Trade and Employment From Myths to Facts - International Labour ...

Trade and Employment: From Myths to Facts
ANNEX 6.A: MEASURING ADJUSTMENT IN OUTPUT AND
CHANGES IN INEQUALITY LEVELS
6.A.1 Measuring adjustment in terms of output
The metrics presented in equations (1) to (4) in the main text to capture adjustment
in labour markets can also be developed to capture adjustment in terms of output.
The starting point is again information on sectoral-level output generated, for
example, with CGE simulations. We will refer to changes in sectoral output as
where the subscript j represents sectors. Total change in output resulting from
trade reform in an economy with n sectors can then be computed using:
250
(Q1)
Where λ j represents the weight of a sector j in the total economy. As in the case
of the employment measure in equation (1), the measure is only of limited use to
reflect the extent of adjustment processes, as positive and negative output changes will
cancel out. It is therefore preferable to use a variance-based measure of the type:
(Q2)
Taking the square root of provides a useful measure of variation across
sectors. This measure can be calculated in a rather straightforward way with standard
CGE models.
An important weakness of this measure is that it does not capture output shifts
across firms within the same sector. For a given change of output within a sector, Q2
would signal the same extent of adjustment if that change entails proportional shifts
in output across firms or if it entails company failures and creation of new companies.
Yet, adjustment of the second type is likely to be more costly for an economy, in
particular if growing and shrinking firms are located in different regions. The importance
of firm-level adjustments has been emphasized in recent trade literature. A more
appropriate measure would therefore take into account within-sector adjustment:
where and weight of firm i sector j
,
(Q3)
A measure for total adjustment, capturing all changes in output, those within
and across sectors, would look as follows:
(Q4)