Affidavit of Mustan Lalani

Transcription

1 ARBITRATION TO DETERMINE THE 2014 STEW ARD OBLIGATION FOR THE BLUE BOX PROGRAM BETWEEN: ASSOCIATION OF MUNICIPALITIES OF ONTARIO and THE CITY OF TORONTO Applicants - and - STEW ARDSHIP ONT ARIO Draft 2: Thursday, March 27, 2014 Respondent Affidavit of Mustan Lalani Table of Contents Introduction... 1 I have not been neutral... 1 My Background and Expertise... 3 MIPC... 3 KPMG Model Cost per tonne... 5 Updating the KPMG Model Baseline Cost Model... 9 CONCLUSION Introduction 1. My name is Mustan Lalani. I am a management consultant focused on providing strategic and analytical support for the development of industry product stewardship solutions to the consumer packaged goods industry. I have an education in statistics and risk management. I am employed by Reclay StewardEdge Inc. My CV is enclosed. 2. I will be a witness at this arbitration because I have been legally compelled to do so by summons. I am not retained or paid to testify by either municipalities or stewards. I have not been neutral 3. I take no position on the amount of the 2014 Steward's Obligation, but I am not neutral on the subject of Blue Box waste funding, as between municipalities and stewards. By

2 stewards, I mean the obligated companies that supply packaging and printed paper into these markets. Generally these would be the brand owners or first importer of the product into the market. 4. From June 13, 2008 until approximately September 2012, I was retained by Stewardship Ontario (SO), through StewardEdge to model, analyse and negotiate the amount of the Steward's Obligation paid each year by SO to municipalities for Blue Box funding. 5. In 2010, the StewardEdge program management contract with SO was terminated, though I and several of our staff continued to provide some Blue Box program management functions and consulting services under contract to SO until approximately September In that capacity, I reported directly to Guy Perry, and indirectly to Derek Stephenson. I understand that Mr. Stephenson will also be providing an affidavit in this matter. 7. Our team negotiated each year, on behalf of stewards, with municipalities, on many topics, including the total amount on an annual basis to be paid by stewards for Blue Box programs. 8. As program managers, we kept SO informed of all our work, and obtained SO board approval whenever required. We always advised SO that they should achieve a negotiated agreement with municipalities prior to setting the fees for each program year and that the outcome of binding arbitration as an alternative to determining these costs would be uncertain. 9. We were successful, on behalf of stewards, in negotiating substantial payment containment agreements with municipalities until StewardEdge stopped representing stewards on Blue Box negotiations in Throughout those years, we were tasked with keeping the cost to stewards of managing waste packaging and printed papers at the lowest possible level. Once the BBPP was developed and approved, we also strove to ensure that the stewards met their obligations under the BBPP, including achieving the recycling targets approved under the plan. 11. My supervisor, Guy Perry, instructed me that stewards were only obliged to pay municipalities 50 percent of "best practices cost" as stewards' share of Blue Box funding. I then read and made my own interpretation of the Cost Containment Plan (CCP) and agreed that this was what it said. 12. When I began work on Blue Box funding, they also instructed me that "best practices costs" meant the cost estimate produced by the computer model (K.PMG Model) set out in Volume 2 of the Operations Improvement: Blue Box Program Enhancement and Best Practices Assessment Project Report, (KPMG Report). I did not question this instruction. 13. I later developed and ran other computer models to calculate what SO called "best practices costs" for municipal Blue Box programs. 14. I continue to be retained by stewards and industry trade associations on waste diversion and product stewardship related matters such as serving as the project manager for a North American based packaging supplier whose objectives are to improve the recovery ofio 2

3 of their packaging formats. I also serve as a technical lead on executing the Canadian Beverage Container Recycling Association (CBCRA) strategic plan in Manitoba. I have increasingly been supporting multinational consumer packaged goods companies on improving recovery of packaging waste in developing and emerging markets globally. My Background and Expertise MIPC 15. I have worked for the same company for almost six years with increasing levels of responsibility. My education and training was at the University of Toronto specializing in Actuarial Science and Statistics. I worked for a short period as an actuarial student at Mercer Human Resource Consulting and have experience with developing statistical and quantitative computer models. 16. On SO's behalf, I sat on the Blue Box Municipal Industrial Program Committee (MIPC) from June 14, 2008 until March 30, 2012, and the Best Practices Committee during the same time period. 17. I was one of four full-time paid consultants working for Stewardship Ontario (SO) on developing, analyzing, presenting and negotiating Blue Box matters with municipalities. 18. SO was able to dedicate significant resources to identifying and promoting what it considered to be recycling "best practices" and to aggressively negotiating lower costs on behalf of stewards. Municipalities generally lacked the resources or the expertise available to the stewards in these negotiations, and we were generally able to achieve annual cost agreements that were less than 50% of municipalities reported costs. 19. Municipalities did not bring any equivalent resources to the table from 2008 until October All municipal MIPC members were volunteers, sitting on the committee in addition to their regular jobs. To the best of my knowledge, no one on the municipal side was a statistician. The municipal volunteers did not have the same capacity to obtain, analyze or present data that SO representatives did. 20. Municipal representatives were generally anxious to show that they were reasonable and willing to compromise. Their negotiations were hampered by the lack of continuity, as municipal representatives to MIPC and to the WDO board had substantial turnover. They therefore tended not to challenge the way SO said things were to be done, or WDO. They did, however, emphasize in each annual negotiation that they believed that the WDA required stewards to pay 50% of their actual BB program costs and that any reduction was a temporary concession to ensure the continuation of the BBPP for the long term. 21. After October , municipalities had a Municipal Blue Box Program Coordinator, who worked on similar issues, named Alec Scott. Mr. Scott did not have a role equivalent to mine, as he was not a voting member of MIPC. KPMGModel 22. The KPMG Model was written in Microsoft Excel 2003 with simple and understandable formulas. The Microsoft Excel file containing the model was transferred to me when I OF JQ 3

4 began work on Blue Box funding, and I was responsible for updating the model on an annual basis from 2009 to The KPMG Model produced a benchmark cost number on the basis of available, however, often incomplete data. It did not claim to be a rigorous engineering analysis or a statistically significant mathematical analysis. 24. For collection costs, KPMG: a. divided municipal Blue Box programs into nine, roughly similar groups or bands. These municipal groupings were later adopted by Waste Diversion Ontario and MIPC; b. picked one of the best performers (based on highest recovery rate and lowest net cost per tonne recycled) in each band as the "analogue" for that band; c. used verified reported cost data from the Datacall, but mainly from the single analogue program in each band; d. "normalized" the verified, reported data, i.e. removed any municipal-specific considerations that were not representative of the municipal grouping. For example, one year the Regional Municipality of Durham had low processing costs on a per tonne basis because of the amortization of the capital equipment in their recycling facility. This was deemed not representative of the rest of the municipalities in Durham's municipal grouping, so there were costs added to Durham's costs to be more representative of the municipal grouping; e. "extrapolated" the "normalized" data to all programs within the band. 25. For processing costs, KPMG developed a cost curve from MRF capital and operating costs from a WDO funded study completed by Entec Consulting in 2001 which considered actual processing operations from the 1990's and early 2000's. The cost curves were used to model processing costs where none were reported, on the basis of a 40,000 tonne per year minimum feasible operation. Processing data from the analogue community was used if it aligned with the cost curve. If it did not align, the cost curve data was used instead. 26. Based on that professional judgment, the KPMG Model produced one or more estimates of costs that would have been incurred, in the entire province, if every municipal Blue Box program had had as effective and efficient a program as the analogue program in their band, and if their processing costs had aligned with the constructed cost curve. 27. The KPMG Report suggested that if all municipal programs adopted "best practices'', and had access to necessary funding for capital improvements, the cost per tonne incurred by all municipalities would gradually approach the cost per tonne incurred by the analogue programs although only after the funding was retired. 28. There is also an inherent tension within the idea of "best practices" - there are simultaneous goals to increase diversion and decrease net cost per tonne. Increasing diversion tends to drive up cost per tonne because the more complicated and expensive OF JQ 4

5 materials are captured (e.g. increased volumes of lightweight packaging being collected and processed). 29. Under the KPMG Model: a. If municipal programs adopted best practices, and became more efficient, and b. If all other factors (including the waste stream mix and commodity prices between all municipalities within each municipal grouping) stayed the same, any gap between the model results and verified municipal reported costs (the Datacall) would likely decrease with time. 30. The KPMG Report did not conclude that all municipal programs could or should have achieved the cost per tonne incurred by the analogue programs, in the year of the relevant Datacall. 31. Every Blue Box program has unique features, and there are wide variations within the bands. Groups 6 to 9, in particular, contained municipal programs with very different demographic, geographic, and economic circumstances that affected their cost per tonne. 32. For example, Barrie and Thunder Bay are both in the same municipal grouping because they have similar population densities and curbside recycling service. Both are small cities, but as an example, the Blue Box cost per tonne is typically much higher in Thunder Bay than in Barrie due to longer collection and hauling distances and population densities. This is not a measure of the effort or attention by the program managers in Thunder Bay. Thunder Bay will always be a long way from major population centres and robust markets for recyclable material. 33. The KPMG Report stated that the City of Toronto was unique, and not analogous to any other municipal Blue Box program. However, it was included in Group 1 in the KPMG Model. 34. The KPMG Model did not measure, or evaluate, why other municipal programs within each band had higher costs per tonne than the analogue program, in the year of the relevant Datacall. 35. The KPMG Model did not measure, or evaluate, whether other municipal programs within each band could have sustainably operated at the same cost per tonne as the analogue program, in the year of the relevant Datacall. 36. The KPMG Model did not measure, or evaluate, what other municipal programs within each band would have had to have done to operate at the same cost per tonne as the analogue program, in the year of the relevant Datacall. 37. The KPMG Model was not designed to be, and was not, a measure of mismanagement by municipal governments. Cost per tonne 38. The KPMG Model measured "efficiency", defined as net cost per tonne of recovered marketed material. OF 10 5

6 39. Cost per tonne is affected by numerous factors that vary from one municipality to another. For example: a. Gross cost per tonne tends to be higher in municipalities with low population densities, narrow streets or remote locations. b. Cost per tonne is strongly affected by the mix of materials collected. Simple, heavy, easily recycled materials such as newspaper and cardboard have a low cost per tonne. Lighter, more complicated materials such as plastic packaging have a higher cost per tonne. All things being equal, municipal programs that accept more light weight materials, may increase their diversion rate, but will have a higher net cost per tonne, provided that the revenue from marketing the baled material was low. Updating the KPMG Model 40. No computer model is perfect. It is, by definition, a simplified approximation. Like any computer model, the accuracy, reliability and relevance of the results of the KPMG Model are determined by the data put into it (inputs) and the calculation method or algorithm. 41. The Model was very sensitive to the precise inputs chosen for key assumptions, particularly the costs and recovery of the nine analogue communities and the investments (or lack thereof) from the nine analogue communities in a particular year. 42. Parts of the KPMG Model (focussing on Blue Box waste collection) were designed to be updated each year with data from the previous year's Datacall. 43. Other parts of the KPMG Model (focussing on Blue Box waste processing) were designed to be updated each year by adjusting the 2005 data. It did not incorporate any actual increases in processing costs experienced by municipalities, even the "analogue" municipality. Instead, the Model used a cost curve constructed based on studies done in 2001 and updated to 2006 presumed values using inflation indicators, a blend of 50% Fabricated Metal Products (now CANSIM Table ) and 50% Non Residential Building Construction Price Index - Warehouse - Toronto Total major sub-trade groups (now CANSIM Table ) (the KPMG factors). 44. We did not do any formal studies to determine how closely changes in Blue Box processing costs tracked the inflation indicators used in the KPMG factors; however Mr. Scott and I often compared and rationalized municipal processing costs to the cost curve based on verbal accounts with municipal staff on their actual operational costs. We concluded at the time that the cost curves were not considerably off the actual operational processing costs that were reported by municipalities into the WDO Datacall. 45. There are many possible reasons why Blue Box processing costs might go up or down at a rate different than the KPMG factors. As an example, the cost curves did not take into account contract renewals and price escalators often levied by waste management service providers. Also, the cost curves made some assumptions about the level of amortization of the processing equipment which could be inconsistent with actual amortization periods from municipal operations. ofio 6

7 The KPMG model always calculated gross costs, not net costs. KPMG explains in their volume 2 report that they were unable to determine a mechanism for determining best practice revenue. 47. In my first year, 2008, I assisted Ms. Liz Parry in populating the collection portion of the KPMG Model with data from the 2007 Municipal Datacall for the previously selected analogue programs. I adjusted the processing portion of the KPMG Model by using the processing cost curves and escalation factors described in paragraph 43 above. 48. Volume 2 of the KPMG report had treated Toronto separately and had included a "normalization" step to remove unique factors and to reflect Toronto's ability to leverage economies of scale which other municipalities were unable to leverage. 49. As there was uncertainty as to which inputs should be used in the KPMG Model, different plausible inputs were tried. This produced a range of $140.3 M to $168.8M. 50. The total verified reported net cost incurred by municipalities in 2007 was $145,888,349, which fell within that range. 51. For the 2009 Obligation, stewards agreed to pay municipalities their 2007 verified reported costs. 52. MIPC did not want to repeat the negotiation process from 2008 which resulted in unproductive discussions about inputs and parameters. Therefore for the 2010 Stewards' Obligation, SO felt that they were paying municipalities too much for their Blue Box programs, and wanted to find a way to reduce the Obligation, and municipalities felt that the KPMG best practice model was no longer linked to real world operations. 53. Despite extensive efforts to assist and encourage higher cost municipal systems to become more efficient, with considerable success, SO remained concerned that municipalities were inefficient, that inappropriate costs were being reported to the Datacall, and that some of the money being paid to municipalities was being wasted. 54. The following year in 2009 MIPC eliminated the range produced by the KPMG Model and generated a single number instead. In the winter of 2008 to 2009, I sat on a working group, the Best Practices Costs Committee, to develop ONE set of major inputs to the model, namely: a. The useful life of a Blue Box container; b. The amortization period for recycling equipment; c. The preferred municipal spend on promotion and education to citizens on a per household basis; d. The cost reduction for municipal programs employing state of the art processing infrastructure and equipment; OF JQ 7

8 e. The cost of capital (or borrowing rate) for determining the interest on municipal capital; f. Etcetera These inputs were used, unchanged, until Based on my professional judgement, we did go to considerable effort to collect realworld data on which to base these numbers, but the available data were not proven to be statistically significant. 57. The numbers chosen were intended to be estimates of provincial averages. Based on my analysis, they were negotiated between the municipal volunteers and the paid steward representatives on the Best Practices Costs Committee. At the time, the municipal representatives did not have the capacity to run the Model for themselves. 58. I then applied these single inputs for each of the key variables in the model to the costs reported by each "analogue" community in a band. 59. With all these adjustments, the amended KPMG produced a calculated "best practice" cost for 2010 that was, for the first time, below the verified reported gross cost. 60. MIPC negotiated and agreed to split the difference. That is, stewards agreed to pay a steward obligation for 2010 of $83,492,926 based on a negotiated gross cost of $267,114,749, midpoint between the reported gross cost of $274,245,169 and the model gross cost of $259,984, The next year, MIPC understood that the KPMG best practice model was reaching its shelf life, and a number of MIPC members were unhappy with the results of the previous year's negotiations. The stewards wanted a bigger delta (gap) between municipalities' verified reported net costs of and the stewards' obligation. Municipalities wanted to be measured against a more representative sample of their members, and not simply one municipal program per "municipal group". That is, they expressed the view that it was not enough to use data from nine individual programs to represent the entire set of municipal Blue Box programs. 62. MIPC then redirected the best practices committee to come up with a principled based new computer model to replace the KPMG model. By this time, municipalities retained a paid staffer, Alec Scott, who worked with me to develop a new model, sometimes called the "best practices model" or "the baseline cost model". 63. Instead of selecting one high-performing analogue community in each band, attempting to normalize it using the KPMG processing cost curves, and extrapolating its results to all programs in the band, this model accepted collection cost data from the top 50 percent of municipal Blue Box programs, defined in terms of: a. cost per tonne collected and marketed b. the number of households participating, and c. best practices scores calculated from the datacall. oflo 8

9 This eliminated the actual collection costs incurred by half of all municipal programs from the model, regardless of why they were incurred. 65. For processing costs, the baseline cost model continued to apply the KPMG factor to the 2005 processing costs curve, and included only reported municipal costs that were below the cost curve value. 66. MIPC (both stewards and municipalities) agreed to the conceptual outline of the model on June 29, 2010, before the calculated outputs were available. Both stewards and municipalities agreed to the conceptual outline without inputting data to eliminate the possibility of negotiating on the "output (i.e. cost)" rather than negotiating on the principles and assumptions. 67. The calculated output was $264,059,308, $20,467,661 less than the verified gross cost. 68. The negotiated result was a gross cost of $269,313,917 resulting in a steward obligation of $91,840, The 2011 negotiations were very acrimonious to the best of my knowledge, although I could not participate fully in the negotiations due to a family emergency which required me to withdraw from the negotiations. I still however, provided counsel to new SO fulltime staff (namely Alex Chan and Daniela Fernandez) who populated and calculated the model as we had in 2010, with some minor changes. 70. The baseline cost model was not used to calculate the Stewards' Obligation in that year. To the best of my knowledge, the Obligation was determined by oral instructions from the Minister, as reported by Lyle Clark, that stewards' fees could not go up. This occurred shortly after public protests had led the Minister to cancel eco fees on household hazardous waste. 71. I was not involved in the 2012 or 2013 negotiations and continue to not be involved in the MIPC Blue Box program cost negotiations. Baseline Cost Model 72. The computer model that Mr. Scott and I developed was not designed to, and does not, represent the costs that municipal Blue Box programs could realistically have achieved in the year of the relevant Datacall. 73. Because of the substantial variability among Ontario municipalities and their Blue Box programs, the Baseline Cost Model will always produce a calculated figure substantially less than the verified costs incurred by municipalities. The gap between the two figures is likely to be greater when there is greater variability in costs, recovery and Best Practice scores within each municipal grouping. This will occur because every municipality was benchmarked against the top 50% of municipalities within a "band". ofio 9

10 CONCLUSION 74. I understand that this affidavit is sworn evidence to be offered to a legal tribunal deciding a question of great public importance. I have done my best to make this evidence as accurate and as truthful as I can. I intend to be bound by it. WITNESS Sworn or Affirmed before me On March 27, 2014 at the City of Vancouver in the Province of British Columbia ~Jf.i.& l\i1ustan Lalani A Commissioner, etc. David Watts, Notary Public West Hastings Street Vcmcouver, BC V6B 1 N2 Tel: Fax: QE.Yl:J (c0davi dnotary.com My Commission Docs No1 Ex~:i; ( OF 10 10