefore, a new round of HES is conducted once every five years in Hong Kong.
David: Why will the survey last for one whole year? Tom: Spending patterns do vary in different seasons of the year. For example, different kinds of food and clothing are purchased by consumers in the summer and winter.
It is, therefore, necessary that the survey covers an entire year in order that seasonal variations in spending are fully taken into account.
David: Tom, do you know how many households will be interviewed in the survey? Tom: The one-year survey period is broken down into 26 cycles and each cycle comprises 14 days. For each cycle, households spreading out in all areas in Hong Kong will be selected and invited to take part.
This will ensure that the results of the survey are representative. A total of some 9,000 households will be involved.
David: What kind of data were you required to provide? Tom: I remember that the questionnaire looked like an expenditure diary. We were requested to keep records of all expenses in a specified period of 14 days.
David: Were you required to record household expenditures only or personal expenditures as well? Tom: All expenditures spent by each household member, whether for the member himself/herself or for the household as a whole were recorded.
David: Were the kids required to record their expenditures as well? Tom: Money spent by children was also recorded. For children less than 12 years old, other household members would supply data on their behalf.
David: I think the 14-day period is adequate to collect information on regular expenditure. But how about infrequent expenditure? Will they be omitted? Tom: They will not be omitted. Infrequent purchases made in a certain reference period reported by households surveyed for that reference period will represent the infrequent purchases made by all households in that period, including those households not actually surveyed. This is accomplished by a scientific survey design in accordance with statistical sampling theory.
For more information on this series of articles, please write to the General Statistics Branch (2) of the Census and Statistics Department at Wanchai Tower, 12 Harbour Road, Wan Chai, Hong Kong or call 2582-4004