Stats NZ has a new website.

For new releases go to

www.stats.govt.nz

As we transition to our new site, you'll still find some Stats NZ information here on this archive site.

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
Reweighting the labour cost index (salary and wage rates)

Information sources

The Labour Cost Index (LCI) wage and salary wage rate indexes have been reweighted to reflect changes in the industry and occupation structures of the labour market. The weights are calculated based on shares of expenditure (quantity times price) by employers on salaries and wages. This has been done using 2006 Census of Population and Dwellings information, Quarterly Employment Survey (QES) figures for the year to June 2008 and figures for the June 2008 quarter, Linked Employer-Employee Database (LEED) figures for the year to March 2006, Business Frame (BF) information for the year to June 2008, and surveyed pay rates from the Labour Cost Survey (LCS) for the June 2008 quarter. Specifically:

  1. The 2001 Census gave (i) the numbers of full-time and part-time paid employees at the detailed (ie five-digit) occupational level within industry groups and sectors of ownership, and (ii) the ratios of arithmetic mean hours worked by part-time employees to arithmetic mean hours worked by full-time employees at the industry group by sector of ownership level.
  2. The QES gave (i) annual ratios of full-time to part-time employees at the sector by industry level, (ii) ordinary time weekly earnings per job at the sector by industry group level, and (iii) ratios of ordinary time earnings to overtime earnings at the sector by industry group level.
  3. LEED gave an indication of the number of jobs filled by paid employees under 15 years of age for the March 2006 month at the sector by industry group level.
  4. The BF gave an annual average number of jobs filled by paid employees at the sector by industry group level.
  5. Surveyed pay rates collected from the Labour Cost Survey (LCS) for the June 2008 quarter, converted to weekly pay rates, gave the price information needed for weight calculation.

top

Adjusting for under 15-year-olds

The employment questions in the Census are restricted to people aged 15 years and over. However, the QES and BF count jobs filled by paid employees of all ages, including those aged under 15. It is also within the coverage of the LCI (as confirmed at the time of the 2001 reweight) to include jobs filled by under 15-year-olds.

Occupation figures at the sector by industry group level were modified due to the exclusion from the Census of under 15-year-olds, using LEED. The private sector was determined to have significant numbers of jobs filled by under 15-year-olds in the following industries:

  • agriculture
  • printing, publishing and recorded media
  • construction
  • wholesale trade
  • retail trade
  • accommodation, cafes and restaurants
  • property and business services
  • health and community services
  • cultural and recreational services.

There were no public industries deemed to have a significant number of jobs filled by under 15-year-olds.

The LEED figures for jobs filled by under 15-year-olds were added to specific occupations within sector by industry groups, based on Census information on occupations typically filled by 15- to 17-year-olds.

top

Quantity benchmarking

Other discrepancies arise between the Census, BF and QES due to people holding more than one job. In the Census people are counted only once and classified for status in employment, occupation and industry on the basis of main job. Part-time/full-time status is on the basis of usual hours worked in all jobs. However for the purpose of this LCI reweight, usual hours in main job were used, as this aligned best with the BF and QES, which count each job separately.

The BF paid-employee numbers were used to determine quantity figures at the sector of ownership by industry group level. The 2006 Census data gave the mix of occupations within industry groups and sectors of ownership, but not absolute levels (that is, the Census full-time and part-time employee numbers at the five-digit occupational level within sectors of ownership by industry groups were adjusted so that industry by sector of ownership totals matched those from the BF).

The part-time to full-time ratios from the QES were used to split the number of jobs filled by paid employees from the BF into full-time and part-time jobs. The part-time to full-time ratios calculated from Census hours-worked information at the sector by industry group level were used to convert the part-time count to a full-time equivalent (FTE) number of paid employees. Multipliers were then calculated to convert Census full-time and part-time figures at the sector by industry group by occupation level to match BF FTE figures at the sector by industry group level. FTE paid employee numbers were used for calculating the ordinary time weights while full-time employees were used for calculating the overtime weights.

A simple example illustrates this:

Sector 1 Census  Quarterly Employment Survey
Full-time employees Part-time employees Full-time (FT) employees Part-time (PT) employees FT/(FT+PT)
 Industry group 1
Occupation 1 22 7
Occupation 2 16 10
Occupation 3 29 11
Total 67 28 72 33

0.6857

 

Sector 1 Business Frame Multipliers
Employees Full-time employees Part-time employees Full-time employees Part-time employees
Industry group 1
Occupation 1
Occupation 2
Occupation 3
Total 96 65.83 30.17 0.98 1.08

 

Sector 1 Benchmarked counts
Full-time benchmarked Part-time benchmarked Full-time equivalent*
Industry group 1
Occupation 1 21.62 7.54 24.57
Occupation 2 15.72 10.78 19.94
Occupation 3 28.49

11.85

33.14
Total 65.83 30.17  77.66

* Part-time to full-time ratio, from Census, is equal to 0.392

The benchmarked FTE counts were multiplied by the wages collected in the LCS to calculate the initial base expenditure levels.

top

Ordinary time wage adjustments at the sector by industry by occupation level

To ensure the accuracy of the LCI weights, they were compared at a sector of ownership by industry group level with the QES gross earnings figures for ordinary and overtime pay for the year to June 2008 and LEED gross expenditure for the year to June 2007. The following LCI industries in the private sector had higher initial expenditure than the other two data sources:

  • agriculture
  • food, beverage and tobacco manufacturing
  • machinery and equipment manufacturing
  • construction
  • wholesale trade
  • retail trade
  • accommodation, cafes and restaurants
  • property and business services
  • education
  • health and community services
  • cultural and recreational services
  • personal and other services.

Central government administration and defence, and local government administration in the public sector were also found to have higher-than-expected initial expenditure levels. Generally, the pay rates collected via the LCS in these sector by industry groups were higher than average across all jobs. This is because jobs tracked in the LCS are selected using the judgement of businesses and tend to be filled by permanent, full-time employees with higher-than-average experience.

A Census average wage (using income from salary and wage information) was calculated for each sector by industry by occupation represented in the LCI sample. The average wages were updated using sector by industry salary and wage rate LCIs for the year to March 2006 to the June 2008 quarter. The Census average wages were further modified to have overtime removed, as only ordinary wage was looked at for this modification.

For the above industries, the LCI average wage figures that were significantly different were modified to reflect the Census averages. For retail trade in the private sector, modifications were done for all occupations collected.

These adjustments helped to ensure the accuracy of the occupational expenditure shares within industry groups.

top

Ordinary time wage adjustments at the sector by industry level

To further ensure the accuracy of the new LCI weights, June 2008 quarter ordinary time average earnings per job from the QES at the sector of ownership by industry level were multiplied by the quantity figures calculated from the BF information to determine the expenditure each sector of ownership by industry group was to represent.

top

Determining the overtime expenditure levels

Using the QES ordinary time to overtime earnings ratios and the LCI reweight ordinary time expenditure (based on the BF FTE quantity figures and QES earnings per job) an overtime expenditure level was derived. The overtime expenditure levels calculated by the LCI reweight (full-time figures multiplied by overtime prices collected from the LCS) were benchmarked to the overtime expenditure levels at the sector by industry group level.

top

Adjusting for annual leave

The final adjustment that needed to be made was to remove the component of the wage and salary payments that represents paid annual leave and statutory holidays. This could not be done earlier as QES and LEED included paid leave in their figures. The leave-adjusted figures were used with the BF-adjusted quantity levels and the QES ordinary and overtime expenditure benchmark-level multipliers to determine the final new LCI weights.

The component of the wage and salary payments that represent paid annual leave and statutory holidays will be used to reweight the LCI annual leave and statutory holidays index.

top

Industry breakdowns

The industry breakdown of the reweighted index, consisting of 27 published industry groups, is based on the Australian and New Zealand Standard Industrial Classification 1996 (ANZSIC96). The ANZSIC-based industry groups are broadly in line with those used for the Producers Price Index.

The industry groups conform with the ANZSIC classification with the exceptions that police services (Q963100), corrective centres (Q963200) and fire brigade services (Q963300) are included in central government administration and defence rather than in personal and other services, and sewerage and drainage services (D370200) is included in personal and other services rather than in electricity, gas and water supply.

top

Sample changes

The LCI sample has not been reselected, but companies that have changed sector of ownership and/or industry has been reclassified.

top

What is next

The next phase involves reweighting the non-wage indexes for the June 2009 quarter.Subsequent phases include implementing more up-to-date industry (Australian and New Zealand Standard Industrial Classification 2006) and occupation (Australian and New Zealand Standard Classification of Occupations) classifications in the LCI. The new industry and occupations classifications have a different structure from the current industry and occupation classifications used in the LCI. The LCI sample will be refreshed to ensure it adequately represents the new classification structures.

First LCI results using the new classifications will be included in the September 2009 quarter salary and wage rates information release. This will be followed by the introduction of the new industry classifications for the June 2010 quarter non-wage index information release.

Back to Price Index News: January 2009

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
Top
  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+