Harry Benson, December 2019 1 www.marriagefoundation.org.uk
The Average Length of Marriage
Harry Benson
December 2019
Using the same method used by Office for National Statistics (ONS) – but adjusting for the
significant number of UK couples who marry overseas – I estimate that the average (median)
length of a marriage beginning today is 40 years before ending in either divorce or death.
This contrasts sharply with the oft-quoted figure of 12 years, which is the average (median)
length of marriages that end in divorce.
Previous analysis by ONS estimated that the average marriage beginning in 2010 would last 30
years. So why the big improvement?
Part of the reason is an adjustment for overseas weddings. Since 1990, some 30-60,000 people
per year have married overseas. Assuming most of these are couples, a conservative
adjustment of half this number boosts the total number of ‘England & Wales weddings’ by 6%
to 17% per year. This produces a slightly lower, but more realistic, estimate of divorce rates.
The biggest reason is falling divorce rates. Back in 2010, ONS projected that 21% of couples
would divorce within 10 years, for example. Adjusting for overseas weddings would reduce that
by 2%. But with eight years data on these couples, actual divorce rates are 4% lower still. So the
final divorce rate figure for 2010 couples over their first decade of marriage is likely to be 15%.
With couples doing so much better early on, lifetime projections for divorce have also fallen as
a result. The original ONS projection for 2010 couples was 42%. Adding in overseas weddings
reduces this to 39%. My latest estimate for 2010 couples is now 34%.
So for couples marrying in 2010, the projected length of time at which half of all marriages have
ended through either divorce or death has thus increased from an original projection of 30
years to 35 years, or from 33 years to 38 years if I adjust for overseas weddings.
For couples marrying in 2017, the average length of marriage is now 38 years if I don’t take
overseas weddings into account and 40 years if I do, which I should.
Percentage of 2017 marriages
projected to end in divorce or death
0
5
10
15
20
25
30
35
40
45
50
0 10 20 30 40 50
Years remaining
%
Divorce
Death
Divorce rate projections
% ending
First 10 years
Lifetime
in divorce
Adjusted for overseas weddings
Without
With
Without
With
ONS original
21%
19%
42%
39%
MF estimate now
18%
15%
38%
34%
MF estimate now
15%
13%
35%
31%
Median length of marriage
Number of years
Adjusted for overseas weddings
Without
With
ONS original
30
33
MF estimate now
35
38
MF estimate now
38
40
Harry Benson, December 2019 2 www.marriagefoundation.org.uk
Falling divorce rates
Divorce rates are falling fast (Benson 2012, 2013,
2019a, 2019b).
I now estimate that projected lifetime divorce
rates have fallen from a peak of 44% for couples
marrying in the late 1980s to 31% for couples
marrying today.
Predicted Lifetime Divorce Rates
(by year of marriage)
25%
30%
35%
40%
45%
1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015
In my previous papers for Marriage Foundation, I
have shown that almost all of this fall in divorce is
due to fewer wives filing for divorce in the first
decade of their marriage. Remarkably enough,
divorce rates attributable to husbands has barely
changed since the 1970s
This highly gendered shift in divorce is not
adequately explained by changes in economics or
social patterns, which ought to cause husbands
and wives together to become both more or less
divorce prone together.
Divorce rate over 10 years
(by year of marriage & party granted)
0%
2%
4%
6%
8%
10%
12%
14%
16%
18%
1985 1990 1995 2000 2005 2010
Wife
Husband
My explanation for this gender shift is that
reducing social pressure to marry means those
men who do marry are more intentional about it.
This fits with commitment theory research
showing that men’s commitment tends to be
related more to decision making than women’s
commitment (Rhoades et al 2006, 2009).
In short, the proportion of newlywed men who
are less committed ‘sliders’ – rather than more
committed ‘deciders’ – has reduced. More
committed men means fewer unhappy wives
filing for divorce.
Methodology
I calculate divorce rates using the same ‘year of
marriage’ method as ONS.
One problem is that there have been large
numbers of overseas weddings since the late
1980s. Because these data are less reliable than
for domestic weddings, ONS do not include them
in their annual figures (ONS 2008, 2017).
Excluding them causes official divorce rates to be
overstated. In my model, I therefore make a
conservative adjustment to England & Wales
wedding numbers by adding 50% of ONS’s
estimated overseas wedding numbers. The effect
is that my figures for divorce rates are slightly
lower than those from ONS.
Calculating divorce rates, by year of marriage and duration of marriage
Year of
marriage
No. of
opposite
sex
marriages
Adjust for
+50%
overseas
weddings
Adjusted
number of
marriages
0-1 year
1-2 years
2-3 years
3-4 years
4-5 years
2012
263,640
23,826
287,466
1.9%
2013
240,854
31,579
272,433
1.6%
2014
247,372
23,154
270,526
1.1%
2015
239,020
23,048
262,068
0.6%
2016
242,774
22,500
265,274
0.0%
Divorces during 2017
44
1,534
3,091
4,446
5,462
A few years back, ONS published a detailed
projection of marriage survival rates for couples
who married in the year 2010 (ONS 2012).
In order to update their model, I have assumed
mortality rates remained much the same but
inserted my projections on cumulative divorce
rates for the 2010 and 2017 wedding cohorts.
References
Benson, H. (2012). Divorce rates have halved for new brides. Why?
Cambridge: Marriage Foundation.
Benson, H. (2013). What is the divorce rate? Cambridge: Marriage
Foundation.
Benson, H. (2019a) Men behaving well. Cambridge: Marriage
Foundation
Benson, H. (2019b) Lifetime divorce risk: Back to the 1960s.
Cambridge: Marriage Foundation
ONS (2008) Marriages abroad 2002-2007. Population Trends, 133,
65-70.
ONS (2012) Divorces in England and Wales: 2011
ONS (2017) British citizens married abroad
Rhoades, G., Stanley, S., & Markman, H. (2006). Preengagement
cohabitation and gender asymmetry in marital commitment.
Journal of Family Psychology, 20, 553.
Rhoades, G., Stanley, S., & Markman, H. (2009). The pre-
engagement cohabitation effect: A replication and extension of
previous findings. Journal of family psychology, 23, 107.
Harry Benson, December 2019 3 www.marriagefoundation.org.uk
Projection for 2017 marriages that end in divorce or death
Duration of
marriage in
exact years, x
(anniversary
number)
Probability of
divorce in
interval to
next
anniversary
Probability of
death of one
partner in
interval to
next
anniversary
Number of
marriages
surviving to
anniversary x
Number of
marriages
ended in
divorce in
interval to
next
anniversary
Number of
marriages
ended in
death in
interval to
next
anniversary
Cumulative %
of marriages
ended in
divorce
before next
anniversary
Cumulative %
of marriages
ended in
death before
next
anniversary
0
0.0001
0.0025
100,000
13
250
0.0
0.3
1
0.0046
0.0027
99,737
456
269
0.5
0.5
2
0.0096
0.0029
99,011
955
287
1.4
0.8
3
0.0135
0.0030
97,768
1,331
293
2.8
1.1
4
0.0167
0.0032
96,144
1,624
308
4.4
1.4
5
0.0199
0.0033
94,212
1,901
311
6.3
1.7
6
0.0203
0.0035
92,000
1,900
322
8.2
2.0
7
0.0182
0.0037
89,778
1,670
332
9.9
2.4
8
0.0181
0.0039
87,775
1,628
342
11.5
2.7
9
0.0178
0.0037
85,805
1,575
317
13.1
3.0
10
0.0173
0.0037
83,913
1,501
310
14.6
3.3
11
0.0171
0.0039
82,101
1,464
320
16.0
3.7
12
0.0172
0.0041
80,317
1,448
329
17.5
4.0
13
0.0151
0.0044
78,540
1,249
346
18.7
4.3
14
0.0137
0.0047
76,945
1,112
362
19.8
4.7
15
0.0131
0.0048
75,471
1,053
362
20.9
5.1
16
0.0130
0.0051
74,056
1,032
378
21.9
5.4
17
0.0114
0.0053
72,646
891
385
22.8
5.8
18
0.0107
0.0056
71,369
829
400
23.6
6.2
19
0.0101
0.0058
70,141
771
407
24.4
6.6
20
0.0093
0.0061
68,963
703
421
25.1
7.1
21
0.0089
0.0065
67,839
668
441
25.8
7.5
22
0.0082
0.0069
66,731
611
460
26.4
8.0
23
0.0077
0.0071
65,660
568
466
27.0
8.4
24
0.0069
0.0077
64,625
508
498
27.5
8.9
25
0.0066
0.0080
63,620
478
509
27.9
9.4
26
0.0060
0.0086
62,632
432
539
28.4
10.0
27
0.0053
0.0091
61,662
376
561
28.8
10.5
28
0.0046
0.0096
60,724
329
583
29.1
11.1
29
0.0042
0.0101
59,812
301
604
29.4
11.7
30
0.0037
0.0107
58,907
263
630
29.6
12.3
31
0.0032
0.0114
58,014
227
661
29.9
13.0
32
0.0029
0.0122
57,125
202
697
30.1
13.7
33
0.0026
0.0128
56,227
182
720
30.3
14.4
34
0.0022
0.0135
55,325
151
747
30.4
15.2
35
0.0022
0.0142
54,427
150
773
30.6
15.9
36
0.0019
0.0152
53,504
130
813
30.7
16.8
37
0.0017
0.0157
52,561
119
825
30.8
17.6
38
0.0014
0.0168
51,617
96
867
30.9
18.4
39
0.0011
0.0169
50,653
74
856
31.0
19.3
40
0.0011
0.0176
49,723
75
875
31.0
20.2
41
0.0010
0.0188
48,773
71
917
31.1
21.1
42
0.0007
0.0202
47,785
51
965
31.2
22.1
43
0.0006
0.0220
46,769
44
1,029
31.2
23.1
44
0.0005
0.0241
45,696
37
1,101
31.3
24.2
45
0.0005
0.0266
44,558
32
1,185
31.3
25.4
46
0.0004
0.0294
43,341
30
1,274
31.3
26.6
47
0.0003
0.0325
42,037
23
1,366
31.3
28.0
48
0.0003
0.0356
40,647
19
1,447
31.4
29.5
49
0.0002
0.0390
39,182
17
1,528
31.4
31.0
50
0.0002
0.0428
37,637
15
1,611
31.4
32.6