Explaining exits from unemployment in the UK, Labour Force User Meeting - 2 nd December 2010 Peter Stam Household Labour Market and Developments Branch
Background Previous work Long (2009) Interest and development Objectives Data
LFS – Characteristics of individuals Methodology – Heckman modelling 1.Modelling the probability of being unemployed 2.Modelling the length of unemployment 3.Modelling the probability of finding employment
LFS – Reference group Female Aged 35 through 49 Living in West Midlands Metropolitan County Not classified as an ethnic minority Unmarried with no dependant children Qualified to Below GCSE Renting privately Previous occupation Elementary
Modelling the probability of being unemployed Marginal effect Statistical Significance Age 18 through 244.5*** Age 25 through *** Age 50 through *** Age 60 plus - 4.5*** Male3.2*** Ethnic minority4.6*** Married - 5.5*** Dependent child and female2.2*** GCSE0.9** LFS - Results
Modelling probability of having a spell of unemployment – Housing -5% -4% -3% -2% -1% 0% 1% 2% 3% 4% Rent-freeMortgageOwn homeHousing Association
LFS - Results Modelling the length of unemployment Coefficient Statistical Significance Constant3.7 months** Age 18 through months* Age 50 through months** Male months*** Dependent child and male months*** Job Seekers Allowance months*** Administrative and Secretarial months* Skilled Trades months** Sales and Customer Service months** Process, Plant and Machine months**
LFS – Results Modelling the probability of finding employment Marginal Effects Statistical Significance Male-3.7 * Married9.1 *** Dependent child and male-1.7 * Dependent child and female-10.7 *** Job Seekers Allowance19.2 * GCSE5.3 *** Further Education13.6 *** Degree12.1 ***
LFS - Results -25% -20% -15% -10% -5% 0% 5% 10% 15% Age 18 through 24Age 25 through 34Age 35 through 49Age 50 through 59Age 60 plus Modelling the probability of finding employment – Age
LFS - Results Spell length (months)Marginal effect 6 or less Base
BHPS – Exit destinations Motivation Economic states Data Methodology Inactive ? Employed Unemployed
BHPS - Results Into employment Elapsed Duration (Months) Hazard
BHPS - Results Into inactivity Elapsed Duration (Months) Hazard
Conclusions - Methods Modelling on the LFS provides consistent results with BHPS As duration of unemployment lengthens… so chance of re-employment decreases (LFS / BHPS) As duration of unemployment lengthens… so chance of inactivity increases (BHPS)
Conclusions - Messages Men more likely to become unemployed… but for a shorter time JSA increases the length of time unemployed… but increases the chance of finding employment Education has positive effect on finding employment (peaks at A-level) Housing association tenants more likely to be unemployed Mortgage holders increases chance of finding re-employment Older people have longer spells and are less likely to find re- employment
This work contains statistical data from ONS which is Crown copyright and reproduced with the permission of the controller of HMSO and Queen's Printer for Scotland. The use of the ONS statistical data in this work does not imply the endorsement of the ONS in relation to the interpretation or analysis of the statistical data. This work uses research datasets which may not exactly reproduce National Statistics aggregates. Peter Stam Household Labour Market and Developments Telephone ext