More precisely, the increase in d increases the intensity of multi-tasking and
the output level immediately after the shock, and this effect persists also in the
long run. The fraction of the workforce in the human resources service, the level
of human capital and the allocation of time in favour of production, on the other
hand, increase immediately after the shock on d, but then return to their initial
levels since this shock does not affect their long run values. Intuitively, an increase
in d corresponds to a reduction in coordination costs and represents therefore the
same kind of qualitative productivity shock as a rise in the technological parameter
A.
The last situation analyzed is an increase in the efficiency of human capital E .
In this case, we have shown analytically that a rise in E has a long run effect on
the number of tasks performed per worker, on the level of output and on the level
of human capital (that all increase).
In the short run (Figures 2.1 - 2.4), the timing of the effects of the shock on
the different variables is different from the previous cases. In effect, the number of
tasks per worker n gradually increases after the rise in E and reaches, in the long
run, a level higher than the initial one (while in the other cases examined there is
immediately the jump in the variable, that then remains at the new value). Following
the rise in E , output also increases gradually reaching the new long run level, and
the same happens for the level of human capital. On the contrary, the fraction of
the workforce in the human resources service and the allocation of time in favour
of productive activity initially increase but then return to their initial levels, since
their long run values are not affected by the shock on E .
22