relative employment growth) and productivity growth, the former representing a social
problem and the latter explaining its causes and the future prospects of the local economy”.
Following their methodological approach, we define relative employment growth (REG) of
region r as:
E02 E98
Errr Er /T-
r = EO ∕e'98
And, in the same way, relative productivity growth (RPG):
P 02 P98
RPG = r I г
Gr P 02 Ip 98
nn
Where n denotes national average variable, and E and P are employment and productivity,
respectively.
Adapting Camagni & Capello (1999) methodology to the Spanish case for the period we are
considering, four possible patterns of regional evolution emerge from the combination of
REG and RPG9:
a. Virtuous cycle, when a higher than average productivity growth generates good
performance in both employment and output;
b. Restructuring, when higher productivity growth is reached through a less than average
increase in employment10 accompanied by good output performance;
c. Industrial take-off, when regions grow higher than average in employment terms, but they
are not able to reach national productivity evolution.
d. Deindustrialisation11, when lower than average increase in employment is not able to
compensate the poor evolution of output, leading to losses in productivity.
Then, the economic performance of each region can be newly charted on a cartesian graph, as
it is done in Figure 6.
9 There is no regions in the deindustrialisation group defined by Camagni & Capello (1999). No one of the
Spanish regions experienced an increase in employment and productivity under average at the same time during
the period.
10 Remember that there is not decreases in employment in the period, nor in output.
11 The definition is different from Camagni & Capello’s, as it can be seen in note 10.