Marriage Rates in Europe since 1960

by Robert Whiston FRSA   Aug 28th 2013

If you’re not married, then you must be single.

Much has been made in politics and the media about marriage numbers declining and as an institution, marriage is said to be toppling into irrelevance if not obscurity in our ‘brave new world’ where we are brash enough to think we can re-write the rules.

Such is the arrogance of youth that we have to accept that they feel immortal, indestructible and invincible, whereas in reality – and perhaps in this instance – it is more a case of immaturity posing as maturity.

But what is the reality ? What do the numbers tell us ? Is a “Tsunami” about to engulf society and if it is from which direction will it come ? Will it overwhelm marriage as the de fault position of traditional society or will the tsunami devastatingly strike the liberal view that more cohabiting is best ?

If we in Britain thought we alone were the exception in having falling marriage numbers, increases in ‘fatherless’ children, and more births out-of-wedlock, then think again !

Our nearest neighbours, the EU nations, all have very similar patterns in marriage decline as measured by per 1,000 of inhabitants.

Having just completed a review of ‘same-sex marriage’ in Canada, it was clear from their datasets that Canada has seen a collapse in heterosexual marriages far some decades. It was a decline that started in 1945 and then became an upturn in 1960 but which reversed into a near terminal decline in 1970 lasting until the present day (see ).

French_2006Before looking at actual numbers and trend lines for the EU it might be useful if we take one exemplar to explain situation – France in 2006.

From Table 1 (shown right), at the age bracket 15-19 yrs one would expect very few women to have married. And so the figure of 99.75% for single women (highlighted in blue), in this category is not exceptional.

However, journey back 40 years and a different picture emerges. By looking at those women aged 55 or older (highlighted in blue), we can see that only about 8% are single indicating that, over time, spinsters have become married wives. To this extent of 92% of women in 2006 (and assuming they have not remarried) one can presume that 30 years ago they would also have been married.

Thus, the likely percentage of married women could be said to be at a 92% at age 20 – 24. Compare that with 2006. In 2006 it is single women that number around 92% (93.7%).

This transition from single-unmarried to married womanhood can be measured by the ‘average age of first marriage’ data which is collected by all government statistical agencies.

In the 1960s and 1970s it was common for couples to marry in their late teens or early twenties. If you were still unmarried by the age of 27 or 32 you  were considered overlooked,  too old, had  ‘missed the boat’ and been ‘left on the shelf.’ Presently, in the UK the average age of marriage is about 31 years of age, i.e. the age at which women first marry, with men a few years older at the time of first marriage, e.g. aged 34.

So if marriage patterns were today the same as they were in 1970 we would not expect to see the 99% and 93% of French women aged between 15 and 24 in the single status column  (see in Table 1). They would more likely be 15% and 8%.

Real life numbers show that only a small collapse occurs at age 29 – 32 (when they marry). Single women number decline to  71% (meaning that only 30% are marrying) and then to 45% by age 34. Thus, ‘fertility’ for women is compressed into a 10 year window (30 – 40 years old)  rather than a 20 or 25 year window. At age 30 – 34 yrs barely half of eligible single women (54%) are married with 45% remaining single. The consequences of this are manifold. Those who marry have fewer children because the women’s ‘biological clock’ usually expires at 40, and the state’s abandonment of “the family wage” means that two incomes are continually necessary for a basic standard of living. Similarly, unmarried mothers have only the smallest of families, 1 child usually, with state benefits or part-time earnings insufficient to finance two children and a home and nursery care costs.

EU Marriage Rates

The following graphic (Graph 1) depicts the number of marriage trend lines since 1960 among a randomly selected group of EU countries; Austria, Belgium, Czech Rep, EU_SelectDenmark, France and Germany.

Graph 1. Composite trend lines

Each country has its own national pattern but the overall downward trend is clearly visible, if a little chaotic.

Germany in the 1960s had one of the highest rates of marriage at over 9 per 1,000 inhabitants. The Czech Republic had seen a see-sawing, up and down, from 1960 until 1990 when marriage rates finally collapsed.

Denmark too experienced something of a recovery in the decades from 1980 to 2000. Yet in this small sample, Belgium and France have seen nothing but decline. With the exception of Denmark in the selected group of countries for this comparison study, there is a distinct flattening-off the trend line in marriage rates after 2000. This might indicate what whatever was the factor(s) influencing lower marriage rates had by 2000 ceased or have lost their power.

No where is the ‘flattening’ effect more vividly on display than in the data from Portugal (see graph below) as it shifts from one normalised ‘plateau’ to another. Sweden on the other hand has neither surged nor collapsed since 1980 but muddled along at a steady hardly fluctuating pace.

To better illustrate the overall trend across the EU since 1960 each of the countries listed in the Table below (Table 2) will later be shown individually as a graph.


Previous Policy Failures

One might have expected that the fall off in marriage rates and numbers would have been made good by increased ‘cohabitation’ rates and divorces followed by re-marriage. But this is not the case as this graph for England & Wales trends_in_marriage2demonstrates.

Graph 2. England & Wales

This particular  ONS graph (left) begins in 1931 and extends to 2011.

A surge in marriage numbers can be seen in the early war years (1938  – 41), followed by a collapse and then a recovery but which then declines from 1945 to the early 1950s. The numbers recover in the late 1950s and continue recovering throughout the 1960s – peaking in 1971-72. From then onwards marriage numbers fall every year regardless of economic impetus. At 250,000 weddings per annum in 2011 the Canada_Marrtotal is barely over 50% of the 1971 level (and the lowest since the 19th century).

Graph 3. Canada

The Canadian experience of peaks and troughs closely mirrors that of Britain (see Graph 3 left).

The immediate post war period, namely 1945 – 1959, was one of economic recovery from a war ravaged Europe to one of sustained increases in living standards and disposable income.

American and Canada had escaped wartime  industrial devastation and social dislocation and upheavals of millions of DPs (stateless or ‘displaced persons’). Yet whereas the marriage rate stabilised in the UK it fell sharply in Canada in this period (1945 – 1959). In part this could be off-set or attributed to the influx to Canada after 1945 of immigrants from Europe but it would have to have been on a massive scale to provoke such a  decline. Discussion of these aspects will have to be reserved for another occasion together with the impact of the Married Man’s Allowance and Child Benefit – state benefits for children born.

However, they can be mentioned here in passing.  In Britain, which was literally bankrupt in 1945 (having sold off all its 1,000 tonnes of gold bars to finance the war against Nazi Germany), the new 1945 Labour Gov’t of Attlee nevertheless embarked on much-needed social subsidisation. [1] High among these were the Family Allowances, Child Benefit and Married Man’s (tax) Allowance.

Child Benefit, for instance, was only payable to mothers and was worth 5 shillings (5/-) in 1946 paid per week for each child in a family, except (but crucially) for the eldest. [2] The “Family Allowance Act, 1945” (enacted 1946) assumed at the time that the ‘family wage’ paradigm would persist and therefore it was reasonable to assume that:

  • “ . . . ..  the family income would be sufficient to stand the cost of one child without hardship”

By 1947 Child Benefit was being claimed by 2-3 million families with more than one child and the cost the government, i.e. the tax payer, was £59m – a huge amount by the standards of the day.

Fostering a stable society led Attlee’s government to launch the Married Couple’s Allowance which by the time of its cancellation by another Labour government, was worth £6,535 to each recipient in its last year (2008). The Married Man’s Allowance of 1945 was an adjustment made on the husband’s tax coding independent of whether he was a father or not. It was a personal tax allowance set at the man’s highest rate of tax. The allowance was scrapped in April 2000 – again under a Labour government.

Individual countries

Below are depicted the disaggregated trends in marriage rates peculiar and unique to each country listed above (see Graph 1, and Table 2).

Austria – Austria has seen a seemingly never-ending decline year-on-year. Only in 2009 did the decline halt and the rate level off.


Belgium – Belgium’s rate held fairly steady until 1990 when it fell markedly only to ‘plateau’ out from 2000 onwards.


Czech Rep – This is a most unusual graph with a saw-tooth profile from 1960 to 2000. By 2009 the rate in this country also appears to have plateaued.


Denmark – From 1960 to 1980 the marriage rate crashed and then just as unexpectedly is soared. for two decades in 2000 – before crashing once again.


France – A small rally can be sen in 1960 to 1970 but there after it is all downhill with 2009 marking yet a gain a change in fortunes to a steady state pattern.


Germany – Is it something to do with the degree of industrial advancement that some countries have a more persistent decline ? From one of the highest marriage rates at over 9 per 1,000 it falls to under 5 per 1,000 by 2009 and then stabilises.


Greece – Is marriage culturally stronger in Catholic and Orthodox countries ? Apparently, if ever once that was true it is not now. The decline persisted from 1960 to 2000, rose, then stabilised.


Holland – The Dutch have a laissez faire reputation in the social and domestic spheres, so one is not surprised by an overall decline in marriage. Yet between 1960 and 1980 there was a significant increase followed by an equally large decline in marriage rates, followed in turn by a levelling off.


Italy – This is perhaps the most surprising country’s profile given its is the sear t of Catholicism. It to has seen an unrelenting decline and there appears to be no redeeming levelling off in the latter years, e.g. 2009 to 2011.


Portugal – The characteristics of Portugal are of two plateaus divided by a collapse after 1990. There was no decline in marriage rates from 1960 to 1990 but in the 10 years from 1990 to 200 they fell about one third to 5 per 1,000. Thereafter, the second plateau shows no signs of a further decline.


Spain – Spain may be on the same peninsular as Portugal and share a common cultural heritage in many areas but the decline in marriage could not be more distinctly different. Spain has been on a continuously downwards track from 1960 to 2011.


Sweden – Perhaps marriage has not been so de rigueur in Sweden compared with the traditions of other EU countries. This might explain why the dramatic fall seen in other countries is not so apparent in Sweden (limited from 1960 to 1980). If any thing there has been a ‘steady state’ situation followed by a rallying after 2000.


Turkey – Accepting that Turkey is not in the EU, it nonetheless provides us with an interesting comparison. Yearly data is not available for Turkey but we can see – and would expect in a Muslim country – that marriage rates are steadier than in the more ‘decadent’ West where ‘moral relativism’ is rife. Lacking consecutive data, a polynomial trend line has been introduced (shown as a curving black line). This gives the anticipated levels of marriage where the data is missing.


U K – Technically this graph represents only England & Wales but as the overwhelming majority of the population lives in England & Wales, it provides an accurate guide. Graph 2 above illustrated the rises and falls from 1931 to the present day. However in the depiction, here, we see the turning point as 1971. From then on marriage rates have slumped with only a brief resurgence in 2010.


However, data supplied by ONS paints a quite different picture when All_marriagesinterrogated for “marriages per 1,000 population” as the Table below and Graph 4,  demonstrate.

Table 3 clearly shows that the normal way ONS collects data for marriage rates must be different from the method it adopts for “Eurostats.” Ranging between 10 and 8 per 1,000 it is significantly higher than the 6 and 4 (per 1,000) in the graph above.

UK_ons_marGraph 4. ONS based data (1998 – 2010 marriage rate)

It looks very much as if Eurostat’s methodology  results in an under-reporting of marriage rates in Britain and this could be echoed many times over in other EU countries.

Elephant in the room

One key aspect not tackled by the official sources of data is the influence of immigration on birth rates and marriage rates.  Common to all the data is the lack of any ‘ethnic dimension’. ‘Nation states’ are no longer that – they are now no longer homogeneous entities with a shared heritage. For Census purposes, many countries have 20% of their population which are non-European, e.g. Sweden with African and Muslim immigrants.

Gap_marrAnother wrinkle is the time interval used.

Graph 5.

Earlier, we saw at Graph 1 the Marriage Rate depicted for a selected number of countries but the decade sequence soon gave way after 2000 to yearly intervals, i.e. 2009, 2010 and 2011 (see Graph 5). When a blank, or interval, is inserted we see that the unrelentingly downward trend is not quite what it seems. From 2009 onwards the marriage rate levels off and, with the exception of Denmark which continues to fall, the overall trend could be said to have stabilised.

  E N D

Appendix 1

The following table is an extract from “Eurostats” data used in this article. If we look at, say, Iceland or Romania, we again see the same rapid decline from 1960 to 2000.

Bucking the trend is Latvia,  climbing from 11 per 1,000 to 16 in 1970, but then collapsing to just over 3 per 1,000 in 2000. Also  swimming against the tide and an ‘extreme’ case is Cyprus, which rose from 8 per 1,000 in 1970 to over 13 in 2000.

Malta can be seen to be an exception because, unlike the rest of Europe, it did not have divorce legislation until a few years ago.



[1] Gold plundered from conquered countries e.g. France, Belgium, Holland etc, was returned after 1945 but whether Britain had its’ gold returned or reimbursed is unclear.

[2] The mistake in later decades was to pay the Benefit to each and all children. The result was a decline from 2.4 children per family as the average, to the present day 1.7 figure.

2 responses to “Marriage Rates in Europe since 1960

  1. Nice work.While I agree that marriage as an institution has been declining, it’s difficult to get a clearer handle on this without availability of age-adjusted data. Propensity for marriage occurs at younger ages, even allowing for trend to marrying later.Spike in marriage rates in1960-70s reflects maturation of Boomers after which marriage propensity declined, especially as fertility dropped and population aged.

    • Thanks George for your reply which is, as always, thoughtful and much appreciated. I agree but did not comment in this article about maturity and the baby boom of the late 1940’s (ending in 1951). I understnad in the US ‘Baby Boomers’ is a term applied to many born after that date (does it end in 1961, I wonder ? ).
      However, what would be of interest would be the average age of marriage prior to the ‘Baby Boomers’ and prior to the impending war of 1939 – which distorted figures. I suspect it was not until 1980 that a deferrment of marraige was seen in the UK – but that is only a guess.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s