Trends in feminization of the teaching profession in OECD countries 1980-95

Working papers are preliminary documents circulated to stimulate discussion and obtain comments

The aim of this report is to examine the feminization of the teaching workforce in primary and secondary education in OECD countries from 1980 to the present, and to see whether the present level of feminization or any changes in the proportion of women teachers over the period are associated with changes in aspects of employment, including salary, or with changes in government spending on education.

The Working Group of the European Trade Union Committee for Education (ETUCE) used the term feminization -

This study was commissioned by the ILO to provide further analysis of material gathered in a preliminary report, and any further statistical material available to the author between mid-September and mid-November 1999. This is mainly already published material, including the invaluable material collected by the European Commission in its series Key data on education, OECD figures, and some material supplied by Education International member unions. In addition, the study draws on two analyses of occupational segregation and feminized occupations (OECD 1997, OECD 1998), and the ETUCE Colloquium on Equal Opportunities, held in 1995.

It proved difficult to track down relevant statistical material in the time available. Material on some relevant factors does not appear to be accessible, particularly attrition and return rates and employment status (permanent/limited-term/casual; collective contract/individual contract). A further difficulty is that statistics on workload, income and positions of responsibility are not generally disaggregated by gender, particularly over time, and statistics on teaching salary are not disaggregated by employment status and part-time/full-time status. Comparisons with other occupations are often made piecemeal, or with other feminized occupations, rather than the occupations which might be attracting men away from teaching. Data on these may also be newer, which does not help trend analysis through time.

A further limitation is that comparable material is not available for every OECD country, or for the same years. While the study endeavours to compare only what appears comparable, individual country experts may encounter assumptions made in the study with which they would not necessarily concur.

The study starts by providing an overall picture of changes since 1980 in the OECD countries in terms of feminization of the school teaching workforce. This yields a set of categories which are then used to analyse other aspects of the teaching profession to see whether there is any association between them, rates of change in them, and feminization levels and rates of change.

By 1980, women made up more than half the primary teaching workforce in most OECD countries, and were at least 40 per cent of the secondary teaching workforce in most of the 12 countries for which this information is available. The main reasons why teaching had become a highly feminized profession in comparison with others by 1980 are usually given as:

A 1995 ETUCE colloquium on equal opportunities raised a number of issues related to the comparatively high and increasing proportion of women in the teaching profession:

Other issues raised included questions of whether increased feminization of teaching would have any impact on student behaviour and achievement, particularly for boys.

There are two ways in which changes in feminization can be compared across countries. First, by looking at changes between levels of feminization, in terms of absolute proportions of women in teaching. The second way is to compare rates of change, in terms of increases as a proportion of the base level in 1980. Both methods are used in the following analysis, comparing the patterns of change in three different groups of countries, in terms of whether their level or rate of feminization is high, medium, or low.

Women teachers made up more than half of the primary teachers in most OECD countries by 1980. By 1995, just over half the 19 countries for which information on feminization levels was available had a primary teaching force that was predominantly female, over 70 per cent. The highest level of feminization was in the Czech Republic, 93 per cent, and the lowest, 44 per cent, in Turkey.

The OECD countries fall into three categories in terms of the feminization of their primary teaching profession in 1980. Countries in the high category had a feminization level of 70 per cent or more. Medium category countries had a feminization level of 50-69 per cent, and low category countries, less than 50 per cent. The countries which increased their proportion of women in the primary teaching profession over the 15 years between 1980 and 1995, and moved up a category, are in bold. ⁽¹⁾

High feminization - 70 per cent or more of the teaching profession is female


7 out of 19 countries in 1980	11 out of 20 countries in 1995

Australia	Australia
Austria	Austria
Hungary	Belgium
Ireland	Czech Republic (no data available in 1980)
Italy	France
Portugal	Hungary
United Kingdom	Ireland
	Italy
	New Zealand
	Portugal (no data)
	United Kingdom

None of the originally highly feminized countries' primary teaching professions became less feminized over the period. Italy and the United Kingdom showed little change.

Medium feminization -- 50-69 per cent or more of the teaching profession is female

Again, there was no decrease in feminization of the primary teaching profession, although Canada and Luxembourg showed little change. Three of the eight countries increased their feminization over this period to the extent that they were counted among the "high feminization" countries in 1995.

Low feminization - below 50 per cent or more of the teaching profession is female


4 out of 19 countries in 1980	1 out of 20 countries in 1995

Greece	Turkey
Republic of Korea
Netherlands
Turkey

Three of these four countries had increased feminization over this period sufficiently to join the "medium" category.

Looking at the rate of change ⁽²⁾provides a slightly different grouping. As one would expect, countries whose levels of feminization were already high in 1980 did not have high rates of increase. Four countries had major or high rates of increase in feminization, all with low or medium levels of feminization in 1980, four had medium rates of increase, and ten had low or very low rates of increase. Countries with medium levels of feminization in 1980 varied most in their rate of increase: two had high rates of increase, two medium, two low, and two very low rates of increase.


Major increase in feminization:

Republic of Korea	(Low level 1980)	Rate of change 1980-95:	62 per cent
Netherlands	(Low level 1980)	Rate of change 1980-95:	41 per cent

High increase in feminization:

New Zealand	(Medium level 1980)	Rate of change 1980-95:	23 per cent
Belgium	(Medium level 1980)	Rate of change 1980-95:	22 per cent

Medium increase in feminization:

Greece	(Low level 1980)	Rate of change 1980-95:	17 per cent
France	(Medium level 1980)	Rate of change 1980-95:	16 per cent
Hungary	(High level 1980)	Rate of change 1980-95:	14 per cent
Norway	(Medium level 1980)	Rate of change 1980-95:	12 per cent

Low increase in feminization:

Austria	(High level 1980)	Rate of change 1980-95:	10 per cent
Spain	(Medium level 1980)	Rate of change 1980-95:	10 per cent
Australia	(High level 1980)	Rate of change 1980-95:	9 per cent
Italy	(High level 1980)	Rate of change 1980-95:	8 per cent
Japan	(Medium level 1980)	Rate of change 1980-95:	8 per cent
Turkey	(Low level 1980)	Rate of change 1980-95:	7 per cent
Ireland	(High level 1980)	Rate of change 1980-95:	6 per cent

Very low increase in feminization:

United Kingdom	(High level 1980)	Rate of change 1980-95:	3 per cent
Luxembourg	(Medium level 1980)	Rate of change 1980-95:	2 per cent
Canada	(Medium level 1980)	Rate of change 1980-95:	1 per cent

Proportions of women are much lower overall in secondary education. A high level of feminization in 1980 was 55 per cent or more, rather than the 70 per cent in primary education. The medium category is defined as 41-54 per cent, and the low category, 40 per cent or less.

Growth in feminization was more marked in secondary education, perhaps because of its lower starting base. By 1995, women made up more than half of the teaching profession in secondary schools in two-thirds of the OECD countries for which information was available. The highest rate was 67 per cent, for Canada, and the lowest, 40 per cent, for Turkey (see table 2, appendix, for more details).

High - 55 per cent or more of the teaching profession is female


3 out of 13 countries in 1980	12 out of 18 countries in 1995

France	Austria
Italy	Canada
Portugal	Czech Republic (no data in 1980)
	France
	Greece
	Hungary (no data in 1980)
	Ireland
	Italy
	New Zealand
	Portugal (no data in 1995)
	United Kingdom (no data in 1980)
	United States (no data in 1980)

Medium - 41-54 per cent of the teaching profession is female


5 out of 13 countries in 19880	3 out of 18 countries in 1995

Australia	Australia
Austria	Germany (no data in 1980)
Canada	Spain
Greece
Ireland

Low - less than 40 per cent of the teaching profession is female


5 out of 13 countries in 1980	3 out of 18 countries in 1995

Japan	Japan
Republic of Korea	Republic of Korea
New Zealand	Turkey
Spain
Turkey

The rate of feminization over the period was greater at the secondary level for some countries than the rate of feminization in their primary teaching service. Those countries which started with (relatively) high levels of feminization in 1980 showed low rates of increase. Rates of increase were mixed for those countries which had medium levels in 1980, with two of the six countries showing major or high rates of increase, three showing medium rates of increase, and one, Ireland, with a low rate of increase. Of the four countries with low feminization levels in 1980, two had major rates of increase, one high, and one medium.


Major increase in feminization:
New Zealand	(Low level 1980)	Rate of change 1980-95:	71 per cent
Canada	(Medium level 1980)	Rate of change 1980-95:	52 per cent
Republic of Korea	(Low level 1980)	Rate of change 1980-95:	48 per cent

High increase in feminization:

Spain	(Medium level 1980)	Rate of change 1980-95:	30 per cent
Japan	(Low level 1980)	Rate of change 1980-95:	24 per cent

Medium increase in feminization:

Turkey	(Low level 1980)	Rate of change 1980-95:	15 per cent
Austria	(Medium level 1980)	Rate of change 1980-95:	13 per cent
Australia	(Medium level 1980)	Rate of change 1980-95:	12 per cent
Greece	(Medium level 1980)	Rate of change 1980-95:	12 per cent

Low increase in feminization:

Ireland	(Medium level 1980)	Rate of change 1980-95:	9 per cent
Italy	(High level 1980)	Rate of change 1980-95:	9 per cent
France	(High level 1980)	Rate of change 1980-95:	6 per cent

By 1995, nine out of the 14 OECD countries for which data were available for primary and secondary education sectors had similar levels of feminization, in terms of the categories, high, medium, and low. Half of the 12 countries for which data were available for both 1980 and 1995 for both sectors show different rates of increase in feminization over the period for both the primary and secondary sectors. (see appendix, table 5). There were no clear trends in terms of consistency between feminization levels and rates of change in feminization over the period. This raises the question of why different rates of increase for primary compared with secondary would be experienced in the same countries. In some countries, cultural perceptions of status and gender roles appear to be playing a part (e.g. Turkey). Other factors are likely to include pay, conditions (particularly hours of work), and qualifications. Although this study includes information about changes in salary over the period for some countries (see tables 6 and 7), such information was only available for a few countries, ruling out any analysis of (in)consistency between sectors over the period in terms of any differences in salary movements.

In this section, some of the hypotheses about changes in employment, pay, and spending on education in relation to feminization of the teaching profession are examined, using the categories derived in the section above to compare countries with different levels and rates of feminization over the period 1980 to 1995, in the light of the data available.

Is an increase in the feminization of the teaching profession visible in an increase in the proportion of women headships?

Although women now dominate primary teaching in terms of numbers in all the OECD countries except Turkey, they have yet to gain headships in proportionate numbers. In the late 1980s, the variation was from 3 per cent in Japan to 45 per cent in France, with proportions particularly low in countries with lower numbers of women in teaching, Denmark, the Netherlands, and Finland (OECD, 1998b, pp. 117-118, citing the European Commission's 1996 report Key data on education in the European Union).

Ground had been lost for some countries even though feminization had increased, for the early 1970s to mid-1980s; in the case of Canada, this was thought due to declining enrolments, school closures and the average size of schools. But by the mid-1990s, most European Union countries had an increased proportion of women in headships.

Table 1 shows changes over time in the proportion of school headships occupied by women, and the proportion of headships occupied by women in 1993, the latest date available for most countries for which information was available.

Of the six OECD countries for which both 1985 and 1993 data about primary headships was available, a higher proportion of women gained headships in 1993 than in 1985 in four countries. The proportion remained static in the Netherlands, and fewer women had primary headships in Belgium than eight years previously, probably reflecting the greater loss of schools (or headships) over this period in Belgium than in other countries. The rate of decrease in the proportion of headships held by women in Belgium was around half the proportion of schools closed, indicating that they were not more vulnerable than men in the primary sector. However, the rate of decrease in the proportion of secondary headships held by women in Belgium was twice the rate of school closure.

It is difficult to conclude anything from the available data about any links between increased feminization of the teaching profession and increased likelihood of women taking on headships. Some hypotheses are possible, but they would need testing against at least three points in time rather than the two we have available, and for a wider range of countries, with more information about women's application and success rates, routes from part-time work, and the impact of taking time out of teaching on promotion.

If we look at changes in primary headships, for example, the two countries which had the highest feminization rates in 1980, also had two of the highest proportion of headships filled by women (Italy and the United Kingdom (England and Wales)). However, Sweden and New Zealand had similar levels of feminization in 1980, but, though each country rapidly increased their proportion of women heads by 1993, their proportions were quite different.

Table 1. Women as a proportion of head teachers, 1980-95


Country	Primary						Secondary

	Women as % of teachers 1980	Women as % of teachers 1995	Women as % of head teachers 1985	Women as % of head teachers 1993	Rate of change of women as % of head teachers 1985-93	Rate of change in No. of schools 1988-93	Women as % of teachers 1980	Women as % of teachers 1995	Women as % of head teachers 1985	Women as % of head teachers 1993	Rate of change of women as % of head teachers 1985-93	Rate of change in No. of schools 1988-93

Belgium	59	72	40	36	-10	-20			29	21	-28	-11
Netherlands	46	65	13	13	0	-6
UK (England/ Wales)	78	80	44	50	+14	-5		56	16	22	+38	-5
Italy	87	94	34	46	+35	-4	58	64	27	29	+7	+3
New	66	81	14	27	+93	-3	39	57	15	16	+7	+1
Sweden	67	70	10	46	+460	+91	43	45	16	27	+59	+35
Canada		65		34				67		19
France							55	58	23	29
Luxembourg*									8	14	+57	+12

Note: Differences in years for column heads reflect differences in the years for which data are available for each aspect. * Caution is needed with the Luxembourg figures, given that in 1993 the country had only 29 secondary schools.

France and Italy started with similar levels of feminization of the secondary teaching profession in 1980, and had much the same level of women in secondary headships in 1993, though the gain was more rapid in France than Italy. Sweden and New Zealand started with similar proportions of feminization and women head teachers, but Sweden had a much higher level of women principals in 1993, even though New Zealand's secondary teaching profession had a higher proportion of women.

One could hypothesise from this that while some "critical" mass of women is necessary in each country for women to become more fairly represented among headships, other factors are also playing a part, such as turnover rates, the age profile of principals, central government support for improving the position of women, changes in the role itself, and features of pay and employment conditions which may make the principalship less attractive to men.

While women are not occupying headships in proportion to their numbers in the teaching profession in any of these countries, comparison with other occupations, both female-dominated and not, would probably show that teaching appears to provide women with more opportunity to gain management positions - and increasingly, as countries decentralize education administration, the role of head teacher has emphasised management more than teaching.

Although it was hypothesized that an even greater emphasis on management and financial responsibility would deter women from applying, or being appointed, in fact women have continued to increase their share of headships in two countries with strong decentralization to the school level, United Kingdom and New Zealand. Decentralization to the school level in England and New Zealand has greatly increased workloads as well as management responsibilities for head teachers. Decisions on principal appointments are also being made by school boards of governors, though it is not clear whether they favour women more, or less, or whether more women are now applying.

In 1980, only 5 per cent of primary headships in New Zealand were taken by women; by 1998, the proportion was 35 per cent, and 23 per cent of secondary principals. In England, 57 per cent of primary principals were women, and 27 per cent of secondary principals. One study of principal appointments found that during the 1980s and early 1990s, men were selected for just over one-third of primary headships; by 1993-96, they were selected for 24 per cent, and by 1998, for less than 12 per cent, close to their proportions in the primary teaching force. The author of the study, Alistair Ross, suggests "it is also possible that male teachers no longer see headship as the career opportunity that it once was. Headship has changed a lot over the past eight or nine years". (Budge, 1999). The change is less dramatic in New Zealand, with one study of 1997 appointments showing that women made up 49 per cent of new appointments (Wylie, 1998).

An analysis was also done to see whether women were more likely to be appointed to headships in countries which had higher numbers of schools, per head of population, thus indicating smaller schools, whose heads possibly had less status than their peers in other countries who head larger schools. This did not show any relationship (see appendix, table 3).

However, looking at the rate of change in the number of schools over time (table 1) in relation to the rate of change in the proportion of head teachers who are women suggests that in the primary school sector women may do better when the number of schools expands, or contracts only slightly. This trend is also visible in the secondary sector data with the anomaly of the United Kingdom, where women principals increased markedly as school numbers fell. This anomaly may be related to the change in the role of principal in the United Kingdom as a result of decentralization, discussed above.

While actual class hours make teaching attractive to women needing to combine employment with family responsibilities, there is less part-time work available in teaching than in other female-dominated occupations. The table below shows great variation between countries, and little apparent relation between the degree of feminization of the primary teaching profession and the proportion of teaching jobs that are part time. This may be because there are "major differences in national definitions of part-time workers" (OECD, 1997, p. 8).

Table 2. The role of women in primary teaching, 1994
(in percentages)


Country	Female share of primary teachers			Part time among

	All	Full time	Part time	All primary teachers	Female primary teachers

Austria		83.7
Belgium	71.8	68.2	85.8	20.5	24.5
Canada	66.1	64.8	77.3	10.2	11.9
Czech Republic		92.5
Denmark	58.0	57.9	58.4	27.0	27.2
France	77.8	74.6	95.4	6.2	7.5
Germany	85.0	76.4	96.4	43.1	48.9
Greece		55.4
Ireland	77.2	77.2	88.2	0.1	0.1
Italy	92.9	92.9	94.8	0.0	0.0
Japan	60.2	60.2	59.6	1.9	1.5
Republic of Korea	58.6	58.6	84.4	0.0	0.0
Netherlands	64.5	48.2	91.3	37.9	53.4
New Zealand	81.0	77.0	94.0	23.5	27.2
Spain		72.0
Sweden	78.9	82.3	73.4	38.6	36.0
Turkey		43.0
United Kingdom	81.4	79.3	89.5	19.2	21.1
United States	86.5	86.7	83.5	10.2	9.9

Source: OECD (1998, p. 102).

For example, Belgium has 21 per cent of its primary teachers in part-time jobs, with women comprising 72 per cent of its primary teachers, compared with Ireland, which has women comprising 77 per cent of its primary teachers, but only 0.1 per cent of its primary teachers in part-time jobs.

Table 3 shows quite a mixture of patterns of change in the proportion of teachers in part-time jobs in secondary education in nine countries between 1985 and 1994. In one-third, Australia, Denmark and the Czech Republic, the proportion of part-time teachers declined over this period.⁽³⁾ In another third it stayed much the same, France, Poland, and Sweden, and it increased for the final third (Hungary, Spain, and the United Kingdom).

Table 3. Employment trends for general secondary schoolteachers
in selected countries: Percentage part time, 1985-94


	Percentage part time

	1985	1994

Australia	24.2	20.5
Czech Republic	18.4	14.8
Denmark	3.5	0.5
Finland	8.8
France	17.0	16.6
Hungary	6.3	8.9
Republic of Korea	-	0.2
Poland	12.1	11.4
Portugal	-	3.4
Spain	0.6	1.9
Sweden	39.0	40.4
United Kingdom	7.2	12.5

Source: ILO, 1998

Information on the rate of change in feminization of secondary teaching positions was available for only four of these countries. It is therefore difficult to draw any conclusions from it. However, like the primary sector data, it also shows no clear trends. New Zealand had a major increase in feminization of the secondary teaching profession between 1980 and 1995, yet part-time positions actually declined. Spain had a high rate of increase in feminization, and also saw an increase in the proportion of part-time secondary teaching jobs. However Australia had a medium increase in feminization, but a decrease in part-time work. France had a low increase in feminization, and its part-time positions as a proportion of teaching positions stayed much the same.

Without information on a larger group of OECD countries, and trend data on the proportion of women in part-time jobs over this period, it is difficult to know whether any overall changes in feminization of the teaching profession are reflected (or led by) part-time work figures: in other words, whether feminization at the overall level in teaching is accompanied by fewer opportunities for full-time work. However, there are no clear trends indicating that this is the case.

While part-time work can be more attractive to women with substantial family responsibilities, or who are also undertaking further education or upskilling, there have been concerns that it can hinder women from promotional and professional development opportunities, and that many may be working part time not from preference, but because full-time jobs are not available. In some countries, it is also of lesser status, offering less secure employment, and less access to leave or retirement provision (ETUCE, 1996, p. 18, and pp. 36-41).

There is very little information about the differences between countries in these respects, making it difficult to determine the relative value of part-time teaching work for women, or how these aspects of part-time work are related to different levels and rates of feminization of the teaching profession. ⁽⁴⁾

Data on contractual working time is available for a number of OECD countries only for 1992-93. Contractual work hours at the primary school level were somewhat lower for six countries, ⁽⁵⁾ with high levels of feminization in 1995, where average work hours were 1,114 hours, than for four countries with medium levels, ⁽⁶⁾ where average work hours were 1,250 hours.

Countries whose rate of increase in feminization in the primary teaching profession between 1980 and 1995 was medium or high also had lower average contractual hours (an average of 1,120 hours, for four countries),⁽⁷⁾ compared with those countries whose rate was low or very low (an average of 1,200 hours, for six countries).⁽⁸⁾

Looking at class hours (or teaching time) shows a similar picture. Using 1996 teaching hours for primary schools,⁽⁹⁾ the average for five countries with high feminization levels in 1995 was 833 hours, compared with 840 hours for five countries with medium to low feminization levels in 1995 (see table 4, appendix).

It is not clear which of these measurements of work hours is more reliable, since teachers may work more than their contract hours, and different countries have different expectations of the time teachers should give over and above actual class hours.

The OECD concluded that between 1990 and 1996 "teacher salaries have often been stable or increased in real terms", but that "the increases have generally been slower than growth in GDP per capita" (OECD, 1999, p. 265).

Teachers in OECD countries "fare relatively well when compared with other full-time wage and salary workers" (OECD, 1999, p. 268), with the exceptions of the Czech Republic, Hungary, and Sweden, where primary teachers' pay is comparatively low, and Belgium and Switzerland, where secondary teachers' pay is comparatively high.

However, comparisons with salaries in other high-skilled occupations show that teachers' salaries after 15 years' experience are "lower than the average earnings of university graduates" (ibid), other than for Australia and, for secondary teachers, France and Switzerland.

There are two kinds of data on salary movements available to compare with the levels of feminization in the teaching profession in 1995, and the rate of change in feminization between 1980 and 1995. First, an inter-country comparison of change in teacher salaries measured in terms of per capita wealth is given. This gives some indirect indication of teacher salary levels, but related to per capita wealth, and not other occupational salary levels.

The second comparison across countries is of change in teacher salaries measured in terms of real salary. Court, Morris, Reilly and Williams (1995) cite four studies showing relationships between salary levels and the choice of teaching as a career. These studies were done between 1979 and 1993. Of particular note is the study by Zabalza, done in 1979, using 1963-71 data, which showed that men were more responsive to increases in earnings than women; a 10 per cent increase in average teacher salaries led to a 24 per cent increase in male entrants to teaching, but only a 3 per cent increase in female entrants. It also showed that continued progression and pay played a role in the decision to become a teacher, as did higher unemployment rates.

Court et al. updated the Zabalza study, using 1986-92 United Kingdom data related to newly qualified graduates.⁽¹⁰⁾ They found that neither salary progression nor the unemployment rate significantly affected the proportion of graduates entering teaching, whether male or female. But relative starting salaries remained a significant influence. They suggested that salary progression may be less important than previously because of changes in the labour market which have changed expectations of a "job for life".

They estimated that a 1 per cent fall in relative teacher starting salaries would result in a drop of 4 per cent in the relative supply of graduates into teaching, with, unlike the previous study, no difference between men and women.

However, after holding salary variables and graduate unemployment levels constant, women were more likely to enter teaching than men, indicating that pay was only one of the factors which influenced their decision. Other factors that positively influenced a choice of teaching were marital or cohabiting status, and degree subject, which was influenced by gender.

In this study, relative data for a comparable range of occupations were not available. Nor were data on graduate earnings, which would be preferable, since such data do not involve judgements of which occupations might appeal to those who also consider teaching. This study has therefore used the next best option, minimal and maximum salary rates (which are likely to be closely related to starting rates and rates after maximum progression).

The ratio of the average 1995 minimum primary teacher salary in relation to average per capita wealth was higher in three countries with medium levels of feminization in 1995, with a ratio of 1.32 compared with 1.06 for five countries with high levels of feminization in 1995.

Table 4. Movement of primary teachers' salaries in relation
to per capita wealth


Country		1980-85	1993-95	% change 1980s-90s	Feminization level 1995	% change in feminization, 1980s-1995

Austria	Min	1.05	0.99	-6	High	Low
	Max	2.19	2.16	-1
Belgium	Min	1.25	1.05	-14	High	High
	Max	2.07	1.06	-49
France	Min	0.88	0.83	-6	High	Medium
	Max	1.42	1.29	-9
Greece	Min	1.23	1.22	-1	Medium	Medium
	Max	3.57	1.84	-48
Ireland	Min	1.67	1.38	-17	High	Low
	Max	2.71	2.67	-1
Netherlands	Min	1.08	0.96	-11	Medium	Major
	Max	1.91	1.71	-11
Spain	Min	1.79	1.78	-1	Medium	Low
	Max	2.86	2.30	-20
United Kingdom	Min	0.87	1.04	+20	High	Very low
	Max	1.37	1.76	+28
Germany	Min	1.97	1.59	-19	High	-
	Max	2.62	2.11	-19

Data on salary levels drawn from analysis in ILO (1998) which used data from the Key data in education in the European Union from Eurydice, the ILO October Inquiry 1985-95, OECD National Accounts, and UN Demographic Yearbook.

A similar pattern was evident in relation to maximum primary teacher salaries, with a ratio of 1.95 for the countries with medium levels of feminization in 1995, compared with 1.79 for those with high levels of feminization.

There was no difference in changes in teacher salary movement in relation to per capita wealth related to the rate of feminization over this period. However, a pattern was evident in looking at maximum salary rates, with an average loss of 29.3 per cent for the countries with high or medium feminization rates, compared with an average loss of 12.6 per cent for countries with low feminization rates. There was a gain of 28 per cent for the United Kingdom, with very low feminization rates of change over this period, the only country to show an increase in teacher salaries over this period.

It is harder to draw conclusions from the data for secondary schools, since data on 1995 levels of feminization was available for only six countries, all but one of which had high levels of feminization. Data on rates of change in feminization was available for five countries only, too small to make any valid comparisons using averages.

Table 5. Movement of secondary teachers' salaries
in relation to per capita wealth


Country		1980-85	1993-95	% change 1980s-90s	Feminization level 1995	% change in feminization

Austria	Min	1.13	0.99	-12	High	Medium
	Max	2.54	2.16	-15
France	Min	1.07	0.86	-20	High	Low
	Max	2.06	1.84	-11
Germany	Min	2.21	1.71	-22	Medium	-
	Max	2.92	2.54	-13
Greece	Min	1.33	1.27	-5	High	Medium
	Max	3.88	1.89	-51
Ireland	Min	1.67	1.38	-17	High	Low
	Max	3.88	2.67	-31
Spain	Min	2.07	2.08	0	Medium	Medium
	Max	3.39	2.72	-20
United Kingdom	Min	0.87	1.04	+20	High	-
	Max	1.37	1.76	+28

This section looks at real salary changes over time for OECD countries with available data, using 1985 as the base year (1985 salary = 100). It uses salary data from 1980 or the early 1980s to compare with the feminization data from the same period.

The six countries whose level of feminization in 1995 was high tended to have somewhat higher real minimum salary rates in 1991-93, an average of 112.42, than the three countries with medium levels of feminization in 1995, whose average was 101.5 .


Country		1980-83	1991-93	% change 1980s-90s	Feminization level 1995	% change in feminization

Austria	Min	98.77	119.76	+21	High	Low
	Max	98.28	123.94	+26
Belgium	Min	110.48	111.40	+1	High	High
	Max	108.69	101.82	-6
France	Min	90.67	102.38	+13	High	Medium
	Max	95.96	105.37	+10
Greece	Min	80.94	81.27	0	Medium	Medium
	Max	125.28	65.20	-48
Ireland	Min	96.46	124.16	+29	High	Low
	Max	83.07	127.65	+54
Italy	Min	90.28	86.84	-4	High	Low
	Max	104.83	88.25	-16
Netherlands	Min	163.90	105.50	-36	Medium	Major
	Max	162.00	106.20	-34
Spain	Min	87.18	117.58	+35	Medium	Low
	Max	104.71	114.06	+9
United Kingdom	Min	85.56	129.99	+52	High	Very low
	Max	85.26	140.14	+64

A different pattern was evident for maximum salary rates. Countries with high levels of feminization in 1995 had an average maximum salary which was 119.8 in relation to the 1985 base, compared with an average increase to 105.9 for countries with medium feminization levels in 1995.

Those countries whose rate of change in feminization had been very low or low between 1980 and 1995 had an average increase of 26.6 per cent on the base minimum salary rate of 1985; those countries whose rate of change in feminization had been medium or higher, had an average decrease of 5.5 per cent on the 1985 base salary rate.

A similar pattern was evident in looking at changes in maximum salary rates. Countries which had had a medium or high rate of increase in the feminization of their primary teaching profession had an average decrease of 19.5 per cent on the base maximum salary in 1985, compared with a gain of 27.4 for the countries whose rate of increase in feminization had been low or very low.

Ideally, one would compare changes in real salary rates for male and female teachers over the time period, to test the hypothesis that teaching becomes less attractive to men as the rate of feminization increases, because of changes in pay. In this respect, it is interesting, and unsupportive of the hypothesis, to find that male primary teachers made greater real salary gains than female teachers (100 per cent compared with 64 per cent) in the Republic of Korea, which had a major increase in feminization over the period.⁽¹¹⁾

Data is only available for seven countries, with insufficient numbers in the low-medium category of feminization levels, and in the major-high category for changes in the rate of feminization. This makes any analysis speculative.

Table 7. Real salary index rate changes for secondary teachers


Country		Early 1980s	1993	% change 1980-90s	Feminization level 1995	% change in feminization

Austria	min	98.93	111.15	+12	High	Medium
	max	98.83	107.27	+9
France	min	98.11	95.87	-2	High	Medium
	max	98.11	105.79	+8
Greece	min	87.72	84.37	-4	High	Medium
	max	136.16	66.84	-51
Ireland	min	96.46	124.16	+29	High	Low
	max	83.07	127.65	+54
Italy	min	89.69	88.29	-2	High	Low
	max	106.03	87.51	-17
Spain	min	84.74	115.05	+36	Medium	High
	max	102.82	111.51	+8
United Kingdom	min	85.56	129.99	+52	High	-
	max	85.26	140.14	+64

The next table shows the level of GNP spent on education in the OECD countries for which information is also available on the proportion of women in the teaching profession.

Table 8. Changes in expenditure on education as a proportion
of GNP and feminization levels and rates

Education
spending as %
GNP change
1980-95

1995 % GNP
spent on
education

Primary

Secondary

Rate of
feminization
increase

Feminization
level 1995

Rate of
feminization
increase

Feminization
level 1995

Australia

5.6

Low

High

Medium

Austria

5.7

Low

High

Medium

High

Belgium

-48

3.1

High

Canada

7.0

Very low

Medium

Major

High

France

+22

6.1

Medium

High

Low

High

Greece

+45

2.9

Medium

High

Hungary

+13

5.3

Medium

High

Ireland

-3

6.1

Low

High

Low

High

Italy

4.7

Japan

-38

3.6

Low

Medium

High

Low

Republic of Korea

3.7

Major

Medium

Major

Low

Netherlands

-32

5.2

Major

Medium

New Zealand

+17

6.8

High

Major

High

Norway

+25

8.1

Medium

Spain

+113

4.9

Low

Medium

High

Medium

Sweden

-10

8.1

Turkey

2.2

Low

Medium

United Kingdom

-4

5.4

Very low

High

Portugal

+45

5.5

High

Source for education spending data: ILO (1998) report using UNESCO database.


8 out of 19 countries in 1980	8 out of 20 countries in 1995

Belgium	Canada
Canada	Greece
France	Japan
Japan	Republic of Korea
Luxembourg	Luxembourg
New Zealand	Netherlands
Norway	Norway
Spain	Spain

Trends in feminization of the teaching profession in OECD countries 1980-95

Cathy Wylie

Trends in feminization of the teaching profession
in OECD countries 1980-95