Scottish Government

Scottish Economic Statistics 2004

A4 Measuring Labour Market Dynamics

Gayle Johnston and Paul Teasdale, Scottish Executive

Introduction

Descriptions of change in the labour market most frequently use levels of employment and unemployment. Labour market statistics provide snap shots which show levels on a monthly, quarterly or annual basis. However, assessing the change in levels of employment does not give a full account of the extent of change taking place in the labour market.

In order to understand the factors that underlie changes in employment it is necessary to examine the gross flows of people into and out of employment. A small change in employment can be the result of a small difference between a large number flowing into employment and a similarly large flow out of employment. Therefore for policy makers who wish to influence change, an understanding of flows is important as it is these flows that many policies aim to influence.

Nature of flows in the Labour Market

Flows in the labour market may follow a number of sequences ⁷. These are:

• Changes in the population - the population of working age changes as people pass sixteen or pass the state pension age, but also because of deaths and migration into or out of the area.
• Changing Economic Status - People move between the three main categories of economic status i.e. moving between unemployment, employment and inactivity.
• Changes within employment - People making job to job moves.

This article examines some of the available data on (short term) movements in the labour market. This is possible by exploring recall data on changes in the economic circumstances of individuals between two dates and by looking at measures of labour market flows i.e. separation rates and claimant count flows.

Data Sources

Information analysing the dynamics of people in the labour market in Scotland can be obtained from various sources:

• The Labour Force Survey (LFS), carried out by the Office for National Statistics, is designed to provide a quarterly picture of the labour market. (Further information about the LFS can be found in Box 4.1.) It can also provide indicators of change and this article makes use of questions in the survey that ask respondents to recall previous economic activity or duration of current activity. The design of the LFS has a longitudinal element as individuals are included in the LFS for five quarters before they are replaced. However this data has not yet been analysed for Scotland.
• The Claimant Count, an administrative series maintained by Department for Work and Pensions (DWP) provides more detailed information about flows into and out of employment.

Labour Force Survey

Changing population

In Spring 2003 the population of working age in Scotland was 3,130,000, three thousand more than in Spring 2002. Change comes from several sources:

• Population ageing - In Spring 2003 there were approximately 70,000 16 year olds moving into the labour market, while there were 50,000 people approaching state pension age (i.e. women aged 59 or men aged 64).
• Migration - The LFS shows about 30,000 people moving into Scotland from UK and outside, since Spring 2002 and about 26,000 leaving in Spring 2002 for elsewhere in the UK. Many people moving into Scotland are students and many leave after their studies.
• Emigrants or deaths - Additional information provided by General Register Office for Scotland (GROS) shows that each year about 10,000 emigrate and 11,000 people die aged 25-64.

Changing Economic Status

In Spring 2003 2,391,000 people aged 16 years and above were in employment. This was 50,000 more than a year earlier. Focusing on net changes ignores the fact that in this year many people left employment and other people moved into employment. For example 185,000 people who had not been working a year earlier were now in work.

LFS respondents are asked to report their economic status twelve months earlier - whether employed, inactive or unemployed. Table A4.1 shows responses, for those respondents who gave information for current and recalled economic status.

Table A4.1: Current economic status shown with recalled economic status a year earlier for people aged 16 and above

The above table indicates that:

• 113,000 people left employment between spring 2002 and 2003, 40 per cent moving to unemployment and 60 per cent to inactivity.
• These outflows were replaced by 187,000 people who moved into employment, 134,000 of whom were from the inactive group and the remainder from the unemployed group.

The net difference between numbers moving into and out of employment (74,000) is greater than the net change in employment mentioned earlier (50,000) because; (a) the LFS sample includes immigrants in the last year, but not emigrants or deaths, (b) some missing responses.

Among those inactive a year earlier were 66,000 people who were then under 16 years of age. The inactive group looks very static but that is because it includes 834,000 people past the state pension age (55 per cent of all inactive people). Among people of working age there is a fairly high probability of moving out of inactivity.

Job to job moves

In Spring 2003 429,000 people in employment (aged 16 years and above) said that they had been in their current job for less than twelve months. In addition to those who had moved from unemployment or inactivity (e.g. people who had previously been in education or carers), there were about 230,000 who had moved from another job. Chart A4.1 shows the division between these three categories over time.

In all years, the number of people moving within the employment category, i.e. from job to job, has been greater than the number moving into employment from non-work. Chart A4.1 shows that the number employed for less than a year increased in the mid-1990s with increases in both the number moving into employment and the number of job to job changes. The number changing to employment from unemployment has fallen as the total number unemployed has fallen.

Chart A4.1: Number of people in current job for less than 12 months, by previous activity, 1995-2003

It is possible to further examine those who moved from job to job and also to identify people who have changed the nature of their job without changing employer. The LFS shows that:

• of those that changed employer 150,000 people also changed industry and 93,000 changed occupation;
• a further 30,000 changed occupation but not employer;
• 45,000 moved from full-time to part-time work and 43,000 changed in the opposite direction;
• 90,000 became managers or supervisors;
• 30,000 moved into self-employment while over 20,000 moved out of self-employment.

The jobs newly filled could be split between those that were previously filled by someone else (who has left employment or changed job) and those that are completely new. The Inter Departmental Business Register (IDBR) shows that between 2002 and 2003 12% of enterprises increased employment. These enterprises along with new enterprises and firms moving into Scotland increased their workforce by about 230,000 (including public sector). Much of this is likely to have been through acquisition of existing establishments, so would not have involved people changing jobs.

Separations

All of the LFS data above compare economic status at two dates 12 months apart. This approach underestimates the total flows as some people will have changed states several times in the course of the year. For example a person employed currently and a year ago could have had a short spell of unemployment inbetween. The LFS includes some questions on separations that help us to describe these flows.

Job separations can be described as 'voluntary' or 'involuntary'. A voluntary separation is where the employee decides to leave the employer for another job or for inactivity; an involuntary separation is where the employer dismisses the employee or makes them redundant. The LFS identifies people who have left a paid job in the past three months and asks them to give the reason for leaving that job. Some (even those experiencing involuntary redundancy) may move quickly to other employment after separation. The LFS therefore captures information on changes for people whose economic status has not changed.

The total number of separations reported in Spring 2003 was 83,000. Chart A4.2 below shows voluntary and involuntary separation rates. The job separation rate is the number of people aged 16 and above who separated from a paid job in the three months before interview divided by the number of people working then (i.e. those who said they were in employment for more than 3 months plus those who had separated from a paid job in the last 3 months). The chart shows that the number of voluntary separations consistently exceeds involuntary separations.

Chart A4.2: Job Separation Rates for Scotland 1995-2003

There is a seasonal pattern to separations. The spring quarter tends to be lower than the others. In the year to spring 2003 the LFS identified 418,000 people reporting separations, of which 136,000 (33%) were involuntary. There may also be a cyclical pattern, with the number of separations rising when there is less unemployment.

Redundancy measures

Attention is frequently focused on redundancies. These are a type of separation for which ONS publishes quarterly figures. In spring 2003 the LFS estimated the number made redundant in the preceding three months to be 12,000. Chart A4.3 shows redundancy rates for Scotland. The rate is calculated as the number of redundancies in the last 3 months per thousand employee jobs (three months ago). Rates in Scotland tend to be slightly higher than the UK average, but the number appears to be fairly stable. Comparison with numbers in Chart A4.2 show that in Spring 2003 redundancies accounted for only 14 per cent of all separations (46% of involuntary separations).

Scottish Employers Skills Survey

Data on engagements

Chart A4.1 noted the estimate of 429,000 people in a new job (i.e. employed for less than 12 months). However, the number of engagements (taking account of short term moves) will be larger than this. The LFS does not measure the total number of engagements, but does record 418,000 separations. Adjusting for non-response, emigration and deaths produces an estimated total of approximately 440,000 separations. This is possibly still an underestimate as the figures do not include movement from second jobs (of which there were 106,000 in Spring 2003) and could miss migrant seasonal workers.

Chart A4.3: Numbers of redundancies and redundancy rate in Scotland 1995-2003

We can complement this with surveys of employers that provide an estimate of the number of engagements. In the 2003 Scottish Employers Skills Survey ⁸, establishments that had been in existence for over a year reported staff turnover of over 25 per cent. They reported 486,000 recruits and 506,000 leavers. Adding the 95,000 people employed in new establishments takes the total number of recruits to 580,000, equal to 29 per cent of the total number of employee jobs in the economy.

Claimant Count Series

Administrative data for people claiming Job Seekers Allowance (JSA) enables detailed measurement of flows on and off the claimant count. The number of claimants in Scotland (not seasonally adjusted) in March 2003 was 107,200. One year later, in March 2004 the number of claimants was 103,500, a fall of almost 4,000. However the number flowing off the count over this period was 312,700. These off flows were replaced with on-flows of an almost similar number. Only 13,900 of the claimants in March 2004 had been claiming for 12 months or more.

Chart A4.4 below shows in March 2004 the number unemployed, at 103,500 was only 43 per cent of the figure ten years earlier. However, the flow into the count in the year to March 2004 was only 19 per cent lower than 10 years earlier. This indicates that unemployment levels have fallen, in part because fewer people become unemployed but much more because they spend less time unemployed. Median duration on the count has decreased from 12 weeks for people flowing off in March 1997 to 8.9 for off-flows in March 2004.

Chart A4.4: Number of claimants, on-flows and off-flows of Job Seekers Allowance in Scotland 1994-2004

Of those leaving unemployment benefits, 60 per cent leave for a job or because their working hours have increased. This has increased by 10 percentage points since 1994. Other destinations include going onto other benefits or losing entitlement.

Conclusion

We have presented a variety of data, some of it relating to change over twelve months, some to three months and some one month. We have shown that while the overall employment level increased by 50,000 between Spring 2002 and Spring 2003, the number of people entering jobs was nearly ten times that number, and that inflow was matched by a similar number of job separations. Similarly a 4,000 fall in the claimant count between March 2003 and March 2004 was associated with flows off the count of over 300,000.

We conclude by trying to draw the strands together to present a stylised picture of flows in the labour market in Scotland.

Each week

• employers take on almost 9,000 people;
• at the same time about 9,000 people leave their job - around half of them for another job;
• about 6,000 people join the claimant count; and
• approximately 6,000 people leave the claimant count, around half of them going into employment.

Additional Sources

The use of the LFS and Claimant Count Series to examine changes in status over the course of one year or three months enables estimates of labour market flows to be made at a Scotland level.

Other sources have been developed to look at changes over a longer period are, for example, the British Household Panel Survey (BHPS), which was boosted in 1999 to provide additional information for Scotland. The Department for Work and Pensions (DWP) announced to Parliament, in December 2003, the development of the Work and Pensions Longitudinal Study (WPLS). The WPLS links benefit and programme information held by DWP with employment records held by the Inland Revenue. The WPLS is a 100% administrative database and will include all individuals who come in contact with DWP benefits and programmes. This will allow analysis of flows to and from all benefits and programmes. Strict safeguards will be put in place to ensure responsible use of the WPLS database. More information on the WPLS can be found at: http://www.dwp.gov.uk/asd/longitudinal-study/ic-longitudinal-study.asp