October Live Register Figures

Surely we cannot even start reading the entrails of the unemployment figures until all the necessary corrective action has been taken and the effects of this have filtered through into the labour market.

All this “bottoming out” talk is just “tóin le gaoth”. More “sans culottes” to come, surely?

there are 19,932 JB total (12,514 male and 7,418 female)

the labour market chapter in the OECD report posted up by Philip is worth reading in connection with this.

We have too little information about the characteristics of people who are unemployed. I would analyse such data in a heartbeat if it were available. We know something about regional distribution of course, but it is important that we collect some further survey data soon on this topic.

As has been pointed out before, by Brendan amongst others, “signing on” is a measure of entitlement rather than unemployment. Nonetheless, in the short run, it might be a leading indicator of the flows into unemployment. Has anyone ever attempted to use the Live Register data to forecast the quarterly unemployment data?

@Brendan
I hope you keep posting on this. I think society is beginning to wake up to the fact that all the East Europeans are not likely to go home and everyone in the building related sectors are not going to be able to stay in them. Personally, given our Euro membership, I think the only answer is a once-off downward adjustment to private sector wages.
Retraining programmes & other iniatives are good but I don’t remember anyone claiming our unemployment fell so far in the mid-nineties because of the BTWA scheme or FAS programmes.
We need a big initiative to make the labour market clear.
That’s my (probably overvalued by me) two cents anyway.

Do we have to wait until the next QNHS to see whether there has been a change in the workforce? (Whether through emigration/discouragement/exhaustion of benefits).

The trend in the live register over the last the last 3 months opens the possibility that unemployment has already peaked. While this might still not be the most likely scenario, the assumption built into every forecast that unemployment will continue to rise 6-12 months after the end of the recession just may not be true.

This lagged effect of UE has only been a feature since the 1991 recession – prior to that, unemployment would stop rising within a couple of months of the recession ending. This time, we’re not even sure if the recession has ended yet!

As much of economics is storytelling, how would we tell the story of a quick end to the rise in unemployment (if it turns out to be true)? With some difficulty, thats how. The two main arguments for lagging labour markets are that:

1. Firms hoard labour, given the costs of hiring and firing, which results in unemployment rising sometime after the recession begins, and continues rising after the recession ends.
2. Structural rather than cyclical recessions make reallocation of labour after a recession ends much harder;

On the first question, while hard to measure, the NCC estimate that Ireland’s labour market (along with pretty much most of the world) has become less free since 2000. This is based on survey data and may not be very reliable. On the other hand, unemployment started to rise in Ireland very early, in fact on the eve of the recession in Q4 2007, rather than after the recession had already started. However we are clear on the answer of the second question, that this is a deeply structural recession.

Might an explanation be around whether the difference between a recession that is dominated by (employment heavy) domestic demand rather than (employment light) international demand? For example, a slowdown in export income in early 2000 took time to filter through to lower domestic demand, and the recovery when it came also took time. This time, the recession is directly in the domestic sector, so the beginning/end of the recession is closely linked to how the domestic economy is doing?

A question on how useful the live register is.

I know two categories of unemployed who don’t show up on the register

– people who have been receiving UB for 12 months and don’t bother applying for UA because they figure they’ll fail the means test due to having savings or a working spouse
– sole traders/company directors/partners/sub-contractors etc who were paying the lower class of PRSI and aren’t entitled to UB at all.

Since they just disappear from all official record keeping – not signing on but not in employment, I suppose their number is hard to ascertain except perhaps through the CSO employment stats? In other words, is it more useful to look at the drop in employment rather than a rise in the live register to decide how many people are unemployed? Perhaps some feel its not, because those unemployed but not signing on don’t cost us anything – well except for loss of taxation.

@ Sarah

Another factor worth considering is the occupational catagories of the unemployed.

What is the break down of architects etc vs. site labourers etc.

Arent the site labourers etc more deployable as a labour force than the specialised professions?

Another question there is also the long term value of the return from education from the labour force, assuming x percentage of x level of education unemployed over the long term?

Al

@Sarah Carey

You are right to point these out – a rather large hole from what I understand – though as you point out, difficult to quantify.

I know many who fall into both categories. The second in particular – lots of one man limited companies such as IT contractors who haven’t worked for over a year but are not entitled to benefits and are living on savings until they run out. Almost all of the people in these two categories seem to have mortgages too. A potential time bomb I suspect.

I think the big picture is the switch from JB to JA, especially among males 25 and over. I would therefore expect the next QNHS to show a significant increase in long-term unemployment. Even as total unemployment stabilises, the intractable components of the problem will continue to grow. Also, as Antoin Daltún points out, given that employment has been falling, the fall in unemployment suggests that the labour force is contracting.

I left the figures for those claiming JB aged under 25 out to keep the graph simple – in fact, the number of males in this category peaked in March this year and is now down by 37%, while the number of females peaked in July and is now down by 29%. This is the category that would be affected by change that Frank Walsh mentioned.
Many of you raise the issue of linking these LR numbers to the better measure of unemployment provided by the QNHS. I grappled with these issues during the last unemployment crisis – see the ESRI’s QEC Spring 2003. The topic is worth revisiting. One thing that was true then is that the overlap between those classified as “unemployed” in the two surveys is surprisingly small! Obviously there are many on LR who are not unemployed in the ILO sense (actively seeking work), while there are others (as Sarah Carey points out) who are seeking work but not eligible for any social welfare payment.
The QNHS publishes a useful table (Table 20) giving three “indicators of potential labour supply” (to supplement the headline unemployment rate and capture “discouraged worker” effects, etc). In QII 2009, when the conventional ILO unemployment rate was 12%, these alternative measures varied from 12.5% to 15.8%. But more significantly, they have all increased less rapidly than the headline unemployment rate.