Markets for Road Use—Eliminating Congestion through Scheduling, Routing, and Real-Time Road Pricing

>> John Taylor: Rick is gonna talk to us on markets for transport, eliminating congestion through scheduling, routing, is that correct?
>> Rick Geddes: Yep.
>> John Taylor: And real-time pricing, so thank you so much. You're coming all the way from Cornell where you're the director of the Cornell program infrastructure policy for 12 years.

>> Rick Geddes: Almost 13.
>> John Taylor: Almost 13, yeah, so you're welcome here. Some of the people will be Jim Mattis for example there on the screen. And other people around the table. But we're looking forward to what you have to say. We'll interrupt you.
>> Rick Geddes: Yep, okay.
>> John Taylor: Okay. I'm going to stand up because I project better.

Okay.
>> Rick Geddes: But if that affects the microphone for the people online, let me know. It's great to be back here. I was actually a national Fellow here in 99, 2000 over in the other building. This building did not exist at that time. So it's really wonderful to be back and to see friends.

I saw John Cochran not too long ago when I was visiting and I said professor Cochran, it's great to see you again. I'm Rick Geddes. You taught me macroeconomics at Chicago. And John said, gee, I'm sorry to see. I'm sorry?
>> John Taylor: He never says that to us.
>> Rick Geddes: I'll bet he doesn't.

So I've been here since mid September. I'll be here through the end of May. I'd love to take advantage of being here physically and meeting as many people in person. Buy you coffee as possible. I do wanna give a shout out to Professor Josh Rao and the state and local governance initiative that is hosting my visitors.

I've done a few sabbaticals. I don't think I've ever been made to feel so welcome as being right in the middle of that wonderful team and hosting my visit while I'm here. And it's really a nice sort of plug in for me to have that instant group. And of course the infrastructure work that I do is mainly a state and local issue because they're the owners of the infrastructure.

The federal government actually owns relatively little civil infrastructure. They own a ton of military infrastructure. And we could talk about how that's going to affect our road pricing plan.
>> John Taylor: Funds a lot of the capital but then it turns it over to the states.
>> Rick Geddes: They do, they do through gas taxes on gas and diesel fuel that are per unit taxes and there's other things like truck tire taxes that we could talk about.

I do want to say a thanks because I think this sabbatical worked in the sense that I was able to clear my mind of administrative work. People know Cornell, it's big red. We call it big red tape. You get sucked into committees a lot. Students are wonderful, but they soak up a lot of time too.

So you need time to clear your head. And I think that the sabbatical work in the sense that I was able to clear my head and sort of refocus things. And I think I convinced myself of the importance of this type of work that I've been doing with two esteemed co authors.

And you may know Peter Crampton, who was at University of Maryland for a long time and is now down in La Jolla working on the consulting firm that a new effort actually was. Peter Crampton Associates, that I'll mention. But this is a series of papers we did a few years back that focused on one proud to have a short paper in Nature.

But the longer paper published in Jite and several co authors, well, I don't know. People know Kara Kochelman at UT Austin Civil Engineering, but she does a lot in this area. And I suggested you read the final working paper. We got some coverage in mainstream outlets like the Wall Street Journal and the New York Times.

But the point is I kind of like to read, you'll notice this, a little dated. I'd like to pick up this work again as a result of this sabbatical. I think it's become a more salient issue in some ways and I'll explain why. So that's very exciting and I think a productive outcome of the sabbatical.

Many thanks to Hoover for hosting me here once again. I have an ambitious agenda. I might get bogged down a little bit, but I wanted to use this talk a little bit differently to try to motivate things. And this is a result of reflecting, you know, on this issue and traffic congestion in the Bay Area and sort of how we got to this position.

So I'm gonna maybe use those guinea pigs to a little bit to talk about some of the benefits we have of high mobility that the automobile and the truck have given us for hauling freight. But we've gotten to a point where it's kind of disruptive and there's some big social costs.

And I wanted to launch a vision, if I could, for a new thinking about a new structure, a new system for the transportation network. And I say that because I think the tech, I'm pretty certain the technology is ready. So in the times of earlier commentators on this, there was a technological barrier to the adoption of real time network wide road pricing.

That's no longer the case. I wanna embed that in the intellectual background of this which is very cosine in our ideas. And of course, we have to talk about William Vickery in this context coming from Columbia. You have to do that. And we use mainly the structure of wholesale electricity networks under the control of an independent system operator and ISO to clear the market in wholesale power in a real time basis as a template or a benchmark or a model for thinking about how that network might work in road pricing and transportation.

We'll see how far we get before John cuts me off. We only got one. Very ambitious another John, yeah, go ahead.
>> John Cochrane: Topics I noticed we're missing that I just want to invite you to comment on which is that the political economy of it in the sense, congestion pricing.

We're all on for it. And look what a disaster in New York.
>> Rick Geddes: Yep.
>> John Cochrane: That now in part cuz they sold it, where they sold it is this gonna save the climate and fund the MTA, and I don't want that. On the other end, San Francisco, we actually have real time tolling.

I never thought. But the poor people can't whatever so and carbon taxes are, you know just a grand failure of selling something to people. So how to get people to sign up for it, I hope will to agree to it as I hope part of this thought.
>> Rick Geddes: So I guess, John, I'd like to separate the economics and policy from the politics if I could and kind of think about that's the most difficult challenge to say to people about yes, we've done New York congestion pricing.

You say it wasn't a political football, it was a political Super Bowl, right? That became so political and it has been over time. I think I'll get to this if we have time. But I'm not expecting this idea to really take place first in the United States. First of all we have pretty polarized politics you may have heard.

Second of all, we have a lot of legacy infrastructure. So just drive around, you'll see and I'll get to that in a second based on the automobile, right? And a lot directing the driver by line of sight, road signs, line paint, you yield signs. All of this stuff that I'm going to argue is obsolete.

So it might occur first in a place like the Middle east or in Africa where they don't have a lot of that. And also our third problem is that we have a balkanized ownership structure. So the road network is owned by a patchwork of states. The interstate highway system is completely owned by states.

No, federal, but then we have local streets that are owned by municipalities. We also have county roads. And so to coordinate all that in a road pricing sense with an ISO up above, it is a bit challenging to say the least. So I think the political economy might be a little bit better in other countries where you could use those as test beds than it would be in the US Anyway.

I know that's a cop out, but let me think. There's also ways of making compensation, right. We don't medically think about how you're going to use the revenues from this and people who operate these dynamic toll lanes. Confidentially, public owners tell me it raises a lot of revenue, right?

It's very valuable. Willingness to pay is high. So how you use that to get the buy in of various political groups is maybe something we could chat about, but that's not my point. Sorry.
>> Speaker 4: The self driving car.
>> Rick Geddes: Yep.
>> Speaker 4: So we're going to be able to sit in the back seat and read our manuscripts.

>> Rick Geddes: Yes.
>> Speaker 4: So why do we care whether.
>> Rick Geddes: So can you hold that for like two slides? Maybe three? Because my other benefit of being here in the Bay Area is I got to take a Waymo, and it was like. It was a revelation, right. So hang tough and I'll start with some history here.

We'll start at the beginning. In the beginning. I wish Oliver. Well, Oliver's not here, but my co author, Axel Oakenfels is in Cologne now and this was in. In Germany. So this is an auspicious year, 2025, for many reasons. But first, it'll be 140 years since the. In what we in my world accept as the invention of the modern automobile in Mannheim, Germany, by a guy named Carl Benz.

And you might recognize that name in automotive technology. Ben Bends. Yes, it's the same guy. So that is what we consider to be the first sort of series production car brought into service. It shares a number of commonalities with the modern automobile. It is steered by the drivers holding a steering rod, in this case by line of sight.

It had an internal combustion engine. It was put in series production. I want to focus on the picture for a minute to motivate the talk. Because if you look at that picture, everything we've done, and I'm using, we in our species, humans have done for the past 40 years, are just changes to that basic fundamental design from Carl Benz.

What have we done? It's a tricycle. It has three wheels. We added a fourth wheel, okay. We added a passenger compartment. You can actually drive one of those replicas, I think in, in Stuttgart, where the Mercedes museum is. It goes pretty fast, so it's dangerous. So we added a passenger compartment, we added much more powerful engines, we added a better braking system, we added a collapsible steering column, padded dash.

We added a trunk and we added a lot more bulk to it that you'll see out on the road today. But that is the basic thing. The other thing that's even more important, roads and cars are strong complements. So we've added a huge system around the world of roads for that thing to drive on, made of concrete or asphalt that go everywhere.

And they have traffic signals and they have stop signs and yield signs and everything else to allow you to take that on a trip from point A to point B. One of my subtexts of the talk today is to suggest that this surface transportation sector is really ripe for disruption in a main way.

And we can't. Well, maybe that's too normative. I don't know. We shouldn't be in a position of just continuing to build that out. We've added lanes to intercede highways. People around here know that. People in Los Angeles know that even more. And we've had some problems with this.

I do say though, and there's some compliment to the automobile highways. So I tell my students that they should, should study infrastructure for their careers. Of course I want to see them, them get into this area. But I say one of the reasons you should do it, we'll give you an attitude of gratitude.

One of the most beneficial things in life you will be understand infrastructure and be grateful for the infrastructure that our parents, grandparents, some case great grandparents built. That was really a lot of work to build and has given us unprecedented mobility that people 100, 200. Harold.
>> Harald Uhlig: If you imagine the cities had grown the way they did, we didn't have the automobile.

We would still have horse drawn carriages, right and we'd still need, roads to drive those on. I mean we maybe would have looked slightly different, maybe more attuned to horses rather than cars.
>> Rick Geddes: But they don't go as fast, right? So that's true.
>> Harald Uhlig: I like a car more, that's the point.

>> Rick Geddes: But we would certainly have some roads. But I don't know, if we were still riding horse drawn carriages, we would have built that. The clover leaves. Point well taken. I take your point.
>> Speaker 4: Manure.
>> Rick Geddes: Well being a little manure problem.
>> John Taylor: All right, let's go on.

>> Rick Geddes: That's true disease. I'm not blaming Carl Benz or the car, or we know Don Miller was his buddy and it evolved into Mercedes Benz. But so I'm saying we should be grateful for the mobility we've had for a long time that people a couple hundred years ago could only dream of.

You know, I'm going to drive down to LA today. That would have been crazy to say a few hundred years ago. However, I submit we're at the point where pushing this basic model has caused some, some pretty serious social problems. The first that photograph could be anywhere in the United States, in the developing world, and you'd see basically the same problem.

The inrichs is kind of the keeper of the data globally on the costs of traffic congestion. And it just came out not too long ago. I was interested to see that the US and Germany match each other for the number of hours per year lost, basically a full working week in traffic congestion.

I was also interested to see that the United Kingdom beats us, if you want to put it that way, in terms of hours lost to traffic congestion. A couple things. There's a cottage industry of literature on this. I guess we could argue about how clean the identification is and so forth, but finds pretty convincing evidence intuitively.

The traffic congestion is closely related to a number of negative outcomes besides just lost time and fuel and etc. Frustration worsens air pollution, as we know, there's a lot of evidence on fine particulates from diesel fuel, PM2.5, which your lungs cannot filter out. That is bad for you.

So there's a lot of bad health outcomes, increases in obesity, divorce rates, et cetera. One thing to note about this, I think, is that for a long time, and particularly in developing countries, this problem has been growing worse as countries develop. One of the first things people want to get is an automobile and they want to move off the farm and into the city.

You have greater densification and you have worse traffic congestion, Peter?
>> John Cochrane: Yeah,
>> Nick Gebbia: I'm thinking about this picture and thinking another problem is that maybe the cities are too big and there are distortions that, right. You get megacities that emerge in developing economies, for example, because that's where the government is.

And so you need to interact with. Right and so it may be that thing you're saying that road network is messed up, but it may be the systems of cities that are funneling people. Places that are too congested.
>> Rick Geddes: Yep, could be. I wonder, of course, my mind is wondering to that to what degree is that due to the mispricing, specifically the under-unit pricing of road use.

And that is, I think, a core problem that drives a whole bunch of distortions. Well, I'll get to one of my favorite Vickery quotes from 1963, but John.
>> John Taylor: Time wasting traffic is endogenous.
>> Rick Geddes: I'm sorry.
>> John Taylor: It is endogenous. It's also the fault of zoning laws. If people could live near work, they wouldn't be driving in the first place.

>> Rick Geddes: Yeah, and I think our. No, that's a good point.
>> John Taylor: Can I partially rain on this parade for a second, please?
>> Rick Geddes: I'm sorry.
>> John Taylor: I'd like to partially rain on this parade for a second.
>> Rick Geddes: Okay.
>> John Taylor: We've adopted so many things inside cars that allow you to do things inside your car that you couldn't do before.

You're really not losing all that time because you're doing other things. Let me just finish. Okay. And in fact, the studies that have tried to look at the shadow value of an hour spent in traffic congestion are well below wage levels. They're like half the wage rate. So something's going on.

So I'm not saying it's not a problem. I'm just saying to lay out all the costs without saying that we partly endogenously adapted to this with lots of things inside the cars, with our use of technology, be able to be on the phone while we're driving in traffic ingestion, listen to music we prefer instead of wasting money on a concert, etc.

All those things. So, I think that's just kind of a useful kind of corollary to what you're doing or.
>> Rick Geddes: Yeah, I wish I had brought in the data on that. But people in my world, it hopes that technology inside cars, like increase airbags, stronger protections against rollover and so forth.

>> John Taylor: Safety is a big issue. I'm talking about the value of your time.
>> Rick Geddes: I know, but that's related to this. And that's caused a lot of distracted driving. And you can, you can see people on the highway doing this. And that has resulted in higher fatality rates on the roads than what we had expected given some of the technological improvements in safety of the vehicle itself.

So I don't disagree. And I've heard estimates of half the wage as well, but it hasn't come. It's not free. Right. Of trying to be productive while you're driving is a challenge. Because our brains can't multitask that way.
>> John Taylor: The point is we ought to net all these things we're talking about.

>> Rick Geddes: Okay? So anyway,
>> John Taylor: rationing with time is inefficient. Compared to rationing with money. We're all economists.
>> Rick Geddes: Point well taken. But, well, let me get into. Here's just a quote on air pollution. And again, that could be anywhere in the world. Air pollution, traffic congestion, 83 nations of the United States contributed 2,200 premature deaths.

So hopefully we're all in agreement that there are pretty negative effects from traffic congestion. Health is one. I know living here that people consider this area to be pretty congested in terms it is the Bay Area, but it doesn't meet the top 10 in terms of places from the recent Inrix data.

And this gives you some flavor for the social losses associated with traffic congestion. There are certain things like stress and frustration that I don't think INRIX includes in the data set. So, these may be low estimates in some sense anyway, and it's growing. The other thing is, of course, motor vehicle accidents, deaths.

And again, this gets back to the issue where there was a hope that it would decline, but it still is pretty high. So in terms of 2023 data from the National Safety Council, a very high number of people continue to die. Of course, injuries aren't included. It's easier to measure deaths because you have a death and that's pretty clear.

Injuries should be added to this. Death rates peak around 20.2 per 100,000 population per year for 21 year olds. And again, when you're 91, leading preventable cause of death for every age group from 5 to 22. I was surprised about the very young children who were dying in traffic accidents.

Second leading preventable cause of death for every age from 20 through 67. Those are numbers. What's the trend line over time?
>> John Taylor: Can you show us a trend line over time?
>> Rick Geddes: I can in a second.
>> John Taylor: And that number does not seem particularly high relative to the historical one.

>> Rick Geddes: That's kind of hang tough. Again, I did. I want to, maybe people can give me advice on how to put the human perspective on this. And my dear friend Barry Strauss, who is now a fellow at Hoover, is a historian and we were talking about the Vietnam War recently and how it was a 10-year period and that people from his and kind of my generation had a sense there was a lot of.

It was quagmire. A lot of deaths. Takes us about 16 months to kill the same number of people on Our roads who died in the Vietnam War. For younger people, Maybe they think 911 was a clearly a traumatic day. A lot of people died. Takes about 24 days to kill people on our roads.

So, we have all these safety devices. We have all the bulk around the passenger compartment, so forth. There's a literature on this. But my point is that to the degree we think reducing congestion through real time network wide road pricing reduces traffic congestion, I'm going to argue increases the adoption of autonomous vehicles and thus reduces traffic accidents.

It's a good thing to promote. One of the interesting things from the data on congestion pricing in New York City, very limited time when it was implemented was the reduction in pedestrian car accidents. People from New York know it's very much a pedestrian city. So, if you have the less congested traffic, you have a fall in injuries associated car pedestrian.

And it was something like half. So, it was traffic congestion does cause more accidents.
>> John Taylor: Because the cars are moving a lot faster.
>> Rick Geddes: I would expect the cars in New York aren't going anywhere moving a lot faster when you get rid of the congestion.
>> John Taylor: So not obvious that.

Offsetting effects. So we netted out.
>> Rick Geddes: But what we observed for that short window, I admit of time was a reduction in pedestrian harm. So this is the fatality rates per 100 million miles traveled just up to 2020. I thought this was interesting because Covid.
>> Speaker 8: Compare this to other forms of transportation.

>> Rick Geddes: No, I don't have time today. You and I can chat about that if you think that would be useful. I mean transit's obviously going to be more safe than in your own. But if you mean bicycle or something that would be. Would be. I just thought this was interesting because people again thought that when everybody was sort of in lockdown during COVID there would be a reduction in this.

But people kind of went out and went joy riding and so they were actually more per mile driven an increase in the risk associated with their driving. So it's very psychological. I just thought that was interesting. And we have seen again, I'll try to in future talks get something to indicate the increasing effect of distracted driving and people trying to work while they're driving.

It's one thing if you're in the backseat, but if you're the driver that's a problem. I want to argue next that we're at a pivotal moment. In technology that has changed since the day of Carl Benz, okay? And that we're now in a situation where. And even since Vickery's time where we're now in a situation where it would be possible technologically to address many of these problems through the allocation of the use to John Floyd use of road space through unit variable unit pricing of roads.

Those changes specifically are autonomous and driverless vehicles. Waymos, we could all take a bus and take a Waymo up in San Francisco. Now is level four autonomy where it still has a steering wheel and has a brake and has all this other stuff that's going on at own.

Level 5 would be to get rid of all of that. And I guess you just have a big red button that says stop or pull over. But what we have it's amazing, right? I urge everyone to do this. I've sometimes thought one of the best things we could do with our subsidy money for transportation is buy everybody a Waymo ride so that it takes you about five minutes to adjust to that and realize it's a better system.

And then you can really work and be productive even if you're in the front seat. The other thing is location technologies. This gizmo can in real time, whether you like it or not, figure out your location in the X, Y and Z plane to the sub meter level.

And it that has gotten much more important. There's also this discipline I like to read more about called kinematics, which is the study of an object object's motion in space without rep. So my arm is causing my phone to move. Physics would normally study that my arm pushing the phone.

But kinematics just studies the movement of an object in space without reference to the force that caused the object to move. You might think that for real time road pricing that would be very important. Also I want to include alternative fuels. I'm not saying electric because I think there's a consensus that long distance heavy trucks do not work well on batteries because the batteries become too heavy at some point.

The alternative I include hydro in that. There's just a sort of a revolution going on in alternative fuels developments in market.
>> John Taylor: Before you go on with that again, in the spirit of netting, electric vehicles are generally heavier.
>> Rick Geddes: Right.
>> John Taylor: And the evidence now is most smog type particulate matter is coming from tire wear, not from the.

Yes, they're worsening smog. So when we knit all this, we need to.
>> Rick Geddes: You mean because the added weight.
>> John Taylor: Added weight causes greater wear on the tires which generates particularly more metric.
>> Rick Geddes: Yeah, yeah.
>> John Taylor: So I think there's been an obsession on just looking up a tailpipe, not looking at things from end to end.

And that is very dangerous, for example.
>> Rick Geddes: I'm sorry, could you say the.
>> John Taylor: Now there's such an overemphasis on just what comes out of the tailpipe.
>> Rick Geddes: Yeah.
>> John Taylor: Electric vehicles are clean. Well, where are the batteries produced? How dirty is the production of batteries? You have to dispose of the stuff.

You got the wear and tear in the tires. And I think right now if you look at a fairly high mileage gasoline powered car, it's about end to end, about neutral with respect to electric cars and carbon emissions. Now that'll get better. There's a larger fraction of electricity comes from lower carbon intensive fuels.

Just want to get back to netting everything.
>> Rick Geddes: Yeah, yeah. So that just to be clear, that's why I use the term alternative fuels instead of electric, because I want to. Hydrogen. Right now you can't get a permit to go under a tunnel or a tunnel or over a bridge with a hydrogen powered automobile.

Yeah, we need to think about blowing.
>> John Taylor: Up on a bridge or in a tunnel.
>> Rick Geddes: Yeah.
>> John Taylor: Also Mike is a lot more combustible. And flammable than gasoline. So I'm not saying those are impossible to deal with. But several of us here were advising Arnold when he was governor of California.

He was big on a hydrogen highway. And we're no closer to hydrogen highway than we were in 2008. So just.
>> Rick Geddes: Yeah. And hydrogen trucks I think are the. I understand there's problems with hydrogen. It doesn't. It's a small molecule, doesn't like to be stored. It's hard to transfer, it's hard to pipe it.

Anyway, we could talk about that. But I think. So this is part of my thinking on the future. Yeah.
>> Speaker 9: If you want to save the planet, not everything has to be zero emission. There are a few great use cases for fossil fuels, long distance trucks and airplanes.

>> Rick Geddes: Right.
>> Speaker 9: What's everything else? But there's also two other things that, which you may have on later. Bullet points. Real time. Every single parking space ought to be charged for. Not just charging for moving, charging for parking. And the other thing is the possibility which hasn't come to.

Most cars sit idle 98% of the time.
>> Rick Geddes: Yeah.
>> Speaker 9: Why do we have to own a car? We should be able to, you know, not just summon a waymo, but rent cars without all the transactions costs of Hertz, which are enormous transactions cost and therefore not have to own or maybe own one and then rent the Other one, you know, that would also help the parking and dramatically reduce the capital cars.

>> Rick Geddes: So that's a couple of slides. The question, John, that I had, the interesting estimate would be how much would your transportation overall transportation cost be reduced if we could flip a switch on the wall over here and switch to that system versus what we have. Like would it be half, you know, of our cost now of owning a private automobile that sits in your driveway and you pay insurance which is higher because of all the accidents and well.

>> John Taylor: It depends on how many costs you get to shift on others. If I have to pay for my parking, all of a sudden the, the incentives to not own so many cars rise.
>> Rick Geddes: Yep, yep. So hang tough.
>> John Taylor: Cost including shadow values. Yes.
>> Rick Geddes: Yeah. For.

Well, I don't know if it's for Peter's benefit, but I have to include developments in market design right here at Stanford that assist with thinking about how this would work. In fact, I'm not gonna show the website even though I have policy developments. We have the ability right now to implement transparent, efficient pricing and transportation.

Peter just got back from Bangkok where he was helping the Thai people to set up their, their spectrum auctions and I think electricity as well. And he created this new entity, I guess out of his old firm called Forward Market Design, which has a team that is ready to do this.

The platform is built, maybe a little bit of advertisement for my buddy. Platform is built and available open access, open source and any country that wants to try this could get help and assistance and advice with Forward Market Design. So in the interest of space all. Well, Shelby's here from cab new visiting.

Thank you very much. Correct me if I'm wrong, but basically a 40 mile drive. So this is about innovations. Innovations that are occurring, that are very important in thinking about the next generation of the surface transportation. I'll leave aviation out surface transportation system basically between downtown Detroit and Ann Arbor.

It's along Interstate 94. As part of it, I think I know the people, both COO, the CEO and Carl Reichelt who works with Tyler duvall who's the CEO opened in July of 2024 and I think I can just advance this. And it'll play. So here we go.
>> Speaker 10: At Cavnue, we are building the world's most advanced roads.

Cavnues are dedicated lanes that enhance connected and autonomous transportation, making travel faster, safer, and more accessible. Our infrastructure will enable vehicles to talk to each other. Cavnue will improve how vehicles collectively perceive and react to their environment. Cavnue is optimized public transit with the efficiency of dedicated corridors.

Letting you travel safely, quickly, and autonomously. Cabinet building the future of roads.
>> John Taylor: So it doesn't have the same problem that trains have. They go from one place to another. But what do you do when you get there?
>> Rick Geddes: You mean with this particular.
>> John Taylor: Yeah. Can take you the last mile.

>> Rick Geddes: What's that, John?
>> John Taylor: Driverless car can take you the last mile.
>> Rick Geddes: Yeah, the driverless car would solve that problem.
>> John Taylor: But then why do you need this? If it's a driverless car, why can't you use the roads we have? Why do you need that?
>> Rick Geddes: Well, this, I view this as a step in toward that.

I mean this is revolutionary in the sense that they had dedicated lanes on an existing United States interstate highway which, where you have to get permission from in this case, Michigan dot to do that and to run these tests. Where the infrastructure is talking to the cars is something bit of a holy grail in this world.

It's called VII, which is vehicle infrastructure integration to make this happen. The other thing you'll notice from the video is something called platooning, where you get the vehicles to be very close together without an increased risk of accidents, which is a human thing. And therefore you make better use of the much better use of the available road capacity.

I've heard of estimates up to 50% more vehicles, throughput vehicles per lane, mile per hour could get through when you have that sort of platooning effect in there. But, but if I were to be able to pull the lever on the wall, I would of course price the entire road in real time and eliminate the traffic congestion by having real time prices fed into a.

An app like Waze or Google Maps that would tell you the way to go, you know, based on road prices. And that would, I think, address your problem.
>> Harald Uhlig: Harold, someone watching this, I couldn't help. But think that this is the lead into a major Hollywood disaster movie.

>> Rick Geddes: If you watch John Oliver, the commentator, he has a hilarious segment on infrastructure. And the first part of his segment is about how Hollywood loves to blow up infrastructure. Like creatures from outer space come and the first thing they do is just destroy our infrastructure. I don't know why that is, but it's True, John.

>> John Taylor: Caltrain the other day, the other that it is to be able to use railroad right of ways. You can't really use railroad now, but what could you use it for that thing?
>> Rick Geddes: Yep. Yep, and right of way.
>> John Taylor: To be able to spend $27 billion to put electricity on Caltrain.

>> Rick Geddes: Getting right of way is a big issue. And that's another reason for doing what Crampton, Gettys, and Hogan Bells are arguing for pricing in real time. Okay, John, here's the inspirational vision. So I know I wanted to show this now. Agreed, you can use fossil fuels. I'm not, you know, attacking fossil fuels, but I do see this.

Whatever makes sense. What's that?
>> John Taylor: Whatever makes sense. It may not be fossil fuels 20 years from now.
>> Rick Geddes: Yeah, it may not.
>> John Taylor: Yeah, may not be fossil fuels down the road, just have to see.
>> Rick Geddes: Okay, okay, so I'm just using the term alternative fuels, which I think is a lot of innovation and power.

I think I wanna sort of divorce the transportation services from the fuel use. I maybe could talk about that a little bit. I don't want to get too much time on that. There is innovation going on in the way we power vehicles right now. And I think that's very interesting.

In the upper left, the blue one is cav connected in autonomous vehicles. I can promise you that there's innovation going on in that. It's like the OEMs, the original equipment manufacturers, Tesla, Google. I mean, they're all working hard to increase the feasibility of more autonomy. And that will, I am pretty convinced, reduce that fatality rate that we see today disproportionately young people who are dying on our nation's roads.

So the engineers are pushing out on both the yellow and the blue. Where are we not seeing as much innovation? Well, with all due respect, it's us. It's us economists, right, all due respect to PhD economists in the room, is working on the feasibility of pricing road use in real time, network-wide, just like utility, yeah.

>> John Taylor: How do you sort of elaborate on this?
>> Harald Uhlig: So it seems- to me that once you have autonomous vehicles, the value of time changes dramatically, as Michael was saying, that you can use your time for other purposes. So therefore, the costs of congestion decline. And also, congestion will decline because autonomous vehicles can be bunched together, as you've already said.

So why doesn't that just solve the problem by itself? What problem is left?
>> John Taylor: More people say, I can go in my driverless car and spend three hours commuting. I don't care. The amount of congestion could skyrocket. It's not obvious unless maybe, but I'd rather be at home on my sofa.

>> Rick Geddes: A better world if we like. I don't think people say, gee, I love being in this enclosure.
>> John Taylor: Better technology is always better, yes.
>> Rick Geddes: What's that?
>> John Taylor: Better technology is always better.
>> Rick Geddes: Better technology, it moves the production possibilities. May have learned that in graduate school, yeah.

But I think these things are strong compliments. That's kind of my takeaway. I think that if we could get the real time road pricing right. And part of the takeaway is from our work is, is we don't think it's that hard to do. It's hard politically to John's point, but technologically not that hard to do.

It would push the adoption of more connected vehicles. Yeah, go ahead.
>> John Taylor: We have some examples and some history in the Bay Area and it was pretty clear for a long time they weren't following our basic optimal pricing rules. First we had HOV lanes which were mostly empty and there was greatly increased congestion in the remaining couple of lanes on the highway, which clearly can't be optimal.

And now we have some real time pricing, which you're advocating, and some in Washington DC, etc. And that's probably a step forward. But it's unclear that the pricing is being done properly.
>> Rick Geddes: Yep, yep.
>> John Taylor: And so it's not just of getting people to adopt it, but it's actually doing the right pricing.

>> Rick Geddes: Yep.
>> John Taylor: And the confidence of the people. And when you're talking about a vast move to economy, all that sort of stuff and being very close together. I'm sure they're getting it really gonna get pretty darn good for you and the vehicle in front of you and behind you.

But somebody has to be paying attention to the whole system.
>> Rick Geddes: Yep.
>> John Taylor: And that's gonna be some government employee somewhere. I don't know your equivalent of your ISO, but that makes me a little nervous. Paul Milgram, Nobel prize winning economist, department has worked with the FAA on an analogous problem of air condition for aircraft.

And he says he's built a model, it's got 130,000 constraints.
>> Rick Geddes: My God.
>> John Taylor: Now from here to San Francisco isn't going to be 130,000, maybe a thousand.
>> Rick Geddes: Right.
>> John Taylor: Okay. So I'm just trying to figure out who's going to get all those shadow prices, right? Who's going to get it all organized.

Even if we have an autonomous vehicle that won't bump into the car in front of us or the car behind. Us, that's the role of the independent system operator. You're right about that. And that's the role it plays in wholesale power markets. And there's an organizational governance structure for that that we talk about a little bit in the paper.

So I think our benchmark model would be what happens pretty successfully in wholesale power markets and see how that could be transposed into surface transportation.
>> Rick Geddes: I don't disagree with anything you said, but I think that would, I'd lean heavily on that.
>> John Taylor: I think it would be really, really beneficial if you could look at the sensors and the traffic patterns where we're using real time pricing right now.

For example, highway 237 in the south bay. I think that would be an interesting comparative case for you to look at.
>> John Cochrane: Why is this such a hard problem? I thought that's what markets do. And so if you just raise the price until cars are going, you're set.

You don't need to solve this huge planning.
>> John Taylor: Well, actually it's more complicated.
>> Ellen McGrattan: Can I give personal evidence on this? Because my house sits on a dynamically priced highway and here's what I do. I look at the price and if it's over $2, I know somebody in the government office screwed up and it's a parking lot.

So it works very well in certain normal situations. But anything goes wrong and the constraints get violated and all things go wrong.
>> John Cochrane: Well, part of it is, I guess you don't yet have futures prices.
>> Ellen McGrattan: There's no future prices. Exactly, exactly. Something like a crash happens or something like that, and they don't deal with it well, and then it screws up the whole thing.

And then you have to be an experienced driver on this road to know, aha, there's been a crash.
>> Rick Geddes: Ford markets are coming up. But John, this gets to your, this is your point is if you think about the oval in the middle, you would, you know, in the future, it's nice to think about the future.

You would. And I think we economists need to offer an inspirational vision of how we can add to social welfare. And you could pay for a subscription service where you're paying for transportation and an autonomous vehicle to come to your house when you have to commute to work.

Maybe it has a couple other. Maybe it's a small pod with other people in it. And we have the technology now in the evening, it would go away to some other parking space, maybe recharge at night. And you would just pay for that service monthly, right? And my guess is it'd be interesting thing to really get into that.

That would. Your overall transportation bill would be much lower as. Than owning your own vehicles that sit in your driveway all the time. So that's kind of what, let me.
>> John Taylor: Just add where we're trying to be constructive here. So don't take. I think we more or less, all of us think that this could be a big improvement.

>> Rick Geddes: Okay.
>> John Taylor: The question is, in the real world, what will happen? So I think we're trying to be constructive. Don't take. This is criticism of the idea. Take it as attempts to try to be.
>> Rick Geddes: Yes, I appreciate that.
>> John Cochrane: So let me go back to my autonomous vehicle.

The great thing about the autonomous vehicle is that you don't have to change too much of the legal structure or the governmental structure. You don't have the same political opposition that you have to congestion pricing. The one thing we know about congestion pricing is, is that it's extraordinarily hard to implement on an old infrastructure.

If you're building a new lane or new highway or something like that, you can do it. You did it actually with the Pennsylvania toll road a long time ago, but it was a new entity. So once you have a system in place which was free, people are not going to want to pay for it.

That's what's happening in New York City. And it encounters this resistance time and time again. Now, the thing about an autonomous vehicle is that even if nobody else adopts it, you benefit from it yourself because you can use your time for other purposes. As more and more people choose to do this on their own, then you get social benefits, you get spillover effects.

So you can see a political process that can lead to general adaptation. But I don't quite see how you're going to get congestion pricing at scale in an old infrastructure, in a legacy infrastructure, which we're going to have for another 100 years at least.
>> John Taylor: Well, that's a political objection, not a technological.

>> Rick Geddes: Yeah, that gets you used.
>> John Cochrane: To have something called political economy. When the economists decide they didn't have to talk about politics. I think you've given them too early. So take New York's congestion pricing. How did they sell it. They said, this will save the climate and we'll pour more money down that literal rat hole called the npa.

If they had said instead, we're going to clean up congestion in New York and we're gonna cut your taxes, that might have a lot better chance of getting some.
>> Rick Geddes: And John, they would have. Instead of saying $15 an entry, which was the original proposal, now it was nine, but heaven knows what's going to happen Next.

But that 15 was not driven by the optimal price. That's gonna clear the. It was driven by the MTA's revenue numbers. Yeah, right.
>> John Cochrane: So, so that I'm identifying the political issues that come up every single time you talk about congestion pricing. People want to take that money and use it for other wonderful good things like saving the.

>> Rick Geddes: Let me, let me say that my response to John initially was that I don't expect this to take place in the United States first. I expect it to be in a place where a lot of, where there's not a lot of legacy infrastructure. And some, you know, some other places like Hong Kong, Singapore are more open to these ideas.

And then those will be test cases where you can see how this works. And then maybe, maybe we'd be last. I don't know. But I don't disagree. The other thing I'd say is that I'm curious about whether if you just have a transportation bill and you're in a connected, an autonomous vehicle, if you have less resistance to road pricing, if that's just a cost of you using your, your, your transportation service.

Right. It's less salient in some sense. And would that enhance public acceptance of road price? I don't know. But, but I think that you're getting behind. I'm getting behind. Sorry. Okay. He Insights connected autonomous vehicles, alternative fuels. Value of progress is limited. If those vehicles are stuck in traffic, I'll stick by that.

Progress needs to be made in this area. That's called for further research. And I do think that free flow, traffic and autonomy are likely to be complementary values. Okay, just a simple analysis of this would say here's one of my favorite infrastructure facilities is the New Jersey Turnpike at 4am Basically, you have free flow.

The market clears. In some sense there's a marginal use of cost of use. Demand is relatively low, supply curves relatively inelastic because it's hard to change capacity for infrastructure in the short term. Visited the Golden Gate Bridge. They do have a reversible lane on the Golden Gate. So incoming.

Obviously you want more lanes in the morning coming into the city, but it's pretty inelastic. And then this is not to scale. But when demand shifts out at 8am for the new Jersey Turnpike, we would think of that as a shortage of road space. Or colloquially traffic congestion, right?

But the key thing with traffic congestion is layered onto that is something much more profound. In some sense this is Herbert Mooring, who's famous guy in this added complexity uses of road and other transportation networks not only experience congestion, they created deciding how and when to travel, they take into account the congestion they experience but not the negative externality they impose on others.

And this is highly non-linear. And we've all seen this driving around a little bit. Social costs go up at an exponential rate once you hit some critical level of traffic congestion. The term gridlock was created by a New York guy named Gridlock Schwartz who talked. It happened in New York City at one point.

So we all kind of know this as on the horizontal vehicles per hour increases. But then you hit a critical rate or critical rate of throughput and the travel time and congestion costs go up exponentially. And that's actually a pretty profound graph. Tell the students again, they're used to looking at it from left to right where it goes up, but it also works in the opposite direction.

So if you can shave 5, 6% of the vehicle and the morning peak is more sharp than the afternoon peak, if you can shave 5 to 6% off the morning peak with accurate pricing of road use with John's Point, both spot and forward markets that tell you the direction of change in prices, then you can ride down the curve and get back to free flow traffic without too much difficulty.

>> Ellen McGrattan: And your argument about the pods is it flattens.
>> Rick Geddes: What's that?
>> Ellen McGrattan: Your argument about the podding that could produce the effect is that it flattens that curve.
>> Rick Geddes: Absolutely. You mean platooning.
>> Ellen McGrattan: Yeah, yeah, yeah, sorry, my apologies.
>> John Cochrane: Curve is, is backward bending the freeways, you get the maximum flow through at about 40 miles an hour.

>> Rick Geddes: Yes.
>> John Cochrane: And then if you try to put one more car, it all goes down to zero. So that actual vehicles per hour goes down.
>> Rick Geddes: Yeah, that's the problem is.
>> John Cochrane: Yeah, you're able to make the nose of that thing move out a little bit. But then you say you got now 10 times, you get maybe a third more cars per hour.

But still you get to that point where put one more car in, boom, it all stops.
>> Rick Geddes: Absolutely. I should probably, that's real simple. I should probably put that on because you're right. And that would be the gridlock where it just collapses. Right. And you have to write a.

>> John Cochrane: Great paper on this. Freeways, it's people, how tightly do people go?
>> Rick Geddes: Right.
>> John Cochrane: So like if you're going 65 miles an hour, people Then will slow down but get closer to each other. So if you add more cars, you can just slow down and get more cars per hour.

I think 45 is about the amount where you're not gonna get any closer to the car in front of you. Now you add one more car, we slow down to take care of one more car. Boom, it all stops.
>> John Taylor: It still stops. Yeah. That's future versions of this, John.

>> John Cochrane: I will do that and demand traffic per hour and actual throughput of traffic per hour. Your supply and demand graphs had that problem too. More people want to go through our but they can't.
>> John Taylor: Yeah, they can't. Yeah.
>> Rick Geddes: So I have seen those graphs. I'll put that up.

>> John Taylor: It would also be instructive if you have a long list of attempts to use markets from my own experience. I'll say something at the end. But right now there are two that I think are relevant. One is we've tried for a long time to get the bridges to use congestion pricing.

The bridges is in the Bay Area. Across the bay, across the.
>> Rick Geddes: Yeah, just eight bucks no matter what time it is.
>> John Taylor: Yeah. Ye been an abysmal failure. I mean testimony at hearings, etc people nodding. And it just. The price reflects their maintenance costs and their projected maintenance cost, nothing else.

And then the biggest fiasco was high speed rail, which is now blended speed rail and probably not even going to get built in the Central Valley. Was sold on the assumption there'd be a massive increase in congestion on California's roads. Yeah, they're predicting 60 million people would be living in California in 20 years.

Stuff like that. And this was how it was heavily justified. If you actually look at the data, they were projected to decrease vehicle miles traveled by 1% in the state. And maybe that 1% was at the peak of congestion. So there's this problem. There's this problem of something sounding good.

And then actually when you get down towards how it's actually implemented being at violent odds with economic.
>> Rick Geddes: So, Hank, I give a slide. Well, I think it will address at least the first point you made on the use of markets in different contexts. But we have to talk about Vikram before that.

And I did go back and read the May 63 AER paper and this is the beginning, igrht? I begin with the proposition that no other major areas pricing practices so irrational, so out of date and so conducive to waste is in urban transportation, two aspects are particularly deficient.

The absence of peak off peak differentials and the gross underpricing of some modes. So the gross underpricing is road space and that leads to lack of exploration of other modes. You could write this today and it would be equally true. So that's my point about disruption needs to be made both in surface transportation, both in policy and in technology.

So Vickrey was really a visionary, maybe.
>> John Taylor: Could I just add one point about the intellectual history?
>> Rick Geddes: Sure.
>> John Taylor: So the original important paper about marginal cost pricing was by Harold O'Telling in the late 1930s, number one, and actually economists in general was Marcel Dupuis in France about marginal cost.

>> Rick Geddes: What was his name?
>> John Taylor: Marcel Dupuis in France about marginal cost pricing of utilities, especially electricity.
>> Rick Geddes: So just a footnote, I've had some emails with Eddie, Ed Glazer at Harvard, who we all know and love and he's doing some intellectual history of congestion pricing. Alan Walters, who worked with Margaret Thatcher actually wrote, according to Ed, a lot of this up first and wrote, I want to go back and look at that, but I'll ask him about hoteling.

I don't know if Ed has written it up, but Vickery gets a lot of the benefits and he also. So here I just have put the benefits on demand side and supply side. And really, I think simplifying consumer decision making down to one price is really, really critical and you can include externalities.

And of course the vicri's point about optimizing across modes is really important and I think just there's a whole bunch of margins. If people knew the price and they knew the direction of price change for the use of the road via forward markets, that would be revolutionary. So you know whether you should wait 15 minutes, have a coffee, answer some emails and then go or should you put.

Put your pants on and go immediately as fast as you can. We don't know that because of the uncertainty associated with travel time, Harold.
>> Speaker 4: So why isn't the natural starting point not really fully cosy on perspective where we say we all have already property rights to drive on streets.

As of this point, we just endow people with that. We use license plate readers in lots of places to just measure on which road you're driving, right? So it's really easy to determine these days where you're driving, when. And then people just sell and buy these rights to drive on the street.

Now how would you do this? You do this on iPhone. We create an app and, and there's trading taking place and probably put in your preferences of time versus money and. Your app will solve for you the problem when you can try it and at what price.
>> Rick Geddes: You're reading my mind.

>> John Taylor: Exactly, because some people would get money.
>> Speaker 4: You can make money on this, sell your right to drive.
>> Rick Geddes: You can make so much money. I do want to emphasize the supply side. You absolutely agree that I've launched some ideas about that in a couple slides with my remaining time.

So. So I wish I had time to go, one of two things from this. Technological innovation is not occurring as fast as one might hope, in part because you don't have pricing of road capacity. If you had this and it was real business, I think you'd see a lot more innovation in this space.

Again, that gets my point about Carl Ben's original invention. The other thing that I think is really important, Adam Smith wrote about how spending on infrastructure tends to get politicized. And it's a famous quote. I put it on my syllabus for the students to read. Back in 1776, probably before that, the time of the Romans, infrastructure spending was being politicized.

Finally, we could say there's a market signal as to where you should put the next dollar of capacity. Where the price is highest in the peak, that's where the next dollar of investment should go. It takes the politics or reduces the politics in spending on infrastructure. And I think that's.

I'm jumping up and down about that because I think it's an underappreciated benefit of network wide.
>> John Taylor: Yeah, I obviously you have to jointly maximize capacity and pricing and usage, right? So same thing in electricity generation. Same thing, anything else? You wanna build more power plants, you want to decongest.

>> John Cochrane: The political problem of equity. Every time this topic comes up, then some people say, but this is inequitable. So how do you. How do you go?
>> Rick Geddes: You want to go to the equity slides or do you want me to go on? I mean, I understand. So, so let me.

What's that?
>> Speaker 7: Get through this slide.
>> Rick Geddes: Let me march through. Okay, so. So coast for those who aren't living and breathing coase. I've boiled down coasing ideas on how to solve a property rights problem into three easy steps. Okay, first, clearly define use rights to the use of the resource.

To be clear, I'm not talking about selling the roads. I'm not talking about privatization. It's just a use, right? This goes back to the time of the Romans. I'm doing the Latin, right? It's ius, utendi, and the Romans understood stripping out a use right from the actual property, right.

And having that tradability. There's the Latin for making it tradable. And then you have a market. Once you have a market you have prices. And I think there's a consensus here, I hope that we like prices for directing resources. The third step is to once you have the market is to try to make the market more efficient.

Try to make prices more accurate and transparent. And that is a huge benefit. Maybe this is Mr Boskin's point. There's a whole bunch of examples that we don't have time to go through today but where coasing analysis has been applied in various contexts with varying degrees of success.

I understand the electrons moving along a wire is different than cars moving along a road. But the economic principles are pretty similar. I like this stuff. I think about individual transferable quotas to address the commons problem in fishing in off Massachusetts for example have been pretty successful radio spectrum auctions.

We're probably familiar with tradable permits for sulfur dioxide. That was when we were burning dirty coal and nitrogen oxide emissions. And cap and trade had some issues over time. Karok Ray, who was just here, he just left wrote a paper with Eddie Lazear. The last time I was here I used to ride on the back of Eddie's new BMW motorcycle and hold him around the waist as we went up and down 280.

So I thought it quite important to mention both. He worked with becker on a 2013 paper when they had this worked out. And then Karok wrote a working paper more recently on immigration slots. So kind of.
>> John Taylor: Let me interject here for a moment. Since when John and I and Dick Smallcy were the CA, we put the emissions trading into the Clean Air act amendments of 1990.

>> Rick Geddes: Great.
>> John Taylor: And let what was involved, EPA didn't want to administer. EPA did not want to administer what the President decided. So I had to get the President to take all the power away from the EPA to price it to cost it and issue a statement of administration position that he would veto any bill that cost Morris scored by the cea.

And the way we kept the cost down is the administrative EPA signed a document agreeing to 10 different things that would basically move towards sensibility including allowing a futures market.
>> Rick Geddes: Yeah.
>> John Taylor: New Source Review wouldn't require every factory to go to best available technology if they repaired a pipe, etc.

And that's the state of my office now for a while that was done. And it turned out that that reduced the cost of compliance by Schmalensee and Stevins estimates I think by 55%.
>> Rick Geddes: Yeah, it's a good.
>> John Taylor: So it's a good example, there are a couple of others.

There's another good example, which is I did this for NASA on trading different characteristics of experiments on unmanned space probes. Because what used to happen is scientists would build their stuff and fight to get on and then had no mechanism to trade characteristics to get more stuff on each flight to reduce the cost.

And we got Caltech to run that. It worked for a while, the first time it was used was on Cassini, the probe of Saturn 2 that didn't work. California water we had a drought when I was at CA. We brokered actually pricing water briefly and the next year there wasn't a drought and there was no consensus to continue at a lower price.

Just FYI there was a strong objection to pricing. The other one was trying to get two governors of California talking about congestion. It now takes the median bachelor's degree at the Cal State University system is six plus years. And UC system is quite a bit longer than it used to be.

And a big problem is they can't get a course for this major etc. So we tried to get the university system to wheel units so you could take a course at San Jose State rather than San Francisco State if you needed it. And there was just immense hostility to that.

So there are times this has worked in times it hasn't. I don't know if there's anything that can be learned from that. I haven't drawn any general principles. Then you keep trying and you try to get higher than checks free throw percentage
>> Rick Geddes: Would like to get a catalog of these things so maybe you and I can chat.

>> John Taylor: There's also a book called Better Living Through Economics that John Siegfried, the secretary treasurer of the AA put together that has a bunch of these things in it.
>> Rick Geddes: Okay, take a look. Yeah, just to close the loop on the auctions of airport takeoff landing slots. Tyler Duvall, who's the CEO of Cavnu was at the USDOT in 2008 when they were trying to get this.

It's actually the slot through the air that is the constraint implemented. And there's challenge. I'm not minimizing the challenges with this. So let's just quickly go with how this might work where we use an independent system operator as our benchmark or template in energy that plans the market.

And it's very simple and it's mandate to ensure the safety and reliability of the electric system that uses markets to allocate wholesale power. In case you don't have the Map of the ISOs in North America committed to memory. Here's what it looks like. I was surprised to see that my state of New York ISO is its own kind of like Texas, which does.

We may remember a few winters back when they had the meltdown and they don't interconnect with other states. I was also surprised to see California ISO is pretty much in there but within the borders of the state of California. The other ones, you can see how big the ISOs are.

They're going across states, they're huge. Mexico has its own. They're just very massive grids that are being managed with wholesale markets that update every five or ten minutes the price of electricity. So this would be the idea. I'm not suggesting that this would somehow map into a transportation network and again I'm not suggesting that this would take place first in the United States but it does give us some idea for how these things successfully operate for very large grids where you have a market for wholesale power.

So are you gonna say something about retail pricing?
>> John Taylor: Because we had a disaster in California. We got Governor Wilson to deregulate electricity but they only deregulate wholesale, not retail.
>> Rick Geddes: Yeah, I remember that.
>> John Taylor: And we had big problems cost of that.
>> Rick Geddes: Yeah.
>> John Taylor: So are you going to say anything about that?

>> Rick Geddes: No, because I don't want to. I don't want to talk about power too much. But the retail customer in transportation would be fed the prices that are coming out of this market and whether it's wholesale retail. So I don't want to get too much into the electric thing.

>> John Taylor: Or hope there won't be any regulation of those prices that is noneconomic.
>> Rick Geddes: No, no. We're hopeful that the ISO would have the freedom to let those markets clear. So we create road slots, ius utendi. Right. Which are right to use a particular section of road maybe I don't know what interval five, maybe a lane mile over five or ten minutes make the slot, create a slot market that's tradable and this prices the use of the road networks.

Really profound. And you have both spot and forward markets. I can't think of a situation where forward markets will be more valuable than knowing currently what is the future price of the road network going to be. So you know whether you should rush to work or whether you should wait to in your travel decisions to face a lower price.

Yeah.
>> John Cochrane: What do you do when there's a crash?
>> Rick Geddes: When there's a crash I bought my.
>> John Cochrane: Right to travel to San Francisco at 2:32 PM, then at 2:20 PM, boom, someone's crash on.
>> Rick Geddes: Yep. So with all due respect, John, I think we're probably all familiar with a road network that is undervalued because it's not priced.

So I can't tell you the number of times been on the D.C. beltway, for example. And there'll be a fender bender, there's a backup that lasts for miles. You wait, there's rubber necking and you get there and there's, there's like three or four. Whether it's police or whatever.

I'm not talking about a fatal accident here. Right? And the delay, the cost of that delay is tremendous compared to the resources that are spent in clearing the accident off.
>> John Cochrane: Totally with you. I just wanted to know how. I mean one thing you could do is just spike the prices.

Something goes off in your car and says why don't you pull over and not, you know, take the next exit and get off. Or how the prices react.
>> Ellen McGrattan: What Waze does, Waze does exactly that.
>> Rick Geddes: Yeah, but don't do the price. And so get off at this exit.

If you just had this information, if you think time is a price, that's.
>> John Taylor: Sometimes Waze has so many people coming off, you're more congested following ways that experience too.
>> Rick Geddes: Let me quickly-
>> Ellen McGrattan: Again, if it did it right, it would incorporate that.
>> Rick Geddes: Okay, so let me just say that what we're saying is not radical in the sense.

I mean it is radical, but it's an extension of current policies. I have to give a nod to the state of Oregon that a number of years ago in 2015 implemented Arrago, which is a network wide road per use, per mile road charge instead of the state's gas tax.

It's like 2 or 3 cents per mile. But the key is it's network wide. And it was the first Rock road usage charge program in the world from what I am told. The problem is they're flat, but there's about 60, I think the number is correct. Price managed lane corridors in the United States.

If you were to take those two concepts and mash them together, you would have. What we're trying to do, John, I mean-
>> John Cochrane: Gotta be a little worried about. So we would all have some object that reports to the government exactly where we are at all times, charge us appropriately.

And I'm worried about having some object that reports to the government exactly where I am at all times.
>> Rick Geddes: It's already happening. I think it's. I'm sorry. So I kept my slides on equity because of his question. But when I talk to young people, they, they don't seem to care about their privacy any anymore.

So I took those slides out. But there's kind of a cottage industry in that. Well, the prices would be set in, fed into the uso. That would clear the market. They would, then the information would be deleted. I know nobody believes that, but, but it's kind of like your big computer, right?

>> John Cochrane: Your vision is, you know, you carry some object exactly where your car is at all times.
>> Rick Geddes: So anyway, I'm not going to get bogged down in the uso. This is how the, the integrated, if it were just the, the independent system operator dealing directly with you as a user, that's the simple integrated market model.

That's probably not gonna be the model that we observe. Service providers, SPs, would come in and they would integrate. They would work with the end user off, you say, a monthly plan for road price usage. And they would deal with the independent system operator. They might have an interest in hedging.

They could be big rogue users that might do this like United Parcel Service, FedEx, Amazon, DHL, United States Postal Service and so forth. And they would operate like this. So you, the user would not have the complexity of dealing directly with the independent system operator. You'd be dealing with with the service provider.

I don't think I missed anything. So the key of course here is we would break forward markets. That not only reveals prices, but the market's expectation of where the prices are going, but also mitigates risk over time. So this is Peter's interpretation, John, of how this would look on your app.

Notice that this is a month. So this would be, Peter is, when he's at University of Maryland, suppose he has to get the Tyson's Corner from Maryland for a meeting at 9am and he wants to buy a road use trip. And so he's buying the month here. And you can see that the forward prices are telling you if you, if you go later, you wait 20 minutes, how much you're going to save, if you leave earlier, how much you're going to save in brackets.

So that's the key price signal that is getting to users. And then as we know, people are going to adjust on a whole set of margins as to how they, whether they want to call their buddy who's next door who commutes and say, hey, let's commute together today, whether they want to take their bike, whether they want to wait whether they want to take the train, blah, blah, blah, on down the road.

So this. Obviously identifies you. Peter called it forward transport and this is what it would look like on. On your app.
>> John Cochrane: We're gonna have traders that buy and sell.
>> Rick Geddes: Absolutely. Of course. Yeah. They're going to arbitrage between is about how soon in Chicago. I know the commodities man.

Yeah.
>> John Cochrane: Typically, you know, I would think, you know, if you get up at 4 o'clock in the morning and say I'm gonna buy my 9:00am you know, not everybody else is out there. So the price is going to be low.
>> Rick Geddes: Yep.
>> John Cochrane: So a lot of it is not.

Just about when you drive. It's when you commit to when you're.
>> Rick Geddes: Absolutely true
>> Ellen McGrattan: That game now with airline pricing.
>> John Cochrane: Absolutely.
>> Rick Geddes: I'm sorry, you say that again with.
>> Ellen McGrattan: Pricing airlines, when we decide to book a flight for a trip we know. We'Re going to take in advance, you.

>> John Cochrane: Can'T buy it and sell it so.
>> Rick Geddes: But there be. People are being. John. These. Across these different markets and in the paper, I don't have time but talk about how we think the forward markets would work and how far in advance you. You need to lock in arbitrage.

Right. You absolutely do. This is your current map. Yeah. Nick, could you say a little bit.
>> Nick Gebbia: More about how the initial allocation would be done? Because you have to. It can't be. Everyone has a claim to the same thing.
>> Rick Geddes: Yeah.
>> Nick Gebbia: Some people have the initial rights to that thing and then they solve.

>> Rick Geddes: I would like to, but I have to think about that more because in phishing ITQs and phishing, that was a huge issue. It's a huge issue in radio spectrum. Right. And I don't want to just shoot off Nick without thinking about it some more.
>> Nick Gebbia: Yeah.
>> Rick Geddes: So if you.

It's a weak answer, but everyone has. A right to call them points. Everyone has a right to 100 points.
>> Speaker 8: Or hundred dollars worth of stuff.
>> Rick Geddes: And there's probably a literature but it might be in civil engineering. But let me. Can I get back to you? It's a great point that I need to think about.

>> Harald Uhlig: I think it's an important point. Right. I mean, of course we could start with a concentric circuit. People that live further out, they drive a long time on the road to reach the inner city. So in some ways right now they have more rights to the road than somebody who lives already downtown always uses a bike.

And if you start with an equal distribution, implicitly redistributing from the implicit property rights that people have right now.
>> Ellen McGrattan: But the counterargument to that is people who live closer paid more to live. Closer in terms of Taxes? Well, no, but for the price of the land, for example, itself.

And therefore you're sort of screwing them for the decision.
>> Rick Geddes: When you said that, Harold, I think Kara Kokelman at UT Austin Civil Engineering has written because she's written on credit based congestion price. We haven't said anything about the use of the revenues here. Right. And her idea is to credit it back to people.

And I think she said the initial allocation could be based on like your last year's treasury driving mileage or something like that. I'd have to get back. It's a good, it's a good point. Let's just put a hold on that. So this, this would be the commute today where you have red and yellow and you have a wide variation in travel time is uncertain under current unpriced road space.

And this is what it would look like if. So the point of this is that it's relatively easy. You could rethink. Nothing's easy in life, but you could integrate this into current apps and get to something where the user interface would be relatively simple without a lot of pain.

Okay, notice that this one identifies you its vehicle, MD 012. You're identifying your vehicle. Notice also that it gives you a price, but it gives you the direction of price in the forward markets. So the price is going up. Gives you a hint. But notice that it's blue.

Right. You're paying for the certainty through a price signal. So these are the comparisons of the two situations.
>> Harald Uhlig: It looks only superficially nice, but this reminds me of in Chicago, the prices for parking your car downtown went up massively when Mayor Daley sold the rights to the parking spot for the Olympic game financing.

Never mind the background. But you know, it's now a lot more expensive to park your car downtown Chicago than it used to be, which I think is great because if I go downtown, there's always an empty parking spot. I don't mind paying for it. But obviously people that for whom this really matters now will be loathed to go to downtown.

They probably hate the high prices. So there are people driven off the roads here, right? Then are no longer stuck in traffic and they have to sit at home because they can't afford to take the road.
>> Rick Geddes: Yeah.
>> Harald Uhlig: So that's back to your equity point in many ways.

>> Rick Geddes: Yeah. So let's equity. How much time do I have? I'm going to deal with this in two minutes. So obviously rich and poor benefit from free flow. Market congestion is regressive. Okay, let's you could go back to the New York City again if you want to use that example.

You see buses and et cetera stuck in traffic with very uncertain travel schedules because of the traffic congestion. Traffic congestion is not good to lower income people that rely on public transit. It seems like this is overall pareto improving in that the well all users benefit from improved travel time and the poor have the option.

They also see prices and they have the option of switching to less expensive travel times. So they're gonna benefit from the additional certainty. The other thing about this that's weird is I've had people say well lower income people commute longer to work so this is going to be a regressive tax.

The rich live closer. It's a different world that you're talking about because different roads have different values, right? And we're just not used to thinking about that. It is and it becomes more difficult of an exercise to predict the distributional effects of real time road pricing in a network because the road prices are going to vary with value.

So it could easily be the case that richer people currently travel. Maybe poor people travel farther distances but richer people travel on higher valued roads. So the effect of congestion pricing like this would be progressive in the sense that they, they lower income travel people travel.
>> John Cochrane: Poor people get up at 5 o'clock in the morning, drive to rich people get up at 9:30.

>> Rick Geddes: Good point. So yeah, so they're gonna have have a lower. Yeah I hear that they, they have more shift work so they have to be there. They don't have as much flexibility. So I'm trying to respond to that criticism.
>> John Taylor: Thanks.
>> Rick Geddes: John has become our resident sociologist.

>> John Taylor: The issue that people don't see it that way.
>> Rick Geddes: Okay.
>> John Taylor: People see the equity problem in a different way in the political world that you can't impose this on poor people. That becomes the issue that defeats this politically. It's not that equity can't be thought you can't, you can't win.

You know it is very difficult to. Get a consensus unless you can give sort of a break to those like we do with utilities. So those people who don't have. Much money can purchase their, their electrical. Rich nimby's whine about the poor people. The actual poor people have often very.

Different opinions that that the rich poor people support this and surveys or they support this but they support dynamic lanes.
>> Rick Geddes: Okay. Anyway I'll just point you to finally paper from presented at the aeas. Joe and our group pointed me to this by Cook and Lee study the district.

It's not network wide totally, but it's dynamically priced highway toll lanes and they find that it's actually progressive. Right relative to world number of highly wins are all free status quo tolling, which is dynamic tolling increases aggregate welfare benefits drivers in all income quartiles, driven in large part by the option value.

Moreover, we find the drivers in the bottom income quartile gain the most under status quo, which is dynamic tolling. Anyway, I'll just conclude by saying that lower income groups are also going to benefit from this market information and facing the trade offs and simplification of decision making. All right.

Anyway, hopefully that addresses.