absorb some of the size effect - see van Biesebroeck (2005, section 5) for evidence on
this. Nevertheless, Girma et. al. (op. cit.) found that the short-run impact of ‘learning-
by-exporting’ on growth was important, although the impact was generally quite
small.
Table 5: Average ‘learning-by-exporting’ effect, UK 1996-2004
Weighted average all
industries a
Weighted average all
industries excl. Retail and
Wholesale trade
|
IV model | ||
|
EXP_entry t+1 |
-0.040 |
0.048 |
|
EXP_entry t |
0.191 |
0.343 |
|
EXP_entry t-1 |
-0.025 |
-0.035 |
|
EXP_exit t+1 |
-0.126 |
-0.085 |
|
EXP_exit t |
-0.053 |
-0.073 |
|
EXP_exit t-1 |
-0.034 |
0.032 |
|
EXP_both t+1 |
-0.066 |
-0.047 |
|
EXP_both t |
0.135 |
0.186 |
|
EXP_both t-1 |
-0.047 |
0.048 |
|
Control function | ||
|
EXP_entry t+1 |
-0.124 |
-0.171 |
|
EXP_entry t |
0.919 |
0.842 |
|
EXP_entry t-1 |
-0.008 |
-0.011 |
|
EXP_exit t+1 |
-0.103 |
-0.077 |
|
EXP_exit t |
-0.374 |
-0.023 |
|
EXP_exit t-1 |
0.020 |
0.027 |
|
EXP_both t+1 |
-0.062 |
-0.035 |
|
EXP_both t |
0.815 |
0.609 |
|
EXP_both t-1 |
-0.037 |
-0.051 |
|
Matched sample | ||
|
EXP_entry t+1 |
-0.057 |
0.050 |
|
EXP_entry t |
0.100 |
0.211 |
|
EXP_entry t-1 |
-0.020 |
-0.027 |
|
EXP_exit t+1 |
-0.124 |
-0.076 |
|
EXP_exit t |
-0.049 |
-0.068 |
|
EXP_exit t-1 |
-0.036 |
0.022 |
|
EXP_both t+1 |
-0.047 |
-0.021 |
|
EXP_both t |
0.133 |
0.149 |
|
EXP_both t-1 |
-0.033 |
-0.018 |
a Average of all estimates in Table 4 that are significant at the 15% or better level (weighted by
industry shares of total real gross output in all industries).
31