As good as it gets? The UK labour market in recession and recovery.

Blanchflower, David G.

Labour markets, both in the UK and internationally, have surprised both in recession and recovery. Contrary to the prior literature summarised in OECD (1994) that argued that unemployment was caused by the levels of unemployment benefits, job protection and other labour market factors including high levels of union density, unemployment grew most in countries a) with large financial sectors (the UK and the USA); b) that had house price bubbles (e.g. Ireland, Latvia, Spain, the UK and the USA); c) that had low levels of job protection and unions (the UK and the USA). By contrast, countries with high levels of job protection and high levels of union bargaining coverage have had relatively small rises in unemployment (e.g. Germany and Austria). (1) The prior literature on the causes of unemployment essentially has nothing to say about this crisis.

In the UK, unemployment did not rise nearly as fast as I, or others, had expected, based upon past performance of the UK labour market and the scale of the shock; and, since a modest recovery began in 2013, it has fallen considerably faster than expected. Overall employment growth has also been healthy and the employment rate is now back to roughly the 2003-7 level, although growth now appears to be flattening off.

Figure 1 plots annual unemployment rates using data from the ONS from 1984 and from 1963-83 using data from the CEP-OECD Institutions Dataset (Nickell, 2006). In the recessions of the 1980s and 1990s when unemployment rates hit double digits, this time, the peak was 8.0 per cent at the start of 2010 when there were just over 2.5 million unemployed.

[FIGURE 1 OMITTED]

This relatively positive experience reflected both sustained structural improvements in the functioning of the labour market over the past 30 years, and a sensible macroeconomic response to the financial crisis: interest rates were cut to the zero lower bound, the Monetary Policy Committee undertook a total of 375bn [pounds sterling] of quantitative easing (QE) and the Labour Government introduced a huge fiscal stimulus.

The flipside of the employment numbers has been the astonishing, and virtually unprecedented, fall in real wages. This implies that while the unemployment (or employment) rate might have been a sufficient statistic to summarise the level of labour market slack before the Great Recession, that is no longer true post 2008. That arises because of the increase in underemployment levels to unprecedented levels, which combined with the unemployment rate keeps wage pressure down.

Table 1 reports the main labour market quantities and rates at the start of the recession in January 2008, May 2010 when the coalition took office, and the most recent data available at the time of writing for September 2014. (2) I also report wages according to the national statistics, the monthly Average Weekly Earnings (AWE); the quarterly Labour Force Survey (LFS) and the Annual Survey of Hours and Earnings (ASHE). Index numbers for the two main inflation indicators, the CPI and the RPI are also shown. In the final two columns I report percentage changes.

A number of points stand out.

* There has been a sharp increase in the size of the 16+ population, up 5.3 per cent since 2008 and 3.2 per cent under the coalition.

* Employment has risen 5.4 per cent since May 2010 but the employment rate, where the 16+ population is used as a denominator, is still below pre-crisis. While the number of people in work is at record levels, so far this simply reflects population growth.

* Self-employment numbers have risen by nearly 600,000 since May 2010 to 14.7 per cent of total employment. The wage data exclude the self-employed. The latest Family Resources Survey shows that real median weekly earnings of the self-employed are down 22 per cent between 2008/9 and 2012/13 and down from 236 [pounds sterling] to 207 [pounds sterling] or 12 per cent on the latest year. (3) So while the self-employed are a heterogeneous group, the sharp fall in average self-employment earnings suggests that much new self-employment is low paid/ low productivity. Internationally, self-employment is generally negatively correlated with development, and there is little evidence that higher self-employment rates are a positive indicator (Blanchflower, 2004).

* Employment has risen by nearly 4 per cent since May 2010 among the UK born, but by just under 22 per cent for the non-UK born. The numbers of workers from the Accession countries has risen by 60 per cent over this period.

* The unemployment rate has fallen to 6.0 per cent, but is still above the starting rate of 5.2 per cent. 35 per cent of the unemployed have been without a job for at least a year, up from 32 per cent in May 2010.

* The number of part-time workers who want full-time jobs--a measure of underemployment--is up by over 200,000. Figure 2 shows this rise is unprecedented; as the unemployment rate fell, the proportion of those in part-time jobs who wanted full-time jobs, expressed as a proportion of all workers, continued to rise. This is consistent with work reported in Bell and Blanchflower (2014), who developed an underemployment index based on desired hours for both full-timers and part-timers.

* The employment rate and the labour force participation rates have remained broadly flat, in direct contrast to the US where the participation rate has fallen sharply (Blanchflower and Posen, 2014). Workers have not left the labour force as they have in the United States, where a major question is whether these individuals will return when labour market conditions improve. There is ongoing debate over the extent to which this increase is cyclical (Aaronson et al., 2014).

* Over the period since the coalition took office the RPI is up 15.0 per cent while the CPI is up 12.1 per cent. Real earnings, deflated by the CPI, have fallen by 4.5 per cent in the AWE (which is likely to be biased upward, see Blanchflower and Machin, 2014), 7.3 per cent in the LFS and 8.6 per cent if ASHE is used and larger still if the RPI is used.

* Blanchflower and Machin (2014) also show that between 2010 and 2013 the fall in real wages in the UK is the 5th largest in the EU, with only Spain, Portugal, Ireland and Greece faring worse.

[FIGURE 2 OMITTED]

Looking forward, there is little evidence that nominal pay growth is picking up, although the sharp fall in inflation means that real wage growth may be positive. The Chartered Institute of Personnel and Development (CIPD) recently asked employers about their pay expectations for the year ahead (in other words, September 2014 to September 2015). (4) For those employers who expected to conduct a pay review in the coming year and who felt able to make a prediction about the likely outcome, the median pay increase was 2 per cent a year, the same as it has been since the winter 2013/14 survey. According to CIPD, one reason why employers are not expecting higher pay rises in 2015--if they feel the need to raise pay at all--is 'because many vacancies still attract plenty of suitable applicants'.

So the big question is why wages have not risen even though there has been a fall in the unemployment rate --and at what point can we expect a return to 'normal' levels of real wage growth? There are at least half a dozen factors we can conjecture that are behind this. While some are cyclical, some are likely to be structural:

1) there is a lag between the fall in the unemployment rate and a rise in wages.

2) Low interest rates have meant that home owners have seen declines in their payments on their variable rate mortgages which has insulated the shock of weak wage growth. Hence reservation wages may have fallen.

3) The impact of globalisation. Union bargaining power is weak and union density has declined sharply, especially in the private sector. Firms can move their production abroad, including to China or other parts of the European Union, including to Eastern Europe.

4) Perhaps more importantly, the rise in the flow of workers, from Eastern Europe, and especially from Poland, has kept wage pressure down (Blanchflower and Shadforth, 2009). The potential for this number to increase further if wages were to rise is high; in the first three quarters of 2014 there have been 509,000 National Insurance number registrations from the A10 compared with 617,000 in the whole of 2013. This keeps wage pressure in check.

5) There is no evidence that the long-term unemployed have any different impact on wage pressure than the short-term unemployed (Bell and Blanchflower, 2014). The MPC is making an adjustment downwards on wage pressure because of high levels of long-term unemployment despite the fact that its own research shows this is inappropriate (Speigner, 2014).

6) Looking forward, the economy and labour market appear to be slowing; there has been a downturn in most business and consumer confidence surveys. The latest REC/KPMG Jobs Report in November finds that the rate at which permanent contracts are being signed is rising at the slowest rate in 18 months, and official ONS data also suggest the labour market is slowing; the consistent upward trend in the employment rate of the last few years appears to have tailed off.

[FIGURE 3 OMITTED]

Of course, the most fundamental problem driving real wage weakness is the 'productivity puzzle'. UK labour productivity remains about 2 per cent below its level prior to the economic downturn in 2008. This is 16 per cent lower than it would have been had productivity maintained its pre-downturn trend. Median wages seem to have become 'decoupled' from productivity growth because of rising inequality, which means that a growing share of the value from productivity growth is absorbed by pensions and higher salaries for top earners (Bell and Van Reenen, 2014).

This is something that the US experienced earlier than the UK, and where real wage performance for the typical worker has remained poor for over 30 years. For significant real wage growth to re-emerge, productivity would need a sharp increase of the kind experienced much earlier in the UK recessions of the early 1980s and early 1990s. There are few signs of this happening, and the problem has been magnified by the UK's dismal investment rates.

Even if productivity was to rise rapidly, the tendency for longer-run inequality trends to cause an unequal division of wages from productivity gains to the top (like bankers' bonuses) would need to be addressed. Until that happens or until policy starts to address these issues seriously, it seems that the prospects of significant, rather than modest, wage increases for typical workers are bleak.

My best guess is that nominal wage growth is not going to rise before the level of slack in the UK economy declines sharply. Figure 3, from Blanchflower and Machin (2014) provides an illustrative example. It suggests that the wage curve, written in underemployment and wage growth space, is flat everywhere from an underemployment rate of 5 per cent and higher. So a reduction in the underemployment rate from 10 per cent to 9 per cent or 8 per cent has no impact on wage growth. Wage growth only starts rising, in this stylised example, when the underemployment rate falls below 5 per cent. We are still a considerable way from full employment, and now it looks like the UK labour market is slowing again. This may be as good as it gets.

Keywords: unemployment; underemployment; wage growth

JEL Classifications: J11; J21; J31; J61

REFERENCES

Aaronson, S., Cajner, T., Fallick, B., Galbis-Reig, F., Smith, C.L. and Wascher, W.L. (2014), 'Labor force participation: recent developments and future prospects', Finance and Economics Discussion Series 2014-64. Board of Governors of the Federal Reserve System, Washington D.C..

Bell, D.N.F. and Blanchflower, D.G. (2014), 'Labour Market Slack in the UK', National Institute Economic Review, 229 August, pp. F4-F11.

Bell, B. and Van Reenen, J. (2014), 'Bankers and their bonuses,' Economic Journal, 124, pp. F1-F21.

Blanchflower, D.G. (2004), 'Self-employment: more may not be better', Swedish Economic Policy Review, 11 (2), Fall, pp. 15-74.

Blanchflower, D.G. and Machin, S. (2014), 'UK real wages: a long way to go', CEP Real Wages Update, December. http://www. piie.com/publications/papers/blanchflower201412.pdf.

Blanchflower, D.G. and Posen, A. (2014), 'Wages and labor market slack; making the dual mandate operation', Peterson Institute Working Paper 14-6, September.

Blanchflower, D.G. and Shadforth, C. (2009), 'Fear, unemployment and migration', Economic Journal, I 19(535), February, pp. F136-F182.

Dustman, C., Fitzenberger, B., Schonberg, U., Spitz-Oener, A. (2014), 'From sick man of Europe to economic superstar', Journal of Economic Perspectives, 28(1), Winter, pp. 167-88.

Nickell, W. (2006), 'The CEP-OECD Institutions Data Set', (1960-2004), Centre for Economic Performance, LSE, September.

OECD (1994), The Jobs Study, Paris, OECD.

Speigner, B. (2014), 'Long-term unemployment and convexity in the Phillips curve', Bank of England Working Paper No. 519.

NOTES

(1) Dustman et al. (2014) argue that part of the explanation of low unemployment in Germany is due to flexibility in the industrial relations system, especially in regard to real wage setting.

(2) Data are available for three month averages so they relate to December 2007-February 2008; April-June 2010 and August-October 2014. It remains unclear why the UK cannot produce timely single month estimates as all other major advanced countries do. An example of this is the monthly Employment Situation Report of the Commissioner of the BLS in the US (http://www.bls.gov/news.release/pdf/empsit.pdf). In the most recent EU data release on unemployment rates, only the UK and Greece produce unemployment data for August; Estonia, Latvia and Hungary published it for September whereas every other major EU country (and the US) produced it for October. http://ec.europa.eu/eurostat/documents/299552116155576/3281 12014-AP-EN.pdf/69clee9f-lblf-4bec-b5cd-5e94eacdbe86.

(3) 'Self-employed workers in the UK, 2014', ONS, August, http:// www.ons.gov.uk/ons/dcp 171776_374941 .pdf.

(4) 'Labour Market predictions for 2015', December 2015, CIPD.

David G. Blanchflower, Dartmouth College and the University of Stirling. E-mail: David.G.Blanchflower@dartmouth.edu.

Table 1. Changes in UK labour force, January 2008-September 2014

                                         Jan. 2008

A) Numbers (000s)

16+ population                   49,160
Employment                       29,662
Self-employment                   3,892
Part-time wants FT                  705
UK born                          25,873
Non-UK born                       3,722
A10                                 571

B) Rates (%)
Employment rate                      60.3
Unemployment rate                     5.2
Per cent unemployed >12 months       24.5
Activity rate                        63.6

C) Wages and prices
LFS (FT weekly)                     511 [pounds sterling]
AWE weekly                          433 [pounds sterling]
ASHE (median weekly earnings)       389 [pounds sterling]
ASHE (mean hourly earnings)          14.0 [pounds sterling]
CPI                                 105.5
RPI                                 209.8

                                          May 2010

A) Numbers (000s)

16+ population                   50,156
Employment                       29,275
Self-employment                   3,949
Part-time wants FT                1,077
UK born                          25,202
Non-UK born                       3,901
A10                                 641

B) Rates (%)
Employment rate                      58.2
Unemployment rate                     7.9
Per cent unemployed >12 months       32.3
Activity rate                        63.2

C) Wages and prices
LFS (FT weekly)                     541 [pounds sterling]
AWE weekly                          449 [pounds sterling]
ASHE (median weekly earnings)       404 [pounds sterling]
ASHE (mean hourly earnings)          14.7 [pounds sterling]
CPI                                 114.4
RPI                                 223.6

                                         Sept. 2014

A) Numbers (000s)

16+ population                   51,758
Employment                       30,852
Self-employment                   4,535
Part-time wants FT                1,319
UK born                          26,132
Non-UK born                       4,748
A10                               1,024

B) Rates (%)
Employment rate                      59.5
Unemployment rate                     6.0
Per cent unemployed >12 months       34.9
Activity rate                        63.3

C) Wages and prices
LFS (FT weekly)                     567 [pounds sterling]
AWE weekly                          483 [pounds sterling]
ASHE (median weekly earnings)       418 [pounds sterling]
ASHE (mean hourly earnings)          15.2 [pounds sterling]
CPI                                 128.2
RPI                                 257.1

                                 2008-2014   2010-2014

A) Numbers (000s)                  Percentage change

16+ population                      5.3          3.2
Employment                          4.0          5.4
Self-employment                    16.5         14.8
Part-time wants FT                 87.1         22.5
UK born                             1.0          3.7
Non-UK born                        27.6         21.7
A10                                79.3         59.8

B) Rates (%)
Employment rate                    -1.3          2.2
Unemployment rate                  15.4        -24.1
Per cent unemployed >12 months     42.4          8.0
Activity rate                      -0.5          0.2

C) Wages and prices
LFS (FT weekly)                    1 1.0         4.8
AWE weekly                         11.5          7.6
ASHE (median weekly earnings)       7.5          3.5
ASHE (mean hourly earnings)         8.4          3.5
CPI                                21.5         12.1
RPI                                22.5         15.0

Source: ONS.

Notes: A10 Accession countries are Bulgaria; Czech Republic; Estonia;
Hungary; Latvia; Lithuania; Poland; Slovakia; Slovenia and Romania.