Recall(i)ng the ex(i)st(i)ng (i)nternat(i)onal standards of stat(i)st(i)cs of wages and hours of work conta(i)ned (i)n the resolut(i)ons of the F(i)rst, Seventh, Tenth and Eleventh Internat(i)onal Conferences of Labour Stat(i)st(i)c(i)ans and (i)n Convent(i)on No. 63 concern(i)ng stat(i)st(i)cs of wages and hours of work (1938),

Recogn(i)s(i)ng that the need for rel(i)able (i)nformat(i)on on wages and other forms of employee (i)ncome has (i)ncreased (i)n recent years both (i)n developed and (i)n develop(i)ng countr(i)es, part(i)cularly for purposes of measur(i)ng soc(i)al and econom(i)c progress among the var(i)ous groups (i)n the populat(i)on,

Recogn(i)s(i)ng further the need to rev(i)se, broaden and (i)ntegrate the ex(i)st(i)ng standards (i)n order to prov(i)de gu(i)del(i)nes for the product(i)on of comprehens(i)ve and mutually cons(i)stent stat(i)st(i)cs of wages ((i)nclud(i)ng salar(i)es) and the need to co-ord(i)nate these stat(i)st(i)cs w(i)th other econom(i)c and soc(i)al stat(i)st(i)cs,

Bel(i)ev(i)ng that, wh(i)le there are d(i)fferences among the needs of the var(i)ous countr(i)es, (i)nternat(i)onal gu(i)del(i)nes for an (i)ntegrated system of wages stat(i)st(i)cs w(i)ll promote development of these stat(i)st(i)cs along sound l(i)nes and contr(i)bute to (i)mprovement of the(i)r (i)nternat(i)onal comparab(i)l(i)ty,

Adopts th(i)s twenty-f(i)fth day of October 1973 the follow(i)ng resolut(i)on:

General object(i)ves and scope

1. Each country should a(i)m to develop (i)ts stat(i)st(i)cal programme (i)n the f(i)eld of wages, hours of work and related matters to prov(i)de (i)nformat(i)on for the var(i)ous users of the stat(i)st(i)cs, tak(i)ng (i)nto account part(i)cular nat(i)onal needs and c(i)rcumstances.The programme should prov(i)de for the needs of users (i)n connect(i)on w(i)th the measurement of levels of l(i)v(i)ng of employees, wage determ(i)nat(i)on, collect(i)ve barga(i)n(i)ng, soc(i)al, econom(i)c and manpower plann(i)ng, analys(i)s of econom(i)c cond(i)t(i)ons and market cond(i)t(i)ons, formulat(i)ng and (i)mplement(i)ng wage pol(i)c(i)es and (i)ncome pol(i)c(i)es and stud(i)es of (i)ncome d(i)str(i)but(i)on.

2. ((i) ) In order to real(i)se the above object(i)ves, nat(i)onal programmes of wages and related stat(i)st(i)cs should, (i)n pr(i)nc(i)ple, cover all sect(i)ons of the economy, should be developed w(i)th(i)n the general framework of an (i)ntegrated system and should compr(i)se two parts: (a) current stat(i)st(i)cs programme to meet short-term needs and (b) non-current stat(i)st(i)cs programme to prov(i)de benchmark data as well as other deta(i)led data to meet long-term and cont(i)nu(i)ng needs.

((i)(i)) In establ(i)sh(i)ng any nat(i)onal programme of wages and related stat(i)st(i)cs, the collaborat(i)on of organ(i)sat(i)ons of employers and workers should be sought.

3. The d(i)fferent (i)tems and ser(i)es of wages and related stat(i)st(i)cs (i)ncluded (i)n the nat(i)onal programme should be comp(i)led (i)n such a way as to be mutually cons(i)stent and reveal the relat(i)onsh(i)ps between them. Cons(i)stency w(i)th other econom(i)c and soc(i)al stat(i)st(i)cs should also be ensured to the fullest extent poss(i)ble.

4. ((i) ) The current stat(i)st(i)cs programme should cover: (a) stat(i)st(i)cs of average earn(i)ngs and hours of work ((i)nclud(i)ng, (i)f poss(i)ble, hours actually worked) and (b) stat(i)st(i)cs of t(i)me rates of wages and normal hours of work.

((i)(i)) The non-current stat(i)st(i)cs programme should (i)nclude: (a) stat(i)st(i)cs of wage structure and d(i)str(i)but(i)on and (b) stat(i)st(i)cs of labour cost.

5. Stat(i)st(i)cs of hours of work relevant to wages stat(i)st(i)cs should be comp(i)led, so far as poss(i)ble, (i)n accordance w(i)th comprehens(i)ve gu(i)del(i)nes g(i)ven (i)n the resolut(i)on of the Tenth Internat(i)onal Conference of Labour Stat(i)st(i)c(i)ans (1962) concern(i)ng stat(i)st(i)cs of hours of work.

6. Labour cost stat(i)st(i)cs should be comp(i)led, so far as poss(i)ble, (i)n accordance w(i)th the resolut(i)on of the Eleventh Internat(i)onal Conference of Labour Stat(i)st(i)c(i)ans (1966) concern(i)ng stat(i)st(i)cs of labour cost.

7. In v(i)ew of the spec(i)al problems of collect(i)on of the data, a separate programme of wages stat(i)st(i)cs should be drawn up for the agr(i)cultural sector, w(i)th(i)n the scope of the general framework of an (i)ntegrated system of wages stat(i)st(i)cs, (i)n accordance w(i)th the general recommendat(i)ons below, so far as they are appl(i)cable and, so far as (i)t (i)s pract(i)cable to do so, also (i)n accordance w(i)th the spec(i)al recommendat(i)ons on agr(i)cultural wages stat(i)st(i)cs wh(i)ch follow.

Concepts and earn(i)ngs

Earn(i)ngs

8. The concept of earn(i)ngs, as appl(i)ed (i)n wages stat(i)st(i)cs, relates to remunerat(i)on (i)n cash and (i)n k(i)nd pa(i)d to employees, as a rule at regular (i)ntervals, for t(i)me worked or work done together w(i)th remunerat(i)on for t(i)me not worked, such as for annual vacat(i)on, other pa(i)d leave or hol(i)days. Earn(i)ngs exclude employers' contr(i)but(i)ons (i)n respect of the(i)r employees pa(i)d to soc(i)al secur(i)ty and pens(i)on schemes and also the benef(i)ts rece(i)ved by employees under these schemes. Earn(i)ngs also exclude severance and term(i)nat(i)on pay.

9. Stat(i)st(i)cs of earn(i)ngs should relate to employees gross remunerat(i)on, (i).e. the total before any deduct(i)ons are made by the employer (i)n respect of taxes, contr(i)but(i)ons of employees to soc(i)al secur(i)ty and pens(i)on schemes, l(i)fe (i)nsurance prem(i)ums, un(i)on dues and other obl(i)gat(i)ons of employees.

10. ((i) ) Earn(i)ngs should (i)nclude: d(i)rect wages and salar(i)es, remunerat(i)on for t(i)me not worked (exclud(i)ng severance and term(i)nat(i)on pay), bonuses and gratu(i)t(i)es and hous(i)ng and fam(i)ly allowances pa(i)d by the employer d(i)rectly to h(i)s employee.

• (a) D(i)rect wages and salar(i)es for t(i)me worked, or work done, cover: ((i) ) stra(i)ght-t(i)me pay of t(i)merated workers; ((i)(i)) (i)ncent(i)ve pay of t(i)me-rated workers; ((i)(i)(i)) earn(i)ngs of p(i)eceworkers (exclud(i)ng overt(i)me prem(i)ums); ((i)v) prem(i)um pay for overt(i)me, sh(i)ft, n(i)ght and hol(i)day work; (v) comm(i)ss(i)ons pa(i)d to sales and other personnel. Included are: prem(i)ums for sen(i)or(i)ty and spec(i)al sk(i)lls, geograph(i)cal zone d(i)fferent(i)als, respons(i)b(i)l(i)ty prem(i)ums, d(i)rt, danger and d(i)scomfort allowances, payments under guaranteed wage systems, cost-of-l(i)v(i)ng allowances and other regular allowances.
• (b) Remunerat(i)on for t(i)me not worked compr(i)ses d(i)rect payments to employees (i)n respect of publ(i)c hol(i)days, annual vacat(i)ons and other t(i)me off w(i)th pay granted by the employer.
• (c) Bonuses and gratu(i)t(i)es cover seasonal and end-of-year bonuses, add(i)t(i)onal payments (i)n respect of vacat(i)on per(i)od (supplementary to normal pay) and prof(i)t-shar(i)ng bonuses.

((i)(i)) Stat(i)st(i)cs of earn(i)ngs should d(i)st(i)ngu(i)sh cash earn(i)ngs from payments (i)n k(i)nd.

Wage rates

11. The data on t(i)me rates of wages should relate to an appropr(i)ate t(i)me per(i)od such as the hour, day, week, month or other customary per(i)od used for purposes of determ(i)n(i)ng the wage rates concerned.

12. Wage rates should (i)nclude bas(i)c wages, cost-of-l(i)v(i)ng allowances and other guaranteed and regularly pa(i)d allowances, but exclude overt(i)me payments, bonuses and gratu(i)t(i)es, fam(i)ly allowances and other soc(i)al secur(i)ty payments made by employers. Ex grat(i)a payments (i)n k(i)nd, supplementary to normal wage rates, are also excluded.

13. Stat(i)st(i)cs of wage rates f(i)xed by or (i)n pursuance of laws or regulat(i)ons, collect(i)ve agreements or arb(i)tral awards (wh(i)ch are generally m(i)n(i)mum or standard rates) should be clearly d(i)st(i)ngu(i)shed from stat(i)st(i)cs referr(i)ng to wage rates actually pa(i)d to (i)nd(i)v(i)dual workers. Each of these types of wage rates (i)s useful for part(i)cular purposes.

14. T(i)me rates of wages for normal per(i)ods of work should be d(i)st(i)ngu(i)shed from spec(i)al and other rates such as p(i)ece rates, overt(i)me rates, prem(i)um rates for work on hol(i)days and sh(i)ft rates.

Wage payments (i)n k(i)nd

15. In v(i)ew of the dual nature of wages as cost to the employer and earn(i)ngs of the employee, (i)t may be necessary to evaluate wage payments (i)n k(i)nd accord(i)ng to both of these concepts.

16. In pr(i)nc(i)ple, for earn(i)ngs stat(i)st(i)cs, payment (i)n k(i)nd should be measured on the bas(i)s of the value accrued to the employee, s(i)nce earn(i)ngs refer to the remunerat(i)on or (i)ncome of a spec(i)f(i)ed group of employed persons, whereas for stat(i)st(i)cs of wage rates and labour cost the evaluat(i)on should be done on the bas(i)s of cost to the employer s(i)nce these data refer to the cost of un(i)ts of work t(i)me.

17. Evaluat(i)on of remunerat(i)on rece(i)ved (i)n k(i)nd on the bas(i)s of reta(i)l market pr(i)ces generally prov(i)des a reasonable est(i)mate of the value accrued to the employee. Th(i)s method (i)s frequently followed when (i)ncome data, (i)nclud(i)ng wages, are collected through household surveys.

18. When earn(i)ngs data are furn(i)shed by the employer, (i)t (i)s generally eas(i)er for h(i)m to express the money value of any payments (i)n k(i)nd wh(i)ch are (i)ncluded as equal to the cost to h(i)m of the goods or serv(i)ces concerned but, (i)f the employer (i)s unable to report the actual cost (i)ncurred, (i)t (i)s conven(i)ent to use producers' sell(i)ng pr(i)ces, or wholesale pr(i)ces. However, to evaluate the level of earn(i)ngs of employees, and adjustment should be made for the d(i)fference between cost to the employer and reta(i)l pr(i)ces wherever payments (i)n k(i)nd are s(i)gn(i)f(i)cant.

Current wages stat(i)st(i)cs programme

19. The current programme should be des(i)gned to prov(i)de essent(i)al data at frequent (i)ntervals (i)n order to measure trends and short-term changes (i)n average earn(i)ngs, hours of work, t(i)me rates of wages and normal hours of work.

20. In determ(i)n(i)ng the scope and content of the current programme, the need for speed (i)n the collect(i)on and (i)n the process(i)ng of data at reasonable cost should be a major cons(i)derat(i)on.

21. In determ(i)n(i)ng the scope of part(i)cular wages stat(i)st(i)cs ser(i)es, the need to perm(i)t the computat(i)on of a sat(i)sfactory (i)ndex of real wages should be borne (i)n m(i)nd. In pr(i)nc(i)ple, consumer pr(i)ce data and wages data relat(i)ng to the same group of the wage-earn(i)ng populat(i)on should be obta(i)ned for th(i)s purpose.

Stat(i)st(i)cs of average earn(i)ngs and hours of work

22. The current programme of stat(i)st(i)cs of average earn(i)ngs and hours of work should cover all (i)mportant categor(i)es of wage earners and salar(i)ed employees (i)n all major econom(i)c act(i)v(i)t(i)es (i)n the country.

23. Stat(i)st(i)cal (i)nqu(i)r(i)es on earn(i)ngs and hours of work may often be advantageously comb(i)ned w(i)th a current survey of establ(i)shments carr(i)ed out for purposes of employment, us(i)ng the whole of the sample of establ(i)shments or a sub-sample. Some other current surveys may lend themselves to the collect(i)on of data on earn(i)ngs and hours of work.

24. Wh(i)le comp(i)lat(i)on of stat(i)st(i)cs of average earn(i)ngs and hours of work on a monthly bas(i)s (i)s des(i)rable, th(i)s (i)mposes a heavy burden on establ(i)shments and on stat(i)st(i)cal off(i)ces. On the other hand, half-yearly or annual collect(i)on of data does not prov(i)de adequate (i)nd(i)cators of short-term seasonal var(i)at(i)ons or trends. As a m(i)n(i)mum, the current programme should (i)nclude the comp(i)lat(i)on of quarterly stat(i)st(i)cs of average earn(i)ngs and hours of work.

25. The t(i)me reference per(i)ods for data and the arrangements for collect(i)on of data should reflect the typ(i)cal cond(i)t(i)ons of d(i)sbursement of earn(i)ngs by establ(i)shments and the usual payroll per(i)ods used (i)n the d(i)fferent (i)ndustr(i)es covered by the (i)nqu(i)ry.

26. In a quarterly survey, data could be collected for a typ(i)cal payroll (i)n respect of only one month, or of each month, thus prov(i)d(i)ng (i)n the latter case a monthly ser(i)es and perm(i)tt(i)ng calculat(i)on of quarterly averages.

27. Wherever poss(i)ble, monthly or quarterly stat(i)st(i)cs of average earn(i)ngs should be publ(i)shed exclud(i)ng (i)rregular or (i)nfrequent payments such as year-end, half-yearly and s(i)m(i)lar bonuses (i)n order that the short-term trend (i)n regular earn(i)ngs w(i)ll be reflected.

28. If the nat(i)onal monthly or quarterly stat(i)st(i)cs of average earn(i)ngs exclude certa(i)n (i)mportant components, such as the annual or other (i)nfrequent payments ment(i)oned above, the current programme of earn(i)ngs stat(i)st(i)cs should (i)nclude the comp(i)lat(i)on once a year, wherever poss(i)ble, of stat(i)st(i)cs of average earn(i)ngs (i)nclud(i)ng these part(i)cular components.

29. For (i)ssu(i)ng data (i)n nat(i)onal publ(i)cat(i)ons, the t(i)me un(i)t (i)n wh(i)ch average earn(i)ngs are expressed, e.g. hour, day, week, or month, should depend ma(i)nly on how mean(i)ngful the f(i)gures would be (i)n the country concerned and also on the feas(i)b(i)l(i)ty of collect(i)on of the requ(i)red data.

30. For purposes of (i)nternat(i)onal compar(i)sons, (i)t (i)s des(i)rable to present stat(i)st(i)cs of average earn(i)ngs and hours of work on a per week bas(i)s. Data on earn(i)ngs and hours for per(i)ods other than a week should be converted to a per week bas(i)s by apply(i)ng the rat(i)o between the number of work(i)ng days (i)n the per(i)od covered and the number (i)n a normal week ((i)nclud(i)ng fract(i)ons of a day (i)n each case).

31. For certa(i)n types of compar(i)sons, stat(i)st(i)cs of average earn(i)ngs per hour are preferable s(i)nce hours of work per day, week or month vary over t(i)me and between (i)ndustr(i)es and reg(i)ons and also from one country to another.

32. In countr(i)es wh(i)ch comp(i)le stat(i)st(i)cs of average earn(i)ngs per hour pa(i)d for, (i)nformat(i)on on hours actually worked not be(i)ng ava(i)lable regularly, add(i)t(i)onal (i)nformat(i)on should be collected, wherever poss(i)ble, to determ(i)ne the rat(i)o between the number of hours actually worked and the number of hours pa(i)d for.

33. Wherever the ava(i)lable data perm(i)t, the stat(i)st(i)cs of average earn(i)ngs and hours of work should be comp(i)led for males and for females for all employees and for wage earners and salar(i)ed employees separately, by (i)ndustry and by reg(i)on.

Stat(i)st(i)cs of t(i)me rates of wages and normal hours of work

34. Nat(i)onal current stat(i)st(i)cs programmes should make prov(i)s(i)on, wherever appropr(i)ate, for the comp(i)lat(i)on of stat(i)st(i)cs of t(i)me rates of wages and normal hours of work cover(i)ng wage earners (i)n each of the (i)mportant (i)ndustr(i)es. The stat(i)st(i)cs may take the form of t(i)me rates of wages and "normal" hours of work actually (i)n force (i)n representat(i)ve establ(i)shments or, alternat(i)vely, of the rates and normal hours of work f(i)xed by, or (i)n pursuance of, laws or regulat(i)ons, collect(i)ve agreements or arb(i)tral awards. The stat(i)st(i)cs should be comp(i)led (i)n respect of adult males and adult females (i)n (i)mportant occupat(i)ons, or groups of closely related occupat(i)ons, (i)n each of the pr(i)nc(i)pal (i)ndustr(i)es. Wherever appropr(i)ate, correspond(i)ng stat(i)st(i)cs should be comp(i)led for the pr(i)nc(i)pal occupat(i)ons among salar(i)ed employees.

35. For comp(i)lat(i)on of t(i)me rates of wages actually pa(i)d by establ(i)shments and the correspond(i)ng hours of work, sample surveys of establ(i)shments normally should be undertaken at yearly (i)ntervals (i)n respect of (i)mportant occupat(i)ons (i)n the pr(i)nc(i)pal (i)ndustr(i)es (i)n the country, wherever the fac(i)l(i)t(i)es and resources ava(i)lable perm(i)t. It (i)s poss(i)ble to cover the same ground by undertak(i)ng an annual cycle of monthly surveys each cover(i)ng a certa(i)n select(i)on of occupat(i)onal groups. The data thus obta(i)ned prov(i)de valuable up-to-date (i)nformat(i)on to supplement the results of wage structure and d(i)str(i)but(i)on surveys undertaken at longer (i)ntervals.

36. Where the stat(i)st(i)cs of t(i)me rates of wages do not g(i)ve the rates per hour but refer to a day, week or other customary per(i)od, the stat(i)st(i)cs of normal hours of work should relate to an (i)dent(i)cal per(i)od. However, (i)f the data collected refer to days of work, (i)n place of hours, (i)nformat(i)on also should be obta(i)ned on the average durat(i)on of the work(i)ng day, for purposes of calculat(i)ng rates per hour.

Stat(i)st(i)cs of wage structure and d(i)str(i)but(i)on

37. Wage structure and d(i)str(i)but(i)on surveys, (i)nclud(i)ng wage censuses, large-scale ad hoc occupat(i)onal wage surveys and s(i)m(i)lar (i)nqu(i)r(i)es, prov(i)de comprehens(i)ve benchmark data for use (i)n the comp(i)lat(i)on of current stat(i)st(i)cs of average earn(i)ngs, hours of work, t(i)me rates of wages and normal hours of work and deta(i)led data perm(i)tt(i)ng the comp(i)lat(i)on of:

• (a) stat(i)st(i)cs of wage rates, earn(i)ngs and hours of work of wage earners and salar(i)ed employees to (i)nd(i)cate wage d(i)fferent(i)als between branches of (i)ndustry, geograph(i)c reg(i)ons, occupat(i)ons, males and females, establ(i)shments of d(i)fferent s(i)zes and poss(i)bly also age groups, educat(i)onal levels and types of vocat(i)onal tra(i)n(i)ng or qual(i)f(i)cat(i)ons of employees;
• (b) deta(i)led data on the compos(i)t(i)on and components of earn(i)ngs and wage rates;
• (c) stat(i)st(i)cs show(i)ng the d(i)str(i)but(i)on of wage earners and salar(i)ed employees accord(i)ng to levels of wage rates, earn(i)ngs and hours of work respect(i)vely, class(i)f(i)ed by var(i)ous (i)mportant character(i)st(i)cs of employees.

38. Sample surveys of establ(i)shments generally const(i)tute the most su(i)table means for collect(i)on of data on wage structure and d(i)str(i)but(i)on. It (i)s part(i)cularly (i)mportant that the des(i)gn and s(i)ze of the sample of establ(i)shments be adequate and (i)n part(i)cular that all s(i)zes of establ(i)shments w(i)th(i)n the scope of the survey, all (i)ndustr(i)es and reg(i)ons should be correctly represented.

39. In pr(i)nc(i)ple, all categor(i)es of wage earners and salar(i)ed employees, (i)nclud(i)ng full t(i)me and part t(i)me, permanent and temporary, should be covered (i)n the survey. Informat(i)on may be obta(i)ned (i)n respect of all el(i)g(i)ble employees (i)n the selected establ(i)shments, or a representat(i)ve sample of these employees. The (i)nformat(i)on collected concern(i)ng each employee should (i)nclude wage rates, earn(i)ngs, hours of work, age, sex, occupat(i)on and, (i)f poss(i)ble, educat(i)on, vocat(i)onal tra(i)n(i)ng or qual(i)f(i)cat(i)on, per(i)od of serv(i)ce and, where des(i)red, other relevant part(i)culars. Manager(i)al staff remunerated predom(i)nantly by a share of prof(i)ts should be excluded.

40. In v(i)ew of the breadth and complex(i)ty of wage structure and d(i)str(i)but(i)on surveys, countr(i)es wh(i)ch undertake them should normally do so only at three- to f(i)ve-year (i)ntervals.

41. Although a t(i)me reference per(i)od of one year (i)s (i)deal for certa(i)n data collected (i)n wage structure and d(i)str(i)but(i)on surveys, (i)n pract(i)ce (i)t (i)s usually necessary to select a shorter reference per(i)od cons(i)dered to be suff(i)c(i)ently representat(i)ve for the purpose of obta(i)n(i)ng data on earn(i)ngs as well as other (i)tems.

42. Data on the compos(i)t(i)on of earn(i)ngs and wage rates should be cons(i)stent w(i)th the class(i)f(i)cat(i)on of components of earn(i)ngs ment(i)oned (i)n paragraph 10. Supplementary (i)nformat(i)on may be collected on the scales for p(i)ece rates, overt(i)me rates, prem(i)um rates for work on hol(i)days, sh(i)ft d(i)fferent(i)als and s(i)m(i)lar payments.

43. Stat(i)st(i)cs of wage rates, earn(i)ngs and hours of work by occupat(i)on should be comp(i)led (i)n as much deta(i)l as poss(i)ble.

Agr(i)cultural wages stat(i)st(i)cs

44. The concepts and def(i)n(i)t(i)ons g(i)ven above would be generally appl(i)cable to the agr(i)cultural sector. Such adaptat(i)ons as may be necessary (i)n part(i)cular countr(i)es to meet spec(i)al cond(i)t(i)ons (i)n agr(i)culture would normally not be of a fundamental character. Although the (i)nternat(i)onal def(i)n(i)t(i)on of labour cost (i)s also appl(i)cable to the whole agr(i)cultural sector, stat(i)st(i)cs of labour cost (i)n the trad(i)t(i)onal subsector of agr(i)culture would not be very mean(i)ngful s(i)nce h(i)red labour const(i)tutes only a m(i)nor part of total labour (i)nput.

45. For purposes of wage stat(i)st(i)cs, the agr(i)cultural sector should compr(i)se major groups 111 (Agr(i)cultural and L(i)vestock Product(i)on) and 112 (Agr(i)cultural Serv(i)ces) of the Internat(i)onal Standard Industr(i)al Class(i)f(i)cat(i)on of All Econom(i)c Act(i)v(i)t(i)es. In certa(i)n c(i)rcumstances, (i)t may be des(i)rable to comp(i)le data separately for each of these major groups.

46. As the factors affect(i)ng wages, hours of work and other cond(i)t(i)ons of employment (i)n hunt(i)ng, trapp(i)ng and game propagat(i)on (major group 113 of the ISIC), forestry and logg(i)ng (d(i)v(i)s(i)on 12) and f(i)sh(i)ng (d(i)v(i)s(i)on 13) are generally d(i)fferent from those concern(i)ng agr(i)culture, these act(i)v(i)t(i)es should normally be excluded from the scope of agr(i)cultural wages stat(i)st(i)cs.

47. The def(i)n(i)t(i)on of "agr(i)cultural work" establ(i)shed for the 1970 World Census of Agr(i)culture should be adopted for the purposes of wages stat(i)st(i)cs, so far as poss(i)ble. By agr(i)cultural work or agr(i)cultural act(i)v(i)t(i)es (i)s meant any farm work or plann(i)ng necessary to the operat(i)on of the hold(i)ng.

48. Wherever poss(i)ble, separate data should be comp(i)led for permanent or regular employees and others, such as temporary and casual workers.

49. In developed countr(i)es and for the modern agr(i)cultural sector (i)n develop(i)ng countr(i)es, stat(i)st(i)cs of earn(i)ngs based on payrolls and other records of employers should be collected, us(i)ng the agr(i)cultural hold(i)ng as the report(i)ng un(i)t. However, agr(i)cultural employees, apart from permanent employees, are not necessar(i)ly attached to one part(i)cular agr(i)cultural hold(i)ng. Notably (i)n develop(i)ng countr(i)es, more espec(i)ally (i)n the trad(i)t(i)onal sector of agr(i)culture, prolonged cont(i)nuous employment w(i)th the same employer (i)s relat(i)vely rare and, (i)n add(i)t(i)on, some workers alternate between work(i)ng for wages and self-employment, or between agr(i)cultural and non-agr(i)cultural employment. In such s(i)tuat(i)ons, the agr(i)cultural hold(i)ng (i)s not the most su(i)table report(i)ng un(i)t for collect(i)on of certa(i)n types of data, espec(i)ally for comprehens(i)ve stat(i)st(i)cs of average earn(i)ngs dur(i)ng the year from agr(i)cultural employment and for stat(i)st(i)cs of the correspond(i)ng hours worked.

50. Stat(i)st(i)cs of agr(i)cultural wages (i)n the trad(i)t(i)onal subsector may be obta(i)ned through household sample surveys. The pr(i)nc(i)pal act(i)v(i)ty cr(i)ter(i)on should be used for def(i)n(i)ng agr(i)cultural employees. A person would thus be cons(i)dered to be an agr(i)cultural employee (i)f the pr(i)nc(i)pal source of the (i)ncome accrued to h(i)m dur(i)ng a spec(i)f(i)ed reference per(i)od was agr(i)cultural wages.

Programme of agr(i)cultural wages stat(i)st(i)cs

51. Nat(i)onal programmes of current and non-current agr(i)cultural wages stat(i)st(i)cs should be developed w(i)th(i)n the framework of an (i)ntegrated system cover(i)ng stat(i)st(i)cs of earn(i)ngs, wage rates, hours of work and labour cost, so far as poss(i)ble. S(i)nce the resources requ(i)red for the product(i)on of agr(i)cultural wages stat(i)st(i)cs are substant(i)al, part(i)cularly (i)n develop(i)ng countr(i)es, pr(i)or(i)t(i)es should be carefully determ(i)ned to ensure smooth and sat(i)sfactory progress of the nat(i)onal programme of wages stat(i)st(i)cs.

52. In the develop(i)ng countr(i)es, (i)n order to take account of the problems spec(i)f(i)c to them, separate subprogrammes of wages stat(i)st(i)cs should be developed for trad(i)t(i)onal and organ(i)sed agr(i)culture, where d(i)fferent methods and data collect(i)on techn(i)ques are needed. So far as poss(i)ble, the data collected (i)n the two subsectors should be su(i)table for comb(i)nat(i)on to produce stat(i)st(i)cs for the agr(i)cultural sector as a whole.

53. Selected stat(i)st(i)cs should be comp(i)led separately for agr(i)cultural employees who are pa(i)d wholly (i)n cash, those pa(i)d wholly (i)n k(i)nd and those pa(i)d partly (i)n each med(i)um. In add(i)t(i)on, (i)nformat(i)on should be g(i)ven on the amounts for major components of payments (i)n k(i)nd, such as food and hous(i)ng, wh(i)ch are (i)ncluded (i)n publ(i)shed stat(i)st(i)cs of average earn(i)ngs of agr(i)cultural employees.

54. Wherever poss(i)ble, the data should be class(i)f(i)ed accord(i)ng to occupat(i)on and accord(i)ng to broad types of agr(i)cultural hold(i)ngs (da(i)ry, poultry, l(i)vestock, f(i)eld crops, m(i)xed farms, etc.).

Current agr(i)cultural wages stat(i)st(i)cs

55. S(i)nce there (i)s seasonal var(i)at(i)on (i)n agr(i)cultural act(i)v(i)ty, and s(i)nce the relat(i)ve (i)mportance of agr(i)culture d(i)ffers from country to country, the (i)nterval for collect(i)on of current data on earn(i)ngs and t(i)me worked (i)n agr(i)culture should be determ(i)ned (i)n the l(i)ght of the needs of each country. The data should cover all categor(i)es of agr(i)cultural employees, (i)nclud(i)ng those pa(i)d wholly (i)n k(i)nd. However, those employees whose remunerat(i)on (i)s not pa(i)d regularly at da(i)ly, weekly or monthly (i)ntervals (but m(i)ght cons(i)st, for example, of a share of the crop, w(i)th or w(i)thout some cash wages) would have to be excluded from the current stat(i)st(i)cs of earn(i)ngs.

56. Where a country has a cont(i)nu(i)ng and frequent household sample survey for obta(i)n(i)ng labour force data, add(i)t(i)onal (i)nformat(i)on m(i)ght be collected, at a reasonable cost, on earn(i)ngs of agr(i)cultural employees and hours or man-days worked. However, attempts to obta(i)n rel(i)able data from th(i)s source, (i)n respect of earn(i)ngs (i)n agr(i)culture, m(i)ght encounter sampl(i)ng and other pract(i)cal problems.

57. Each country wh(i)ch does not regularly comp(i)le current stat(i)st(i)cs of earn(i)ngs and hours of man-days (i)n agr(i)culture should undertake surveys of t(i)me rates of wages actually pa(i)d and hours worked (i)n agr(i)cultural establ(i)shments, preferably at (i)ntervals of not more than s(i)x months. These surveys should cover the pr(i)nc(i)pal occupat(i)ons (i)n agr(i)culture.

58. Where labour contract rates quoted (i)nclude both a wage rate and h(i)re charges for the worker's own equ(i)pment, (i)mplements or work(i)ng an(i)mals, adjustments should be made to exclude the h(i)re charges.

Non-current agr(i)cultural wages stat(i)st(i)cs

59. Stat(i)st(i)cs of wage structure and d(i)str(i)but(i)on and of labour cost should be comp(i)led for the organ(i)sed subsector of agr(i)culture at f(i)ve-yearly (i)ntervals, based on results of surveys of agr(i)cultural establ(i)shments.

60 S(i)nce agr(i)cultural hold(i)ngs commonly employ temporary and casual workers and the total earn(i)ngs of these persons from agr(i)cultural work for the data reference per(i)od of a wage structure and d(i)str(i)but(i)on survey may not be fully reflected (i)n the payroll records of the establ(i)shment, data could be collected from the employees currently work(i)ng on the hold(i)ng on any add(i)t(i)onal earn(i)ngs they had rece(i)ved (i)n respect of pa(i)d work on other agr(i)cultural hold(i)ngs dur(i)ng the reference per(i)od.

61. Informat(i)on on the structure and d(i)str(i)but(i)on of earn(i)ngs and hours of work (i)n the trad(i)t(i)onal sector of agr(i)culture (i)n develop(i)ng countr(i)es could be obta(i)ned through household surveys cover(i)ng agr(i)cultural households, espec(i)ally large-scale surveys of agr(i)cultural labour (i)ncomes where data are obta(i)ned on earn(i)ngs throughout the year. It (i)s des(i)rable (i)n large-scale sample surveys of agr(i)cultural households to des(i)gn the sample (i)n such a way as to perm(i)t monthly or quarterly est(i)mates to be made of average earn(i)ngs and hours or man-days worked.