APPENDIX 8 (continued)
The UoD data used in this paper is taken from
the government's DfEE website http://www.dfee.gov.uk/ndimprove/
Core performance tables relating to the New Deal. The data refers
to all 144 Units of Delivery in the England, Wales and Scotland.
Our information begins with the sample period before the introduction
of New Deal ie. Q4 in 1997 is t=0. The time period is then indexed
from the beginning of the Pathfinder operation in January 1998,
i.e. our t=1. The second quarter of the data relates to the cohort
who began the New Deal between April and June 1998 at time t=2.
Each subsequent quarter's data relates to the cohort who began
New Deal in succeeding quarters, so for example the t=3 cohort
began New Deal between July and September 1998. The available
Core Performance data goes up to period t=5 which relates to the
period April to June 1999.
The website provides details of the numbers
in each of the New Deal options as well as the performance of
these New Deal recipients in the sense of the number who leave
to unsubsidised jobs, the number who leave to subsidised jobs,
and the number who do not return to JSA within 13 weeks. In addition
the website provides useful covariate information which constitute
the characteristics of the UoD locality and the New Deal population
in that area. There are details of : the proportion who are men,
the proportion disabled, the proportion attaining NVQ level 2
and the total unit cost by UoD. In addition extra information
is provided include whether or not the DfEE regards the area as
a "tight" labour market or not.
The data from the DfEE website was supplemented
from NOMIS to provide unemployment numbers broken down by age
and duration for each UoD. This enables us to calculate: exactly
how many other people are unemployed in the UoD apart from those
on New Deal; the proportion of unemployed who are between the
ages of 18-24, and, the proportion of the unemployed who are long
term (over 6 months duration). Unfortunately the vacancy level
is not directly available for each UoD. Hence the best match was
sought from existing NOMIS vacancy data by county, employment
office, ward and other available geographies. This resulted in
a close match for 114 out of the 144 UoD locations. Many of the
missing locations are rural areas in Scotland or Wales with spare
populations and very low numbers of New Deal. Full details are
provide in the Data Appendix of this paper.
| Variable definition|
|Number of Unsubsidised Jobs||177.09 (104.96)
||Number of New Deal participants moving into Unsubsidised Jobs. From Core Performance Table A.
|Number of Subsidised Jobs||27.38 (27.31)
||Number of New Deal Participants moving into subsidised Jobs. From Core Performance Table B.
|Number not returning to JSA within 13 weeks
||136.46 (104.96)||Number of New Deal participants not returning to JSA from jobs for 13 weeks. From Core Performance Table D.
|New Deal Cohort||431.67 (533.12)
||The number of people entering the New Deal cohort in quarter t.
|Unit Cost||2,751.79 (837.28)
||The per head unit cost of a New Deal person in the UoD for the intake. From Core Performance Table C.
|Percentage Men||72.41 (5.71)
||The proportion of New Deal cohort who are men. Reported in Core Performance Table E.
|Percentage Ethnic Minority||87.19 (17.80)
||The proportion of New Deal cohort who are not from any ethnic minority. Reported in Core Performance Table E.
|Percentage Disabled||14.40 (5.35)
||The proportion of New Deal cohort declaring a disability. Reported in Core Performance Table E.
|Percentage Employment Option||7.69 (4.88)
||The proportion of New Deal cohort who have been in the Employment Option. Calculated from Core Performance Tables A and I.
|Percentage FTET Option||16.65 (6.52)
||The proportion of New Deal cohort who have been in the Full Time Education and Training Option. Calculated from Core Performance Tables A and I.
|Percentage Voluntary Option||14.53 (10.44)
||The proportion of New Deal cohort who have been in the Voluntary Option. Calculated from Core Performance Tables A and I.
|Tight Labour Market||0.49 (0.50)
||Dummy variable which takes the value 1 if the UoD labour market is "tight" in terms of low unemployment. This is clusters A, C or E in the Performance Tables.
|London||0.97||Dummy variable taking the value 1 for all UoD in the Greater London area and 0 otherwise. Note that London is one TTWA and therefore has the same values on various labour market variables like earnings.
||Dummy variable taking the value 1 if the UoD was a Pathfinder area and 0 otherwise. The pathfinder areas are: Black Country, Cornwall, Cumbria, Eastbourne and Hastings, Harlow, Lambeth, Newcastle and Gateshead and South Tyneside, Rotherham, Southern Derbyshire, Swansea, Tayside, West Wales, and Wirral.
|Unemployment UoD||8,660.87 (6,230.57)
||The total number of unemployed by UoD by quarter. This data is taken directly from NOMIS data for each quarter of the panel.
|Vacancies UoD||6,256.90 (9,718)
||The total number of vacancies by UoD by quarter. This data is derived from NOMIS data for each quarter of the panel. Note that NOMIS does not as yet provide vacancy data by UoD. Hence this date had to be constructed for the nearest match on UoD constructed from other county, city, ward and LEA geographical data. (See Vacancy data note below.)
|Percentage Long Term Unemployed||40.76 (7.55)
||The proportion of those unemployed who have been unemployed for more than six months in their present spell. This data is derived from NOMIS data for each quarter of the panel.
|Percentage Young Unemployed||25.05 (3.97)
||The proportion of those unemployed who are aged 18-24. This data is derived from NOMIS data for each quarter of the panel.
|Average Earnings||375.53 (59.72)
||Average Gross weekly full time earnings, April 1998 (pounds) in nearest match to UoD. From Regional Trends 1999 Tables 14.5, 15.5 and 16.5. This variable is not time varying.
Note that NOMIS does not as yet provide vacancy data by UoD.
Hence this data had to be constructed for the nearest match on
UoD constructed from other county, city, ward and LEA geographical
data. In the present regressions the UoD there are 30 UoD areas
for which no exact match is found in NOMIS data. They are: Channel,
Ceredigon, Chatham, Croydon and Bromley, East Lancashire, Harlow,
Havering Barking and Dagenham, Hereford and Worcester, Knowsley,
Moray Strathsprey, North Devon, North Essex, North Lancashire,
North West Wales, Powys, Ross and Cromarty, Sefton, Solihull,
South Essex, Stockport, Sutton Merton and Esher, Kingston, Epsom,
West Lancashire, West Lothian, West Norfolk.
Peter Dolton and Yvonne Balfour
University of Newcastle upon Tyne, Department of Economics