Scottish Economy Watch

Good news on Scottish jobs, but why?

The latest labour market data published today reveal that conditions in the Scottish labour market continue to improve. In the quarter January to March unemployment fell by just under 7,000 to below the 200,000 mark to 199,000, or a rate of 7.3%. The chart below shows that unemployment is now faring better in Scotland than the UK.

What is striking about these latest jobs market data is the surge in employment in Scotland. Jobs rose by 54,000 in the first quarter of this year compared to a fall of 43,000 in the UK as a whole. This brings Scottish jobs performance since before the recession more into line with the UK as the following chart shows.

Scottish jobs are now -1.3% below their pre-recession peak, whereas in the previous quarter they were more than 3% below the pre-recession peak. This is still worse than the UK, where, despite the loss of jobs in the recent quarter the total is 0.6% above the pre-recession peak.

So what is going on here?

First, we cannot discount the possibility that the Scottish data reflect measurement or sampling error, more so than the UK data where samples are relatively bigger. The latest 'surge' may reflect a correction of earlier errors, or a large error in the latest measurement. Only later data will help clarify that.

Secondly, if we assume no measurement error, the explanation could be as follows.

Since the fourth quarter of 2009 when there was a large loss of jobs in Scotland, the Scottish jobs market has been weaker than the jobs market in the UK as a whole. Yet, GDP growth has been very similar in Scotland and the UK. On the face of it, that meant that productivity worsened more in the UK than Scotland. Again we also cannot discount measurement error affecting the GDP estimates both in Scotland and the UK. There may also be different industrial/structural reasons, including the relative number of 'zombie' firms in Scotland and the UK, which would affect the relation of employment to output in Scotland and UK - I assume that the contribution of such firms to GDP is much less than their contribution to employment, other things equal.

But setting all those possible explanations aside, maybe, just maybe, Scottish employers took a more pessimistic view of the course of the recession and recovery than their UK counterparts. Hence, employment was cut back here but nor in the UK. Now we may be seeing a correction as Scottish employers revise their expectations up considering that conditions are not as bad as they expected. Conversely, UK employers may be in the process of revising their expectations downward: the recovery is not as strong as they anticipated, hence cutback on employment.

Only a theory, but it does reconcile the different Scotland-UK jobs market behaviour with the similar output/GDP behaviour.

What do you think?

Stronger signs of recovery in Scottish labour market

The latest Scottish labour market statistics published this morning serve further to reinforce the view that a more sustained recovery is now underway.

Employment during the quarter to Dec - Feb 2013 rose by 39,000, or 1.6%, and by 20,000 over the year, 0.8%. In the UK, employment was almost flat, falling slightly by just over two thousand. But over the year, the growth in UK jobs was strong than in Scotland, rising by 1.7%.

Despite this improvement, which is significant, jobs in Scotland are still -2.1% below their pre-recession peak, whereas in the UK jobs are 0.4% above their pre-recession peak. The Scottish jobs position is shown in the chart below

Unemployment fell in Scotland during the quarter by -11,000, or -5.2%. In the UK, in contrast, unemployment rose by 70,000, or 2.8%. The measured ILO unemployment rate is now 7.3% in Scotland compared to a much higher rate of 7.9% in the UK. Over the year, unemployment has fallen by -21,000, or -9.8%. In the UK the number unemployed fell by -71,000 over the year, or -2.7%, a much slower rate of decline than in Scotland.

One factor that has tended to flatter the Scottish unemployment figures compared to the UK has been the rise in the number of inactive workers, which has been stronger here. This has tended to depress the measured Scottish unemployment rate compared to the UK. However, in the most recent quarter the number inactive fell by -23,000, or -1.5%, whereas in the UK inactive numbers rose by more than 25,000, or 0.1%.

The fall in inactive numbers has reduced my estimate of the 'real' level of unemployment; see earlier blogs, as the chart shows

'Real' unemployment is now 55,000 higher, at 9.1%, than measured unemployment compared to 78,000, at 10.4% in the previous quarter.

The latest jobs and unemployment data are encouraging and taken with the latest GDP data, see forthcoming blog, do suggest that the Scottish economy is recovering if slowly from the recessionary period.

However, it is worth sounding a note of caution. From the latest data, to December 2012, it appears that full-time employment continues to fall, with jobs growth driven by the growth of the part-time and self-employed. As I noted here we can only be certain of a sustained recovery in the Scottish labour market when the number of full-time employees begins progressively to rise.

Whither full-time employment?

The latest labour market data for Scotland allow us to update the analysis of the composition or type of employment change.

The updated chart to Oct 2011-Sept 2012 is shown below

While these data lag the latest total jobs and unemployment data to Nov-Jan 2013, they offer a pretty stark reminder of the significant change in composition that lies beneath the overall employment numbers.

Full-time employment has been the main casualty falling by -7.2% from its pre-recession peak. Part-time and self-employment, in contrast, are +8.7% and +11.3%, respectively above their pre-recession heights.

While there are clearly structural changes occurring in the labour market on the demand side, with more workers expressing a preference for part-time and self-employment, the bulk of the change is largely driven by supply-side rather than demand side change. Put more simply this means that the switch is more a reflection of the wishes of firms hiring workers than the choice of workers themselves - see the STUC here.

I suspect we can only be certain of a sustained recovery in the Scottish labour market when the number of full-time employees begins progressively to rise.

Recovery or false dawn in Scottish jobs market?

Last week the latest Scottish jobs and unemployment data were published and somewhat lost amidst the media coverage of the Budget.

These labour market numbers were clearly better than they have been for some time. During the quarter to Nov-Jan 2013 employment rose by 19,000, unemployment fell by 4,000 and the numbers inactive also fell, by 11,000. The 16 plus activity rate rose slightly from 62.3% to 62.6%, the unemployment rate fell from 7.6% to 7.4%, and the inactivity rate fell from 37.7% to 37.4%.

All good news.

Moreover, given the fall in the numbers inactive my estimate of 'real' unemployment also dropped relative to the measured number and rate of unemployment - see chart

Clearly, one swallow doesn't make a summer.

The estimate of real unemployment is still 63,000 higher than the ILO unemployment measure for Scotland; that is 2.2% points higher than the official rate of 7.4%.

The jobs numbers also continue to look weak when compared to the pre-recession figures as this chart shows

After nearly 5 years employment remains 68,000 below the pre-recession peak, a short-fall of -2.7%.

Added to this, with the working population rising, the ratio of employment to working population still looks like an economy in recession:

One would like to hope that we are beginning to see the start of a more sustained recovery in the labour market in Scotland. But several more quarters of positive data are required before we can conclude that hope has finally triumphed over experience.

Jobs Market Weakens Again

The latest jobs market data provided further evidence that the labour market in Scotland continues to weaken. The positive element in the latest figures was a strong fall in unemployment of 14,000 or -6.4%. In the UK, unemployment also fell by 37,000, a much weaker fall of -1.5%.

Yet, the jobs market was much weaker in Scotland. Employment fell by 24,000, or -1%, in Scotland. But in the UK net new jobs were created. Employment rose by 90.000 in the UK. An increase of 0.3%.

So, falling unemployment in Scotland provides an illusory view of the absolute and relative state of the Scottish labour market.

It is also incorrect for the media to report the latest statistics as yet another fall in unemployment in Scotland as the BBC did here. In the data published last month a fall of 19,000 was reported. Now there is a fall of 14,000. But we are dealing with quarterly data not monthly data. The latest quarterly data published today compares the quarter September to November with the previous June to August quarter. Last month's data compared August to October with the previous quarter. Hence in the quarterly data published this month and last month, two of the three months are the same: September and October! The latest figures add November and exclude August.

So, while unemployment fell this quarter, it rose by 7,000 in the previous quarter after falling successively by 3,000 and 13,000 in the two quarters before that.

However, as I noted in earlier posts on the Scottish labour market - here and here - one reason for the apparently stronger unemployment performance in Scotland has been the rising number of inactive workers. In the latest quarter the numbers inactive in Scotland rose by 42,000 or 2.7%. In the UK in contrast the inactive number rose by absolutely less, 39,000, an increase of only 0.2%.

My hypothesis is that the rising inactive numbers in Scotland is a consequence of disproportionate weakness in the Scottish jobs market. Workers are discouraged by the lack of available jobs to cease looking for work and leave the labour market.

If, as before, I assume that the labour market was in equilibrium at the start of the recession then we can compute a 'real' level and rate of unemployment by adding back in this 'discouraged worker' effect. This is done in the chart below.

The hypothesised 'real' level of unemployment is some 80,000 higher than the official rate and on a rising trend.

It is what is happening in the jobs market that, I think, explains this rising trend in 'real' unemployment.

The latest fall in employment is included in the chart below which shows Scottish employment since the start of the recession.

With the latest data, Scottish employment is 96,000 or -3.7% below its pre-recession peak. It currently is only 23,000 above the trough of the 2008-2010 recession. Any recovery that there was in Scottish jobs faltered after the March - May 2011 quarter and has essentially stagnated ever since.

However, that is not the end of the story because it gets worse!

The working population is rising in Scotland so if you express employment over the total number of workers economically active and inactive we get the next chart.

The situation is now worse than at the worst of the recession with employment- working population ratio now -5.7% below its pre-recession peak compared to -5.6% in the trough of the recession.

I know that we must not put too much store by one month's or one quarter's figures. I know also that the Labour Force Survey has a fairly small sample and so is subject to sampling errors which are large.

Yet, my judgement leads me to believe that what we are dealing with here is not measurement error but a real problem in the Scottish jobs market.

 

'Real' Unemployment is Rising

The jobs market data published this morning show both employment and unemployment falling in Scotland. Total employment fell by 27,000 (-1.1%) in the quarter and by 7,000 (-0.3%) over the year. Yet, unemployment fell by 19,000 in the quarter and 25,000 over the year. The unemployment rate now sits at 7.6 percent, or 204,000, from 8.2% in the previous quarter.

The UK did better on employment with jobs increasing by 40,000 or 0.1% and rising by 1.7% over the year. But the drop in unemployment in the UK was 82,000, which was relatively less than in Scotland and so the UK rate only fell to 7.8 percent from 8.1 per cent.

However, the favourable Scottish unemployment performance is more apparent than real.

In the latest quarter there was a surge in the numbers inactive. That is there was an appreciable exit from the labour market with the inactive numbers rising by 50,000 or 3.2 percent. In the UK, too, there was an increase in the inactive numbers by 134,000. But that amounts to a much smaller rise than in Scotland of 0.7 percent.

The chart for inactive numbers in Scotland is shown below

Inactive numbers have been on a rising trend since almost the start of the recession, increasing by 102,000. This suggests as I noted in this earlier post that underlying unemployment is also rising as workers are discouraged by the lack of available jobs to cease looking for work and leave the labour market.

If, as before, I assume that the labour market was in equilibrium at the start of the recession then we can compute a 'real' level and rate of unemployment by adding back in this 'discouraged worker' effect. This is done in the chart below

On this basis, 'real' unemployment in Scotland is on a rising trend and rose in the latest quarter to 294,000 or 10.7 percent, not the 204,000 and 7.6 percent official figures.

That is a more true measure of the state of the labour market in Scotland.

Labour Income Shares in the Scottish Economy

The STUC's valuable report A Just Scotland see here, raises many interesting questions which the two campaigns for and against independence need to address. One issue they raise in the report is the labour market and the lack of a debate on the likely structure and regulation of the labour market in different constitutional scenarios.

The STUC clearly fears that the current UK approach favouring flexible labour markets has been accepted by both sides of the constitutional debate when they believe the issue should be debated.

I agree.

The STUC argue that flexible labour markets produce "an economy where work is less well paid, less secure and less permanent." In addition, they note the "completely overlooked" macroeconomic consequence of a falling wage share, which, they go on to suggest:

If the wage share in the economy of an independent Scotland continues to fall then the only way in which the products and services the economy produces can be consumed is through rising debt. This hardly provides for the foundations of a stable, prosperous society. (Page 13)

While welcoming further debate, I think the issue is a little more complicated than this.

First, there is an issue about flexibility. A labour market lacking in regulation seems certain to produce 'sweat shop' conditions, low wages and insecurity. Conversely, a labour market that is heavily regulated seems likely to lead to high hiring costs and low job creation. I'd like to see what the evidence shows on this.

The issue then is can we find an 'optimal' degree of labour market regulation and flexibility that protects workers from vindictive and exploitative employers, while at the same incentivising firms to hire workers and so create jobs? That is the question where I would like to see more theory, evidence and debate.

Secondly, the data on labour shares do not, as far as I can tell, support a view that labour's share of income has been consistently falling in Scotland. The latest Scottish National Accounts Project (SNAP) - experimental - data on compensation of employees is shown in the chart below

Clearly, the data only start in 1998 so I am not sure if the share was falling during the 1990s after the so-called 'Thatcher revolution'. But what the SNAP data show is a rising labour share from 59 per cent to 66 per cent between 1998 and 2001. Then the share fell gently to 62% at the beginning of 2008 when the recession was beginning to bite. During the recession and after the share has oscillated around 63 per cent.

So there is no downward trend during this period of data and other factors seem to be in play as a determinant of labour income share in addition to the UK's fairly flexible labour market regime.

Scottish entrepreneurial buzz, or needs must?

There seems to be a buzz. A buzz that Scotland has become more entrepreneurial. We have the exciting new website published by my good friend Bill Jamieson, Scot-Buzz. On Friday, in the Scotsman, the imaginative nationalist columnist George Kerevan, in a very partial use of statistics to defend both Scottish government policy and the robustness of the Scottish labour market, asserted

In the year to March 2012, the number of enterprises in Scotland increased by a record 10 per cent – fully 30,835 new firms. Not only was this the biggest annual increase since 2000, the growth in new enterprises in Scotland happened at roughly double the rate for the UK as a whole. .... this is hardly indicative that Scotland was under-performing compared to the UK as a whole.

And George Kerevan is correct. The data he uses come from the latest Businesses in Scotland publication which was published on 7 November.

Unfortunately, I do not believe that these data are sufficient to draw the conclusion that we have a new entrepreneurial buzz in Scotland, or that the Scottish economy and labour market is doing very well thank you.

First, we need to say a few words about the data. The businesses that comprise the database are registered and unregistered businesses. The registered businesses are those that are registered for VAT and/or PAYE. The unregistered businesses relate, in the words of the relevant Scottish government web page

a substantial number of very small enterprises which have no employees and operate below the VAT threshold, and are therefore not included on the IDBR (which contains the registered businesses). A modelling procedure that combines data from the IDBR with data from the Annual Population Survey (annual version of the Labour Force Survey (LFS)), Family Resources Survey (FRS) and HMRC Self Assessment tax returns are used to estimate the number of unregistered enterprises.

So, this is not a simple count. There is a lot of estimation and to be blunt: educated guesswork in the series, which nevertheless does have the National Statistic kite mark.

This is how the data look when we separate out the registered and unregistered businesses

Yes, it is the unregistered businesses that are growing. The registered business numbers have been fairly stable over the first decade of 2000s. And the recent jump between 2011 and 2012 is in part to do with HMRC discovering more firms - see here page 2. The statisticians have tried to adjust the earlier numbers to be comparable but there must clearly be a health warning as to whether we are really making a like-for-like comparison with the earlier years of registered businesses.

Now, it is clearly the growth of the unregistered businesses that is interesting. These are the ones with no employees and which are sufficiently small to have a turnover less than the VAT threshold - currently £77,000 per annum - and are not registered for PAYE. Most of them will not have business bank accounts either - see below.

Over the period 2000 to 2012 their numbers have grown by more than 90,000 or 99 per cent. The number basically doubled. But there was something of an acceleration after 2006 with nearly 60,000, or two-thirds, of the rise occurring during the years of recession and limited economic recovery. An even rate of growth would have delivered about 45,000.

While there is now a large body of theory and evidence linking increased entrepreneurship to growth - see here and here - one would also expect a growing economy to raise business opportunities and hence stimulate entrepreneurship - see here.

But what we appear to see in Scotland is an apparent rise in entrepreneurship that has had no effect on growth to date. And this has occurred when the economy has been contracting and stagnating.

A reasonable person might conclude that these data do not indicate that there has been an increase in entrepreneurship of any sizable magnitude in Scotland.

This conclusion is supported by the Scottish clearing bank data on the opening of business bank accounts, which is taken as a one proxy for entrepreneurship. The latest data are shown in this chart

No suggestion of an entrepreneurial resurgence in these data.

So what has been happening?

The answer I think lies in the weakness of the Scottish economy coupled to structural and demographic changes in the labour market favouring part-time, multiple job holding and self employment as this post shows. And also a growing awareness that tax bills can be legally reduced by creating companies and paying corporation tax instead of income tax. But while the creators of the companies may have several different income streams from individuals' portfolio work, they have no employees and can't in truth be considered to be 'proper' businesses.

The chart below shows what has been happening to full-time employment in Scotland.

It has been falling since 2007 and is over 7 percent below its 2007-2008 peak compared around 3 per cent for total employment. Many of these full-time workers that lost their jobs will have gone into self-employment and there is a significant evidence base from across the world to support this "recession-push" hypothesis - of being pushed into self-employment but not necessarily entrepreneurship - for example see here, and here.

The latter study, by Dawson, Henley and Latreille (2009), which analysed the UK labour force survey between 1999 and 2001, did not find much support for the 'push' hypothesis but drew this interesting conclusion

Whether this conclusion would be as robust during the current period of severe economic downturn and rapidly rising unemployment is open to debate and, with suitable data, further future analysis. Indeed it would give cause for considerable concern if the proportion choosing self-employment because it represents the only alternative to economic inactivity was to rise significantly in the next few years. Such 'forced' choices may not lay solid foundations for well-resourced, successful new business ventures.

Indeed. And that appears to be what we are witnessing in Scotland.

Needs must, not entrepreneurial buzz.

Scotland's unemployment differential (Revised)

A revision to the chart presented in this 18 October post. I should have included wage rates as an influence on the supply and demand of labour, which I did in the latest Commentary and now include here

Clearly, differential movement in wage rates between Scotland and the UK could affect both the demand and supply of labour. But we have no evidence so far that this is material in the Scotland - UK unemployment differential.

Where have all the jobs gone?

Today's labour market data highlight the continuing divergence between Scotland and the UK. The jobs market is clearly deteriorating in Scotland while apparently markedly improving in the UK. Employment in the UK rose by 100,000 in the quarter to July - September, an increase of 0.3 per cent. In contrast, Scottish jobs fell by 27,000 a drop of 1.1 per cent.

Looking back on employment since the recession we have the following jobs picture in Scotland and UK.

The UK jobs market has now returned to, and slightly passed, its pre-recession peak, whereas in Scotland employment is still 3.1 per cent below the peak.

These data suggest that output growth in Scotland will have been weaker in the third quarter and I think primarily because of the Olympic 'bounce' effect. Output might be stronger than these jobs figures imply because of the apparent labour productivity effect that has seen output hold up much better than both jobs and the volume of hours, which we noted in the recent Fraser of Allander Institute Commentary here.

But the overall labour market picture is worrying from a Scottish standpoint.

Unemployment continues to rise in Scotland as it falls in the UK. Unemployment rose by 4,000 in Scotland in the quarter to July - September, a rise of 1.8 per cent. In the UK, unemployment dropped by 49,000 a fall of 1.9%. Unemployment now stands at 8.1% in Scotland compared to 7.8% in the UK as the chart below shows

It also worth remembering where we have come from. At peak activity before the Great Recession Scottish unemployment had fallen to 3.9 per cent compared to 5.2 per cent in the UK. While this was a boom period and almost certainly unsustainable it does give a pointer on how far the Scottish labour market has to travel to get back to a more normal or natural rate of unemployment. This is likely to be between 5 per cent and 6 per cent.

Moreover, we cannot say that the rise in Scottish unemployment has been worsened by an increase in labour supply. The chart below shows that the activity rate fell in Scotland in the latest quarter and has been on a downward trend since the start of the Great Recession.

We don't know why this is. It may in part be due to a discouraged worker effect as people leave the labour market frustrated by a lack of job opportunities. This may also link to a rise in outmigration as job opportunities have risen in the UK compared to Scotland.

Unfortunately, the data are not available to provide a detailed understanding of what is currently happening on the supply side of the Scottish labour market. Nor, do we really understand where all the jobs have gone.