29/06/2011: Live Register - June 2011

Live Register figures for June are out, so let's updated the charts.

Headline numbers are for implied standardised unemployment rate which is up in June to 14.2% from 14.1% in May. Remember, this change is reflective of the past adjustments made to the Live Register-implied standardised unemployment rate made on the foot of latest QNHS-derived official unemployment rate which in Q1 2011 stood at 14.0%.

So:

• Standardized unemployment rate was driven up by 2,900 new entrants to seasonally adjusted LR (+0.7% mom)
• Overall in June 2011 there were 457,948 people signing on the Live Register (not seasonally adjusted) -- an increase of 5,066 (+1.1%) yoy. In May 2011, yoy increase was 3,025 (+0.7%)
• Increase in the first 6 months of the year was 37,420(+9.0%) relative to the first six months of 2010. Obviously, that is down from a massive jump of 194,651 (+88.2%) in Live Register for the first six months period between 2009 and 2010

In seasonally adjusted terms: there was a 5,000 (+1.13%) increase in the numbers on Live Register in June 2011 relative to June 2010. May 2011 increase on May 2010 was 2,900 (+0.66%). Taking 3-mo MA: current 3mo MA (Q2) rose 0.604% on previous 3-mo period (Q1) and current (Q2) 3mo average is up 0.925% yoy. This suggests renewed acceleration, albeit weak, in LR numbers. Monthly increases in LR were 2,900 (seasonally adjusted) for June and May, a decrease of 600 in April and a rise of 1,000 in March. Not an exceptional level of volatility in the series to suggest general upward trend. Since January 1, 2011 we have added 5,300 to the seasonally adjusted Live Register figures.

Some other results from the latest release:

• Seasonally adjusted, there was a monthly increases of 2,000 males and 900 females, but year on year the number of females on LR has risen by 3.8% (+5,975) while the number of males is down 0.3% (-909). In the six months through June 2010, 15,797 (+11.3%) females were added to the LR against 21,623 (+7.9%) males
  
• Perhaps expected, but extremely worrisome is the rise of long term claimants by 49,448 in the year to June 2011, in the words of CSO: "bringing to 40.8% the number of claimants that have now been on the Live Register for one year or more. In June 2010 long term claimants made up just over 30% of the total Live Register." The problem with is that long duration of unemployment spell implies structural unemployment, with requisite loss of skills, continuity of experience and subsequently reduced employability.
• Youth unemployment has eased somewhat in June. LR for those under 25 years of age has fallen 4,500 year on year (-5.18%), but zero percent mom. In May 2011, yoy decline in this category was 4,800 (-5.50%), so not exactly an improving trend here. But at least some good news. In contrast, LR for those 25 years of age and over increased 9,500 yoy (+2.68%), against May 2011 yoy increase of 7,600 (+2.15%).

• Numbers of casual and part-time workers on LR increased 6,526 yoy (+8.23%) in June, up from the increase of 6,058 (+7.68%) in May, suggesting that the overall quality of employment gained by LR signees is low. Current 3mo average yoy rise is 8.04% against previous 3mo average rise of 7.82%. Not dramatic, but certainly not signaling any improvement in jobs creation quality.
• Lastly, the numbers of non-Irish nationals on LR has risen by 76 in June 2011 relative to June 2010 (+0.1%) while the number of Irish nationals was up 4,990 (+1.33%). In May 2011, yoy changes in these two groups were: -1,091 (-1.40%) and +4,116 (1.14%). 3mo averages also suggest strong divergence in LR in favour of lower non-Irish nationals participation and higher Irish nationals participation.

2/02/2011: Live Register for January

Live Register was out today for January with some positive news. The headline figure shows a seasonally adjusted decline in LR of 6,900. Unadjusted, LR rose 5,598 over the same month. It's hard to argue just how accurate are seasonality adjustments off the trend line, but let's stick to seasonal figures, so we can at least feel somewhat better about our predicament.

Charts below shows the slight decline in LR-implied rate of unemployment and the drop in LR itself:
So LR-implied unemployment rate is down from 13.6% in December to 13.4% in January.

Average weekly dynamics are looking pretty strong (in part, these are due to 4 weeks month - December, for example covered 5 weeks):
Monthly rate of change:
Also looks strong.

These are all headlines so far. Now, let's dig a little deeper:
Number of LR signees who are non-Irish nationals rose 1,882 to 78,527 in January 2011 - an increase of 2.46% mom and a decline of 3.52% yoy. Number of Irish nationals on LR rose 3,716 in January 2011, a rise of 1.03%mom and 2.42%yoy. So as chart above clearly shows, the two series are moving in the opposite directions. There were 4.64 Irish nationals on LR per each non-Irish national in January 2011. The same number for January 2010 was 4.37.

Another alarm bell is ringing for the increasing number of those in part-time and occasional employment (presumably not all happy to undertake less than full employment). This number rose from 82,058 in December 2010 to 83,232 in January 2011, with monthly increase of 1.43% and a year on year rise of 8.29%.

Economics 2/12/10: What PMIs tell us about the job market

An interesting additional point of view on jobs market. Today's Manufacturing PMIs suggest no improvement in November jobs outlook in Manufacturing sectors:
So far, there are absolutely no signs of jobs creation here with employment PMIs indicators:

• Services - October reading (latest so far) at 46.2 - well below expansion 50+) and declining on September reading of 49.8; and
• Manufacturing - November reading at 49.3, signaling worsening performance from already contractionary 49.8 in October.
  

Economics 1/12/10: Live Register

Live Register data was out today, throwing some positive news into the generally adverse newsflow. The headline figure is that November LR has declined 4,200 in seasonally adjusted terms month on month.

This follows declines of 5,400 in September and 6,200 in October. In 11 months through November we are still clocking and increase of 9,900. Expressed in weekly terms, chart below illustrates the dynamics.


Now, net average and monthly changes:
Seasonally-adjusted implied unemployment rate dipped slightly again, for the third month in the row:
Unemployment, as estimated by the LR, now stands at 13.5%, having slipped from the high of 13.8% back in August. It is impossible to tell, based on LR, whether the moderation is driven by contracting labour force (with LR dropouts) or emigration (ditto) or outflow of LR recipients to education, or all three. However, some reduction in new jobs destruction can be expected over a period of time of 3-5 months, given the level of jobs destruction prior to mid 2010. Whether this is sustainable trend or a 'dead cat bounce' effect is a matter of time.

One possible glimpse at what is going on relates to the males LR numbers, which has fallen by a larger proportion than female in November. Males unemployment was much faster to rise and started to do so earlier in the cycle, which means that males are now more likely to come off LR and also to emigrate. However, the emigration story might be overplayed here. There was a monthly decrease of 4,698 (-1.3%) in Irish nationals on LR and an increase of 147 (+0.2%) in non-Irish nationals. So, with non-nationals more likely to emigrate (return migration or movement to another third country for employment), these numbers suggest that emigration is most likely not a significant contributor to the LR changes.

On the other hand, based on occupational groups, the encouraging signs are clearly evident:

• The largest percentage decrease was in the Professional group (-6.0%), followed by the Clerical and secretarial group (-3.9%) - potentially, a sign that professional services are starting to stabilize

In contrast,

• In the year to November 2010 the largest percentage increase was in the Other occupations group (+11.2%), while the next largest increases were in the Personal and protective service (+8.8%) and Sales (+7.1%) groups.
  
• The largest percentage decrease was in the Managers and administrators group (-3.7%).
  

So overall, the numbers would be cautiously optimistic, at least as far as potentially signaling a bottoming out of the jobs destruction cycle.

One point of pressure that remains is the duration of unemployment:

• There was a monthly unadjusted decrease of 7,270 (-2.6%) in short term (less than one year) claimants on the Live Register in November, while the number of long term claimants increased by 2,719 (1.8%). This clearly shows that transition into long-term unemployment continues.

Likewise of concern is the quality of employment (although, of course, having at least a part-time job is much better than none at all):

• In the year to November 2010 the number of casual and part-time workers increased by 6,578 (+8.9%).

Economics 3/11/10: Live Register update

As promised, here are the updated charts for the Live Register:
Unemployment (implied rate) above clearly shows the relative size of adjustment over the last 3 months. Chart below shows the last 4 years worth of Live Register:
Next, rates of change
Monthly series clearly showing some serious decreases over the last 2 months.

Economics 3/11/10: Live register

Being away from Dublin this week means I am missing both the Exchequer returns and Live Register data. I will, of course, update the charts on both in due time - probably closer to the weekend.

While I am away, here is the best analysis of the LR data I've seen so far (well done, Brian!) issued earlier today by the NCB Economics team:

" On a seasonally adjusted basis there was a monthly decrease of 6,600 in the Live Register (unemployment claims) in October 2010, following a fall of 5,400 in September. This is the largest monthly fall in the numbers on the Live Register in the last ten years

[I seem to think it is in 14 years, but I might be wrong - again, need my trusted database off my trusted Apple to check]

In terms of flows in/out, which are not seasonally adjusted figures, there were 49,827 new registrants on the Live Register in October. 62,691 persons left the register in October.

It does appear as if job shedding is easing in the economy with redundancies (separate statistics from Live Register) in September down 30% from September 2009 levels . In Q3 2010 redundancies were down 24% from Q3 2009, but despite this the rate of inflows into the Live Register is still elevated highlighting that net job creation is still anaemic given the growth in the labour force.

We have no timely data on employment creation and emigration. In other words it is impossible to decipher whether emigration rather than job creation is causing the large outflows from the live register. It is likely a combination of both, as even in the good times Ireland was characterised by a large amount of churn in the labour market, with for example approximately 13% job gains in 2006 versus 10% job losses for a net gain of 3%. This points to the flexibility of the Irish labour market, which is ultimately required for Ireland to dig its way out of its problems.

The standardised unemployment rate in October was 13.6%, down from 13.7% in September and the peak of 13.8%."

So my two cents to add are:

• Decreases in long term claimants numbers (173 yoy) are small compared to unadjusted decreases in short term claimants (36,008 yoy) suggesting that we might be witnessing some exits from the long term list of older LR recipients (by duration, not age) and simultaneous transfer of newer vintage LR recipients into the long term lists. If true, then it is more likely that as older LR claimants lose their benefits or migrate or both, newer recipients move into their places.
• Net decreases in LR claimants can be accounted in part by the terminations of JB claims and failures to secure means-tested JA status.
  
• The numbers of part-time and casual workers on LR is still rising (+ 1,045 mom), suggesting that quality of employment (remember, we are after higher quality jobs in this country, aren't we) is deteriorating.
• 3,100 exits from the LR are accounted for by workers of age 25 or less, in other words the very demographic that is more likely to engage in education or is at a higher risk of emigrating, suggesting that a significant proportion of the LR decrease might have little to do with net jobs creation in the economy.
• Lastly a quick comment on labour force flexibility referred to in NCB note. In my view,w e do have much more flexible flows out of the country (disregard for now inflows into the country, as these hardly matter in our current economic environment). However, in contrast with previous recessions and certainly in contrast with 2006, the little data we have shows that foreigners and younger Irish are dominating the outflows through emigration, while the longer term unemployed of older age and middle-aged families are stuck either due to lack of skills (the former) or due to negative equity (the latter). This implies that if the current trends continue, we are at a risk of encountering significant drain of talent and human capital out of this country. Of course, our bankruptcy laws will make it impossible for those who emigrate alongside defaulting on a mortgage to come back into the country when recovery takes place.
  

Economics 29/9/10: Live Register

Live Register figures are out for September with the latest CSO numbers showing both an improvement and a deterioration in the labour market. Here are the details.

"On a seasonally adjusted basis there was a monthly decrease of 5,400 in the Live Register in September 2010. The number of persons on the Live Register now stands at 442,417 which represents an annual increase of 22,563 (+5.4%) in the unadjusted series. This compares with an increase of 30,198 (+6.9%) in the year to August 2010."

It is worth noting that the latest decline in the LR is not a "new event", in so far as there have been other events of declining LR (mom in seasonally adjusted series). It will require more than 1 month move down to establish any sort of confidence about the underlying trends reversals:

In terms of weekly and monthly changes, the latest results are certainly welcome:
But the underlying causes of the decline in LR are certainly continuing to point to further pressures in the job market. Per CSO (comments and emphasis are mine):
"In September there were 39,960 new registrants on the Live Register, which compares with 36,194 in the previous month [new registrants to LR rose significantly month on month in September, implying clearly that decline in overall LR was not driven by improvement in the labour market, but by exits from LR by workers who run out of benefits].

"New registrants consisted of 16,952 JB claims (42.4%), 21,276 JA applications (53.2%) and 1,732 other registrants (4.3%). It should be noted that the number of new registrants is not the same as the overall change in the number of people on the Live Register which is affected by
closed claims and the movement of people between schemes. [It is also worth noting that training schemes increases imply reduced LR, but hardly lead to actual jobs gains]

"On average 9,990 new registrants joined the Live Register each week in September, comprising 5,905 males (59.1%) and 4,085 (40.9%) females. By comparison, in March 2010 9,935 new registrants joined the Live Register each week, consisting of 6,340 males (63.8%) and 3,595 (36.2%) females."

These trends - reflecting in my view exits from LR, rather than new jobs gains - is reinforced by the data that in September there was a monthly decrease of 2,412 (-3.1%) in the number of non-Nationals on LR, which suggest strong outflow of workers due to net emigration.

Implied unempoloyment rate now stands at 13.7%, down from 13.8% in August.
This means I will not be changing my forecast for Q3 unemployment to reach 13.9%.

Economics 28/9/10: Live Register v QNHS measures of unemployment

As promised last night - here are comparatives on Live Register and QNHS measures of unemployment.

First raw data:
Pretty close?
Actually 98.8% close. But recall, QNHS is quarterly, LR is monthly, so what about average quarterly numbers for LR and QNHS?

So 99.2% close.

Economics 1/9/10: Live Register

Live Register came in with no surprise - a moderate increase in the calm of August was so predictable, my forecast for unemployment to reach 13.8% in August made back in May came in bang on.

Here are few charts and some analysis.
Per chart above, seasonally adjusted LR rose from 452,500 in July to 455,000 in August, an increase of 2,500. Added cost to the Exchequer - ca €63mln per annum. Added cost to the economy - 2.5 times that. Added cost to the society - much greater than the latter. Added cost to those who lost their jobs and their families - incalculably high.

So far, year on year to August 2010 there was an unadjusted increase in the Live Register of
30,198 (+6.9%). This compares with an increase of 34,403 (+8.0%) in the year to July 2010. Now, that doesn't look like a stabilization to me.

There was an increase of 700 males and 1,900 females in the seasonally adjusted series in August. Which means services are tanking faster than manufacturing.

The average net weekly increase in the seasonally adjusted series in August 2010 was 625, which compares with a weekly increase of 1,700 in the previous month. Chart below illustrates
Both monthly and weekly increases are now below 6 months moving average lines, both of which are still trending up. Momentum suggests moderate mean reversion in September/October. Which brings us to unemployment levels.
Standardised unemployment rate in August was 13.8%, and all indications are it will continue to rise.

Finally, charting the changes together:

My suggestion that latest breakdown between men and women joining LR shows jobs destruction in services sectors is supported by the CSO analysis of detailed data.

Overall,

• Craft and related (25.5%) was the largest occupational group on the Live Register in August,
  
• Plant and machine operatives (15.4%) second largest, followed by
  
• Personal and protection service (10.7%) and
  
• Clerical and secretarial (10.7%).
  

Six of the nine occupational groups showed monthly Live Register increases in August:

• Largest percentage increase was in the Professional group (+1.2%)

In the six months to August 2010 Professional (+26.8%) group also showed the largest increase, followed by Clerical and secretarial (+14.6%) and Sales (+ 11.9%). The smallest percentage increase was in the Managers and administrators group (+1.0%).

Per CSO: "There were increases in six of the nine occupational groups in the six months to August for males on the Live Register. The largest percentage increase was in the Associate, professional and technical group (+12.0%), with the largest decreases in the Managers and administrators and Personal and protective service groups (both -0.8%). For females there were increases in all occupational groups in the six months to August 2010. The largest percentage increase was in the Professional group (+43.3%) followed by the Other occupation group (+25.8%)."

Live Register duration also rose, for males and females in 6 months to August 2010, suggesting severe pressures in the jobs market continue.

Of greater interest will bee changes in the labor force participation - to be shown in QNHS results. I suspect we will see severe contraction in overall number of people working or seeking work in the country.

Economics 5/7/10: The toll of un- & under-employment

Two interesting charts on the ratios of full-time employed and live register signees to total working age population (note, this combines quarterly data from QNHS with monthly LR data, all expressed in quarterly terms):
I can't spot any turn around in either chart, yet both reflect the extent to which the burden of unemployment and under-employment is impacting this economy - on both sides of equation: for those who lost their work (the truly tragic outcome) and for those who have to cover the nation's bills while remaining in employment (also having tougher times).

Economics 1/07/2010: Live Register - no recovery here

Live Register figures for June are truly depressing, folks. Regardless of what our QNA numbers telling us about real GDP growth, unemployment is continuing to climb.
We are now at 444,900 and climbing. In the year to June 2010 there was an unadjusted increase of 37,420 (+9.0%), down from an increase of 43,788 (+11.1%) in the year to May 2010. But that offers little in terms of consolation - most of people on LR in 12 months to May are still there - unemployed or underemployed.
A snapshot of weekly numbers above. Depressing. The average net weekly increase in the seasonally adjusted series in June 2010 was 1,450, which compares with a weekly increase of 1,650 in the previous month. But unadjusted things are looking much worse (figure above).

There was an increase of 4,800 males and 1,100 females in the seasonally adjusted series in June. Undoubtedly strengthening contraction in construction activity in June is not helping here.
The standardised unemployment rate in June was 13.4%. This compares with 12.9% in the first quarter of 2010, the latest seasonally adjusted unemployment rate from the Quarterly National Household Survey. We are firmly on track to reach 13.7% before the end of this year.

Rates of change in LR are also accelerating - a disheartening feature:
As I said in my previous post on QNA data (here): we are having a fake recovery.

Economics 02/06/2010: Live Register - no longer flatlining

Liver Register numbers are out today, erasing much of the optimism that might have been building up about unemployment figures over the last couple of months. Here are the updated charts:
Unemployment rate is now rising again, reaching 13.7% after staying flat for the last four months at 13.4%. My forecast in January was that we will hit 13.5% in May. I was wrong. Well, not as wrong as some of our 'official' forecasters who were saying that we are turning the corner on unemployment...

Let's put today's news into perspective:
Notice 'missing' bubbles before May? That is because we had no growth in LR figures (they were actually shrinking) in February-April. Since the beginning of the year, mom, LR increased by 5,800 in January, fell by 2,300 in February, rose by 600 in March and contracted by 500 in April. With May increase of 6,600, we now have a net addition to the LR over the first 5 months of the year of 10,200. The Exchequer cost of these will be around €325 million per annum. In fact it will be probably higher as contractions in employment most likely have taken place in higher value added sectors, given that construction sector has virtually no jobs left to lose.

Taking a slower snapshot of the LR:
You can clearly see an uptick in the series in May. Given how relatively 'sticky' (trend-driven) LR is, this suggests that we might be heading for a new acceleration.

Next consider average weekly series. These are not seasonally adjusted:
However, taking seasonally adjusted data and extracting monthly series shows that we are firmly above the 'stabilization' line of zero change in the LR mom:
So here you have it - labour markets beg to differ from the Government's official line that 'all indicators point to a recovery'...

Economics 22/02/2010: Detailed analysis of Live Register

Updated (below)

CSO published its analysis of the Live Register Data for 2009 which shows some interesting details.

Per CSO data, reproduced below, the highest risk of unemployment by sector was found in:

• Construction (with LR contribution from the sector reaching 170% of the sector own contribution to total employment);
  
• Hotels and Restaurants (with Live Register contribution from the sector standing at 161% of the sector weight in overall employment);
• Other Production Industries (136%);
• Financial & Other Business Services (131%) and
• Wholesale & Retail Trade (120%).
  

All state-dependent or provided jobs were the safest ones (see above marked in blue bold).

Update: since both Health and Education sectors are heavily reliant on public sector workers, we can consider a broader definition of the Public Sector to include the above sectors together with Public Administration & Defense. In this case, broader PS accounted for 23.1% of total employment in Q4 2008 and 6.9% of total number of new LR signees in Q1 2009, implying a 29.9% relative incidence of unemployment by sector - a number that is more than 3 times smaller than the average for the entire economy.

The above relative incidence number for the broader PS is actually biased in the direction of overstating the overall incidence of unemployment in the PS, as a number of employees who lost their jobs in Health and Education sectors were most likely from private firms providing these services.


And here is another table, also slightly adjusted by me. This time around, I am adding several categories together - people who are left on the Live Register (aka the Unemployed), people who moved from the LR to illness benefit (aka also the Unemployed), people who have retired from the Live Register to a state pension and people who are unaccounted for (aka - emigrants who left Ireland, immigrants who left Ireland and people who just dropped off Live Register into gray economy 'entrepreneurship').

Notice couple of things here - virtually the same number of foreigners and Irish who have joined LR in Q1 2009 stayed in some sort of 'Unemployment' by the end of Q2 2009. Actually, this percentage was slightly higher for the Irish LR signees, but the difference does not appear to be statistically significant.

Those over age 25 tended to remain on LR with higher probability than those who are under 25. The trick part here is that many under 25-year olds went off to training and education, dropping off the LR. One hopes they will have a job to go to, once their Fas-run courses and college programmes end.

Males were more likely to remain broadly unemployed (83.46%) than females (80.26%) but the difference is small and there are several factors here. One might wonder how the birth rate increase affects this number and also how it depends on transition to single parent family supplement. Also, younger women are more likely to undertake new training and education than younger males. Can these three factors explain the difference between men and women in re-employment rates?

Once we look at differences across sectors, one striking detail shown in the table above is that sectors with higher wages and better jobs are suffering the largest non-returns to jobs by the Live Register Signees. Table below details:
So in the nutshell - the jobs our LR signees are getting after they lose their primary occupation are of poorer quality and in less productive sectors.

Economics 03/02/2010: Live Register for January

You've heard the numbers on Live Register results by now, no doubt. A summary, courtesy of CSO:
Here are few charts:
Having breached 430,000 marker this time around, the LR is back on the upward trajectory. As predicted. And with it - unemployment rate:
At 12.7%, we are now in December 1994 territory. Officially, all Celtic Tiger gains in terms of reduced unemployment are now gone. Some 14 years worth of hard labour gone within a span of just 21 months.

Oh, and in case you've heard that we are now doing soooo much better than in January 2009:
It is true - in January 2009, unadjusted LR was rising by 7,251 per week, this January it was rising by 'only' 2,668 per week. Yet, three things worth mentioning:

1. In January 2008 the rate of increases in LR was 2,768 per week - just 100 shy of January 2010;
2. This January saw the highest unadjusted increase per week since July 2009; and
3. Remember - the latest increases are ameliorated by two factors not present in previous years - already high unemployment (meaning that the number of jobs to be cut should be really declining) and high rate of workers outflow from the labour force.