How many mortgage-holders are faced with unemployment?

A couple of weeks ago, I discussed the likely extent of the problem of negative equity and homes worth less than when they were bought. This had led to a rich vein of suggested blog posts as extensions, including last week’s look at which counties have suffered most from “unexpected” unemployment since the start of the recession.

That post was a first step towards estimating how many households are faced with both unemployment and negative equity. Today’s post is an intermediate step: how many mortgage-holders are unemployed? How does this vary across counties? And what would it look like if the Live Register were to hit 500,000, as some have suggested it might?

There are about 1.7 million households in the country – almost 600,000 were mortgage-holders in the Census of 2006 and they have been joined by another 90,000 or so first-time buyers since then. (Landlords and buy-to-let investors are of course another issue, but I’ll leave them out for the moment.) At the same time, since the Census, the number on the Live Register has increased from 155,000 to 385,000, meaning there are about 230,000 “unexpected unemployed” around the country, many of whom would have bought property at some point over the last decade or two assuming a stable employment situation.

Working out how many of those two groups intersect is not a precise science. Given the broad nature of the economic downturn in Ireland, I have assumed that unemployment has been indiscriminate across working households, i.e. of the 230,000 new unemployed, 55% are in homes with a mortgage, the same ratio in the broader labour force. The map below gives the approximate percentage of households with a mortgage where one person has become unemployed since the recession started. The national average is about 7% of households with a mortgage (or one in fifteen) are currently faced with unemployment.

Unemployment among mortgage-holders in Ireland by county

Unemployment among mortgage-holders in Ireland by county

It might be useful to walk through one county to explain in more detail. In Louth, where there are 44,000 households, about 14,000 of them are more than likely households with retired (and mortgage-free) inhabitants. Of the remaining 30,000 households, just under 20,000 are owner-occupiers with mortgages. At the same time, almost 9,000 people have been added to the number of unemployed people in Louth in the last two years. Assuming that the spread of unemployment was not related to home ownership status, that would mean that 60% of the new unemployed – or just over 5,000 people – are mortgage-holders. If those figures are at least in the right ballpark, that means that one in eight households with a mortgage in Louth is dealing with unemployment.

If you go to the original Manyeyes visualization, you can also look at the 2010 scenario of 500,000 on the Live Register, which assumes that the future increase in unemployment is distributed the same way the increase in the last 24 months has been. Because of that assumption, the regional dynamics don’t change – Leinster is still clearly worst affected – but the national headline naturally worsens. In that scenario, 10% of mortgage-holders would be faced with the problem of unemployment.

The final piece of the puzzle – next week’s post – is estimating how many of those who are unexpectedly unemployed and who have a mortgage are faced with the loan on their property being greater than that property’s current value.

Where in Ireland has seen the biggest increase in unemployment?

My recent attempt to put some figures on the scale of negative equity in Ireland – which concluded that about 40% of Irish homes are worth less than when they were bought and that as many as 20% of homes may be in negative equity – sparked some discussion here, on thepropertypin and most thoroughly on

The original post was designed just to put some numbers on the potential problem of negative equity, leaving aside for the time being the implications. Two important strands of discussion have arisen about the implications. The first relates to financial consequences, as mentioned by Karl Whelan, particularly in relation to the proposed NAMA and the fate of the banks. The second broad strand of discussion, being led by Liam Delaney, relates to how negative equity has labour market implications, particular when unemployment is on the rise. (Unemployment and negative equity are mirror images of the home ownership/labour mobility discussion being led in the US by Richard Florida.)

I’m currently working on estimates of how many households are affected by the dual problem of unemployment and negative equity. Combined with the likelihood of falling rents over the coming two/three years, rents being the alternative income a homeowner could get from their house, this is a cocktail for widespread misery currently partially staved off by all-time low interest rates and therefore mortgage repayments.

A next step in working out where both negative equity and unemployment will strike is looking in more detail at the problem of unemployment. The CSO provides very detailed statistics on unemployment by county/town and more occasional detail on the age profile and duration of unemployment. The map below gives an idea of ‘unexpected’ unemployment (original visualization here). It show the increase in those signing on by county in April 2009, compared to the average of 2005 and 2006, meant to indicate a natural level of unemployment (whether long-term or just switching jobs).

Unemployment in Ireland by county, April 2009 compared to 2005/2006

Unemployment in Ireland by county, April 2009 compared to 2005/2006

Those looking with relief at counties in a light brown – such as Waterford, Louth, Donegal and Mayo – should be aware that in all counties, the April 2009 was at least twice the 2005/2006 average. What’s more worrying, though, is that there are a number of counties where unemployment is three times what it was three years ago. In Meath and Kildare -stalwarts of Dublin’s commuter belt – unemployment has more than trebled. Likewise in Cavan and Laois.

The next part of the puzzle is to revisit county-level estimates of negative equity based on comments on the last set of figures and then try to put some numbers on how many households finds themselves faced with both unemployment and with a house worth less than their debt to the bank.

Westmeath says ‘Watch out below’! An updated heat-map of Ireland’s property market

A little behind schedule, given that the report is out a couple of weeks at this stage, but the latest Manyeyes visualization of Ireland’s property market is up here. The overview snap is below.

Heat map of price changes in Ireland's property markets

Heat map of price changes in Ireland's property markets

As you can see, all counties have notched up two consecutive quarters of price falls by this stage (Limerick was last to fall). Some counties are now on six quarters. It seems that those that fell first have fallen hardest – in the Midlands (defined loosely enough), Laois and Longford were among the first counties to register falls in asking prices. They have now been joined by neighbouring counties, which are among the worst affected so far by falling prices. Take Donegal, for example, which was among the last to give up rising prices, where they are now 17% lower than a year ago. In Westmeath, the figure is even higher (18.1%), which marks a huge slide of more than 10% in the year-on-year change from the previous quarter. Longford and Louth are also in the same range close to 17%.

Now, as for Tipperary and Waterford (and Limerick and Mayo, the other two counties where falls are still single digits)… Are sellers there living in a mild form of cloud cuckoo land? Even looking at fall-from-peak figures, rather than year-on-year, they’re still in single digit territory. Or perhaps they think that they’re more sheltered, because the overhang of property is not as severe as it is in the Midlands/North-West? Answers on a postcard…

(PS. Do people think that this heatmap should change from year-on-year changes to one masuring the fall from the peak instead? That might give a better idea of total adjustment. Biggest adjustment so far is still Westmeath, down 20.0% exactly.)