As much as for any other social program, the demand for education is influenced vitally by demographic factors. After a long period of expansion, the education system of most of the countries has entered or is about to enter a period of decline in the number of students as a consequence of low fertility rates. In principle, this decline should yield savings in expenditure to offset some of the increased pension and health costs associated with the aging of the population. This chapter estimates the magnitude of such possible savings, but argues that the potential savings implied by projected demographic developments may not be fully realized because of the specific cost structures and structural rigidities of the education systems in some countries.
The Structure of the Education Sector
At first glance, the structure of the education systems of all seven major industrial countries appears to be similar: 12–13 years of schooling are offered, of which 8–10 are compulsory. In the past two decades, there has also been a tendency to lengthen the years of compulsory schooling, either to unify different school systems or to alleviate the pressure of the growing population on the labor market. Differences across countries are primarily reflected in the number of years associated with different levels of schooling (Table 29). Nevertheless, there are common features other than the number of years of schooling offered. Public schools predominate, and the government exerts an important supervisory role over the curriculum and the examination system in private institutions. Private sector activities in regard to education play a subordinate role within the overall system and have a sizable share, if at all, only at the preprimary and postsecondary or tertiary levels (primarily in Japan and the United States). Public policies on education in all countries, with the exception of France, are the concern of provincial or state governments and not of the central government. Furthermore, teachers usually have lifetime tenure in the public sector, and in a number of countries teachers’ benefits have been adjusted over time to the level received by other civil servants. The special employment status of teachers may introduce downward rigidities that may constrain employment policy options in the education sector in the face of declining enrollment.
Past Developments
Expenditure on education grew rapidly as a percentage of GDP in all countries under study between 1960 and 1975, with the expenditure ratio increasing by more than 50 percent, to 5 percent of GDP, by 1975 in most countries. The growth was especially pronounced in Canada, with the ratio of education expenditure to GDP reaching 8 percent. After 1975, education expenditure declined somewhat in most countries except for Italy and Japan, where education expenditure expanded as a percentage of GDP throughout the 1970s.
Two main factors appear responsible for past developments. First, a substantial increase in enrollments occurred (Table 30), reflecting the movement of the postwar “baby boom” generation into school age, compulsory schooling, and a rising demand for non-compulsory higher education (Table 31). Attendance at preprimary education facilities increased continuously in most countries between 1950 and 1980, primarily as a response to the higher labor force participation rates of women. Second, the cost per student increased during the 1970s. This was principally the result of increasing teacher-pupil ratios, as demographic pressures abated, but also was influenced by increases in teachers’ salaries. Several aspects of these developments are worth noting.
The expansion of the education system in the 1960s and early 1970s took place at a time of generous public-spending policies. Education was perceived as a vehicle for advancement and opportunity for the poor. Education was also considered to be a key factor in sustaining and promoting steady economic growth. With the rapidly increasing size of the student age group, governments accommodated the increased demand for education by expanding educational facilities and increasing the teaching staff. Expenditure on education began to decline only after 1975, coinciding with the decline of the total school population in all countries except Japan and with a more restrictive fiscal policy stance. Although there was a decline in overall enrollment by 1975, different school levels were affected at different times by the movement of the postwar “baby boom” generation through the education system.
The expansion of the education system was reflected in increased investment during the 1960s and early 1970s, when more facilities were provided. This expansion was paralleled by an increase in the number of teaching staff. Recurrent education expenditure continued to grow even when investment declined sharply. Owing to a lack of teachers in the 1950s and 1960s, teacher remuneration improved substantially over time to attract more students to this profession. Consequently, over the last 30 years, the teacher-pupil ratio increased markedly at the primary level. Between 1950 and 1975 the improvement in this ratio at the primary level was due to the number of staff increasing faster than the number of students, while thereafter the ratio improved further because of the decline in the number of students.
The employment of teachers was not adjusted to the lower number of students, since in most countries teachers have lifetime tenure and thus cannot be readily laid off. Most governments, except for the United States, used the opportunity of the abating demographic pressure to deliberately allow for an improvement in the teacher-pupil ratio as a way of strengthening the quality of the educational system. At the secondary level, an overall improvement in the teacher-pupil ratio was not recorded, as increases in the number of staff could barely keep pace with the growing number of students in the 1970s. However, because of the special status of teachers, declining enrollment at the secondary level in the 1980s is likely to result in an improvement of the teacher-pupil ratio at this level as well.
Education Expenditure Projections, 1980–2025
The projections of the baseline demographic scenario suggest that only Canada. France, and the United States will be able to maintain the present size of their school populations over the next 45 years (Table 11). All other countries will experience a sharp reduction in enrollments between 1980 and 2025. In the next 20 years, the lowest number of students will be recorded in the Federal Republic of Germany and Italy (by 1990) and in the United Kingdom (by 1995). Thereafter, overall enrollment will increase only moderately and reach local peaks in all countries except Japan between 2000 and 2005. Japan will experience this peak somewhat later. From then on a continuous decline in enrollments will take place in the Federal Republic of Germany, Italy, Japan, and the United Kingdom until the end of the projection period. By 2025, the school population in the Federal Republic of Germany will be almost 35 percent lower than in 1980, in Italy 24 percent lower, and in Japan and the United Kingdom 15 percent lower. These movements reflect a combination of diverse trends in enrollment at different levels of schooling.
Overall School Enrollments, 1980–2025
(Index: 1980 = 100)
Overall School Enrollments, 1980–2025
(Index: 1980 = 100)
Country | 1980 | 2000 | 2010 | 2025 |
---|---|---|---|---|
Baseline scenarios | ||||
Canada | 100 | 99 | 97 | 100 |
France | 100 | 101 | 99 | 103 |
Germany, Fed. Rep. of | 100 | 84 | 74 | 66 |
Italy | 100 | 82 | 84 | 77 |
Japan | 100 | 81 | 88 | 86 |
United Kingdom | 100 | 86 | 86 | 85 |
United States | 100 | 101 | 99 | 103 |
“Greater aging” scenarios | ||||
Canada | 100 | 99 | 97 | 99 |
France | 100 | 97 | 94 | 92 |
Germany. Fed. Rep. of | 100 | 76 | 64 | 52 |
Italy | 100 | 78 | 74 | 64 |
Japan | 100 | 80 | 81 | 68 |
United Kingdom | 100 | 85 | 80 | 73 |
United States | 100 | 94 | 85 | 77 |
Overall School Enrollments, 1980–2025
(Index: 1980 = 100)
Country | 1980 | 2000 | 2010 | 2025 |
---|---|---|---|---|
Baseline scenarios | ||||
Canada | 100 | 99 | 97 | 100 |
France | 100 | 101 | 99 | 103 |
Germany, Fed. Rep. of | 100 | 84 | 74 | 66 |
Italy | 100 | 82 | 84 | 77 |
Japan | 100 | 81 | 88 | 86 |
United Kingdom | 100 | 86 | 86 | 85 |
United States | 100 | 101 | 99 | 103 |
“Greater aging” scenarios | ||||
Canada | 100 | 99 | 97 | 99 |
France | 100 | 97 | 94 | 92 |
Germany. Fed. Rep. of | 100 | 76 | 64 | 52 |
Italy | 100 | 78 | 74 | 64 |
Japan | 100 | 80 | 81 | 68 |
United Kingdom | 100 | 85 | 80 | 73 |
United States | 100 | 94 | 85 | 77 |
Since the cost per student increases with the level of education (Table 32), it is useful to examine briefly the changes projected over time in the distribution of students at the different school levels (Table 12). By 2025, Canada, France, and the United States will experience a higher enrollment at the preprimary and primary levels, and a decline at the secondary and tertiary levels, compared with 1980. Japan will experience an increase in enrollment at the tertiary level, while at all other educational levels, particularly the preprimary and primary levels, it will register a lower enrollment by 2025, compared with 1980. The decline in the overall enrollment in the United Kingdom between 1980 and 2025 is concentrated at the secondary and tertiary levels, while preprimary enrollment can be expected to be higher at the end of the projection period, compared with 1980. The Federal Republic of Germany, which might have the strongest decline in overall enrollment among the seven major industrial countries, will experience this decline, like the United Kingdom, at the secondary and tertiary levels. However, in the Federal Republic of Germany, enrollment at these levels will be half the size, compared with 1980, while in the United Kingdom enrollment at these levels will be only 20 percent lower.
School Enrollments by Level, as Projected in the Baseline Scenarios, 1980–2025
(Index: 1980 = 100)
School Enrollments by Level, as Projected in the Baseline Scenarios, 1980–2025
(Index: 1980 = 100)
Country | 1980 | 2000 | 2010 | 2025 |
---|---|---|---|---|
Preprimary level | ||||
Canada | 100 | 108 | 111 | 113 |
France | 100 | 108 | 106 | 110 |
Germany, Fed. Rep. of | 100 | 104 | 85 | 82 |
Italy | 100 | 100 | 88 | 88 |
Japan | 100 | 79 | 85 | 79 |
United Kingdom | 100 | 109 | 103 | 106 |
United States | 100 | 107 | 109 | 112 |
Primary level | ||||
Canada | 100 | 110 | 106 | 110 |
France | 100 | 102 | 97 | 102 |
Germany, Fed. Rep. of | 100 | 101 | 80 | 76 |
Italy | 100 | 85 | 82 | 76 |
Japan | 100 | 76 | 86 | 79 |
United Kingdom | 100 | 93 | 89 | 90 |
United States | 100 | 107 | 102 | 109 |
Secondary level | ||||
Canada | 100 | 94 | 92 | 94 |
France | 100 | 100 | 98 | 98 |
Germany, Fed. Rep. of | 100 | 63 | 64 | 51 |
Italy | 100 | 73 | 83 | 72 |
Japan | 100 | 83 | 92 | 91 |
United Kingdom | 100 | 78 | 82 | 79 |
United States | 100 | 100 | 95 | 100 |
Tertiary level | ||||
Canada | 100 | 82 | 87 | 83 |
France | 100 | 82 | 100 | 111 |
Germany, Fed. Rep. of | 100 | 61 | 70 | 57 |
Italy | 100 | 85 | 85 | 83 |
Japan | 100 | 103 | 90 | 103 |
United Kingdom | 100 | 80 | 86 | 83 |
United States | 100 | 85 | 91 | 89 |
School Enrollments by Level, as Projected in the Baseline Scenarios, 1980–2025
(Index: 1980 = 100)
Country | 1980 | 2000 | 2010 | 2025 |
---|---|---|---|---|
Preprimary level | ||||
Canada | 100 | 108 | 111 | 113 |
France | 100 | 108 | 106 | 110 |
Germany, Fed. Rep. of | 100 | 104 | 85 | 82 |
Italy | 100 | 100 | 88 | 88 |
Japan | 100 | 79 | 85 | 79 |
United Kingdom | 100 | 109 | 103 | 106 |
United States | 100 | 107 | 109 | 112 |
Primary level | ||||
Canada | 100 | 110 | 106 | 110 |
France | 100 | 102 | 97 | 102 |
Germany, Fed. Rep. of | 100 | 101 | 80 | 76 |
Italy | 100 | 85 | 82 | 76 |
Japan | 100 | 76 | 86 | 79 |
United Kingdom | 100 | 93 | 89 | 90 |
United States | 100 | 107 | 102 | 109 |
Secondary level | ||||
Canada | 100 | 94 | 92 | 94 |
France | 100 | 100 | 98 | 98 |
Germany, Fed. Rep. of | 100 | 63 | 64 | 51 |
Italy | 100 | 73 | 83 | 72 |
Japan | 100 | 83 | 92 | 91 |
United Kingdom | 100 | 78 | 82 | 79 |
United States | 100 | 100 | 95 | 100 |
Tertiary level | ||||
Canada | 100 | 82 | 87 | 83 |
France | 100 | 82 | 100 | 111 |
Germany, Fed. Rep. of | 100 | 61 | 70 | 57 |
Italy | 100 | 85 | 85 | 83 |
Japan | 100 | 103 | 90 | 103 |
United Kingdom | 100 | 80 | 86 | 83 |
United States | 100 | 85 | 91 | 89 |
In the “greater aging” scenario, the downward trend in enrollment is accentuated relative to the baseline scenario. In the Federal Republic of Germany, school enrollment by 2025 is projected at less than half its 1980 level, while in Italy, Japan, and the United Kingdom, a 30 percent decline in enrollments can be envisioned.
Table 13 illustrates how the development of overall enrollment and the changes in the distribution across educational levels will influence absolute expenditure on education, assuming that costs per student by educational level and educational participation rates remain at their 1980 level. Structural changes in enrollment in France show no impact on overall expenditure. In contrast, in Canada and the United States expenditure on education is projected to be somewhat lower by 2025, compared with 1980, because of lower enrollment at the tertiary level, notwithstanding unchanged overall enrollments. In the Federal Republic of Germany, absolute expenditure will fall more than the overall enrollment level, because enrollment at the secondary and tertiary levels will be markedly lower by 2025 than in 1980. In other countries, the structural changes are not projected to have a significant impact on the level of overall absolute expenditure.
Demographic Effect on Education Expenditure, 1980–20251
(Index: 1980 = 100)
In absolute terms.
Demographic Effect on Education Expenditure, 1980–20251
(Index: 1980 = 100)
Country | 1980 | 2000 | 2010 | 2025 |
---|---|---|---|---|
Baseline scenarios | ||||
Canada | 100 | 96 | 96 | 97 |
France | 100 | 98 | 99 | 102 |
Germany, Fed. Rep. of | 100 | 76 | 72 | 62 |
Italy | 100 | 81 | 83 | 76 |
Japan | 100 | 81 | 88 | 86 |
United Kingdom | 100 | 97 | 89 | 88 |
United States | 100 | 95 | 95 | 98 |
Greaser aging-scenarios | ||||
Canada | 100 | 95 | 91 | 85 |
France | 100 | 95 | 93 | 91 |
Germany, Fed. Rep. of | 100 | 71 | 63 | 50 |
Italy | 100 | 78 | 74 | 64 |
Japan | 100 | 80 | 81 | 68 |
United Kingdom | 100 | 86 | 83 | 76 |
United States | 100 | 90 | 83 | 74 |
In absolute terms.
Demographic Effect on Education Expenditure, 1980–20251
(Index: 1980 = 100)
Country | 1980 | 2000 | 2010 | 2025 |
---|---|---|---|---|
Baseline scenarios | ||||
Canada | 100 | 96 | 96 | 97 |
France | 100 | 98 | 99 | 102 |
Germany, Fed. Rep. of | 100 | 76 | 72 | 62 |
Italy | 100 | 81 | 83 | 76 |
Japan | 100 | 81 | 88 | 86 |
United Kingdom | 100 | 97 | 89 | 88 |
United States | 100 | 95 | 95 | 98 |
Greaser aging-scenarios | ||||
Canada | 100 | 95 | 91 | 85 |
France | 100 | 95 | 93 | 91 |
Germany, Fed. Rep. of | 100 | 71 | 63 | 50 |
Italy | 100 | 78 | 74 | 64 |
Japan | 100 | 80 | 81 | 68 |
United Kingdom | 100 | 86 | 83 | 76 |
United States | 100 | 90 | 83 | 74 |
In absolute terms.
These demographic trends may be expressed in terms of their impact on the ratio of education expenditure to GDP, given assumptions on the likely growth in real educational costs per student over time. With the exception of Japan, it is assumed that the teacher-pupil ratio is held constant at the 1980 level. In Japan, this assumption is made only for the preprimary and tertiary levels of education; at the primary and secondary levels, government policy seeks an increase in the teacher-pupil ratio through the year 1991, such that the ratio of expenditure on primary and secondary education to GDP is assumed constant through that year. Thereafter, the teacher-pupil ratio is assumed to remain constant.
It is also assumed that the cost per student (i.e., salaries, equipment, and supplies) grows at the same rate as productivity in the economy. This last assumption is based on the high labor intensity of the education production function and the limited opportunity for capital-labor substitution. It also reflects the assumption that relative wage rates between teachers and other professions in the economy are unchanged. No differential is assumed between the growth in cost per student and productivity, as all countries, except the United States, suffer from an excess supply of professionally trained teachers. Thus, in the foreseeable future, it does not appear necessary to offer special incentives to attract secondary school graduates to this field.
By assuming constant teacher-pupil ratios, these projections also effectively assume that a decline in enrollment relative to the size of the labor force is reflected immediately in a reduction of expenditure. However, such an assumption reflects neither the high fixed costs present in the education sector nor the institutional restrictions limiting the ability to lay off teachers. Thus, while these calculations may overstate the potential savings over the next 10–15 years, they might capture the correct trend in the long run or represent the maximum savings that can be achieved over time.
The results suggest that the decline in the expenditure ratio will not be as dramatic as the demographic trends might suggest (Chart 13 and Table 14). In fact, the prospective changes in the expenditure ratios are more similar across countries than are the trends in enrollments because of offsetting trends in the size of the labor force in the different countries. Countries such as France and the United States, which according to the baseline demographic scenario will have the same overall school enrollment by 2025 as in 1980, show a reduction in the ratio of education expenditure to GDP because of an increasing active labor force. In contrast, the Federal Republic of Germany shows a 20 percent savings, although the overall enrollment is projected to be 35 percent lower, because the labor force will drop in size over time. However, despite the decline in enrollment in most countries, savings in terms of GDP turn out to be at best 1 percentage point. In the “greater aging” demographic scenario, because of even lower enrollment levels, savings can be expected in the range of 1.0–1.5 percentage points of GDP.

Government Education Expenditure as a Percentage of GDP, 1980–2025

Government Education Expenditure as a Percentage of GDP, 1980–2025
Government Education Expenditure as a Percentage of GDP, 1980–2025
Policy Issues
The seven major industrial countries do not share the same problems in the field of education, as they are affected differently by demographic developments. Since Canada and France will not face dramatic changes in enrollment in the near future, demographics and education are not a political issue. In Japan, issues of the character and quality of education appear to be far more important in terms of the political agenda. Major changes in the curriculum and examination system are the subject of considerable public discussion. Moreover, Japan has adopted a medium-term policy of raising the teacher-student ratio by about 10 percent for elementary and secondary schools over the 1980–92 period. This underlies the assumption that the share of output allocated to primary and secondary education will remain unchanged, despite decreasing primary and secondary school enrollments through 1991.
In contrast, the Federal Republic of Germany and the United Kingdom have had to deal with declining enrollments for a number of years, in particular at the primary level. The authorities have sometimes faced the dilemma of either closing schools because of the low number of pupils or accepting rising costs per pupil. In both countries, the minimum enrollment standard for closing small schools was revised downward in order to maintain a dense network of educational facilities. This policy action might suggest that the authorities would not make use of savings that emerge at times of declining enrollment. However, as demonstrated in the Federal Republic of Germany, such an argument cannot be sustained.
In the early 1980s, the German authorities limited the number of newly hired teachers to a minimum because of declining enrollment and tight fiscal policy at the federal and state levels. While in 1979 about 84 percent of graduates of teacher training institutions were hired, in 1983 this fraction fell to about 25 percent. Owing to the marked decline in enrollment, it would not be necessary to hire even this small fraction of new teachers in order to maintain educational standards. However, by hiring new teachers, the authorities have sought to limit the aging of the teaching staff and further improve the teacher-pupil ratio. This in turn has led to higher increases in costs per student than was assumed in the above projections. However, even if no new teachers had been hired, the size of the teaching staff could not be quickly reduced because of lifetime tenure rules. A reduction of the staff through attrition can occur only slowly. In 1983, 44 percent of the staff was in the 30–40 age group, and only 14 percent was in the 50–65 age group. In the Federal Republic of Germany, the question is not so much how to achieve savings but how to employ additional staff without exceeding budgetary limits—a seemingly impossible goal as the status quo of salaries and benefits of the existing staff is considered inviolable.
In Italy, the adjustment problem is more difficult because of the strong unionization of teachers. Owing to pressure from labor unions, all auxiliary teachers were made regular staff with lifetime tenure some years ago. Furthermore, teachers receive indexation adjustments to their salaries twice a year. This means that, contrary to the assumptions made for Italy, cost increases are likely to exceed productivity growth and inflation.
In the United States, the problems are of a different kind. In the 1970s, initial declines in enrollment were not matched by a reduction in the number of teachers employed. This did not occur until the late 1970s, and was also accompanied by an erosion of remuneration. In 1982, the annual supply of newly qualified teachers fell by half relative to 1970. If students continue to choose careers in other fields because of perceived better salaries and working conditions, a shortage of new teachers is expected to emerge by the end of the 1980s, when the decline in enrollment will bottom out. This shortage might lead to a marked upgrading of teacher remuneration that will increase costs in the education sector above the assumed levels.
Apart from the individual problems that the countries face, some common challenges are likely to emerge. For example, it has often been argued that the real cost of stagnant or declining enrollment is a reduction in the number of new faculty hirings. Since in some countries (e.g., the United States), new faculty at the tertiary level of education are responsible for much research, this could have some effect on the rate of technological progress. Another such challenge might be an increase in the demand for adult education, which might absorb some of the projected savings in education expenditure. These examples show that the assumptions made in the projections may turn out to be optimistic regarding the savings that can emerge in the education sector because of future demographic developments. If there are savings, they are likely to be small and will compensate only to a limited extent for rising costs in other social expenditure categories.