![Sample our Urban Studies journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110826/TOC-Subject-Sample-2021-Urban-Studies.jpg"})
Abstract
Abrate G., Erbetta F., Fraquelli G. and Vannoni D. The costs of disposal and recycling: an application to Italian municipal solid waste services, Regional Studies. The paper investigates the costs of waste disposal and recycling services by using a well-behaved Composite cost function model. The estimates on a unique sample of more than 500 Italian municipalities highlight that the refuse collection technology exhibits constant returns to scale as well as scope economies between disposal and recycling. As far as the size of the municipality increases, scope economies rise up to 20%, but they are accompanied by overall diseconomies of scale. The findings suggest that, on the one hand, joint management of disposal and recycling should be encouraged and, on the other hand, that strategies aimed at increasing the share of waste sent for recycling would not imply a considerable increase in total costs.
Abrate G., Erbetta F., Fraquelli G. and Vannoni D. 垃圾处理与回收的成本：意大利市政层级固体废弃物处理服务之应用，区域研究。本文运用优化的复合成本函数模型，探讨废弃物处理与回收服务的成本。超过五百个意大利市政层级特殊样本的评价，突显垃圾收集技术在垃圾处理与回收之间展现出规模及范畴经济收益不变。一旦市政的面积增加，范畴经济将增长至 20%，但却会伴随着全面性的规模不经济。本研究发现建议，我们一方面需鼓励垃圾处理与回收的共同管理，另一方面，强化分担回收废弃物的策略并不意味总成本的显著增加。
Abrate G., Erbetta F., Fraquelli G. et Vannoni D. Les coûts de l'élimination et du recyclage des déchets: étude de cas des services municipaux de gestion des déchets solides en Italie, Regional Studies. A partir d'un modèle Composite robuste de la fonction de coût, cet article étudie les coûts des services d'élimination et de recyclage des déchets. Les estimations sont basées sur un échantillon unique auprès de plus de 500 municipalités italiennes et soulignent que la technologie de l'enlèvement des ordures démontre des rendements d'échelle constants aussi bien que des économies de gamme entre l'élimination et le recyclage. En ce qui concerne la taille de la municipalité, les économies de gamme augmentent jusqu'à 20%, mais elles sont assorties des déséconomies d'échelle globales. Les résultats laissent supposer que, d'un côté, on devrait encourager la gestion commune de l'élimination et du recyclage des déchets et, de l'autre côté, que les stratégies qui envisagent une augmentation de la proportion des déchets envoyée au recyclage n'impliqueraient aucune augmentation importante des coûts totaux.
Abrate G., Erbetta F., Fraquelli G. und Vannoni D. Die Kosten von Entsorgung und Recycling: eine Anwendung auf die kommunalen Dienste für festen Abfall in Italien, Regional Studies. In diesem Beitrag untersuchen wir die Kosten von Abfallentsorgungs- und Recyclingdiensten mit Hilfe eines sich wohl verhaltenden kombinierten Kostenfunktionsmodells. Aus den Schätzungen anhand einer eindeutigen Stichprobe von mehr als 500 italienischen Gemeinden geht hervor, dass die Technik der Abfallsammlung konstante Skalenerträge sowie Verbundvorteile zwischen den Bereichen Entsorgung und Recycling aufweist. In dem Maß, in dem die Gemeinde wächst, erhöhen sich die Verbundvorteile um bis zu 20%, was aber generell mit fallenden Skalenerträgen einhergeht. Aus den Ergebnissen geht hervor, dass einerseits eine gemeinsame Verwaltung der Entsorgungs- und Recyclingdienste gefördert werden sollte und dass andererseits Strategien zur Erhöhung des Anteils des zur Wiederverwertung eingesammelten Abfalls zu keiner beträchtlichen Erhöhung der Gesamtkosten führen würden.
Abrate G., Erbetta F., Fraquelli G. y Vannoni D. Los costes de la eliminación y el reciclaje de residuos: una aplicación en los servicios italianos municipales de desechos sólidos, Regional Studies. En este artículo investigamos los costes de los servicios de eliminación y reciclaje de residuos utilizando un modelo eficaz de función de costes combinados. Los cálculos a partir de una muestra única de más de 500 municipios italianos ponen de relieve que la tecnología de recogida de basuras muestra devoluciones constantes en el rendimiento a escala así como economías de escala entre la eliminación y el reciclaje de los residuos. En la medida en que el tamaño del municipio aumenta, las economías de escala crecen hasta un 20%, pero vienen acompañadas por deseconomías de escala en general. Los resultados indican que, por una parte, debería fomentarse la gestión conjunta de los servicios de eliminación y reciclaje de residuos y, por otra, que las estrategias destinadas a aumentar el porcentaje de los residuos enviados para reciclar no implicarían un aumento considerable de los costes en total.
Acknowledgement
The authors would like to thank two anonymous referees; Thomas Kinnaman, Massimo Filippini; and the participants at the 6th International Industrial Organization Conference (IIOC), George Mason University, Washington, DC, USA, 16-18 March 2012, who provided helpful comments. The financial support of HERMES is gratefully acknowledged. The usual disclaimer applies.
Notes
1. For example, see Kinnaman (Citation2005, Citation2006). In particular, Kinnaman Citation(2005) tries to understand why municipalities are operating cost recycling programmes designed to reduce the external costs of refuse disposal. The results pointed towards the presence of altruistic tastes for recycling on the part of households, so that policies aimed at setting specific recycling goals might be expensive but not effective at reaching the required target.
2. For a comprehensive survey on the first issue, see Bohm et al. Citation(2010); and for updated references on the second issue, Bel et al. Citation(2010).
3. See Reeves and Barrow Citation(2000) for Ireland, Ohlsson Citation(2003) for Sweden, Dijkgraaf and Gradus (Citation2003, Citation2007) for the Netherlands, Bel and Costas Citation(2006) for Spain, and Sorensen Citation(2007) for Norway.
4. For another example of a paper dealing with input prices, see also Ohlsson Citation(1996).
5. For example, Hirsch Citation(1965), working on a sample of twenty-four cities and municipalities in the St. Louis area in 1960, suggested the presence of constant returns to scale. In a similar vein, Stevens Citation(1978) estimated a Cobb–Douglas cost function (including the price of labour among the regressors) on a sample of 340 US public and private firms, and found that while private operators were better performers, economies of scale were exhausted at population sizes above 50 000 inhabitants.
6. The authors presented estimates where input prices were included among the regressors, too, but the usual microeconomic theory properties ensuring well-behaved cost functions (that is, Shephard's Lemma, homogeneity of degree one in input prices, concavity) where neither imposed on the estimation nor checked after having estimated the model.
7. Unfortunately, the authors were not providing estimates of scale and scope economies for different output levels and for different combinations of outputs.
8. In the author's own words:
Therefore, the data show no evidence of the existence of cost complementarities (which are a sufficient condition for the presence of scope economies), even if this does not imply that scope economies are absent.
9. Cost-shares are computed as:
By Shephard's Lemma as:
where Xr is the input demand for the rth input, so that:
10. For a twice continuously differentiable cost function, cost complementarities are present at Y′ if:
for all Y′ ∈ [0,Y]. Cost complementarities between two products imply that the marginal cost of producing one output decreases as the quantity of the other good is increased. Baumol et al. Citation(1982) have shown that a multi-product cost function characterized by weak cost complementarities over the full set of outputs up to the observed level of output exhibits scope economies.
11. For more details on CS-type models and for some applications to the cost analysis of multi-product firms, see Piacenza and Vannoni Citation(2004) and Piacenza et al. Citation(2010).
12. Unfortunately, data limitations prevent the authors from disentangling corporations which are owned by private operators from limited companies whose shares are still in the hands of the local governments. Therefore, the subsequent cost analysis will not be able to separate the effects of corporatization (that is, the transformation of the juridical form without implying a change in the ownership) from the ones stemming from privatization. For an attempt to measure the impact of corporatization on the costs of a sample of Italian local public transport firms, see Cambini et al. Citation(2011).
13. For example, official data report that, in 2005, 11% of the population (22% in the South) was receiving refuse collection services from municipalities by means of in-house arrangements. In the same year, the share of recycling over total refuse collection was 24% (38% in the North, 19% in the Central Regions and 9% in the South).
14. Consistently with the large majority of empirical papers in this field, this paper relies on municipal data. It must be acknowledged that the reported cost data might in principle overstate the actual costs in the case in which the local public administrations are contracting with private firms for the provision of the service. Stevens (Citation1978, p. 441) tackled this issue and argued that the cost approach could be relied on (see also Carroll, Citation1995, p. 219; and Hirsch, Citation1965, p. 91). This issue should, however, be mitigated by the completion of the reform geared to the exclusive assignment of the service through competitive tendering procedures.
15. A previous version of the paper followed Antonioli and Filippini Citation(2002) in assuming that the price of fuel was the same for all municipalities in the sample. As requested by a referee, data were collected on provincial fuel prices, so that this revised version can now present more precise estimates of cost-function parameters. The authors thank the anonymous referee for his valuable suggestion.
16. The Translog model is estimated with NLSUR so that it is straightforward to make comparisons with the Composite model. However, the TS model was also estimated by using iterated generalized least squares as well as maximum likelihood estimators. As expected, the results are virtually unchanged across the three different estimation procedures.
17. A similar pattern can be observed by comparing the estimated sums-of-squared errors (SSE) of the cost and input-share equations.
18. The average municipality (the point of normalization) corresponds to a hypothetical council operating at an average level of production for each output and facing average values of the input price variables.
19. The results of the Translog specification show the presence of cost complementarities, since the coefficient on the parameter αDR of equation (1) is negative (–0.12) and statistically significant.
20. The presence of constant returns to scale for relatively small municipalities is consistent with Carroll Citation(1995), who was using a sample of municipalities of an average population size of 26 284 inhabitants. In addition, the finding of scale diseconomies in correspondence with higher output levels is consistent with the analysis conducted by Antonioli and Filippini Citation(2002), as far as disposal is concerned, and with the outcomes obtained by Bohm et al. Citation(2010), as far as recycling is concerned.
21. For example, when moving from λ = 1 to λ = 8, costs increase by a factor of 9.2 for disposal and by a factor of 11.1 for recycling.
22. The results of the Translog specification are remarkably similar also with respect to the estimates of scale economies for municipalities larger or smaller than the sample average.
23. For municipalites with a population above 100 000 inhabitants, ShareR is on average 18% with a maximum value of 42%. Therefore, particular caution should be used when interpreting results for large councils (that is, when λ = 4 and λ = 8), because the curves in partially rely on out-of-the-sample simulations.
24. As stated by the authors:
Household source reduction efforts presumably complement recycling practices. Households that increase recycling may simultaneously seek ways to reduce the use of shopping bags and beverage containers.
25. The estimates of the TS model, which are available from the authors upon request, are very similar. Similarly to what has been found for the baseline model, moreover, the CS cost function turns out to be the specification that fits the data better.
26. Unfortunately, data limitations prevent other variables that have been used in the literature – such as the frequency of pick-ups, the distance to landfills, the level of tourism activity, and so on – from being included.
27. The authors have used also the number of homes per km2, or the number of buildings per km2, as alternative measures, obtaining identical results.
28. While data limitation prevent the authors from disentangling the three subcategories of Corp, official data report that in 2005 11.1% of the Italian population was served by municipalities through in-house arrangements, 58.5% by state-owned firms, and only 30.4% by private operators. Therefore, a large part of municipalities classified as Corp (66%) organize refuse collection by relying on publicly owned firms.
29. For example, Chiades and Torrini Citation(2008) showed that in the Southern regions of the country, the population served with in-house arrangements has reduced from 11 million to 4 million persons between 1998 and 2007. The authors found, for a sample of Italian municipalities, a negative impact of In house arrangements on costs, too.
30. Bel and Fageda Citation(2010), when working on Spanish data, argued that inter-municipal agreements could be used as a way to reach scale economies for relatively small municipalities; while Sorensen Citation(2007), when working on Norwegian data, underlined the difficulties of managing the service when the ownership was very dispersed, as in the case of inter-municipal joint ventures. Consistently with Sorensen's analysis, Garrone et al. Citation(2010) found for a sample of Italian utilities operating in gas, water, electricity and refuse collection in the years 1997–2006 a positive and significant impact of a proxy of Intermun on total costs.
31. See notes 12 and 28.
32. Compare, for example, the comments offered by Bohm et al. Citation(2010) and Callan and Thomas Citation(2001), who all found a positive coefficient on Density (measured as persons per square mile and the number of homes per square mile, respectively).
33. For instance, the presence of narrow streets may reduce the ability to use large, specialized equipment. In addition, the extent of on-street parking may involve difficulties in using some automated machinery, with the consequence that operators are forced to use more manual labour.
34. For example, consider a municipality localized in the North, which is observed for the year 2004 and which takes on the organizational form Corp (that is, both Intermun and In house are set equal to zero). For such a municipality type, scope economies are estimated at 0.017 when λ = 0.25 (that is, YD = 4281, YR = 943, Medium = 0, Small = 1 and Density is the average density for the subsample of small-sized municipalities) and at 0.092 when λ = 4 (that is, YD = 68 488, YR = 15 080, both Medium and Small are set equal to zero, and Density is the average density for the subsample of large municipalities).
35. Moreover, the estimates of scale and scope economies do not change if one considers the years 2005 or 2006 instead of 2004; cities localized in different regions of the country (that is, in the Centre or in the South), or inter-municipal partnerships or in-house provision as organizational forms.