TOWARD COST-EFFECTIVE TRANSPORTATION

 

Transit is much more expensive to build than highways. It’s politically correct to focus on transit. But is it the best use of our tax dollars? Let’s look at the numbers.

Transportation planners in our region look at many. At the most recent Transportation Planning Board (TPB) meeting, there was a presentation on the ways that transportation plans are measured and approved factors – social equity, air quality, and many more. But when I asked if there was a cost-benefit analysis, it became clear that this did not appear to be on anyone’s list of measures.

By cost-benefit, I mean this: when you build a new transportation project, how much money does it cost to move people?

Over the last few weeks, I went back through some presentations and found the two slides shown below that have the numbers to tell an important story. I spent a lunch hour on the phone with TPB staff to verify that what I was seeing was accurate, and what it might mean. Here is what I learned from TPB’s data:

The DMV region plans to spend $42 billion to expand transportation capacity over the next 25 years, split between $27 billion on highway expansion and $15 billion on transit. This will result in 2.7 million more daily trips by auto and 300 thousand more daily trips in transit. By simple arithmetic, this means that it costs just over $10,000 to add capacity for another auto trip, and more than $53,000 to add another transit trip. Building transit capacity currently costs more than 5 times as much as highway capacity!

 

If this was the only factor that was important, then decisions would be easy. Any CEO would immediately allocate more money into adding highway capacity. Of course, it’s not the only factor. Not everyone can afford to travel by auto – we want lower-income people to be able to get to their jobs, so we need transit. Transit trips are less polluting than autos, although TPB’s data shows a steady decrease in auto pollutants thanks to greater efficiency and the growing number of electric, zero-emission vehicles.

The other key is that, for parts of our region, building new roads or even expanding existing ones is terribly difficult. Where would you put a new thoroughfare in DC, or in the close-in suburbs?

The costs I focused on so far are the capital costs for new projects. The same TPB information can be used for operating costs – how much it costs for each trip. It turns out that we’re going to spend $130 billion over the next 25 years on transit operations and repairs, about $5.2 billion annually, with capacity growing to 1.5 million daily trips, for a per-trip cost of about $9.50. Each time someone takes a transit trip, the government subsidizes the trip by that amount. We’ll spend $72 billion to maintain roadways during the same period, about $2.9 billion annually, to move up to 16.6 million trips/day. That comes to just under 50 cents per trip.

The operating cost information is useful in a couple of ways. At the same TPB meeting, the Commuter Connections presentation unveiled a new program, piloted in Howard County MD, where auto commuters can receive a $10 stipend for taking a rider along with them. That number is almost exactly right – it is comparable to the cost of putting someone on transit instead, but we don’t need to build more transit lines.

That is the kind of thinking we need. When we look at a new project or a new idea, does it move people more effectively than how we’re doing it now? Is it better for some reason, is it faster, is it cheaper?

For example, the TPB recently recommended that we find ways to encourage employers to let more people work from home. What if the government provided an incentive to the employers? With these numbers, we can make informed judgments about how much of an incentive makes fiscal sense.

The amount of money we have to transport people is limited, so we need to think carefully about optimization strategies to move people cost-effectively as well as focusing on all the other factors.

Neil Harris is a member of the Gaithersburg City Council and the Metropolitan Washington Council of Governments Transportation Planning Board.

One comment on “TOWARD COST-EFFECTIVE TRANSPORTATION

  1. Staff voteforneil Reply

    Comments on this post appeared on a number of Facebook pages, and are pasted here:

    Seth Grimes on his Facebook page:
    Neil Harris has written an interesting analysis, “Toward Cost-Effective Transportation,” for Seventh State. Yet the opening statement: “Transit is much more expensive to build than highways” lays out an unbalanced comparison, given the focus exclusively on tax-funded expenditures. What about personal expenditures? Consider:
    Transit is both infrastructure and vehicles, and highways are just infrastructure. If we include the cost of private cars and their fuel and upkeep in the comparison, and of 35,000 auto fatalities and of auto insurance etc. – the full “cost to move people” by car – how would the road-trips vs. transit-trips cost equation change? We will find that highways aren’t so cost effective after all.
    In any case, as Neil recognizes, we need both transit and roads. Neil reports from the Transportation Planning Board meeting, “the Commuter Connections presentation unveiled a new program, piloted in Howard County MD, where auto commuters can receive a $10 stipend for taking a rider along with them.” I agree, that’s cool, as is finding ways to encourage employers to let more people work from home, a TPB recommendation. Myself, I’d like to see more-aggressive investment in smart roadways technology, which will both add capacity to our existing road network and create local business opportunity. (I represented Takoma Park for three and one-half years on the Transportation Planning Board.)
    Creativity, optimal use of our transportation infrastructure, and consideration of ALL relevant data provide the best path forward toward cost-effective – and affordable, congestion-relieving, and green – transportation.
    Neil Harris Seth, I appreciate your response, but I will point out that you didn’t actually provide any data to back up the assertion that auto trips may be as costly as transit trips. I’m not disputing the assertion, but I haven’t seen that data. At a little over $.50 a mile, the government reimbursement rate might be a useful guideline to the trip costs in autos, since it is intended to cover pro-rata fuel, insurance, wear and tear, purchase cost, etc. But I’m sure there’s better data somewhere if you would look, and I think anyone making these decisions ought to have the facts. That’s the discussion I was trying to start with this piece.

    Seth Grimes Of course you’re right. I’ll think about doing some research and posting in longer form. Your article is indeed a good discussion-starter!

    Neil Harris Aside from starting a discussion, I would really hope that we evaluate any new transportation projects on a cost-benefit basis as well as for the other factors. That’s really what I see lacking today. And I would hope that any analysis of the effectiveness of new proposals would be evaluated independently, such as by TPB staff.

    On the Seventh State Facebook page:

    Louis Koutras This is an oversimplified analysis. Transit takes cars off the road and helps reduce gridlock, which saves our workforce time and money and improves quality of life for all drivers. It expands people’s options for work, recreation, shopping, etc.

    Highways also require large amounts of land, land that could otherwise be developed and used for commercial activity, or left developed to promote environmental health.

    The article seems to suggest that a rational analysis would lead us to put *all* of our transportation budget into roads. This would 100% lead to unusable, overcrowded roadways that completely dominate our landscape. No walkability. Poor quality of life. Say goodbye to our ability to attract and retain top talent.

    Transit is not “politically correct,” it’s not an ideological position, and it’s not welfare. IMO, transit is vital to the economic health of a dense urban/suburban region such as ours.

    Neil Harris All good arguments. The analysis can easily be used to show the cost of everything you suggest, and then a decision can be made for whether it’s worth it.

    I would suggest that there should be a rebalancing between roads and various transit modes. I suspect that passionate arguments like yours have tilted the decision makers. Over-emphasis on roads leads to what you describe — but over-emphasis on transit leads to worsening congestion because we cannot afford enough transportation to meet the demand if we’re going to pay >5X the cost of alternatives. 60% of our transportation funds are going to go to transit in the current plan, to move a small fraction of the riders.

    I’m not advocating one or another, but rather I’m trying to balance the transportation budget.

    Jen Fischetti It would be helpful if the occasional analysis refrained from conflating system types, thereby ignoring a massive cost factor, not accounted for in that analysis.

    I read a fair number of discussions and critiques for both roads and transit. THAT statement alone is a misidentification wherein I proffered the term roads as a stand-in for an open transportation system wherein typically governmental units acquire land to construct and maintain roadsways of various length, access and capacity and private and public interests make the acquisition of the vehicles to be operated, repaired and maintained, to use on the roadways owned and kept in good keep by governmental units (with rare exception, ownership is near always a governmental unit).

    Transit is a closed system, wherein from a user’s perspective, all of the considerations to offer that user the freedom to choose where they wish to go, is encapsulated inside of a fare, a generally simplified cost mechanism. I can give the transit agency $5 and receive a prescribed access to the closed system. I cannot buy a bus and have the transit system allow that bus into their system, the way I could purchase a minivan and access roadways in the open system, subject to prescribed terms.

    So when comparing costs, what is typically missed, and at the expense of the analyst’s conclusions, is the true and full cost of open transportation systems which use our public roadways.

    Whether this cost is expressed in the cost of acquiring, operating and maintaining a fleet of vehicles by a company providing a product on the market along with the effect in the price of goods and services, or in the individual or family buying/renting/leasing vehicles and using roadways to express their freedom to choose where to go.

    Whatever those costs may be, we can all agree those costs are removed from the rest of the economy, wherein the company or household may find those costs better served, in housing, education, long term investment in the markets. There are any number of tools to compare what might be the best and highest use of those resources, we and citizens of a roadway-centric society, could in turn benefit from; especially if compared to the cost of using a public transit system, where it is less expensive for the individual to subscribe to use on a need basis, not on a full time, just in case one.

    So I would like to see the cost of roadways include the cost of vehicle acquisition, operation and maintenance as expressed in terms of TOTAL costs for the open system, compared to the otherwise disclosed total cost for closed transit systems. While that is not the only adjustment needed to compare open and closed transportation system, it is one which is frequently ignored.

    Until that is performed in an analysis, that analysis is incomplete and without real value.

    Neil Harris It’s actually pretty easy to add what you suggest. The IRS has already determined the cost-per-mile for auto ownership and operation, pro-rated over the life of the vehicle and which includes purchase, fuel, insurance and maintenance. So if we know the average trip length and multiply by 54 cents per mile, we know the cost of the trip to the consumer.

    I agree that my analysis, meant to be a first step and only a blog post after all, was not complete, but I do hope it adds value to the discussion.

    Aside from adding costs to auto ownership — and I still think my numbers are quite relevant since they focus on the socialized cost for the system — there are more detailed and nuanced next steps required. Perhaps the right questions could be shared with TPB staff, especially since this is the year for them to update their traffic survey — to make sure they are gathering enough data to answer all the questions.

    From my analysis, another insight is that time of day is an important factor — transportation capacity is “driven” by peak usage times (rush hour, obviously) but the data available is for all trips and work trips — but not all work trips are at rush hour and not all rush hour trips are for work, therefore we don’t have numbers for the peak times. Since transit usage is more heavily weighted to rush hour, perhaps the costs are less unbalanced.

    Geography is another key factor — I expect the costs to be very different in denser areas.

    Mode preference is another factor, impacting the demand side rather than supply. Would more people use nicer buses — if we added amenities like wifi and nicer seats, more soundproofing — or studied what factors consumers of transit would want — could we get more people to use them? Right now we are investing more in transit and yet ridership is on a steady decline. If it’s faster, more convenient, more pleasant, and less expensive to take an auto than a bus or other transit, then what would you expect people to choose?

    An alternative is providing a variety of draconian measures to discourage auto usage, and I would oppose those plans. That’s not the way I want to see us governed. We need to make choices better, not worse.

    Jen Fischetti The IRS allowance is far from the cost each specific user of an open transportation system experiences.

    I tend to rely in AAA’s calculations which ranges from 93 to 38 cents per mile based on vehicle type and annual miles. I experience per mile rates in excess of $1.40 because of my extremely low annual mileage.

    The practical reality is the threshold to gain access to the open transportation system is insurmountable for well over 30% of Baltimore city households.

    So per mileage rates are disproportionate in historically marginalized communities with higher density as compared to privileged low density communities constructed around vehicle access.

    So let’s express access in terms of annual costs which range from $$6700 for small sedans to $11,000 for large sedans according to AAA.

    When we add $8000 annual costs for access to open transportation systems to EVERY vehicle on the road PLUS the annual cost of the commercial fleet of vehicles our comparison is far from “dramatic”.

    Jen Fischetti We could also express those 55 billion Vehicle Miles Travelled against the AAA per mileage averages and we exceed $35 Billion dollars annually in costs of roads ways typically ignored in analysis.

    Jen Fischetti The $35B goes up dramatically when we account for the commercial fleet portion of our 55 billion annual VMT

    Jen Fischetti Just to bring the true cost into the 25 year projection framework you proffered vis-a-vis the NCRTPB’s CLRP Budget costs over a 25 year period.

    Spending $35B (our current 56.5B VMT x62 cents per miles based on an average of AAA per mileage cost estimates) we will have spent $875B, that’s BILLION to place, operate, maintain and repair the largely private vehicles on the open transportation system in addition to CLRP’s projection of $99B. This means we spend nearly $1 TRILLION dollars in that 25 year window for the policy outcome of an open transportation system that allows private interests utilization of that system to provide them the freedom of mobility. THAT $1,000 Billion is compared to the paltry $145B over the same period and we have to ask, which makes more sense?

    The $1 TRILLION investment in more gridlock or shifting a mere 20% of the $1 TRILLION toward transit expansion on dedicated Class A Right of Ways and expansion of public cycling infrastructure?

    No investment in our open system will relieve gridlock, EVERY attempt in the past has been met with only a near immediate response to increased road capacity with still more maximum capacity vehicle travel.

    Anyone with a basic econ background can tell you the market is begging for subsitutes, yet as a society, we somehow believe it is always an “either or” proposition.

    Maybe it is because we are constantly bombarded with faulty cost-benefit analysis which excludes the biggest cost of them all, the actual vehicle to operate on open systems, leaving us will the false conclusion that “building roads” is cheaper than investing in multi-modal transportation solutions.

    Neil Harris Transit usage has been steadily declining in the region and elsewhere in the US, despite increasing expansion of transit. Does that make a case for more investment?

    I am not against transit, but perhaps we need to rethink what kind of transit we’re providing. From your numbers, private citizens are overwhelmingly making the choice of how to spend their time and money. Why would you say that would be? Is it rational?

    From Jud Ashman’s Facebook page:

    Cathy Carlson Drzyzgula I presume these costs are for projects that have at least some preliminary approval, which means the proposed road cost estimates are not in the same places as the proposed transit projects. I wonder if there are estimates for road projects that were not approved because transit in the same corridor is a better alternative. That would be interesting to compare.

    Neil Harris These costs are all for projects in the last CLRP, which means they are all expected to be funded.

    Cathy Carlson Drzyzgula Neil Harris that is what I figured. Is there info available on land acquisition costs? That must be a significant factor in initial cost.

    Neil Harris The numbers here are an average for the entire region, and they include land acquisition. Since they are an average, some projects are more or less expensive. Obviously land is more expensive closer in, which has an impact on all projects. Adding a bus to a road in DC costs less than building a new road — but adding a transit line would require acquiring rights-of-way as well.

    Neil Harris The rules of thumb I’ve found is that building heavy rail costs $250 million per mile — and that matches the cost of the Silver Line, and MDOT uses $4.4 million as a cost per highway lane mile — although the numbers that they shared regarding the proposed widening of our highways (I-270, I-495, and the BW Parkway) come out considerably higher than this.

    Some more arithmetic, now that I think of it — at $250 million per mile, and at $53,000 per added trip, we would expect each mile of heavy rail to add ~4700 trips per day of capacity. The Silver Line at $2.77 billion should add over 52,000 trips per day, if it’s an average transit project on a cost efficiency basis.

    We could obtain the expected increased trip capacity of any transportation project by knowing either the length or the cost. If the models showed better or worse expected performance, then agencies could look at decisions.

    Cathy Carlson Drzyzgula My point was that the average cost per mile or trip is affected by where the projects are located and that makes for part the higher cost for transit projects. I would be interested in comparing the cost of building a mile of highway in the SAME place to building a mile of transit. I imagine this would be hard to do as highways aren’t proposed for dense areas. So the average costs for building transit are higher in part because the projects are in places where the costs of building a road to carry the same traffic is considered too high to even propose.

    Neil Harris Of course, the answer is, it depends. It depends on the mode: you still need land to build fixed transitways like metro or light rail, but to build city buses you just need available roadways that exist.

    Then there’s the question of mode preference, which I don’t see an any models yet. By that I mean that, the more money people have, the more they prefer higher quality modes like autos or metro rail. The more money you have the less likely you are to ride a bus.

    Wouldn’t it be great if we could find a transit mode that was more cost-effective? Aside from making a cheaper, another way would be to make something that people really like and prefer to use.

    Cathy Carlson Drzyzgula Neil Harris thanks. I regularly rode the bus but I know many people won’t. Just like living in a diverse (income wise) neighborhood. The fact most people won’t choose those options makes them unavailable in the quantities that would probably be better overall.

    Neil Harris I wonder if you can take a mode that’s less attractive and make it more attractive, for example by taking a bus and adding Wi-Fi, television, amenities of other sorts. Maybe one answer is just nicer buses?

    From the Neil Harris, Gaithersburg City Council Facebook page:

    Joseph Allen While interesting, this article arrives at a false conclusion about what roads purchase versus transit. Even while acknowledging costs of air pollution, those costs are not quantified. More importantly, transit dollars encourage very different land uses than expanding roads. The dollars spent on roads induce demand in far flung areas (requiring new infrastructure (sewers, schools, police, etc) that is unsustainable in less dense areas). Roads also require massive amounts of land for the road itself, parking upon arriving at the destination, and lack of physical exercise by those using them. The use of the land by roads reduces its availability for other more productive uses. There also are maintenance costs for both transit and roads that require lifetime costs to be listed here. All of these costs are not taken into account by this article. A real analysis would require more than a basic spreadsheet of cost/users. The analysis is a start but a very crude and less than useful comparison. Outside of costs, there are real social and cultural impacts of safety, community, and what the role of transportation is when one thinks roads verus transit instead of community building. Even car companies like Ford get this today:https://www.recode.net/…/smart-vehicles-cars-ford… and https://corporate.ford.com/innovation/city-of-tomorrow.htmlManage

    Ford CEO: Together, let’s take back the streets for living
    RECODE.NET

    Neil H. Harris, Gaithersburg City Council. Joe, those are all good arguments that are being made by you and others, but I’m not sure where the data is to support decision-making in them. The maintenance costs for rugs and transit are part of the calculations are used in the article.

    In terms of land use reducing availability for more productive uses, those decisions are made by governments all the time. The ag reserve it is a good example of a community deciding to make less productive use of land for other reasons.

    Joseph Allen You are asking good questions and they could be even better informed with urban planning scenario planning and data. TPB started with their analysis but it can go further. I would welcome the opportunity to add data variables to your analysis and collaborate on answering the questions more completely. There are well established metrics for the tax burden of different land uses, acceptable commute times, induced demand, etc. All these variables figure into cost/benefit analysis. The ag reserve is not a poor land use, but rather a thoughtful effort to drive development to existing infrastructure that is near job centers. In addition, there are only so many places to add roads in the region. I support road or transit construction in the appropriate places that are cost effective in supporting economic and social connections.

    Neil H. Harris, Gaithersburg City Council. Joseph Allen I’m always happy to work with you, Joe.

    The data in my high-level analysis was readily available and verifiable and I thought it was a useful start. A deeper dive would definitely be the right next step. Perhaps the right questions could be shared with TPB staff, especially since this is the year for them to update their traffic survey — to make sure they are gathering enough data to answer all the questions.

    From my analysis, another insight is that time of day is an important factor — transportation capacity is “driven” by peak usage times (rush hour, obviously) but the data available is for all trips and work trips — but not all work trips are at rush hour and not all rush hour trips are for work, therefore we don’t have numbers for the peak times. Since transit usage is more heavily weighted to rush hour, perhaps the costs are less unbalanced.

    Geography is another key factor — I expect the costs to be very different in denser areas.

    Mode preference is another factor, impacting the demand side rather than supply. Would more people use nicer buses — if we added amenities like wifi and nicer seats, more soundproofing — or studied what factors consumers of transit would want — could we get more people to use them? Right now we are investing more in transit and yet ridership is on a steady decline. If it’s faster, more convenient, more pleasant, and less expensive to take an auto than a bus or other transit, then what would you expect people to choose?

    An alternative is providing a variety of draconian measures to discourage auto usage, and I would oppose those plans. That’s not the way I want to see us governed. We need to make choices better, not worse.

Leave a Reply

Your email address will not be published. Required fields are marked *