Sectoral Energy- and Labour-Productivity Convergence

Iron and Steel and Machinery, allowing for individual country-effects in explaining labour-
productivity growth yields statistically less significant or even insignificant estimates of β.

This result suggests that in terms of labour productivity the variation in explanatory
variables over time is relatively small as compared to cross-country differences, since
correcting for the latter by means of including country-specific intercepts results in weaker
evidence of a negative relationship between the initial labour productivity level and its
growth over time. Nevertheless, the regression results suggest that both energy- and labour-
productivity convergence depend to a large extent on individual country-effects, indicating
energy- and labour productivity to be conditional rather than absolute in virtually all sectors.
The latter is illustrated by the fact that the speed of conditional convergence is substantially
higher than of unconditional convergence: for energy productivity the half life, as it follows
from the implied λ, now lies between 1 year (Transport Equipment) and 14 years (Total) and
for labour productivity it has been reduced to a period in between 47 years (Transport
Equipment) and 77 years (Non-Ferrous Metals).

Of course, this brings back the question as to which are the country-specific variables
driving energy- and labour-productivity growth and, hence, determining the country-specific
steady states? Recall from the introduction that several mechanisms may be at work, causing
‘followers’ to grow faster than ‘leaders’: advanced economies may suffer from diminishing
returns, lagging countries may benefit from knowledge spill-overs, production processes may
convergence due to increasing competition, etcetera.

In order to explain (persistent) differences in cross-country energy- and labour-
productivity growth we replace in equation (5) the unspecified country-effects μ_i by a number
of country-specific explanatory variables x_i^j, according to:

5

g₁t = a + β ln (y )i∙ t _i + ∑ y jxit + ε_it

j=1

The specified explanatory variables are defined at the sectoral level and include:

1 a   {f^,E, t ^{+ p}E, t-1 ^{+ p}E, t - 2 ⁾

^xit  =--------------3-------------

1 b _ (wt ^{+ w}t-1 ^{+ w}t - 2 ^)
xit =          ^

Investment share:     x2 = I-

<<   12   13   14   15   16   17   18   >>

More intriguing information