The preliminary Irish census results for April 2011 which were released during the week focussed attention on greater than expected overall population growth – the actual population is 100,000 up on estimates of population just 12 months before and the expectation was that growth would be flat as natural increases were expected to have been offset by emigration. But the focus on here is generally housing and the most interesting factoid to emerge was that Irish households have shrunk in size from 2.82 per household in 2006 to 2.68 in 2011. Now on the face of it, that is a tiny change but it has a major impact on housing. In fact between 2006 and 2011, we needed an extra 85,627 dwellings to accommodate the reality of smaller household sizes; yes, a tiny 0.14 change in household size equates to 85,627 dwellings in just five years, or just over 17,000 per annum. This entry examines why our households are shrinking.
The shrinkage in Irish household sizes is to be expected and furthermore is in line with realities in the rest of Europe. According to Eurostat, in 2009 the average EU household size was 2.4. This hides a large variation with Romania, Bulgaria, Cyprus and Malta having 2.9 and at the other end of the spectrum Germany, Denmark and Sweden with just 2.0. Eurostat reported that we had a 2.7 average size here in 2009. It seems though that the long term trend for all countries is average household size is reducing. In 1971 in the UK there were 2.9 people per household with one person households accounting for 18% of households. In 2009 that had fallen to 2.4 per household and one person households accounted for 29% of households. And as forIreland, here’s a table setting out our population, number of households, household size and % of households having a single occupant since the first census in the State in 1926.
The trend from the above is obvious, on average each year household sizes drop by 0.025 which equates to 15,552 new dwellings required based on our present population of 4.6m. But in the past 25 years it has been dropping by 0.04 which equates to 25,356 new dwellings required each year. In fact over the past five years we needed 121,072 dwellings to accommodate population growth (population growth 2006-2011 divided by household size in 2006) which equates to 24,214 new dwellings a year. So in modern times, we apparently need dwellings more for the shrinkage in household size than population growth. Told you this was interesting!
So the key question. Why are household sizes shrinking? The answer requires a sociologist’s input so please forgive this amateur’s views.
Lower birth rate?
Irelandis close to the top of the league of developed countries for producing babies on a birth rate basis with 15.5 births per 1,000 head of population per year. That compares with 12 and below for our EU neighbours. On a fertility rate basis, average number of babies per woman we are also close to the top of the developed world league with 1.96 babies per woman. That compares with 1.82 in theUK for example. So in terms of the developed world, we’re doing our bit to ensure there will be an adequate supply of Irish into the future, though our birth and fertility rates are significantly below those in the undeveloped world. However, from the point of view of shrinking household size, we’re not interested in our world rankings today but in changes to Irish birth rates. Oddly enough our birthrates have been holding up remarkably and although we are no longer at 22 per 1000 that we had between the 50s and 70s, we are up from the lows of 14s in the 1990s.
Now this conclusion is a little unsettling because, in younger days, I can certainly remember families with 14 children, which might be considered a bit of a freak show today because the norm seems to be for smaller sized families. And if we look at the size of households from 1926 onwards, it seems that households with 1-4 members has increased whilst 5+ has declined.
So my tuppence worth at this stage, which might be interesting to research, is that the overall birth rate has held up despite family sizes reducing because there are more families. There are more marriages and co-habiting couples as a proportion of total family units than previously. So fewer bachelors and spinsters, but that theory needs more research. What I can say here is that our marriage rate has stayed the same since the 1960s as revealed in these Eurostat numbers, and we still have 5-6 marriages per 1000 of population per year (5.2 in 2009, 5.5 in 1960) but I can’t find figures on co-habiting couples which might have been frowned upon in the 1960s but seems quite acceptable today, and figures from 2009 suggested that 33.3% of all births were outside wedlock which was most certainly frowned upon in decades past. In Greece just 7% of births are outside wedlock though in Iceland it is 65%, the EU average is 37%.
Marriage break-ups and divorces
It wasn’t until after the referendum in 1995 that divorce was finally recognised in Ireland, so it is perhaps not surprising that we still have the lowest divorce rate in the EU with just 0.8 divorces per 1,000 per annum, that’s about 3,500 in total each year. And presumably each of these gives rise to the need for a new dwelling. We’ve always had annulment and separation of course and it might be that these are now seeking divorces instead. The EU average divorce rate is 2.1 per 1,000 andBelgium has the highest rate of 3.0 per 1,000 per year.
Living longer
Not only does Irelandhave one of the highest birth rates in the developed world but we also have one of the lowest death rates with 7 per 1,000 compared to 9 in France, 10 in the UK and 11 in Germany. Paradoxically we have just average life expectancy rates in the EU at 77 for men and 82 for women (compared to the EU average of 75 and 82, the best Iceland 80 for men and France 85 for women and the worst Estonia 66 for men and Bulgaria 77 for women). The paradox resolves itself with understanding that Ireland has one of the youngest populations in Europe. We are definitely living longer in Ireland today than in previous generations, in 1960 the average life expectancy for a man was 68 and woman 72. I have not seen statistics on healthy life expectancy which is generally defined as life expectancy before a serious illness or condition impairs life to a serious extent but I would expect that this also has increased inIreland. And whereas in times past, the aging parents would live in the family home of the children these days it seems that elderly people are wealthy and healthy enough to live by themselves.
Co-habitation instead of marriage
In the good old days, folks lived with their parents or possibly by themselves before marriage. These days co-habitation before marriage is almost a pre-requisite. So instead of two dwellings belonging to two families, we might now need three dwellings for the two families and the co-habiting couple. I have not seen statistics on the number of households inIrelandcomprising co-habiting couples but if one third of all babies are born outside marriage, it is likely to be significant.
Marrying later in life
In 2006 the average age for a first marriage in Ireland was 32 for men and 30 for women. Outside Scandanavia, Irish women are the oldest in Europe to get married. Irish men get married at about the European average and certainly below Germany (33) Italy (32.8). How does that compare with times past? In 1996 the average age for men was 30.2 and women 28.4. And looking back as far as the 1960s, the average age of marriage today for women is considerably higher than the 24-27 range that applied previously.
The role of women
It’s been many years now since it was frowned upon for women to venture forth beyond the family home without directly entering the married home. Education and career opportunities are undoubted better than in times past. And with that independence from family and spouse comes housing needs.
The single life
It certainly seems to be a trend in western Europe that there are more singletons living alone. I’m not sure that has changed much in Ireland, but whereas in times past, the bachelor son or spinster daughter might remain in the parental home, these days it is not unusual for them to have their own place.
So there you have an amateur’s view on why we have smaller household size. There might well be other headings such as criminality and prison population and patterns associated with migration. The main point of this entry has been to emphasise household size is shrinking and to discuss possible reasons. And finally, what for the future, will our average household size decline further from 2.68? Difficult to say but you might expect divorces to increase and we are very much out of line with the rest of Europe. Our life expectancy should also increase in terms of better medical care and diet though obesity and less exercise might offset those gains. We will probably increase our standards of living so that we can afford to live independently when we age or between the parental home and co-habitation/marriage. The birth rate seems to be holding up remarkably well in Irelandthough it is down from 50 years ago, has it stabilised? So overall difficult to say what will happen in the future but the trend in the past 85 years has overwhelmingly been towards smaller household sizes.
[…] Second, the ever-excellent NAMAWineLake gets sociological on us in a fascinating post on growth of household sizes and why we need an extra …. […]
Most insightful article. But will this be enough to stop the Govt buying the biggest wrecking ball taxpayers money can buy to demolish the ghost estates?
It would be so Irish, would’nt, 10 years time when more houses are required, but they were all demolished when the country was broke?
It’s either a feast or a famine.
Years of healthy life expectancy available here: http://epp.eurostat.ec.europa.eu/portal/page/portal/health/public_health/data_public_health/main_tables
Shows Ireland in a favourable light.
@Brendan, thanks for the link. So Irish females have a healthy life expectancy of 65.2 at birth today considerably better than the EU average of 62, way better than Slovakia at 52 and not too far off the best Malta at 70.6 (who would have thought Portugal would be at 55.9 and Germany at 57.7)
As for men, we’re also pretty good here with 63.7 compared to an EU average of 60.9, far better than 52.1 in Slovakia again, but quite a ways off Sweden at 70.5. We’re way better than Germany at 56.7 and Portugal at 58.
So what’s so special about Sweden that keeps men healthy to over 70?
“So what’s so special about Sweden that keeps men healthy to over 70?”
ammm…..ammmm….. NO I’m not going to say it!
Interesting article, reminds me of a blog entry I read before the 2011 General Election by a North Kildare candidate http://ericdoylehiggins.wordpress.com/2011/01/31/lets-build-homes/
2Pack
Stall the digger, should you not add the temporarily absent numbers ( 30k ) onto the 1.462m occupied households in 2006 to get 1.492m ??
I do recall the headline figure in the preiminary data in 2006 was 1.49m occupied but not all on the night!!!
The 1.71m ‘occupied’ number for 2011 also includes temp absents I’ll wager but we don’t know how many of the 1.71m till next year.
I therefore put it to you again that household sizes in 2006 were more like 2.83 per and that the rate of shrinkage in household sizes has temporarily abated ….thanks to that recent sustained birth rate blip.
The long term average shrinkage 1991-2006 was around 0.04 persons per household per annum and it was around 0.03 persons per household per annum 2006-2011.
G’wan, ya know 2Pack has been onto this one for along time :)
@2Pack, I think we’re both on the same page with this – 0.3 or 0.4 drop per year from 2.8 or indeed 2.7 would still be a 1%+ drop in household size which means a 1% increase in number of dwellings needed. And if we have 1.7m occupied dwellings, that 17,000 per annum.
@2Pack, you’ve lost me. What I did for both 2006 and 2011, was take the total dwellings and subtract vacancies to get households. So in 2011, there were 2,004k dwellings in total and 294k were vacant and I deduced that there were therefore 1,710k households (2,004k minus 294k). Used the same approach in both Census years to be consistent.
The results were surprising to me but I guess if you have a 0.03 reduction in household size per year and say we’re at 2.7 now, then the 0.03 seems tiny but it means we simplistically need 0.03/2.7 or over 1% more dwellings each year, and if we’re now at 1,710k that’s 17,000 new dwellings just to accommodate that 0.03 reduction.
In addition we need homes for population growth and what I would think will be very limited obsolescence. And population growth is tricky. Our births and deaths are well documented and we are growing naturally 50-60k per year needing 20,000 new homes. I believed that we had net outward migration – truth be told, I still think that because of what I see every day – but the Census last week said we had substantial net inward migration in 2006-2011. That might still mean we have heavy outward migration today but I was surprised last week, and lost quite a lot of confidence in CSO/ESRI migration estimation
We are, it is just that Temporary abesnts are counted as occupied in the prelim results and the temp absents only shown in the detailed housing report ( due next year)
If you look at 2006 again we had 1.462m occupied and 275k empty in the prelim results
http://census.cso.ie/Census/TableViewer/tableView.aspx?ReportId=76536
see column C ( end)
When the detailed results came out we found that 29,946 extra units were pseudo occupied and were in fact NOT occupied but Temporarily Absent.
We really had 300k empties on the night, D+E+F and +C
Temp absent is a new category starting 2006 so in previous Censii temp absents were not detailed and were included in occupied.
The real eyeopener of course is the increase in single person households. That is irrefutable.
[…] preliminary Census results . Second, the ever-excellent NAMAWineLake gets sociological on us in a fascinating post on growth of household sizes and why we need an extra 17,000 houses a year. Third, Seamus Coffey thinks about what an extra […]
NWL – the per capita incomes in the countries with smaller households are larger than in those with larger households. Also, the declines in household size over time corresponds to increases in per capita incomes. Simply put, if more people can *afford* to live in smaller households, they will, as this means a more comfortable existence. Materialism is just as valid in sociology as it is in economics.
@l’ennui, you might be right. But not always. Saudi Arabia would probably have seen an increase in family size with wealth. So I think local culture comes into it as well.
Why on earth would Saudi Arabia have seen an increase in family size with weath? Am intrigued.
@l’ennui, more wives, children, servants (often overseas). Women still don’t generally work outside the home, or apparently drive without a male companion, so you’re unlikely to find singleton women, better medical care means that lower infant mortality. That’s off the top of my head.
According to the graph and other indicators on page 7 of this paper: http://research.ncl.ac.uk/forum/v6i1/al-otaibi.pdf, population divided by number of households was 5.14 in 2002 and was 4.8 in 2007. I would guess that this was a time of rising per capita GDP. I find it difficult to find figures relating to Saudi Arabia to be honest.
“@l’ennui, more wives, children, servants (often overseas). Women still don’t generally work outside the home, or apparently drive without a male companion, so you’re unlikely to find singleton women, better medical care means that lower infant mortality. That’s off the top of my head.”
Most of those things are sad, but they don’t mean that people are more likely to live in larger households. You’re (in my humble opinion) confusing lifestyle choices with economic imperatives. In any case, it’s probably a moot argument, because Saudi Arabian household size is probably reducing.
@l’ennui, you might be right. I can’t find any research that includes live-in servants in the household for example. Much of this area seems to be based on hunches which are then researched.
But what is clear is that in Ireland in the past 85 years there has been a relentless segmenting of households from nearly 5 in 1926 to just 2.68 today. Attaching causes to these changes is fun, and a lot of what is written above is conjecture, though I think informed conjecture.
Might we not see the household size to revert somewhat between Census 2011 and Census 2016, 1) as the singletons emigration gathers apace over that time frame 2) young adults remain in the home for longer for financial reasons 3) reduced birth rate over financial security concerns. Just a few idle thoughts.
@joseph, I’ll happily admit this is outside my field and the above simply reports the facts and gives an amateur’s view of why we’ve societally changed, at least as far as household segmentation is concerned. Certainly some causes should have plateaued, eg woman studying, working. Others will probably intensify, eg life expectancy and healthy life expectancy. I’m puzzled by our high birth rate particularly because it seems that children per family is reducing from the good old days with 12-15 siblings. Divorce rates will probably tend to increase and we are practically the lowest in Europe. But economic constraints may indeed lower birth rates. So overall, can’t say I’m qualfied to judge.
But I will make the observation that the reduction in household size has been relentless in the past 85 years and we are still a ways above EU averages.
“But I will make the observation that the reduction in household size has been relentless in the past 85 years and we are still a ways above EU averages.”
I very much doubt that the “the reduction in household size has been relentless in the past 85 years”. Certainly our household size is smaller now than it was 85 years ago but it’s been by no means a smooth or relentless decline.
NWL – you are saying on the one hand that you are not a sociologist and on the other you’re making sweeping statements that we need 17,000 new households a year and there’s been relentless drops in household. Which is it? If you’re not, as you say, qualified, why are you making these predictions?
I think your blog is excellent, but I think this post is ridiculous. I have a BA in Sociology, although I’m by no means an expert.
@l’ennui, I reproduce above the population, no of households both from the CSO and household size is calculated. I’m qualified to do that!
I examine the household sizes at each point where there are household sizes available and there is a fall from 4.77 in 1926 to 4.17 in 1961. Between 1961 and 1979 there is a 0.06 increase which I would characterise as minimal both in quantum and over 18 years out of 85. And then from 1981 there is a 0.2 drop per census period of roughly 5 years. All told I’m happy to characterise that as relentless or consistent.
And in response to your prickly question, there is more than enough competence here to assemble statistics and analyse them. And as for exploring reasons for the phenomenon, well I state clearly I have given an amateur’s view but the suggested reasons are evidenced and linked to sources. But I would welcome a sociologist’s views and additions or corrections.
I’m sorry you think the “post is ridiculous”. Perhaps you could amplify so as to improve standards. Is it the statistics you have a difficulty with? The suggested causes? Are there causes missing, do you think? Is there something of value that might be contributed with respect to future projections?
As to the main point of the post. It points out that between 2006-2011, some 17,000 homes per year were needed for segmenting households. That trend is certainly consistent since 1981 and is consistent with the overall trend since 1926. So it may reverse next year but why? It might enhance all our pools of knowledge to get an academic sociologist’s views.