Fri, 22 Jan 1999 16:21:20 -0500


[Forwarded from Peter Anderson. NOTE: To send directly to
GreenYes, address message to] -bill s.

Subject: Tierney & Recycling Economics: REQUEST FOR

In the course of reviewing the Tierney piece, I requested from NYC a
copy of their Comprehensive Waste Management Plan cost study
which Tierney references as the basis for his figures on recycling
economics. (Even though most people would find it unreasonable to
use NYC's "unique" conditions as the basis to evaluate the
applicability of any kind of program for the rest of the country, I
thought it might be useful to look at the details of their accounting, to
see what infirmities lie within Tierney's own foundation.)

A quick review of section 3 of that study suggests substantial
problems in cost computation, allocation and assignment. However,
I have not lived in NYC for 26 years and do not have any "on the
ground" knowledge of the actual situation currently in the city.

Can someone with a close-in view of NYC look at these
PRELIMINARY comments to evaluate whether they are valid.
Thank you. Please DO NOT publicly distribute them until we have
some peer comment. Another caution is that, although I disagree
with the conclusions that Tierney drew from the report, its details
suggest that the authors were probably not seeking to discredit
recycling. Rather they were largely describing the atypical things that
go on in NYC.

Peter Anderson
RecycleWorlds Consulting
ph: (608) 231-1100
fx: (608) 233-0011


One of the major determinants of whether the addition of a curbside
recycling program incrementally increases (or decreases) overall
solid waste handling costs is how much landfill tipping fees can be
avoided by the diversion that recycling achieves.

NYC's residential (and some minor commercial) refuse is disposed
of at the city's own Fish Kills landfill. According to the NYC study,
the authors are assigning a cost of $14 per ton as the tipping fee plus
an annualized accrual to reflect closure and long term care which is
avoided due to recycling (excluding transfer station costs accounted
for elsewhere). See Table 3.5.

No information is provided as to whether $14/ton accurately reflects
historic costs incurred. However, some information in the footnotes
strongly implies that it does not fully reflect the future amounts that
should be currently recognized in the line provided for accruals of
future costs associated with the expense of closing the facility and
providing long term care.

The "accrued liability" listed on the annual funding account
statement for this purpose is $9.5 million (see Table 3-7). Specifically,
the NYC study begins with an estimate made by Roy Weston in 1994
that closure and care would be $557 million in 1994 dollars. Table 3-1
at note (c). I have not seen the Weston report, but past experience
with these types of things would suggest that the final cost will be
substantially greater. In addition, the proper economic treatment of
that figure would be to compute a discounted present value of such a
future stream. Since the closure and clean up costs would be front-
ended, I would guess the discounted present value would be over
$600 million, assuming the $557 million is accurate, but more
complete year-by year data is needed to calculate the true figure.

Beyond that probable (but not established) underestimation is the
manner in which the future costs for closure and care is currently
recognized. It is first divided by the air space in the, apparently, two
open cells at Fish Kills (cells 1/9 and 6/7) containing 190,000,000
cubic yards, or $2.93 per cu. yd.

Second, they convert yardage to tonnage by assuming that the
density of their compacted refuse at Fish Kills is 2,439 pounds per
cubic yard. This compares with a national average of less than 1,000
pcy, and a high of about 1,200 pcy for those with the new large 96,000
pound compactors. That is to say, it seems twice as high as would
seem possible (unless they're counting subsidence of the refuse body
into the underlying muck). It also is not clear whether their
190,000,000 cy of air space made a subtraction for daily cover.

Third, they multiply this $2.93/cy figure by the 3.95 million tons of
MSW landfilled in 1994, adjusted for compaction at 2,439 pounds per
cubic yard.

Apart from the problems with the density conversion noted in the
previous paragraph, the real question is whether the proper divisor is
the nominal space in the open cells or the number of tons that are
expected to be disposed of between the report's date and the
termination date for Fish Kills of 2001. I do not believe that it is
economically responsible to amortize a cost over more units of an
operation that will occur before the operation ends, yet that appears
to be what the City is doing. If, as I believe, the only appropriate
manner to amortize the closure and care costs is the number of tons
that are expected to be disposed of prior to termination, then the
divisor becomes 6 years times 3.95 million tons (instead of the
remaining space in the open cells). Changing the closure and care
cost to $600 million and the divisor to 24 million tons adds $25 per
ton to the $14 per ton being booked for landfill costs. (Because the
City's report accounts for this factor as a separate line for accruals, in
order to make our statement of the tipping fee that includes closure
and care equivalent to the City's tables, this $25/ton should be
reduced by $2.40/ton to $23/ton.)

IMPACT OF CORRECTION: In the final chart in the study, the City
report estimates that in 1996 the incremental cost of curbside
recycling is $55 million or $112 per ton, after accounting for avoided
tipping fees at $14/ton. Table 3-6. Increasing the avoided tipping fee
by $23/ton for the 462,000 tons processed and sold by the City's
MRFs reduces the incremental cost by $11 million to $44 million and
the cost per ton to $95/ton. This all assumes that $557 million cost
for closure and care is accurate, which is exceedingly unlikely. Even
the resulting tipping fee of $29/ton, after more accurately accounting
for the City's estimated closure and care costs, still seems like a non-
market rate for that part of the country. If the average Northeast
tipping fee of $73/ton is used (Waste Age, May 1996, at p. 180)
instead of $14+$25/ton, or $29/ton, then the greater avoided tipping
fees decreases the incremental cost of recycling by another $20
million to $24 million, or $52/ton. (At !the same time, it should be
noted that report approaches this construct from a different angle, p.
3-26, that is too complicated to describe here, but the net cost effect of
which is to increase the benefits of recycling by approximately $12
million, or $26/ton.)


One of the major savings from curbside recycling that is widely
unrecognized is the reduction in the need for refuse packers because
approximately a fifth of their load is diverted to the second fleet. We
analyzed the issue in detail in "Debunking the Two Fleet Myth,"
Waste Age (Oct. 1995). Typically, this can eliminate the need for the
mid day trip to unload, and can make it possible to use less
expensive single axle instead of twin axle chassis. This can reduce the
refuse collection costs by 25% or more.

In the City report, this concept is commendably recognized, but the
amounts recorded for the impact are 7%, or $27 million, of what is
apparently a $362 million refuse collection bill. Table 3-3 and 3-6.
The report notes that the City's labor environment limited its ability
to reducing refuse routes (p. 3-21). If the pattern found elsewhere
unencumbered by work rules is prorated to the City's low diversion
rate is used, the avoided packer costs would be doubled from that
shown. Certainly, I would not want productivity gains to be
implemented by layoffs, but limited downsizing to attrition is a fairer
compromise than perpetual featherbedding. In any event, I do not
believe that it the best basis for an economic analysis to assign costs
caused by work rules to recycling when one of the purposes of the
analysis is to draw or imply lessons for other communities.

IMPACT OF CORRECTION: In the final table, the avoided refuse
collection cost is shown to be $27 million. Doubling it, would
increase the offset by $27 million to $54 million, turning the city's
reported incremental cost of recycling in 1996 of $55 million or $112
per ton, into a break even point. Table 3-6. Combining the correction
to the avoided tipping fee with this correction would show an
incremental savings of $10 million, or $23/ton.


The report indicates that recycle collection is done in two person
crews (p. 3.23). Unfortunately, no information is provided to indicate
what proportion of that collection is containerized, rather than
curbside, collection.

While there a few larger cities which use two person crews for
curbside recycling, there is no performance basis to justify the extra
crew member since the advent of low entry, right side cabs and side
loading boxes, although there is for containerized collection which
loads from the rear. Without information on the proportion which is
curbside, and on the cost of the second person on those trucks, it is
impossible to estimate the savings.

Nor is there any way to estimate the total second level impact of two
person crews. For, in order to avoid having both operators sitting in
the vehicle when it travels off route to unload, a relay system is used
in which the vehicle first returns to the garage and then one person
takes the truck to the MRF (p. 3-23). The cost of these relays by
themselves is indicated to be $13 million. In addition, because the
recycle vehicle now must return to the garage before being taken out
again to go to the MRF, and because this wasted travel is on heavily
congested NYC streets, more trucks will be required than would
have been the case if there were one person crews which did not use

IMPACT OF CORRECTION: As noted, insufficient data is provided
to estimate an impact from this correction, but the total collection cost
is $84 million and the total relay cost is $13 million.


The costs of waste prevention and reuse and composting are added
to the costs of the paper and container recycling program (p. 3-21).
However, it does not appear that the diversion achieved by these
efforts is quantified and the savings are not used to offset those costs.
The avoided cost benefits are calculated based upon just those
tonnages diverted directly through at the MRFs.