National Environmental Taxes and Industrial Waste in Countries across Europe

Abstract

The use of economic instruments within environmental policy has become a challenging topic for policymakers, governments and scholars. Environmental taxes have emerged as a prevailing preference in developed countries to promote sustainability. Recently, a particular focus has been given to waste generation and disposal, shifting the attention from greenhouse gases to another important source of environmental pollution. This paper investigates the effect of national environmental taxes and policies on industrial waste. A fixed effects model is used for 34 countries across Europe from 2004 to 2022. The results suggest that environmental taxes and energy policies reduce industrial (hazardous and non-hazardous) waste. However, environmental tax reforms should take into consideration the deterioration in environmental quality, the increase in economic costs and undesirable social consequences.

1. Introduction and Motivation

The EU aims to become the first climate-neutral continent by 2050 with net-zero greenhouse gas emissions. This climate neutrality is also in line with the EU commitment to global climate action under the Paris Agreement [1]. To meet this goal, the European Commission has set a goal for the overall energy mix composition to comprise at least 40% renewables, reducing the net greenhouse gas emissions by at least 55% by 2030 compared to 1990 levels and revising the current EU waste laws to make producers responsible for the full life-cycle of products.

As industries constitute major drivers of economic growth, they have a significant impact on the environment. In developed countries, concerns about the environment and sustainability are increasing and industrial waste generation has become an important topic of discussion and analysis. Industrial waste is the largest source of waste [2] and is even more dangerous for human health when it is classified as toxic and hazardous [3]. Due to the fact that industrial waste is processed differently from urban waste [4] and industries are responsible for its disposal or recycle, environmental policies have been established focusing on the industrial sector [5]. Hence, the effective management of industrial waste has risen to prominence as a pivotal component of sustainable development agendas worldwide.

Generally speaking, the objectives of environmental policy focus on achieving sustainable development and environmental targets. In order to minimize the costs associated with environmental resolutions, address externalities and increase revenue for designated goals, policymakers employ incentive-based instruments. These instruments include environmental taxes, fees and charges, permits, deposit–refund systems, and subsidies, among others. Environmental taxes, sometimes referred to as green taxes, pollution taxes, or eco-taxes, are a broad category of charges imposed by law on companies and citizens with the goal of curtailing activities that harm the environment. Thus, in line with Regulation (EU) No 691/2011, environmental tax can be defined as “the tax whose base is a physical unit (or a proxy of a physical unit) of something that has a proven, specific negative impact on the environment and is identified in the European System of Accounts (ESA) as a tax” [6]. Compared to the United States, the majority of European countries present relatively high levels of environmental taxation [7]. Unlike regulatory or administrative techniques, the ability of taxes to affect markets in a cost-effective manner is economically rational and is the reason they are preferred for use. Since environmental harm affects a large number of people and has little to no direct cost to the polluter, financial incentives for businesses or households to account for their undesirable outcomes do not exist when the government is not involved. Moreover, as green tax reform [8] (green tax reform involves an increase in taxes regarding the environment and a reduction in taxes on other tax bases) has recently gained a lot of attention, data on environmental tax revenue as a share of all taxes revenue are important for the estimation of environmental impact. Overall, the utilization of environmental taxes “prices in” the environmental costs due to market failure and gives customers and companies the freedom to choose the least expensive strategy to diminish their environmental damage. As a result, even if a considerable reduction in emissions or waste has already taken place, environmental taxes (i) offer a constant incentive to reduce undesirable outputs, (ii) increase demand for low-emission waste alternatives and (iii) provide incentives for firms to develop new innovations and adapt existing ones due to the increase in the cost for the polluter [9].

Economic instruments, such as taxes on the environment, pollution, industrial waste and energy, have been consistently integrated into environmental policies on the continental, national and regional level [10]. Since industrial waste has become a negative consequence of industrial processes that threatens humanity, environmental taxes are imposed on companies in order to mitigate their waste generation, but they also serve as an additional economic revenue for governments. Taxes can directly address the incapacity of markets to include environmental effects in prices. In this sense, environmental taxes have been successfully used to address a wide range of issues including waste disposal, water pollution and air emissions [9,11,12,13,14,15,16]. Moreover, taxes on environmental pricing give firms and consumers the freedom to choose the best way to mitigate their “footprint” on the environment. In the literature, even though there is quite a number of studies that analyze the effects of tax policies on urban waste (i.e., [17,18,19,20,21,22]), there is limited empirical evidence for the case of industrial waste. More specifically, studies focus mostly on the role of economic growth, the stringency of environmental policies, energy mixes, and innovation, amongst other factors, regarding reductions in atmospheric emissions (i.e., [23,24,25,26,27,28]). Other scholars estimate environmental/energy efficiency and productivity and their effect on sustainability [29,30,31,32,33].

Hence, it is imperative to center attention on industrial waste generation and determine the effects of different economic and environmental mechanisms in waste abatement might cause. Environmental protection regulations and spatial dependence [34,35], legislative frameworks, effective management and diverse tax reforms [36] can provoke quite immense shifts in attitudes and change industrial waste quantity. However, environmental taxes can be considered as an effective regulatory tool when accompanied by transparency and targeted abatement strategies [37]. In this sense, their implementation can mitigate the negative environmental impact by promoting energy conservation and emission and waste reductions. Furthermore, as [38] indicated, low levels of environmental taxation do not sufficiently prompt the adoption of green technologies or an increase in green investments. The novelty of this study lies in the examination of the impact of environmental taxes on industrial waste for a set of countries. Environmental taxes can neither prompt results regarding industrial waste and their abatement. Based on the level of their adoption, an attempt to determine how different countries across Europe try to regulate the increasing quantities of waste is explored in this research work. We also take a step further and shed light into understanding this relationship by using additional variables emphasizing energy, innovation and population characteristics. Knowing which factors can mitigate waste pollution is imperative. To our knowledge, this is the first work that examines the impact of environmental taxes on industrial waste at such a wide national level by also taking into account a variety of explanatory variables. The climate crisis needs to be controlled and economic science plays an important role in this endeavor.

To this end, an analysis is performed targeting the identification of the effect of national environmental taxes on industrial waste. A unique panel dataset of 34 countries from 2004 to 2022 is created to explore the aforementioned relationship by additionally incorporating other factors such as energy production, its intensity, and research and development of hazardous and non-hazardous waste. The method employed in this study, the fixed effects (FE) model, is a well-known econometric technique that accounts for endogeneity that is present in a dataset [39]. (In this study, the endogeneity lies in the existence of data from many countries over many years, and an unobserved time-invariant individual effect exists.) The results confirm the initial assumption that taxes are an effective tool to mitigate industrial waste. Environmental taxes on goods and services are in favor of the environment [40], encouraging eco-friendly practices and new green technologies that would abate emissions and waste.

The remainder of the paper is structured as follows. Section 2 develops the methodology and Section 3 describes the data, while Section 4 presents and discusses the empirical findings concerning the relationship between the dependent variables and the examined independent ones. Lastly, Section 5 summarizes and concludes the paper.

2. Methodological Framework

2.1. Fixed Effects Model

Economic science has changed rapidly during the twentieth century. The most radical change was the introduction of econometrics in the field. Econometrics involves a quantitative analysis of phenomena that are related to economics, and it heavily depends on statistical and mathematical techniques, known as regressions [41]. Thus, regression is a method that attempts to determine the strength and character of a relationship between one dependent variable (denoted as Y) and a series of other variables (known as independent variables $X_i$). The existence of different types of data and the analysis of their economic relations on a micro and macro level has resulted in the emergence of many econometric techniques. (We owe this to an anonymous reviewer.).

The fixed effects model addresses and captures intricate dynamics related to panel data analysis. When analyzing panel data, standard regression linear models fail to capture the correlation due to the presence of grouped and correlated observations corresponding to single units repeated over time [42]. At the same time, the fixed effects model allows one to take into account the influence of time-invariant unobserved effects or omitted variables [41]. This becomes even more important when investigating the relationship between industrial waste and environmental taxes since unobserved effects might introduce bias, making the analysis unreliable. Thus, the supervision of individual-specific heterogeneity or region-specific characteristics is necessary in order to understand the true relationships between the variables.

The model considered in the empirical analysis allows for a better focus on time-varying explanatory variables. This aspect is crucial when attempting to understand which factors can influence early and medium-term decisions given that the time invariant factors are fully captured, by definition, in the individual fixed effects. The general form of the model can be described as follows:

$$y_{it}={\alpha }_i+{\beta }_z{z}_{it}+\gamma {\mathit{X}}_{it}+u_{it},i=1,\dots ,N,t=1,\dots ,T,$$

(1)

where $y_{it}$ represents the industrial waste for a specific unit i at a given time t, ${\alpha }_i$ captures the fixed effect associated with unit i and ${\beta }_z$ is the coefficient corresponding to our variable of interest, namely environmental taxes ($z_{it}$) for unit i at time t. Furthermore, $\gamma$ represents a vector of coefficients associated with the vector of specific explanatory variables ${\mathit{X}}_{it}$ of unit i at time t, while $u_{it}$ represents the standard error term. This model structure enables us to explore the relationship between environmental taxes and (hazardous and non-hazardous) industrial waste while accounting for unit-specific effects, other explanatory variables and potential unobserved factors.

To evaluate the choice of adopting a fixed effects model against the alternative of a random effects (RE) model [42] in the presence of panel data, the well known Hausman test [43] was performed. The random effects model assumes individual effects as part of the error term, so it treats them as stochastic components that are certainly uncorrelated with the regressors. The results of the Hausman tests indicate that the fixed effects model is the most appropriate model (results are available upon request). Thus, its robust capability to accommodate and control for individual-specific heterogeneity strengthens its suitability in addressing the research objectives with enhanced precision and insight.

2.2. Empirical Model Specification

We consider a series of nested models to deeply investigate which factors can influence industrial waste generation. More specifically, the models are classified into two primary groups: one that encompasses the total industrial waste generation as the dependent variable, including both hazardous and non-hazardous waste, and another that focuses solely on non-hazardous industrial waste.

For the total industrial waste (hazardous and non-hazardous) case, model (1) can be presented as follows:

$$\begin{array}{c}wast{e}_{it}=\alpha +\beta EnvTax{R}_{it}+\gamma EnIn{t}_{it}+\delta EnPro{d}_{it}+\hfill \\ ϵR{D}_{it}+\zeta Po{p}_{it}+\eta GD{P}_{it}+{\theta }_t+u_{it}\hfill \end{array}$$

(2)

where $EnvTaxR$ measures the environmental tax revenues, $EnInt$ is the energy intensity, $EnProd$ is the productivity, $RD$ is the research and development (R&D) expenditure, $Pop$ is the population, and $GDP$ is the gross domestic product (GDP) for country i in year t. We also consider a time-specific effect represented by the variable year (t) [39,44]. The purpose of the employed model is to capture the collective impact of these factors on total industrial waste generation by evaluating how different characteristics of economic, technological and demographic dimensions contribute to the dependent variable.

Similar to model (2), we adopt a nested model approach for the non-hazardous industrial waste model:

$$\begin{array}{c}nhwast{e}_{it}=\alpha +\beta EnvTax{R}_{it}+\gamma EnIn{t}_{it}+\delta EnPro{d}_{it}+\hfill \\ ϵR{D}_{it}+\zeta Po{p}_{it}+\eta GD{P}_{it}+{\theta }_t+u_{it}\hfill \end{array}$$

(3)

Considering a different model estimation for non-hazardous waste separately allows for a comparison between total waste and non-hazardous waste while gaining insights into the specific factors/drivers affecting a specific part of this variable. Overall, taking into account the aforementioned nested models, a systematic exploration of the effect of the different factors on industrial waste generation is feasible whilst the inclusion of various independent variables at each time point in the original model provides valuable insights into waste management policies and practices.

3. Data and Variables

To empirically investigate the relationships described in Section 2.2, a unique panel dataset was constructed based on several international sources like Eurostat, the Organization for Economic Co-operation and Development (OECD) and the World Bank. This dataset includes 34 countries (Austria, Belgium, Bulgaria, Croatia, Cyprus, Czechia, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Liechtenstein, Lithuania, Luxembourg, Malta, Netherlands, North Macedonia, Norway, Poland, Portugal, Romania, Serbia, Slovakia, Slovenia, Spain, Sweden, Turkey, United Kingdom), covering an expansive time span from 2004 to 2022 with annual observations at its core, each a distinct pairing of a country and a specific year, for a total of 646 unique observations. Data pre-processing required a specific effort for the refinement of the variable representing industrial waste. Industrial wastes are defined as contamination stemming from industrial processes [45]. A prudent aggregation of hazardous and non-hazardous waste variables was needed, harmoniously creating the dependent variable. Notably, countries exhibiting a substantial number of missing values concerning the variable of waste were excluded from the analysis, ensuring the dataset’s robustness.

Table 1. Descriptions of the employed variables.

Variable	Description	Type
Country	The country for which the data observations are recorded	Categorical
Year	The year in which the data observations were recorded	Numeric
Waste	Total waste (hazardous and non-hazardous) generated by the industrial sector	Continuous
EnvTaxR	Revenue generated from environmentally related taxes	Continuous
EnInt	Energy intensity level of primary energy consumption relative to gross domestic product	Continuous
EnProd	Productivity of energy consumption in terms of economic output	Continuous
RD	Gross domestic expenditure on research and development for all sectors as a proportion of GDP	Continuous
Pop	Proportion of the population aged 65 years and older	Continuous
GDP	Gross domestic product, a measure of the country’s economic performance	Continuous

presents a description of the variables employed in this study which have also been used in similar research works. The nexus between waste and taxation is a big concern for policymakers [35,46]. The optimal taxation needs scrutiny and careful consideration of other factors that might limit waste generation. Taxes are selected as they constitute an effective tool to control the generation of waste, while energy dynamics reveal how the use of energy can impact sustainability. Expenses in research and development promote environmentally friendly technologies and demographic and economic variables offer insight into waste generation management and waste disposal.

The variable of waste generation was derived from Eurostat and shows the total waste generated by the industrial sector, measured in tonnes. Waste management and reduction in the context of sustainable industrial practices are pivotal in our research. Aligning with prior research studies, as [35] have pointed out, waste generation is not just a random process; it can be influenced by both economic and environmental factors. Thus, understanding these factors is crucial for devising effective environmental policies and sustainable industrial strategies.

Environmental tax revenues, drawn from the OECD, represent the total environmental taxes as a percentage of the total tax revenue. According to the literature, there is evidence of a strong connection between economic policies, such as environmental taxation, and waste management strategies [47], and, thus, it is crucial to examine this relation. The applied methodological approach allows us to evaluate the impact of environmental taxation on waste generation patterns, contributing to the existing discussion on the effectiveness of economic instruments in environmental policy.

Energy intensity, derived from World Bank, is measured in constant 2017 USD at PPP and shows the ratio of energy use to GDP. As highlighted by [48], the efficiency of energy usage plays a pivotal role in shaping environmental policies and waste generation patterns. Indeed, an energy intensity reduction indicates a more effective use of energy resources, potentially resulting in lower waste generation. Moreover, the productivity of energy consumption, obtained from Eurostat and measured in EUR per kilogram of oil equivalent (kgoe), indicates how efficiently energy is converted into economic output. The literature suggests that if an increase in energy productivity occurs, a more economic output will be generated with the same amount of energy, consequently leading to reduced waste generation.

Gross domestic expenditure on research and development (R&D), as a percentage of GDP, was collected from Eurostat and plays a clear role in fostering innovation and economic growth. Finally, in order to account for the demographic and economic factors affecting industrial waste generation and sustainability, the proportion of the population aged 65 years and older and GDP in current USD, collected from Eurostat and World Bank, respectively, were considered as key indicators of the national economic performance.

The summary statistics, as presented in

Table 2. Summary statistics. Waste and GDP values are displayed in thousands. The descriptive statistics contain the mean value, the standard deviation, the minimum and maximum value of each variable.

Statistic	Mean	St. Dev.	Min	Max
Waste	183,518	493,387	312.180	2,338,530
HazWaste	7664	19.840	2.649	107,850
NonHazWaste	175,864	473.923	303.826	2,512,550
EnvTaxR	7.159	2.258	1.000	17.000
EnInt	3.990	1.825	1.000	16.000
EnProd	6.537	3.166	1.426	24.453
RD	1.683	0.901	0.340	3.730
Pop	18.158	2.386	7.200	23.080
GDP	46,983,827	656.099	568,315	313,137,776.2

, reveal some initial meaningful insights that shed light on their dynamics and significance. The average waste quantity of all countries is 183 million tonnes, while the minimum and maximum value is 312,180 and over 2 billion tonnes, respectively. The relatively high standard deviation indicates significant variability around the average, reflecting diverse waste production levels within the dataset, as emphasized by its extensive range. For a better understanding of the variability among countries, in Figure 1, this large variability among countries is more evident. The sum of hazardous and non-hazardous waste is equal to the total waste quantity. The summary statistics for the non-hazardous waste are very close to those of total waste. Hazardous waste presents much smaller values, indicating a successful attempt towards their mitigation in Europe. Large-scale differences are also observed in GDP, where the minimum and the maximum values are quite different.

On average, environmental tax revenues (EnvTaxR) constitute approximately 7.159% of the total tax revenue. This average underscores the notable contribution of environmentally related taxes to the overall tax revenue across observations with small variations, with a standard deviation equal to 2.258. Among countries, this indicates that different entities exhibit similar dependencies on environmental tax revenues as a share of their total tax income. However, the range of environmental tax revenues as a percentage of the total tax revenue spans approximately from 1% to 17%, as also shown from their distribution among countries in Figure 2, highlighting the diversity in the dependency on environmental taxation relative to overall tax revenues. In Figure 3, the histograms of the employed variables show that most of them exhibit a noticeable skewness towards the left side.

4. Results and Discussion

The results of the FE regression for the total industrial waste (hazardous and non-hazardous) are presented in

Table 3. Estimation results of FE models for total waste generation. All models include year and country fixed effects. Standard errors are clustered in parentheses. *** $p<0.01$; ** $p<0.05$; * $p<0.1$ Each column represents a different model. The variable of interest is total industrial waste. Model 1 considers only tax revenues as the independent variable. In each model, we test the impact on waste generation by adding explanatory variables. Model 4 contains all the explanatory variables.

Variables	Model 1	Model 2	Model 3	Model 4
EnvTaxR	−1.091 ***	−1.087 ***	− 0.948 ***	−0.852 ***
	(0.187)	(0.186)	(0.185)	(0.179)
EnInt		−0.311	−0.420 *	−0.524 **
		(0.221)	(0.219)	(0.210)
EnProd		0.693 ***	0.550 **	−0.530 *
		(0.248)	(0.246)	(0.274)
RD			0.575 ***	0.470 ***
			(0.125)	(0.121)
Pop				0.054 ***
				(0.022)
GDP				0.778 ***
				(0.101)
Country FE	YES	YES	YES	YES
Year FE	YES	YES	YES	YES
Constant	19.94 ***	18.95 ***	18.63 ***	−0.683
	(0.417)	(0.832)	(0.822)	(2.551)
Observations	646	646	646	646
$R^2$	0.1104	0.1344	0.1626	0.2404

. Each of the four columns of the table displays the results of a specific model that includes different variables. More specifically, model 1 considers only the impact of environmental tax revenues on the total quantity of industrial waste. In model 2, energy intensity and the productivity of the consumed energy are also included as independent variables, while in model 3, expenditure on R&D is incorporated. Finally, in model 4, the variable that accounts for demographic characteristics and the one that depicts a country’s economic performance, namely Pop and GDP, respectively, are integrated.

In model 1, environmental tax is a negative and statistically significant estimator of waste at the 1% level. More specifically, for every unit of increase in environmental tax revenues, there is an associated decrease in waste generation of approximately 1.091%, advocating that higher environmental tax revenues can lead to more sustainable industrial practices [46]. In fact, EnvTaxR presents the same behavior in all four models in Table 3. In model 2, it exhibits a very similar value to that of model 1, implying that industries are motivated to reduce waste generation and benefit from tax incentives. The energy intensity is negative but not statistically significant whilst productivity is positive and statistically significant at the 1% level. This suggests that while a higher productivity may lead to economic growth, it might also result to higher waste outputs. Energy productivity denotes how efficiently energy is converted into economic output and its positive sign deviates from the existing literature.

In model 3, RD is used as an additional variable to the variables in model 2. Once again, taxes remain negative and strongly significant at the 1% level. Energy intensity becomes statistically significant at the 10% level, showcasing a negative impact on industrial waste. In other words, a higher energy intensity is associated with lower waste generation [48], while the EnProd value is similar to the value in model 2. The estimator of expenditures on R&D activities exhibits a positive and statistically significant impact on industrial waste, indicating that higher investment in research and development activities is linked to increased waste generation.

Finally, in model 4 in Table 3, a consistency regarding the environmental taxes on waste generation is observed. The negative effect of EnvTaxR is present, as in all previous models, and is statistically significant at the 1% level. Therefore, the initial hypothesis that environmental taxes and industrial waste have a negative relationship was confirmed in all four models. This means that if environmental taxes increase, then industrial waste will decrease due to their negative relationship. The EnInt variable has a negative and statistically significant impact on waste generation at the 5% level, denoting that a 1% increase in energy intensity will decrease the total waste generation by 0.524%. The coefficient of the variable depicting the productivity of the consumed energy in model 4 becomes negative at −0.530, albeit with lower significance (10%).

The fact that the coefficient was positive in models 2 and 3 but converts to positive in model 4 could indicate a potential non-linear relationship that needs in-depth investigation. Moreover, the RD coefficient exhibits a positive and statistically significant relationship with waste generation. More specifically, a 1% increase in R&D expenditures results at an increase in waste generation of 0.470%, which is close to the value that the R&D estimator had in model 3. Conversely, the results of model 4 indicate that the share of the population that is older than 65 years plays a positive and statistically significant role, albeit the magnitude is very low. This means that a 1% increase in the proportion of the population that is 65 years old and above increases waste generation by 0.054%. In other words, when society is composed of many people older than 65 years, waste generation is larger. Also, the coefficient of GDP reveals the positive nexus between country economic factors and waste generation [34]. More specifically, a 1% increase in terms of the GDP increases the total waste generation by 0.778%, and the result is statistically significant at the 1% level. Nevertheless, information about pricing behavior, about how much of the tax is passed on to consumers and about the price elasticity of demand for relevant products is also necessary for estimating the impact of tax [6,49]. Furthermore, stringent environmental policies, like environmental taxes, can minimize the adverse effects of pollution by promoting environmentally friendly technologies and by discouraging environmentally “dirty” technologies. Strict regulations have the ability to potentially change the behavior of producers and consumers towards eco-friendly production and consumption of energy products [50,51]. (Price elasticity and behavior, social consequences and environmental policies are out of the scope of this study.)

Table 4. Estimation results of FE models for non-hazardous waste generation. All models include year and country fixed effect. Standard errors are clustered in parentheses. *** $p<0.01$; ** $p<0.05$; * $p<0.1$. Each column represents a different model. The variable of interest is non-hazardous industrial waste. Model 1 considers only tax revenues as independent variable. In each model, we test the impact on waste generation by adding explanatory variables. Model 4 contains all the explanatory variables.

Variables	Model 1	Model 2	Model 3	Model 4
EnvTaxR	−1.093 ***	−1.093 ***	−0.946 ***	−0.853 ***
	(0.186)	(0.185)	(0.184)	(0.178)
EnInt		−0.375 *	−0.489 **	−0.593 ***
		(0.220)	(0.218)	(0.208)
EnProd		0.719 ***	0.568 **	−0.515 *
		(0.247)	(0.245)	(0.272)
RD			0.603 ***	0.496 ***
			(0.125)	(0.120)
Pop				0.057 ***
				(0.021)
GDP				0.781 ***
				(0.099)
Country FE	YES	YES	YES	YES
Year FE	YES	YES	YES	YES
Constant	19.93 ***	18.98 ***	18.65 ***	−0.783
	(0.417)	(0.829)	(0.817)	(2.553)
Observations	646	646	646	646
$R^2$	0.1138	0.1422	0.1732	0.2520

presents the results of the FE regression when non-hazardous industrial waste is the variable of interest. Following the same pattern as in Table 3, a distinction between the four models is present. A common finding in all regression models is the strong and negative impact of environmental taxes on non-hazardous waste generation. The consistent negative relationship implies that higher revenues stemming from environmental taxes operate as motivation for industries to adopt cleaner and more sustainable practices. We can further elaborate on this by assuming that businesses that face financial incentives to reduce their environmental impact become more efficient in terms of waste generation, leading to diminished non-hazardous waste outputs. Similar evidence was given in the results of Table 3. Hence, the analysis confirms that environmental taxation can effectively encourage industries to adopt more sustainable practices, resulting in a more positive environmental impact.

Model 2 in Table 4 includes the variables of EnInt and EnProd in contrast to model 1. The results indicate a negative impact of energy intensity on non-hazardous waste, which is statistically significant at 10%, and a positive and strongly significant impact of energy productivity on the dependent variable at the 1% level. Thus, when an increase of 1% occurs, non-hazardous waste also increases by 0.72%, indicating that a higher energy productivity is associated with greater non-hazardous waste generation.

With the inclusion of R&D expenditures in model 3, the variables tested in models 1 and 2 remain almost the same in magnitude and significance. The R&D estimator showcases that higher investments in research and development are linked to increased non-hazardous waste generation. Similar evidence was found when the total waste was taken into consideration. This finding could be attributed to the fact that R&D experiments generate waste.

Finally, model 4 takes into account variables that depict demographic changes and economic factors. Environmental taxes continue to have a negative impact on non-hazardous waste generation. Specifically, a 1% increase in environmental taxes leads to a decrease, of 0.853, in non-hazardous waste. The energy intensity also has a negative impact, which is not only statistically significant at the 1% level, but the coefficient also has the largest magnitude compared to the values of models 2 and 3. Energy productivity becomes positive with a lower significance level (10%). Therefore, a higher energy productivity corresponds to lower waste generation. R&D expenditures remain a positive predictor of non-hazardous waste, as in model 3. The coefficient for population presents a statistically significant relationship with waste generation. A 1% increase in the share of the population over 65 years causes an increase in waste of 0.057%. Lastly, GDP also exhibits a positive and statistically significant relationship with non-hazardous waste. More specifically, a GDP increase of 1% increases waste by 0.78%. This specific result aligns with the common understanding that a growing population and economic growth often correspond to an increased economic activity, which, in turn, can lead to larger amounts of industrial waste.

5. Concluding Remarks

Nowadays, the world is confronting environmental challenges that result in adverse consequences on national economic growth. Waste generation and disposal constitute a cumulative important environmental issue, becoming prominent from both policy and academic perspectives. This study intends to fill a significant gap in the literature regarding the effectiveness of environmental taxes in reducing total and non-hazardous industrial waste by taking into account demographic- and energy-related variables as well as research and development expenditure and establishing a solid framework regarding the waste generation determinants. Moreover, it is the first study that employs countries from Europe covering the 2004–2022 period and empirically investigates the above-mentioned relationship.

Based on the findings of the study, some indicative policy recommendations are proposed in order to help policymakers, researchers and governments reduce industrial waste. More specifically, environmental tax policies adopted in countries across Europe enables them to comply with the “polluter pays” principle and, thus, they diminish the total non-hazardous waste. In this sense, governments should continue and even intensify their environmental taxes policies, as it appears that these actions act as a robust and standard-based tool. Also, the results corroborate a relative non-decoupling situation for the demographic and economic variables. Demographic variables such as population are linked to waste generation and disposal, prompting sustainable management policies. Policies related to improving the recycling infrastructure, like computing facilities, monitoring waste patterns with economic tools, tax breaks for life-cycle assessments and eco-design development that reduce waste, should be at the top of the agenda for policymakers and authorities.

Conversely, the evidence we provide for energy intensity and productivity indicates that we are still far from the turning point needed to achieve absolute decoupling from waste generation. Based on this specific result, a guaranteed price paid by the authorities to producers of renewable energy from waste, adopting a feed-in tariff policy, should be implemented at national, regional and municipal levels, supporting energy conversion from waste transformation. Moreover, authorities and citizens should regard waste as a “new” resource, requiring concentrating more on R&D efforts in order to create technologies for this transformation, as production inputs and energy are more difficult reduce in high quantities. Relatively speaking, the technological context and innovation dynamics associated with waste disposal options are stronger than in landfill management. Thus, the development of related innovative technologies aiming at waste reduction and the creation of waste-to-energy systems is significant. In addition, governments, authorities and individual organizations should also develop information and education campaigns on waste’s environmental impacts, waste reduction technologies and consumer behavior regarding waste generation and treatment.

Finally, several important limitations should be considered. Beyond economic and technological factors, social attitudes and perception influence industrial waste practices. In addition, spatial dependence among regions could result in different findings in this study. The location of waste disposal facilities is a rather critical factor for municipal waste production and recycling. Nevertheless, this paper could become a reliable tool for policymakers and governments and pave the way for future research on waste reduction and sustainability.