1. Introduction
The purpose of the paper is to examine what the main factors of recycling municipal waste at the European level are, using a linear regression model in order to discover and analyze the relationship between several economic factors and the recycling rate of municipal waste (RRMW). The question of the research is what is the entrepreneurial innovation impact on the RRMW?
Previous studies proved several impacts and answered in different modes to this specific question. Still, the study intended to demonstrate the impact of entrepreneurial innovation interpreted as business expenditure on R&D, resource productivity, and environmental taxes on the RRMW.
The motivation for this approach is given by the fact that entrepreneurial innovation is no longer a desiderate, but rather a constant key activity of our society. Both public policy and business environment have an interest in all of the innovative solutions for the improvement of their activity, as well as in the positive impact of the innovation on the return of business. Constant improvement of the activity is due to the implementation of the innovation results.
Entrepreneurship is being recognized as the main solution for economic development and social welfare [
1,
2]. Still, the economic growth may be determined by many other responsible factors, and some of them are quite unexpected [
3], especially those in the Asian development model. Nevertheless, entrepreneurial manifestation has the government’s or local public authorities’ support [
4], and business environments sustenance, too. Starting with Schumpeter’s complex definition [
5] on entrepreneurship, there were many conceptual approaches regarding the role of an entrepreneur, from the person who bears the uncertainty, to that person who allocates resources with multiple uses [
6]. Therefore, entrepreneurship seems to be the foundation of recycling activities even in the municipal waste field.
The link between innovation and entrepreneurship [
7] crossed from the historical perspective to economic growth and innovation entrepreneurship, with two directions: innovation in product-market and technology innovation. Precursor studies already have demonstrated that innovation and entrepreneurship are positively related to each other [
8] and together will have an impact on economic growth. Plus, the question still remains: Do they have an impact on the RRMW?
Innovation nowadays becomes important in waste management, too, as it is needed in order to resolve the issue of the huge annual quantities of municipal waste all over the world and to maintain urban and rural areas as clear as possible. Taking into consideration that current environment policy is based on sustainability [
9], waste management has at least three dimensions: economic, ecological, and social [
10], especially in highly-urbanized centers. From this perspective, it is very important to see what would be the solutions for better recycling the municipal waste. The municipalities and public policy makers try to identify which would be the most effective solution [
11,
12] for urban waste, no matter the type—domestic or waste from municipal services, solid, e-waste, dust, food, or any debris from cleaning activities. Obviously, there are studies that demonstrate an important difference between urban and rural areas in managing waste [
13,
14], while taking into consideration the higher number of solutions in waste management for rural households, legal or less legal, such as burning, composing, using them as fertilizers for vegetable cultivation, etc.
In this regard, innovation was used and, for example, even green channels were proposed for e-waste [
15] in order for the countries to reach green economy principles. The conclusions of different other studies demonstrate that even e-waste management is not as developed and as well prepared as the high-tech industry is and produces [
16].
The approach might be slightly different and more effective for the food industry [
17], as the innovation brought to light very effective recycling solutions to this sector. This is the place where entrepreneurs may intervene in an innovative way [
18] and boost the link between entrepreneurship and innovation in order to generate better and more functional recycling solutions.
Therefore, conceptual and empirical studies were run in order to better demonstrate how much entrepreneurship determines economic growth [
19,
20], concludes towards a circular economy [
21], boosts innovation or vice-versa [
22,
23,
24], and even show the link between these three concepts: innovation, entrepreneurship, and economic growth [
25]. Moreover, empirical studies were run in order to find out innovative solutions for smart partnerships [
26] and sustainable development [
27]. Considering all these issues, in order to understand the impact of entrepreneurial innovation on the RRMW, it is important to determine what the business expenditure on R&D and the GDP expenditures on R&D (as main supportive factors for innovation) are, then private investments in general as key factors for entrepreneurship and recycling rates.
The connection between entrepreneurial innovation and recycling municipal waste has been studied by many economists. It was proved that there is a close entrepreneurial link between innovation and recycling of municipal waste [
28]. Other authors [
29,
30,
31] concluded that business expenditure on R&D have a positive impact on the RRMW.
Moreover, while some researchers [
32,
33] underline that private investments have a significant impact on the RRMW, other scientists [
34,
35] argue that GDP expenditures on R&D by the business enterprise sector have a positive impact on the RRMW in EU member states. Nevertheless, da Cruz et al. [
36] and Busu [
37] conclude that the environmental taxes and productivity of the resources have a direct and significant impact on the RRMW. Resource productivity is the gross domestic product (GDP) divided by domestic material consumption (DMC), where DMC measures the total amount of materials directly used by an economy. It is defined as the annual quantity of raw materials extracted from the domestic territory of the focal economy, plus all physical imports minus all physical exports.
Starting from the above-mentioned empirical results, we state our research question: “What is the entrepreneurial innovation impact on the RRMW?” Besides what is already known in this area, we will try to make an estimation on which of the five independent factors (i.e., business expenditure on R&D, private investments, GDP expenditures on R&D, resource productivity, and environmental taxes) has the greatest impact on the dependent variable of the quantitative model.
Nowadays, EU has 28 member states which joined the Union at different times. In the past 20 years, there were three moments when new countries joined EU: In 2004, when 10 new member states joined EU; in 2007, two new states adhered to EU; and in 2013, when a new state joined EU. Due to data availability, our analysis covers the period between 2010 and 2017.
This paper has the following structure. Firstly, we make a descriptive analysis of the macroeconomic key indicators of the RRMW at the EU level. RRMW measures the share of recycled municipal waste in the total municipal waste generation and the ratio is expressed in percentage points. Secondly, the relationship between entrepreneurial innovation and RRMW is analyzed. Research hypotheses are formulated and then tested through the regression analysis. Further research, limitations, and conclusions are summarized in the last section of the article.
3. Results
For the quantitative analysis, RRMW was set as the endogenous variable (Y), determined by five exogenous variables. They were: Business expenditures on R&D (X1); private investments, jobs, and the added value related to the recycling sector (X2); GDP expenditure on R&D (GERD) by the business enterprise sector (X3); resource productivity (X4); and environmental taxes (X5). Linear regression analysis was made through the following steps: performing the quantitative analysis, estimating the model parameters, and checking the results.
A description of the statistical indicators used in our study (min, max, median, mean, and standard deviation) is given in
Table 2. The values of median and mean are useful indicators of how close the data is to normal distribution. If the median and the mean approximate each other, we could assume that the data has a normal distribution [
40].
In
Table 2 we can see that the median and mean values are close to each other; therefore, we could conclude that the variables in our model are normally distributed.
The multicollinearity test among the independent variables (X
1, X
2, X
3, X
4, and X
5) used in our model was performed by the Pearson correlation analysis. In
Table 3 we can see the values of the pairwise correlation coefficients. Since these values are smaller than ±0.30, we could assume that there are no multicollinearity issues among the exogenous variables [
41].
Now we will perform the Lagrange multiplier (LM) Breusch–Pagan test and the F-test to determine whether the research model we used in our analysis, given by Equation (1), was pooled data, fixed effects, or random effects.
The F-test was used for testing the validity of the pooled model against the fixed effects model [
42]. In order to perform this test, we will consider the unrestricted and restricted models.
- (i)
- (ii)
Unrestricted model: .
The fixed effects estimator or within estimator of the slope coefficient β estimates the within model by ordinary least square (OLS) analysis:
The null and alternative hypotheses are:
If the null hypothesis is accepted, then the restricted model is accepted. Otherwise, the fixed effect model would be suitable for our analysis.
The F-test results can be seen in
Table 4.
Since the probability p-value (probability = 0.190) is greater than the 0.05 threshold, we will accept the Null Hypothesis and conclude that random effect model should be used in our study.
Now, in order to make a choice between random models and pooled data we will use the LM Breusch–Pagan test [
43].
By means of this test, we analyzed the existence of the kth order autocorrelation of residual values. We assumed that the errors regression model is given by the following equation:
In order to assess the presence of the kth order autocorrelation, we tested the following null and alterative hypothesis:
H0: ;
H1: ;
If the null hypothesis is accepted, then the pooled model will be suitable for our analysis. The results of the LM Breusch–Pagan test can be seen in
Table 5.
Upon analyzing the results of the random effect test in
Table 5, we accepted the null hypothesis, since the probability
p-value (probability = 0.091) was greater than the threshold 0.05. Hence, we concluded that the pooling of the model in Equation (1) was suitable for our analysis.
The five statistical hypotheses formulated in the previous section were tested with a multiple regression equation using the pooled least square (PLS) method. We used this approach in order to analyze the impact of the entrepreneurial innovation on the RRMW at the EU level between 2010 and 2017.
The evolution of the RRMW between 2010 and 2017 in EU countries was analyzed with a regression model and we obtained the following results (see
Table 6):
According to the table above, the regression equation is:
where:
Y = recycling rate of municipal waste;
X1 = business expenditures on R&D;
X2 = private investments, jobs, and gross added value related to the recycling sector;
X3 = GDP on R&D by the business enterprise sector;
X4 = resource productivity;
X5 = environmental taxes.
Therefore, we could conclude that all independent variables in our model had a significant impact on the dependent variable and, hence, all five hypotheses were validated.
4. Discussion
In this section we discuss the results of the multiple linear regression analyzed by the PLS method. The method was used by the authors to estimate the impact of entrepreneurial innovation on recycling municipal waste.
Analyzing the RRMW evolution in the 27 EU countries, from 2010 to 2017 through independent variables (business expenditures on R&D, private investments, jobs, and value added related to recycling sector, GDP on R&D by business enterprise sector, resource productivity, and environmental taxes), through the multifactorial linear regression analysis, we obtained the following equation (see
Table 6): Y = −2.237 + 0.658X
1 + 0.764X
2 + 0.385X
3 + 0.276X
4 − 0.187X
5, with the standard error coefficients (1.235), (1.414), (1.152), (1.667), and (1.162). Moreover, since the value of R-squared was 0.729, we could conclude that 72.9% of the variability of the endogenous variable is explained by the variability of the exogenous variables. Additionally, the value of the Durbin–Watson statistic (DW = 2.06) is close to 2 and; therefore, we could affirm that the regression errors are not autocorrelated.
The positive coefficients of X
1, X
2, X
3, and X
4 in the PLS model reveal the fact that business expenditures on R&D, private investments, jobs, and added value related to the recycling sector, GDP expenditure on R&D by the business enterprise sector, and resource productivity have a positive impact on the RRMW, while the negative coefficient of X
5 leads to the conclusion that any increase in environmental taxes would lead to a decrease in RRMW. The regulatory framework plays an essential role in modeling RRMW. This explains the fact that an increase of the expenses related to the environmental taxes could be an impediment for potential investors in ecological projects. No matter how peculiar this conclusion may appear at a first glance, there are situations when the increase of the environmental taxes in the field of the waste can lead to the decrease of RRMW. First of all, a unitary environmental tax in the field of waste does not lead directly to the increase of the RRMW, but a differentiated tax system is required depending on the category of waste for which it is applied. Then, there is the entrepreneurs’ perspective: Decreasing the economic facilities of waste generators and increasing their contributions will just demotivate them to get involved in recycling actions. Moreover, the opportunistic behavior of entrepreneurs may be developing, as they will look for other ways to elude the law and to be able to strengthen their profit. Additionally, the
p-values associated with the independent variables in
Table 6 (prob.), compared to the 0.05 threshold, give us the conclusion that all independent variables of the regression model were significant in their relationship with the endogenous variable, the RRMW.
The quantitative analysis concluded that the model was valid and the independent variables were significant for RRMW in all 27 EU countries, since the values of the estimated coefficients of the regression model were statistically significant. The results of the paper confirm recent studies of entrepreneurship innovation for recycling municipal waste [
44,
45].