What I learned from Professor Per-Olof Gutman about two-loop controls on HOT lanes

On Monday, Metrans hosted Professor Per-Olof Gutman, who discussed some of his work in control engineering for the HOT lane connecting Ben Gurion airport and Haifa (Highway 2).

The HOT lane, like most, is operating under two political mandates. First, those who pay to enter the HOT lane are guaranteed to be allowed to travel 70 kph for the stretch of the HOT lane, and there is also a limit that tolls can go no higher than 30 shekels, which is about $10. Given those constraints, it’s not likely that the franchisee can get to a profit-maximizing strategy–whether there’s a profit at all is the question.

The original automatic control algorithm for setting tolls didn’t function properly: it overestimated the costs and contained a measurable lag: so cars and congestion would be clearing on the tollway and on the free lanes and the automatic control algorithm, responding to previous conditions, would raise the price as congestion was going down. Not what you want with a dynamic pricing model where you want to give people the right price signal when they are confronted with the decision to take the HOT lane or the free lanes.

Professor’s Gutman’s improvement suggested a two-loop control: an interior loop that monitors entry and an exterior loop that monitors changes in flow that cascades back to the interior loop as speeds increase or decline. That enables the company to maintain the floor speed of 70 kph.

The kicker on this–the rickety part of policy–is going to be the 30 shekels, not Professor Gutman’s controls. Because there will be a time, if demand grows, when it’s going to become impossible for the company to attain that speed floor with a $10 toll–if it isn’t there now. One of those performance constraints–the speed floor or the price–have to allowed to vary more once the HOT lane faces higher demand.

I can’t find Professor Gutman’s manuscript online, so he’s probably working on it now. I’ll post a link when we see it. One of my wonderful colleagues, Barak Fishbain, said that there are Youtube videos of some of Professor Gutman’s system control work on robotic motorcycles, but I can’t find those. I’ll post them if I find them.

Aha! edited, thanks to Barak, links to the YouTube of the Unmanned Motorcycle project!

Prices, simply, work

For 40 years, the US has spent billions investing in transit systems hoping to get people out of their cars. We have obdurately ignored economists who note that pricing gasoline more appropriately with a gas price floor or carbon tax would raise the costs of driving, would give us revenues to invest in public transit, and would do what everybody wants everybody else to do—stop driving gas guzzlers and stop driving so much.

Instead, we’ve built and built transit that has underperformed for years simply because driving is still so cheap. But we haven’t invested probably enough to prepare for the demand for public transit because we don’t have the revenues to do so, partially because we’ve stuck to the policy of keeping gas cheap.

Argh.

Stupid, short-term thinking.

The Financial Times has a series of very good articles recently on where we are.

Gasoline consumption shrinks vis-a-vis higher prices (what? REALLY?? HOW CAN THAT POSSIBLY BE?) as US Congress questions $2 billion in tax candy handed out Big Oil.

Gas prices spur inflation (of course they do, if everybody is getting around by car, everything from labor to other inputs are higher in prices)

And it’s not helping the trade deficit (again, basic math)

Gas taxes are bad bad bad, private investment is good good good

Polls, don’t you love them? There are so many ways that polls can be done badly.

Like this one cited in the WSJ, with a great revelation: people think it’s important to spend money on transport, but they don’t want gas taxes, either. And it would be ok if more private investment went into transport.

Awesome with awesome sauce on top.

This popularity of private funding raised for transportation infrastructure can be seen the universal love that people have for paying tolls and parking charges, and the long history of extremely successful private franchising arrangements we’ve had.

Not.

Maybe we should try to start our very own philanthropic organization: Fill a Pothole for Humanity. We could have sincere-looking spokespeople come on, in between the miserable child and miserable animal charity commercials, and talk in urgent terms about how swell it would be if you were to voluntarily send in your money right now, if you feel like it, in your spare time between driving too much and using inefficient vehicles while worrying about US national security policy.

Just so long as nobody is actually expected to pay for stuff they use.

Edited to add: Columbia’s David King writes that the pothole charity idea has been tried in Germany.

The Expo Line and Who is Overpaid in Transit Construction Contracts

Richard Green notes on his blog that:

A light rail line going by USC–the Exposition Line–has been under construction for some time now. For a considerable time, the site featured a sign that said the line would open in 2010. Now the estimates are that it will open some time in 2011 or 2012. At the same time, when I walk by the project, I can’t say that the workers building it show a great deal of, shall we say, urgency about getting the thing done.

At the same time, I don’t hear a lot of people who are upset about how far behind schedule the project is. Maybe this is because no one is planning to take the Expo Line. Maybe it is because peoople have such low expectations of LA Metro that they are not surprised, and therefore not outraged. Either way, it suggests a problem.

First off, it’s a bad idea to conclude anything about work effort based on what you observe by walking by. That’s like the people who judge professors by saying we “only teach two hours a week.” It’s not a valid sample, and it’s very had to evaluate other people’s work effort when you have never done the job yourself— and that’s particularly true of white collar workers passing judgment on blue collar workers engaged in dangerous and often tiring work–during a recession, no less, where anything that extends their work hours has direct implications for their family’s ability to eat and pay rent (unlike salaried work).

More to the point, Richard is mistaken when he concludes that people are not upset. The LA Weekly recently published a story called L.A.’s Light-Rail Fiasco which eviscerates the CEO of the Exposition Metro Line Construction Authority, Rick Thorpe, for salary and his conduct. Rick Thorpe is exactly the sort of transit guy who becomes a free agent and CEO: relentlessly self-promotional and confident, any previous successes get attributed to his leadership. So he picks up stakes, gets recruited away, commands an enormous salary, and builds a brand for himself that he delivers projects on time and on budget.

From the LA Weekly Story:

The reasons behind the fiasco are as numerous as they are complex. But at its core, it’s a simple story: Somebody had a clever new idea, and it backfired.

In this case, that somebody is Rick Thorpe, CEO of the Exposition Metro Line Construction Authority and one of the leading lights in light rail. He sold elected officials on a new type of contract, which he said would bring the project in cheaper and faster than it could be done by traditional means.

Colleagues from other transit agencies warned that the idea might not work. In the name of holding down costs, it could inadvertently create incentives that would drive costs up. But Thorpe pressed ahead anyway, and the elected officials charged with overseeing the line put their faith in his expertise.

Now, four years into the project, the results are plain.

“It just doesn’t work,” says Dan Peterson, an arbitrator with 50 years of experience in public works projects. “They’re trying to save 20 cents and it’s costing them $20 million.”

Thorpe and the MTA board argue that the contracting approach does, in fact, work. It is a process of negotiated design-build that, in Thorpe’s mind, prevented contractors from getting windfalls as they sometimes did under the design-bid-build process that has been industry go-to contracting process for a long time. However, the project where Thorpe’s innovation is supposedly working is not just behind schedule: it’s now 40 percent overbudget (to the tune of $260 million).

Now, you would think that Thorpe would know better than to be too fussy about contractors making money, based on this bit of info:

As CEO of the Expo Authority, Thorpe oversees a staff of 15 and earns a salary of $334,000. He makes more than the CEO of the Metropolitan Transportation Authority, who is responsible for 8,000 employees.

Thorpe’s self-branding that captured this salary is one of the major flaws of leadership both in the public and in the private sector. Once shareholders or board members believe you are some kind of magician, and they will if you are fortunate and if you blow your horn hard enough, competence is no longer enough. And as anybody who has built anything or completed any public project knows: 1) given all the barriers and problems, it’s amazing that anything gets built, ever, let alone on time and on budget, and 2) nobody, no matter how smart, confident, or charismatic gets anything built all by themselves. It takes a team, and while teams need managing, it is really easy to overstate and overcompensate the contributions of management when things go right and to make excuses when things go toes up.

There is another side. Why shouldn’t public managers good at their jobs make more than doctors or other professionals? Thorpe is a CEO, after all, and this is a major project, and major projects are tremendously hard to deliver, and private-sector CEOs make much much more.

However, the management and incentive contradictions emerge quickly. If you are paid $300+K a year to run something, your desire to finish it on time–particularly when you still have a board that rushes to your defense and you are no longer a hungry young guy building your brand–is low. Unless there are bonuses in Thorpe’s contract for on-time performance, he has little reason to protect the project from delays at this point, given all many reasons why projects get delayed in construction and the halo surrounding him personally.

Moreover, if you make your money because you are such an excellent manager, there is also the desire to innovate practices that reinforce the need for *management* on this project and future projects.

And from the what the LA Weekly reports, this “make management work” approach is right at the center of the problem. Instead of hammering out the details of the contacts up front–as with traditional design-bid-build contacts–Thorpe’s “negotiated design-build” requires the agency to keep their hands in project management more so than under traditional design-build where they would have been managing the contracts primarily–instead of negotiating as they go along.

In the end, I’ve always argued that there is very little wrong with the design-bid-build process.The US built most of the Interstate system with the approach and most existing transit, water, sewer, and other infrastructure this way. Where Thorpe sees the potential for “windfalls”, I see an incentive for construction companies to keep costs down so that they can increase profits. In the hands of a competent agency contract manager keeping track of the as-builts and project specifications, you shouldn’t end up with a poor-quality project. Instead, you provide an incentive on the part of the construction companies to keep employees hopping and to strike hard bargains with suppliers in the hopes of getting in under budget so that you can walk away with that the built-in cushion. The bidding process keeps companies from building in too much of a cushion. Innovations here have almost always struck me as cases of fixing something that wasn’t broken–to the overall detriment.


the MTA budget cuts and the need for long-term transit finance reforms

David King pretty much sums up how I feel on the cuts, except for one thing: I was surprised they weren’t worse. However, the MTA is a sharp outfit: if they have to, they’ll announce 4 percent this month, another 4 percent three months from now, and another 4 percent after that until they can meet their payroll. It’ll be…interesting…to see where we are a year from now.

There has to be long-term finance reform in transit, and no, the 30/10 doesn’t get us there. The 30/10, wonderful as it is, would enable us to build more stuff that, ultimately, we won’t be able to operate.

These cuts are going to get worse and worse, not just in LA, but everywhere, and eventually, those cuts are going to hit the train operations, too. Um, yeah. Without long-term finance reform, eventually we’re going to get to the point where the bus cuts that all the transit blogs treat as little more than a passing headline will be joined by rail service cuts.

New York is already there.

Train cutbacks might actually make people in the streets blogging and transit-advocating world actually care …maybe…about operations.

Then maybe they will spend as much time talking about the hardship that ensues from cutting services as they appear to want spend gossiping about which of their favorite celebrity transit advocates might work in Los Angeles to replace retirements.

Nah. Talking about transit’s operations problems–like getting back and forth to work in a world where services are getting cut–would be a distraction from treating transit like the jungle gym in the ultimate urban playground for twentysomethings raised on a steady diet of Friends rather than as a place where transit needs to work for people other than trust funders on pub crawls.

Whatever, right? Everybody knows operations are just a waste of money in transit. Building is where the professional and political payoff is. Advocacy for projects gets you in the position where you get to be part of the buzz. Who was the last bus operations drudge you saw featured on a transit blog, or given an high-profile award for their work?

So what to do–other than grow the hell up? I haven’t run any numbers yet, but it’s pretty clear that the Feds are not going to hand out candy for operating subsidies. Which means we are probably left with local funding sources. A place to begin thinking in terms of reform:

a) When we find new a dedicated source of funding–and we’ll have to–we’ll have to have the self-discipline NOT to promise to spend it on our addiction to building new transit projects.

New sources of funding = political nightmare. But some suggestions: Ratchet back on property tax insanity in California–ideally. So how about a 3-cent gas tax increase on local pumps? We can tell everybody we are solving congestion even though we aren’t. Sure, let’s have parking charges a la Don Shoup. But whatever funding it is, it’s not for geegaws or fripperies or extensions or new services in far-flung suburbs, at least not right now. It’s about keeping existing routes and frequencies.

With whatever funding victory may come, we can’t fall the into the political expedience trap of earmarking new funds for projects. We love to do that because taxpayers like to be able to point at a map and see what they think their money goes to.

We can’t keep indulging that. We have to start learning how to make the social marketing case for operations.

b) Create a dedicated pot for operations at the local and regional level, and expect to put money into that pot forever for transit as the municipal public service it is.

Nobody expects street sweeping to pay for itself; ditto police protection or libraries or snow clearance. Transit in cities should be treated the same way as those services. We don’t talk about “garbage collection subsidies.” It’s time to talk about funding services instead of obsessing about subsidies.

(And yeah, in case you can’t tell, I’m pissed.)

Add to: Facebook | Digg | Del.icio.us | Stumbleupon | Reddit | Blinklist | Twitter | Technorati | Yahoo Buzz | Newsvine


Spring 2010 ACCESS

From its website:

ACCESS magazine reports on research funded by the University of California Transportation Center. The goal is to translate academic research into readable prose that is useful for policymakers and practitioners. Articles in ACCESS are intended to catapult academic research into debates about public policy, and convert knowledge into action.

Brian Taylor and I are in this issue of ACCESS discussing the relative merits of user tolls versus sales taxes.

Add to: Facebook | Digg | Del.icio.us | Stumbleupon | Reddit | Blinklist | Twitter | Technorati | Yahoo Buzz | Newsvine


The New Federalism v. 4.0 in transport?

Over dinner the other night, a political scientist and an economist dismissed my financial worries about HSR. “The government wastes money all the time” one said said. “It might as well be on trains rather than military spending.” The political scientist, in particular, said that Congress can just appropriate more to transportation to cover HSR; I was making a faulty assumption that the money for HSR will displace other transportation funding needs like the Big Dig did in the state of Massachusetts.

I’m still not convinced that the US Congress is ready to pull out their checkbooks for HSR any time soon, and I’ll be really interested to see the day that military spending gets cut in favor of infrastructure spending (or anything else)–but I am not holding my breath. The $2B allocated to California is hardly chump change, but it’s still $7B short of what the HSR says it expects the Feds to kick in (which means it will probably need closer to $10B to $12B of additional federal sources).

As early as 1974, comments on the “the New Federalism” began emerging as the Federal role in transportation began change from what it had been after the mid-1950s. The easy part of the interstate system had gotten built; the rest was becoming increasingly more nightmarish to implement as urban neighborhoods resisted being paved on. From there, the federal involvement in planning for transport became much more associated with “fiscal federalism” for just about everything, and Reagen’s “New Federalism” was ostensibly about a diminished federal responsibility in domestic programs. As Brian Taylor noted, however, it’s nearly impossible to argue in transport that finance is separate from projects or planning. Nothing gets implemented without resources, and the resources usually comes with strings. ISTEA was also noted as break from the previous round of fiscal federalism, where localities, MPOs and states got much more discretion and block-grant funding. Nonetheless, there were real variations in how much discretion and control over funding MPOs had from state-to-state.

It has been widely bandied about that the Obama administration and the economic crisis he’s got on his hands has worked to re-Federalize policy across the board, from banking to housing to urban policy more generally. We have for years had persistent calls for Federal funding for walking and biking projects, which I have to say I’m not sure I can get behind: why, exactly, can these not be paid for out of local taxes sources and, in particular, property taxes given all the studies that want to show me how TODs and New Urbanism, etc increase property values? Ditto with street projects, btw.

More compelling, the inter-city nature of HSR suggests the same type of Federal role it had during the Interstates. Yet with the Interstate we had a new Federal tax that went straight into a trust for that system. And that’s where I am pausing before I can call HSR or ARRA the new New Deal.

By contrast, the 30-10 plan from Los Angeles, recently passed, does promise to be a very new approach to Federalism in the US–one where the states are largely bypassed in favor of local fiscal capacity hedged against US federal financial capital and risk reduction. The feds are paying up front, but are being paid out of a dedicated local revenue source; it’s an infrastructure banking plan. That is a lot different than a simple act of appropriations, and my suspicion is that if states want their piece of the HSR pie, that’s the direction they are going to have to go. How far they have to go in will depend on how powerful their Congresspeople are.

References that informed this post:

Blumenberg, E. & Schweitzer, L. (2002). Devolution and the new federalsim: The role of the federal job access program in improving the mobility of welfare recipients. Planning Theory and Practice.

Dinan, J. & Gamkhar, S. (2009). The state of american federalism 2008-2009: The presidential election, the economic downturn, and the consequences for federalism. Publius: The Journal of Federalism.

Gage, R. W. & McDowell, B. D. (1995). ISTEA and the role of mpos in the new transportation environment: A midterm assessment. Publius: The Journal of Federalism, 25(3), 133.

Pagano, M. A. (1986). Old wine in new bottles? An analysis and preliminary appraisal of the surface transportation assistance act of 1982. Publius: The Journal of Federalism, 16(1), 181.

Mertins, H. & Jr. (1973). The “new federalism” and federal transportation policy. Public Administration Review, 33(3), 243-252.

Taylor, B. D. (2000). When finance leads planning: Urban planning, highway planning, and metropolitan freeways in california. Journal of Planning Education and Research, 20(2), 196-214.


Unfair fares

I have to admit, this story from the Washington Post

Bus riders see inequities in proposed Metro fare increases – washingtonpost.com

has my blood up. Bus riders have a median income of $70K–which is very high for bus riders in general–but rail passengers have a median income of over $100K. Now, it takes real differences in distribution to produce those kinds of differences in medians, but what makes me angry is this comment:

Metro Board member Chris Zimmerman of Arlington County played down the importance of the relative increase in bus vs. rail fares, calling it “a distraction.” “To minimize the impact on the lower-income, we need to not cut service and get the jurisdictions to mitigate the impact of the fare increases,” he said.

Um, yeah, I bet you see this as a mere distraction, buddy, because your constituents are the more affluent rail riders. So of course a smaller percentage increase to them seems fair to you.

The first rule of poverty: never listen when somebody who has money tells you not worry about money.

That said, I’m not actually sure what would be fair here in terms of percentages, but I have the following thoughts:

1) I suspect that transit companies are way more likely to pull buses off the road to lower costs than they are to cut rail service, on the longstanding argument that rail operating costs are lower per passenger than on the bus (supposedly; this only works if you have riders on your trains). This works out well because I also suspect that most transit companies, having had to fight like the devil to spend the millions they have on capital investment on the rail side, thus have no intention of scaling back the rail service they’ve fought so hard to put out there.

That means the bus riders would be more likely to be subject to service cuts; I also think that many, many transit companies would be happy to lose bus passengers via higher fares but would dread losing any rail passenger because low ridership on rail looks infinitely worse under public scrutiny due to the higher project costs associated with rail.

2) I also suspect that bus riders are more transfer-dependent passengers. That is, I suspect that most of them use the rail services in DC, too, but also rely on bus feeders while rail-only passengers are park-and-ride or kiss-and-riders. This means that bus riders are charged for lower quality service, up front, and that when the bus side is cut they lose even more value in the network. In addition, the across-the-board fare increases on both sides are going hurt here, too.

In any case, I had to do some talking to convince some of my colleagues that transit affordability is a real issue. Just because something is cheaper than keeping a car doesn’t mean it’s objectively affordable. Yes, hamburger is cheaper than sushi but if you have no money, the price difference between the two is irrelevant.

Not happy news for DC area commuters, in any case.


The Transport Politic on the 30/10 Plan

I haven’t commented on LA Mayor Antonio Villagairosa’s 30/10 plan, which has been to go to the Federal government to ask for an up-front loan to fast track (see what I did there?) the rail projects currently programmed for 30 years into 10 years of construction. The revenue from Measure R, a local option sales tax, would then be used to retire the debt. This is an infrastructure banking strategy, and we don’t yet have that type of setup in the US. Perhaps its time that we did. I haven’t commented on it largely because I can’t see much to object to; the voters have already approved Measure R and the lines are decided (for better or worse). The Transport Politic calls this one right: it’s an innovative financing plan more than anything else:

How Feasible is Antonio Villaraigosa’s 30/10 Gambit for Los Angeles Transit? « The Transport Politic

Should the rest of the country go for it? I think so. The major risks associated with the plan are cost overruns, which basically just mean LA will end up with less rail than envisioned, and that risk is always there with project development. This upfront financing probably lessens that risk. As an innovative financing tool, this may be one of the most important steps forward in funding regional transit services rather than relying on lukewarm federalist arguments that transit in LA is such a worthy good that senators from Iowa and Nebraska and North Dakota should want to see federal dollars going towards it. These types of local self-help financing tools, then, mean that the regional transit agency could tap into LA’s enormous tax base directly rather than having to deal with the crap-shoot of trying to get the feds to pay for local projects. There’s a lot to recommend this strategy both practically and theoretically.


Californians wake up to the reality of major projects like HSR

Cities not on track with California’s plan for high-speed rail system – latimes.com

This story is about a Buena Park Mayor faced with the loss of a TOD and station area he fostered because of the high speed rail. So not only do we have to throw money at a new system, we’re going to use those fund to undo a project that we just threw money at. Is there really, seriously, no better design than one that takes out a TOD? .