Over the year to 30 June 2002, Scotland’s population fell by 9,400 to 5,054,800, a fall of 0.2 per cent. This relatively slow rate of decline continues the general downwards trend in Scotland’s population evident since 1974. In recent years the main reason for the fall in population has been the decline in the number of births, so that a natural decrease (more deaths than births) has been recorded. In contrast, net migration from Scotland, while fluctuating from year to year, has not had the same marked downwards effect on Scotland’s population which it had in the 1960s or even the 1980s.
The number of births recorded in 2002 was 51,270, an all-time low, while the number of deaths increased slightly to 58,103, an increase of 721 compared with 2001. There were 278 stillbirths and 270 infant deaths (children aged under one) in 2002, both the lowest totals ever recorded. There were 29,826 marriages and 10,826 divorces in 2002, both slightly up on the equivalent figures in the previous year.
While the rate of decline of Scotland’s population in recent years has been relatively slow, much more pronounced changes are occurring in the age structure and geographical distribution of the population. These changes, for example an increasing proportion of older people, and a shift away from most of the larger cities towards the surrounding areas, are projected to continue in the years ahead.
KEY POINTS
POPULATION
The latest estimate of Scotland’s population (30 June 2002) is 5,054,800. Nineteen per cent of the population was aged under 16 and 19 per cent was of pensionable age (60F/65M+) with the remaining 62 per cent of working age (16-59F/64M).
In the 12 months to 30 June 2002, Scotland’s population was estimated to have fallen by 9,400. Two-thirds of this decrease was attributable to natural decline, i.e. more deaths than births. The remaining decline is a result of net out-migration. The migration figures include movements of asylum seekers to Glasgow City and an adjustment for unmeasured migration.
The fall in Scotland’s population in the year to 30 June 2002 should be seen in the context of the relative stability of the population over the last 50 years as shown in Figure 1.1. The population reached a peak of 5.24 million in 1974 and since then has been on a gradually declining trend with some fluctuations.
Figure 1.1 Estimated population of Scotland, actual and projected, 1951-2026
1 2001-based projections
It can be seen from the trends in natural change and migration presented in Figure 1.2 that the population increase up to 1974 was the result of natural change being greater than net emigration from Scotland. But, since 1974, natural change has fallen dramatically as a result of sharp declines in the number of births (over 100,000 in the 1960s to less than 60,000 since 1996), while the number of deaths has remained fairly constant. This fall in natural change was accompanied by a reduction in net emigration from Scotland, but net emigration remained higher than the levels of natural change during the late 1970s and 1980s, causing the population to decline.
Figure 1.2 Natural change and net migration, 1951-2002
Figure 1.3 Estimated population by age and sex, 30 June 2002
Age structure
The age/sex composition is one of the most important aspects of the population, as changes in different age groups will have different social and economic impacts. For example, increases in the elderly population are likely to place a greater demand on health services.
Figure 1.3 shows the age structure of the population for both males and females in 2002. Using past trends in fertility and mortality as a guide, it is possible to explain the peaks and troughs at different ages. Peaks at ages 55 and in the mid-30s reflect the baby boom after the Second World War and in the 1960s. Declining births in more recent years are evident by the tapering of the population under the age of 30. The more stable levels of 10-20 year olds reflect a levelling off of the decline in births during the 1980s when more women were passing through their childbearing ages.
At the more elderly ages, particularly over 75, the higher number of females reflects the longer expectation of life for women, partly as a result of higher rates of male mortality during the Second World War. The effects of the post war ’flu epidemics and lower levels of fertility during the First World War are also evident, as seen in the sharp decline in population aged over 82.
The changing age structure of the population since 1991 is illustrated in Figure 1.4. Of particular note are the decrease of 7 per cent in the number of children under 15 and the increase of 11 per cent in the numbers aged 75 and over. During this time the median age (age at which half the population is older than the value and half is younger) has grown by 4 years from 35 in 1991 to 39 in 2002. The ageing of the population is evident in the large rise of the 45-59 age group and the fall in the 15-29 age group.
The ageing of the population is not unique to Scotland. This pattern of change over the last twenty years is consistent with other countries in the UK and Europe, although the rate of ageing varies.
Figure 1.4 The changing age structure of Scotland’s population, 1991-2002
Changes within Scotland
The map at Figure 1.5 shows the percentage change in population between 1991 and 2002 for each Council area. For comparison purposes it is better to compare over a time frame longer than one year, as population change tends to fluctuate from year to year, particularly for smaller areas. In general, the larger urban areas (apart from Edinburgh) are declining, while areas around the bigger cities and many rural areas are increasing. Generally, urban areas tend to have lower levels of fertility, higher mortality and more out-migration. The areas with a growing population tend to experience net in-migration and an excess of births over deaths.
Figure 1.5 Percentage population change by Council area, 1991-2002
The Council areas which showed the largest relative decreases over this period were Eilean Siar (-11%), Inverclyde (-9%), Glasgow City (-8%) and Dundee City (-7%). The largest relative increases in population occurred in West Lothian (+10%), East Lothian (+8%), Stirling (+7%) and Perth & Kinross (+6%).
Projected population
The latest 2001-based population projections take preliminary account of the results of the 2001 Census, which showed that the base population used in the previous 2000based projections was overestimated. These projections incorporate a revised assumption of net international migration to Scotland, informed by early results of the 2001 Census and taking account of more recent migration information. As a result, long-term migration is assumed to be a loss of 1,000 persons a year, compared with an assumption of nil net migration in the previous projections. Long-term fertility and mortality assumptions are unchanged. The next full set of population projections which take full account of the results of the 2001 Census are due to be published in October 2003.
The overall trend of a slowly declining population is projected to continue with the population projected to fall below 5 million in 2009. In making these projections, assumptions have been made about future levels of fertility, mortality and migration based on past trends. Figure 1.6 shows a widening gap between births and deaths with a natural decrease of over 14,000 a year after 2026. This is projected to be the main reason for population decline in future.
Figure 1.6 Births and deaths, actual and projected, Scotland, 1951-2026
1 2001-based projections
Within this overall decline, significant changes to the age structure are projected (Figure 1.7). The proportion of children under 16 is projected to fall by 19 per cent by 2026, while the proportion of people aged 60 and over is projected to increase by more than a third.
Figure 1.7 The projected percentage change in age structure of Scotland’s population, 2001-20261
1 2001-based projections
The projected population decline by 2026 for Scotland is proportionately higher than that currently projected for any other European country. Indeed, most countries (including other countries within the UK) are projected to increase in population over this period. The underlying difference is that in other countries assumptions about future levels of migration offset projected declines in natural change (births minus deaths).
Longer-term projections to 2050, such as those produced by the United Nations, show the population peaking after 2025 in most European countries and then declining to 2050. Indeed in many EU countries (e.g. Italy, Germany, Spain, Sweden) the population in 2050 is projected to be lower than in 2000. These declines are the result of the natural decrease (more deaths than births) growing to exceed the assumed level of in-migration.
BIRTHS
In 2002, there were 51,270 births registered, the lowest total since civil registration began in 1855. This is a fall of 24 per cent since 1991 and 43 per cent since 1951. Births play a significant role in population change and if recent declines continue this will ultimately have a significant impact on the future level of the working age population.
The number of actual and projected births is shown in Figure 1.6. It shows a peak in births in 1964 then fell dramatically in the late 1960s and early 1970s. After levelling off in the 1980s as a result of the larger number of women, who were born in the baby boom of the 1950s and 1960s, passing through their childbearing years, the decline in births continued, reaching its lowest level in 2002.
More information on births and fertility is given in Chapters 2 and 3. Chapter 2 focuses on recent trends in Scottish fertility, comparing these trends with the rest of the UK and Europe. Chapter 3 places the Scottish fertility experience in a wider geographical context, discusses reasons for low fertility and addresses the scope for policy intervention.
DEATHS
The number of deaths registered in Scotland in 2002 was 58,103, 721 more than 2001. However, this is still one of the lowest totals recorded since civil registration began in 1855 and 5 per cent fewer than 1991 and 12 per cent fewer than 1951.
As would be expected, the majority of deaths occur at older ages – about 58 per cent of deaths were to people aged 75 and over, and a further 27 per cent between the ages of 60 and 75.
Up to 1994, the number of deaths has remained relatively stable at about 60,000-65,000 a year and since 1994 has remained below 60,000 (see Figure 1.6). However, the number of deaths will almost certainly increase over the next 40 years as the large number of people born in the baby boom after the Second World War and the 1960s grow older. While improvements in mortality rates are likely to continue it is unlikely that they will improve at such a rate as to offset the projected increase in deaths resulting from a sharp growth in the number of the elderly.
Stillbirths and infant deaths
The biggest improvements in mortality have been seen in stillbirth, perinatal and infant death rates. Figure 1.8 shows rates of stillbirths, perinatal deaths and infant deaths since 1951. The stillbirth rate has reduced from 26.6 per 1,000 total births (live and still) in 1951 to 5.4 in 2002, despite a change in the definition of stillbirths in 1992 which reduced the minimum period of gestation from 28 weeks to 24 weeks thus increasing the numbers classified as stillbirths. The rate of perinatal deaths (stillbirths and deaths in the first week of life) fell from 44.2 per 1,000 total births in 1951 to 7.6 in 2002, an improvement of 83 per cent. Finally, the infant death rate (deaths of children under one) has improved by 86 per cent from 37.4 per 1,000 live births in 1951 to 5.3 in 2002.
Figure 1.8 Stillbirth, perinatal and infant death rates, Scotland 1951-2002
* Change in definition of stillbirths from 28 to 24 weeks
The infant death rate of 5.3 per 1,000 births in Scotland is the same as the UK rate but is still above the EU rate of 4.6.
Mortality by age
The relative stability in the number of deaths over the last 50 years masks large improvements in age-specific mortality. Figure 1.9 shows the age-specific mortality rates over the last twenty years relative to the 1981 rate for both men and women in the age groups 45-64, 65-74 and 75+ (which between them accounted for about 95 per cent of all deaths in 2002).
There have been greater improvements in male mortality for these ages than females. For the 45-64 age group both males and females experienced an improvement of about 35 per cent. Whilst in the 65-74 age group males showed an improvement of 34 per cent compared to 28 per cent for females. The greatest differential is still in the 75 plus age group where male mortality improved by 21 per cent compared to only 5 per cent for females.
Figure 1.9 Age specific mortality rates as a proportion of 1981 rate, 1981-2002
Although mortality rates have been improving in Scotland, the rate of improvement is slower than elsewhere in the rest of the UK. Since 1976, age standardised mortality rates have improved by 41 per cent in Northern Ireland and 35 per cent in both England and Wales compared with 32 per cent in Scotland.
Improvements in mortality rates in Scotland have generally been slower than in the rest of the UK and elsewhere in Europe. But the improvements are still considerable and the impact is demonstrated in the steadily rising expectation of life for males and females over the last 50 years.
Life expectancy
Life expectancy is a commonly used measure to show the effects of current levels of mortality on the length of time people at various ages can expect to live. It is very useful in comparing the ‘health’ of a nation through time and with other nations as it takes account of the age structure of the population which can differ significantly and so affect the overall crude mortality rate. This is particularly true where countries have a higher proportion of young or elderly within their populations.
Since 1951, there have been considerable improvements in life expectancy at birth as shown in Figure 1.10. Males born in 1951 could, at that time, expect to live to 64.4 years, compared with 73.3 years for those born in 2002. Similarly, females have experienced an improvement of ten years from 68.7 years in 1951 to 78.8 in 2002. Figure 1.10 also illustrates that improvements in life expectancy at birth are projected to continue, rising to 77.4 for males and 82.3 for females by 2041.
Figure 1.10 Expectation of life at birth, Scotland, 1952-2041
1 2001-based projections
2 Provisional
However, despite the improvements in expectation of life, Scottish females have the lowest expectation of life at birth in the EU and males the second lowest. For both sexes, the expectation of life is more than four years lower than the countries with the highest expectation of life and over two years below the EU average.
Regional variations within Scotland
Standardised mortality ratios (SMRs), which compare local death rates with death rates in Scotland as a whole, taking account of the different population structure of each area, are presented in Figure 1.11. In all, 14 out of 32 Council areas have a higher standard mortality ratio than the Scottish average of 100, and 8 of these are in west central Scotland. The worst, Glasgow City, is 22 per cent higher than the Scottish average which itself is about 15 per cent higher than the UK average.
At the other end of the scale the mortality rate in East Renfrewshire was 18 per cent below the Scottish average. Aberdeenshire, East Dunbartonshire, Orkney Islands and Perth & Kinross were all 14 per cent below, or better than, the Scottish average.
Figure 1.11 Standardised mortality ratios, by Council area, 2002
Cause of death
In 2002, the two most common causes of death in Scotland were cancer (26 per cent) and ischaemic heart disease (20 per cent). However, since 1981 the proportion of deaths caused by ischaemic heart disease has fallen from 29 to 20 per cent, whereas the proportion caused by cancer has risen from 22 to 26 per cent. Since 1995, there have been more deaths from cancer than ischaemic heart disease. Of the 15,051 deaths from cancers in 2002, trachea, bronchus and lung was the most common site, accounting for over a quarter (27 per cent) of all cancer deaths.
Death rates, by sex, for the most common causes of death are shown in Table 1.1. Over the last 20 years or so, death rates for men from lung cancer have fallen by 20 per cent (119 per 100,000 population in 1980-82 compared to 95 in 2002). In contrast, rates for women have increased by over 60 per cent (41 per 100,000 females in 1980-82 compared to 66 in 2002), but are still considerably lower than the level experienced by men.
Table 1.1 Death rates from selected causes, by sex, Scotland, 1950-2002
Males – rates per 100,000 population
Year |
Cancer |
Ischaemic heart |
Cerebrovascular |
||
|---|---|---|---|---|---|
All sites |
Trachea, bronchus |
Prostate |
|||
1950-52 |
206 |
48 |
13 |
276 |
155 |
1960-62 |
241 |
86 |
16 |
360 |
166 |
1970-72 |
272 |
112 |
14 |
407 |
158 |
1980-82 |
291 |
119 |
19 |
408 |
139 |
1990-92 |
314 |
111 |
27 |
367 |
119 |
2002 |
320 |
95 |
32 |
255 |
99 |
Females – rates per 100,000 population
Year |
Cancer |
Ischaemic heart |
Cerebrovascular |
||
|---|---|---|---|---|---|
All sites |
Trachea, bronchus |
Breast |
|||
1950-52 |
185 |
10 |
31 |
203 |
213 |
1960-62 |
195 |
13 |
35 |
262 |
230 |
1970-72 |
218 |
24 |
40 |
289 |
226 |
1980-82 |
247 |
41 |
45 |
304 |
210 |
1990-92 |
278 |
57 |
48 |
297 |
191 |
2002 |
278 |
66 |
42 |
210 |
164 |
The next most frequent site for cancer deaths was prostate for men (775 deaths of which 63 per cent occurred in men aged 75 and over) and breast for women (1,105 deaths). Death rates for the former continue to increase whereas those for the latter have shown a slight fall in recent years.
In contrast to the rise in death rates resulting from cancer, death rates for ischaemic heart disease and cerebrovascular disease (stroke) have shown significant declines. Since 1981, males have experienced the larger improvements, 40 per cent for ischaemic heart disease and 32 per cent for stroke, compared with improvements of 33 and 24 per cent, respectively, for females.
Using the latest comparable data available, 2001, Figure 1.12 compares the death rates for the constituent countries of the UK for selected causes after adjusting for differences in age structure. The Scottish rates for cancer, ischaemic heart disease and cerebrovascular disease are well above the rates for the other countries of the United Kingdom for both men and women.
Figure 1.12 Age-adjusted mortality rates, by selected cause and sex, 2001
Source: ONS
In 2002, deaths from intentional self-harm (suicide) numbered 635 (480 males and 155 females), 26 more than in 2001. To allow for any under-recording of suicides, a more robust measure is to combine deaths classified as events of undetermined intent with those for intentional self-harm, as is done in international comparisons. The total number of deaths classified to these two groups was 899 in 2002, compared with 887 in 2001 and 728 in 1981. For men the most frequent cause of these deaths was hanging, strangulation and suffocation, whereas for women it was poisoning.
Summary of main causes of death by age and sex
The main causes of death vary in importance by age and sex (Figure 1.13). Accidents were the most important category of death in the very young (aged 1-14), accounting for 29 per cent of deaths in boys and 22 per cent of deaths in girls in 2002. Cancer, at 15 per cent for boys and 22 per cent for girls, was the next largest cause.
For males aged 15-34, the main category was accidents followed by intentional self-harm (suicide) and mental disorders (entirely due to drug and alcohol abuse). For females in this age group intentional self-harm (suicide) was the largest category followed by cancer.
For almost all age/sex groups above age 35, the main cause was cancer. For women, cancer was responsible for a higher proportion of deaths in almost every age group than for men. Conversely, ischaemic heart disease accounted for a higher proportion of deaths in all age groups for men than for women.
Figure 1.13 Deaths, by cause and age group, Scotland, 2002
MIGRATION
In addition to births and deaths, migration is the other component of population change. However, unlike births and deaths, there is no comprehensive source for estimating migration and hence it is the most difficult component of change to measure. Migration and the reasons for migrating are also much more susceptible to short-term changes in social and economic circumstances than births and deaths. These factors and the fluctuating nature of migration make it very difficult to estimate.
Regular information on the other characteristics of migrants (for example, educational qualifications, country of birth and ethnic group) is limited, although more will become available when detailed results about migrants from the 2001 Census are published later in 2003.
Migration – Revisions to the 1982 to 2000 series
The results of the 2001 Census indicated that the previously published 2000 mid-year estimates had been overestimated by some 50,000 as a result of cumulative errors in estimating migration (mostly young men) during the 1980s and 1990s. As a result, the population estimates for 1982-2000 have been revised with implied changes to migration estimates. It is likely that these migration errors are the result of an underestimation of young male migration from Scotland to the rest of the world during this period, but if moves from Scotland to the rest of the UK have been under-recorded, the implied increase in overseas migration would be less.
Further work is being undertaken to review the quality of the method and data sources used to estimate migration, in particular to reduce the level of unmeasured migration. This includes analysing the detailed migration results from the 2001 Census due out later this year. In addition, a National Statistics Quality Review of International Migration Statistics is being carried out by the Office for National Statistics. This review, to be published later this year, will recommend ways of improving the quality and accuracy of international migration data.
A complete set of migration revisions, which takes account of this further work, will be available at the end of 2003.
Trends in migration since 1951
Estimates of net migration to and from Scotland since 1951 are shown in Figure 1.14. Historically, Scotland has tended to be a country of net out-migration rather than net in-migration, that is more people leave Scotland to live elsewhere than move to live in Scotland. However, since the 1960s the level of net out-migration, some 30,000-40,000 a year, has reduced significantly and in recent years has been less than half of the peak net migration losses in the 1960s. Indeed, in some years during the late 1980s, early 1990s and in 2000-01, Scotland experienced net migration gain rather than loss. There has been an underlying, long-term trend of decreasing net emigration from Scotland over the last 50 years as can be seen from Figure 1.14.
It should be noted that estimates of net migration are the difference between much larger gross flows of migrants into and out of Scotland. In the last 10 years these have typically been of the order of 70,000 both in and out of Scotland. The level of net migration can be significantly affected by relatively small changes in these gross flows from year to year, particularly if one flow rises while the other falls.
Migration to and from Scotland comprises moves to/from the rest of the United Kingdom and moves to/from the rest of the world. In net terms, the contribution to the overall total of moves with the rest of the UK and with countries overseas is roughly similar. This broad relationship has been maintained over the last 50 years, and the net loss from Scotland has reduced since the 1960s and early 1970s both in relation to the rest of the UK and the rest of the world. More information on recent trends will be published with the migration revisions to be published at the end of 2003.
Figure 1.14 Estimated net migration, Scotland, 1951-2002
Migration by age and sex
The age/sex pattern of migrants tends to remain relatively constant from year to year. Figure 1.15 illustrates the age/sex distribution of migrants for males and females moving into and out of Scotland from the rest of the UK between 2001 and 2002. The peak ages for migrating are the late teens to mid-20s reflecting moves out of the parental home to attend higher education or take up employment. There also tend to be smaller peaks for moves of the very young, under the age of five. This reflects migration of parents who move home before their children have started school. The pattern of migration is very similar for men and women though women tend to have much larger peaks in their early 20s than men. However, this may reflect different patterns of re-registering with an NHS doctor after a move rather than different patterns of migration.
Figure 1.15 Movements between Scotland and the rest of the UK, by age, mid 2001–mid 2002
Source: NHSCR
The peaks in migration for males and females in their late teens and early 20s create marked net migration gains at ages 19 and 20, and net migration losses at ages 23 and 24. These patterns are consistent with an influx of students from the rest of the UK and overseas starting higher education followed by a return to elsewhere after completing their education.
Similar age/sex patterns of migration occur for moves between Council areas in Scotland with peak ages occurring in the late teens and early 20s for both males and females.
Migration and the distribution of population within Scotland
Migration has a large impact on population distribution within Scotland and at local level is often the most important component of population change. Net migration rates, that is the amount of net migration between 1991 and 2002 as a proportion of the 1991 population, are a useful indicator when comparing migration between areas of different sizes. Information on net migration rates for Council areas is shown in Figure 1.16.
Generally, areas in the south, east and north of mainland Scotland have experienced migration gains over the last ten years (indicated by a positive rate in the chart), with the largest relative gain taking place in Perth & Kinross (80 migrants per 1,000 population). In contrast, the majority of migration loss was in the west of Scotland with the largest relative rate of migration loss (-72 migrants per 1,000 population) in Glasgow City. For comparison, the rate of migration loss for Scotland as a whole over the period was 2 migrants per 1,000 population.
Figure 1.16 Net migration rates per 1,000 population for Council areas, 1991-2002
The General Register Office for Scotland also registers marriages, divorces, and adoptions in Scotland, and this information sheds light on changing trends in society.
MARRIAGES
In 2002, there were 29,826 marriages in Scotland compared with over 41,000 in 1951. Figure 1.18, however, shows that the decline in the number of marriages may be levelling out at around 30,000 a year. Over 70 per cent of the marriages in 2002 were to couples where at least one partner was resident in Scotland.
Marriages of non-residents
The information on marriages in this chapter covers all marriages which were registered as having taken place in Scotland regardless of the usual residence of the parties involved. For almost 30 per cent of the marriages registered in 2002 neither the bride nor the groom was resident in Scotland and half of these took place at Gretna. For some demographic purposes users might wish to limit analyses to specific categories of residents. Further details of available information may be obtained from the GROS Customer Service address given in Contact Points. Conversely, a number of couples who are resident in Scotland now go abroad to be married. These marriages are not included in this chapter, and only some come to the attention of the Registrar General through notification to British Consular authorities.
Marital status at marriage
Figure 1.17 gives the percentage of marriages by marital status at the time of marriage between 1951 and 2002. The percentage of people marrying who had been divorced rose from only 3 per cent in 1951 to just under 6 per cent during 1971, but by 2002 over a quarter (28 per cent for males and 27 per cent for females) of those marrying were divorcees. The majority of this shift reflects a reduction in the proportion of marriages where one of the partners was a bachelor/spinster. However, the proportion of those marrying who were widowed has also declined – in 2002 the proportion was about 2 per cent or about half of what it was 50 years ago.
Figure 1.17 Marriages by marital status of persons marrying, 1951-2002
When looking at the marital status of persons marrying it is also interesting to look at the remarriage rates, i.e. the number of marriages where one of the participants was widowed or divorced per 1,000 population of widowed or divorced people. It shows a pattern which is consistent with the declining rates of first marriages, although there are differences in the patterns for men and women.
Table 1.2 shows that widowed or divorced men have a much higher propensity to remarry than widowed or divorced women (45.7 per 1,000 population of widowed or divorced men compared with 18.9 for widowed or divorced women). This is the reverse of the pattern for first marriage rates where single women have a higher propensity to marry (35.4 per 1,000 population) than single men (31.7 per 1,000 population). This suggests that women are more likely than men to marry someone who has already been married when first marrying, and that widowed or divorced males are more likely than widowed or divorced females to marry someone who is single when remarrying.
Table 1.2 Marriage rates by sex and marital status, Scotland, 1981-2002
|
Men |
Women |
||
|---|---|---|---|---|
Bachelors1 |
Widowed and |
Spinsters1 |
Widowed and |
|
1981 |
56.1 |
56.7 |
63.1 |
16.6 |
1991 |
41.8 |
46.9 |
49.0 |
18.2 |
20023 |
31.7 |
45.7 |
35.4 |
18.9 |
1 First marriage rates per 1,000 population (single only)
2 Remarriage rates per 1,000 population (widowed and divorced only)
3 2002 rates are based on 2001 Census marital status figures
Marriages by type of ceremony
Civil marriages accounted for nearly 40 per cent of all marriages in 2002. As illustrated in Figure 1.18, this is more than twice the proportion fifty years ago. The trend reflects a move away from religious marriages to civil marriages, particularly during the 1970s and 1980s when the proportion of civil marriages reached current levels. There was a further increase in the early 1990s reaching a high of 46 per cent but this has since fallen back to just under 40 per cent, reflecting an increase in religious marriages, of which a significant proportion were carried out at Gretna. More detailed information on marriages in Gretna is available in the GROS Occasional Paper No. 4 Marriages at Gretna, 1975-2000.
Figure 1.18 Marriages by type of ceremony, Scotland, 1951-2002
1 Includes irregular marriages
Trends in the type of marriage ceremony are likely to be influenced in future by the Marriage (Scotland) Act 2002 and associated Regulations and Guidance which came into effect in June 2002. This changed the law to give people a wider choice of venues for civil marriage in Scotland. By June 2003, 365 venues had been approved for civil marriages outwith registration offices, and over 1,200 civil marriages had been conducted in approved places.
DIVORCES
The number of divorces in 2002 was 10,826, around 200 more than in 2001.
Information on divorces, which relates to divorces granted under the Divorce (Scotland) Act 1976, are for all divorces granted in Scotland regardless of where the marriage took place. Figure 1.19 shows the number of divorces between 1951 and 2002. There was a very marked increase in the number of divorces up to the early 1980s, since when the level has fluctuated. Since 1994, when there were over 13,000 divorces, there has been a steady fall.
Increasing levels of co-habitation may be relevant to the recent decline in divorces since the breakdown of cohabiting relationships is not subject to divorce proceedings.
Figure 1.19 Divorces, Scotland, 1951-2002
Grounds for divorce
Figure 1.20 shows the trend in grounds for divorce between 1981 and 2002. The Divorce (Scotland) Act 1976 introduced new grounds for divorce, principally noncohabitation, meaning that couples separated for two or five years could file for divorce on grounds of non-cohabitation.
Figure 1.20 Divorces, by grounds for divorce, Scotland, 1981-2002
In 2002, non-cohabitation was the most frequent reason for divorce, accounting for 80 per cent of all divorces. Non-cohabitation (2 years and consent) increased from 25 per cent of all divorces in 1981 to over half of all divorces in 2002; non-cohabitation (5 years) increased from 14 per cent to 24 per cent; and adultery as the stated reason for divorce fell from 17 per cent to 4 per cent.
Divorces by marital status
Of those divorcing in 2002, 16 per cent of men and 15 per cent of women had divorced previously. This compares with 8 per cent for males and 7 per cent for females in 1981. This is consistent with the increase in the proportion of all marriages where one or both participants was divorced previously (now 2 in 5 marriages compared with 1 in 4 twenty years ago).
Duration of marriages that ended in divorce
Figure 1.21 compares the cumulative duration of marriages, which ended in divorce in 1981 with 2002. The median duration of marriages ending in divorce is shown by the horizontal line at 50 per cent, indicating that half of marriages ending in divorce lasted for more than this duration and half for less. In 2002, the median duration of marriage was over 13 years, whereas the comparable duration for 1981 was nearly nine years. This increase may be affected by the changing balance between cohabiting relationships, and marriage.
Figure 1.21 Duration of marriages ending in divorce, 1981 and 2002
Divorce by age at marriage
In 2002, 30 per cent of all divorces were to couples where at least one of the partners was aged 20 or under when they married. This is a significant fall from 60 per cent in 1981, but not unexpected given that the proportion of marriages where at least one of the partners was under 20 has fallen from 36 per cent in 1981 to 4 per cent in 2001.
ADOPTIONS
The number of adoptions recorded by the Registrar General during 2002 was 385, the lowest number since the first full year of recording in 1931. This is nearly half the level in 1991 and a quarter of the number adopted in 1951.
Nearly 35 per cent of the children adopted in 2002 were adopted by a step-parent, while over 60 per cent were adopted by non-relatives of the child. 17 per cent of children adopted in 2002 were aged under two, the majority of whom were adopted by non-relatives. In contrast, of the 99 adoptions of children aged over ten, only 21 per cent were adopted by non-relatives.