Official Report (Hansard)

PDF version of this report (236.35 kb)

Committee for Employment and Learning

NISRA Quarterly Labour Market Statistics: DEL/DFP Briefing

The Chairperson: I welcome Dr James Gillan, the head of economic and labour market statistics in the Northern Ireland Statistics and Research Agency (NISRA), and Mr Victor Dukelow, the deputy director of analytical services in the Department.  Gentlemen, you are very welcome.  Over to you.

Dr James Gillan (Department of Finance and Personnel): Good morning.  Thank you very much, Chairman.  I am going to start this morning by giving you a summary of the most recent labour market statistics that were published on 15 May.  I will go across the main outputs from the labour force, claimant count, quarterly employee jobs and a new output index.  I will then go into some of the key groups that might be of interest to the Committee.

Our first slide summarises the latest labour force survey (LFS) estimates.  As you know, this is the household survey that is conducted in accordance with International Labour Organization (ILO) statistics.  It shows an increase for Northern Ireland over the quarter and over the year to January to March 2013.  There was not a marked increase — it was 0·3 percentage points over the quarter — but there was an increase over the year.  That is in contrast to what happened in the UK, which, while up over the quarter, was down over the year.  An estimated 70,000 persons are unemployed in accordance with the LFS definitions.

Quarter on quarter or year on year does not tell you that much; you really need to look at the trends.  The next slide details the unemployment rate and shows two lines.  The blue line is Northern Ireland, and the red line is the UK.  From the low point of 3·3% that was reached in May to July 2007, we have now increased to 8·1%.  There was a steady climb and peak where the Northern Ireland line crosses that of the UK, at about 8% at the end of 2010.  In 2011, we saw a bit of a decline in the unemployment rate, but in 2012, it started to climb again, and it is back up to 8·1% in the most recent period.  For much of the period since the start of the downturn, we were lower than the UK, but we are now just edging higher.  These are non-overlapping series, but one thing to note is that, in the last three months of the published figures, there has been a fall in the rate of increase from about 8·5% to 8·4% to 8·1% on the LFS.

The next slide gives another source by which we measure unemployment, which is jobseeker's allowance and benefit claimants.  This is a census, and it has come up with a figure of about 64,300 people who are in receipt of jobseeker's allowance.  Interestingly, this is the first time since August 2007 that we have had a decrease in the claimant count for three consecutive months.  It is not much of a decrease.  It is a fall of 600.  We have had one-off months before in which there have been decreases, and we still have the second highest claimant count unemployment rate across the UK regions.  However, it is worth noting that there is at least a flattening out of the unemployment rate on the claimant count.

On the next slide, you will see that, contrary to the other picture of unemployment in the LFS, our unemployment rate on jobseeker's allowance claimants relative to the UK has shown an increase over the period since the low point was reached.  It was a very rapid increase to start with and then a steady, slow climb in claimant count unemployment; whereas, the UK, after a rapid increase, stayed relatively flat.  We will talk a little bit later about why the two measures might be giving us a different picture.  So, even though there has been a slight fall in the unemployment rate, in that longer time series it is not particularly marked.

The next slide looks at the surveys of businesses.  The information was published in March, but it helps to triangulate what is going on.  Overall, we saw a decrease in jobs, from September to December 2012, of about 750 or minus 0·1%.  Again, that is not particularly marked.  Over the year, there was an increase in jobs.  When you break that down into the public and private sectors, you can see that the increase in jobs occurred in the private sector.  It is only by 0·5%, but it ties in with the slowdown in some of the unemployment.  We are seeing private sector jobs begin to increase.  I think that we have had an increase in private sector jobs in three of the last four quarters.  Some of those have not been very marked increases; nevertheless, there have been increases.

The next slide shows graphically how we compare with the UK.  You will see that we have fallen considerably from our peak — down 5%, or a total of 39,680 jobs, since quarter 2 of 2008 — whereas the UK is only down 2%.  So, again, we are relatively worse off in the fall from peak.  However, from another point of view, you could argue that the peak was an unsustainable level of growth in jobs.  You can see that we have fallen back to the levels that were recorded in 2004 and 2005.

The next slide breaks jobs down by public and private sector.  As we summarised earlier, you can see a considerable fall from peak in both sectors.  The private sector shows some signs of recovery, although not terribly marked, in the last three quarters out of four.  That is our employee jobs measures.  Since we were last at the Committee, we have another measure, which is the Northern Ireland composite economic index.  That is shown on the next slide.

There is always a bit of difficulty in that we have piecemeal measures of what is going on, so we have tried to combine some of the output measures for the index of production, the index of construction, the index of services and agricultural output, which are basically sales by businesses.  We have weighted those together with a measure of public sector employee jobs on the basis of their share of GVA.  This is an attempt to give us a frequent and timely measure of how the whole economy is doing in Northern Ireland.  The slide is a bit wordy, but I will draw your attention to the second bullet point, which states that three rises in the last four quarters has contributed to an increase of 0·3% over the year.  Again, that is not very marked, but it is in line with the quarterly employee jobs measure.  In the last two bullet points, it states that the fall from the maximum is still of the order of 11% — so the increases have not brought us back anywhere near where you would like to be — and we are back to levels previously recorded in quarter 1 of 2004.

The next slide illustrates that graphically.  The red line is the Northern Ireland composite index.  You can see the fall from the peak of quarter 2 of 2007 up to quarter 4 of 2012.  There has been some increase, but it has been pretty flat over the last year and beyond.  The slide is useful if you compare it against what has been happening in the UK, which is the blue line.  You can see that there has been a gradual increase from the UK's trough, but, even in the UK, the growth beyond that, in quarter 1 of 2013, is about 0·3%.  The consensus is that it has been very flat over the last 18 months in the UK.  We are back to levels of quarter 3 of 2011.  So, there is not a lot of growth happening there.  The green line is quarterly GDP for the Republic of Ireland, which you can see is much more volatile, but it is showing evidence of growth in the last couple of years.  So, our trajectory is out of kilter with what is going on in the UK.  I should point out that this is not a quarterly measure of GDP.  We do not have all the expenditure sources and income sources.  This is just a proxy measure for GDP, but it is the closest that we can get.

The next slide summarises some of those across all the measures.  Even though some of them might differ, you are seeing unemployment on the LFS up 40,000 since April to June 2007; the claimant count is up 40,800 since September 2007; and employee jobs is down nearly 40,000 since quarter 2 of 2008.  Both the quarterly employment survey (QES) and the composite are showing changes in three out of the last four quarters, so there is broad agreement between the different measures of considerable decline from peak activity and limited evidence of growth.

There are a lot of numbers there.  I will take a break in case anybody wants to raise anything before we go on.  The next part of the presentation is to talk a bit more about the differences between the two unemployment measures, the inactive, the long-term unemployed and the youth unemployed, but I do not want to go hurtling on if members have questions.

Ms McGahan: My question is on the back of the point that you have just made.  I am finding all of this very confusing.  How different are the labour market statistics from the figures for those who are economically inactive?  Would it be fair to say that the figures for those who are economically inactive are more accurate, given the fact that anyone who carries out at least one hour's work is considered as being employed under the LFS figures?  I am finding it all very confusing.  Maybe you will address that in the next session.

Dr Gillan: Yes, we have a couple of slides that look at the differences between the two measures.  The labour force survey measure definition of unemployment, employment or inactivity is the lead measure because it is the International Labour Organization’s definition of what constitutes unemployment.  If we want to be able to compare ourselves with the Republic, the UK, Spain or Germany, this is the best way to do it because they are measured on exactly the same basis.  However, I take your point:  it is a fairly generous measure of what constitutes employment.  You have to have done one hour's work in the reference week, and then you are considered as in employment.  That is why things like the employee jobs measure, which measures people who are on the PAYE system, is a better measure. 

There is that wider pool of those persons who are inactive and who have been available for work but did not actively seek work, and that has been in and around 50,000.  Those people are constituted as economically inactive, but that is because they were not available to work and were not actively seeking work in the reference period.  They are defined out of the equation, if you like, and they could probably be constituted as discouraged workers.  They could say that they have been searching for work for a year and a half and they see that the long-term unemployed figure is up.  They may feel that there is limited benefit in actively seeking work and that there is nothing out there.  We publish that figure; we say that it is out there, and we track how it changes.  So, if you want to look at the slightly broader definition, it is there, but you would not be comparing like with like.

Mr Victor Dukelow (Department for Employment and Learning): If you do not mind, I will add to that point, and it is around the issue that you raise on underemployment — people who have a job but would maybe like to work more.  In the downturn that we are working our way through at the minute, there is that sense of slack in the labour market, and that is coming through in the figure work that James publishes.  Certainly, one of the measures that you can find through the LFS is the proportion of people who are working part-time and would like to work full-time.  About a quarter of those who are part-time are reporting here that they would like a full-time job if one were available.  There are different ways of looking at these figures and making sure that we can uncover as much information as possible about the realities in the labour market currently.

The Chairperson: I have a point about economically inactive people.  One of the phrases confused me, and I would like you to explain it.  There are 54,000 people who are economically inactive and who actually want work but are neither actively seeking work nor available for employment.  How do you measure somebody who wants work but does not go looking for it?

Dr Gillan: The first contact that the labour force survey has with a household is a face-to-face interview.  There is a structured questionnaire, and the person is asked questions about what they were doing in the previous week.  Were you employed for at least one hour or more — yes or no?  Were you available to start work — yes or no?  Were you actively seeking work — it gives a list of things that might constitute actively seeking work — in the previous four weeks?  Are you available to start work in the next two weeks?  If you were not actively seeking work or available to start, you are classified as economically inactive, because you are not an active member of the labour market.  I do not have the exact question but it is along the lines of:  "If you are economically inactive, would you like to work?"  We can break that down into those who would like to work and were available but were not actively seeking work.  It is a face-to-face interview in the first instance and via phone calls in subsequent quarters.

Mr F McCann: It is amazing how Ministers can derive so much joy from quite depressing information and statistics.  There was an increase in the number unemployed according to the labour force survey but a decrease in those on jobseeker's allowance according to official statistics.  Those do not match.  Was the decrease of 600 jobs actual jobs or government training schemes?  Were those people who worked one hour a week?  That is difficult to gauge because the statistics give a false impression that we seem to be on the up and up.

Dr Gillan: There are two different measures because they are measuring two different things.  You cannot look at either in isolation.  Are they saying different things?  The jobseeker's allowance survey is a census of everybody, so it is a sensitive measure.  What we published in May refers to April.  So, you are right:  it is a very temporary measure.

The labour force household survey of about 2,500 is an estimate of what is happening.  There is a statistical margin of error around that estimate of, say, plus or minus 1·8% on the unemployment margin, so the LFS will not pick up a fall of 600 people on the claimant count.  It is not sensitive enough.  That is well within the statistical random error on the LFS survey.  Rather than focusing on one quarter, quarter on quarter, over the year on the LFS, you really need to see it in the longer term.

We put out both sets of statistics so that people could make an informed decision.  The difficulty with jobseeker's allowance, which the introduction of universal credit may highlight, is that definitions will change.  People could flow off jobseeker's allowance, and we will say that that is great, but they may be going onto other benefits.  However, the labour force survey will give a consistent measure over time.  You get consistency, but a margin of error is associated with the estimates.

Mr F McCann: The survey refers to a fall of about 600 in the claimant count.  That gives the impression that 600 people left benefits and may have gone into employment.  The fact is that they may have become ill and gone onto employment support or onto other benefits.  That does not really answer the important question:  where did they go?  If there are jobs, the survey does not make it clear where they were created or in which areas.  It does not give a picture of what is happening.  I appreciate that all you have to work with are the statistics that you are given.  For us, however, it is a question of trying to work out what is happening in the broader labour market, and those statistics do not answer that question.

Dr Gillan: Our monthly publication has some information about where people go when they come off jobseeker's allowance.  For example, 7,600 came off the jobseeker's allowance count in April, and fewer went onto it, which gives the 600 decrease.  Those are unadjusted figures and not seasonally adjusted.  That shows that 39% found employment.  We give a comparison, so there is a breakdown of the destination of people who came off jobseeker's allowance.  That compares with 37·6% who left in April 2012.  So, you are able to see just how well people are faring and whether they are going into employment, benefits, inactivity or just ceasing to claim.  We have some slides on that for youth unemployment as well.

Mr F McCann: So that identifies it.  The last time that you were here, we spoke about the location of jobs, and we heard that jobs were available in the north-west.  I do not know whether it was Alastair or Thomas who asked what industry those jobs were in and whether they were in the service industry or elsewhere.  That was to try to get us an understanding of the calibre or type of jobs that are available.  It is interesting that, if you declare that you are working for one hour, you are no longer classed as unemployed.

Dr Gillan: That is the labour force survey definition; it is not part of the claimant count.  We will see the overlap between the two in a later slide.

In our publication, there is a graph for each month that shows the percentage that found work, that went into education and training, that failed to sign, and all others.  That is not in your packs; we publish that information monthly.  I do not think that we get down to the occupation that someone goes into.

The Chairperson: Fra, that is in your pack from page 59 onwards.  The monthly survey is already there.

Mr Douglas: James, can you give us an idea of whether the increases in private sector jobs were full-time or part-time?  Can you also give us a breakdown of what sectors those jobs were in, as Fra was asking?  I want to find out where those jobs came from.

Dr Gillan: The growth was in the service sector, which was up by 3,000 jobs or 0·5 %, and in some of what we call "other" industries.  Some decreases happened at the same time.  Construction was the worst affected and was down by 5·5% or 1,700 jobs over the year.  Manufacturing fell by about 1,100 jobs over the year.  I do not have a breakdown of the part-time and full-time jobs.

Mr Douglas: I do not know whether you will be able to answer this, but I read recently that the Republic of Ireland created some 37,500 jobs last year, mainly because of its lower rate of corporation tax.  Can you compare that with the Northern Ireland statistics?  How are we faring?

Dr Gillan: We do not measure jobs created.  We measure the net change in jobs.  We go to a company, and it will tell us that it had 1,000 jobs in the last quarter and 1,200 in this quarter.  We do not know whether that was because it created 500 jobs and shed 300 jobs in another part.  We get only the balancing figure or the net change.

Mr Douglas: We should probably ask Invest Northern Ireland about that.

Dr Gillan: Yes.

The Chairperson: You talked about — [Inaudible.] — against the UK GDP and ROI.  How far away are we from getting a Northern Ireland GDP figure or calculation?

Dr Gillan: It is quite a difficult task to put that together, and the statistical framework that has to be put together is quite intensive.  We are planning a feasibility study this year that will identify the data sources that would help to populate a system of national accounts, where our data gaps are and what resources would be needed to try to fill those gaps.

I should say that we have taken this measure and the equivalent UK sources and created what would be the UK equivalent of our measure.  That tracks UK GDP quite well.

This is just sales, and it does not take into account the costs of doing business.  GVA is the difference, and it takes profits and wages into account.  We are using sales as a proxy for it.  Against that, we could not go to businesses and ask them what their costs were for the quarter.  We would grind businesses to a halt if we started to ask them for that.  It is quite a long-term job to try to put that together.  You would need national accounts to get a true measure of quarterly GDP.

The next slide shows the areas that I was going to look at and picks up on some of Bronwyn's points.  It shows the differences between the labour force survey and the claimant count.  The red circle is the labour force survey measure of unemployment, which is 21,000 and 48,000 and comes out at 69,000.  I said that the LFS measure of the unemployed was 70,000, but this is an unadjusted figure and rounding comes into play.  The blue circle is the number of persons who are in the claimant count or jobseeker's allowance measure of unemployment, and it comes out at 64,300.

Where the red and blue circles overlap — this is estimated information that is taken from the labour force survey, so it is not an exact match — you can see that over half of the people are counted as unemployed under both measures, so there is a fair degree of overlap.  The left-hand side of the red circle are those people who are unemployed under the labour force survey definition but who are not in receipt of jobseeker's allowance benefit.  It captures those who are not in receipt of benefit.  They may not be in receipt of benefit for a variety of reasons:  their spouses may be employed and their income may be above a certain threshold.  The labour force survey is valuable in that it picks up something that is not in the jobseeker's allowance data.  The green circle represents those in the claimant count who are employed — 8% or an estimated 7,000 people — and the rightmost part of the blue circle are those persons who are in receipt of jobseeker's allowance but who are defined as inactive.

The red circle on its own is about 70,000 persons, and the blue circle on its own is 64,000 persons.  However, if you take out the double counting, you will come out with an estimate of about 86,000.  That picks up some of the discouraged workers who are not on the labour force survey but who are in receipt of jobseeker's allowance.  It broadens the definition and reconciles the two measures.  If, on the unemployed measure, we were to take out people who are inactive and in receipt of benefits and those who are employed and in receipt of benefits, we would be back to the LFS estimate.  They are in good agreement.  That gives you that broader measure of unemployment.

The next slide asks whether there is anything about the economic status of our people in receipt of jobseeker's allowance that is different from recipients in the UK.  The differences are not big, but the green bar at the top left-hand side that is labelled "Northern Ireland" refers to the economically inactive in Northern Ireland, and there are slightly more people who are in receipt of jobseeker's allowance in Northern Ireland who are economically inactive than in the UK.  We have 15% of people on jobseeker's allowance who would be defined as economically inactive, whereas the UK has only 14·2%.  It is not a big difference — it is one percentage point — but this is a census of people, so that could be contributing to the difference.  We tend to have fewer employed people in receipt of jobseeker's allowance compared with the UK.  If a proportionately bigger share of our population is economically inactive, and if we have more people in receipt of jobseeker's allowance who are economically inactive, that is possibly why our rate is markedly higher than that of the UK for jobseeker's allowance.

In the next slide, the blue line is the labour force survey, the red line is the claimant count measure for Northern Ireland and the bars are the statistical margins of error that I was talking about.  We said that the claimant count had fallen by 600, whereas the LFS is showing an increase in unemployment.  You can see that the decrease in the claimant count falls within the margin of error of the labour force survey.  It is a relatively blunt tool, and that is why the labour force survey is not picking up the minor changes in the LFS.

The next slide tries to put that into a longer historical context and runs back to 1971.  The red line is the claimant count measure, and the blue line is the labour force survey measure.  In October 1986, the red line peaked with about 123,000 people on the claimant count measure.  We are currently at 64,000.  Although unemployment is bad at the minute, you can see that, in the historical context, we are still some way shy of the worst of unemployment back in 1986 and 1993.

The next area of interest to the Committee is the long-term unemployed and the youth unemployed.  To give you a feel for the scale of the issue, we have an estimated 69,000 persons unemployed aged between 16 and 64, and 42,000 of that 69,000 — the purple part of the pie chart on the right-hand side — are long-term unemployed.  Therefore, 61% of the unemployed are long-term unemployed, which is a year or more, and that has gone up in recent months.  That has been an issue for some time.

The next slide has the same figure of 42,000, but it is presented as a rate.  The blue line is Northern Ireland, and the red line is the UK.  This shows the percentage of the economically active who are long-term unemployed.  You can see that our long-term unemployment has started to accelerate since the end of 2011, whereas there has been a gradual rise in the UK but not as marked as that for Northern Ireland.

The next slide takes a similar approach to the youth unemployed.  Just to give you a flavour for the scale of the youth unemployment problem:  23,000 young people on the labour force survey who are aged between 16 and 24 are unemployed.  That represents just under one third of the total number who are unemployed.  I have split that 23,000 further into those young people who are long-term unemployed, which is 11,000, and those young people who are not long-term unemployed, which is 12,000.

Mr P Ramsey: You previously referred to NEETs without giving any figures for that.  At present, are there any numbers on those in the NEET bracket, separate from the numbers for youth unemployment?

Dr Gillan: Yes.  That is dealt with just a little bit further on in the pack.  This slide gives you a sense of the scale of unemployment.  That figure of 11,000 is a proportion of the 42,000 long-term unemployed.  By comparing the light purple section with the dark purple section, you can see that about 26% of the long-term unemployed are youth unemployed aged 16 to 24.  That shows the scale of the issue.

The next slide shows the trend in youth unemployment rates for 18- to 24-year-olds.  Again, the blue line represents Northern Ireland, and the red line represents the UK.  You will see that 21% of 18- to 24-year-olds in Northern Ireland were economically inactive or unemployed compared with 18% in the UK.  So, the difference is not as marked for youth unemployment as it is for long-term unemployment.

The next slide shows youth unemployment in respect of the claimant count rate.  I want to point out that the figure here is higher than in the UK.  The rates tended to part company in and around the end of 2010.  The Northern Ireland rate on this measure is about 7·5%, and the UK rate is 5·4%.

The next slide presents youth unemployment on the basis of the claimant count by district council area.  I will not go into this in detail, but I want to let you know that the information is there if you want to refer to it at a later stage.  The worst performing areas in that respect are the same as those for the 16-64 population.

You asked us where people go when they leave a benefit.  This next slide shows the destination for youth claimants aged 16 to 24 when they leave jobseeker's allowance.  An average of 39,000 16- to 24-year-olds have left jobseeker's allowance over the past year.  About 34% of them, which is roughly 13,000, went into employment; 20% went into another benefit; 4% went into training, 3% went abroad; and 1% did "other".  There is quite a large group there —

Mr F McCann: Sorry, James.  You said that 3% went abroad.  Does that mean that they emigrated?

Dr Gillan: That would have been the reason that they gave.

Mr F McCann: How many people is that?

Dr Gillan: It is about 1,000, but we do not know how accurate that is.  Does somebody tell the jobs and benefits office the real reason they are leaving?  It is a stated reason in any case.

You can see from that 37% that we do not know where quite a few people have gone.  That is either because they failed to register, did not turn up or ceased to claim, or because we do not know.

Mr Ross: Is the definition of "abroad" leaving the UK?

Dr Gillan: I would need to check that out for you.

Mr F McCann: On that point, I know quite a number of young people who do not tell those at the dole that they are going away; they just head off, whether it is across the water or to Europe, Australia or America.  All the indications you have given are that there has been a serious increase in the number of people going to other countries.  It has been difficult to get a ballpark figure for how many people that is.

The Chairperson: A large swathe of that could be those "37% unknown".

Dr Gillan: That issue was raised at an earlier Committee meeting.  A note in your briefing pack on youth migration shows that among 18- to 24-year-olds, there was a net migration of 215 into Northern Ireland from outside the UK.  Taking into account people coming from or going to GB, there was a net outward migration of 2,292.  Some use a residence in GB for the purposes of higher education, so some of those will be in the statistics.  They are not all leaving to work in GB.

Mr F McCann: Do those figures include people going to work in Britain?

Dr Gillan: Yes.  If they went to find work in GB, they would be included.

Mr F McCann: Would they be included in the emigration figure?

Dr Gillan: Yes.  The end is nearly in sight, so I will keep going.  The next slide relates to the economically inactive.  There are 69,000 unemployed people in Northern Ireland, which gives a sense of the scale of the situation.  An additional 51,000 are classified as inactive but would like to work.

The next slide gives the main reasons for economic inactivity in Northern Ireland.  Students account for 32%, while 24% say that they are inactive because of illness.  Another 24% look after the home or family, and 13% are retired.  The remaining 7% are "other".  A lot of people are economically inactive because they are full-time students.  Outside that, one quarter of the economically inactive report themselves as sick.

As I said, 24% of economically inactive persons said that is due to long-term sickness or disability.  In Scotland, that figure is 29·5%.  The Northern Ireland percentage is also lower than Wales, and the north-east and north-west of England.

Mr P Ramsey: Where do you get the figure of 24·4% for Northern Ireland?

Dr Gillan: The labour force survey, which interviews householders, showed the reasons for inactivity.  The interviewer will go to or phone a householder and ask whether somebody was working in the reference week available to work.  If they are classified as inactive, the interviewer will ask their main activity in the reference week:  whether a student, looking after the home, or whether they are long-term sick or disabled.

Mr P Ramsey: How many surveys took place, and how many were by telephone?

Dr Gillan: We issue around 2,200 surveys every quarter, and the response rate is in the order of 65% or 70%.  The status of everybody in the household is taken, so I think that we are coming out with about 4,000 or 5,000 surveys each quarter, which is what this is based on.  Obviously, this is a proportion of that.

Mr P Ramsey: I ask in the context that, historically, in the lead-up to and during the welfare reform debate, we argued that we have more long-term sickness because of the legacy of the Troubles, but the figure seems low compared with other regions.

Dr Gillan: We are above the UK average.  The figure for the UK is 22·3%, and Northern Ireland is at 24·4%.

Mr P Ramsey: We are below Wales and Scotland.

Dr Gillan: We are.

Mr P Ramsey: That is surprising.

Dr Gillan: It is surprising, and I think that, historically, it might be interesting to look at the figures over time as that might have changed.

Mr Dukelow: Overall, our inactivity is higher than that of any other region of the UK.  Two bits are really out of kilter with the rest of the UK as a whole:  we have more students as a proportion of the inactive; and we have more inactivity because of sickness and disability.  The comparison with other regions is an interesting one.

Mr P Ramsey: Two weeks ago, DEL gave a presentation to the Committee on the economically inactive, and one official suggested that two thirds of those on long-term sickness benefit were not really sick.

Mr F McCann: This may be an unfair question, but I thought that, if you were on incapacity benefit, you were not classed as economically inactive:  is there a level of benefit, not for long-term sickness, that disqualifies you from being classed as economically inactive?

Dr Gillan: Whether someone is employed, unemployed or economically inactive, we also ask them what benefits they receive.  This relates to the definitions of inactivity.  If someone was not available or was not seeking work and in receipt of incapacity benefit, we would classify them as economically inactive, so they would be included.  The whole 16-plus population is broken down into three categories:  employed, unemployed or economically inactive, and that is irrespective of what benefits someone is in receipt of.

Mr P Ramsey: The other point is that, presumably, there will be a dramatic shift because we are informed that tens of thousands of people will migrate from sickness benefit to other benefits.  When will you next review and analyse these areas?

Dr Gillan: Irrespective of which benefits people move on to, the questions will stand:  are you long-term sick or disabled; and are you economically inactive, employed or unemployed?  That is the beauty of the LFS.  We also ask people about their benefits.  It is not terribly precise when you get down to that level because you are disaggregating a disaggregated group, but you would be able to see in the LFS any trends in long-term sickness according to a standard definition.   We can go back and look at what the reason for inactivity was over time and see whether that has changed compared with now.

Mr Dukelow: I do not mean to dwell on this point, but some people in employment also declare themselves sick or disabled.  What we find, and there has been some research on this in the past, is that a smaller proportion here declare themselves sick and disabled when in employment than is the case in other parts of the UK.  I suppose that you would have to do a bit of work to discover the reasons for that.  Proportionately, we also have more people declaring themselves as sick and disabled than in the UK as a whole.  There are a few issues to be considered.

The Chairperson: We will move on to the NEETs category.

Dr Gillan: The next slide is similar, but it deals with 16 to 24-year-olds.  It shows that, in the younger population, we are below the UK average for those who are long-term sick or disabled.

The next slide deals with the NEETs population.  NEETs is a category that crosses a lot of different boundaries.  You can be inactive and NEET or unemployed and NEET.  This slide looks at all the different categories of economic status, but I draw your attention to the two slices in the top left-hand quadrant of the pie chart, coloured light blue and orange, which show that there are 21,000 people in each group.

The light blue slice represents those young persons who are inactive, not in full-time education and not in employment.  The orange slice represents people who are unemployed and not in full-time education.  If you add the two, the total represents the group of people who are not in education, employment or government training.  It shows that about 9·6% of the age cohort are NEETs unemployed and 9·6% are NEETs inactive.  In total, about 19% of the age cohort are considered NEETs.

The next slide gives that information in tabular form.  The total number of NEETs is 41,000, or 19%, relative to the other groups.  The following slide gives the times and trends; the blue line on the graph represents Northern Ireland and the red line represents the UK.  The graph represents the 16 to 24-year-old NEETs group and shows that about 19% of the age cohort in Northern Ireland are NEETs, and the UK figure for the same cohort is 17·8%.  Northern Ireland is a bit more volatile, but the difference is not that marked.

The final slide is the summary that you saw earlier and is for reference.  I am happy to answer any questions that you may have.

Mr P Ramsey: I am bewildered by your NEETs figures.  Your analysis shows that 41,000 young people are NEET, but the LFS estimate is 51,000.  Last week, OFMDFM told us that the number was between 44,000 and 46,000.  I recall DEL telling us that there were 54,000.  Who is accurate?  There are big fluctuations in the numbers.

Dr Gillan: The fluctuations may arise because different people report their estimate at different times.  These are the most recent estimates for January to March 2013.  When we were in front of the Committee on previous occasions, the estimate would have been slightly different.

Mr P Ramsey: Do you believe that your figures are the most accurate?

Dr Gillan: We can say that they are accurate to within a certain percentage.  The figure is an estimate — so it is 42,000 plus or minus x per cent.

Mr P Ramsey: This is from your briefing paper:

"For the period, October - December 2012, the LFS estimated that there were 51,000 young people".

That was just over a year ago.  Even taking it for granted that that figure was accurate, what happened to the other 10,000, given that we know that 10,000 young people have probably been added to the figure?  I am at a loss, James, to be quite honest.  If I put in a written question today to DEL, I know that its evidence will be more accurate.  I think that it will come back and tell us that it believes that 54,000 people are NEET.  So where are they getting the figures from?

Dr Gillan: The LFS is one of the key sources for measuring those not in education, employment or training.  You said that there were 51,000 in October to December 2012.  That comes from the LFS, and this comes from the LFS.  They refer to two different periods, but you have put your finger on one of the limitations of the LFS.

Mr P Ramsey: You are not suggesting that the LFS now tells us that 10,000 fewer people are NEET?

Dr Gillan: What I am saying is that the estimate for October to December and the estimate for January to March are based on sample surveys.  There is a sample of 2,500 each quarter.

Mr P Ramsey: I am not being awkward, but could it be that far out?  Based on a sample survey a year ago, it is out by 10,000.

Dr Gillan: Let me talk about the unemployment rate.  When we measure the unemployment rate, we say that it is 8·1%.  The statistical margin of error is plus or minus 1·4%.  So we are saying that it is in the range 6·7% to 9·5%:  if we went out and counted everybody who was unemployed, the true measure would be between 6·7% and 9·5%.  That is what you get from a sample survey, and ditto with the estimate of NEETs and the margin of error.

Mr P Ramsey: I accept that, but your evidence shows an increase in youth unemployment.  Based on the figures that you have presented, is that accurate?

Dr Gillan: Yes, over the month.

Mr P Ramsey: There has been an increase, but you also suggest that there has been a substantial decrease in the number of young people who are NEET.  Chair, I think that we need to get our research staff to look at this again.  We held a full, thorough inquiry into NEETs, and we know the figures.  I am at a loss to understand these figures and where they come from.  I know that they are based on samples, but the samples cannot be that far out.  Maybe our research staff could do a bit of work on that, Chair?

Dr Gillan: It is a valid point.  I said earlier that when you look at LFS unemployment rates, quarter on quarter changes, unless they are really dramatic, are within the random error associated with a survey of that magnitude.  Over the longer period of a year, a change of two percentage points in the LFS unemployment rate is statistically significant and beyond what would be expected from random variation, and ditto with NEETs.  If a shift is marked enough, we can say that it is beyond what would have been expected from random variation.  If you wanted to achieve precision, and you said that you would like to be able to measure, using the LFS, changes of 5,000 or more —

Mr P Ramsey: Do you not accept that there is a huge variance?

The Chairperson: It is restricted by the sample size.

Dr Gillan: The variance is a function of the sample size, and the sample size is a function of the cost of interviewing people on the doorstep or phoning them.  If we wanted more precise estimates, we would have to invest in a larger sample size.

Mr Ross: I want to return to the number of young people leaving the country for work.  I accept that 3% said that they were going abroad, and there may be a section of the 37% who are abroad, but to make sense of the figures, we need to know how the situation differs from previous years because many people will go abroad voluntarily to get work or on a gap year.  If you do not have any figures now, it might be useful to supply us with those later to show how 3% compares with previous years.  Otherwise, we do not really know whether they are being forced to go abroad for work or are doing so for another reason.

Dr Gillan: We can get you a time series on that.

Mr Ross: That would be useful.

Mr Douglas: James, will you e-mail your presentation to us?

Dr Gillan: Sure.

The Chairperson: Gentlemen, thank you for your time and for answering our questions.