BUS606: Operations and Supply Chain Management

From a general point of view, we can observe that ten years after the introduction of the WEEE regulation system, 70% of provinces shows a relevant collection performance being in the HWC state (kgpc ≥ 4). Only 17.3% and 10.9% of these provinces are respectively in Southern and Central Italy. Consistently, the percentage of provinces in LMC and MWC states at the end of the period is low and mainly concentrated in the South and Central area. Moreover, Northern provinces progressively moved both from LWC (2008 = 28.2; 2017 = 0) and MWC (2008 = 13.6; 2017 = 0.9) states state (2008 = 0.9; 2017 = 41.8).

It is worth highlighting the huge growth in the HWC state in the first four years of the period (2008–2011), whereas North, Central and South provinces increase respectively of41.8, 13.6, and 9.1. In particular, by observing the trend of the HWC state, it is worth highlighting the presence of two different dynamics for the three territorial areas. In fact, a first increase is visible for the period 2008–2012, then, after a slight decline, a new increase is registered in the period 2014–2017. Probably the first increase is derived from the impact of the digital switchover, which involved the replacement of televisions sets and the increase of the R3 group of WEEE that includes televisions sets, screens and monitors. While the second is caused by the entry into force of LD 49/14 that receipts Directive 2012/19/EU and produced a new impetus on the collection process. In the following section, both of these regulatory effects are more deeply analyzed.

Transition Probability Matrix Results

In this section, the results deriving from the transition matrix analysis are presented.

Table 5 shows the transition probabilities for all the provinces for the years 2008–2017. At the time of introduction of the WEEE's management system (2008), as expected, the probability of a province being in LWC state was very high (80%). At the end of the period (2017), such probability dropped to 13.6%, while the probability of a province to evolve into the MWC state was pretty high (20.5%), as for moving to HWC (H1WC + H2WC + H3WC = 63.6%).

Table 5. Collection Centers: transition probabilities matrix of Italian provinces, 2008–2017.

2008–2017	LWC	MWC	H1WC	H2WC	H3WC	Total (a)
LWC	14 (15.9%) *	18 (20.5%)	35 (39.8%)	15 (17%)	6 (6.8%)	88 (80%)
MWC	1 (4.8%) **	1 (4.8%)	9 (42.9%)	8 (38.1%)	2 (9.5%)	21 (19.1%)
HWC	0 (0%)	0 (0%)	0 (0%)	1 (100%)	0 (0%)	1 (0.9%)
Total (b)	15 (13.6%)	19 (17.3%)	44 (40%)	24 (21.8%)	8 (7.3%)	110 (100%)

The MWC state was quite stable. The probability to remain in an MCW state is low (4.8%), while it is very high for evolving in the HWC states (in particular, 42.9% in H1WC, 38.1% in H2WC and 9.5% in H3WC). Thus, we can say that provinces that have overcome the initial obstacles in organizing the WEEE collection system have reached relevant performance.

Looking at the initial HWC state, as expected, the probability of being in this state is almost null (0.9%), while the probability to move in the higher states (69.1%) is very high in the period considered. In particular, only one province evolved from H1WC (4.2 kgpc) toH2WC (6.2 kgpc) state after ten years. As well as the probability to remain in the HWC state being 1 (100%), thus this can be considered the sole absorbing state for the period under analysis, even if limited by the presence of only one province.

Carrying out a more detailed analysis (Table 5), we have identified three groups of provinces on the basis of the best and worst WEEE collection performance. Table 6 reports the three groups according to the number of positions gained or lost in the states of the probability transition matrix. The first two groups are the groups of provinces that have collected more than 8kgpc and gained more positions. In particular, the best performing provinces group contains six provinces which have gained 5 positions starting conditional on being in the LWC state. This group includes two Southern provinces (Nuoro an Isernia, this last in 2017 was the best WEEE collector with 14.57 kgpc) and four Northern provinces (among these was Bologna, which collected 12.1 kgpc in 2017 with a growth of 35% compared to 2016). These provinces are referred to as the "most virtuous" ones and considered that their collection performance is more than three times the 4 kgpc EU target.

Table 6. Group of best and worst performing provinces.

The last group includes the worst performing provinces of the class. In 2008, these provinces were in the LWC and after ten years still remain in the same last position. Eleven provinces belong to this group, which are almost all located in Southern Italy. They are labeled as "structurally blocked" because of their consolidated inability to progress in the WEEE collection.

The data in Table 6 highlights the important feature of the transition matrix method which provides the possibility of having information on the dynamics of any specific province in order to identify customized initiatives and policy actions to improve the efficiency of the WEEE collection system.

The next step in the analysis of the transition matrices was addressed to shed light on the reasons of two different dynamics for the three areas in the periods 2008–2012 and 2013–2017. As mentioned in the descriptive results (Section 5.2.), an interpretation of these trends was the impact of two legislative events occurred in the period 2012–2014 on the performance of the WEEE collection system. Table 7 shows the transition matrices for the two sub-periods considered (2008–2012; 2013–2017) the first event is the digital switchover occurred in 2012, which, by law, decreed the technological obsolescence and the replacement of a large part of the installed base of televisions sets. The second event is the entry into force of the Legislative Decree 49/2014 which followed the introduction of the EU directive 2012/19/EU, thus providing a potential impulse to the e-waste collection.

Table 7. Transition matrix in Italy, 2008–2012 and 2013–2017.

The transition probability matrix highlights that the first period is featured by a higher mobility probability since the principal diagonal probabilities are lower when compared with the second period (Table 7). Moreover, when we consider the mean of the sum of probabilities which are over the principal diagonal (which represents the improvement trend), we can see that this mean is slightly higher for the first period (20.1%) than for the second period (15.2%). Furthermore, the probability of moving to the MWC a final time was moderately higher (34.1% vs. 31.3%) for the first period. Similarly, the probability of moving to the HWC a final time, conditional on being in the LWC at initial time, was very relevant for the first period (51.1% vs. 6.3%). Focusing on the MWC state, we can observe that the probability of moving to the HWC a final time, was appreciably higher for the first period (95.2% vs. 51.4%), while the probability of moving back to the LWC a final time, conditional on being in the MWC at the initial time was higher for the second period (+8.6%). Thus, we can conclude that the legislative/technological event represented by the digital switchover has provoked a higher positive impact on the 2008–2012 period when compared with the effect of the legislative event represented by the introduction of the LD 49/2014 on the period 2013–2017.