Quality Report. The Labour Cost Survey Norway

Transcription

1 Quality Report The Laour Cost Survey 2004 Norway

2 Tale of contents 1. Relevance Accuracy Sampling errors Proaility sampling Non-proaility sampling Non-sampling errors Coverage errors Measurement and processing errors Non-response errors Model assumption errors Punctuality and timeliness Punctuality Timeliness Accessiility and clarity Accessiility Clarity Comparaility Geographical comparaility Comparaility over time Coherence Coherence with the Laour Force Survey (LFS) Taular comparison with the LFS Coherence with National Accounts Taular comparison with the NA Appendix A - Frame population and sample size in LCS Appendix B - Coefficient of variance Appendix C Description of variales in the LCS List of tales and figures Tale 2-1. Sample fraction y enterprise... 3 Tale 2-2. Sample fraction y employees... 3 Tale 2-3. Coefficients of variations. Total laour cost. By size-class... 5 Tale 2-4. Coefficients of variations. Total laour cost. By industry... 6 Tale 2-5. Coefficients of variations. Total laour cost per paid hour. By size-class... 6 Tale 2-6. Coefficients of variations. Total laour cost per paid hour. By industry... 6 Tale 2-7: Percentage of correction... 8 Tale 2-8. Response rate... 9 Tale 6-1. Distriution of employees y industry. LFS and LCS Tale 6-1. Distriution of compensation of employees, hours and numer of employees y industry.... NA and LCS Figure 3-1. Data-flow. Aggregated response

3 1. Relevance The purpose of the statistics is to provide an overview of the total costs of having an employee. Statistics are provided for each industry separately, roken down on cost components. Users are the Technical Reporting Committee on the Income Settlement, research institutes, employee and employer organizations, Eurostat, the media, usiness and industry. Based on feedack Statistics Norway assumes that the main user, Eurostat, is satisfied with the quality of the main results of the Norwegian LCS. 2. Accuracy 2.1. Sampling errors 1. The population is ased on the Norwegian Business register where all enterprises with local units that have employees in the reference period are included as the total population. The final population used for sampling, in the following called frame population, is limited y use of cutoff (see chapter Bias). Sampling errors can arise where we have great differences etween the frame population and the population used for post-stratification in the following called the target population. The frame population eing the population most readily availale at the time of sampling, while the target population is an updated version of the population that is a etter description of the actual population we want the survey to represent. Sampling units included in the frame population can have faulty information due to time lags in registration or other reasons. The vital updated information in the final target population includes not just the stratification variales ut also information on the lifespan of the sample unit in the reference period. Tale 2-1. Sample fraction y enterprise Size class of enterprise (numer of employees) Numer of enterprises in universe Numer of enterprises in the sample Sample fraction employees % % % % % All % Tale 2-2. Sample fraction y employees Size class of enterprise (numer of employees) Numer of employees in universe Numer of emplyees in the sample Sample fraction employees % % % % % All % More details aout frame population and sample in appendix A. 3

4 Proaility sampling The method of sampling in the Norwegian LCS is stratified random sampling. The stratification variales are industry and employment which is used to stratify into size-groups. Calculating weights can e descried through two stages the first is a calculation of inverse inclusion proailities (1, elow), and secondly a post-stratification procedure (2, elow). Notation i Enterprise (Sampling and analyses unit) Strata c Industry Numer of enterprises in strata K k N ˆ N ni * w w Numer of enterprises in sample from strata Total numer of employees in strata Sum of weights for i enterprises in strata Numer employees in enterprise i in strata Inverse sample proaility Final post-stratified adjusted weight The inverse inclusion proaility is defined as: 1) w * = K k While the finally adjusted post-stratified weights are defined as: 2) w w = N n * * as shown aove gives the following calculation i = = i i i K w = gives an estimation of the numer of enterprises in a strata in the population. K k i K k k, which Following after the aove descried a post-stratification procedure is initiated where we seek to adjust * and end up with he final set of weights called. w w N (Total numer of employees in strata ) is ased on information for the most updated version of the Norwegian Business register, an extraction which est descries the reference period of the survey. The aim of the final adjusted weights is to find Nˆ and adjust the weights through a post-stratification procedure in all cases where the following is true: i * wi ni = Nˆ N. 4

5 The final weights can therefore e descried as: w * N w Nˆ * N w * wi n K k N K i k n K k N K n k = = = = = i i i N n where i Therefore the main aim for the weights is an estimation of how many employed there are in the target population in the reference period. The main aim is of course that the weights make it possile to calculate an estimation of laour costs for the target population Bias The statistics on laour costs are as all other sample ased statistics suject to ias, which arises when the distriution on some variales in different parts of the sample is not the same as the corresponding distriution in the population. Dividing the population into groups (strata) according to certain stratification variales reduces the possiility of imalances in the sample. Partial non-response in several of the items collected y form and used in the statistics can in some cases e recalculated on the asis of other information given on the form or extracted from the register on End of the Year Certificates for the year Post-stratification adjusts most imalances arising in the distriution etween the stratification variales due to non-response. The weights are additionally adjusted for any imalances due to nonresponse. Non-response that is not randomly distriuted may ias the sample, and this may have some influence on this statistics. Non-response in the statistics is 4 per cent and varies etween 0 and 12 per cent for the different divisions. (See tale 2-5: Response rate ). The use of cut-off may also e a source of ias. In all industries the cut-off is 10 employees Variance Variance of interest in this case is variance that arises ecause of the size and composition of the sample, more specifically the sampling model, so-called sample variance. Below in tale 2-3 and 2-4 we see coefficients of variations for total laour costs roken down to size-classes and industry. Tale 2-4 and 2-5 shows the same just for average laour costs per hour. See more aout coefficients of variations in appendix B. Tale 2-3. Coefficients of variations. Total laour cost. By size-class Size-class Average total laour cost Variance total laour cost. Million NOK Coefficients of variations Total ,038 1 E ,014 2 E ,016 3 E ,017 4 E ,019 5 E ,080 5

6 Tale 2-4. Coefficients of variations. Total laour cost. By industry Variance total laour cost. Industry Average total laour cost Million NOK Coefficients of variations Total ,038 C ,381 D ,039 E ,201 F ,092 G ,039 H ,063 I ,154 J ,211 K ,065 M ,071 N ,065 O ,104 Tale 2-5. Coefficients of variations. Total laour cost per paid hour. By size-class Size-class Average total laour cost per hour Variance total laour cost per hour. NOK Coefficients of variations Total 211,4 1,6 0,005 1 E ,0 3,1 0,008 2 E ,4 5,3 0,010 3 E ,4 9,7 0,013 4 E ,6 8,6 0,012 5 E ,6 6,3 0,090 Tale 2-6. Coefficients of variations. Total laour cost per paid hour. By industry Industry Average total laour cost per hour Variance total laour cost per hour. NOK Coefficients of variations Totalt 211,4 1,6 0,005 C 358,2 628,2 0,069 D 205,3 2,1 0,007 E 320,3 601,8 0,076 F 197,0 8,2 0,014 G 202,6 5,6 0,011 H 157,3 10,5 0,020 I 204,6 12,2 0,017 J 309,1 177,0 0,043 K 267,8 29,6 0,020 M 218,6 111,3 0,048 N 189,8 21,9 0,024 O 207,3 23,0 0, Non-proaility sampling Not applicale 6

7 2.2. Non-sampling errors Coverage errors The population is made up of all enterprises in Statistics Norway's Central Register of Estalishments and Enterprises, with the exception of small enterprises with fewer than 10 employees. Each enterprise covers one or more local units. The samples in each section consist of enterprises drawn from the population, dependent of industry and the numer of employees. All section C-O not including L and pulic services within health, education and social services run y state or municipal authorities. Errors in the stratification variales, activity (Nace Rev. 1) and numer of employees, in the frame population could e sources of errors. Also actual differences etween the frame population and target population may lead to prolems such as over-coverage or under-coverage in su-populations. To cope with this potential prolem, the local units in the sample are asked to control the pre-printed code of activity and information from the register on End of the Year Certificates for the year 2004 on the form. If this information is elieved to e incorrect, the enterprises are asked to give a correction. In each specific case, this information is assessed individually to seek a most correct result on each unit this is relevant for. Some under-coverage may e expected due to a time-lag in the registration of new units in the Central Register of Estalishments and Enterprises. Over-coverage may also e present due to the same reason, i.e. the time lag in the registration process when enterprises no longer have employees ecause the usiness has een closed, sold or taken over y new owners, has gone ankrupt or has een merged in the time period etween the selection of the sample and the time of the census or within the reference period which can ee frequent in a survey where this period is long. As long as these errors are rather constant, the effect on the statistics is minimal Measurement and processing errors Measurement errors Measurement errors mainly occur ecause the respondent misunderstands what is included in and consequently reported in each column on the form or ecause it is very difficult for the respondent to find the information requested. As far as possile the questionnaire used the most common ookkeeping terms. All variales collected and that are directly or indirectly included in released statistics are checked, either in logical controls or y asolute limits for what is considered valid. If important data are missing or suspected incorrect in the received reports, the data are otained or corrected in several ways. This was either y returning the form, y a phone call to the respondent, y use of other administrative sources or y imputation. To simplify the questionnaire the respondents were asked to give the numer of employees at the end of the year. This did not in all cases match with the level of costs accumulated through the year. Statistics Norway has tried to compensate for this in the cases where register information have indicated a mismatch etween average numer of employees during the year and numer of employees at the end of the year. Processing errors Data that are received are registered either y optical scanning or manual recording. Several controls are carried out on the material. The tale shows that the variales that have een corrected most often are hours paid and employers social contriution. These are all difficult variales for the respondents. A common error in paid hours are that the respondents did not report figures that corresponded with the numer of employees. 7

8 Many of the corrections are caused of respondents not summing up variales to a total. Tale 2-7: Percentage of correction Variale Definition % of cases that has een corrected A1 Total numer of employees 24,9 % A11 Full-time employees (excluding apprentices) 20,0 % A12 Part-time employees (excluding apprentices) 20,9 % A13 Apprentices 1,2 % C1 Total hours paid 43,4 % C11 Paid hours for full-time employees (excluding apprentices) 41,7 % C12 Paid hours for part-time employees (excluding apprentices) 39,5 % C13 Paid hours for apprentices 4,5 % D Total laour costs 66,3 % D1 Compensation of employees 54,6 % D11 Wages and salaries 35,0 % D111 Wages and salaries (excluding apprentices) 35,0 % D1111 Direct remuneration, onuses and allowances 29,6 % D11111 Direct remuneration, onuses and allowances paid in each pay period 29,6 % D1113 Payments for days not worked 34,8 % D11142 Staff housing (optional) 0,2 % D11143 Company cars (optional) 1,1 % D112 Wages and salaries of apprentices 0,4 % D12 Employers' social contriutions 50,5 % D121 Employers' actual social contriutions (excluding apprentices) 48,5 % D1211 Statutory social-security contriutions 49,9 % Collectively agreed, contractual and voluntary social-security D1212 contriutions 16,1 % D122 Employers' imputed social contriutions (excluding apprentices) 10,7 % D1223 Payments to employees leaving the enterprise (optional) 0,2 % D1224 Other imputed social contriutions of the employer (optional) 10,6 % D123 Employers' social contriutions for apprentices 0,2 % D2 Vocational training costs 19,5 % D3 Other expenditure paid y the employer 0,6 % D4 Taxes 23,2 % D5 Susidies received y the employer 16,7 % Non-response errors Unit non-response Unit non-response refers to the fact that the respondent, in this case enterprises, has not completed and returned the statistics questionnaire. In the statistics the unit response is etween 88 and 100 per cent (tale ellow). The main reasons for non-response are that units have ceased to exist, een sold or transferred to a new owner, gone ankrupt or have een merged. Furthermore, there is also a small group reporting too late to e included in the statistics, or providing data of a quality that cannot e used for statistical purposes. In case of unit non-response, the weights of the units on which the statistics are ased are adjusted to compensate for the non-response. 8

9 Tale 2-8. Response rate Numer of enterprises Division responded Rate of respons Percentage of enterprises used in statistics Nonrespons % 100 % % 88 % % 100 % % 91 % % 88 % % 100 % % 92 % % 100 % % 100 % % 92 % % 94 % % 92 % % 93 % % 98 % % 96 % % 96 % % 93 % % 94 % % 100 % % 96 % % 91 % % 80 % % 94 % % 95 % % 88 % % 90 % % 98 % % 91 % % 88 % % 91 % % 89 % % 75 % % 89 % % 86 % % 88 % % 91 % % 82 % % 98 % % 95 % % 89 % % 95 % % 84 % % 91 % % 92 % % 85 % % 86 % % 90 % % 100 % % 84 % % 84 % % 80 % 9

10 Partial non-response The type most typical for a sample survey is that the sample unit, enterprise, has not reported on all the necessary items in the questionnaire. Some of the items can often e calculated on the asis of other information and possily imputed from previous surveys or other sources Model assumption errors Statistics Norway has chosen to use information concerning the end of the year on the case of employees and industry. To ask the enterprises to calculate average levels of employment in the reference year or other complicated procedures could have given rise to other just as difficult prolems to handle in the compilation of the statistics. The accounting and fiscal year is identical with the calendar year in Norway. Hence, this is not suject to any errors regarding the Laour Cost statistics. The sample model used is ased on stratified samples. Dividing the population into groups (strata) according to certain stratification variales reduces the possiility of imalances in the sample and assures a etter coverage of certain units or group of units. The sample consists of enterprises drawn from the population. The population is asically all active enterprises in the section, with the exception of small enterprises with fewer than ten employees, which are not included in the frame population. Large enterprises (sample units), where the definition of large varies etween industries, receive a sampling proaility of 1. While strata that cover small and medium size sample units are given a lower sampling proaility. The stratification is made according to industry and size (numer of employees) of the enterprises, on the assumption that laour costs and composition off these costs in large enterprises differ from those in small ones, and that there are differences according to industry. In each stratum, this sample model ensures a minimal dispersion in the main variales measured, i.e. Laour costs and especially when it comes to compensation of employees where supplementing sources exist. The numer of employees is an important feature regarding the stratification. Some assessment of this size is done through the sampling-process and serves as guidance for ongoing improvement. In each stratum the mean numer of employees is calculated along with the standard deviation. This is done do ensure a est possile stratification that reflects the differences etween the strata, also information on compensation of employees is taken into account where possile. The enterprises that were drawn in the sample were matched with the register on End of the Year Certificates for the year Some information from this register was pre-printed on the questionnaires. 1.1 per cent of the enterprises in the sample did not match the register on the identification numer for enterprise. If the pre-printed information is elieved to e incorrect, the enterprises were asked to give a correction of the enterprises made corrections on the figure preprinted for wages and salaries. The purpose of the sample model selection process is asically to get samples that ensure a representative asis for the final statistics and avoid urdening all enterprises in the industry. This limits the size of the samples while focusing on main variales. Another ojective is to ensure that the smallest enterprises are the least possile urdened with reporting oligations. Statistics Norway likes to elieve that all these purposes are well fulfilled. In addition to this, the effect of any known and unknown model errors are reduced to an acceptale minimum through the use of this model. See chapter for more aout the coefficients of variation. 10

11 3. Punctuality and timeliness 3.1. Punctuality The questionnaires were sent 13 th of May 2005, with a deadline of 10 th of June Two rounds of reminders were used. In addition several enterprises were phoned to ensure that their questionnaires were returned. Exact dates are in the graph elow. Figure 3-1. Data-flow. Aggregated response. Data-flow 4000 Dispatch: 13. may Deadline: 10. june 1st reminder: 1. july Deadline, 1st reminder: 5. aug. 2nd reminder: 12.august Deadline, 2. reminder: 14. sept Numer of enterprises Date Aggregated response Of this used in the statistics Statistics Norway informs every year the respondents aout which surveys they might expect to receive during the year. A pre-sample was therefore drawn early The first quarter every year enterprises are occupied with alancing their accounts. The questionnaires where therefore dispatched quite late in the following year, Approximately 80 per cent of the enterprises in pre-sample ended up in the final sample. The statistics are collected in accordance to the mandate given through The Statistics Act of 1989 which for LCS makes response mandatory. The data processing period started July From there it was an ongoing process of analysing and approving questionnaires Timeliness The reference period for the survey is the year The result of the survey was pulished 29 th of June Date for delivery to Eurostat was 1 st of July

12 4. Accessiility and clarity 4.1. Accessiility The statistics are pulished on the Internet, at The results have een sent to Eurostat. No results are sent to the respondents Clarity At the same Internet-address mentioned in the previous chapter, the users can find references to a rief methodical document in the link aout the statistics Comparaility 5.1. Geographical comparaility The laour costs statistics for Norway is regarded as one region, at NUTS 1 level. The data is not roken down y geography Comparaility over time The Norwegian Laour Cost Survey are collected every forth year. There have een surveys for 1996, 2000 and The questionnaires were partly different in these surveys, ut the main variales are comparale for all three surveys. In the 1996 survey Statistics Norway asked the respondents to distriute the laour costs at local unit level. It was found that this was very difficult for the respondents. The result of this distriution given y the respondents was not used. So all the surveys are carried out at enterprise level. This might have effected the results roken down on NACE. The questionnaire for the 1996 survey contained questions aout hours actually worked. The quality of the data was considered to e poor. Experience from contact with the respondents, oth at the LCS and in other occurrences have given us indications that information aout hours actually worked will at est e a rough estimate. Even the quality of the reported variale Hours paid are in many cases poor, and a large percentage of the reported cases on hours paid have een corrected. (see tale in chapter Measurement and processing errors ). The applied methods and models have een suject to ongoing improvements ased on increased knowledge. Major point in this is the extended use of registers that helped us correct and impute data in the survey. It is reason to elieve that the quality of the data affected y this it higher. 12

13 6. Coherence See appendix C for description of variales in the laour cost survey Coherence with the Laour Force Survey (LFS) 2004 This is a short presentation and comparison of the Norwegian LCS and the Norwegian LFS survey. It is important to point out asic differences that possily could e the cause of differences etween the surveys as they are oserved in the tale. The main reasons for different surveys are in most cases, to meet different needs and as a consequence the statistics are uilt up on assumptions that meet these specific user needs. The LFS survey monitors and documents quarterly changes in the composition and distriution of the work force. It is ased on a sample survey covering individuals (the sample unit is family), that report on their status in the work force. Statistics on Laour costs on the other hand are uilt up to answer questions concerning the level and distriution of total laour costs. The source is as earlier descried a sample of enterprises that report on for the whole enterprise. The population for the two surveys does overlap, very much, ut the source of information is different and so are the sampling models. Further more the two surveys have different reference periods, and utilize different sources for control, verification and finally dissemination. Both statistics are none the less used for explaining different properties of the same field of interest and in this capacity we can use the LFS to understand some aspects of the distriution and composition of the employed within laour force. Discrepancies should where they occur e explained and understood as a consequence of overlapping information. Population and sampling units LFS LCS Population All individuals aged All enterprises with more than 10 employees Sampling unit Families Enterprises Analysis unit Individuals Enterprises Reporting unit Individuals Enterprise Frequency Quarterly Every 4 years Variale definitions LFS LCS Employed Persons on sick-leave included Working time Full-time-37 hours or more, Numer of full-time employees if not defined otherwise y the reported y the enterprise reporting unit Ojective of the the LFS and LCS statistics LFS Provide statistics on employed and unemployed and laour force participation LCS Provide statistics on the level and composition of Laour Costs 13

14 Taular comparison with the LFS Tale 6-1. Distriution of employees y industry. LFS and LCS 2004 Industry Distriution numer of employees, LFS Distriution variale A1 Total numer of employees, LCS Total 100,0 100,0 C-D 22,7 28,3 E 1,2 0,6 F 12,2 10,0 G-H 31,7 28,7 I 11,4 12,2 J 3,7 4,2 K 17,0 16, Coherence with National Accounts 2004 The national accounts (NA) statistics are designed to provide a consistent and comprehensive survey of the overall national economy. The annual national accounts give oth a summarised description of the economy as a whole and a detailed description of transactions etween different parts of the Norwegian economy. Compensation of employees, hours and numer of employees are compared etween NA and LCS in the tale elow. The definitions of the variales in NA are: Compensation of employees = Wages and salaries + Employers' social contriutions Where wages and salaries are remuneration to employees in respect of work done in domestic production. Wages and salaries are oth in cash and in kind. Wages and salaries in cash include pay for overtime, and sickness and maternity allowances paid y employers. Wages and salaries in kind consist of goods and services, or other enefits, provided free or at reduced prices y employers that can e used y employees at their own discretion. Wages and salaries in kind include, inter alia, the services of vehicles, value of the interest forgone y employers when they provide loans to employees at reduced rates of interest, and free transportation for employees in some transport industries. Employers' social contriutions are social contriutions incurred y employers, paid to central government and to autonomous social security and pension funds as well as non-autonomous pension funds. They include the following su-items: employers' contriutions to National Insurance, employers' other actual social contriutions (contriutions to the Pulic Service Pension Fund, Municipal Pension Funds, other social security schemes, and other social contriutions), and in addition, employers' imputed social contriutions. The latter item coincide with social enefits actually paid through unfunded arrangements - from employers to present or former employees, for instance AFP-pensions. Hours worked are hours worked y employed persons (employees and self-employed) in domestic production during one year. The hours worked refer to production within effective and normal working hours, with addition for overtime while deducting asences due to sickness, leave of asence, vacations and any laour conflicts. 14

15 Hours worked are also influenced y the calendar effect (movale holidays and leap years). Numer of working days may vary until three days from one year to next. The LCS includes only paid hours. Numer of employees include employed persons who, y agreement, work for another institutional unit and receive a remuneration recorded as compensation of employees. Owners of corporations (joint-stock companies etc.) if they work in these enterprises, are counted as employees Taular comparison with the NA Tale 6-1. Distriution of compensation of employees, hours and numer of employees y industry. NA and LCS 2004 Industry Distriution compensation of employees, NA Distriution of D1 - compensation of employees, LCS Distriution of hours worked, NA Distriution of C1 -paid hours, LCS Distriution Of numer of employees, NA Distriution of A1 Total numer of employees, LCS Total 100,0 100,0 100,0 100,0 100,0 100,0 C 5,1 7,6 2,5 4,0 2,2 3,8 D 21,6 25,5 21,3 26,3 20,3 24,5 E 1,4 0,9 1,0 0,7 1,0 0,6 F 10,0 10,3 11,6 11,1 10,4 10,0 G 21,0 16,9 23,9 19,3 25,7 22,0 H 3,5 3,5 4,2 4,9 5,0 6,7 I 13,8 11,8 15,3 12,7 14,2 12,2 J 5,2 6,2 3,7 4,6 3,6 4,2 K 18,3 17,3 16,4 16,4 17,7 16,1 15

16 7. Appendix A - Frame population and sample size in LCS 2004 Strata 1 Strata 2 Strata 3 Size of Numer Numer of Numer Size of Numer Numer of Numer Size of Numer Numer of Numer Industrial enterprise of Sample enterprises of enterprise of Sample enterprises of enterprise of Sample enterprises of Classification (>=) enterprises proaility in sample employees (>=) enterprises proaility in sample employees (>=) enterprises proaility in sample employees 1='CA' ='CB' ='DA' ='DB' ='DC' ='DD' ='DE' ='DF' ='DG' ='DH' ='DI' ='DJ' ='DK' ='DL' ='DM' ='DN' ='E' ='F ' ='F-45.3' ='F ' ='G-50' ='G-51' ='G-52' ='H-55.1' ='H-55.2' ='H-55.3' ='H-55.4' ='H-55.5' ='I-60' ='I-61' ='I-62' ='I-63' ='I-64' ='J-65' ='J-66' ='J-67' ='K-70' ='K-71' ='K-72' ='K-73' ='K-74' ='M-80' ='N-85' ='O-90' ='O-91' ='O-92' ='O-93'

17 8. Appendix B - Coefficient of variance In chapter 2.1. we have made use of coefficient of variance (CV). The following is a short summary of the CV as it is used in the main text and tales. CV sometimes called relative standard deviation is defined as the ratio of the standard error of the estimator to the expected value. In our case we aim to calculate a CV for the estimated average or estimated total of annual laour costs and average laour costs per hour. Based on the general definition of an average as X N N X i i= = 1. We can define the estimated average for the target population y using the earlier definition of the weights in chapter k i i i= 1 ˆ = where k xi X wi i= 1 w x sample and wi The standard deviation: is the oserved average laour costs per hour in enterprise i in the is the calculated weight (Final post stratified weight) for enterprise in the sample. k wi x The CV can therefore e calculated as ( ) i x 2 i= 1 s = where n is the numer of enterprises in the sample. n 1 cv expressed as a percentage 100% percentage. s X ˆ =. The CV as commented in chapter is often s cv % =. In text and tales our CV is not expressed as a X ˆ 17

18 9. Appendix C Description of variales in the LCS 2004 A.* Numer of employees A.1 Total numer of employees A.11 Full-time employees (excluding apprentices) A.12 Part-time employees (excluding apprentices) A.121 Part-time employees converted into full-time units (excluding apprentices) A.13 Apprentices A.131 Apprentices converted into full-time units C. Paid hours C.1* Total hours paid C.11 Paid hours for full-time employees (excluding apprentices) C.12 Paid hours for part-time employees (excluding apprentices) C.13 Paid hours for apprentices D.* Laour costs D.1 Compensation of employees D.11 Wages and salaries D.111 Wages and salaries (excluding apprentices) D.1111* Direct remuneration, onuses and allowances D.11111* Direct remuneration, onuses and allowances paid in each pay period D.11112* Direct remuneration, onuses and Allowances not paid in each pay period D.1112 Payments to employees savings schemes D.1113 Payments for days not worked D.1114 Wages and salaries in kind D Company products (optional) D Staff housing (optional) D Company cars (optional) D Stock options and share purchase schemes (optional) D Other (optional) D.112 Wages and salaries of apprentices D.12 Employers' social contriutions D.121 Employers' actual social contriutions (excluding apprentices) D.1211 Statutory social-security contriutions D.1212 Collectively agreed, contractual and voluntary social-security contriutions D.122 Employers' imputed social contriutions (excluding apprentices) D.1221 Guaranteed remuneration in the event of sickness (optional) D.1222 Employers' imputed social contriutions for pensions and health care (optional) D.1223 Payments to employees leaving the enterprise (optional) D.1224 Other imputed social contriutions of the employer (optional) D.123 Employers' social contriutions for apprentices D.2 Vocational training costs D.3 Other expenditure paid y the employer D.4 Taxes D.5 * Susidies received y the employer E. Information on units E.1 Local units, universe E.2 Local units, sample 18