| |
 |
In
1999 TriMet initiated an experimental shuttle in the Cedar Mill neighborhood
on Portland’s west side. Cedar Mill is a low-density area with many
hills and cul-de-sacs, making it impractical for traditional fixed-route
bus service. TriMet contracted with Sassy Cab Company to use two vans
to provide specialized service within the neighborhood. Riders were required
to call a day ahead of time to make reservations, and they would be provided
door-to-door transit anywhere within the neighborhood. The price was the
same fare required for a TriMet bus or light rail ride. A primary (if
unstated) purpose of the service was to boost ridership on TriMet’s
Westside light rail line by bringing riders to the Sunset Transit Center.
Sassy received $12,700 a month for which they provided service throughout
the Cedar Mill region for 14 hours a day, five days per week. The company
made a small profit from this work.
Although the private sector service was working well, TriMet brought it
in-house in September 2002. This
was a requirement of its labor contract with the Amalgamated Transit Union
(ATU). Expenses immediately increased by 18 percent and service was cut
by 50 percent. Under Sassy the service was all-day; now it’s only
during the morning and afternoon peak periods. The Sassy service cost
$24 per driver-hour; the unionized service is costing $49 per driver-hour.
Given TriMet’s willingness to use labor priced at 100 percent above
the market rate, one might surmise that the agency has a lot of cash.
In fact, however, the opposite is true; TriMet is facing a fiscal crisis.
The agency currently has a bill pending in the Oregon legislature to raise
its regional payroll tax from the current rate of $6.22 per thousand of
payroll to $7.22 over a ten-year period. That tax is TriMet’s largest
source of revenue and brought in more than $155 million last year.
Why would the agency refuse to use less expensive outside labor? Because
TriMet managers don’t really have a choice. They gave away their
rights to decide a decade ago.
THE SUMMER OF ’94
The roots of TriMet’s financial crisis go back to February 1994,
when TriMet management and the ATU reached a verbal agreement on a new
contract that would vastly increase costs for the agency over the next
10 years. The agreement, negotiated behind closed doors, reduced the agency’s
ability to contract out work to lower-cost firms, increased wages, and
dramatically raised the long-term cost of employee benefits.
Senior TriMet staff negotiated the contract. The terms of the agreement
were so secret that for months even the TriMet board did not know what
had been promised to the union. When details became known, it immediately
created tension between Loren Wyss (Board Chairman for 8 years) and General
Manager Tom Walsh, who had approved the settlement. Wyss thought that
the ATU had received most of the benefits of the contract while giving
up nothing in return.
The rift grew over subsequent months. Walsh sent a four-page memo in May
1994 to Bob Stacey of Gov. Barbara Roberts’ office: “Tri-Met
needs either a new board president or a new general manager.” In
early August, Wyss was summoned to the governor’s office to discuss
the conflict. Midway through the meeting
Wyss resigned.
At the time, Walsh claimed to have had no prior contact with the governor,
who appoints the seven TriMet board members. Six days later, after the
Oregonian published a story quoting from the May memo, it was clear that
he had helped orchestrate Wyss’ ouster.
In December of that year, both the ATU and the TriMet board ratified the
contract. A few days later, in a scathing letter to the editor of the
Oregonian, Wyss predicted that the labor agreement would doom TriMet to
rising costs and declining productivity. He wrote:
“...the contract just approved by Tri-Met union employees will protect
all its members from additional contributions to their pensions for 10
years. It will also guarantee 3 percent minimum wage increases in the
future, no matter what happens to the economy or to other public budgets.
“And that’s not all. Unused sick leave should add annual year-end
bonuses of $500 to $1,000 for those with perfect attendance records; every
single dollar of health, welfare, dental and vision plans will be paid
for by the public employer; retirement age will decline to 58 within 10
years; and nonunion operators who have helped keep down the cost of shuttle
lines will be forced to join the union, at large increases in wage expense.
“It now appears that the Amalgamated Transit Union has succeeded
in pulling off one of the greatest coups in the history of public employment
in our city, while the press is focused on the voting of a few public
officials. Who is guarding the public interest in this issue? Where is
the discussion?
“If there is one predictable reason for transit to fail its mission,
it is the burden of fixed costs, which this contract guarantees.”
THE DEATH SPIRAL BEGINS
During the mid-1990s, as the Portland economy went through the most sustained
period of growth in history, cash from the payroll tax poured into TriMet’s
offices in record amounts. At the time of the union contract signing in
1994, the payroll tax was bringing in $89.5 million per year. Within eight
years it had gone up to $155.5 million. This accounted for 57 percent
of all operating revenues in Fiscal Year (FY) 2001-02.
TriMet opened Westside light rail in 1998 and Airport MAX in 2001, initiated
construction on the North Interstate rail line, received political approval
for three more rail lines (at a likely cost of over $2 billion), and began
operating the Portland Streetcar. The good times seemed to be rolling.
Unfortunately, the economy began tanking in 2000. TriMet’s payroll
tax revenues peaked in FY 2000-01, then declined by three percent the
following year. Revenues increased by only 1.1 percent in FY 2002-03.
Yet labor costs were growing rapidly. According to agency records, between
1994 and FY 2003-04 the number of employees increased 25 percent, total
salaries and wages went up 75 percent, and the cost of health benefits
increased 179 percent. Annual pension costs for union members went up
242 percent between 1993 and 2002.
The total cost of all fringe benefits was 60 percent of payroll in 2001
and 65 percent in 2002; they are likely to exceed 70 percent in 2003.
In essence, employee benefits are rapidly cannibalizing the general fund
budget.
There is also a growing problem with unfunded pension liability. TriMet,
like many public employers, operates a “defined benefit” retirement
plan for most of its employees (separate from the state program, PERS).
This means once TriMet employees retire, they are eligible to receive
certain benefits such as monthly income and health insurance for as long
as they live. TriMet is required to put money into the pension fund every
year to ensure that enough cash is available to pay for the obligations
to current retirees. However, if the fund does not have enough money to
pay for all the estimated obligations to future retirees, that creates
what is known as an unfunded accrued liability.
In 1993 the unfunded pension liability for the bargaining unit plan was
$38 million. By 1997 it had grown to $61 million, and by June 30 of 2002
it was $112.4 million. The growth in liability has outpaced TriMet’s
growth in employment, as evidenced by the fact that in 1993 the unfunded
pension liability per TriMet employee was $18,630 and by 2002 it had increased
to $43,230.
TRIMET’S CITIZEN
ADVISORY COMMITTEE:
FROM WATCHDOG TO LAPDOG
Each year TriMet convenes an outside group called the Citizen Advisory
Committee (CAC) for the budget. The CAC is a seven-person group of volunteers
that reviews financial data each year prior to formal budget adoption
by the TriMet board. Individuals generally serve three-year terms, and
the committee meets approximately two times per month from December through
March. Each year the committee releases a report in April or early May,
commenting on various aspects of the proposed budget.
In 1995 Tony Rufolo, a professor from Portland State University who teaches
transportation economics, chaired the committee. When the committee’s
report was released that year, it had a harsh assessment of TriMet’s
performance. On the opening page the report states, “It does not
appear that simply providing additional service will achieve the ridership
goals that have been set by TriMet.” The report continues, “While
ridership has been growing, other forms of travel have been growing more
rapidly, so that TriMet serves a smaller share of regional trips. This
trend is clearly
the opposite of expectations generated by the
Strategic Plan.”
The committee report criticized the 1994 labor contract for raising costs
and reducing flexibility: “The CAC continues to believe that the
agency must look to alternative service provision for low-density areas
and must work more closely with local governments, employers, and others
in developing innovative methods to improve service. The recent labor
agreement that changed alternative service from a contracted status to
an in-house service is seen as a roadblock to achieving these goals. In-house
provision is more costly than contracting and greatly reduces flexibility
in providing alternative service.”
The committee was also concerned that TriMet was increasingly relying
on non-passenger sources of revenues to fund operations. The report notes
that TriMet had a stated goal of getting 30 percent of its operating revenue
from passenger fares, but that the goal had never been met and there appeared
to be no commitment to making it happen.
Professor Rufolo presented the report to the TriMet board at a public
meeting in April 1995, and he didn’t sugarcoat things. He said,
“We’ve seen the goals, and we’ve seen the strategic
plan. We just don’t see the connection between the two.”
The TriMet board was stunned by this review. Board member Shirley Huffman
told the Oregonian, “This
is the most frank of the budget reports that I have
ever read.”
Unfortunately, the CAC advice did not lead to any noticeable change at
TriMet. Apparently some people on the board rejected the message because
they didn’t like the messenger. Nita Brueggeman, a TriMet board
member at the time, says that although the CAC played “an incredibly
influential role in the formation of the budget,” Tony Rufolo “was
anti-public transportation so his dissertation wasn’t very helpful.”
Rufolo’s term on the CAC expired in 1995 and he left the committee.
Never again did the CAC provide such a blunt reality check. In 1996 the
report mentioned concerns (again) about the high costs of the 1994 labor
contract, and reiterated the desire to recover 30 percent of operating
revenues from the farebox (it was about 24 percent at the time). In each
succeeding year, the CAC reports became increasingly optimistic about
the future and less critical of the past.
The 2003 report, released on May 14, is essentially a puff piece that
shows little in the way of independent thinking. For instance, in a section
on the Interstate MAX entitled “Business Vitality,” the report
states, “It was exciting to learn that more businesses exist today
along the Interstate MAX alignment than before construction began. All
of this in spite of the fact that construction has taken place in a depressed
economic environment! This demonstrates (a) the effectiveness
of TriMet efforts to support existing local businesses during the construction
period and (b) how light rail can serve as a catalyst encouraging economic
development, even during very difficult economic times.”
There is no mention of the multiple businesses that have been forced to
shut down due to the disruptive effects of rail construction. Nor is there
any apparent understanding of how many subsidies have been required to
entice development along Portland light rail lines, including 10-year
property tax abatements, grants, and reductions in system development
charges.
Elsewhere the report describes the Washington County commuter rail line
planned along the HW 217 corridor as a “nice ‘test case’
where a high-speed rail alternative closely parallels a major highway
and Interstate system.” In fact there is no “high-speed”
rail being planned in Washington County or anywhere else in Oregon. The
commuter line will average 36 mph with a top speed of 60 mph. The 18-mile
trip will take 30 minutes, not including waiting time.
Moreover, we already have a test case for rail paralleling an Interstate
Highway: Eastside MAX. After 16 years of experience, we know that it’s
not cost-effective and it doesn’t relieve highway congestion. It’s
not really necessary to learn this again.
Rather than looking out for taxpayers, the CAC was so excited about TriMet’s
proposed payroll tax increase that they even reprinted a glowing editorial
from the Oregonian in the report—the first time an editorial has
ever been included.
From 1996 to 2003, none of the CAC reports look at the problem of rising
labor costs. Mary Fetsch, TriMet’s communication director, confirms,
“The CAC has not discussed the issue of employee benefits.”
This is hard to fathom given the widespread media attention in the past
12 months to the $15 billion unfunded pension liability of the Oregon
Public Employee Pension System.
Today, Professor Rufolo comments, “Tom Walsh was furious with the
article that Gordon Oliver (reporter for the Oregonian) wrote and I was
told afterwards that the TriMet staff were instructed to ‘manage’
the CAC process better to avoid such critical statements in the future.”
If that’s the case, the strategy has clearly worked.
OPTING OUT OF TRIMET: SMALL IS BEAUTIFUL
The experience with the Cedar Mill Shuttle provides a rare glimpse into
the real-world differences between TriMet’s cost of service and
the competitive market. Additional insights can be gained by looking at
the track record of small jurisdictions that have used special provisions
of TriMet’s enabling statute to leave the service district. Wilsonville
withdrew from TriMet in 1988, Molalla and Happy Valley in 1989, Sandy
in 2000, and Canby in 2002. The results for these communities since then
show that smaller transit districts are far more cost-effective than TriMet.
Wilsonville provides a much higher level of service than was formerly
provided by TriMet, and customers get to ride without paying fares. Operating
revenues are paid through the same type of regional payroll tax that TriMet
uses, yet Wilsonville’s tax rate is less than one-half that of TriMet’s—$3
per thousand of payroll.
After Wilsonville first withdrew, it signed a contract in 1989 with Buck
Ambulance to provide door-to-door transit service. In 1991 the first passenger
van was purchased, and gradually over the course of the decade additional
equipment was bought and services expanded. Annual boardings were less
than 20,000 in 1992; they exceeded 160,000 in 2000. Service is provided
to Salem, Oregon City, Canby, and to connection points with TriMet buses.
Wilsonville has been able to expand without raising the payroll tax in
part because the districts cost per route has been declining. The cost
per hour of operations in 1992 was $78. It dropped to $62 in 1995 and
dropped further to $58 by 1998.
In Sandy the service between Gresham Transit Center and Sandy went from
peak-hour only under TriMet’s management to all day plus Saturday
service when the city took it over, and fares were eliminated. Operating
revenue is paid for by a local payroll tax that is slightly lower than
the former TriMet tax. In the three-plus years of operation, Sandy ridership
has gone up over 600 percent.
When Sandy
set up its service, it took competitive bids from private firms to operate
the shuttles. The city has one employee who works full-time as the transit
coordinator. Costs are low because there are few employees, virtually
no management bureaucracy, and the system is entirely road-based. That
means the city does not have special responsibilities to maintain infrastructure,
as it would if it used fixed rail; the road system is paid for by millions
of other users through their gas taxes and weight-mile fees.
[ ]
|
|
TriMet
cannot remotely compete with Wilsonville, Sandy or the other smaller jurisdictions
because it is a big, inefficient monopoly. The supposed “economies
of scale” that many people thought would result from the government-mandated
takeover of Rose City Transit in 1969 never happened. What we’ve
learned is that government monopolies create diseconomies of scale, primarily
in the form of administrative overhead (e.g., fulltime administrators,
labor relations personnel, lobbyists, and human resource specialists),
long-term liabilities in pensions and benefits, and expensive upkeep of
fixed-rail right-of-way whose costs must be borne entirely by the transit
agency.
One can actually trace the transformation of TriMet from a relatively
small, user-financed agency to a bloated, tax-supported bureaucracy by
looking at several numerical trends. For instance, in 1971 drivers constituted
73 percent of TriMet’s work force. By 1999, this had declined to
49 percent. It’s hard to say what the other 51 percent of employees
are doing to actually improve transit.
In 1971, passenger fare revenue paid for 38 percent of all costs. Passenger
fares declined to 25 percent of costs in 1981, 17 percent in 1991, and
9 percent in 2001. As users paid less of the real costs, they actually
became less important to TriMet. The payroll tax is now much more important
than ridership, because in a good economy the money flows in regardless
of whether TriMet provides service that anyone cares about.
TRI-MET RESPONDS
TriMet has taken several measures in recent years to reduce the high costs
of labor, most notably in pension reform. In July 2001 the agency developed
a new defined contribution pension plan for management and union employees
and submitted it to the IRS. The plan provides for TriMet to make a fixed
contribution of eight percent of the employee’s salary into their
pension account. According to TriMet General Manager Fred Hansen, “There
is no guaranteed return on investment; each employee will have the ability
to direct his/her investments in a variety of investment options, and
the ultimate retirement benefit will depend on each employee’s account
balance at the time of distribution of the account.”
In recommending approval of the plan to the TriMet board earlier this
year, Hansen said, “Not only does this new plan give employees control
over their own investments, it also provides TriMet with a great deal
more certainty in the funding of the retirement plan.” He also noted
that the portable nature of the plan, whereby each individual owns their
retirement account and can take it with them when they leave, is more
appealing to many contemporary workers who do not envision themselves
staying with the same employer for 30 years or more.
The defined contribution plan will be mandatory for all new management
and staff employees hired after April 27, 2003 and optional for current
employees.
While this was an important step forward, in many respects it’s
too little and too late. Three trends are driving TriMet towards financial
ruin. The first is the long-term financial obligations already incurred
to current employees. Regardless of the new pension plan, TriMet is still
on the hook for expensive fringe benefits that will be owed to current
employees for decades to come.
The second is the inability to contract out work to lower-cost labor.
Previous budget committee reports noted that the expanding market of suburb-to-suburb
travel is not one easily served by TriMet’s traditional reliance
on fixed-route service. As regional land-use patterns continue to decentralize,
the agency will need to experiment with more flexible, small-scale shuttles
using outside labor. But TriMet’s union agreement makes that almost
impossible.
The final trend is TriMet’s suicidal commitment to fixed rail, including
the regional light rail program, the Portland Streetcar, and now the Washington
County commuter rail line, which will run from Beaverton Transit Center
to Wilsonville. Even though the agency receives federal grants to help
pay for construction, the operation and maintenance of fixed rail is extremely
capital-intensive. Costs of track maintenance, overhead wire repair, and
upkeep associated with the many park-n-ride facilities (which are built
mostly for rail, not bus service) must be borne exclusively by TriMet.
The revenue simply isn’t there from passenger fares to pay for all
this.
HISTORY REPEATING ITSELF
Although 1994 was a landmark date in TriMet labor relations history, it
was not the only instance where politics trumped common sense. In 1968,
when transit service was still being provided by the privately owned Rose
City Transit Company (RCT), the mayor of Portland intervened in labor
contract negotiations because of fears of a possible strike. The drivers
wound up with a 45-cent per-hour raise in one year, 4.5 times the average
annual increase they had received in the previous three years.
That by itself may not have been so bad, but when Rose City subsequently
sought to raise passenger fares from 35 cents to 40 cents to cover the
increased wages, the Portland City Council, which regulated rates as a
condition of the RCT monopoly franchise, denied the rate increase and
then terminated the franchise agreement. The City Council then succeeded
in getting a bill passed in the state legislature authorizing the creation
of a government-run transit district to take over service. TriMet used
the powers of eminent domain to take over Rose City’s assets, and
implemented a payroll tax in order to help fund the system.
This was supposed to be a small, temporary tax used to jump-start the
new agency. According to an official TriMet account of that period, “After
looking at seven taxing options—of the array authorized by the legislature—Tri-Met
adopted a payroll tax of one-half of one percent, to become effective
on January 1, 1970, and hopefully to be reduced to 1/10 or 2/10 of one
percent as soon as possible” [emphasis in original].
Of course, things never worked out that way. The new tax rate was permanent,
and after the dust settled, TriMet was receiving the revenue equivalent
of what would have been a 25-cent fare increase for Rose City Transit,
if the City Council had been willing to grant such an increase. That should
have been the first clue for local taxpayers that having the government
run a “non-profit” transit business was going to be a costly
affair.
Moreover, much of the money was immediately used to increase TriMet wages.
According to the TriMet history, “For some time, driver wages [at
Rose City Transit] had been below those of comparable west coast cities.
TriMet’s first decision was to conclude a new 18-month contract
with the transit workers’ union, bringing their pay scale to
a status comparable to other West Coast transit contracts.”
It’s clear that neither the TriMet board nor local elected officials
have ever been overly concerned about protecting taxpayers. TriMet’s
current effort to raise the payroll tax rate simply continues a trend
that began the day the Portland City Council used the government’s
power of eminent domain to put a private bus company out of business.
WHAT TO DO
TriMet’s request for increased tax revenue could hardly come at
a worse time for Portland businesses. Some are already leaving Portland
to escape its anti-business climate, and a bump in taxes for transit services
that many don’t even use will push others over the edge. Before
a tax rate is authorized, a number of other measures should be considered.
One is to abolish Fareless Square. The policy of free transit in downtown
Portland was originally motivated by compliance with the Clean Air Act,
but urban air quality is no longer a pressing concern. The air is better
in Portland than it’s probably been in 50 years and it’s likely
to get even better due to technological innovation.
Fareless Square presents a multitude of problems: it artificially inflates
ridership; it slows down service by increasing the time it takes to let
passengers on and off; it encourages Portland’s population of transients
to ride back and forth to Lloyd Center, to the detriment of legitimate
passengers; and it costs millions of dollars in foregone revenue.
If transit is actually a valuable service, it’s worth paying for
it at least in part by its users. Getting rid of Fareless Square would
help TriMet reach its long-stated goal of raising 30 percent of operating
revenues from the farebox.
Another step toward fiscal sanity would be to abandon the quixotic rail
building campaign. All of the rail lines currently on the planning agenda—light
rail to Clackamas Town Center, Milwaukie and Clark County, along with
the Wilsonville-to-Beaverton commuter rail line—are projects in
highly suburbanized communities where there is no natural ridership base.
Rail is irrelevant to a region that is steadily decentralizing, and the
effort to prop up this 19th century system will require massive increases
in tax subsidies.
Another change worth considering is mandating certain levels of competitive
bidding for transit routes. The TriMet board could actually be split off
from the agency and given a new mission of setting routes and then taking
bids for the service. TriMet should be required to bid on every line,
and if a private competitor has a lower-cost bid, they should get the
job. This would introduce much-need market discipline to future labor
negotiations.
Although these may seem like long-shot proposals, the PERS crisis has
sensitized many political leaders to the problem of unfunded long-term
liabilities in public employment contracts. Gov. Kulongoski has been willing
to embrace a “tough love” approach to his union backers, despite
their outrage, so there’s no reason why the legislature couldn’t
take the same approach with TriMet.
On the other hand, it’s hard to see why legislators would take this
step when the lobbying organizations representing many of the large employers
in the Portland region refuse to stand up and defend themselves. At a
May legislative hearing in the House Transportation Committee, not a single
business representative spoke against the TriMet tax bill. In fact, Lynn
Lundquist of the Oregon Business Association and Mike Salsgiver of the
Portland Business Alliance endorsed it.
The Westside Economic Alliance, a group representing employers in the
most economically productive county in the state—Washington County—has
remained silent publicly, even though the payroll tax forces large suburban
employers to subsidize downtown interests because of TriMet’s routing
system that directs most service to the central city.
As of this writing, the TriMet legislation has passed the Senate and is
still pending in the House Rules Committee. Given the difficulty of getting
new tax measures passed this year, the fate of the bill may not be decided
until the final days of the session. The outcome will dictate whether
transit in Portland becomes more innovative, or simply more expensive.
(Several TriMet board members were contacted for this story. None returned
phone calls.)
John A. Charles is a contributing editor to BrainstormNW and is environmental
policy director at Cascade Policy Institute, a free-market think tank
in Portland. Previous essays on transit policy can be found at www.cascadepolicy.org.
SIDEBAR: UNION COMPENSATION
Early proponents of TriMet were confident that a government monopoly would
provide more cost-effective transit than a private firm because it would
not have to pay taxes or generate dividends to stockholders. While these
may provide some advantages, they are dwarfed by the fiscal effects of
unionization in the workforce, which is much more prevalent among public
employees. According to the Bureau of Labor Statistics:
· In March 2002, overall total compensation costs were 44 percent
higher among state and local government employers ($31.29 per hour worked)
than among private sector employers ($21.71 per hour worked).
· State and local government employers? wages and salary costs
were 40 percent higher than among private sector employers and benefit
costs were 55 percent higher.
· In 2002, 13.2 percent of wage and salary workers were union members.
Within government, 37.5 percent of workers were union members, compared
with 8.5 percent of employees in private sector industries.
· In 2002, full-time wage and salary workers who were union members
had median usual weekly earnings of $740, compared with a median of $587
for wage and salary workers who were not represented by unions.
According
to TriMet’s budget for FY 2003-04, the cost of benefits in the coming
year will be:
· Health and welfare benefits: $10,061 per full-time union employee,
$9,255 per management employee.
· Disability and life insurance: $347 per union employee, $414
per management employee.
· Pension expense: $9,767 per union employee, 13.7 percent of gross
income per management employee. The pension expense also includes the
current year portion of
30-year funding of unfunded actuarial accrued liability.
· Sick and vacation payoff expense: $260 per union employee.
|
