*4.2. Comparison of Green GDP across Countries*

Green GDP comes from and is tightly related to the traditional GDP; however, it serves as an irreplaceable measure for overall economic development different from the traditional indicator, because it takes into account the energy and environmental costs. As indicated in Table 1, we found that in 2014, there were wide gaps for different countries and regions after comparing their standardized GDP per capita with the standardized green GDP per capita. We chose the year 2014 because the data then were the most complete. It is evident that countries with higher green GDP per capita usually have highly developed tertiary industries, which consume few natural resources and do not cause much damage to the environment. For example, in Switzerland, whose green GDP ranked much higher than its traditional GDP (i.e., positive *GAP*), the financial industries, as well as other services, contribute to the vast majority of its GDP, while heavy industries and agriculture account for only a negligible share. By contrast, countries with negative *GAP* are often characterized by imbalanced industrial structures. For instance, the economic growth for nations in the former Soviet Union relies heavily on the energy-consuming and polluting industries, while those oil exporting countries like Iran are dependent on national natural resources. In fact, after comparing the *GAP* of the countries and regions, we learn that countries with outstanding performance in green GDP are not necessarily highly developed in economy and technology. Many newly emerging countries perform quite well with moderate economy scale and stable social development. In this way, whether a nation's economy is green is not merely determined by its power in economics, science, and technology, but relies more on its development pattern. For some nations that achieved industrialization earlier, economic stagnation refrained them from timely industrial restructuring and technological innovation, which, inevitably, set back their green economy development. On the other hand, the latecomers may have learned the lessons and benefited from current technological advances.



**Note**: *GAP* is an indicator to determine whether a nation's economy is green; *GAP* (per capita) equals standardized green GDP per capita minus standardized GDP per capita.

#### *4.3. The Role of Higher Education to Green GDP*

In order to examine the role of higher education, we performed the two-step analysis elaborated in the research design section. At the first stage, we ran the linear regression of Equation (9) based on the per capita data to verify the first hypothesis that higher education has positive influence on building green economies. The dependent variable in the model was *GAP* (the difference between standardized green GDP per capita and standardized GDP per capita), while the explanatory variables were capital per capita, labor per capita, and the gross enrollment rate of higher education. After checking with the Hausman test, we used the country fixed effect model with the panel data, and the regression results are shown in Table 2. The coefficient π<sup>3</sup> demonstrates the relationship between *GAP* and the gross enrollment rate of higher education. It is apparent that the enrollment rate of higher education had a statistically significant positive influence on the outcome *GAP*, indicating our first hypothesis to be valid.


**Table 2.** The linear regression of *GAP* on higher education.

**Note:** t statistics in parentheses; \* *p* < 0.1, \*\* *p* < 0.05, \*\*\* *p* < 0.01.

As for the second hypothesis that green GDP is more responsive to changes in higher education than the traditional GDP, we ran log–log regressions of Equations (10) and (11) separately at the per capita level. The variable higher education was still calculated by the gross enrollment rate of higher education. According to the regression results, a percentage increase in the enrollment rate of higher education can significantly lead to 0.2% of growth in GDP per capita, while green GDP per capita can significantly rise by 0.33% with one percentage increase in the enrollment rate of higher education. Furthermore, the comparison of estimations from the two regression models (suest), combined with the chi-square tests, show that the coefficient of education in the green GDP model was significantly larger than that in the GDP model at the per capita level (Table 3). Considering that the regressions were modified from the Solow growth model, the coefficients can be interpreted in the same way. Therefore, we can conclude that green GDP is more sensitive to changes in higher education than the traditional GDP, which verifies the second hypothesis.

**Table 3.** The log–log regressions of gross domestic product (GDP) and Green GDP on higher education.


**Note:** t statistics in parentheses; \* *p* < 0.1, \*\* *p* < 0.05, \*\*\* *p* < 0.01.
