204
Tạp chí
Kinh tế - Kỹ thuật
that have demonstrated a stronger regard for the environment (Sarkiset al., 2010). Firms’ decisions
on their operation location at the regional level provide the information for explaining the social
pressure experienced by a company (González-Benito and González-Benito, 2010). Responding
to community stakeholders’ demand can improve or maintain relations with the community
stakeholders (Raines, 2002; Delmas and Toffel, 2004; Zhanget al.,2008). Furthermore, community
stakeholders’ power on organizational environmental management practices may vary depending on
the community’s characteristics (Kassinis and Vafeas, 2006). Sometimes the location of production
facilities may not produce significant environmental pressure as perceived by industrial companies
(González-Benito and González-Benito, 2010). Compared with residents in poor regions, residents
in wealthy regions may pay more attention to the non-economic aspects of their living conditions
(Frooman, 1999; Pfeffer and Salancik, 2003). However, those residents located in poor regions lack
the power to influence corporate policies (Kassinis and Vafeas, 2006). Therefore, residents located
in wealthy regions may demand firms to adopt green product as a commitment to the environment
(Perkins and Neumayer, 2010). We hypothesize that:
Hypothesis 5a:The higher the income in a geographic community, the higher adoption rate of
horizontal green product diversification of firms will be in that community.
Hypothesis 5b:The higher the income in a geographic community, the higher adoption rate of
vertical green product diversification of firms will be in that community.
3. METHODOLOGY
3.1. Data
The dataset was based on a survey undertaken by the General Statistics Office (GSO) and
the Central Institute for Economic Management (CIEM). This was a survey on the the Vietnam
Technology and Competitiveness that has been conducted every year since 2010, concluding with
2014. The General Statistics Office (GSO) conducted the survey of 2600 firms in 10 provinces.
Though the choice of the 10 provinces was based on the principle of representation rather than
on a random basis, they constitute a reasonable representation of Vietnam in terms of geographic
locations, economic, social, and legal indicators. The survey mainly included manufacturing firms
and accounting information from 2010 to 2014. Based on the previous study , the observations
were eliminated if the firms concerned had already adopted ISO 14001or eco-label in the previous
year. We have added information of regional environmental regulations and regional GDP per
capita during this period based on the Vietnam Statistics Yearbook and regional environmental
management indicators computed by the Vietnamese Academy of Science, and the final data used
for analysis in our study is composed of 8156 observations. The questionnaire includes questions
on the state of certification of ISO 14001 or eco-label. If firms passed the certification of ISO 14001
or eco-label, they were further asked the year that they had obtained the certification. Once a firm
passed the ISO 14001 certification or the eco-label certification, we dropped the observations of
the firm after these years.
3.2. Measurement
3.2.1 Dependent variables
Green product diversification is, for the purpose of this paper, defined at the enterprise level
as a production of more than one 4-digit ISIC product whit the greening certification status.
We consider an enterprise to be diversifying if it started producing a new product (at the 4-digit
ISIC level) during the past tree years (denoted vertical diversification) or if it made significant
improvements of existing products (denoted horizontal diversification). Thus, certification status of
