Following EU accession of the “EU8” countries in 2004, we have witnessed large scale immigration from Poland to the United Kingdom (UK). Immigration levels peaked in 2007, but the size of the Polish population remains large. In 2012, there were approximately 646,000 Polish individuals living in the UK, Poland was the second most common non-UK country of birth, and Polish the most common non-UK nationality.
For UK population change, the childbearing of Polish migrants has been a particularly interesting feature. Data collected at birth registration shows that Poland first entered the top ten countries of birth of non-UK born women having live births in the UK in 2005, positioned in ninth place. By 2010, Poland had moved to the first place where it has remained through to 2012 (most recent data available). Also consistently found in high positions across this period are Bangladesh, India and Pakistan – population sub-groups who typically have higher fertility levels in the UK than does the UK-born population.
In case of the Polish group, originating from a low fertility scenario, this raises interesting questions. For example, does their fertility remain low, with the large number of births attributable to the size of the Polish population? Alternatively, is the fertility of Polish individuals higher in the UK relative to Poland? Recent studies and newly released census estimates provide some insights into this topic, each with strengths and limitations.
Recently published UK 2011 Census estimates give the Polish Total Fertility Rate (TFR) as 2.13 – higher than both the TFR of UK-born women (1.84) and the TFR in Poland (1.30), but lower than the average for all non-UK born women in the UK (2.21) (Dorman, 2014). The census provides robust estimation of fertility for population sub-groups for a snapshot in time, but since it is collected only decennially it cannot inform us about inter-censal patterns and short-term trends over time.
To establish patterns for the inter-censal period, we need to use other data sources. Janta (2013) documents the increase in births to Polish women over the 2004 to 2011 period as recorded at birth registration, showing the composition of these births by whether or not both parents were Polish. Over this time period there were also substantial changes in the size of the Polish population and birth registration data alone do not allow for the calculation of fertility rates since a denominator (i.e. estimate of the ‘population at risk’) is unavailable.
Zumpe et al. (2012) combine birth registration data with Annual Population Survey data to overcome this lack of a denominator. Producing General Fertility Rates (GFRs) for Polish women, they find an increase from 93 births in 2007 to 106 in 2010, before a decrease to 93 again in 2011. Since GFRs do not standardise for the age structure of the population, Zumpe et al. suggest that some of this fluctuation may be attributable to the changing age composition of the Polish migrant population in the UK over this time. Nonetheless, the GFRs found are higher than those for UK-born women in the same time periods, suggesting potentially higher fertility amongst the Polish group.
Another study by myself and colleagues (Waller et al., 2014) uses UK Labour Force Survey (LFS) data to indirectly estimate age-standardised fertility. We use the ‘Own Child Method’ (Cho et al., 1986), which reverse-survives women and their children (assuming that they are living together) to calculate births in each year. A limitation of this approach is that the available sample sizes require pooling of several years of data for reliable estimation. With these data, we produce an average TFR of 1.4 for Polish women across the 2004-2012 time period, which is much lower than that found in the 2011 census and more comparable to that seen in Poland.
Comparing our results with the 2011 census estimates suggests a dramatic underlying change in the Polish fertility patterns since 2004. Indeed, if we disaggregate our TFR into two-year periods (omitting 2012), we can see these changes emerging, as shown in the figure below.
Source: Based on analysis from Waller et al. (2014), Own Child Method estimation with UK Labour Force Survey data, 2004-2012
We gain some further insight into these changes by using information on the year of arrival to the UK collected in the LFS to explore the association between migration and birth timing. Our paper shows that many of the Polish women arriving to the UK at young childbearing ages are observed as childless at arrival. Relatively small proportions go on to have first births within three years of arrival, increasing to approximately one third of those arriving at peak childbearing ages (during their late twenties). We also find a later childbearing profile amongst Polish migrants in comparison to women in Poland, which could be associated with some postponement of childbearing amongst migrants.
Our results are supported by Lübke’s (2012) study combining LFS and European Social Survey data from 2008 to 2011 to study migration and fertility timing amongst Polish migrants to the UK, which finds depression of childbearing prior to migration. Some elevation of fertility is then shown for the first two years after migration (with comparable proportions to our paper), but the largest odds of experiencing a birth are found with slightly longer durations of residence at four or more years after arrival.
Thus, it could be that the low TFRs, shown above, reflect some postponement of fertility in the period prior to migration in the earliest period. Significant increases across the period may occur due to a sub-group having births soon after arrival, and as a combination of those who delayed their childbearing for longer periods after arrival now also having births as the period since EU accession, and the duration of stay, increase. Whether some of the Polish women migrating at childbearing ages in more recent years will follow this pattern of delay, we are not able to determine.
In summary, the UK data available for studying this topic are limited but taken together these different studies further advance our understanding. Our findings suggest that the period fertility of female Polish migrants to the UK was quite low in the period immediately after EU accession and has gradually increased to considerably higher levels over the past decade. This could be largely due to important changes in fertility timing, which are reflected in TFR measures of fertility, due to the complex relationship between fertility, migration and age at migration. However, it is too early to tell whether Polish women in the UK will have higher completed fertility than women in Poland.
Cho, L., Retherford, R. D., & Choe, M. K. (1986). The own-children method of fertility estimation. Honolulu: East-West Center, University of Hawaii.
Dormon, O. (2014). Childbearing of UK and non-UK born women living in the UK – 2011 census data. Available at: http://www.ons.gov.uk/ons/rel/fertility-analysis/childbearing-of-uk-and-non-uk-born-women-living-in-the-uk/2011-census-data/article–childbearing-of-uk-and-non-uk-born-women-in-england-and-wales-using-2011-census-data.html
Janta, B. (2013). Polish migrants’ reproductive behaviour in the United Kingdom. Studia migracyjne – Przegląd Polonijny. 3, 63-96.
Lübke, C. (2012). How migration affects the timing of childbearing: Examining Polish women in Britain. Poster presentation at the Pairfam conference ‘Fertility over the life-course’, September 12th-13th 2012, Bremen. Available at: http://www.pairfam.de/fileadmin/user_upload/redakteur/Tagung/Programme/ppt_pdf/Bremen2012_L%C3%BCbke.pdf
Waller, L., Berrington, A. and Raymer, J. (2014). New insights into the fertility patterns of recent Polish migrants in the United Kingdom. Journal of Population Research. DOI: 10.1007/s12546-014-9125-5
Zumpe, J., Dormon, O. and Jefferies, J. (2012). Childbearing among UK born and non-UK born women living in the UK. Available at: http://www.ons.gov.uk/ons/dcp171766_283876.pdf
 The EU8 group includes: Czech Republic, Estonia, Hungary, Latvia, Lithuania, Slovakia, Slovenia and Poland.
 The TFR is a synthetic measure reflecting period fertility levels, and represents the average number of children a woman would have in her lifetime if she was subject to the age-specific fertility rates for the period in question. As such, the TFR is sensitive to changes in the timing of fertility, which is particularly important here where there may be complex interrelations between fertility timing, migration timing and age at migration. The TFR reflects the intensity of childbearing in the period of interest.
 GFRs represent the number of children born per 1,000 women of childbearing age in a population.
 This may be a more questionable assumption in the case of migrant mothers, as discussed in our paper.