The international standards on employment and unemployment statistics are embodied in Resolution I concerning statistics of the economically active population, employment, unemployment and underemployment, adopted by the Thirteenth International Conference of Labour Statisticians (October 1982)1.
Countries in transition to market economy are increasingly using these standards in their effort to put in place a system of statistics on employment and unemployment based on labour force surveys which can be compared internationally. The list in Annex 1 indicates that of the 25 transition countries of Central and Eastern Europe and Former Soviet Union, ten are already conducting annual or quarterly surveys and six have carried out or are in the process of carrying out pilot tests.
In applying the international standards in these surveys, certain issues have been raised for which no explicit provision could be found in the 1982 ICLS Resolution. These include, in particular, the statistical treatment of persons on various types of long-term leave, namely: (i) maternity leave; (ii) education or training leave; (iii) paid leave initiated by the employer or administration; (iv) unpaid leave initiated by the employer or administration; and (v) unpaid leave initiated by the employee.
At the ECE/ILO/OECD Work Session on Labour Statistics and Issues of Concern for Transition Countries (Paris, December 17-18, 1992), the ILO agreed to examine these issues and, if necessary, prepare draft guidelines which could be submitted to a future International Conference of Labour Statisticians.
Subsequently, a study was commissioned to identify the main issues involved and to review national legislations and country practices in labour force surveys, establishment surveys or administrative records in 15 transition countries.2
The present paper is a follow up to this study. It examines one of the five issues listed above, namely, (i) maternity leave, and makes a proposal on the basis of the 1982 ICLS Resolution. This proposal together with corresponding ones concerning the four remaining issues (ii)-(v) would form the subject of a debate at a special meeting to be convened by the ILO at the Czech Statistical Office in Prague towards the end of 1995. It is envisaged that after further discussion, draft guidelines could be developed for consideration by the Sixteenth International Conference of Labour Statisticians, planned to be held in 1998.
Maternity leave3 is the length of time off from work for childbirth, usually determined as a minimum period considered necessary for the rest and recuperation of the mother. Depending on national circumstances and legislations, this period can vary from a few weeks to several months, and is often divided into two parts before and after the birth. A period of employment prohibition or compulsory leave may be stipulated for a part or the whole of the maternity leave entitlement to ensure that women have sufficient rest. It is common to extend the period of maternity leave for difficulties associated with certain pregnancies or the outcome of a birth, such as illness due to the pregnancy or the birth, premature or late birth, and multiple births. The leave period may also be shortened in cases such as miscarriage, stillbirth or infant death.
Child care leave is a period of leave available to mothers (optional maternity leave) or to the parents (parental leave) in order to take care of their child or adopted child (adoption leave), usually on an unpaid basis, which is additional to the regular maternity or paternity leave, and can be taken either consecutively after maternity leave or at any other time, possibly split in different periods, until the child is below a certain age.
In the second half of 1993, the ILO despatched a questionnaire requesting 16 transition countries to provide information on their national legislations and statistical practices with respect to the various types of leave listed in the introduction. The answers from 15 countries received during the fourth quarter of 1993 and the first quarter of 1994 were analysed by Jaroslav Kux, ILO Consultant, and reported in the paper cited earlier.1 The information concerning maternity leave, formulated slightly differently, is presented in the Table 1 below.
It can be observed that both national legislations and statistical practices with regard to maternity leave vary greatly among countries.
National legislations. The normal duration of maternity leave prescribed by law in transition countries is generally longer than that decreed in the legislations of other industrialised countries. It is also generally longer than the 12 weeks minimum period set in the ILO Convention concerning maternity protection (No. 103, Revision 1952).
It can be noted from the above table that there is wide variation in the normal duration of maternity leave in the 15 transition countries covered. It is as low as 56 days in Lithuania and as long as 365 days in Albania. Except for Hungary where maternity leave includes a 6-week period after birth of compulsory leave, the other countries do not specify a compulsory period within the duration of the maternity leave.
Legislation Statistical
treatment
Country
Normal Employment Labour
duration Cash Source reporting force
(days/weeks) benefits system survey
Albania 365 - - E -
Belarus 126 100% Soc. Ins. N -
Bulgaria 120-180 100% Soc. Ins. E E
Croatia - - - E
-
Czech Rep. 28 w. - Soc. Ins. N E
Hungary 24 w. 100% Soc. Ins. E
E
Latvia 112 - Soc. Ins. E
-
Lithuania 56-70 - Soc. Ins. N -
The former
Yugoslav
Republic
of Macedonia 270 - Soc. Ins. E
-
Poland 16-18 w. 100% Soc. Ins. E
E
Rumania 112 50-94% Soc. Ins. E -
Russian Fed. 140 100% Soc. Ins. N -
Slovak Rep. 28 w. - Soc. Ins. N E
Slovenia 105 - Soc. Ins. E E
Ukraine 126 100% Soc. Ins. E -
Note: E = Employed; N = Not economically active.
In all the 15 countries, women on maternity leave are paid during the prescribed period of leave, although the amount of payment may in certain cases be less than 100%, as in Rumania which varies from 50% to 94% depending on the length of service and the number of children of the woman worker. The cash benefits received during maternity leave in transition countries are almost invariably paid from the state social insurance fund, generally paid to employees directly by employers and then deducted from the employers's contributions due to social insurance.
National legislations regarding child care leave also vary greatly among countries. The duration is generally expressed in terms of age of the child and ranges from 1 in the case of Romania to 3 in most other transition countries. The prescribed duration may change depending on the rank of the child, i.e., whether it is the first child, the second or subsequent child, as Bulgaria which is up to 2 years for the first, second and third child and 6 months for any subsequent child.
In relation to cash benefits, the duration of child care leave in most the countries covered can be divided into two parts: the first period during which beneficiaries are eligible for some payments, generally in the form of an allowance fixed by regulations, typically calculated as a fraction of past earnings or in relation to the national average monthly wage or minimum wage; the second period which is unpaid. For example, in Slovenia for the first 260 days of the child care leave, an allowance is paid from the state social insurance fund and from then on, until the child reaches three years of age, the leave is unpaid.
In certain cases, national legislations permit the mother or the parent to take up part-time work or work at home during leave on child care. This is the case, for example, in Belarus and Hungary.
Statistical practices. The statistical treatment of persons on maternity leave differ not only among countries but also between sources of data collection in the same country.
In employment statistics based on establishment reporting systems (administrative records), women on maternity leave are classified as employed in 10 transition countries listed in Table 1 above and as not economically active in 5 others. Each country applies the same statistical treatment to persons on child care leave as to persons on maternity leave.
In labour force surveys, the situation is somewhat different. Persons on maternity leave are classified as employed in all the six transition countries with such surveys as reported in Table 1 above. But, persons on child care leave are classified the same way (employed) only in three of these countries (Czech and Slovak Republics and Hungary). In the other three countries (Bulgaria, Poland and Slovenia), these persons are classified as not economically active if unpaid.
According to the international definition of employment (ICLS 1982), as well as virtually all existing national definitions, includes among the employed certain categories of persons who were not at work during the reference period. These are persons who were temporarily absent from work , with or without leave, for various personal or economic reasons such as illness or vacation, temporary disorganisation or suspension of work.
The international definition of employment specifies certain principles for ascertaining temporary absence from work. It differentiates between paid employment and self-employment. In the case of paid employment, these principles are based on the notion of "formal job attachment", which is to be determined, depending on national circumstances, according to one or more of the following criteria:
(a) the continued receipt of wage or salary;
(b) an assurance of a return to work following the end of the contingency, or an agreement as to date of return;
(c) the elapsed duration of absence from the job which, wherever relevant, may be that duration for which workers can receive compensation benefits without obligation to accept other jobs.
It should be noted that the wording of the international definition implies that the three criteria of job attachment need not be met simultaneously in every situation; appropriate combinations may vary "in the light of national circumstances".4
Regarding temporary absence from self-employment, the international standards are less elaborate than for paid employment, and simply specify that "persons with an enterprise, which may be a business enterprise, a farm or a service undertaking, who were temporarily not at work during the reference period for some specific reason" should be considered as employed.
Persons who are considered to be without "formal job attachment" are not to be classified as employed, but, depending on their current availability for work and/or their recent jobsearch activity, could be classified as unemployed.
In applying the international standards in particular cases, it is necessary to interpret the principles in order to develop more specific guidelines. with respect to the notion of "formal job attachment", it is generally agreed that the second criterion (b) should be the essential one, since in effect it means a return to normal conditions: a return to the same job, or, more generally, to a job with the same employer. Furthermore, by "absence from work" in every day parlance, one generally means a time interval in between two periods of employment with the same employer. Therefore, there should be an assurance of a return to work (or at least a reasonable expectation of a return to work) for the absence to be considered as part of employment.
With regard to the other two criteria (a) and (c), an approach that could be adopted is the following:
For example, if a leave is unpaid, then the duration of absence should be fairly short for the person on leave to be considered as having a "formal job attachment" (assuming of course that there is in any case an assurance to a return to work). On the other hand, if the leave is fully paid, then a longer duration of absence may be considered as acceptable.
The above approach is now applied to the case of maternity leave and of child care leave, and for the purpose of illustration also to two other types of leave mentioned in the introduction, namely, unpaid leave initiated by the employer (or administration) and unpaid leave initiated by the employee.
For the purpose of statistical measurement, persons on maternity leave should be considered as in "employment" if they maintain "formal attachment" to their job during the reference period.
As mentioned earlier, according to the international definition of employment (ICLS 1982), "formal job attachment" should be determined in the light of national circumstances, using one or more of the following criteria: (a) "continued receipt of wage or salary"; (b) "assurance to return to work"; and (c) "elapsed duration of absence".
In the context of maternity leave, it is proposed that the second criterion (b) be regarded as the essential one, supplemented by either (a) or (c).
The first criterion (a) should be interpreted as meaning that payments received during maternity leave form part of the wage and salary bill of the enterprise or an employers' insurance fund. Payments drawn from other sources such as social security or public insurance funds should not be considered as "wage or salary". The third criterion (c) should be assessed on the basis of the length of time elapsed from the start of the maternity leave to the present time, i.e., to the end of the reference period used for measuring employment. The length of this elapsed time should could be somewhat long, for example, up to 12 or 14 weeks5.
Persons on maternity leave without "formal job attachment" who, during the reference period, were available for work and actively seeking work should be classified as unemployed like other categories of workers. In certain circumstances, the seeking work criterion could be relaxed and in which case persons on child care leave would be classified as unemployed only if currently available for work.
For the purpose of statistical measurement, persons on child care leave should not be considered as having "formal job attachment", and therefore should not be classified as in employment, unless they were engaged in another economic activity during the reference period used for the statistics.
Similar to the case of maternity leave, persons on child care leave who were not engaged in any economic activity during the reference period, but who were available for work and actively seeking work should be classified as unemployed like other categories of workers. In certain circumstances, the seeking work criterion could be relaxed and in which case persons on child care leave would be classified as unemployed only if currently available for work.
Unpaid leave initiated by the employer (or administration) refers to a situation where, for reasons of labour redundancy or adverse economic conditions or for other circumstances the employer or administration gives leave, admitted by law, to an employee for a specified or an indefinite period, during which the employee does not receive any wage or salary. This unpaid leave may in certain cases be an extension to a paid leave granted by the employer (or administration) during which the employee was receiving full or part remuneration.
For statistical purposes, it is proposed that the situation of "unpaid leave initiated by the employer" be assimilated to that of "temporary or indefinite lay-off without pay" as specified by the existing international guidelines.6
A person on temporary or indefinite lay-off without pay is one whose contract of employment, or activity, has been suspended by the employer for a specified or unspecified period at the end of which the person concerned has a recognised right or recognised expectation to recover employment with that employer.
According to the present international standards, persons on "lay- off" are to be classified in any one of the categories "employed", "unemployed" and "not in the labour force", depending upon the nature of the attachment to their jobs, their job search activity and their current availability for work. In particular, (i) lay- offs with formal job attachment are to be classified as employed (with a job but not at work); (ii) those without formal job attachment but seeking and currently available for work are to be classified as unemployed; and (iii) others, i.e. lay-offs without formal job attachment who are not currently available for work or not seeking work are to be classified as not in the labour force.
It is proposed to assess "formal job attachment" in this particular case in terms of the combination of criterion (b) assurance of return to work and criterion (c) the length of the elapsed time of absence which should be fairly short, no more than, say, 4 weeks or one month, but in any case not longer than the duration of the statuary annual leave.
Unpaid leave initiated by the employee is a situation where an employee requests the employer (or administration) to be granted a leave of a specified duration for personal or family reasons. This unpaid leave may in certain cases be an extension of a statutory leave such as annual leave, sick leave or other types of paid leave, or a follow-up to a paid leave initiated by the employer (or administration).
For statistical purposes, it is proposed that persons on unpaid leave initiated by themselves be considered having a formal attachment to their job and therefore classified as employed, if they have an assurance to return to work at the expiration of the leave (criterion b) and the length of the elapsed time of absence is less than a specified period, say, no more than four weeks or 30 days, or in any case not longer than the duration of the statuary annual leave (criterion c).
For the sake of consistency, the upper limit set for the duration of absence in criterion (c) should be the same as that specified in the case of unpaid leave initiated by the employer (or administration). This is important because an unpaid leave initiated by an employee may in fact be a leave initiated in disguise by the employer (or administration). If that is the case, the statistical treatment should be the same whoever has initiated the leave.
Survey Starting
Country frequency date
Albania - -
Armenia - -
Azerbaijan - Apr 95 (Pilot)
Belarus - Sep 95 (Pilot)
Bulgaria A Sep 93
Croatia - Oct 93 (Pilot)
Czech Republic Q xxx 92
Estonia A Feb 95
Georgia - -
Hungary Q xxx 92
The former Yugoslav
Republic of Macedonia - -
Moldova - Sep 95 (Pilot)
Kazakhstan - -
Kyrgyzstan - -
Latvia - Sep 95 (Pilot)
Lithuania A Apr 94
Poland Q xxx 92
Romania A Mar 94
Russian Federation A Oct 92
Slovak Republic Q xxx 92
Slovenia A May 93
Tajikistan - -
Turkmenistan - -
Ukraine - Nov 93 (Pilot)
Uzbekistan - -
Note: A = Annual; Q = Quarterly.
_______________________________
1 Resolution concerning statistics of the economically
active population, employment, unemployment and
underemployment, adopted by the Thirteenth International
Conference of Labour Statisticians (October 1982, ILO Geneva),
reproduced in ILO Bulletin of Labour Statistics, Geneva, 1983-
3, pp. XI-XV.
2 "Treatment of Selected Items in Employment and
Unemployment Statistics in Transition Countries," by Jaroslav
Kux, ILO Consultant, Prague, February 1995.
The statistics issue of maternity (and parental leave, in
general) is also discussed with regard to OECD countries in
"Parental leave: Statistical analysis and impact on the labour
market," Working Party on Employment and Unemployment
Statistics, DEELSA/ELSA/WP7(94)5, OECD, Paris, 5 September
1994.
3 Conditions of Work Digest on Maternity and Work, Volume
13, ILO, Geneva, 1994.
4 Further details on the notion of "formal job attachment"
may be found in Hussmanns, R., Mehran, F. and Verma, V.,
Surveys of economically active population, employment,
unemployment and underemployment: An ILO manual on concepts
and methods, ILO Geneva, 1990, pp. 72-75.
5 These limits are those established, respectively, by the
ILO Convention concerning Maternity Protection (No. 103,
Revised 1952), Article 3, paragraph 2, and the corresponding
ILO Recommendation No. 95, subparagraph 1(1).
6 Hussmanns, R., Mehran, F., and Verma, V. (1990), pp. 103-
4.