Principal assumptions

Assumptions about future divorce rates were based on trends in the relevant time-series during the period 1993-2003. These rates use the revised population denominators by marital status published by the Office for National Statistics on 7 October 2004 which take full account of the results of the 2001 Census and the subsequent revisions to population estimates.

The approach for the projections was to extrapolate the time-series from the last ten years forward but with trends gradually assumed to diminish so that rates are held constant from 2013. In addition, the results from the projection were controlled to agree (approximately) with the known provisional total number of divorces in the first year of the projection (2003-04).

There has been a gradual upward trend in divorce rates at most ages which has generally accelerated since the year 2000. However, the provisional divorce data for 2003-04 suggests that the more rapid increases of the last two or three years may not continue and the longer-term projections are consistent with a continuation of the slower increases seen in earlier years.

Click here for graphs showing actual and assumed divorce rates.

Variant assumptions

The high and low divorce variants assume that divorce rates will gradually diverge from those assumed in the principal projection, so that from 2018 onwards rates, at all ages, differ by ± 10 per cent from those assumed in the principal projection.