Municipal Solid Waste Management services and its funding in Spain

Resources, Conservation and Recycling 107 (2016) 65–72

Contents lists available at ScienceDirect

Resources, Conservation and Recycling journal homepage: www.elsevier.com/locate/resconrec

Municipal Solid Waste Management services and its funding in Spain c ˜ Julián Chamizo-Gonzalez a,∗ , Elisa Isabel Cano-Montero b , Clara Isabel Munoz-Colomina a

Autonoma de Madrid University, Faculty of Business and Economy, Campus de Cantoblanco, 28049 Madrid, Spain Castilla-la Mancha University, Faculty of Social Sciences of Talavera, Av Real Fábrica de Seda S/N, 45600 Talavera de la Reina, Toledo, Spain c Complutense of Madrid University, Faculty of Business and Economy, Campus de Somosaguas, 28023 Pozuelo de Alarcón, Madrid, Spain b

a r t i c l e

i n f o

Article history: Received 27 July 2015 Received in revised form 13 December 2015 Accepted 14 December 2015 Keywords: Municipal Solid Waste (MSW) Public Funds(ing) Sustainability Waste charges

a b s t r a c t Municipal Solid Waste (MSW) generation and management concern many cities. Several implications, mainly resource-consumption, socio-economic and environmental-sustainability, arise. Concurrently, financial-budgetary constraints in some local governments provoke allegations of “misuse” of Wastecollection-treatment-disposal charges and suggestions that they are used mainly to balance budgets. The paper first examines traditional forms of levying charges for Waste-collection-treatment-disposal under the coverage of the Polluter-Pays-Principle in OECD countries and Spanish provincial capitals, finding a prevalence of flat fee systems in Spain. Regarding Madrid specifically, the paper analyses the relationship between its Waste-collectiontreatment-disposal charges and some possibly (in-)dependent variables. Relationships between MSW generated and some potentially-linked variables are identified. Analysis rejects that Madrid waste generation-treatment-disposal charges based on dwelling values had a positive relationship with waste generated (more value of the properties in a district does not imply more waste generated), and reveals/confirms other significant correlations between some variables, it being remarkable that neither age, gender, nationality nor education were found relevant. Conclusions – such as the soundness of the suggested use of the number of dwellings per district as a suitable indicator for the level of waste generated (and its required funding) and the inexistence of a conventional Pay As You Throw system in Madrid – are offered with some policy implications-considerations. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Sustainability is a three-tier concept, “social, economic and environmental”, related to meeting the needs of the present without compromising the ability of future generations to meet their own needs (Bruntland Report, UNWCE, 1987; Machado e Silva et al., 2014; Cardoen et al., 2015). Unwanted impact from waste generation aims to be compensated for by the Polluter Pays Principle (PPP) (UN, 1992; EU, 2008; OECD, 2007, 2013) by means of a charge-tax-fee this charge well might be variable on the weight or volume of waste (Brown and Johnstone, 2014). PPP requires that “the costs of disposing of waste must be borne by the holder of waste, by previous holders or by the producers of the product from which the waste came” (EU, 2008).

∗ Corresponding author at: Universidad Autónoma de Madrid, Facultad de Ciencias Económicas y Empresariales, C\ Francisco Tomas y Valiente, 5, Campus de Cantoblanco, 28049 Madrid, Spain. E-mail addresses: [email protected] (J. Chamizo-Gonzalez), [email protected] (E.I. Cano-Montero), [email protected] ˜ (C.I. Munoz-Colomina). http://dx.doi.org/10.1016/j.resconrec.2015.12.006 0921-3449/© 2015 Elsevier B.V. All rights reserved.

Dahlen and Lagerkvist (2010) suggest the existence of a gap in the relevant literature, disclosing empirical evidence of waste collecting charges applied in Europe. This paper contributes to filling the gap, delivering the findings on the existing funding system in use in the Spanish Provincial Capitals and some additional evidence on waste related variables in the Capital city of Madrid. ˜ et al. (2011) and Spanish Law Echoing Bilitewski (2008), Munoz on waste and polluted soil (2011), it is to be acknowledged that a more than merely adequate costing system is necessary for the optimal implementation of a charging system. Determining the funding of local public services with the lack of a reliable measure of the cost does not inject transparency into the system. Therefore, the possibility of establishing a variable waste charge must be accompanied by the operation of a proper waste management accounting system that provides transparency in the complex world of MSW management and its related costing. Bilitewski (2008) acknowledges the fact that overall, across Germany, 70% of the costs of MSW disposal services are fixed – and so are not dependent on the waste generated by citizens or dwellers. The objectives of this research are to introduce the European and OECD framework on waste charges and then to disclose what

66

J. Chamizo-Gonzalez et al. / Resources, Conservation and Recycling 107 (2016) 65–72

is happening with waste charges in Spain, later to reveal if Madrid waste charges are in any way dependent on the waste generated, to do so will require proof of the existence of a statistical relationship between the value of properties and waste generation and finally to confirm the dependence of waste generation on the number of properties (Household waste generator). In accordance with the objectives later some hypothesis will be formulated. The paper is structured into five main sections. The first section introduces the research issue; the second places it in better focus and context with some key relevant literature that enables the theoretical approach. The third section is devoted to matters relating to the empirical methods-methodology, while the fourth presents findings-results with discussion of the main findings and results, while being supported by an overall summary of key aspects of the paper. The last part includes conclusions and policy implications. 2. Research literature review and hypotheses 2.1. Funding-charging systems Early literature refers to the ancient Roman Empire, to describe a so called “Cloacarium” tax that could well be the oldest referenced waste charge, raised to pay for the maintenance of slaves/criminal convicts doing this labour (Mac Chombaich De Colquhoun, 1851; Barles, 2014). A review of literature on waste funding-charging (Hong and Adams, 1999; Gordon Mackie Associates Ltd., 2007; Bilitewski, 2008; Puig-Ventosa, 2008; Skumatz, 2008; Chamizo González, 2010; Dahlén and Lagerkvist, 2010; Brown and Johnstone, 2014) reveals the application of four main systems of funding waste services. 2.1.1. Undesignated funds system This system is based on funding the service from general city funds with no attempt to relate the cost of the service (either through actual service provided as expressed in either weight or cubic volume of MSW removed). In this system funds when collected (by means of other relevant taxes such as council tax) are not specifically attributed to or connected with MSW. They are placed within the overall general funds and remain undesignated, meaning that all residents are paying for the provision of all local services including MSW services. Evidence of this form of MSW funding was witnessed in virtually all (if not all) the countries considered, Germany (Bilitewski, 2008), France (Le Bozec, 2008), and Canada. The Netherlands, Sweden and Switzerland (Brown and Johnstone, 2014). This system of funding provides no incentive to the citizen to behave accordingly to environmental concerns originated. Within Spain it is apparent in Table 2 below that this system of charging is not prevalent in Spanish provincial capitals with an average of 5.33 out of 52 (10.3%) applying this undesignated funds system. Despite that (essentially) this system of charging prevailed ˜ et al., 2011). in Madrid until 2008 (Chamizo González, 2010; Munoz 2.1.2. Flat fee system The second funding system is the levy of a flat fee for the service again, with no formal attempt to relate to the cost of the service. Such systems do not seek to relate the charges for MSW to the actual waste generated. The flat fee may be a relatively fixed and constant one. This funding system categorises householders and everyone who is included in a given category will be charged with the same amount of money. Councils applying this funding system may settle on a single category or up to seventeen (the case of Madrid). This system is often considered the most easily applicable, as simple as calculating/estimating the global cost of the service of X D , to split up into N taxpayers, so a minimum charge of X/N D per capita).

This form of MSW disposal funding was also seen in virtually all the countries considered. Bilitewski (2008) refers to its existence within Germany, Le Bozec (2008) provides evidence indicating that in 2005 around 10% of the French population paid for their MSW disposal on that basis (including Paris with a population of more than two million inhabitants an Rouen with more than four hundred thousand), and OECD (2013) refers to its presence in Australia, Canada, Chile, France, Israel, Japan, Korea, Netherlands, Spain, Sweden and Switzerland. Kawai and Osako (2013) refer to its presence in Vietnam pointing out that fixed fees do not encourage householders to reduce the waste generated.

2.1.3. Variable (on waste generated) fee system The third funding system is the levy of a fee that is overtly tied into quantifiable aspects of MSW generation (per relevant unit e.g. household). This charging system is also known as a Pay As You Throw (PAYT) system or pay per waste generation. That variability criterion could be the weight of the MSW or its cubic volume or, in some cases, the number of bags/bins. Djemacim (2009) describes the case of three French municipalities (Paris, Rouen, and Besanc¸on) the last one being the only one applying what this author refers to as “incentive fee” based on the number of weekly collected bins. Djemaci discloses that small entities opt for incentive fees whereas the large cities tend to choose a flat tax (flat fee). In terms of the MSW service, in particular, the application of a specific waste charge/fee is a way for certain municipalities to recover the cost of a sustainable provision of the service – shifting the financial burden from the local authority to the consuming citizen-dweller. In countries like USA and Canada, the PAYT system (or a variant of it) has been applied for many years and there is evidence indicating that in 2006, it was in application in over 7100 US communities (United States EPA, 1997; Skumatz, 2008,). Still, there is some evidence to show that such charges frequently fail to cover the full cost of the service – due to the lack of an overt causal relationship between the fee and the cost of the service (OECD, 2013; Brown and Johnstone, 2014). Evidence of this form of MSW service funding (or a variation of it) was witnessed in virtually all countries considered with its most classic expression being seen in PAYT systems. Exceptionally, within the Spanish context, Puig-Ventosa (2008) notes the application of PAYT in a region of Catalonia and the Environment Ministry refers to another experience in the Balearic Islands (MAGRAMA, 2014). In Germany, Bilitewski (2008) identifies its application with the basis of charging being each disposal bag or bin used. Dahlen and Lagerkvist (2010) reveal application in Sweden. Supporting evidence is also seen in the United Kingdom (Dresner and Ekins, 2010). With regard to the United Kingdom it must be said that the practice of charging for the waste services was forbidden (as of 2010) and the referenced cases were pilot experiences. In Sweden as in Spain, MSW management is a local authority duty. Swedish local governments can charge fees for the service but never above cost. While there are several ways to implement these rates within the law, in Sweden the most common form is a PAYT system (based on volume or less commonly weight). Over the decade from 1995 till 2005, only 26 (8.9%) of Sweden’s 290 municipalities had implemented that system or a form of it. Table 1 gives a visual insight as to how intensively variable waste charging prevailed in Europe in 2005. As described by most of the authors referred to, a variable fee system is also a unit price system and hence there must be an individualised per household collection system (e.g. door to door collection) as there is no point in attempting to make an individual charge per household when there is no individual measurement of

J. Chamizo-Gonzalez et al. / Resources, Conservation and Recycling 107 (2016) 65–72

67

Table 1 Waste charges in four OECD countries as declared by householders 2011. Canada PAYT (by weight or volume) Flat charge (includes flat fee + undesignated) No charge (under householder perception) Not known (under householder perception)

Netherlands

6% 51% 31% 11%

12% 71% 3% 12%

Sweden 8% 63% 18% 11%

Switzerland 52% 34% 8% 6%

Source: Brown and Johnstone (2014).

waste generated (Catalonian Waste Agency – Agencia Catalana de Residuos, 2010). 2.1.4. Variable fee system not depending on waste generation The fourth funding system method is the levy of a fee dependent on a measurable value different to the waste generated. Evidence of this form of MSW disposal funding was witnessed in all the countries considered for this paper. Some Spanish evidence points out that in this system when in use the most applied factor was the water consumption per household-dwelling. In the case of the Andalusian City of Seville – Spain the charge is levied upon a variable component relying on water consumption. The tariff of the waste fee/charge in Seville is dependent upon the cubic metres of tap water consumed per dwelling, once the water bill is calculated, and considering only the portion devoted to sanitising (disposal and purifying treatment), a 56% of this sanitising portion is levied as the waste charge to householders. Applications of all four methods (or slight variations of them) were identified in operation across one or more points of time in all the countries considered, OECD (2013) groups the first two categories in the “Flat charge” category, reflecting that status of waste charges by 2011 in eleven different OECD member states, Brown and Johnstone (2014) within the eleven countries selected those four with the relevant variable fee system (Table 1) – called Pay As You Throw (PAYT) – presence (Switzerland 52%, The Netherlands 12%, Sweden 8% and Canada 6%). Remarkably, all eleven except Switzerland have a majority presence of the Flat fee + Undesignated. When waste charging systems are complex and difficult to understand (Bilitewski, 2008) they finally provoke more waste than usual and so tend to discourage sustainable behaviour. Thus, the tariff applied to waste charges should be reasonable and logical while being based on actual data on how the service is being used. 2.2. Waste services and charging in a Spanish context A key piece of relevant legislation related to waste services is the Act entitled ‘Ley 7/1985, de 2 de abril, Reguladora de las Bases del Regimen Local’ (Basic law on local authority regulations).1 Its Article 25 states that local authorities must provide ‘as much and many services as may be required to meet the needs and aspirations of the local community.’ These services include waste collection, its treatment and environmental protection. Spanish legislation identifies the services to be provided by local authorities – one of these services is the collection of waste. The precise form of this service is partly dependent on the number of inhabitants, so that there are increased responsibilities for municipalities with a population of over 50,000 citizens. In such instances, the relevant local authority must also undertake waste treatment and environmental protection. Such responsibilities have been accompanied by intense pressure on all resources – particularly finance. The local government must ensure that citizens receive the service, by means of a physical management (direct management

1

Last updated 30/12/2013, in force since 31/12/2013.

of public services) with public employees and public equipment or a contract with a private entity (indirect management of public services) that will invoice the local government for the use of their own employees and equipment. This has forced many Spanish local authorities to seek out more appropriate, optimal, efficient and effective ways to manage MSW. The specific waste legislation in Spain is the “Waste and polluted soil law” Ley 22/2011, de 28 de julio, de residuos y suelos contaminados, that includes the conceptual framework of waste hierarchy and the polluter pays principle (PPP). PPP is specially set up for domestic waste that requires local governments to calculate a real cost of waste collection, transport, and treatment, including its monitoring and landfill maintenance and closure. Linked to the provision of the services above are financial mechanisms used by the various municipalities. Of particular interest to this paper is the issue of fees.2 Fees are a minor figure in the revenue budget of the Spanish central government, but not in local corporations, where they are very significant. In this (fee) context, the most relevant item of legislation is Law 8/1989 of 13 April, on Fees and Public Prices, wherein it is made clear that finances raised should aim to cover the cost of the service or activity. Accordingly, that statute clearly requires the fee charged to be consistent with the cost of the related service. In Spain in 2009–2014 (Ayuntamiento de Madrid – Agencia Tributaria de Madrid; 2006–2015), the most often applied system was the flat rate fee, as can be observed in Table 2 with an 82.7%. 2.3. Waste charges in a Madrid context Recognising that Madrid is the capital city of the Kingdom of Spain and its most highly populated – with some 3.3 million inhabitants in 2011, it is of special interest in terms of MSW services and the above law. Indeed as one of Europe’s largest cities, one could reasonably assume that the City Fathers of Madrid city council were up-to-date and fully aware of all relevant environmental issues with an EU and-or international genesis. For several years that city funded its MSW service from the general budget (Undesignated funds system). In Madrid, MSW facilities for collecting and treating had adopted most recent environmental requirements (e.g. separate and selective collection bins and trucks; recycling facilities, etc.). However, in 2009 the City of Madrid implemented a general waste fee to be levied on all householders, the fee being based on the municipal (in Spain termed cadastral) value of the residential properties (Chamizo González, 2010) charges ranged in 2014 from four euros per dwelling for the properties with a value below 17,000 D , to 152 D per year for properties with a value between 261,001 and 1,740,000 D , and beyond this 0.000682 D per euro of cadastral value. Several matters of concern arose and, in this context, Batllevell and Hanf (2008) highlight the difficulty of implementing a variable waste charge when one really intends to apply a PAYT approach under the ‘polluter pays’ principle (Table 3).

2

A fee is the most frequent form of a applying a waste charge in Spain.

68

J. Chamizo-Gonzalez et al. / Resources, Conservation and Recycling 107 (2016) 65–72

Table 2 Waste charging systems Spanish provincial capitals 2009–2014. Waste charging system

2014

2013

2012

2011

2010

2009

Average

Flat rate per dwelling by steps (flat fee system) Flat rate per dwelling (flat fee system) General budget (undesignated funds system) Strictly variable Variable-mixed

24 20 4 2 2

26 18 4 2 2

24 19 6 2 1

25 17 5 3 2

24 19 6 1 2

23 19 7 1 2

24.33 18.67 5.33 1.83 1.83

Total

52

52

52

52

52

52

52

% 46.8% 35.9% 10.3% 3.5% 3.5%

Cumulated 46.8% 82.7% 92.9% 96.5% 100.0%

100%

Source: The authors 2015. Table 3 Madrid 2014 waste charge depending on cadastral value of properties. Tariff by Euros of cadastral value of the property To 17,000: 4 D From 17,001 to 26,000: 6 D From 26,001 to 35,000: 10 D From 35,001 to 44,000: 13 D From 44,001 to 52,000: 17 D From 52,001 to 61,000: 23 D From 61,001 to 70,000: 28 D From 70,001 to 78,000: 33 D From 78,001 to 87,000: 38 D From 87,001 to 104,000: 48 D From 104,001 to 122,000:62D From 122,001 to 139,000:76D From 139,001 to 174,000: 90 D From 174,001 to 218,000: 121 D From 218,001 to 261,000: 137 D From 261,001 to 1,740,000: 152 D +1,740,000: 0.000682 D /D cv Source: www.munimadrid.es.

So it was a must to test whether or not the new Madrid waste charge might be firstly waste generated sensitive in any form, secondly a variable system and third and finally a conventional PAYT system. When the Madrid waste charge was checked with the relevant elements of a conventional PAYT charge, it could be realised that the charge to be applied per householder-despite being to some extent waste generation related by means of the householdproperty value-, could not be said under any circumstances to be variable with the waste individually generated. However when globally (the whole city) considered this change in the funding system it may well be a non-conventional approach to a Pay Per Generation system as the tariff of the fee to apply every year is based on the previous year’s generation and cost. A major concern arising between citizens and opposition political parties was the suspicion of being doubly charged, as with the previous system (undesignated) the council tax on properties (the main local tax)3 was the source for funding most of the local expenses including waste services, and the new waste charge did not modify the council tax. Taxpayers tend to believe that they are paying twice for the same service. But not all citizens were aware of the new environmental legislation that has added to the waste services, new tasks such as separate source and selective collection and post collection treatments – like sorting, recycling, energy recovery, sanitised disposal and so on. It was also questioned whether it was fair to charge depending only on the value of the residential properties, regardless of how many people were living in the property, its age or their gender. In 2014 the council had included some exceptions for the social situation of householders taking into consideration circumstances like unemployment, level of income,

3

So Called in Spanish “Impuesto de Bienes Inmuebles de naturaleza urbana”.

or disabilities. In 2015 the waste charge was discontinued and is no longer in use. 2.4. Hypotheses Led by the literature review and given the paper’s empirical focus on the City of Madrid, in order to disclose (in the case of property values) and to confirm relevant relationships of waste generated with other social, economic and environmental variables, the following hypotheses were established and evaluated. With H1 wanting to discover whether or not a waste generationrelated charging (based on linkage between waste generated and value of dwellings.) exists in Madrid since 2009, and H2 aiming to confirm the existing tendency in the literature of the relevance of the number of households-dwellings.4 H1: The amount of MSW generated in Madrid is dependent on the municipal (cadastral) value of properties in a district. H1a: There is a direct relationship between the MSW generated – in absolute terms (Tonnes of waste) – and the basis of the waste fees-charges (Cadastral value of the properties). H1b: There is an inverse relationship between the MSW generated per property – in terms of tonnes of waste/property (dwelling) – and the basis of the waste fees-charges (Cadastral value of the properties). H1c: There is an inverse relationship between the MSW generated per capita – in terms of Tonnes of waste/Inhabitant – and the basis of the waste fees-charges (Cadastral value of the properties). H2: The amount of MSW generated in Madrid depends on the number of properties in a district. H2a: The amount of MSW generated in Madrid – in terms of tonnes generated depends on the number of properties. H2b: The amount of MSW generated in Madrid per property – in terms of tonnes generated/property depends inversely on the number of properties. H2c: The amount of MSW generated per capita in Madrid – in terms of tonnes generated/inhabitant – depends inversely on the number of properties. 3. Research methods and methodology 3.1. Empirical data, evidence, analysis and evaluation This section of the paper is devoted to its empirical contributions in a specific Madrid context. The section first identifies the main elements of data employed in the empirical exercises – i.e.

4 The relevant figures to test hypotheses should be the amount of waste generated, as is the input that local governments must collect, transport and treat. Relevant indicators in use in European countries and proposed by the European Environment Agency however are related to waste generated per capita or per property (Communities and Local Government, 2007; European Environment Agency, 2005), so our waste-related hypotheses are to be tested in absolute terms, per capita and per property/dwelling.

J. Chamizo-Gonzalez et al. / Resources, Conservation and Recycling 107 (2016) 65–72

69

Table 4 Analysed variables – city of Madrid. Absolute terms variables Age 0–19. Population aged 0–19 years old. Age 20–69. Population aged up to 69 years old Age +70. Population aged over 70 years old. Population. Number of inhabitantsa Spaniards. Number of Spanish inhabitants Foreigners. Number of foreign inhabitants Male. Number of male inhabitants Female. Number of female inhabitants Without primary education. Inhabitants without primary education Primary education. Inhabitants with up to primary education Secondary education. Inhabitants with up to secondary education Higher education. Inhabitants with up to higher education Number of dwellings. Number of registered dwellings Cadastral value. Total municipal value of the dwellingsb Tonnes of waste collected. The amount of waste generated expressed in metric tonnes Average calculated variables Average Cadastral value. The total municipal value of dwellings divided by the number of registered dwellings. Tonnes per capita. The amount of waste generated expressed in metric tonnes per inhabitant Tonnes per dwelling. The amount of waste generated expressed in metric tonnes per dwelling Tonnes of waste collected per Euro of value. The amount of waste generated expressed in metric tonnes divided by the cadastral value of dwellings Euros of value per tonne collected. The cadastral value of dwellings divided by the amount of waste generated expressed in metric tonnes a There was a slight difference in total population, and population by gender, age or studies because of people not declaring their age, gender or level of studies as the category “not declared-no data” has not been included in the data analysis. b The cadastral value is applied to properties for tax purposes – mainly the Spanish forms of (US) real estate or (UK) council tax.

the data variables. In doing so it briefly explains from where and how these variables were obtained. Then, it explains the precise statistical tests applied to the variables and to what purposes. This was done in a Madrid context – where data acquisition and analysis were successfully carried out. The available data, despite being gathered in all the city, is per district aggregated data, which limits the number of observations to the twenty-one districts of Madrid, on the other hand, they are said to be all real data, not taken from a survey. The “absolute terms” variables are the raw data extracted from the data bank, with the per capita and per dwelling variables being average values calculated based on other absolute terms variables. An unavailable datum not included in the conducted tests is the waste charge-fee applied per district; this fact has limited the extent of the research.

3.2. Data variables, identification acquisition and analysis. City of Madrid context For analysis of Madrid data, the variables judged (with data being available for all its 21 districts – year 2009) were those disclosed in Table 4. All the population-related variables (age, gender, nationality and level of studies), are social variables used to test whether waste generation might be socially sensitive. The data relating to Madrid – i.e. population, gender, waste collected, fee applied, number of households, cadastral values and similar came from the Madrid Databank (http://www-2. munimadrid.es/CSE6/control/menuCSE), and was available online in Microsoft Excel format. The source for most of the data was the Madrid Statistical databank (Ayuntamiento de Madrid. Banco de datos, 2012). In terms of analyses, by means of the IBM SPSS 19 statistical software, a series of bivariate tests were conducted between the above-mentioned variables and later with those disclosed as relevant and related to the formulated hypotheses, multivariate tests were applied. As a first step towards analysing and being able to appreciate the intrinsic and inherent nature of the variables being considered, they were subjected to a series of (descriptive) tests of normality – prior to their being subjected to appropriate (inferential) statistical

tests. The tests of normality were also undertaken in recognition of the fact that some statistical tests specifically require that the data be normally distributed. The results of the normality tests of Kolmogorov–Smirnov and Shapiro–Wilks performed by the software IBM SPSS 19 showed that all the related variables reflect a normal distribution.

4. Findings discussion The prevalent waste charge system in the Spanish context was the flat fee, with a minority considering variable aspects not depending on waste generation. Flat fee systems do not consider the cost of the service to individually provide, just the amount to globally levy and the criteria to individually apply to taxpayers. For the Madrid context, Table 5 summarises the decision on relevant hypotheses motivated by Pearson’s correlation coefficient and its level of signification. As a result of the exploratory tests conducted the paper reveals statistically significant correlations between some of the selected variables analysed. Related to our research hypothesis H1a the correlation analysis reveals (Pearson 0.488, Sig. 0.025) that we cannot reject that indeed there exists a direct relationship between “Tonnes of waste collected” and the “Cadastral value” of dwellings, that – as stated before – is the base on which is calculated the fee per dwelling, hence the tax applied seems to be directly related to waste generated-collected-treated. This enables us to say that first the Madrid funding system is a “waste generation related system” second that the “Cadastral value” of dwellings in a district could well be – despite the controversy originated with citizens and opposition parties – a suitable indicator of waste generated by the district. The “Tonnes per capita” and “Tonnes per dwelling variable” do not display a correlation and significance level that can suggest a relationship with “Cadastral value”. The literature review provides some evidence to suggest that when fees are levied in relation (weight/volume) to waste generated, then less waste is generated (OECD, 2013), up to 34% in Switzerland, 33% in Korea and 48% in Japan. In doing so it is not feasible to prove that the particular initiative of applying the Polluter Pays Principe (PPP) in Madrid – by means of the 2009 waste charge

70

J. Chamizo-Gonzalez et al. / Resources, Conservation and Recycling 107 (2016) 65–72

Table 5 Relevant correlations and hypotheses. Variable

Number of dwellings Related hypotesis

Tonnes of waste collected H2a Tonnes per dwelling H2b Tonnes per capita H2c

Cadastral value

Pearson correlation

Sig. (bilateral)

Preliminary decission

Related hypotesis

Pearsoncorrelation

Sig (bilateral)

Preliminary decission

0.804** −0.601** −0.473*

0.000 0.004 0.030

Non Rejected Non Rejected Non Rejected

H1a H1b H1c

0.488* −0.429 −0.168

0.025 0.053 0.0466

Non Rejected Rejected Rejected

Source: The authors. Note. Correlations calculated using IBM SPSS 19. All the variables represent values per district, with the 21 districts summarising all the tonnes of waste collected and the number of dwellings of the city of Madrid. The data does not come from a sample or a survey, but the whole city. Hypothesis is NON rejected if **Correlation is significant at 0.01 (bilateral). Or *Correlation is significant at 0.05 (bilateral), Else is rejected.

Fig. 1. The inverse association expressed by correlation coefficients because when the number of dwellings is high the waste collected per dwelling and per capita is usually low.

– works as a result of lack of relevant data about the waste charge applied per district (part of a work in progress) to include in the data analysis. Further results-findings within the specific Madrid context (Table 5), revealed that the identified inverse associations between the “Number of dwellings” on the one hand the (a) “Tonnes per dwelling” (Pearson −0.601, Sig. 0.004) and on the other the (b) “Tonnes per capita” (Pearson −0.473, Sig. 0.030); and the direct association between the “Tonnes of waste collected” with the “number of dwellings” (Pearson 0.804, Sig. 0.000). These results could be strategically employed as they suggest that the “Number of dwellings” well could be a suitable indicator to consider when estimating the waste to be generated per district and the required funding to face the cost. Furthermore, these would suggest that districts with the largest number of dwellings generate less MSW per dwelling. Considering the possible future implementation of a conventional PAYT system in Madrid, it is highly interesting to reveal the (inter-)relationships between several potentially associated variables – most importantly the total amount of waste collected and/or waste collected per capita and per household, with other variables that could be used as an indicator when setting a variable fee that is based on the MSW generated (Fig. 1). When considering links between waste generated (total, per capita and per dwelling) and other social variables such as age, gender, nationality or education the Pearson correlation did not suggest any of them being relevant, hence no further analysis was conducted with these social variables. Following up the previous exploratory correlations analysis a deeper analysis suggests a multivariate focus. Hence, three

Ordinary Least Squares Linear Regression (OLS LR) were conducted including the variables Cadastral Value, Number of dwellings, Tonnes per capita, Tonnes per dwelling, Tonnes of waste collected. With each of them trying to confirm the relationship of the tonnes related variables with the number of dwellings and/or its cadastral value.

OLS LR 1

Relationship of “Tonnes of waste collected” with the number of dwellings and/or its cadastral values. Model 1: OLS, using observations 1–21 Dependent variable: Tonnes of waste collected Heteroskedasticity-robust standard errors, variant HC1 Std. Error t-Ratio p-Value Coefficient 18,981.7 6784.39 2.7978 0.01189** const Number of dwellings 0.407941 0.072106 5.6575 0.00002*** Cadastral value −0.000154588 0.000706366 −0.2188 0.82923 Mean dependent var 56,434.26 S.D. dependent var 17,310.76 2.11e + 09 S.E. of regression 10,832.92 Sum squared resid 0.647547 Adjusted R-squared 0.608386 R-squared 21.91116 p-Value (F) 0.000015 F(2, 18) −223.2764 Akaike criterion 452.5527 Log-likelihood 455.6863 Hannan-Quinn 453.2328 Schwarz criterion

Model 1 suggests that despite the fact that the existing Pearson correlation “Cadastral Value” may not be considered as explanatory of the behaviour of the Tonnes of waste collected (H1a cannot be confirmed).

J. Chamizo-Gonzalez et al. / Resources, Conservation and Recycling 107 (2016) 65–72

Equally Model 1 confirms H2a revealing that the number of dwellings indeed explains the number of Tonnes of waste collected (R-squared 64.76%). OLS LR2

Relationship of the variable tonnes per capita with the number of dwellings and/or the Cadastral Value. Model 2: OLS, using observations 1–21 Dependent variable: tonnes per capita Heteroskedasticity-robust standard errors, variant HC1 Std. Error t-Ratio p-Value Coefficient 0.573559 0.143165 4.0063 0.00083*** const Number of dwellings −2.59954e−06 1.06302e−06 −2.4454 0.02498** Cadastral value 9.03231e−09 7.11079e−09 1.2702 0.22018 Mean dependent var 0.392663 S.D. dependent var 0.149208 0.333021 S.E. of regression 0.136019 Sum squared resid 0.252074 Adjusted R-squared 0.168971 R-squared 4.179709 p-Value(F) 0.032287 F(2, 18) Log-likelihood 13.71506 Akaike criterion −21.43013 −18.29656 Hannan-Quinn −20.75006 Schwarz criterion

Model 2 confirms that H1c has to be rejected, because of nonPearson correlation and not enough significance (22.01% > 5%), but confirms that H2c cannot be rejected (R-squared 25.21%). OLS LR3

Relationship of the variable tonnes per dwelling with the number of dwellings and/or the Cadastral Value. Model 3 OLS, using observations 1–21 Dependent variable: tonnes per dwelling Heteroskedasticity-robust standard errors, variant HC1 Coefficient Std. Error t-ratio p-Value 0.987741 0.178659 5.5286 0.00003*** const Number of dwellings −3.29812e−06 1.45803e−06 −2.2620 0.03631** Cadastral value −4.8046e−09 8.9379e−09 −0.5376 0.59747 Mean dependent var 0.641484 S.D. dependentvar 0.209817 Sum squared resid 0.558415 S.E. of regression 0.176134 R-squared 0.365771 Adjusted R-squared 0.295301 F(2, 18) 2.780484 p-Value(F) 0.088657 Log-likelihood 8.287631 Akaikecriterion −10.57526 Schwarz criterion −7.441695 Hannan-Quinn −9.895198

Model 3 confirms that H1b has to be rejected, because of nonPearson correlation and not enough significance (59.75% > 5%), but confirms that H2b cannot be rejected (R-squared 36.58%). The OLS LR models evaluated disclose that a relationship between number of dwellings and/or its cadastral value as independent variables and Tonnes of waste generated (Model 1), Tonnes per capita (Model 2) and Tonnes per dwelling (Model 3), dismiss the cadastral value as explanatory of any expression of waste generated (Tonnes, Tonnes per capita and/or Tonnes per dwelling). 5. Conclusions and policy implications This paper is just one further step in having a better knowledge of the domain of waste management. In both exploratory and confirmatory terms, it engages with relevant MSW theory based (Polluters Pay and Hierarchy Principles) on economic and financial thinking and similar tools. In so doing, the paper has attempted to make a contribution to the management of economic resources and related unsustainability issues. In part, the paper evaluated/explored, using appropriate data of and from the City of Madrid, as to whether there is evidence that a variable waste charging (and especially PAYT) is in use in that city. The findings within the paper suggest that this is not the case.

71

While not confirming/retaining Hypothesis H1a, results dismiss any relationship between the MSW generated and the base upon which the tariff (fee) for that waste (i.e. the municipal/cadastral value of the dwellings) was predicated. This allows us to conclude that Madrid’s 2009 funding system is not consistent with a variable charge system and consequently lacks the necessary weight or volume elements to be classified as a PAYT scheme and consequently has to be declared as a flat fee, system. In the same way it was discovered that in the case of Madrid, the value of dwellings is not highly related to the waste generated by householders, giving rise to a largely controversial issue. Given that 82.7% of the capital cities of Spain use a flat fee system the findings related to Madrid can be extended to these cities and hence to every city applying it. The PAYT system (or a variant of it) is being more frequently applied within Europe and at present – but not so frequently – within Spain. In Europe and Spain, charges for MSW services are increasingly being applied on the basis of the ‘Polluters Pay Principle’ (PPP) – thus more closely linking the charge for a service (i.e. MSW collection-treatment-disposal) with the costs associated with rendering that service. PPP is frequently seen as an environmental principle implicitly supported within the text of the EU’s Waste Framework Directive (EU 2008). This directive embraced within its scope not only manufacturers and distributors but also household (domestic) waste producers (Gordon Mackie Associates, 2007; Chartered Institution of Waste Management, 2007). Some Spanish local governments have been discreetly employing PPP as a financial management tool and as a justification to enable them to balance their highly strained local budgets, just by making the enforcing of the waste charge, when previously the service was being funded by the general budget, and doing so adding new income items to the budget without changing the service provided and hence relieving the budget pressure. Nevertheless, Price (2001) determined that better results in terms of a reduction of MSW generated were achieved when waste charges were closely linked to the amount of MSW generated. This would suggest that money is a strong motivator for citizens to reduce their individual generations of MSW as opposed to environmental concern (Lakhan, 2015). So even though one might not be entirely certain about the motivation behind the adoption of a waste charge by some Spanish local governments, positive outcomes in both financial and environmental terms appear to be attainable. Aliu et al. (2014) suggest that the simplest is to pay a waste charge, the lower the opposition for implementation, thinking of cities where most householders do not have a bank account or a credit card. In relation with the objective of identifying relevant relationships between the waste generated (and other variables, the results confirm that the “Number of dwellings” is an appropriate indicator to be taken into account for Municipal Solid Waste generation and its funding. Given that the districts with more dwellings generate less tonnes of waste per dwelling, the charge to be applied to these householders should be related to the number of properties in every district, and hence to the density (population and dwellings). Rightly, in all the services provided by the Spanish public administration in general and in particular by the local administration, there is a growing concern for, and recognition of, the ‘full cost’ of these services. We must acknowledge a lack of the cost focus in this paper and we recognise this focus as a new research to be carried out considering the possible relationship between the cost of the waste services and the fee to be levied to dwellers.

References Agencia de Residuos de Catalunya, 2010. Guía para la implementación de un sistema de pago por generación de residuos municipales (Catalonian Waste

72

J. Chamizo-Gonzalez et al. / Resources, Conservation and Recycling 107 (2016) 65–72

Agency. Guidelines for the implementation of a Pay As You Throw MSW charging system). Aliu, I.R., Adeyemi, O.E., Adebayo, A., 2014. Municipal household solid waste collection strategies in an African megacity: analysis of public private partnership performance in Lagos. Waste Manage. Res. 32 (9 Suppl.), 67–78. Ayuntamiento de Madrid – Agencia Tributaria de Madrid (2006–2007–2008–2009–2011–2012–2013–2014–2015). ‘Ranking tributario ˜ de los municipios Espanoles, 2005–2014’. Available at: http://www.madrid.es/ UnidadesDescentralizadas/UDCPortalDelContribuyente/Publicaciones/ RankingTributario/ranking.pdf (accessed 01.12.12). Ayuntamiento de Madrid, 2012. Banco de datos Madrid, Available at: http://www2.munimadrid.es/CSE6/jsps/menuBancoDatos.jsp (accessed 06.06.12) (Madrid data bank). Barles, S., 2014. History of waste management and the social and culture representation of waste. In: Agnoletti, M., Neri Serneri, S. (Eds.), The Basic Environmental History. Springer, Switzerland (Chapter 7). Batllevell, M., Hanf, K., 2008. The fairness of PAYT systems: some guidelines for decision-makers. Waste Manage. 28 (12), 2793–2800. Bilitewski, B., 2008. From traditional to modern fee system. Waste Manage. 28, 2760–2766. Brown, Z.S., Johnstone, N., 2014. Better the devil you throw: experience and support for pay-as-you-throw waste charges. Environ. Sci. Policy 38, 131–142. Cardoen, C., Joshi, P., Diels, L., Sarma, P.M., Pant, D., 2015. Agriculture biomass in India: Part 1. Estimation and characterization. Resour. Conserv. Recycl. 102, 39–45. Chamizo González, J., (PhD thesis) 2010. Aspectos económicos y medioambientales en la Gestión de residuos sólidos urbanos. Calculo de costes e indicadores. El caso del ayuntamiento de Madrid (Environmental and economic aspects in the municipal solid waste management. Cost calculation and indicators. The case of Madrid City hall). Complutense University of Madrid, ISBN 978-84-694-9649-7. Communities and Local Government: London, 2007. National Indicators for Local Authorities and Local Authority Partnerships: Handbook of Definitions. Dahlén, L., Lagerkvist, A., 2010. Pay As You Throw: strengths and weaknesses of weight-based billing in household waste collection systems in Sweden. Waste Manage. 30, 23–31. Dresner, S., Ekin, P., 2010. Charging for domestic waste in England: combining environmental and equity considerations. Resour. Conserv. Recycl. 54, 1100–1108. Djemacim, B., 2009. Public waste management services in France: national analysis and case studies of Paris, Rouen, and Besanc¸on. CIRIEC Working Paper 2009 (02), 1–35, Rouen (Fr). European Environment Agency, 2005. EEA technical report No 1/2005 EEA Core set of indicators, Guide. European Union, 2008. Directive 2008/98/EC of the European Parliament and of the Council of 19 November 2008 on Waste and repealing certain Directives (Text with EEA relevance) Official Journal of the European Union 22/11/2008. Gordon Mackie Associates Ltd, 2007. Direct and Variable Charging for Household Residual Waste – Overview of Key Issues’, Available at: http://www.ciwm.co. uk/web/FILES/Technical/DVC household residual waste 2007.pdf (accessed 08.08.12).

Hong, S., Adams, R.M., 1999. Household responses to price incentives for recycling: some further evidence. Land Econ. 75 (4), 505–514, http://dx.doi.org/10.2307/ 3147062. Kawai, K., Osako, M., 2013. Advantages and disadvantages of a municipal solid waste collection service for citizens of Hanoi City, Vietnam. Waste Manage. Res. 31 (3), 327–332. Lakhan, C., 2015. Exploring the relationship between municipal promotion and education investments and recycling rate performance in Ontario, Canada. Resour. Conserv. Recycl. 92, 222–229. Le Bozec, A., 2008. Implementation of the PAYT system under the condition of financial balance in France. Waste Manage. 28, 2786–2792. Ley 7/1985, de 2 de abril, 1985. Reguladora de las Bases del Régimen Local (Basic law on local authorities’ regime), Article 25. Ley 22/2011, de 28 de julio (2011) ‘de residuos y suelos contaminados’. [Law 22/2011 of 28 April, on waste and polluted soils]. Mac Chombaich De Colquhoun, P., 1851. A Summary of the Roman Civil Law. V. and R. Stevens and Sons, London. Machado e Silva, M.D.F.M., Calijuri, M.L., Ferreira de Sales, F.J., Batalha de Souza, M.H., Sampaio, L.L., 2014. Integration of technologies and alternative sources of water and energy to promote the sustainability of urban landscapes. Resour. Conserv. Recycl. 91, 71–81. Ministerio de Agricultura, Alimentación y Medio Ambiente () Recogida puerta a puerta asociado a un sistema de pago por generación en Esporles. http://www. magrama.gob.es/es/calidad-y-evaluacion-ambiental/temas/prevencion-ygestion-residuos/buenas-practicas/Pago generacion Esporles.aspx (accessed 06.10.14). ˜ Colomina, C.I., Cano Montero, E.I., Chamizo González, J., 2011. Una Munoz propuesta integradora de costes, indicadores y medioambiente. El caso de los residuos sólidos urbanos en Madrid Presupuesto y gasto público (An integrativeproposal of cost, indicators and environment. The case of Madrid Municipal Solid Waste, N 65)., pp. 159–176, ISSN 0210-5977. OECD, 2007. Guidance Manual on Environmentally Sound Management of Waste, Available at: http://www.oecd.org/env/resourceproductivityandwaste/ 39559085.pdf (accessed 30.11.12). OECD,2013. Household waste generation, recycling and prevention. In: OECD, Greening Household Behaviour: Overview from the 2011 Survey. OECD Publishing, http://dx.doi.org/10.1787/9789264181373-11-en. Price, J.L., 2001. The landfill directive and the challenge ahead: demands and pressures on the UK householder. Resour. Conserv. Recycl. 32, 333–348. Puig-Ventosa, I., 2008. Charging systems and PAYT experiences for waste management in Spain. Waste Manage. 28 (December (12)), 2767–2771. Skumatz, A., 2008. Pay as you throw – a tool for urban waste management Pay as you throw in the US – implementation, impacts, and experience. Waste Manage. 28 (December (12)), 2778–2785. United Nations World Commission on Environment and Development UNWCWE, 1987. Our Common Future. Oxford University Press, Oxford, Available at: http://www.un-documents.net/wced-ocf.htm (accessed 22.08.12). United Nations (UN), 1992. The Rio declaration on environment and development. In: The United Nations Conference on Environment and Development, Rio de Janeiro, Brazil. United States Environmental Protection Agency, 1997. Pay-As-You-Throw Success Stories Unites States of America.