Towards Circular Economy: Unveiling Heterogeneous Effects of Government Policy Stringency, Environmentally Related Innovation, and Human Capital within OECD Countries

Abstract

With reference to the existing literature, this paper investigates the heterogenous effect on the attainment of circular economy by government policies in the form of government stringency and government financial support, environmentally related innovations, and human capital. The study was carried out in 26 countries of the OECD from 2010–2019 using the Poisson pseudo-maximum likelihood (PPML) model and data from Eurostat and OECD datasets. Indicators for the independent variables were non-market-based stringency, market-based subsidy, gross domestic expenditure on R&D by source of funds, R&D expenditure intramural, national expenditure on environmental protection environmental protection, environmentally related patents with co-inventors, and employees involved in education and training. The results revealed that a significant effect of government stringency in the form of non-market-based stringency, environmental innovation, government financing on R&D, and national expenditure on environmental protection have significant impact on the attainment of circular economy within OECD countries. Surprisingly, there was no significant effect of the market-based subsidy on domestic material consumption (dmc). We conclude that a blend of government policies is the effective means of achieving a circular economy.

1. Introduction

The OECD member states have experienced heightened growth in their economies coupled with urbanization. This affects the environment and climate change. Insomuch as the modern economy places high stakes on environmental protection, the growing environmental burden evokes pressure of environmental stakeholders especially on firms and policy makers. However, the linear economy concept of ‘take-make-dispose’ cannot be relied on to protect the environment against the reduction in, and the reuse and recycling of, the circular economy. The linear economy tenets mainly include the exploitation of resources from the natural endowment of the environment without recourse to by product and the scarce nature of the natural endowment. Therefore, we can simply say that linear economy poses a great danger to environmental sustainability [1,2,3]. By contrast, considering the rate of economic growth with its associated consequences, it is becoming difficult to achieve the circular economy-oriented goals stated by countries, and required by environmental stakeholders.

Inversely, the circular economy-oriented models can effectively address human and environmental needs through adaptability and a resilient economic system through the minimal usage of the scarce resources and adoption of circular production processes. Moreover, several aspects of circular economy exist, which include the energy use and efficiency, environmental protection, and many other economic activities impact on the environment. Therefore, a concerted effort from the government and private sectors in promoting the movement from linear economy to circular economy is far advanced. The economy becomes more sustainable if there is a reduction in waste generation [4,5] and the material resource used for production and consumption [6,7], as well as an increased rate of recycling waste [8,9]. In line with prior research [5,6,8], in this study we chiefly define circular economy based on the level of materials considered as input factors for economic production, the amount of waste generated because of the extraction and usage of these materials and the rate of recycling of the generated waste. These circular economy variables are assumed to have significant impact on the environment.

Government is a major actor in this circular economy agenda. Through public policies in the form of regulations and financial support, government can positively influence the circular economy by helping in mitigating the impact of economic activities (firms and households) on the environment. Giving some examples of analyses so far, Banacu et al. [8] use the pooled OLS model to investigate the impact of environmental taxes, business expenditure on R&D and private investments on the recycling rate of municipal waste (recyw) in the EU region. Their results show that environmental taxes have a significant and inverse impact on the recycling rate of municipal waste (recyw). Next, using the linear regression model, Egüez [5] also finds a positive effect of stringency enforcement and financing on the compliance with municipal waste management among EU countries. Again, in the EU region, Cainelli et al. [6] use linear regression to show that environmental policy and green demand have a positive influence on resource efficiency through reduced domestic material consumption (dmc). In the Section 2, we further show previous research related to this study to display a more comprehensive picture.

Considering the related literature, it can be grouped into two mainstreams centered around the effectiveness of government regulation and subsidies towards circular economy (for a literature overview, please see Section 2). However, many of these extant literature focuses on the mostly subsidies and taxes (market-based policies), but the impact of the non-market-based stringency is less explored. Whilst this research has incorporated the market-based factors the analysis is extended to include environmentally related innovation and human capital on circular economy, which plays a greater role in the effort of instilling circular economic attitudes within the OECD. More concretely, the purpose of the study is to investigate the heterogenous impact on the attainment of circular economy by government policies in the form of government stringency and financial support, environmentally related innovations, and human capital in 26 OECD countries from 2010–2019 using the Poisson Pseudo Maximum Likelihood (PPML) model.

The study contributes to research into the government action which affect circular economy goals in three ways. First, it extends the body of knowledge on circular economy by assessing the heterogenous impact of non-market-based stringency and market-based subsidy and the environmentally related innovation, human capital, and government financial support on government policy towards circular economy. Second, it provides an avenue for making inference in the OECD territory on the level of circularity of economies in relation to the variables adopted in the study leading to public policy implication. These contributions are important for the fact that governments within the OECD countries have employed various policies in attempt to mitigating climate impact by human activities. It is therefore imperative to find out which aspects of the policies can influence the sustainable goals.

The paper is structured as follows. Section 2 and Section 3 are the theoretical framework and hypothesis, as well as the methodology and model of the study, respectively. The fourth Section presents empirical results and discussion is delineated in the Section 5. The conclusion and implications of the study follows in the Section 6.

2. Related Literature and Hypotheses Development

The waves in the economic booms coupled with population growth have caused much resource exploration for production. The modern economy has a great tendency and a need for material consumption that culminates in the massive extraction of raw materials to fuel the industrial cohorts of the predominantly linear economies of the world for economic growth [10]. The classical linear economy depends on the transformation of the extracted raw materials into finished goods, which then generate waste and are pushed through the production funnel, causing pollution and ecological disequilibrium [11]. The mammoth chase for this raw material has cast a dangerous scare on the environment, which faces a serious threat. Many economies, mostly, advanced, and emerging economies, are at fault for devising unsustainable techniques in their extraction, causing the danger of global warming and its associated emissions (climate change issues), resource depletion, and ecological extinction.

The solution professed to mitigate climate change has been the deployment of circular economy and innovation, which depends largely on the behavioral attitudes of all economic actors and households [12]. Hence, there is a growing literature in the field of circular economy. Consequently, one of the main barriers to the transition of firms from the linear to circular economy is the lack of effective government legislation [13] due to poor policy mix and coordination.

2.1. Public Policy and the Circular Economy: Joining the Theoretical Dots

Based on stakeholder and institutional theories, government is an important player with enormous power (economic and political) to affect adoption of circular economy. In a theoretical framework by [14], concerns about putting pressure on firms to be sustainable create both external and internal stakeholders. A major external pressure on firms is largely by society, which is carried through by public policies. The behaviors of economic actors are checked by government institutions. Hence, we build on the main thoughts of the institutional theory and link it to the adoption of a circular economy to show how government stringency in environmental policy and financial support impact sustainable behavior in the economy.

Furthermore, the coercive nature of institutional pressures through government regulation helps to control unsustainable activities in the economy [15], while it influences economic actors, such as firms, to engage in innovative activities, directly affect the level of impact of economic activities on the environment. Environmental policy instruments developed to show stringency (for instance, the Environmental Policy Stringency Index (EPS) within the OECD) include market-based environmental instruments such as emission trading schemes, DRS, feed-in tariffs, and environmental taxes. Additionally, the non-market instruments focus on R&D subsidies and environmental standards. Generally, based on the weak Porter hypothesis, government environmental regulations stringency promotes intermural research and innovation of the firm on the one hand. On the other hand, government environmental policy may demotivate firms to relocate when they cannot cope with the rigidity of the environmental regulation, which remains the major concern policy makers seek to answer. What is more and less of the stringency of the government policy on sustainability is a policy decision dilemma. In recent related research employing DEA and Quantile regression, Moutinho et al. [16] explain the differing Eco-efficiency scores between EU member states with respect to the disparity among the explored variables (Environmental Taxes Revenues, Resources Productivity, and Domestic Material Consumption). The researchers conclude that taxes have an inverse effect on more eco-efficient countries than less efficient ones, which calls for a policy review to align environmental tax imposition albeit the ecoefficiency levels within the EU.

On the market and non-market financial mechanism policy, Xie et al. [17] found that market-based financial policies control the environmental damage and greenhouse gas emissions affect cost of firms; nevertheless, non-market-based policies enforce existing environmental policies. In effect, a desire for the use of non-tax policies to achieve circularity, such as trading scheme [18], legal sanctions for non-compliance to circularity practices is on the rise, and legal sanctions are in place for non-compliance with circularity practices.

Based on the above arguments, we hypothesize as follows:

H_1a: 

The stringency of government policy has significant and positive effect on the indicators of circular economy.

2.2. Government Financial Support for the Circular Economy

Government financial support takes the form of support for research and development and financial incentives towards the environmentally friendly innovative activities of firms (most importantly, among SMEs within the EU [19]). This constitutes a concerted effort of the government to encourage circular economic activities of the firm and has received greater attention from practitioners and policy makers alike. However, a crowding-out problem may arise when firms substitute public finance for their internal environmental-related research and development [20]. Otherwise, it may also engage in less expensive circular economy activities due to financial constraints. Our argument is supported by [21] whose findings indicate the inability of SMEs to engage in eco innovation activities. In the conclusion of the research of [22], the researchers concur that public finance is crucial for the business model of waste management of SMEs. Likewise, within the public sector, Okuda & Thomson [23] find in their meta-analysis of Japan and the USA that they have an effect on waste management. The authors indicate that strict government control and funds dedicated to municipal waste management (mw) prove to be strong in Japan, whereas the market mechanism is prevalent in the US. This shows the diverse approaches among the OECD countries in the adoption of the circular economy and the drivers that spur its effectiveness.

Within the context of the EU, Busu [24] argues for continuous and considerable investments in the higher rate of material reuse and domestic material consumption within the EU economy. We argue that the more government expenditure on research and development towards a circular economy, the less domestic consumption of materials that consequently impact the recyclability of waste. However, the support of internal research and development has a positive impact on municipal waste management (mw). Hence, the following hypothesis was developed:

H_1b: 

Government financial support positively and significantly affects circular economy indicators.

2.3. Environmentally Related Innovation and Human Capital Effects on Circular Economy Attainment

Innovation related to environmental protection emerged a few decades ago. The tried-and-tested linear economy had less regard for the byproduct and seems to push through the production funnel the so-called waste materials. Such has been the call of international bodies and leaders across the globe to discuss the climate situation resulting in the sustainable development goals. Until then, the principles of circular economy were mere rhetoric except for the environmental protection activist. Innovation has been proven to influence the negative impact of economic activities on the environment [19,25]. Innovation activities involving R&D propel the adoption of novel eco-innovation technologies. R&D activities towards a circular economy include mainly the generation of scientific knowledge about the protection of the environment and relating to energy-efficient processes for energy production, distribution, and consumption.

This enforces eco-innovation activities and circular economy business models among firms as argued by the extant literature [19,25,26]. Environmental related innovation has theoretical underpinning from the human capital theory and the resource-based view, whose main tenants indicate the intellectual capabilities of resources (human and organizational-specific) to influence the generation of knowledge for eco-innovation activities towards circular economy achievement [27]. In the conceptual framework of [25], the knowledge of the circular economy is created through the synergy of basic and applied research and development of interconnected stakeholders to diffuse the same and be used as a leverage for the further development of environmental related technologies [28]. We argue that human capital and innovation within the OECD economies affect the rate of recycling of waste, municipal waste management per capita, and domestic material consumption. At the government level, support of the creation of human capital through the training of employees to improve knowledge, skills, and competencies within firms, civic and employment related perspective towards the achievement of circular economy constitute the interplay of the human capital and resourced-based circular economy nexus for which prior research has failed to avert on. In effect, we assume the hypotheses that:

H₂: 

Human capital stock positively and significantly affect circular economy indicators.

Moreover, we also assume that:

H₃: 

Environmentally related innovation has a significant and positive effect on circular economy indicators.

Please see our conceptual model in the appendix (Appendix A) for further review. In addition, in the

Table 1. Previous studies on variables.

Study	Period	Methodology	Country	Causality
[23]	1990–2005	Meta-analysis	Japan & US	Financing → mw
[8]	2010–2017	Pooled OLS	EU	GERD → recyw
[24]	2008–2017	GEM-general equilibrium model	EU	Financing → mw
[5]	2010–2016	Linear regression	EU	Stringency → mw
[6]	2008	Linear regression	EU	Stringency → dmc
[16]	2001–2012	DEA and Quantile regression	EU	Taxes → Eco-efficiencydmc → Eco-efficiency

Note: → unidirectional causality effect; domestic material consumption = dmc; municipal waste = mw; Recycling rate of municipal waste = recyw.

below, we show the summary findings of previous studies on the relationship between financing and regulation on our selected circular economy indicators.

3. Methodology and Empirical Strategy

We developed three separate models based on the three dependent variables (domestic material consumption, municipal waste generation per capita, and recycling rate of municipal waste) used as indicators for circular economy. Prior research [5,6,8] has strongly linked these indicators as effective means to measure the circular economy. The Poisson Pseudo Maximum Likelihood (PPML) estimation was used for this study. PPML was used because it considers ‘count’ data with zero values in its estimation [29] and provides a robust regression co-efficient, as it overcomes the problem of autocorrelation and heteroscedasticity [30].

Considering the proximity of the selected countries involved in our study, our data are likely to be characterized by cross-sectional dependence. In such analysis, the previous literature has employed linear regression analysis (see Table 1), mainly applying log-transformed dependent variable through an OLS regression. The major issue is that such estimation fails to deal with the sensitivity to the numerical problems of the dataset which result in misleading and biased conclusions on inferences due to the failure to adjust for heteroskedasticity and the distribution of the residual values. Hence, the PPML enables efficient estimations through a non-parametric regression for the performed analysis and is able to correct any inconsistencies of estimators when using OLS.

The PPML estimator ‘falsely’ omits cross-sectional dependence and still produces effective results because there are insufficient observations to obtain reliable estimates of the cross-sectional covariances [31]. Diebold [32] asserts that, the principle of ‘false restriction’ can improve estimator performance. Moundigbaye & Rea [31] further argue that, although the OLS is commonly used by scientific researchers, it performs poorly on both efficiency and inference grounds for small to moderately sized panel datasets. Our panel dataset is moderate with insufficient observations to conduct cross-sectional dependence. However, the adoption of the PPML estimator, ‘falsely’, omits the cross-sectional dependence to perform effectively in producing desirable results. PPML usually makes use of the robust choice for inferences as it is not essential to postulate it [33].

The assumptions of ordinary least squares (OLS) do not account for the adjustment for heteroskedasticity and the distribution of the residual values [34]. This could lead to misleading and biased conclusions on inferences made using the OLS. The independent variables used for our study are count data, with some years reflecting zero values. The use of the OLS model would not have been able to effectively deal with this issue. However, the use of the PPML model dealt with the issue of dropping such observations. The PPML model enabled us to avoid such biases in examining the impact government financing, environmental stringency policies, and innovation have on the circularity of the economy. In examining the impact of government financing policies, regulation policies and innovation on domestic material consumption, municipal waste per capita, and recycling rate of municipal waste used as the indicators for circular economy. The paper used linear panel estimators for the following equation for the three models.

Yit = β₀ + β₁X_1it + β₂X_2it + β₃X_3it +…+β_kX_kit + ε_it

(1)

where:

Y is the dependent variable.

β₀ is the intercepts.

β is the regression coefficients.

X is the various independent variables.

i is the various countries.

t is the time period.

ε is the error term.

The PPML regression is estimated by:

$$Pr\left({\mathsf{{\rm Y}}i=\frac{j}{\varkappa }}_i\right)=\frac{{℮}^{-\lambda }{\lambda }^j}{j^{!}}, j=\mathrm{0,1},2,\dots$$

(2)

where λ is generally specified as λ = ${℮}^{{\varkappa }_{i\mathsf{\beta }}^{\prime }}$= ${℮}^{{\mathsf{\beta }}_0 + {\mathsf{\beta }}_1{\mathrm{X}}_{1\mathrm{i}} + \dots }$.

3.1. Measurement of Variables

The achievement of circular economy goals is the key to economic growth. Environmental care is an important indicator of circular of economy [35]. Moreover, economic growth has adverse consequences for the environment in certain circumstances. Waste generation, recycling of waste [36], and domestic material consumption [37,38] are factors that mostly indicate the level of circularity of an economy. This study adopts the circular economy as the dependent variable and municipal waste per capita [36], domestic material consumption [37,38], and recycling rate of municipal waste [36] as its proxies (indicators). Our study used three independent variables. These are environmental policy stringency, government financing, environmentally related innovation, and human capital. Proxies for environmental policy stringency were non-market-based policies [39,40] and market based policies [40,41], for government financing were gross domestic expenditure on R&D by source of funds [42], R&D expenditure intramural [43,44] and national expenditure on environmental protection [45,46], and environmentally related innovation and human capital proxies were patents to protect collaborative inventions of environmentally related technologies [47,48] and employees involved in education and training [49].

Table 2. Variable description.

	Variable	Description	Reference	Source
Dependent variable	Domestic material consumption per capita (dmc)	DMC refers to the number of materials (extraction of raw materials, consumption of materials) directly used in an economy. It is computed as domestic extraction used minus exports plus imports.	[36]	OECD
Circular Economy	Municipal Waste per capita (mw)	This is the waste collected by or on behalf of municipal authorities excluding waste from municipal sewage networks and treatment. It is measured in kilogram per capita	[37,38]	OECD
	Recycling rate of municipal waste (recyw)	This is the tonnage of recycled municipal waste per total tonnage of municipal waste generated. It is expressed in percent (%) terms.	[36]	Eurostat
Independent Variables	Non-Market-Based stringency (nonmkt)	Is a degree of regulation that put certain obligations on firms by installing non-monetary incentives to change environmentally harmful behavior. Degree of stringency: 0 = not stringent and 6 = highest stringent	[39,40]	OECD
Environmental policy stringency	Market-based subsidy (mkb_str)	Is a degree of regulation that put and explicit or implicit price or payment on environmentally harmful behavior. Degree of stringency: 0 = not stringent and 6 = highest stringent	[40,41]	OECD
Financing	Gross domestic expenditure on R&D by source of funds (GERD)	This refers to Gross Domestic Expenditure by government sector and by source of funds (government, higher education, private non-profit and business sectors).	[42]	OECD
	R&D expenditure intramural (R&D_intr)	This refers to the research and development expenditure by the government and by field of research (environment, engineering, natural sciences, and social sciences).	[43,44]	OECD
	National expenditure on environmental protection (ne_epro)	It refers to the financial resources committed in protecting the natural environment in relation to reduction and elimination of waste and other pollutants.	[45,46]	Eurostat
Environmentally Related Innovation	Environmental patents with foreign co-inventors (env_pat)	This is the percentage of patents to protect international collaborative innovations with OECD countries in environmentally related technologies.	[47,48]	OECD
Human capital	Employees involved in education and training (emp_edtr)	This refers to all learning activities undertaken with the aim of improving knowledge, skills, and competencies within a personal, civic, and employment related perspective. Unit of measure is in percentage of total employees involved in education and training.	[49]	Eurostat

below shows a detailed description of the variables, and their proxies used, as well as existing studies that used similar variables and data sources.

3.2. Description of the Data

This study used secondary panel data from OECD and Eurostat databases. Centralized data on circular economy, government financing, and environmental policy stringency for OECD and European countries. Over the years, the OECD and Eurostat databases have become vital sources of data for researchers globally. Our study considered these sources because they use consistent, ethical, and accurate methods in data collection. This makes the data credible, reliable, and highly valid for any research analysis. This data has been used by prior researchers [5,50] in similar studies. Our study collected data on 26 OECD countries from the European territory. The countries were Belgium, Czech Republic, Denmark, Germany, Estonia, Ireland, Greece, Spain, France, Italy, Latvia, Lithuania, Luxembourg, Hungary, Netherlands, Austria, Poland, Portugal, Slovenia, Slovakia, Finland, Sweden, Norway, Switzerland, United Kingdom, and Turkey.

The data period was from the years 2010 to 2019. The overall observations of the panel data were 334. The reason for choice of these member states was the significant differences observed for municipal waste generation, the recycling rate of municipal waste [51], and the domestic material consumption among them. It will be interesting to know the extent of government financing, environmental policy stringency, and innovation impact on the circularity of these countries.

4. Empirical Results

We used Stata/MP 17.0 for our statistical analysis. We first ran the Pearson correlation matrix for all the indicators in this study (see in Appendix B). The results enabled us to examine the degree of correlation that existed between the study variables. This was adopted to test the multicollinearity among the variables in the model. The values of the pairwise correlation co-efficient indicated moderates (±0.6) to low (±0.1) correlation among variables [8,52]. Our correlation table exhibited higher correlation (>±0.6) for R&D expenditure intramural (0.6928). However, the trace of multicollinearity was resolved with the use of the PPML estimator for the regression analysis [33]. A further test was carried out to check for multicollinearity using the Variance Inflation Factor. The mean VIF of 1.59 shown in the VIF table (Appendix C) indicated the absence of multicollinearity in the model [53]. In addition, summary statistics was conducted and has been provided in Appendix D for further review.

The results presented in the

Table 3. Regression analysis of government support and innovation activities towards the circular economy.

Variables	Model 1 dmc	Model 2 mw	Model 3 recyw
Environmental patents with foreign co-inventors	−0.114 ** (−1.04)	−0.114 ** (−6.87)	−0.018 * (0.47)
Gross domestic expenditure on R&D (GERD)	0.054 *** (−3.87)	−0.024 ** (−1.63)	−0.007 * (−1.84)
Employees involved in education and training	0.035 (3.61)	0.045 ** (0.01)	0.006 (0.15)
R&D expenditure intramural	0.045 *** (0.01)	0.045 *** (3.71)	−0.024 *** (−5.45)
National expenditure on environmental protection	−0.022 *** (−4.08)	0.033 *** (4.15)	−0.026 *** (−10.94)
Non-market-based stringency	0.458 ** (2.52)	0.564 ** (2.78)	0.346 *** (5.92)
Market-based subsidy	0.014 (0.22)	0.287 *** (3.69)	−0.043 ** (−2.01)
_cons	−2.948 * (−2.37)	4.582 ** (2.58)	2.702 *** (5.44)
R2	0.565	0.887	0.800
N	141	141	138

Note: Significant p values denoted by * p < 0.1; ** p < 0.05; and *** p < 0.01. Robust z scores in parenthesis estimated using PPML.

show the extent of impact indicators for the stringency of environmental policies, government financing, and environmentally related innovation on domestic material consumption in model 1, municipal waste per capita in model 2, and the recycling rate of municipal waste per capita in model 3. Domestic material consumption includes materials used for production and consumption domestically. The materials are in the form of metals, non-metallic minerals, biomass, and fossil energy. The use of these materials has economic, social, and environmental consequences beyond borders. Although environmentally related innovation indicators, such as environmental patents with foreign co-inventors, showed a significant negative relationship (β = −0.114; p < 0.05) and a significant impact on domestic material consumption, employees involved in education and training revealed a positive (β = 0.035) relationship (=0.035), but no significant effect on domestic material consumption. This implies that, the more inventions of environmentally related technologies are protected through patents, the less materials used for consumption and production.

In government financial policies towards domestic material consumption, gross domestic expenditure on R&D and national environmental protection show a negative significant (β = −0.054; p < 0.01) and (β = −0.022; p < 0.01) association with dmc, while R&D intramural showed positive significant association (β = 0.045; p < 0.01) with dmc. Thus, an increase in expenditure on research in the field of sustainability and financial resources committed in protecting the environment leads to a reduction in materials used domestically for production and consumption. On the other hand, both indicators for environmental stringency policies showed positive association with dmc. Non-market-based stringency showed a positive and statistically significant (β = 0.458; p < 0.05) impact on dmc. Market-based subsidy showed a positive association with DMC but was not statistically significant on dmc. The value of R² in model 1 was 0.565. This means 56.5% of total variations DMC is explained by the independent variables in the model (government financing, environmental stringency policy, and environmentally related innovation).

In model 2, municipal waste generation per capita was the dependent variable. Municipal waste (including household waste) per capita generated in each country is related to urbanization, pattern of consumption, household revenue and lifestyle per capita. With the exception of environmental patents with foreign co-inventors,

Table 4. Marginal effect estimates.

Variable	dy/dx	SE
Environmental patents with foreign co-inventors	−0.007 (−0.56)	0.013
Gross domestic expenditure on R&D (GERD)	−0.014 (−0.59)	0.024
R&D expenditure intramural	−0.023 (−1.19)	0.019
Employees involved in education and training.	0.052 *** (3.31)	0.015
National expenditure on environmental protection	−0.037 *** (−3.18)	0.011
Non-market-based stringency	0.437 * (1.52)	0.288
Market based subsidy.	−0.037 (−0.23)	0.162

Note: Robust z scores in parenthesis estimated using PPML: Significant p values denoted by * p < 0.1 and *** p < 0.01.

shows that, all exogenous variables used in the model had a significant impact on municipal waste per capita. GERD had a negative but significant (β = −0.04; p < 0.05) impact on municipal waste generation per capita. However, employee’s participation in education and training had a positive significant (β = 0.012; p < 0.05) impact on municipal waste generation per capita. There was also a negative but significant (β = −0.017; p < 0.05) impact of R&D expenditure intramural on municipal waste generation per capita. Additionally, national expenditure on environmental protection (β = −0.033; p < 0.05), had a negative but significant impact on municipal waste generation per capita. Environmental policy stringency measured by non-market-based stringency (β = 0.099; p < 0.05) and market-based subsidies (β = −0.679; p < 0.05), respectively, had a negative and positive significant effect on municipal waste generation per capita. The R² in model 2 was equal to 0.854. This means our model explains 85.7% of the total variations in the municipal waste per capita.

In model 3, we examined the impact of the government stringency policy, environmentally related innovation, government expenditure, and human capital on recycling rate of municipal waste generation. To achieve sustainable cities and communities, towards a circular economy, there should be an increased rate of recycling rate of municipal waste. From Table 4, patents on environmentally related technologies had a negative and significant (β = −0.018; p < 0.1) relationship with the rate of recycling of municipal waste, on the other hand, employees involved in education and training had a positive but not significant impact on the recycling rate of municipal waste. On government financing for circularity, gross domestic expenditure on R&D (β = −0.007; p < 0.1), R&D expenditure intramural (β = −0.0024; p < 0.01) and national expenditure on environmental protection (β = −0.026; p < 0.01), all had an inverse association with municipal waste recycling rate but were statistically significant. Although non-market-based stringency had a positive significant relationship with the municipal waste recycling rate, market-based subsidy had inverse association and a statistically significant (β = −2.702; p < 0.01) impact on the recycling rate of municipal waste generation. By implication, a unit change in environmentally stringent variable leads to an average change in municipal waste generation per capita by an average amount keeping all factors constant. With R² of 0.800, it means that 80.0% of total variations in recycling rate of municipal waste is explained by the independent variables used in our model.

To further test the robustness (see Appendix E) of the model, we used panel correlated standard error estimator (PCSE). This was to correct cross-sectional dependence and heteroskedasticity [31].

Additionally, we ran a test for the average marginal effects (AME). The AME estimates the partial effects of the independent variables on the dependent variables as outlined in the table below.

The results in Table 4 demonstrate the marginal effect of the explanatory variables on the circularity of the economy. Developing environmental patents with foreign collaboration (β = −0.007; p < 0.575) is effective in ensuring the circularity of the economy. This implies that an increase in environmental innovation per unit reduces municipal waste generation per capita and domestic material consumption by 0.7%. This leads to an increase in the circularity of an economy. A unit increase in gross domestic expenditure on R&D (GERD), (β = −0.014; p < 0.556) reduces domestic material consumption and municipal waste per capita by 1.7%, thereby enhancing circular economy. It implies that government expenditure on R&D will lead to efficient exploitation and use of resources, thereby reducing environmental pollution. The more employees are involved in education and trained (β = 0.052; p < 0.001), the better it is to achieve a circular economy. This implies that a unit increase in the involvement of employees in education and training on environmental issues leads to a 5.2% improvement in municipal waste recycling rate. Similarly, a unit increase in intramural R&D expenditure (β = −0.023; p < 0.234) leads to a 2.3% reduction in the generation of municipal waste per capita and domestic material consumption. This, in the long run, improves the circularity of an economy. National expenditure on environmental protection (β = −0.037; p < 0.001) reduces environmental pollution by reducing the generation of municipal waste and domestic material consumption. Therefore, a unit increase in government expenditure to protect the environment increases the circularity of the economy by 3.7%. Non-market-based government stringency (β = 0.437; p < 0.130) and market-based subsidy (β = −0.037; p < 0.815) have a varied impact on the circularity of the economy. With a unit increase in the non-market- based stringency leads to a massive improvement in circular economy by 43.7%. This is made possible by reducing municipal waste generation per capita and the consumption of domestic material consumption. Market base subsidy, on the other hand, leads to 3.7% reduction in municipal waste generation and domestic material consumption. Hence, a unit increase in market-based stringency policy leads to circularity by 3.7%. Effectively, the non-market-based subsidy has the highest marginal effect on the circularity of the economy.

5. Discussion

Non-market-based stringency on the circularity of the economies had the biggest impact among the indicators used in the study. It shows that command-and-control regulation and active technology support policies should be among the most effective means of controlling domestic materials used for production and consumption. The results imply that market-based subsidy alone is not effective in achieving circular economy attainment. However, it enforces such payment of environmentally related taxes [5], and the simultaneously blending market-based subsidy and non-market-based stringency [54] has a great impact on the advancement circularity in the economy. Government financial support has a significant impact on domestic material consumption. Government expenditure on research and development in the fields of environment, engineering, natural sciences, and social sciences, as well as financial resources committed to protecting the environment through reduction in domestic material use, has great potential to achieve circular economy. This supports [13] in the argument that government funding for R&D leads to a reduction in waste created by materials used in production and consumption, thus improving environmental sustainability and the circularity of the economy.

Government should consciously make adequate budgetary allocation towards research and development related to the environment, strengthening of government institutions, and building the capacity of the personnel to recognize the importance of circularity. Innovative ways to reduce the amount of material used in production and consumption come at a cost. SMEs might struggle financially to engage in such innovative ventures. Therefore, government support is needed for such firms to embark on innovative ways of extracting raw materials for production and consumption without endangering the environment. Thus, the role of government in ensuring responsible domestic material use is very important in the long-term attainment of the circular economy.

Additionally, the environment related innovation and human capital have a significant impact on domestic material consumption. Environmental patents with foreign co-inventors encourage inventions and innovations of environmentally related technologies for improving domestic material use. Collaboration with foreigners in similar field of interest opens to innovation [55,56]. Hence, once collaborative innovation is protected through patenting, it encourages more research [55] in the sustainability field. This supports [6] in the idea that environmental innovation is closely related to domestic material consumption. On the other hand, employees involved in education and training with the aim of improving knowledge, skills, and competence did not have any significant impact on domestic material consumption. This contradicts [57] in the assertion that not considering human development or behavioral factors hinders environmental responsibility of firms and individuals. It implies that responsible domestic material use is positively linked to building human capacity through education and training. Municipal waste generation and treatment is a major issue among countries experiencing high levels of urbanization. It also has an association with the pattern of consumption, household revenue and lifestyles. This poses a threat to the circularity of the economy. The reduction in waste generated the better the impact on the circularity of the economy.

Our study reveals that government stringency has a significant effect on municipal waste generation per capita. Non-market-based stringency has more (56.5%) impact on municipal waste generation per capita than market-based subsidy (28.7%). This is consistent with the extant literature [5,19]. However, our findings failed to corroborate with [58] that government stringency in the form of regulations is not the most effective way to promote the generation and treatment of municipal wastewater towards circularity. By implication, government regulations on materials used in packaging products by firms, household treatment of waste in terms of sorting and separation of waste products should be enforced. This could be done through environmental taxes and fines for firms and individuals that flout such environmental regulations.

The study also found a significant effect of government support on municipal waste generation per capita. Thus, national expenditure on environmental protection, non-market-based stringency and market-based subsidy have significant influence on the reduction of municipal waste generation per capita. However, non-market-based stringency has a positive correlation, employees involved in educational, and training have positive association with municipal waste generation per capita. The performance of municipal waste generation per capita among some OECD countries is worryingly low [59]. The onus for effective waste management unto circularity should, therefore, not lie on only the private firms and individuals but the government as well [60] by strengthening the degree of stringency in the environmental protection regulations. Government should allocate more funds to research and development on effective means of municipal waste management especially in the urbanized areas of the OECD countries [61].

In addition, there is also a significant effect of environmentally related innovation and human capital on municipal waste generation per capita. The education and training of employees equip them with the requisite skills and know-how on effective management of waste in the workplace and at home. Open innovation through collaborative patent with foreign co-inventors serves as avenue for transfer and sharing of skills, ideas, and knowledge on the best practices for municipal waste treatment. Additionally, environmentally related patents protect technological inventions and innovations towards circular economy. This assertion is supported by [50], who posits that environmentally related technology development has inverse relation on waste generation per capita. It means, the more environmentally related technologies are developed, less waste is generated.

Therefore, innovation policies in OECD countries should support the design of environmentally friendly technologies towards circular economy. Human capital development in the form of training and re-training of employees to properly handle municipal waste is very important in the waste management process. Green human resources, in the form of training and education of employees, positively enhance environmental performance through employee skills and knowledge in handling waste as found by [62] corroborate our findings.

Regarding the recycling rate of municipal waste, rigorous non-market-based stringency significantly enhances recycling rate of municipal waste. This aligns with the existing literature [63] that strict environmental policy promotes the rate of recycling of waste. Contrary to this, high market-based subsidies in terms of environmental taxes stifle recycling rate of municipal waste. This implies that the use of standard setting and compliance in ensuring high recycling rates of municipal waste is effective as compared to environmental taxes. On the other hand, simultaneous application of both stringent policies yields better results [54]. Surprisingly, government support in the form of government expenditure on research and development and national expenditure on environmental protection has no positive impact on the recycling rate of municipal waste. In addition, environmental innovation did not have any positive influence on the recycling rate of municipal waste. This is inconsistent with studies such as [64,65], which claim that government expenditure on R&D and the provision of machinery promote waste recycling in the circular economy. Employees involved in education and training did not have a significant impact on recycling waste. Park et al. [53] stated that factors such as educational level, attitude, income level, and experience influence recycling habits. The environmentally related technologies, new inventions, and innovations contribute positively to a higher rate of recycling.

6. Conclusions

6.1. Contributions

The main aim of this paper is to examine the heterogenous influence of government stringent policies, government financial support and environmentally related innovations, and human capital on the circular economy using the PPML model. Three circular economy variables were used to depict three models. They are domestic material consumption, municipal waste per capita, and recycling rate of municipal waste. The research sought to test four main hypotheses. These hypotheses include the degree of government regulation or stringency of the government policy, government financial support, human resource and environmentally related innovation have positive and significant effect on circular economy.

The results show that government stringency, such as non-market-based stringency, has significant influence on domestic material consumption, but market-based subsidy did not have significant influence on domestic material consumption. Furthermore, government financial support (GERD, R&D expenditure intermural, and national expenditure on environmental protection) had a significant influence on domestic material consumption. Similarly, environmental patents with foreign co-inventors have significant influence on domestic material consumption but employee involved in education and training did not. Furthermore, stringent government policy, government expenditure, environmental innovation, and human capital contributes significantly to municipal waste reduction. Except for non-market-based stringency, all independent variables did not have a positive influence on the recycling rate of municipal waste. Among the variables employed, non-market stringency had the most significant impact on the circularity of the economy across the 26 OECD used in the study within the 10-year period (2010–2019). Again, municipal waste was the most influenced variable in the circularity impact. Our work therefore contributes to the body of knowledge on circular economy regarding the most significant factor to promote circular economy, thus proving the heterogeneous influence of government stringent policies, government financial support and environmental related innovations and human capital on the circular economy.

6.2. Implications

First, non-market-based stringency in terms of putting non-monetary obligations on firms and individuals is a major key factor to promoting circular economy. Policies on circular economy among OECD countries should place emphasis on environmental regulations more than other factors. This is because environmental regulations are a more proactive measure to ensure a circular economy than market-based subsidy measures, which are more reactive. Second, economic growth is associated with urbanization and is also associated with the issue of municipal waste. Our study points to the fact that one key element to address in achieving circularity is municipal waste. It is the most impacted circular economy in our study. Education and training, financing waste reduction and protection innovations to reduce municipal waste should be encouraged and consciously executed towards a circular economy. Third, achieving a circular economy is costly. It should not be left to only firms and private individuals to embark on the venture. Government, as well as international organizations, must all lend support in diverse ways to achieve circular economy goals. Finally, collaborative inventions among OECD member states are vital in achieving a circular economy. Countries are encouraged to collaborate in the production of environmentally friendly technologies.

6.3. Limitations and Future Research

Our study is not without limitations. The achievement of a circular economy is a global issue. Other researchers have provided factors such as GDP per capita income, population, green human resources, and green procurement as factors influencing the circular economy. Our research focused on the impact of non-market-based stringency, market-based subsidies, environmental patent with foreign co-inventors, government expenditure on R&D, and employees involved in education and training. One of the main limitations of this study may be the fact that we used secondary data. However, this approach allowed us to provide the reader with a more comprehensive picture on this topic. Even so, we recommend that future research focus on single countries and on a combination of quantitative and qualitative research. We also propose that future research should explore institutional policies and demand side policies towards circular economy. Additionally, grouping the OECD countries into the level of circularity will be a novel approach in researching into circular economy impacts.