Identifying the most relevant peers for benchmarking waste management performance: A conditional directional distance Benefit-of-the-Doubt approach

Abstract

The paper demonstrates that the outcome of the robust and conditional directional distance version of the BoD model not only contains information on the overall efficiency score and rank, but also enables a region to identify its genuine peers, disentangle the overall effect of the background conditions on the performance and assess one's relative strengths and relative weakness vis à vis its most relevant peers when evaluating its municipal solid waste management. We illustrate how this information can be retrieved from the outcomes of DEA/BoD based efficiency and performance benchmarking studies by applying the methodology to the case of the Brussels Capital Region. The results show that Freiburg, Darmstadt and the province of Limburg are in general the most interesting peers for the Brussels Capital Region. In addition, the analysis reveals that Darmstadt and Freiburg are identified as the most instructive peers if the region is interested in improving its composting and recycling rate, respectively.

A composite indicator for measuring the environmental performance of water, wastewater, and solid waste utilities

“…In particular, the deterministic nature of the CI leads to three issues: 1. statistical inference is difficult, 2. the scores are sensitive to outliers and 3. to the sample size. To face these problems, we complement the model with a robust and a conditional analysis, by applying the methodology proposed by Cazals et al (2002) and by Simar (2005, 2007) (see also Rogge et al, 2017;Fusco et al, 2020;Lavigne et al, 2019;D'Inverno et al, 2020).…”

“…Then, the mean of the obtained B values of β b,m,z is calculated, and the corresponding Composite Indicators CI m,z is computed by using function (3).The interpretation of the conditional Composite Indicator CI m,z has to go arm in arm with the comparison between this indicator and the robust one, namely CI m . To investigate the source2 The terminology used in the literature is super-efficiency; since we refer to a composite indicator, we prefer to talk of 'performance' instead of 'efficiency, in line withRogge et al (2017) andLavigne et al (2019).…”

Local government size and service level provision. Evidence from conditional non-parametric analysis

“…For the data at hand, imposing such a condition on the drawing does not allow for much variation in the municipal service provision CI scores. Consequently, in the following we propose the same analysis allowing the drawing with replacement from the whole sample (as for example executed in Rogge et al, 2017;Lavigne et al, 2019). Table B.1 shows the estimated municipal service provision CI scores and Table B.2 reports the results of the efficiency analysis including the alternative CI scores.…”

“…In the second case, the more the municipality restricts the sub-indicator the better it is (e.g., the level of crime). To encompass both desirable and undesirable sub-indicators, the model builds on a directional distance function as introduced by Zanella et al (2015, models 7 and 8 p.523) and recently used by other scholars (see for example Rogge et al, 2017;Fusco et al, 2019;Lavigne et al, 2019). As for the second issue, the deterministic nature of the basic BoD model makes each municipality in the sample constitute the performance possibility set.…”

“…With this model specification, a municipality's service provision is not evaluated against the service provision of all the other municipalities in the sample, but rather against municipalities with equal or better performance and with similar background conditions or, put another way, against the expectation of the "best" (Daraio et al, 2020). Following Fusco et al (2019) and Lavigne et al (2019), we can formally define the performance possibility set for the robust conditional BoD model in its directional distance version as follows:…”