, ,f f Filr· "l'l ,. . ~i;~•?r ( ·T-91 ( L ~. :_/ ,_,4,j_) . Erh1b t 1 l: ! \J-0 -- '-A_ '..4.,.w l ;;;~ti 1.1 o>,.?ce S-Q - CT-91/2 0 l Dtpo,if: le/\ / , / Ae;;etru J;!V\il t) f <) ·. , 7 -nm COMPETITION TRIBUNAL · Grettie~ p . _ . r · ·- ' _ 1 ~ c/ '--..iM1-L-"1.· IN THE MATI'ER. OF an Application by the Director of . '- Invesngattbttltlld Research under s. 79 of the Competition.A.ct, R.S.C:, 1985, c.. C-34, as amended; AND fN THE MATIER OF certain practic~ by Laidlaw Waste Systems Ltd., in the communities of Cowichan Valley Regional District, Nana.imo Regional District and the District of Campbell River, British Columbia. BETWEEN: nm DIRECI'OR OF INVFSl'lGATION ...--C-O-MP-E-TIT-ID-N- T-R-IB-UN-A-L ---- ANDD J U . C . .! U 'U A A 3~ D CH TRIBtNAL DE LA COf!Cl:~REMa P Applicant F
t S~? ~o 1991 r R 0 ~ ·-and-D
REGISTRAR - lEGISUAllE - --on-AWA~- -o-N-r.-T " LAIDLAW WASI'E S%T.EMS LTD. REBUT·TAL AFFIDAVIT I, JAMES J. MCRAE, of the District of Norrh Saanich and the Province of British Columbia, make an oath and say as follows:
1. I swore an Affidavit, attaching my original report in this proceeding, on September 12, 1991. 2. Attached hereto as Exhibit 11 A 11 is a true copy of my report which. provides a rebuttal t-0 the report prepared by Mr. Michael Ross et. al of Peat Marwick Stevenson Kellogg.
.- .
3. I make this Affidavit pursuant to Rule 42 (1) of the Competition Tnounal R.uJ.es. SWORN before me at the ) ...... , .. ' City of Victoria, in the . ) Province of British Columbia, ) .. :: ::· .. · on th~ ·?b th day of September, ) /~~:,·\;··l·9·9· F. >.~ ) = } '"- ; : = :-: --: -, ,; . .. . •. ·.. ;. ~ --== 19\\ .---·% - :. . · , . , . -~~ ---, -... G . 'onultis joncr for Takimr " ;;.·~;..··. - . •• ... Aftidavits ..... ------;~.;--~::.~. . . . . .~ . _...-:: ........: . . ~.--~- ~·MADELEINE BARRIE NOTARY PUBLIC 817 A FORT STREET VICTORIA. B.C. VSW 1 H6 382-5331
77ils u E;;~ " f't ~· ;-q'!'Tc°l l~~t Jit c:ff.~.-rr.i: ,,! JOrl\\t!C; ·->. ~it.c. E S:rnr.! ~<';·..rre t.'t:; ct Vf.::c;r~w EXHIBIT "A"
IN THE MATTER OF an Application by the Director of Investiption and Research under s. 79 of the Competition Act~~: ~·~~El£1N.E ·a~~!f: 9 oc: C 34 ded· R.S.C., 1 0-1, c. - ' as amen AND IN THE MATI'ER OF certlin practices by Laidlaw Waste Systems Lt.d., in the communities of Cowichan Valley Regional District, Nanai.mo Regional District and the District of Campbell River, British Columbia.
BETWEEN: TBE DIRECTOR OF JNVESTIGATION AND RESEARCH Applicant
-and-LAIDLAW WASTE SYSTEMS LID.
AFF.IDA VIT EVIDENCE OF 1AMF.s J. MCRAE 1. This rebuttal to the Affidavit and report prepared by Mr. Ross and his associates at Peat Marwick Stevenson Kellogg is organiz.ed into six sections. Section I compares the relevant product market as defined by Mr. Ross and myself, and concludes that the two are nearly identical. Section II compares our respective definitions of the relevant geographic mar~ and reconciles the differences. The next Section re-visits the issue of Laidlaw's market share in the relevant market areas. Section IV notes that both Mr. Ross and I conclude that barriers to entry exist in the markets under examination. Section V loo!G at the empirical relevance of economies of density in light of Mr. Ross' suggestion that exclusivity may be an efficient response to prevailini market forces. The final Section discusses the issue of economies of scale, again in light of the efficiency question.
1
: - . -- . ,.._ : ~ ·.. ', n.9TARY PU81.;!C2 ' ~~7.A ·FORT siRi:fr VICTORIA . B · . -~ "3-s2~~~3;VSW !H
I THE RELEVANT PRODUCT MARKET 2. The report of Mr. Michael Ross, dared September 13, 1991, defines the relevant product market (p. 47) as, "lift-on-board(LOB) services, Le., the provision t.o customers of the service of solid waste pickup, haulage, and disposal, using bins of a siz.e between 2 and 8 cubic yards, lifted and dumped into front, side or rear-load compactor trucks".
3. My report, dated September 12, 1991, defines the product market as solid waste collection and disposal service t.o medium volume (one to ten cubic yards per week) customers who desire the contrlnerl7.cd st.magc of their refuse, and who am most efficiently served by means of front-load vehicles, or less frequently by automated rear or side-loaders which have container lifting ca:pa.city.
4. I believe that the two definitions of the product market aze almost identical. Accordinely, I conclude that the market for commercial containerized solid waste collection and disposal service (my terminology) is the relevant one for the purpose of this proceetfing.
Il THE RELEVANT GEOGRAPllIC MARKET 5. In the report of Mr. Ross, there is a detailed analysis of the 11 cost impact of remote service" which is based on cost information made available by I aidla:w operating staff on servicing the Cowichan Valley directly from Nanaimo. Mr. Ross concludes (p. 44) that, •any pick-up route that bas at least part of its length within 30 miles (50 kilometres) of the base of operations of another provider of the same service is potentially open for competition•.
6. I a:oive at a similar conclusion, i.e. a solid waste collection and disposal firm operating from a hub outside of the local area (an •outside• firm) can remain competitive with locally domiciled (•inside•) competitors so long as the outside firm's operational hub is SO km or less from the operational hub of the inside firm.
2
7. Primarily from this observation on the reach of the firm's competitive "envelope", ie. 50 km from an operation hub, Mr. Ross concludes that the Nanaimo Regional District and the Cowicban Valley Rqional District - with the exception of the City of Duncan and the Village of Lake Cowicban which provide commercial containeri7.cd service as a local government activity - should be considered as a single market. This is incoz:rect..
8. The 50 km competitive envelope would allow a firm with an operational hub in Nanaimo to competitively serve castomers in the southern portion of the Nanaimo Regional Distrl~ plus customers in the northern portion of the Cowichan Valley Regi.ona1 District, i.e. the Town of Ladysmith and sm:rounding Electoral Areas G and H, and population centres in the northern half of the District of North Cowichan. These 8mlS comprise 35 % of the total population of the Cowichan Valley Regional District.
9. Thus, the relevant .ma.t:ket in this area is the Nanaimo Regional District plus the northem portion of the Cowichan Valley Regional District comprising about one-third of the Regional District's total population. This is the area which I rcfermce as the "Nanaimo" market in paragraph 32 of my report. The southern portion of the Cowicban Valley Regional District is a separate aeographic market which I reference as the "Cowichan Valley•.
10. In combining all of the Nanaimo Regional District and the Cowichan Valley Regional District into a single market, Mr. Ross also appeals t.o the observation that Laidlaw considers both regions as a single entity in terms of the regional office functions of operations, equipment maintenance, m.arkcting and adm.i.nistration. This observation is not relevant to the issue of defining the geographic boundaries of the market
11. The placement of geographic boundary lines is accomplished by determining the likely substitution to producers whose production filcilities, i.e. operational hubs in the present context, are located in other regions. The observation that Laidlaw finds it efficient to supply regional office services to two (or three) hubs from a single location
3
bas no bearing on the issue of dctcrmining whether or not a significant number of commercial containeri.7.ed customers can obtain service from firms domiciled in other re&ions if they so de.med.
12. Finally, the practical significance (m terms of market shares) of where the geographic market boundary line should be placed for the Nana.imo Regional District and the Cowichan Valley Regional. District~ is minimal given that Laidlaw has a very high market share in both jurisdictions.
13. In his report, Mr. Ross defines the Campbell River market broadly. The stated reasons (p. 52) for this particular ifOupini of communities into a mare.t area for the purpose of this proceeding range from •convenience• to the observation that existing firms dump waste in the •same Campbell River landfill as Taidlaw cumntly uses•.
14. As ~y discussed, the placement of ~hie market boundary lines should be done on the basis of the likely substi.tuti.on of firms whose operating hubs arc located in other re&ions.
15. Mr. Ross and I have independently estimated that in the Nanaimo an:a, the competitive envelope of a solid waste collection and disposal firm extends SO km or less from its operational hub. Given the distribution of population between the District of Campbdl River on the one hand, and the City of Courtenay, the Town of Comox and Village of Cumberland on the other, I conclude that the estimated siz.e of the competitive envelope is also valid between these population centres.
16. The District of Campbell River is at the outer limit (or beyond) the competitive envelope in terms of the distance from the Courtenay/Comox/CUmberiand area. Evidence of this fact is provided by the observation that Laidlaw in Campbell River, and the firms providing commercial containerized service in the Courtenay/ Comox/Cumbcrland area, do not enter each other's market areas, and end their service at a point approximately one-half way between the two population centres.
4
17. Quadra Island is unique because part of the journey to the Campbell River dumpsite must be made by ferry. Because of the higher cost of travelling by this mode, the competitive envelope will be considerably smaller than the estimated 50 km which is based on road transportation only.
18. The Village of Sayward is also unique due to the fact that it lies 81 km north of the District of Campbell River, but Utldlaw simultmeously provides residential hand bag/trash can servic:e and commercial containerized service from its base in Campbell "River. However, as discussed in paragraph 28 of my report, the siz.e of the competitive envelope depends partially on the size of the market area to be sexved by the outside finn. Small market areas such as Sayward (population 411) can be served from campbcl1 River because the truck must make the •empty 11 segment of the trip relatively infrequently. Larger m.arlcet areu such as the Courtcna.y/Comox/Cumberland area must be served by establishing an operational hub.
19. The practical significance of including or excluding both Quadra Is1aJld and the Village of Sayward is minimal due to the low number of customers ®ring commercial containeriz.ed service in these areas.
20. Thus, the relevant geographic market for the purpose of this proceeding is considerably smaller than the one identified by Mr. Ross. As discussed in paragraph 32 of my report, the relevant area is referenced as "C&npbell River" and includes the District of Campbell River and surrounding Electoral Areas D, E and F.
ill MARKET SHARE 21. Mr. Ross uses three methods to measure Laidlaw's market share in the commercial containerimi (or lift-on-board) solid waste market. His three methods are described on pag~ 54 to 58 of his report He chooses Method 3 as his •bestn estimate in the Nanai1no Regional District ( ~ defined by him); Method 1 (the only one available)
in the Campbell River market; and an average of all three methods for the Cowichan Valley Regional District
22. In my view, his Method 2, which is based on actual weighed volumes of waste, is the most reliable and is also used in my calculations. Method 1 is based on a numerical cotmt of bins rather than recorded weight of refuse. Method 3 is based on a combioation of arbitrary underlying measures.
23. Mr. Ross' estimates usin& method 2 are very similar to mine. My estimate of Laidlaw's (averap) market share for 1991 in the Cowichan Valley Regional. District is 899' as compared to his 86-87%. In the Nanaimo Regional District, my (average) 1991 estimate is 86% while his. is 87%. Because no weight records are kept at the Campbell River dumpsitc, neither Mr. Ro~ nor I is able to use this method to produce market share estimates.
24. MI. Ross' estimate of Iaidlaw's mar.ket share for Campbell .River (46%) is significantly biased downward by bis definition of the Id.evant geographical lll.3Mt (see para. 20 above). Once the Campbell River market is correctly defined, Laidlaw's market share is estimated to be 83% using Mr. Ross' Method 1 (see col. 2, line 9, p.30).
IV EASE OF ENTRY 25. Mr. Ross acknowledges the existence of entry barriers due to economies of scale, and the existence of a contracted customer base which is difficult for new entrants to penetrate (seep. 61 of the Ross report).
26. I make precisely the same points in my report. V ECONOMlES OF DENSITY 27. Mr. Ross points out that in various sectors of the transportation industry, operating costs per pick-up are thought to fall as the distance between pick-up points
-. d~, i.e. the density of pick-up points·increases. Mr. Ross calls this concept the "economies of contiguity" and produces an empirical estimate of the importance of the concept, but not in the conventional economic format of a per unit cost elasticity.
28. Data provided on p. 37 of Mr. Ross' report may be organized into a cost elasticity format to show that a 1 % increase in density (measured in terms of pick-ups per 100 miles) reduces the labour and landfill costs per pick-up by only 0.25 %•
29. Even this relatively small degree of cost responsiveness to increases in density is an overestimate. Mr. Ross does not include the costs of fuel and oil, tire wear and tear, or maintenance and depreciation, all of which increase with the higher levels of stop-and-&0 drivin& associated with increased customer density. Inclusion of these costs will reduce the cost responsiveness estimate of 0.25 % toward zero.
30. Based on my analysis of the empll:i.cal results produced by Mr. Ross on economies of contiguity (or density), I conclude that "~usivity•, Le. I.aidlaw's near monopoly position in the relevant geographic markets, is not an efficient response to the market. The potential cost efficiencies which may accrue to an exclusive commercial containeri7.ed firm due 1D economies of density are in.sufficient in rnqnitude for exclusivity to be an efficient market outcome.
VI ECONOl\.fiFS OF SCALE 3 L Mr. Ross suggests that economies of scale, attributable to better labour and equipment utilization, speciaHZ<lti.on of personnel functions within the firm, volume purchasing and the pooling of overhead costs are present in the solid waste collection and disposal industry.
32. The source of this observation is a 1977 academic analysis of the costs of residential solid waste collection and disposal in a sample of 315 US cities. For residential solid waste service provided under a •market" organizational structure, the
7
author estimates (pp. 132 - 134) that economics of scale are exhausted in cities served by more than two residential collection vehicles.
33. These findings are of limited relevance to this proceeding for several reasons. First, the author analyzes the costs of residential solid waste collection and disposal only. Second, the data are quite stale, and hence cannot incorporate the many changes in ~n technology and management practices which have taken place since 1977. Finally, the functional form of the estimated cost function is a rigid specification which has been replaced in leading econometric studies of the transportation industry.
34. The matter of economies of scale is also examined in demit in my report (see pazagraph 60). For reasons discussed therein, I conclude that there are economies of scale applicable to single truck firms in the commercial containerized solid waste business, but they are almost completely exhausted at the point when a front-load vehicle produces approximately 240 lifts per week.
35. Another important consideration is that the achievement of economies of scale in the solid wast.e collection industry does not necessarily require a firm to hold a specific share of the market in the individual relevant markets under enmination in this ~ Some multi-vehicle economies of scale, arising due to the reasons identified by Mr. Ross, are potentially available to firms who are able to operate in two or more adjoining market areas, and/or other solid waste transportation and disposal product markets. This observation is supported by I 41jdfaw' s organimtion of its operations into Divisions which, in some cases, serve adjoining ~onal markets. For example, both the Cowichan Valley Regional District and the Nanailno Regional District are subsumed within the respondent's Nanaimo Division.
36. I therefore conclude that the asserted economies of scale are not sufficiently strong, relative to the size of each of the three geographic markets, to imply that Laidlaw's overwhelming market shares or its restrictive contractual practices are an efficient response to market forces.
•. 0