Hong Kong, China (1)

Title of the survey

Survey of Employment, Vacancies and Payroll The survey is composed of three different enquiries:

the Survey of Employment, Vacancies and Payroll (SEVP) in the industrial, services and distribution blocks of the private sector,
the Survey of Employment and Vacancies in the Civil Service,
the Survey of Employment and Vacancies at Construction Sites (the Site Survey).

Organization responsible

Census and Statistics Department

Periodicity of the survey

Quarterly.

Objectives of the survey

To measure the number of persons engaged and existing vacancies, broken down by industry and sex. The results provide indicators of short-term and structural changes in labour market; and To measure the changes in per capita payroll at industry level. These data provide information on changes in employee earnings as a result of changes in wage rates and in amount of work available.

Main labour topics covered by the survey

Employment, vacancies and payroll in the industrial, services and distribution sectors.
Employment and vacancies in the civil service and construction sites.

Reference period

In the industrial, services and distribution sectors, statistics of employment and vacancies refer to the last full working day of March, June, September and December, while statistics of payroll refer to the same full months. In the civil service, employment and vacancies refer to the first day of January, April, July and October. A different reference date is adopted for different rounds of the Site Survey. For June 1993, for instance, the survey reference date was 29.06.1993.

Coverage of the survey

Geographical

The whole territory.

Industrial

In the private sector, the survey covers the following major economic activities:

Industrial: major divisions 2 (mining and quarrying), 3 (manufacturing) and 4 (electricity, gas and water);
Construction: major division 5 - construction sites only;
Distribution: major division 6 (wholesale, retail and import/export trades, restaurants and hotels);
Services: major divisions 7 (transport, storage and communication), 8 (financing, insurance, real estate and business services), and 9 (community, social and personal services).

The following economic activities are not covered in the surveys:

agriculture, hunting, forestry and fishing,
construction (other than construction sites),
hawkers, street vendors and retail pitches (other than market stalls),
taxis, public light buses, marine cargo handling services and tow boats,
veterinary services, religious organizations, authors and other independent artists, domestic servants, and miscellaneous recreational and personal services.

In the civil service, the survey covers all government departments, except the armed forces. The Site Survey covers all private sector sites registered with the Buildings Ordinance Office of Buildings and Lands Department, and all public sector sites under the charge of the Works Branch and Housing Department. Sites under the control of the Mass Transit Railway Corporation and Kowloon-Canton Railway Corporation and all public sites under the charge of the various other Government Departments are also covered.

Establishments

Establishments of all types and sizes in the public and private sectors, except in construction. In construction, construction projects for village-type houses in the New Territories and minor alterations, repairs, maintenance and interior decoration of existing buildings are excluded.

Persons

All persons engaged, aged 15 years and over (legal working age), in the public and private sectors, except in construction. In construction: manual workers only.

Occupations

Not relevant.

Concepts and definitions

Employment

In the non-construction private sector, persons engaged include working proprietors and active business partners; all full-time employees paid by the establishment on the reference date, both permanent and temporary, who are either at work or temporarily absent from work (i.e. on sick leave, casual leave or paid vacation); and part-time employees and unpaid family workers working more than three hours on the reference date. Also included are seasonal and casual workers, and persons temporarily absent from work because of industrial dispute or any other reason, whether authorised or not. Excluded from persons engaged are home workers, persons on temporary or indefinite lay off and persons on indefinite leave. Data on the total number of persons engaged are collected by sex; and in the industrial block, the distinction is made between operatives and other persons engaged. Operatives in the industrial sector refer to all manual workers directly engaged in fabricating, processing, assembling and other activities directly auxiliary to these.
In the civil service, persons engaged include all direct employees paid under the personnel emolument vote on the reference date. Data are collected by sex from each government department.
At construction sites, the definition of persons engaged is limited to manual workers, i.e. persons either directly employed by the main contractor, or being called upon by sub-contractors or gang leaders to work in the construction site for more than three hours on the reference date. They include foremen, technicians, craftsmen, semi-skilled and unskilled workers. They may be permanent, temporary, seasonal or casual, full-time or part-time workers, and include apprentices, trainees and workers on probation. Excluded are professional and administrative employees such as architects, engineers, surveyors, contract managers, site agents, clerks of works and general clerical staff. Also excluded are persons temporarily absent from work for any reason, whether authorized or not, and whether paid or unpaid. Employment data at construction sites are collected by sex.

Reported vacancies refer to those unfilled, immediately available job openings reported by the establishment, for which active steps are being made to try to recruit personnel on the reference date. At construction sites, reported vacancies include only job openings for manual workers. Number of vacancies is collected by sex, and by operatives and other persons engaged in the industrial block.

Earnings

Payroll data are collected in the industrial, services and distribution divisions of the private sector and refer to aggregate cash payments made to employees. Payroll includes wages and salaries, overtime pay, shift allowance, attendance and efficiency bonuses, cost-of-living or dearness allowance, food and transportation allowance, year-end and seasonal bonuses, and pay in lieu of leave, retirement gratuities, severance and termination pay, back-pay, etc. Employers' contributions to social security and pension schemes, as well as benefits received by employees under these schemes are excluded. Profits earned by proprietors and business partners, pensions, payments to outworkers and the value of earnings in kind are also excluded. Only the total payroll for each month of the reference quarter is collected, separately for (a) earnings paid regularly in cash and (b) irregularly paid bonuses and gratuities in cash.

Wage/salary rates

Not relevant.

Hours of work

Not relevant.

International recommendations

Payroll data collected in this survey correspond to the international concept of earnings with the following major exceptions: (a) the monetary value of earnings in kind is excluded; (b) irregularly paid bonuses and gratuities, which include severance and termination pay, retirement gratuities and back-pay, are included. In addition, the following remarks apply to the estimates of payroll per person engaged computed from the survey results: payroll per person engaged should not be interpreted as average earnings of employees. First, the time period covered by employment figures (a specified reference date) differs from that covered by monthly payroll figures (aggregate data for the whole month). Secondly, payroll refers to payments made to employees, whereas persons engaged also includes persons who normally are not on the payroll (working proprietors, business partners and unpaid family workers).

Classifications

Industrial

Data on employment, vacancies and payroll are classified according to the Hong Kong Standard Industrial Classification (HSIC) and selected industries are coded at the four-digit level of this classification. The HSIC can be linked to the International Standard Industrial Classification of all economic activities (ISIC), Rev.2, 1968 at the four-digit level, with some slight modifications.

Occupational

Not relevant.

Others

Employment data are classified by sex, employment size, geographic district and by category of persons engaged (paid and unpaid employees, and operatives and other persons in the non-construction private sector). Payroll data are classified by aggregate group of components, i.e. by regular and irregular payroll.

Sample size and design

Statistical unit

Except for construction, the unit of inquiry in the private sector is the establishment, which is defined as an economic unit which engages, under a single ownership or control, in one or predominantly one kind of economic activity at a single physical location (e.g. an individual factory, workshop, retail shop, office). In the civil service, the unit of inquiry is a government department. In the Site Survey, the unit of inquiry is a construction site, which refers to a demarcated locality where one or more stages of construction work are being carried on.

Survey universe / sample frame

The sample frame for the SEVP consists of the Central Register of Establishments kept by the Census and Statistics Department. It covers all non-agricultural establishments engaged in economic activities classified under HSIC major divisions 1 to 4 and 6 to 9, but excludes major division 5 (construction). Some 480,000 establishments are covered, out of which some 338,000 establishments are active. The Register is updated each quarter on the basis of data from the Business Registration Office, and on a yearly basis, from the Lists of Markets from the following leasors: the Housing Authority, the Urban Council, the Regional Council and the Housing Society. For records which are new to the Register, establishment particulars are collected by postal, then field enumeration method. New insertions and amendments are updated on a quarterly basis. Industry updating is also undertaken to collect up-to-date information on business activities of each establishment. Industry updating for a particular block is scheduled in its full enumeration quarter on every alternate year. Since June 1991, the results of the industry updating are added to the sampling frame on a continuous basis. In the Site Survey, the sample frame is the Register of construction sites, which covers some 2,000 units. This register is updated on a quarterly basis, with information obtained from the following departments: Buildings and Lands Department, Architectural Services Department, Civil Engineering Department, Drainage Services Department, Highways Department, Territory Development Department, Water Supplies Department, Housing Department, Mass Transit Railway Corporation, Kowloon-Canton Railway Corporation and Provisional Airport Authority.

Sample design

Construction sites and Government Departments are covered in full every quarter. Surveys in the industrial, services and distribution blocks in the private sector are organized in annual cycles with a full count in one quarter and sample surveys in the following three quarters. Because of resource constraints, the timing of full counts and sample surveys is staggered so that each block is covered in full in one of the four quarters. The sample survey is based on a two-stage stratified sample. In the first stage, a main sample is drawn with stratification by industry and employment size. The sample size for each stratum is determined by first determining the sample size for each industry group. The method of optimum allocation is used, aiming at a coefficient of variation for employment statistics of one per cent at the three-digit industry level and two per cent at selected four-digit industry. Then within each industry group, the sample size is determined for each employment size group, also using the optimum allocation method. Establishments engaging more than 100 persons are enumerated in full, instead of using the sample sizes determined. Sample selection in each stratum is done by systematic sampling and decimal selection interval method is adopted in determining the selection interval. The main sample covers some 40,000 units on average and provides data on employment and vacancies. In the second stage, a subsample is drawn from the main sample, from which establishments provide information on monthly payroll in addition to those on employment and vacancies. The sample size for each stratum is determined by proportionate allocation and systematic sampling with decimal selection is used. The subsample covers some 10.000 units. Altogether, there are some 300 four-digit industry groups by seven employment size classes. Once selected, the sample is used for three consecutive quarters for the first stage, and a full enumeration is carried out on the fourth quarter. The subsample is used for a year and a new sample is drawn when the first stage sample is renewed. New openings made known to the Census and Statistics Department during the survey reference quarter are enumerated in full for data on employment and vacancies; supplementary samples are drawn in subsequent quarters to account for new openings after the main sample is selected. The whole sample is completely renewed each year.

Field work

Data collection

A permanent survey organization exists, and temporary staff are also recruited to assist in data collection in each round, where relevant. For the industrial, services and distribution blocks, the survey is carried out mainly by means of postal inquiry. Data collection starts usually some 15 days after the reference date. Respondents are required to complete and return the forms within a specified date, after which Computer Assisted Telephone Interviews (CATI) and field visits are made to collect data from non-reporting establishments. For public sector construction sites, employment statistics are compiled from their returns forwarded regularly to the Census and Statistics Department, whereas the number of vacancies are obtained through a special survey. Data for private sector sites are collected by field officers who have direct interviews with the responsible site agents of the main contractor or subcontractor of each site. Regular employment and vacancy returns are collected from various Government Departments and bodies, which are then consolidated to produce employment and vacancy statistics in the Civil Service sector.

Survey questionnaire

In the non-construction private sector, the survey questionnaire consists of three or four sections, depending on whether payroll data are collected or not:

Section A provides for the identification of the selected establishment,
Section B is designed to collect data on the number of persons engaged and vacancies, by sex, and where relevant, by category of employees,
Section C (in the questionnaire addressed to the subsample of establishments) is designed to collect data on aggregate regular and irregular payroll outlays, and
the last section provides for remarks concerning significant changes.

In the Site Survey, the questionnaire consists of a single sheet, designed to collect data on:

identification of the site,
number of manual workers,
number of vacancies,
current status of site (active, work suspended, vacant, completed, with or without maintenance period).

Instructions and explanatory notes are provided along with the questionnaires, and a training manual is provided to enumerators.

Substitution of sampling units

Not relevant.

Data processing and editing

Data are processed by computer and coding is done both by computer and manually. Completed questionnaires are put through various editing stages for range and consistency checks, and doubtful returns are verified by field officers.

Types of estimates

Totals of persons engaged and reported vacancies,
Total monthly payroll and average, nominal and real, payroll per person engaged,
Vacancy rate.

Construction of indices

Indices of total monthly payroll and persons engaged referring to a particular month for an industry group are obtained by dividing the amount of payroll or the number of persons engaged in that month by the base period (at present March 1991=100).
The nominal index of payroll per person engaged is obtained by dividing the index of total monthly payroll by the index of persons engaged.
The real index of payroll per person engaged is derived by deflating the nominal index of payroll per person engaged with the "all-items" CPI-(A).

Weighting of sample results

In the sample survey in the non-construction private sector, grossing-up factors equal to the selection intervals are used to estimate the number of active establishments, the number of persons engaged and existing vacancies. Due to the limited size of the subsample, payroll is estimated for each three-digit industry and the ratio estimation is used, with the number of persons engaged as an auxiliary variable.

Adjustments

Non-response

Missing data are imputed. If data for the previous quarter are available, data for the current quarter are imputed by multiplying the stratum growth by the reported data for the previous quarter. If data for the previous quarter are not available, employment and vacancy data are imputed based on the average of reporting establishments in the stratum concerned in the current quarter.

Other bias

Adjustments due to changes in the number of establishments by industry are taken into account during the grossing-up process. In addition, an Industry Updating Survey is conducted in alternate years during the full scale survey, to update the industrial classification.

Use of benchmark data

Not relevant.

Seasonal variations

None.

Indicators of reliability of the estimates

Coverage of the sampling frame

The current sampling frames cover the whole target population, i.e. non-agricultural establishments engaged in economic activities.

Sampling error / sampling variance

Not available.

Non-response rate

In the sample survey in the non-construction private sector, the current non-response or non-contact rate is around five per cent, representing some 2,000 units. In the Site Survey which is a full enumeration, the current non-response rate is less than one per cent, i.e. about 10 to 15 sampling units.

Non-sampling errors

Two types of non-sampling errors may occur in the Site Survey, which concern data collection on vacancies in construction:

Although the definition of vacancies is consistent with those adopted in other surveys, it may not be so appropriate for construction sites and there are substantial substitution possibilities in the employment of workers at construction sites. Consequently, vacancy data supplied by contractors and subcontractors are likely to be subject to relatively large variations in reporting behaviour.
Normally, vacancy data should be collected directly from subcontractors who have first-hand information on the site vacancy situation. However, it is difficult to contact these persons and data are mainly obtained from the general contractors, who may not have accurate information on the number of vacancies in their sites.

(See also the remark under International recommendations, which applies to the estimates of payroll per person engaged.)

Conformity with other sources

Employment data by major division of economic activity are compared with those obtained from the quarterly General Household Survey and the quarterly Business Survey.

Available series

The following time series and quarterly tables are regularly prepared from the SEVP in the private sector:

Number of persons engaged and reported vacancies,
Number of establishments in the private sector,
Indices of employment, total monthly payroll (nominal) and payroll per person engaged (nominal and real) in the private sector,

where the private sector is analysed by three- and four-digit industry group. From the Site Survey:

Number of active construction sites, manual workers engaged (by sex), vacancies and vacancy rate by sector, industry (type of project), employment size and geographic district;
Distribution of the changes in vacancy rate between two quarters, by sector.

History of the survey

An integrated approach to collect data on employment, vacancies and payroll was introduced in April 1980. It replaced the separate surveys on employment and vacancies which had been carried out in various economic sectors. It incorporated the quarterly Employment Survey of Construction Sites (Site Survey) which existed since March 1976, and the Survey of Employment and Vacancies in the Civil Service. A new classification scheme of types of project in construction was introduced in June 1982. As from December 1987, a revised classification scheme was introduced to classify public and private construction projects, and the number of manual workers engaged at sites excludes those watchmen engaged at inactive worksites. In 1988, the questionnaire was modified to collect additional information on the number of vacancies of manual workers. Prior to the first quarter of 1991, breakdowns of employment and vacancy statistics by industry were published according to ISIC. Since March 1991, these statistics are classified according to HSIC. As a consequence, detailed breakdowns in some cases are not directly comparable. To facilitate comparison with the previous year, statistics for the relevant quarters of 1990 were recompiled using the HSIC, by applying conversion factors created by double coding the Central Register of Establishments for the four quarters concerned. Prior to the June 1991 round of the SEVP, results of industry updating (see under Sampling frame) were incorporated into the sampling frame and used to compile statistics at a time lag of one year. Since June 1991, this updating is done on a continuous basis, so that statistics are compiled based on the most updated information in the sampling frame. Since March 1994, the collection of payroll statistics is transferred to the Half-yearly Survey of Wages, Salaries and Employee Benefits. The latter is subsequently renamed "Labour Earnings Survey".

Documentation

Census and Statistics Department: Quarterly Report of Employment, Vacancies and Payroll Statistics (Hong Kong); covers the Civil Service and private sector, and contains summary statistics in the form of time series and quarterly tables for all economic sectors. idem: Employment and Vacancies Statistics - Detailed Tables (quarterly, ibid.); contains more detailed statistics for the particular economic blocks enumerated in full once a year. idem: Quarterly Report of Employment and Vacancies at Construction Sites (quarterly, ibid.). All publications contain some methodological information. Additional detailed statistics are available on request, which include:

number of establishments and persons engaged in the private sector, analyzed by one-digit industry sector and by employment size, and total monthly payroll in the private sector analyzed by three-digit industry group, derived from sample surveys;
number of establishments and persons engaged in the private sector, analyzed by four-digit industry group and employment size, geographic district and tertiary planning unit, from full-scale surveys.
number of active construction sites, manual workers engaged and vacancies, analyzed by geographic district.

Confidentiality / Reliability criteria

The SEVP in the private sector operates under the Census and Statistics Ordinance which prohibits unauthorized disclosure of information collected from respondents. Appropriate measures are taken at various stages of data collection and data compilation to avoid leakage of confidential data. Sensitive data which might lead to unauthorized divulgence of information provided by individual establishments are suppressed from publication.

Other information

Data supplied to the ILO for publication

The following data are published in the Yearbook of Labour Statistics: average employment (all persons engaged) by industry and by sex, in Tables 3B, 4 and 5B. Quarterly statistics of employment (all persons engaged) in non-agricultural activities and manufacturing are published in Tables 2 and 3 of the Bulletin of Labour Statistics.