Light Rail in Milwaukee: Public Policy Forum - Transportation Committee Milwaukee 10 July 1996
THE FUNDAMENTALS: PUBLIC PURPOSES
Any complete public policy discussion should start with the fundamentals --- the public purpose of the public service under consideration. The public purpose of public transit --- the reason that public subsidies are granted to public transit --- is to serve customers --- the riders and the taxpayers. Public transit is subsidized because of the benefits that it is capable of producing, which are expressed in three public purposes:
Mobility for those without access to automobiles (the transit dependent).In attempting to serve these public purposes. transit has the potential to help shape a better community, in which both traffic congestion and air pollution are under control. The test of transit's effectiveness is the percentage of urban travel --- in person miles --- that it attracts. In the cities of the western world --- automobile oriented as they have become --- this is no easy task. It is particularly difficult in the United States, where public transit market share has fallen so precipitously as to have become irrelevant in most metropolitan areas.
While it is true that transit has the potential to improve the urban environment, is it even more certain that contemporary American transit policies cannot deliver on the potential.
Excessive operating costs: With few exceptions, most, if not all of transit subsidies established over the last 30 years have been consumed by inordinate unit cost escalation. Overall, less than $0.25 of each new operating dollar has been used to produce new service. This is a deplorable productivity record --- by far the worst of any transportation industry in the nation. This has consumed resources that should have been used to expand transit service to better serve customers and the community.Transit's costs have risen so starkly that its costs per passenger mile (person mile) are more than four times that of the automobile.
THE SITUATION: TRANSIT IN AMERICA AND IN MILWAUKEE
It is thus no surprise that transit is in steep decline.
Since public subsidies began on a large scale in the 1960s, the percentage of people using transit for the work trip --- the "bread and butter" of transit --- has declined by 60 percent.In Milwaukee, transit is in critical condition.
From 1980 to 1990, work trip market share declined 31 percent.And new money is not the answer. It has not been in the past, and will simply not be available.
The United States has a large budget deficit, and a national debt of $50,000 per household, which and is increasing by $2,500 per year.In short, the demand for public services is great, but the supply of tax revenue is limited. Turning around transit will largely require better use of existing resources.
URBAN RAIL
Much of what I will say today may be interpreted as negative toward rail systems. But I emphasize that there is nothing inherently good or bad about rail systems. The issue is how effective, relative to other options, are new rail systems in a particular environment. In American urban areas, including Milwaukee, the answer is clear --- there are better alternatives.
It is widely assumed that expanded public transit service can play a major role in reducing air pollution and traffic congestion. Urban rail --- heavy rail or subways, and light rail --- is often viewed as a particularly effective strategy for accomplishment of these objectives. Light rail transit (LRT), a resurrection of 19th century streetcar technology, has received particular attention, and has been proposed for Milwaukee. Light rail transit, unlike subways, often operates on city streets, at grade, and provides point to point travel speeds approximately the same as buses, and considerably slower than automobiles.
In America, the performance falls pitifully short of the promise. We are told that one rail line can carry the same number of people as four, eight, or some other impressive number of freeway lanes. But they don't, because new light rail systems attract so few new riders from the driver's seat. The total daily demand on the best new American light rail system could be accommodated in a single hour on an 10 mile stretch of urban freeway. Indeed, none of the new systems own enough rail cars to carry the capacity of a single freeway lane.
Projections and Performance: The disappointing performance of new rail systems is becoming common knowledge. A late 1980s United States Department of Transportation (USDOT) report showed that:
Capital costs average 50 percent higher than projections.Since publication of the USDOT report transit agencies have begun issuing reduced ridership estimates after project commencement, which has made their performance appear better. But, in fact, only two urban rail systems have equaled the projections made when the decision to proceed was made --- San Diego and Los Angeles. I can assure, from a personal perspective, that this is an important point. Once we made the decision to spend $140 million for a light rail project in Los Angeles, there was no turning back --- even though the price tag rose to nearly $900 million.
San Diego constructed its first light rail line without federal funding, and used very conservative cost and ridership projections, which it met. Full accountability to local taxpayers provided a powerful incentive.But things have changed little since the USDOT report was published. Virtually all subsequent system have been plagued by large cost overruns.
Attraction of new riders: Study after study has shown that most urban rail riders are former bus riders --- and that only 25 percent to 40 percent are riders who didn't previously ride transit. The result is exceedingly high costs. A US Department of Transportation report indicates that:
The cost of diverting a single new rider to Atlanta's heavy rail system is $15,000 per year --- enough to lease five new Fort Tauruses for the new rider.Sources of Ridership: But the cost per automobile driver diverted is even more. Not all new transit riders are drivers. Many are riders who would not have previously taken the trip. Others are attracted by free downtown fares, as they forsake walking for transit at the lunch hour. Still others are passengers who have been stripped from car pools --- a highly efficient mode of high-occupancy vehicle travel, which requires no subsidies, and is considerably more energy efficient than either bus or rail..
And there is another dimension. A substantial portion of the ridership on some new light rail systems use light rail in conjunction with their automobiles or at times that air pollution and traffic congestion are not problems.
Some new light rail systems serve as expensive parking shuttles, as people park free at stations adjacent to the downtown area, and ride short distances to their destinations, often in free fare zones. Virtually no traffic congestion is alleviated, because the reduction in vehicle miles is so small. Similarly, almost no air pollution reduction occurs, since vehicle miles remain similar, and further, since much of automobile pollution is attributable to starting and stopping.The parking shuttle function and special event usage take few cars off congestion roadways leading to downtown, providing minimal reduction in vehicle miles traveled. Downtown, as will be discussed later, is the primary rail transit market.
Little Attraction from Autos: And that's exactly what the evidence shows. The number of cars taken off the road by new urban rail systems is barely perceptible.
In St. Louis, considered by many to be the most successful new light rail line in the United States, freeway traffic continued to grow along the corridor at virtually same rate as on other freeways.Market Share Decline Continues: Construction of new rail systems has done virtually nothing to reverse transit's declined. The work trip market share trends in new rail cities is pitiful. During the 1980s, urban rail systems were expanded or opened in 13 cities. Market share declined in 12 rail cities, increased in none, and was stable in one.
Despite building a $2 billion heavy rail system, Atlanta's work trip market share declined by 36 percent.Impact upon transit users: To better use resources, transit agencies routinely truncate bus services at light rail stations, which makes transit trips longer and more circuitous for many existing riders. This is particularly burdensome for lower income people in the inner cities, for whom an "improvement" such as light rail can retard the quality of life by requiring more time to make transit trips.
Urban Rail's Market: Downtown: The urban rail market is very limited --- even more than buses, it focuses on downtown, where employment densities are highest --- significantly higher even that suburban office centers that may have more employment. But while employment densities are higher in downtown as is, the number of jobs located outside downtown is much higher. In most metropolitan areas, downtown employment is less than 15 percent of metropolitan employment, and in virtually all areas the downtown share of employment is dropping. Moreover, a considerable percentage of downtown commuters already use buses or car pools. As a result, the downtown market offers little potential for improving transit market share. At the same time, the sparse spatial development of large suburban office parks makes them virtually inaccessible to significant bus usage --- much less rail usage. In metropolitan areas with heavy rail systems, such as San Francisco and Washington, DC, large suburban office centers located directly on rail lines attract inconsequential rail market shares.
Development: It is often claimed that urban rail generates economic development. The connection between transit infrastructure and economic development is tenuous, at best. Claims are made by advocates of both rail and busway systems --- Ottawa's claims with respect to busways are at least as impressive as the rail related development claims in Portland and St. Louis. It should be noted, however, that much of the identified development is not market driven --- taxpayer financed infrastructure, such as publicly financed convention centers, publicly financed sports arenas, tax abated retail developments, and government office buildings. Developments such as these are frequently occurring in downtown areas, regardless of whether a rail or busway system exists. Downtown development has been particularly impressive in Indianapolis, which is smaller than Portland, and has not built light rail. It would be as valid to attribute Indianapolis' development to the lack of light rail as to attribute Portland's to light rail. A number of other non-light rail cities have experienced greater development than Portland. The light rail development claims of St. Louis bear special scrutiny, where a major enclosed shopping center is faced with tax foreclosure due to insufficient business, and office vacancies are inordinately high. As transit agency commissioned report on the San Diego light rail system noted, the dynamics of urban development are so complex as to render transit improvements an insignificant factor.
Construction Disruption: Rail construction is particularly difficult for property owners, businesses, and residents along the route, due to construction disruptions. Many businesses have failed, while others have been put in jeopardy by sales losses. For residents, the quality of life is hampered by constrained access, noise, and localized air pollution. Such projects are particularly hard where rail alignments follow surface streets, such as the proposed line on Fond du Lac Avenue in Milwaukee.
Can America be Europe? Some have suggested that we should emulate the "successful" transit policies of Europe --- that we should build extensive rail systems.
It is important to understand, however, European urban population densities are much higher than American. American urban areas are decentralized, with multiple centers, while European urban areas typically have a single dominant commercial center. A single commercial center provides the concentration of destinations that permit high transit ridership. But we are much different. For example, Joel Garreau's Edge Cities indicates that the Washington area --- home to the nation's most successful new rail system --- has 16 commercial centers ("Edge Cities") that are larger than downtown Portland. Rail transit is simply incapable of effectively serving such dispersed centers. This is clearly demonstrated, not by theory, but actual experience in city after city. In Washington, the $12 billion subway has not reversed transit's continuing market share decline.
But if American urban areas are characterized by a lack of concentrated destinations, they are even more characterized by an absence of concentrated origins. US urban densities are 30 percent to 90 percent below those of European urban areas. The extent to which US urban densities have decline is illustrated by the fact that in 1990, Los Angeles became the most densely populated urbanized area, overtaking the New York urbanized area in population per square mile.  US urban densities are minuscule compared to Asian cities. For example, Hong Kong is one of the few cities in the world in which transit is profitable. Hong Kong has nearly 250,000 people per square mile. That compares to the 2,000 to 5,000 typical of US urban areas (the Milwaukee urbanized area has approximately 2,400 people per square mile). At Hong Kong's population density, metropolitan population of Milwaukee would fit within half the area of the city of Wauwatosa. At Paris metropolitan densities, the metropolitan populations of Milwaukee, Chicago, and Cincinnati could fit within the Milwaukee urbanized area.
But it is not just that America is not Europe, it is that Europe is becoming more like America. Public transit market share is declining or stable in most large European cities. Like in America, central city population densities are declining. Automobile usage is expanding rapidly, as Europeans become more affluent. Despite $5.00 per gallon gasoline, Europe is using automobiles more, and is suburbanizing. Finally, the streets of European cities are at least as congested by automobile traffic as those of American cities. Indeed some of the most severe traffic congestion in the world is to be found in European capital cities with extensive rail systems.
Cities Rejecting Rail: Trade and general press attention has been focused on the US "rail renaissance." These stories usually overlook the urban areas that have made other choices. For example:
Houston rejected an urban rail system, opting instead for system-wide bus improvements.But these developments are not all that surprising. Too much transit planning is predicated on an assumption that what the customer wants is technology --- an amusement park ride. Urban rail systems cannot make more than a marginal contribution to reducing air pollution and traffic congestion because they fail to provide what customers want --- proximity, frequency of service, and speed. Urban transportation is not Disneyland.
The Test: Auto Travel Trend: Transit can reduce traffic congestion and air pollution only by removing automobiles from the road. The test of urban rail, or and any other major public transit investment, is the number of automobiles that it removes from the road. The evidence is clear --- new urban rail systems remove very few automobiles from the road --- and as a result have no more than a negligible effect upon air pollution. In fact more circuitous trips to park and lots can increase negate even the negligible pollution and energy efficiency savings.
Incentives behind Urban Rail: If new American urban rail systems do not reduce traffic congestion and air pollution, why are they built?. The reasons have little to do with transportation. Urban rail projects, like publicly financed convention centers and domed stadia, are perceived as pre-requisites to achieving world class city status. To put this psychological need in Freudian terms, the rush to build light rail may be characterized as infrastructure envy. Urban rail has made even more attractive by the availability of federal funding --- which study after study has shown is spent less effectively than local money by local governments. As Milton Friedman reminds us, "people are more careful with their own money than with other people's money."
Supporting these incentives to build overly expensive projects is a planning processes that routinely exclude consideration of less costly and more effective options, such as.
Comparable busway and high-occupancy vehicle lane options.Moreover, background assumptions used in the planning process often support overly costly rail capital improvements. Typical deficiencies include:
Unrealistic job growth projections, especially for downtown areas.This is not to suggest that there are no corridors in which new urban rail systems make sense. Urban rail is a very effective strategy for moving large numbers of passengers in dense urban corridors. In the United States, it is likely that there few, if any such corridors that are not already served by rail systems. In the average US metropolitan area, adding a high capacity rail system makes as much sense as installing an elevator in a single story building. Even a world class elevator would provide little benefit. There is no question that over-investment in transit, through expensive rail systems, has precluded the regional transit investments that could have stopped, if not reversed transit's decline.
THE MILWAUKEE LIGHT RAIL PROPOSAL
The Milwaukee East-West Corridor Transportation Study Draft Environmental Impact Statement (DEIS), which analyzes light rail and other options, lists a number of objectives, which include the transportation related objectives of:
Reducing traffic congestionMy discussion will focus primarily on these objectives, on the assumption that the primary purpose of a transportation project is transportation, and that the relevance of other objectives is dependent upon achieving transportation benefits.
Planning Process Deficiencies: The same dynamics that operate nationally are present in Milwaukee. The planning process favors rail options over bus and ridesharing (car pool) options.
1. Inferior bus options: An equivalent bus system is not compared to the light rail options. The bus options proposed provide lower levels of service, and as a result can be expected to attract fewer customers. In fact, studied bus service increases could be nearly tripled without exceeding light rail operating costs. The planners should have considered what could have been accomplished with a bus option requiring the same operating budget as light rail.
2. Overly costly bus options: The studied bus options are more costly than necessary, in that operation would not be through competitive contracting.
3. Overly optimistic downtown employment projection. Implausible downtown job growth is projected, which skews projected light rail ridership higher. While the downtown employment market share dropped by 31 percent from 1970 to 1990, a 14 per cent market share increase is projected for 1990 to 2010. This is simply not credible. Downtown employment market share is rising virtually nowhere. Moreover, downtown oriented employment is particularly vulnerable to the decentralization likely to occur as the telecommunications revolution advances. The projected miraculous recovery of downtown Milwaukee's employment market share is simply not going to happen
 4. Insufficient attention to ridesharing potential: Car pool impacts are given short shrift. While more than 6,000 new daily transit riders are projected for bus and rail services, new car pool ridership is projected at only 1,000. The special lanes on Interstate 94 should provide much improved travel time for car pools (if they did not, there would be little reason to build them). In a community in which car pool use is four times transit use, it seems unreasonable that the increase in car pool use would be one-sixth that of transit. The opposite conclusion would be more plausible. Perhaps the increased transit ridership is accomplished through car pool stripping. This would make little sense, since the car pool energy efficiency advantage over transit is even greater in Milwaukee.
5. Exclusion of ridesharing data from cost effectiveness analysis: The pro-rail bias is further illustrated by the fact that the implausibly small increase in car pool use is excluded from the cost per new riders calculation, disadvantaging the bus and car pool options relative to light rail options. less
6. Bus system park and ride disadvantage: Fewer park and ride spaces are provided under bus and car pool options than under light rail options, biasing results in favor of light rail.
7. Anti-bus development bias: The analysis attributes long term job creation (development) to light rail, but not to bus options, again revealing a pro-rail bias.
8. Overly optimistic bus service baseline: The analysis uses a baseline that already assumes a 50 percent increase in corridor bus service and a substantial increase in transit ridership. The underlying assumption is that transit will maintain its productivity, providing new service with the new money. Yet from 1979 to 1994, Milwaukee County Transit System revenues increased by nearly five percent (inflation adjusted), while service levels dropped by 15 percent --- an annual productivity loss of -1.4 percent per year (bus miles per million dollars of operating expenditure). Without an injection of competition (through competitive contracting), it can be expected that most, if not all, of any new subsidies will be consumed by escalating unit operating costs. 9. Unfair cost loading on special lanes: The analysis disadvantages the busway-car pool special lane option (Alternative 6) by including nearly $600 million for modernization of the general purpose Interstate 94 lanes. Since these costs --- nearly equal in themselves to light rail capital costs --- would be incurred under options not involving special lanes, this biases the analysis against the highest quality, but feeble bus option. Correction of this deficiency, and inclusion of the new car pool users in the analysis would eliminate the cost efficiency per new rider advantage of light rail, even taking into consideration the considerable improvements that would derive from competitive contracting and a level of bus service comparable with that proposed for light rail. 10. Omission of travel time improvement in general purpose lanes: Under the busway-car pool alternatives, no travel time improvement is predicted in the general purpose freeway lanes. This is at odds with the experience in other cities, where special lanes have drawn many car pools away from the general purpose lanes, improving travel times on the general purpose lanes. 11. Exclusion of an freeway expansion option: Finally --- the planning process did not consider the addition of general purpose traffic lanes on Interstate 94. This is the one option --- the addition of one general purpose lane in each direction --- that could indisputably accommodate the increased demand projected on Interstate 94 (chart). Related options could have included studied bus or light rail improvements. Or, the new lanes could used as "commuter lanes," which would be available to high occupancy vehicles (HOV), such as buses and car pools, during peak hours, and to all traffic at other hours. Perhaps the omission of additional freeway capacity as an option derives from the quasi-religious fanaticism that blindly and injudiciously considers any automobile based strategy unthinkable. It would have been one thing to study the freeway expansion option and reject it. But it is indefensible to have kept it "off the table."
 In citing these deficiencies, I do not fault those who have performed the analysis, or suggest that they are biased. I am confident that they have followed existing federal planning requirements. The deficiencies noted above are rather an indictment of those federal planning requirements.
Evaluation: But even with these advantages, the light rail option provides little or no advantage over the less than robust bus and car pool options that were studied. The most effective bus and car pool `options studied provide a modestly better reduction in traffic by 2010 than the light rail option, but both provide scant relief. Light rail reduces traffic by less than two year's growth over the period ---- and considerably less than the time that will be required for construction. Throughout the corridor, light rail would reduce total vehicle trips by less than one-percent, reducing the increase in automobile demand by 1/40th between now and 2010 (chart).
 The light rail options are virtually the same or marginally less effective in reducing air pollution than bus options. The DEIS comments that "... the relative effects of transit projects (even substantial ones) on the air quality of a region or sub-regions are not dramatic." And this is confirmed by the finding that the most effective light rail option would reduce air pollution by only 0.34 percent --- pollution would remain at 99.66 percent of the baseline level. The Ivory soap test is appropriate here --- recall the advertising phrase "99 and 44/100ths percent pure," which was used to characterize Ivory soap as virtually pure. By the same test "99 and 66/100ths percent polluted is virtually polluted. Yet, presumably with a "straight face," the DEIS goes on to say that "the proposed alternatives would result in improvements to air quality because of the diversion of trips from private automobiles to transit..." Incidentally, the most effective of the feeble bus options analyzed would reduce air pollution more than light rail, but like light rail, by a virtually imperceivable amount.
 The most cost effective light rail option would cost nearly $40 per new ride --- or nearly $20,000 annually per new commuter. This is enough to lease a new Ford Tauruses for each new rider, and six of his or her friends --- enough to lease each new rider a new Lexus 400 and pay the average house mortgage. The cost per new lifetime commuter would approach $1 million.  The best light rail option would add fewer than 5,000 new one-way riders per day --- or put another way --- nearly 90 percent of light rail riders would not be attracted from automobiles. In fact, the traffic impact is even less, since many fewer automobiles will be removed from the road. During the morning peak --- at the busiest section of the light rail system, it is estimated that the new transit riders attracted to light rail could be accommodated by the addition of one or two buses. Fewer than 5,000 one-way daily riders --- 2,500 people --- can hardly justify a capital expenditure of more than $600 million. Put in context, 2,500 riders is pathetically few --- nearly twice as many people live in Wisconsin's least populated rural county.
 Moreover, the troubled history of urban rail construction leads to a reasonable expectation of a considerable capital cost overrun --- an obligation that will be up to state and local taxpayers to fund. Frankly, the capital cost estimate for the Milwaukee light rail project looks a bit light.
Light rail can be expected to provide virtually no benefit with respect to the "job and worker mismatch." Just like other cities, urban rail in Milwaukee will be structurally incapable of serving a material share of the "reverse commute" market. On the other hand, the single form of efficient that can provide measurable measurably improved access to suburban office centers, car pools, would be discouraged by selection of light rail over a busway-HOV option. This would further exacerbate the "job and worker mismatch."
Finally, the corridor in which this large capital facility would be built has lost population, and is projected to lose 10 percent of its population between 1990 and 2010. At the same time, the metropolitan area is expected to grow by at least that much. Thus, an over-capitalized option could be developed in a corridor that is in decline.
In Milwaukee, as elsewhere in the United States, light rail cannot be justified based upon transportation objectives. Traffic congestion would not be reduced, because not enough auto drivers would be attracted to light rail. And because automobiles would not be taken off the road, there would be virtually no reduction in air pollution. Other options would be more effective. The excessively high cost of the proposed light rail line makes the proverbial Defense Department $500 toilet seat look like a bargain by comparison.
Alternative estimate: The busway-car pool alternative would be far more cost effective than light rail. If comparable bus service levels were provided, competitive contracting used, incorrectly attributed capital costs removed, and a more defensible assumption made with respect to new car pool use, the cost per new rider under Alternative 6 would drop to 1/4th that of the lowest cost light rail alternative. BETTER AND MORE COST EFFECTIVE OPTIONS
There are far more effective strategies for improving urban transportation in US urban areas. I will discuss these, but will first briefly digress to consider the problems of traffic congestion and air pollution in context.
Traffic Congestion: Traffic congestion is an important concern. Yet there is a good deal of hype and hysteria with respect to traffic congestion. Cities such as New York, Chicago, Los Angeles and San Francisco have serious traffic congestion. But these cities are not typical of America. I recall being asked by a talk show host in Salt Lake City about solutions to "gridlock." I recounted having driven that during that weekday morning peak period by freeway to downtown Salt Lake City without dropping below 50 miles per hour. There are more urban areas like Salt Lake City than San Francisco.
As Alan Pisarski has put it, the American highway system demonstrated a high degree of resiliency during the 1980s. The number of single automobile commuters increased by more than a third --- by 22 million. At the same time there was virtually no new construction of urban freeways. Yet the average work trip time increased by 42 seconds, barely perceivable to the average commuter. It takes more than 42 seconds to transfer from a bus to rail, or from a parked automobile to a rail line.
In the longer term, the problem of traffic congestion could be solved by increased roadway capacity and the development of "smart highways," which I will mention later.
Air Pollution: Very real progress is being made in controlling air pollution. Since 1970, automobile produced carbon monoxide is down by more than 50 percent. Lead is down more than 95 percent. And further improvements will be achieved as cleaner cars replace those being retired. The air pollution problem is well on its way to being solved.
There are other, more effective strategies that are appropriate to the US situation.
Busways & High Occupancy Vehicle Lanes: Busways and high occupancy vehicle lanes are an option to light rail. And they are far more flexible, in that they can provide high levels of conventional downtown oriented transit service, while providing the only significant high occupancy access possible to other employment, by car pools and van pools.
Busways are capable of providing levels of service comparable with theoretical light rail capacities.
Curitaba, Brazil: A single busway is carrying nearly 300,000 passengers daily --- this is more passengers than even the most grandiose projections have predicted for new rail lines in the US, and almost 1.5 times the daily ridership of Milwaukee County Transit.Busways and car pool lanes have been successful in the US as well.
Los Angeles: The El Monte busway and HOV lane carries more daily riders than any new US light rail line. Car pool person trips exceed bus person trips, while total bus and car pool peak hour trips are nearly three times that of Portland's light rail line..Busways and high occupancy vehicle lanes have other advantages as well. A US Department of Transportation report by Harvard's John Kain and colleagues concludes that busway-HOV construction costs per rider average one-fifth those of either light rail or heavy rail systems. And commuter lanes --- which serve high occupancy traffic during peak hours and all traffic at other times are 99 percent less expensive. And bus operating costs are considerably less.
 Buses on Freeways: High levels of bus service can also be effective in the general purpose lanes of freeways. The most cost effective transit service in the nation relies primarily on general purpose lanes. A large New Jersey company provides commuter service into New York City for lower operating costs than the best commuter rail, heavy rail and light rail systems in the nation. A number of competitively operated express bus systems produce service for less than the best light rail systems.
 Corridors: Rail and HOV: There is a fundamental reason for the superiority of busways and HOV lanes compared to new urban rail systems --- it has to do with corridors. A rail corridor is very narrow. To commute by rail, the passenger must either live within walking distance (live inside the corridor), or must drive to a park and ride lot or ride a connecting bus. The transfer from automobile or bus to rail takes time, and is partially responsible for the inferior travel times characteristic of rail. Moreover, people generally seek to avoid transfers between mode, not least because of the additional time consumed. On the other hand, busway and car pool corridors are much wider. A bus may circulate in a neighborhood, and then enter a busway, providing a quicker trip by eliminating the need for a transfer. Similarly, a car pool can originate miles from the HOV lane, enter the lane and provide an expedited trip.
Other Strategies: Technological advances are also on the horizon. As the telecommunications revolution proceeds, more people will telecommute, which could reduce peak period travel demand. The development of "smart" highways will increase the capacity of the present roadway system; first by improving system management, and perhaps ultimately by taking controlling of vehicle movements on congested corridors --- ushering in a consumer friendly combination of automobile and mass transit technology.
Comprehensive Regional Transit Improvements: The last two decades of US transit history make one point crystal clear --- that the best way to increase transit ridership is to increase transit service throughout the urban area. Virtually all of the handful of transit systems that have achieved material ridership increases have done so by significantly increasing transit services throughout their service area, rather than concentrating efforts in one or a few corridors. Examples include Los Angeles, San Jose, Houston, Phoenix, Austin, and San Diego. Indeed, the only two metropolitan areas to improve their work trip market shares in the 1980s relied on significant regional bus improvements, and were rewarded with work trip market share gains --- Houston at 27 percent and Phoenix at five percent. The metropolitan area that preserved its market share --- San Diego --- will be discussed in greater detail, because, it alone has significantly increased ridership without substantially increasing operating expenditures..
A TALE OF TWO CITIES: MILWAUKEE AND SAN DIEGO
Nearly a decade ago, a Milwaukee transit official criticized me in the metropolitan press for having compared the progress that San Diego had made in improving its bus system to Milwaukee. The point of the criticism was that it was inappropriate to compare transit trends in a non-transit oriented metropolitan area like San Diego to those in a transit oriented area like Milwaukee.
How times have changed. In 1986, Milwaukee transit ridership was more than double that of San Diego. Now San Diego carries more riders than Milwaukee. It has achieved this by increasing service --- by putting customers first. Two thirds of San Diego's ridership increase has been on buses. While the percentage of costs covered by fares has dropped more than on fifth in Milwaukee, San Diego now has the highest fare to operating cost ratio in of any metropolitan transit system in the United States. This has been accomplished by controlling costs. What has happened in San Diego holds important lessons for Milwaukee.
 The story begins in 1979. Local officials, concerned at the rising costs of public transit, began to provide some routes through competitive contracting --- the process of awarding contracts to the most responsible and responsive private bidders through competitive procurement. By 1994, 35 percent of San Diego bus service was operated under competitive contract, with contracts awarded not only to private bidders, but also to the former public transit monopoly. Competitively contracted services are 35 percent less costly than services not subject to competitive procurement.
The results have been astounding. Between 1979 and 1994, San Diego bus: • Fare and subsidy revenue rose approximately six percent (inflation adjusted).In response, bus ridership has increased by 25 percent. Annual productivity is improving at 3.6 percent. The public purposes has been served as follows: • Each new dollar of expenditure has produced $8.22 in new service (inflation adjusted).By comparison, between 1979 and 1994 at Milwaukee County Transit: • Fare and subsidy revenue rose approximately five percent (inflation adjusted).And, in response bus ridership has dropped by 35 percent. Annual productivity, as noted above, has declined 1.4 percent annually. The public purpose has been served as follows: • Each new dollar of expenditures has produced a $3.43 in service reductions (inflation adjusted).This comparison is not intended as a criticism of Milwaukee County Transit management. Frankly, the incentive structure in Milwaukee, and in most US transit systems is counter-productive to the public purpose. The US model --- and rapidly disappearing western world model --- for of organizing public transit grants higher stakes in transit to interests other than customers. The most skilled management could do little to make the situation materially better. In San Diego, the organizational structure of transit corresponds to the "separation of policy from operations" model that has emerged in Europe, Australia, and New Zealand. Responding to competition, the former public monopoly operator in San Diego has reduced its costs per mile by 23 percent (inflation adjusted), compared to the nearly 20 percent inflation adjusted increase at Milwaukee County Transit. San Diego system-wide bus unit costs declined 33.2 percent between 1979 and 1995 with annual savings of $34 million and total savings of $275 million (inflation adjusted). These savings not only financed significant bus service expansion, but also provided a major source of funding for what is the nation's most successful new light rail system in terms of ridership, capital costs, and operating costs (which are little more than public agency bus costs, even after the much improved bus cost performance).
The Competitive Contracting Revolution: And San Diego is not unique. A number of US transit agencies are using competitive contracting to lower costs, and improve service quality. Metropolitan areas such as Dallas, Austin, Denver, and Los Angeles now competitively contract 20 percent or more of their transit service, at savings ranging from 30 percent to 60 percent.
Internationally the progress is even more significant, as national and regional governments have recognized that the public monopoly approach to transit is a strategy that leads, at best, to managed decline.
London competitively contracts 55 percent of its service, and will complete conversion by 2000. While reducing its operating expenses by 27 percent (inflation adjusted), bus service levels have been increased by 26 percent. Milwaukee's transit decline over the past 15 years was unnecessary and avoidable. The best hope for increasing transit ridership in Milwaukee is to use the competitive market to reduce costs, thereby generating funding to increase service levels and attract higher ridership. By placing the interests of customers above all others, Milwaukee can be as successful as San Diego, through comprehensive, cost effective regional transit improvements.
CUSTOMERS: THE PUBLIC PURPOSE
What is required is no less than a resolute determination to serve customers, the riders and taxpayers. Customers must be served before managers; before employees; before developers and elected officials; and before interests that would profit by producing unnecessarily costly infrastructure.
Existing and potential transit riders have simple desires. They are summarized in three service characteristics:
Proximity. service that is conveniently close to both their trip origin and destination.Transit service is a means, not an end in itself. A USDOT study has shown that customers are primarily interested in mobility, not in whether the means of mobility is bus or rail. The proclivity toward more costly than necessary infrastructure and more costly than necessary services must cease. Transit can fulfill its public purpose only by using the resources provided by riders and taxpayers to the best effect. .
| ||||||||||||||||||||
