“Firms exports, innovation,… and regions”
E. López-Bazo (UB & AQR-IREA) & E. Motellón (UOC & AQR-IREA)
INTRODUCTION
(New) Trade Literature
Regional Economics Literature
• Firm heterogeneity is crucial for explaining export activity.
•Disparities in volume of exports (absolute and relative)
• Special attention to the role played by innovation: improvement in productive processes
and/or in products increases sales in foreign markets.
•Differences across regions in firm’s characteristics (beyond industrial mix and size)
•Regional heterogeneity are likely to affect cost of exporting, and self-selection into export
80%
80%
80%
70%
70%
70%
Extensive margin exports
Extensive margin exports
• All the literature that have analysed the link between innovation and firm exports so far
have neglected the role of space.
Extensive Margin of Exports
Intensive Margin of Exports
50%
Product Innovation
Process Innovation
45%
40%
60%
“Firms exports, innovation,… and regions”
60%
50%
40%
30%
20%
10%
60%
35%
50%
50%
30%
40%
40%
25%
Enrique López-Bazo (UB&AQR-IREA)
Elisabet Motellón (UOC&AQR-IREA)
20%
30%
30%
15%
20%
20%
10%
10%
0%
10%
5%
0%
0%
0%
0%
10%
20%
30%
40%
50%
0%
10%
Product Innovation
20%
30%
40%
50%
Process Innovation
Descriptive Analysis (Spain)
HYPOTHESIS
Database and Variables (firm-level data)
• Innovation in Companies Survey (ICS), produced
by INE (waves: 2005 and 2009)
• Representative at regional level
• Information on:
 extensive and intensive margin of exports
 technological
and
non-technological
innovation, productivity, human capital, …
• Large regional disparities in export activity, more
pronounced in extensive margin of exports.
Firm’s heterogeneity explains a big deal of
regional disparities in export performance
(extensive and intensive margins). In
particular, there is a key role of differences
across regions in propensity to innovate.
• Disparities in innovation propensity as well
• Regions with high (low) innovation tend to be
those with high (low) extensive margin of exports
• Ambiguous relationship between innovation and
intensive margin of exports
Regional impact of innovation on
the INTENSIVE MARGIN of Export
Regional impact of Innovation on
the EXTENSIVE MARGIN of Export
Methodology: BIPROBIT MODEL controlling for endogeneity of the measures of innovation.
Methodology: IV to control for endogeneity of the measures of innovation.
Results (only for regions with extreme values)
Results (only for regions with extreme values)
Table 1. Marginal effects for the EXTENSIVE MARGIN of exports in Spanish regions
Table 2. Effects from the 2SLS estimates for the intensive margin of exports in Spanish regions.
Spain
Andalusia
Catalonia
Extremadura
Basque Country
Only
Product
0.342***
0.279***
0.354***
0.307*
0.437***
Only
I nnovation
Process (Prod/Proc)
0.327***
0.299***
0.244***
0.220***
0.352***
0.319***
0.308**
0.320***
0.370***
0.369***
Firms in regions with high extensive
margin of exports tended to innovate
more than firms in regions with a
lower margin and, in addition, they
obtained a higher payoff from
innovation in terms of exports status.
*** p<0.01, ** p<0.05, * p<0.1. Marginal effects obtained as the average from the ones for
each firm in the sample. The specification used to compute the marginal effects includes
controls for productivity, size, firm group and industry dummies.
Product & Process
I nnovation
Product
Process
Spain
0.180
-0.075
Andalusia
0.978
-0.778
Catalonia
-0.058
0.249
Extremadura
-0.137
0.118
Basque Country
0.022
0.066
Only
Product
Only
Process
I nnovation
(Prod/Proc)
0.110***
0.152*
0.158***
-0.051
0.090
0.111***
0.109
0.186***
0.054
0.085
0.095***
0.091
0.164***
0.026
0.071
The effect of innovation on
the share of sales abroad is
rather limited and even
negligible in most Spanish
regions.
*** p<0.01, ** p<0.05, * p<0.1. The specification used to compute the marginal effects includes controls for
productivity, size, firm group and industry dummies.
COUNTERFACTUAL regional export margins under different scenarios
Empirical specification
1.
Assigning to each region a common propensity to innovate  Scenarios Table 3
2.
… modifying not only innovation but also firms’ labour productivity  Scenarios Table 4
The benchmark is Catalonia (region with highest share of innovative firm)
Results (only for regions with extreme values)
Table 3. Changes in the margins of exports from COUNTERFACTUAL INNOVATION.
Change in Extensive M argin
I nnovation
Product Process (Prod/Proc)
Spain
4.33
0.85
3.89
Andalusia
5.61
0.37
4.14
Catalonia
0.15
0.24
0.47
Extremadura
9.27
6.97
10.64
Basque Country
3.47
0.56
2.27
Change in I ntensive M argin
Product
&
I nnovation
Process Product Process (Prod/Proc)
0.79
0.70
0.53
0.56
6.83
2.11
0.95
1.26
0.00
0.00
0.00
0.00
-2.22
-1.30
0.60
0.54
0.35
0.58
0.27
0.27
Table 4. Changes in the margins of exports from COUNTERFACTUAL INNOVATION AND PRODUCTIVITY.
The
change
in
the
extensive margin is far
more pronounced than
the change caused in the
intensive margin. This is
so for the entire country
and for most regions.
Spain
Andalusia
Catalonia
Extremadura
Basque Country
Change in Extensive M argin
Change in I ntensive M argin
I nnovation
Product Process (Prod/Proc)
12.69
10.16
12.06
18.75
12.80
16.19
6.04
7.61
6.30
Product
&
I nnovation
Process Product Process (Prod/Proc)
1.01
0.92
0.77
0.78
8.57
2.49
1.20
1.53
0.06
0.14
0.08
0.12
22.33
18.05
22.65
-1.06
0.28
2.28
2.31
8.23
8.41
7.84
1.59
1.70
1.55
1.52
The change in the propensity
to innovate and in the level of
productivity
would
substantially
encourage
firm’s decision to export in all
regions, but would have an
almost negligible effect on
the share of sales abroad
made by exporting firms.
Note: Change in percentage points with respect to the actual extensive and intensive margins in each region. Figures on the share of innovative firms in Catalonia were used as a benchmark. Results obtained from the bivariate probit (extensive margin) and the 2SLS (intensive margin) estimates.
CONCLUDING COMMENTS
• This study provides evidence on the effect of firm innovation on
export performance from a regional viewpoint. As a novel contribution,
the effect is allowed to vary across regions.
• In addition to the effect of regional differences in innovation,
disparities in firm’s productivity are crucial when explaining
geographical gaps in the extensive margin of exports.
• Innovative firms are more prone to export than otherwise similar noninnovative firms in all Spanish regions. However, the effect of
innovation is far from regionally uniform.
• Direct policies aiming at promoting exports in each region must
focus on the group of innovative firms not exporting yet. They are
candidates to become exporters if the locational disadvantages
are compensated in some way.
López-Bazo, E. and E. Motellón (2013): “Firm exports, innovation,… and regions”, WP IREA 2013/08. (http://www.ub.edu/irea/working_papers/2013/201308.pdf )
Contact me for any suggestion or comment: [email protected]