Thinking Outside the Rails on Transit


To many in the transit business – that is, people who seek to profit from the development and growth of buses, trains and streetcars – Southern California is often seen as a paradise lost, a former bastion of streetcar lines that crossed the region and sparked much of its early development. Today, billions are being spent to revive the region’s transit legacy.

Like many old ideas that attract fashionable support, this idea, on its surface, is appealing. Yet, in reality, the focus on mass transit, however fashionable, represents part of an expensive, largely misguided and likely doomed attempt to re-engineer the region away from its long-established dispersed, multipolar and auto-dependent form.

Traditional transit works best when a large number of commuters work in a central district easily accessible by trains or buses. New York and Washington, D.C., where up to 20 percent of the regional workforces labor downtown (the central business district), are ideal for transit. Even in those metropolitan areas, however, the auto is king.

In contrast, less than 3 percent of Southern Californians work in downtown Los Angeles. Overall, despite all the money sunk into new rail lines around the country, Americans’ transit commuting is overwhelmingly concentrated in a few older “legacy” cities. Altogether, 55 percent of transit work trips are to six core cities: New York, Chicago, Philadelphia, San Francisco, Boston and Washington, and 60 percent of those commutes are to downtown.

In contrast, in the Los Angeles-Orange County region, barely 6 percent of workers take transit, one-fifth the rate in New York. Yet we’re a bunch of committed strap-hangers compared with Phoenix, Atlanta, Charlotte, N.C., and Dallas-Fort Worth, where, despite surfeits of new trains and streetcars, 2 percent or less of commuters use public transit. Even in Portland, Ore., widely proclaimed the exemplar of new urbanism and transit investment, the percentage of commuters taking transit is less today than in 1980. Portland is now contemplating cutbacks that could eventually eliminate up to 70 percent of its transit service.

Imposing Past on Future

This miserable record reflects how trains, a largely 19th century technology, have limited utility in a contemporary setting. Indeed, the only way to make it work, planners insist, is if the population is moved from their low-density neighborhoods to high-density “pack and stack” areas near transit stops, while suburban businesses are dragooned to denser downtown locations. This is the essence of the recently approved Bay Area Plan.

Although these kinds of strategies have never materially reduced automobile use – the Bay Area Plan itself says automobile use will still increase by 18 percent over 30 years – the bureaucratic logic here is almost Stalinesque in the scope of its social-engineering ambitions. As Bay Area journalist and plan advocate John Wildermuth puts it, people know they should take transit but don’t because it’s very inconvenient. But by forcing three quarters of new residents into dense housing, some with no parking, he reasons, it then will be “easier for them to either give up their cars or, at least, use them a lot less.”

Yet getting people to change their way of life, as many central planners have discovered, is not as easy as it seems. The highly dispersed San Jose-Silicon Valley area, the economic epicenter of the Bay Area and worldwide information technology, has a commute trip market share barely a third of major metropolitan area average… . Building “one of the longest” light rail systems in the United States in 50 years has barely moved the percentage of transit commuters over the past three decades.

What the Bay Area Plan will probably accomplish is to boost housing prices ever further out of reach, both in urban areas and in the suburbs. With new single-family development effectively all but banned, prices of homes in the Bay Area already are again rising far faster than the national average and now are approaching two and half times higher, based on income, than in competitor regions such as Salt Lake City, Phoenix, Dallas-Fort Worth, Austin, Tex., Houston or Raleigh, N.C.

Environmental Imperative?

Greens and their allies in the high-density housing lobby long have suggested that “peak oil” and rising prices will inevitably drive suburbanites out their cars. But, clearly, recent advances in U.S. oil and natural gas production may have already made this moot. Transit activists increasingly have focused on climate change to justify massive spending on expanding transit and forcing recalcitrant suburbanites from their cars.

This logic is largely based on the notion that suburbanites must travel greater distances to work. Yet, a study by McKinsey & Co. and the Conference Board found that – largely because of the impact of higher energy standards for cars forecast by the Department of Energy – sufficient greenhouse gas emission reductions can be achieved without reducing driving or necessitate “a shift to denser urban housing.”

The fundamental limitations of transit in dispersed cities further weakens environmentalists’ claim. Ridership on some transit systems is so sparse that cars are more energy efficient. Then, there’s the oft-mistaken assumption that higher-density housing will reduce congestion and travel. But in multipolar areas like Southern California, traffic congestion and resultant pollution generally becomes worse with higher density.

There may be other, more technologically savvy ways to reduce emissions and energy use. People have cut automobile use the past three years but their reduced travel is not showing up so much in transit usage, but, rather, is driven by other factors such as unemployment and the high price of gasoline.

But, arguably the biggest reduction can be traced to the rise of telecommuting. Over the past decade, the country added some 1.7 million telecommuters, almost twice the much-ballyhooed increase of 900,000 transit riders. In Southern California, the number of home-based workers grew 35 percent, three times the increase for transit usage. By 2020, according to projections from demographer Wendell Cox, telecommuting should pass transit, both nationally and in this region, in total numbers.

What About the Poor?

Perhaps the most compelling argument for transit stems from serving those populations – the poor, students, minorities – who often lack access to a private car. Yet, for workers in newer cities, public transit often is not an effective alternative. Brookings Institution research indicates that less than 5 percent of the jobs in the Los Angeles and Riverside-San Bernardino areas are within reach of the average employee within 45 minutes, using transit. The figure is less than 10 percent in the San Jose metropolitan area, the same percentage as for cities nationwide. Moreover, 36 percent of entry-level jobs are completely inaccessible by public transit.

Not surprisingly, roughly three in four poorer workers use cars to get to work. Recent work by University of Southern California researcher Jeff Khau finds that car ownership is positively correlated with job opportunities; no such relationship can be proven with access to transit.

At the same time, we should look at more-flexible systems, notably, expanded bus and bus rapid transit, which work better in dispersed areas and are less costly. Most rail systems tend to cannibalize most of their riders from existing bus lines, which explains the small net increases in total transit ridership.

Transit too expensive

Costs matter, and will become more important as cities and counties face the looming threat of fiscal defaults. In this respect, rail systems essentially steal from other transit – notably, the buses used mostly by the poor – and from hard-pressed city and county general-fund budgets. Gov. Jerry Brown’s outrageously expensive high-speed rail, which will principally serve the affluent, takes this unfairness to an extreme.

Instead, we should push far more cost-effective ways to provide transportation options, including those from the private sector, such as the successful Megabus, which provides efficient, quicker and far-less expensive transport between cities than either existing rail or short-haul airline flights. USC’s Khau suggests the private sector also could enhance solutions for lower-income commuters through car loans and car-sharing services such as ZipCar and and Lyft, a mobile app that links riders with drivers.

As we attempt to figure out ways to improve both the environment and people’s economic prospects, innovative 21st century solutions – from telecommuting to car-sharing – may prove more effective than relying on the 19th century technology of rail. We should not blindly follow transit ideology but focus on how to improve people’s mobility in ways other than the overpriced, inefficient and often far-less-equitable solutions being bandied about today.

Joel Kotkin is executive editor of and Distinguished Presidential Fellow in Urban Futures at Chapman University, and a member of the editorial board of the Orange County Register. He is author of The City: A Global History and The Next Hundred Million: America in 2050. His most recent study, The Rise of Postfamilialism, has been widely discussed and distributed internationally. He lives in Los Angeles, CA.

This piece originally appeared at The Orange County Register.

Photo by biofriendly, Metro Bus Campaign, Los Angeles

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Why is transit so often funded from taxes on petroleum products?

I have often wondered why "Peak Oil" is brought up as a reason to not build highways, but it seems perfectly O.K. to raise taxes and in some cases tolls on highway users in order to fund rail projects.

Never any mention of "Peak Oil" when taxes on motor fuels are raised to fund train projects, as they recently were in my state of Maryland.

Nor was there any discussion of "Peak Oil" when tolls were massively raised on the Dulles Toll Road (Va. 267) to fund an extension of the Washington Metrorail system to Dulles Airport.

Self-driving cars

Everything will change when cars are driven by electronic chauffeurs.


Self driving cars will have arguably a greater impact in dense cities where they can move around in close proximity without hitting each other or people. Traffic efficiency doesn't change the basic fact that fuel needs to be consumed to propel the vehicle. The shorter the distances, the better the efficiency especially once stop and go is eliminated. This basic law of physics applies to all fuels and energy sources. The train's great advantage is it's ability to move great amounts of mass most efficiently between points on a line. There is a price to pay for decentralization and personal mobility. Gas, mortgage subsidies, infrastructure and Standing Armies are very real and very dangerous costs.

The price of concentration is little understood

The price of decentralisation is nowhere near as high as the price of concentration, in actual dollar living costs.

The more concentrated an urban economy is, the higher the economic land rent. People end up paying more and more per square foot of space, and getting less and less space, and the impact is highest on those with the least ability to pay their way out of the space rationing system.

I am always bemused that the advocates of constraint either do not understand this, or they understand it only too well and have vested interests in boosting economic land rent.

If we really have an existential crisis that requires urban concentration, then sell the public on the necessity to nationalise or at least compulsorily acquire the land on which the concentration is to take place. Otherwise, the results of the planning and zoning always is that a few property owners make a killing in capital gains, and only a derisory amount of actual concentration of population occurs before the great majority of the population is "priced out" of the efficient locations anyway.

I also constantly point out in discussions like these, that the former USSR had urban economies based on trains and apartments, and it was an economic and socio-economic disaster. It was not something that was an advantage to the Communist economy, that it still failed "in spite of", it was an inherent part of the gross inefficiency of the whole. Land markets do allocate resources most efficiently when left to themselves. The only effect of "subsidies" to "sprawl" is that it might happen a little bit faster, but it is absolutely absurd to suggest that spatial growth per se is unnatural and needs to be explained by "subsidies". What is unnatural and needs to be explained, is efficiency-eroding containment of the kind the UK has been practising for 60 years.

Paul Cheshire, of the Spatial Economics Research Centre at the London School of Economics has been stating the amazing insight in recent months, that the massive inflated levels of land rent in UK cities (300+ times higher per square foot than the typical US city) represents foregone agglomeration economies. He is quite right. Planners have this back to front, thinking that "constraint" = increased agglomeration economies. It does not. It represents increased economic land rent and decreased agglomeration economies.

Besides this, there is no evidence that infrastructure costs are higher in the long term for sprawl. Infrastructure is much more expensive to maintain and renew and expand in high density conditions. The cost per person mile of commuter rail travel even in the most ideal local density conditions is not a lot lower than small efficient private cars, and no public anywhere has shown willing to pay the true costs in fares (whereas the majority of travel is still by car even in nations where fuel taxes cover the cost of roads and externalities many times over).

Standing armies would be needed anyway. Getting the resources literally means being the world's major power. People who don't want the USA to be the world's policeman and at least allow international markets in resources, either haven't thought through the geopolitical implications of the alternative, or are Quislings who quite like the idea of an ascendant global Communism or fascism of a kind inimical to our existing freedoms.

Real Costs

Agreed that cheap land should in theory relieve price pressures in the central cities and allows options however lets be clear that "cheap" exurban land is anything but cheap. Massive subsidies go into the mortgage machine to allow those cul de sacs to propagate, State and Federal tax money is used to extend the highways, arteries and utilities to these sites. Then they must be maintained, on ongoing cost that doesn't even address the legacy costs of the admins and workers receiving bloated government pensions to oversee the whole operation.

I don't think those arguments are based on the facts

"....lets be clear that "cheap" exurban land is anything but cheap. Massive subsidies go into the mortgage machine to allow those cul de sacs to propagate, State and Federal tax money is used to extend the highways, arteries and utilities to these sites....."

Mortgage subsidies are not location-specific.

The "Costs of Sprawl 2000" Report suggested that the costs of infrastructure for sprawl came to around $80 per household per annum.

This is a lot less than the thousands of dollars per household per annum inflated costs of housing that results from containing the growth. And there is no guarantee that transport costs will be saved due to the "pricing out" effect and the increased congestion.

".....Then they must be maintained, on ongoing cost that doesn't even address the legacy costs of the admins and workers receiving bloated government pensions to oversee the whole operation....."

Sounds like you are describing the situation in dense city cores. Municipal Utility District suburbs avoid all this bloat and wastage.

The "Costs of Sprawl 2000" paper did not take into account the higher costs associated with infrastructure maintenance and renewal in dense cores - the higher costs of access and disruption and land acquisition. The one paper that has attempted to quantify this is Helen Ladd (1998). It finds rising costs of infrastructure as densities rise, except that ultra low densities are slightly more expensive than not quite so low densities.

Hybrids work well in stop and go traffic

Actually in stop and go traffic a hybrid makes far more sense than in the country. If you have regenerative braking the energy used to speed the car up is recovered (or most of it) by the brakes, and re-charges the battery. This is why hybrids have better city mileages than highway mileages. n particular this is true if you don't jackrabbit start.