Reason critique of 3-Profs report on Penn Pike - and the ensuing controversy
Posted Sat, 2008-04-12 22:23
A Reason Foundation report on the Pennsylvania Turnpike by Robert Poole (Bob) and myself (Peter Samuel) has got some attention. Bob did all the work and I got all the controversy! Well not quite. My major contribution was the discussion of the costs of the Pennsylvania Turnpike (see below) which generated publicity and several other small additions to Poole's work. Poole wrote a devastating critique of many other aspects of the 3-Profs report, which has been largely overlooked in the media focus on costs.
The 3-Profs report compares a longterm private sector lease/concession with the Turnpike Commission's Act 44 plan for an expansionist state turnpike devoted to generating profits for subsidizing the state's free roads and transit.
"3-Profs" report is my shorthand for the report titled "For Whom the Road Tolls: Corporate Asset or Public Good" (FWTHT) authored by three academics Gary J Gray, Penn State, University Park, Patrick J Cusatis, Penn State Harrisburg and John H Foote, Harvard. The Reason report calls it FWTRT. It was sponsored by the Democratic Caucus of the Pennsylvania state house of representatives and published in February.
It is available on the Turnpike's website
http://www.paturnpike.com/I80/pdf/For_Whom_the_Road_Tolls_Final_2-23-081_FINAL.pdf
Poole discovered three kinds of flaws in 3-Profs:
(1) using arbitrarily different assumptions when comparing private and public sector alternatives
(2) assuming away genuine differences between public and private sector operations
(3) failure to use risk-based discount rates on projected future revenues of the Expansionist Turnpike
Different traffic growth
One of the ways the 3-Profs rig the assumptions to favor the public toll authority model is assuming traffic growth of 2.5% a year for 20 years, then 2% thereafter in the public toll authority model versus 1%/yr throughout under a concession. Compound interest over 50 years means traffic grows 197% for the public toll authority and 65% for the concessionaire.
Not surprisingly the public toll authority generates more revenue and more profit!
In order to hustle up a scare about toll rates 3-Profs assumes toll rates under a concession would be allowed to grow 5.5%/yr based on that being a possible outcome of the Indiana Toll Road formula of caps rising each year by the greatest of three measures - GDP/person, CPI, 2%. Then they assume the caps will be fully exploited regardless of loss of traffic and revenues!
The Profs' assumption chicanery
Such assumption-chicanery of course yields sky-high toll rates tailormade for scaremongering talking points.
Poole discovers that one table (Table 5) uses an unexplained 4% annual cost growth while purporting to use a 3% cost growth.
Another example of selective use of assumptions to "prove" the desired conclusion in 3-Profs is their notion that only the private sector needs to add a risk premium to the discount rate used to estimate the present value of future toll revenues. In the best of circumstances public sector tollroads need to discount for risk too.
Risk misassigned
Indeed from a public standpoint a private sector lease involves zero risk if the state gets its payment in one upfront sum. The risk premium needed is of concern only to investors and lenders. The state has its money out of the concession upfront and if things go bad on the road that money isn'ty going to be given back.
By contrast public-public concessions like that between the PTC and PennDOT under Act 44 contain no provisions for cancellation of the concession due to non-compliance, and reversion of control to the state - as will occur from default with a private concessionaire. So there's a serious risk of a defaulting state turnpike in a public-public concession remaining in charge of the tollroad.
Private sector focus on efficiency due to shareholders
When it suits their advocacy purposes the 3-Profs assume away striking differences between the public and private sectors - such as the differing incentives to minimize costs and maximize profits. Return on capital is a private sector obsession that leads to constant efforts to improve efficiency and maximize traffic whereas in the public sector with no shareholders to satisfy efficiency and profitability are much more occasional concerns.
3-Profs cites studies of socalled 'privatization' in transit operations to claim that toll concessionaires won't be able to reduce capital costs. Poole points out these studies are irrelevant to tollroads because they refer to management contracts for transit in which the private sector has no incentive to contain capital costs because they are being borne by the government.
Poole points out that the 3-Profs make the preposterous assumption that there is zero risk in the Act 44 plan to toll I-80, discounting future toll revenues from it by a mere 4.5%/yr. Given that Act 44's plan to milk I-80 tolls for supporting other state roads and transit is totally at odds with federal law which only permits tolls to be implemented on projects that use the proceeds within the corridor, the risk factor is extremely high – not negligible as the 3-Profs assume.
Costs of capital misrepresented
3-Profs is plain wrong in claiming that cost of capital is higher for the private sector. The private sector borrows on an international basis, can tap equity as well as
debt and requires smaller reserves and coverage ratios, and can claim tax deductions on interest. And now via USDOT-approved 'private activity bonds' it can issue tax-free debt in the US just like the public sector.
Morgan Stanley target return
Morgan Stanley's suggestion made in studies for Gov Rendell that investment of the concession fee could get a yield for the state of 7 to 9% is ridiculed as absurdly high by 3-Profs. Poole points out that many public pension funds and endowments get over 9% including the Pennsylvania public employees fund PSERs which earned 9.67% over the last ten years. Why shouldn't the investment of the concession fee do as well?
Diversion doesn't pay
3-Profs is big on the dangers of a private concessionaire's toll structure diverting large amounts of traffic to local tax-supported roads. This overlooks the fact that the concessionaire has a powerful bottom-line incentive to minimize diversion because every vehicle diverted is business lost and a toll not collected.
Efficiency opportunity the value driver for a concessionaire
The Reason report says: "The great 'value driver' for a toll concession is not the opportunity for a private operator to raise tolls, as FWTRT wrongly claims, but the opportunity to dramatically improve upon the cost structure of a politically driven Turnpike Commission by eliminating patronage hiring and patronage deals. A dollar saved on costs will be a dollar pocketed as profit whereas a dollar in higher tolls will be at least partly lost to other roads. Depending on how the concession agreement is written, it may not even be a lawful option. Toll caps can focus the toll road company on earning profit through greater efficiency."
There are no toll caps in the public-public concession done under Act 44, just a statement by the Turnpike Commission that it will try to keep toll increases to an intiial 25% in 2010 and 3% thereafter.
Costs controversy
3-Profs assumed that costs of operation for a private concessionaire would be similar to those of
the Turnpike Commission. I did a comparative survey of operating costs of different tollroads in order to see how much scope there may be a concessionaire to make money through greater efficiency on the Pennsylvania Turnpike.
In all cases I worked from audited financial reports often called Comprehensive Annual Financial Reports (CAFR). Where depreciation is included in operating expenses I deducted it, because there is a certain arbitrariness to depreciation charges, and also because it can be argued it's not an operating expense. I called the ratio of operating expenses or costs (minus any depreciation) to toll revenues the "cost-take".
The results can be seen in the table nearby.
The Penn Pike is the third highest cost of 35 tollroads sampled. The Massachusetts Turnpike is tops with 79% of revenue going to operating costs. West Virginia is next at 64.5% (because tolls have been frozen to 1988 levels) and the Penn Pike is 62.4% compared to the average for public toll authorities of 42.6%.
Private concessionaires average 27.6% including the large and recently concessioned ASF in France which is relatively high cost at 57.2%. Without ASF the private sample - a small sample to be sure - is 27.6%.
Turnpike Commission reacts
The Pennsylvania Turnpike Commission reacted to our report with a claim that their operating expenses are not $369.9m as stated in their CAFR FY2007 but "at least $120m less" than that. They initially said their operating expenses for FY2007 were $249.8m (number given to the Pittsburgh Tribune Review) then in an email to me they said the number is $243.5m.
That would reduce their cost-take from 62.4% to 41.1%, about average for public sector tollroads, but still well above the private sector costs.
Trouble is neither number appears in their audited CAFR. We have repeatedly asked PTC for a detailed breakdown of their operating expenses of the kind many toll authorities routinely lay out in their annual financial report - toll collection, policing, maintenance, administration, penions etc. It is not available. They won't provide it.
Expenses directly related to capital projects are capital expenditures and if they are included in operating expenses that is an accounting error - a misclassification which auditors should have picked up. If they want to "expense" some portion of capital expenditures then honest financial accounts would report this expensing, if not in a line item separate from Operating Expenses, at least in a footnote. There is no such line item or footnote in the PTC's CAFR - no indication at all that they are expensing part of their capital projects, as they now claim.
Here is the Pennsylvania Turnpike Commission's CAFR from which we took our numbers:
http://www.paturnpike.com/pdf/PTC_CAFR_07-06.pdf
The PTC said to us in an email: "Non-operating expenses (shown in the CAFR within Operating expenses) include such items as costs associated with E-ZPass transponders, bank charges, self-insurance, legal contingency, etc." (email Capone to Samuel 09:15 Apr 11)
But bank charges, insurance, and legal costs are correctly operating expenses.
Transponders though they last five years are commonly 'expensed' as they are purchased by toll authorities, and therefore included in operating expenses. If other toll authorities expense them too then it would be misleading in a comparative study to only deduct them for the Penn Pike.
In the financial report there is a reference to 'cost of services'
(the Penn Pike term for Total Operating Expenses less Depreciation, our Operating Expenses) as including "non-capitalizable projects for maintaining the road and facilities." If projects are "non-capitalizable" then their costs are being borne as operating expenses and they are properly assigned as such.
The PTC effort to move these items out of the Operating Expenses category only came after the Reason report was published. It looks to be an opportunistic ploy to make their operating costs look lower than reported in their own official CAFR.
If their official report is wrong they should withdraw it, and prepare a new FY2007 report with accurate numbers showing a full breakdown of their expenses.
In the meantime it would be naive to accept numbers generated by the Turnpike Commission in a public relations effort to discredit a paper that points up inconvenient facts they report in their own official audited financial report.
see the Reason report here
http://www.reason.org/pb70.pdf
Text of a press release from PTC on Reason report:
Text of April 11 press release.
HEADLINE: Invalid report reflects author's skewed agenda
HARRISBURG, PA (04/11/2008; 1428)(readMedia) -- A new "study" of the PA
Turnpike's operating budget is fundamentally flawed and reaches an
erroneous conclusion that clearly reflects its author's political
agenda, Turnpike CEO Joe Brimmeier said today.
Yesterday's Reason Foundation report hinges on the assumption that the
Turnpike's cost of services includes operating and maintenance costs
only - a false assumption. The Turnpike includes both operating and
non-operating expenses in this line item.
The report showed the commission's operating costs more than doubled
during the last seven years from $181 million in 2000 to $370 million
last year. But, in reality, its actual operating-cost figures are $164.5
million in FY00 and $243.5 million in FY07 - a 48 percent increase.
As a specific example, consider that the Turnpike includes project
expenses associated with its rebuilding effort in this line item. In
2007, project expenses represented $122.6 million of non-operating
expenses. These are non-capitalized expenses related to the commission's
FY07 capital plan.
"The authors either made a fundamental error - which they should
immediately go back and correct - or they decided in advance of their
analysis what conclusion they had to reach to justify their support for
privatizing a vital public asset. Either way, the bottom line is that
the 'analysis' is worthless," said Brimmeier.
Brimmeier pointed out that the major Wall Street bond rating agencies
have consistently ranked the Turnpike as one of the most efficiently
operated toll agencies in the nation. The Turnpike earns some of the
highest ratings for its bonds as well.
Moody's Investor Service found that:
* The Commission has the lowest operations, maintenance and
administrative expenses per roadway mile of any comparable turnpike
agency in a national peer group assembled by Moody's.
* The Commission has the lowest debt per roadway mile of any similar
turnpike agency in the Moody's study.
* The Commission's five-year toll revenue growth rate for toll revenue
and total transactions is well above the median for peer group turnpike
agencies.
"The experts - the real experts - are constantly reviewing our finances
and transactions, and our record speaks for itself," Brimmeier said.
"Wall Street has reviewed our financial projections that are the
underpinnings of Act 44 and, once again, have given us favorable
ratings."
The Turnpike remains committed to implementing Act 44, which was signed
into law in July 2007 by Gov. Ed Rendell. The agency will shortly
re-submit its amended application to the Federal Highway Administration
(FHWA) seeking approval to toll I-80.
"We are working to answer all of the questions raised by FHWA. We expect
that there will be more back and forth with the agency as they do their
due diligence and conduct a thorough review," Brimmeier said.
To date, the Turnpike has provided PennDOT with $520 million in new
funds for the state's transportation systems. Upon completion of a
fourth payment to the commonwealth at the end of this month, the
Turnpike will have provided $750 million in this fiscal year alone.
"We at the Turnpike have consistently - often voluntarily - shared with
the public our financial documents, and we've freely discussed and
circulated information on Act 44 related contracts, records and
expenditures," Brimmeier concluded. "I call upon the Reason Foundation
to publicly divulge its funding sources. Perhaps then they might be
taken with some degree of sincerity."
Contact: Carl DeFebo, 717-920-7176 [end of PTC press release]
TOLLROADSnews 2008-04-13
ADDITIONS, EDITS 2008-04-14 00:30