Table A4 Variable definitions and... | Emerald Publishing

Table A4

Variable definitions and descriptions

Variable	Description	Calculation
Herfindahl index	The Herfindahl index measures concentration in a particular industry. A higher value indicates higher industrial concentration in a region, while a lower value indicates a lower level of industrial concentration in the region (van Egeraat et al., 2018). The index of ${H H I}_{i}$ measures the degree of industrial concentration. Using standard measurements consistent with Power et al. (2019) and Antonietti and Cainelli (2011) a measurement for sectoral concentration within regions is developed	${H H I}_{i, 2017} = {\sum_{j}^{N} (\frac{y_{i j}}{y_{i}})}^{2}$ Where HHI_i is the Herfindahl index for region i; y_ij is the number of firms in region i in NACE four-digit industry j; and y_i is the number of firms in region i
The location quotient (LQ) variable	The LQ variable compares the concentration of employment in a sector in a region with the concentration of the same sector nationally	${L Q}_{s, r} = \frac{E_{s, r} / E_{r}}{E_{s, n} / E_{n}}$ Where ${L Q}_{s, r}$ is the concentration of firms in a sector s and region r and E is the share of firms in sector s and region r while $F_{n}$ is the firms nationally
Related variety (RV) and unrelated variety (UV)	RV and UV are variables which proxy levels of industrial diversification within regions. Higher values of these indices indicate higher levels of unrelated variety or higher levels of related variety. These calculations of these variables are consistent with those used by Frenken et al. (2007)	${R V}_{i} = \sum_{j} P_{j i} (\sum_{h \in (j i)} \frac{P_{h i}}{P_{j i}} \log_{2} (\frac{1}{P_{h i} / P_{j i}}))$ ${U V}_{i} = \sum_{j} P_{j i} \log_{2} (\frac{1}{P_{j i}})$ Where ${R V}_{i}$ is the measure for related variety and ${U V}_{i}$ is the measure for unrelated variety. The 4-digit NACE classification h falls exclusively falls under the 1-digit NACE classification j in region i. Where $P_{j i}$ is the number of firms in region i and $P_{h i}$ is total number of firms in all regions in industry j. And where $P_{j i} = \sum_{h \in (j i)} P_{h i}$ indicates the 1-digit shares
Population Density	The density is measured via persons per squared km	Persons per km²
Log of Average Firm Size	Log of average firm size acts as a control variable for the scale of businesses within the sub-county regions	The natural log of average firm size within sub-county regions
Log of Income per person	Log of income per person acts as an economic control variable	The natural log of income per person within sub-county regions

Variable	Description	Calculation
Herfindahl index	The Herfindahl index measures concentration in a particular industry. A higher value indicates higher industrial concentration in a region, while a lower value indicates a lower level of industrial concentration in the region (van Egeraat et al., 2018). The index of ${H H I}_{i}$ measures the degree of industrial concentration. Using standard measurements consistent with Power et al. (2019) and Antonietti and Cainelli (2011) a measurement for sectoral concentration within regions is developed	${H H I}_{i, 2017} = {\sum_{j}^{N} (\frac{y_{i j}}{y_{i}})}^{2}$ Where HHI_i is the Herfindahl index for region i; y_ij is the number of firms in region i in NACE four-digit industry j; and y_i is the number of firms in region i
The location quotient (LQ) variable	The LQ variable compares the concentration of employment in a sector in a region with the concentration of the same sector nationally	${L Q}_{s, r} = \frac{E_{s, r} / E_{r}}{E_{s, n} / E_{n}}$ Where ${L Q}_{s, r}$ is the concentration of firms in a sector s and region r and E is the share of firms in sector s and region r while $F_{n}$ is the firms nationally
Related variety (RV) and unrelated variety (UV)	RV and UV are variables which proxy levels of industrial diversification within regions. Higher values of these indices indicate higher levels of unrelated variety or higher levels of related variety. These calculations of these variables are consistent with those used by Frenken et al. (2007)	${R V}_{i} = \sum_{j} P_{j i} (\sum_{h \in (j i)} \frac{P_{h i}}{P_{j i}} \log_{2} (\frac{1}{P_{h i} / P_{j i}}))$ ${U V}_{i} = \sum_{j} P_{j i} \log_{2} (\frac{1}{P_{j i}})$ Where ${R V}_{i}$ is the measure for related variety and ${U V}_{i}$ is the measure for unrelated variety. The 4-digit NACE classification h falls exclusively falls under the 1-digit NACE classification j in region i. Where $P_{j i}$ is the number of firms in region i and $P_{h i}$ is total number of firms in all regions in industry j. And where $P_{j i} = \sum_{h \in (j i)} P_{h i}$ indicates the 1-digit shares
Population Density	The density is measured via persons per squared km	Persons per km²
Log of Average Firm Size	Log of average firm size acts as a control variable for the scale of businesses within the sub-county regions	The natural log of average firm size within sub-county regions
Log of Income per person	Log of income per person acts as an economic control variable	The natural log of income per person within sub-county regions

Source(s): Author's own work