KEA

Komanoff Energy Associates

11 Hanover Square • 21st Floor • New York, NY 10005
www.komanoff.net • komanoff@gmail.com
212 260 5237

Brian M. Rosenthal
Investigative Reporter
The New York Times
620 Eighth Avenue
New York, NY 10018
May 22, 2019
Dear Brian —
I’m writing to invite you to work with me in retracting and repairing what I believe is a
serious and consequential misstatement in your Sunday May 19 New York Times front-
page exposé, “Driven to Despair: How Reckless Loans Devastated a Generation of Cab
Owners.”1
In that story, you wrote:
City data shows that since Uber entered New York in 2011, yellow cab
revenue has decreased by about 10 percent per cab, a significant bite
for low-earning drivers but a small drop compared with medallion
values, which initially rose and then fell by 90 percent. (emphasis
added)
In actuality, yellow cab revenue appears to have decreased by a far greater
amount over that period; I put the decrease at 36 percent. If true, that would
constitute far more than “a significant bite” for cab drivers and medallion owners — it
would be a financial catastrophe.
As I describe below, I have drawn my conclusion from the same Taxi and Limousine
Commission data source you relied on for your assertion. My result differs from yours
for two reasons: (i) in tabulating industry-wide 2019 revenue, you failed to net the
surcharges that are collected under state and city mandates and are not retained by
the drivers, and (ii) rather than letting the decline in revenue from 2011 to 2019 stand
by itself, you calculated the revenue loss per taxicab and then attenuated that loss by
dividing industry-wide revenue by the shrunken number of taxicabs in use in 2019; this
maneuver had the effect of canceling much of the financial distress your story
otherwise addressed and described.
The first issue involves an error in accounting; correcting it reduces the 2019 industry-
wide revenue figure you relied on by 15 percent, and, with no further correction,
enlarges the per-cab revenue loss from around one-tenth as you reported to nearly
one-fourth. The second issue is perhaps open to interpretation, but I believe the
appropriate construct is not the loss in revenue per vehicle but the loss in revenue for
the taxi industry as a whole. Even so, if you insist on calculating revenue loss per
vehicle, the right approach — one that captures rather than overlooks the mothballing
of thousands of yellow cabs because demand has dried up — is to employ the same

1The story was posted online under the headline, ‘They Were Conned’: How Reckless Loans
Devastated a Generation of Taxi Drivers, with the link, https://www.nytimes.com/2019/05/19/
nyregion/nyc-taxis-medallions-suicides.html

1
denominator (number of taxicabs in use) for both 2011 and 2019. For either
calculation — 2011-2019 industry-wide revenue decrease or 2011-2019 per-taxicab
revenue decrease — the correct figure is 36 percent.
Needless to say, the difference between a 10 percent taxicab revenue decrease and a
36 percent taxicab revenue decrease is extremely consequential. The lower figure
could be consistent with a narrative that largely attributes the extreme financial
distress of yellow cab drivers and medallion owners to the predatory lending you
described in your story and its next-day follow-up, “As Thousands of Taxi Drivers Were
Trapped in Loans, Top Officials Counted the Money.” In contrast, the higher figure
points inescapably to a broader narrative that places the cannibalization of the taxi
industry by Uber and Lyft as a destructive force that has been every bit as impactful as
the predatory lending depicted in your series.
If my re-analysis is correct, it seems to me that the remedy is for The Times to publish
a story to this effect. This would not be an ordinary “correction” but a new story
that retracts the “10 percent” assertion in your original story and makes clear
that the decimation of driver income has been three to four times as great as you
asserted.
I am directing this request to you rather than to an editor because of my regard for
your work and because I believe working this out with you will yield a better process
and outcome. (I’m sending a copy of this letter to Emma Fitzsimmons because I have a
good working relationship with her and because she contributed reporting to your
story.) Because events following from your story are moving rapidly, I would like
to hear back from you soon; I also ask that the new story be posted sometime this
week.
Before I turn to my re-analysis of your assertion, I want to state for the record that I
have done this re-analysis, and am writing to you, entirely on my own behalf. While it
is true that from late October 2018 to late January 2019 I worked for and represented,
as a paid consultant, a taxi medallion owner, that work occurred in a different context
than what I have undertaken here (it concerned the inequities of the new state-
mandated taxi congestion surcharge, and alternatives to it). I undertook the analysis
outlined here on my own after reading your Sunday story yesterday and being struck
by the assertion that per-cab revenue had declined by only around 10 percent from
2011 to today.
Analysis of Declining Yellow Cab Revenue
After I emailed you yesterday evening with my preliminary calculations of declining
yellow cab income, you replied and kindly shared your data source, which was a Taxi
and Limousine Commission spreadsheet file, “Data Reports Monthly Indicators.”2 You
advised me that you had compared the March 2011 and March 2019 rows from that
file, which I agree function nicely as before and after indicators for the yellow cab
industry.
I’ve placed the important data from those rows in the table on page 3. All figures are
per day.
Table 1: TLC data as interpreted and used by Brian Rosenthal

2The TLC spreadsheet file is available via the link, https://www1.nyc.gov/assets/tlc/

downloads/csv/data_reports_monthly_indicators.csv.

2
 Revenue per Vehicle Change in
Trips Per Farebox (calculated by Rev. per
Day Revenue Vehicles Komanoff) Vehicle

March 2011 518,212 $5,717,207 12,852 $444.85

March 2019 252,634 $4,289,036 10,814 $396.62 minus 11%

From the figures in the last column, revenue per vehicle appears to have fallen by
(only) 11 percent from 2011 to 2019. That is the basis of the assertion in your story I
quoted at the top.
My first point of contention goes to the March 2019 farebox revenue figure,
$4,289,036. My understanding is that that figure includes two surcharges that were
instituted after 2011 and, thus, would not have affected the March 2011 revenue
figure ($5,717,207) that is used for comparison. I base my understanding on a
glossary3 for the TLC spreadsheet that defines the “Farebox per day” data entry as
“Total amount, across all vehicles, collected from all fares, surcharges, taxes, and
tolls. Note: this amount does not include amounts from credit card tips.”
The first surcharge added since 2011 is the Taxicab Improvement Surcharge of 30 cents
per trip that has been collected since January 1, 2015. The second and more recent
surcharge is the New York State Congestion Surcharge of $2.50 for all yellow taxi trips
that begin, end or pass through Manhattan south of 96th Street; it has been collected
since February 2, 2019.
Since these surcharge revenues are passed through in their entirety to government,
they should not be counted as part of driver revenue. To fairly compare 2019 revenue
with 2011 revenue, then, it is necessary to deduct their amounts from the March 2019
farebox revenue figure of $4,289,036 that you relied on to conclude that the per-
vehicle revenue decrease to 2019 from 2011 was only around 10 percent.
I have estimated these deductions, as follows:
• Each of the 252,634 daily trips in March 2019 included the $0.30 Taxicab Improvement
Surcharge. The associated revenue (to be deducted) is $75,790 (calculated as 252,634 x
$0.30).
• I have previously estimated that nearly 95 percent of all yellow taxi trips begin, end or
pass through Manhattan south of 96th Street on the east side or south of 110th Street on
the west side.4 That is a slightly larger area than the zone for which the NY State
congestion surcharge is collected; to adjust for the smaller zone, I have reduced the
nearly-95 percent share to 90 percent. The associated revenue (to be deducted) is then
$568,426 (calculated as 252,634 x $2.50 x 90%).
With these deductions, the reported daily March 2019 farebox revenue figure of
$4,289,036 that you relied on becomes $3,644,820. I have made the alteration in Table
2. It shows that when the surcharge revenues included in the 2019 figure are
netted out, the decrease in revenue per vehicle from 2011 becomes 24 percent.

3The TLC glossary is available via the link, https://www1.nyc.gov/assets/tlc/downloads/pdf/

data_dictionary_monthly_indicators.pdf.
4Derivation of my nearly-95% figure is available in the Taxis tab of my “Balanced
Transportation Analyzer” Excel spreadsheet, in Rows 249-261, Column K. The spreadsheet is
publicly available and may be downloaded via the link: http://www.nnyn.org/kheelplan/
BTA_1.1.xls.

3
Table 2: TLC data, with 2019 farebox revenue netted for revenues from surcharge imposed
after 2011
Revenue per Vehicle Change in
Trips Per Farebox (calculated by Rev. per
Day Revenue Vehicles Komanoff) Vehicle

March 2011 518,212 $5,717,207 12,852 $444.85

March 2019 252,634 $3,644,820 10,814 $337.05 minus 24%

Cells changed from Table 1 are marked in yellow.

My second point of difference from you concerns your adoption of the shrunken
number of yellow cabs in operation in 2019 as the denominator for calculating
per-taxicab revenue for this year.
The heart of this discussion is a comparison of the top and bottom figures in
the third column in Table 2. Daily farebox revenue (after surcharges have been
netted) shrank from $5,717,207 in 2011 to $3,644,820 in 2019 — a decline of 36
percent. I believe the calculations should end there. There is no need to
compute the decline per taxicab.
The reason for ending the calculation there should be clear: the issue before
the public — which your stories this week have done so much to amplify — is
the financial distress that is devastating the yellow taxicab industry. There is
no starker evidence of this devastation than the withdrawal from daily service
of some two thousand taxicabs, as shown in the fourth column of Table 2. To
calculate per-taxicab revenue without those mothballed taxicabs, as you did in
your Sunday story, is to whitewash this destruction. That would be akin to
measuring the effect of famine on a herd of livestock by measuring weight loss
only among the survivors while ignoring the sizeable fraction (16 percent in this
case) of those who perished.
Nevertheless, for the record I have recalculated the prior table but with the
number of vehicles in use in 2011 retained for 2019. The 36 percent decline
shown in the last cell in Table 3 is, of course, the same figure as the decline in
industry-wide revenue calculated earlier from the respective farebox revenues
in the third columns of Tables 2 and 3.
Table 2: TLC data, with 2019 revenue netted for surcharges and number of vehicles held
constant
Revenue per Vehicle Change in
Trips Per Farebox (calculated by Rev. per
Day Revenue Vehicles Komanoff) Vehicle

March 2011 518,212 $5,717,207 12,852 $444.85

March 2019 252,634 $3,644,820 12,852 $283.60 minus 36%

Cells changed from Table 2 are marked in light green.

(Note that, if anything, the number of vehicles for 2019 should have been
increased to reflect medallion sales since 2011, which I believe numbered 250.
This would have increased the decline in per-vehicle revenues still further.)
Before I conclude, I would like to sketch the financial import of the revenue
declines I’ve calculated here. The respective figures in the penultimate column

4
of Table 3 indicate that the average yellow taxicab is today generating, per
day, approximately $160 less in revenue than it generated in 2011 (calculated
as $444.85 less $283.60). If we apply that revenue loss to 300 days of service
per year, the annual per-taxicab drop in revenue is on the order of
$48,000. Simple arithmetic suggests that had this revenue not been lost — had
the yellow cab sector not been ravaged by Uber and Lyft — the lost revenue
could today be supporting a debt of nearly half-a-million dollars financed at a
10 percent interest rate. This could have significantly eased the financial
burden of shouldering the rise in the price of a medallion to more than $1
million in 2014 from $200,000 in 2002 that you reported in your first story.
This does not in any way lessen the culpability of the predatory lenders whose
exploitative practices you uncovered and depicted in your stories. What it does
do is demonstrate that the destruction of the yellow taxi franchise by Uber
and Lyft has operated alongside those practices to wreak widespread
financial ruin on the taxi industry.
I hope you will accept my criticisms and corrections of your assertion that I
have addressed in this letter. Assuming that you do, I would like to hear back
from you soon regarding your plans for retracting that assertion and replacing it
with an accurate appraisal that reflects the true extent of the revenue losses
suffered by the taxi industry in recent years.
Thanks and all best,

Charles Komanoff