Cliff Winston on Transportation
Oct 14 2013

Cliff Winston of the Brookings Institution talks with EconTalk host Russ Roberts about his recent article in the Journal of Economic Literature on the U.S. transportation system. Winston argues that the while the United States has a very good transportation system overall, it is extremely expensive and poorly organized. What is needed, Winston argues, is not more money, but to spend the money already allocated more wisely. He discusses the evolution of the U.S. transportation system, government's role in transportation, dramatic innovations that might transform aviation and driving, and the potential for privatizing airports and roads.

Clifford Winston on Market Failure and Government Failure
Clifford Winston of the Brookings Institution talks about the ideas in his book, Market Failure vs. Government Failure, with EconTalk host Russ Roberts. Winston summarizes a large literature on antitrust, safety regulation and environmental regulation. He finds that government regulation...
Steven Teles on Kludgeocracy
Steven Teles of Johns Hopkins University talks with EconTalk host Russ Roberts about kludgeocracy, a term Teles coined in a National Affairs article to describe what Teles sees as the complex and unproductive state of political governance in the United...
Explore audio transcript, further reading that will help you delve deeper into this week’s episode, and vigorous conversations in the form of our comments section below.


Richard W. Fulmer
Oct 14 2013 at 11:31am

Excellent podcast, as usual. Texans are getting ready to vote on a bond issue to raise money to improve the public water supply system. I’d love to hear a similar podcast on water privatization.

Oct 14 2013 at 11:37am

Traffic lights and stop signs have never been safety devices. They’re always traffic flow efficiency devices – it you’re in the favored direction, you don’t have to slow to check the intersection.

Thinking of them as safety devices slows change.

If anything, they reduce safety by offering a chance for failure that ruins your assumption, like if the other guy’s red isn’t working when your green still is, or his stop sign has fallen over : places where you normally would have slowed to check the intersection.

The exception is a light at a busy road and an unbusy crossing street, where a chances to cross are scarce and the crossing driver finally starts taking risks to get across. A light saves him that.

Mads Llindstrøm
Oct 14 2013 at 1:39pm

I am wondering if self driving cars will change the desire to own our own cars. Image this:

You are about to get to work, but instead of getting into your personal vehicle you grab your smartphone. You bring up the taxi service app and press “To Work”. The taxi-app then contact all local taxi services, taking into account your preferences for comfort, price, … the taxi services automatically gives you their best offer. The app then chooses the best deal. The app then tells you when you can expect the taxi cap to arrive.

The advantages of using “robo-taxies” in stead of owning your own car would be:

  • No need for finding a parking place when you arrive at your destination. Simply get off and let the robo-taxi go off to its next assignment. This will also save companies and cities billions of dollars in not having to build parking places.
  • Choosing a car for your current needs. When you go to work you may opt for an economical small one person car, whereas when you take your wife to dinner you may opt for a more luxurious vehicle. When you take your family to a ball game you would choose the SUV.
  • Sharing of capital costs. If you only drive 10.000 miles per year, owning a big luxurious Mercedes Benz will be very expensive per mile. However, when you share a Mercedes Benz with other robo-taxi customers, the cost per mile would be manageable. Cars that are expensive to buy but cheap to drive in, such as some electrical cars, may start to make sense when you can share the car with many other customers. Self-driving cars themselves may initially be so expensive, that the vast majority of people will only be able afford them by sharing though a robo-taxi service.
  • People looking for big cost saving could share a single car. A car could pick up three different persons at three different locations. This would of cause make your commute longer, but may be beneficial to poor people, environmentalists or simply people who at a given day have plenty of time.

I don’t think that it would make sense for all people to use a robo-taxi service, but it may be for a majority.

Google is rumoured to be considering a robo-taxi service.

Shawn Barnhart
Oct 14 2013 at 4:07pm

With regard to airports, one thing not discussed was the relationship between government-monopoly airports and the cost of flying. It was alluded that gate restrictions limits competition, but the implication was that this was a result of inflexibility on the part of government-run airports.

As merely an observer, this seems to be more a function of regulatory capture. Airlines pay big gate fees to airports and these funds are a major source of revenue for the airport, resulting in the airport catering to the airlines as their principal customer (since fliers don’t directly pay to use the airport).

It strikes me that this is the major driving force in airport organization and behavior, not control by the government, and it’s difficult to see this changing in a positive way regardless of whether the airport is publicly or privately held.

Furthermore, it’s not hard to see where airport management may actually argue that, in an era of airline mergers and declining flights, that by giving a specific airline, especially one using an airport as a hub, a sweetheart deal they are in fact promoting the public good by ensuring the airline keeps as many flights (and jobs!) in the community even if the long-term result is increased cost to the public and less flight competition, especially when a second-tier city may be at risk of losing a hub (look at St. Louis without TWA).

I also think that airport operations for airlines are far more physical capital intensive than merely landing and taking off an airplane. Computer systems are proprietary, ground equipment is specialized and maintenance facilities are usually carrier-specific as well. This would also seem to limit the value of competition between airports, at least for airlines. Geographic and environmental limitations also hamper the idea of competing commercial airports. A competing airport may have potential customers, but environmental or geographic limits inhibit where you can put it and travel time is a major factor in airport selection for flyers. Nobody will drive an additional 100 miles to potentially save $50 on an airfare.

Oct 14 2013 at 4:46pm

an illustration of Russ’ description of the driverless car idea

Oct 14 2013 at 8:30pm

As a Civil Engineer I cringe at the Engineering Mentality being defined as spending more money, because it’s true and becoming more and more prevalent every year.

When engineer’s do illustrate some economic anlaysis, they add non-monetary benefits like environmental benefits, in order to select a more costly solution over another.

Billions of Federal and State tax dollars are spent on county and city street improvements withouth available funds to operate and maintain the improvements. When local residents point that out, the public officer and the engineer’s state that if we don’t spend the money by a certain time, it’ll be given to another City.

Drive around any metro area and you will find highways with various road standards. The older highways are being upgraded to new standards as a requirement for Federal money. That was probably a good requirement for roads that were built pre WWII, but having a slower speed and better driver, would allow more effeictive spending of taxpayer money. The safety features in cars far outwieght the benefits of upgrades to highways.

Imgaine how we can stretch public expenditures, if we built better drivers.

Again, I cringe because it’s true. We need more true cost-beneift analysis. The last true CB analysis I was n the 80’s where a group of professors and other experts recommended spending increased on bus infrastructure versus light rail. Their recommendations were placed in the recycling bin. The fix was in and the subsidyies have been increasing ever since.

Jason Scheppers
Oct 14 2013 at 11:06pm

Well, you definitely made us civil engineers happy. Dr. Winston help make a great podcast and his paper was a very good summary of potential ideas, with good reasons why many should be implemented.

The best part of the interview was when Dr. Roberts challenged Dr. Winston regarding why people don’t embrace Congestion Pricing. Dr. Winston replied with one the most important facts that it is the heterogeneity of people’s value of time and even people’s variation in value of time from trip to trip that provides an opening for congestion pricing. A superb and important reply.

While I am complete agreement on the heterogeneity, I propose the following challenge:

Traffic congestion in aggregate is not an externality because the congestion an average driver causes is equal the congestion the same average driver experiences. It is the variation in value of time that makes congestion a positive externality to some and a negative externality to others. I challenge the concept that we can charge people with low value of time for being in the way, but instead must use Coase’s logic and turn the proposition that the high value of time people need to pay a euvoluntary price to the lower value of time folks to allow them to pass.

Democratically, the median value of time (VOT) is far lower than the mean VOT, supporting Dr. Roberts point that congestion pricing is not supported by the typical driver. I would add that another of Coase’s tenents was that is was about a transaction or bargin. Setting the price correctly rarely creates fairness unless the payment is made to the aggrieved. Congestion pricing as currently proposed does not seem to address the need for a balanced transaction.

I thought Dr. Winston made excellent points about not needing trillions upon trillions of addition infrastructure spending, but still makes reference to crumbling infrastructure. One can note that fatalities per vehicle mile driven are at material lower rates than 20 years ago (as was discussed in the podcast). Polk indicates that the average of a car has never been older, and despite the hard work of advocates at TTI to make us believe traffic congestion is terrible, The TTI travel time index is the same in 2011 as it was in 1998.

We are living in the good new days, where more access is provided by transport and by the new transport of the internet. Opportunities abound, but I believe it will not be the government that tries privatization but privatization that sneaks in before the government can regulate it that will make the biggest difference.

Oct 15 2013 at 8:14am

Very enjoyable podcast because it illustrated in various ways “the economic way of thinking”. Costs and prices matter. Who among the academic advocates of “infrastructure spending” take into account the lost time and aggravation of road congestion thanks to “Your Federal Road Dollars at Work”? Maybe it sounds better if you can walk to your tenured campus job.

Winston spoke about the recent advances in microeconomic work involving heterogeneity. The comments were provocative but how about an Econtalk episode on that, or more generally on advances in microeconomics. Macro has been leading us into confusion for almost 80 years now. Can Micro show us a way out?

This guy makes so much sense I’m going to give his last Econtalk visit a re-listen.

Morgan Dubiel
Oct 15 2013 at 12:08pm

Why sell an airport to a single owner? Why not treat airports like we would condominium buildings?

This way you’d be able to sell the gate to a new owner and treat the public portion of the airport as you would the public portion of a condominium unit.

The costs would be divided perhaps by passenger throughput or passenger miles or some other usage metric. Even the individual restaurants/sales shops/sales cars could be sold off as condominium units where you’d own the underlying area.

This way no municipality would have an overbearing influence or control over how an airport is managed. You’d see laws about cabs v. chauffeurs eliminated. Costs would come down and political employment would be ended.

Keep in mind the goal of America is preserving individual liberty. Torts or alleged torts (like noise pollution) caused by airports would simply be adjudicated.

Oct 15 2013 at 9:45pm

Heard that the Hong Kong metro is paid for by the real estate development along the stations. The real estate development is done by the Metro itself. Have to wonder what business the metro people are really in.

This suggests that the businesses in some dense places in US, like New York, should take over the subway and pay for part of the subway as an overhead cost, much like the costs of their headquarters building, since almost all their workers get to work using the subway.

Oct 16 2013 at 1:10pm

Interesting discussion. I agree with Shawn Barnhart’s comment regarding the gate issue being more of a contest of who the customer is (airlines vs. passengers) than necessarily being about regulation. This thinking makes a lot of the public / private assumption behind this discussion questionable, in my opinion.

I also had the same thought as Jason Scheppers regarding the mean vs. median VOT, although I do think that congestion pricing is an interesting proposition, especially considering the possibility that a reduction of a relatively small percentage of cars during congested periods could result in significant benefits for the remaining drivers in time savings. If this was offset by a corresponding reduction in the gas tax, it would provide a benefit for those with the “median” lower VOT, offsetting their payment penalty somewhat. The public choice issue rises here with actually reducing one tax when the other is raised, but it’s a possibility to make it politically feasible, perhaps.

One topic I hoped was going to be discussed was the relative change that transportation has undergone. It used to be that a large separator between markets was the extremely high difficulty, and cost, of transportation, making things have widely variable availability from one place to another. Now, it seems that the transportation problem is transforming to be that of such easy and otherwise cheap transportation, that the congestion of goods and services is one of the major problems in the developed world. Really, developed transportation might be one key distinguishing factor in determining what places are considered part of the “developed world.” Perhaps a topic for future discussion?

Ari Tai
Oct 16 2013 at 7:50pm

I’ve done business in the U.S. and Europe. I can ship a ton of non-perishables from most anywhere on the west coast to most anywhere on the east coast for much less than shipping the same goods a fraction of the distance from Paris to Nice.

The U.S. has an amazing (local)truck-plus-(long-haul)train system, and in Europe it’s all on trucks. Why is this? (valuing people over freight?) Yet we seem to envy Europe and its train system. I wonder how much of Europe’s 50%+ markup of popular consumer goods is just the higher cost of shipping?

David Nehme
Oct 17 2013 at 10:45pm

If you put a tax on commuting, you are going to make drivers worse off.

A properly implemented congestion pricing scheme would make drivers much better off. For one thing, the capacity of the roads vs number of cars on the road is not only non-linear it’s not even monotonic. If you take 10% of the cars off the beltway between 5pm and 6pm, more cars will get where they are going between 5 and 6pm. It’s like the exit door of a theater with everybody trying to leave at once. A fee of $5, could push 10% of the drivers to wait until after rush hour to leave (and with a shorter rush-hour, they might not arrive home much later than they do today), or nudge them into car-pooling or buses.
If the peak pricing tools went back into the road, then the driver would benefit as well. The money could be used not only for “engineering” solutions like eliminating bottlenecks, but on more immediate things like queuing up police and tow trucks to quickly clear accidents.

Ed Hatch
Oct 18 2013 at 9:25am

Excellent podcast. Cliff Winston rightly pointed out that the U.S. freight rail system is the envy of the world – cheaper, safer, more productive, etc. Not coincidentally, it’s also almost entirely private – private companies own, build, and maintain it, something a shockingly high number of Americans don’t seem to know.

Glenn Mercer
Oct 18 2013 at 11:46am

Great podcast! I agree with 95% of the ideas. Have downloaded paper, will read. A few comments:

1. Driverless car. ENORMOUS efficiency benefits. I would stress those. Safety? Less clear. Last year Americans drove 3 trillion miles, had 35,000 fatalities. That is one for 85,000,000 miles. Any fatality is bad of course, but are we sure the autonomous vehicle system can beat 1 in 85,000,000? I can’t say I trust even Google that much.

2. Parking. There was talk about it being wrong to tax gas (to maintain roads) when we want to tax usage, really. Agreed. Let’s go a step further: cars are idle roughly 95% of their lives – parked. It would seem that parking fees might therefore be an even bigger lever for transport pricing than road tolls. Don’t get me wrong, I support congestion pricing, but I follow D Shoup’s view that the urban congestion problem is as much due to having provided underpriced parking as underpriced roads.

Oct 18 2013 at 4:28pm

It befuddles me how the speaker can recognize the severe under-pricing of roads and think that this does not impact society’s bias towards cars at the expense of public transit. Dollar for dollar, public transit is a far better investment, and would be fully utilized if not for massive government subsidies of roads, parking, and other driving infrastructure.

Jason Scheppers
Oct 19 2013 at 1:21am

In response to David Nehme:

To take 10% people off of the road requires a substantial charge, and the equilibrium has to be reached. If a $5.00 charge in a static case removes 10% of the traffic but provides a $7.00 benefit in time savings to those that stay. Not all the traffic would leave because of the gains in value of time. The price of the congestion charge must on average be higher than the average savings in value of time to make any significant difference, meaning many people are worse off.

I agree that spending on many critical aspects of congestion could improve the flow, but the politics of that are tricky. Elected officials are not always happy when a $50K solution eliminates the $5 million project that her contributors were hoping to build.

In response to William:

I understand that private vehicles are now in the game of politics and the percentage of user fee payment for road building and maintenance projects is now at about 65%. But remember that 90% of private vehicle costs are private operating costs paid exclusively by the owner. Meaning they are paying close to 90% of the costs including a 5% cost for environmental externalities.

Transit user fees (fares and ads) pays on average 20-25% of the operating costs when you take into account depreciation of capital. It takes more time to use transit to commute which to most commuters is a cost.

My understanding is that per user dollar paid, the transit subsidy is 10 times the private vehicle subsidy. Also, consider the environmental cost of transit. Transit has a higher negative environmental externality per user fee collected when compare to private vehicles.

Many aspects of transit are beautiful things like NYC or even the LA bus network, but significant expansion of transit since they got a piece of the gas tax has not yielded more market share.

Why do you believe dollar for dollar transit is a far better investment?

In response to Glenn Mercer:

As for Don Shoup’s parking work. I was surprised he was not directly referenced in the podcast. He has done amazing work not only coming up with a theoretical parking price, but getting it implemented in many jurisdictions with happy consumers, businesses and governments. That is what I call good work. I am glad to join you and Dr. Shoup on pricing parking efficiently.

Don Rudolph
Oct 19 2013 at 1:12pm

In this and other podcasts it might be interesting to ask the question, What entrenched interest might take up this cause so these improvements can happen. I believe change usually takes place when large economic interests collide and rules are rewritten.
The wisdom of economists only then come to light for purposes of debate.

Oct 19 2013 at 5:44pm

a TEDTALKS with Janette Sadik-Khan, Transportation Commissioner of New York City.

New York City Transit advertises Bus Operator jobs with top pay rate of $29/hr. Really nice pay. Wonder how long that can last.

Oct 20 2013 at 9:17am

Here in NYC, we can’t wait to rid ourselves of Janette Sadik-Khan. She is a perfect example of everything wrong with this city and our mayor. The mayor could have picked a transportation commissioner who was knowledgeable in the field of civil engineering, automobiles, or commuter behavior. Instead, he foists upon us a poli-sci major/lawyer, crusading bicycle activist who also happens to be the daughter of a big campaign contributor. Her reign has been terrible. She is actually making traffic congestion worse here. A great example is Brooklyn, where all of the N/S multi-lane roads (which during commuting hours were already packed to capacity)are being reduced to 1 lane each way so as to install wide concrete medians down the center. Then there is Queens. Here in the Rockaways we have a road, Shorefront Parkway, that (prior to Hurricane Sandy) was 2 lanes in each direction with a nice grassy median and a 40mph speed limit. This was ideal as it allows those of us who live here to easily pass the slower moving traffic of tourists gawking at our beach. It has now been reduced to 1 lane in each direction with 4′ tall concrete barriers running down what was formerly the dividing line between the 2 Eastbound lanes so as to turn 1 lane into a bicycle and pedestrian walkway because the already extant sidewalk is apparently not hip enough for Hizzoner (we used to have a 5 mile long boardwalk which also served this purpose but the city is dragging its feet on post-Sandy repairs so here it is a year after the storm and we still have no boardwalk but we now have lots of new traffic congestion). Lord save us from the technocrats and those who fall for the fallacy of Top Men.

William F.
Oct 20 2013 at 11:19am

@Jason Scheppers

I am curious about the sources for you numbers because I am currently trying to better understand and quantify the subsidization of automobiles and the automobile lifestyle.

There are many reasons why I think that public transit is a better value than the automobile, but I will limit my response to the economic reasons.

First of all, I think you underestimate the subsidization of automobiles and the environment that facilitates, or worse, requires their use. For one, I don’t believe that 65% figure covers the full cost of road building and the payment of debt necessary to finance those roads. Perhaps you are familiar with Chuck Marohn’s work with Strong Towns. In one very compelling series, he illustrates how the cost of suburban road construction is primarily financed through massive subsidies from the state and federal governments, with the pretense that the investment in this infrastructure will pay for itself through the property taxes collected from those who live or work along the street. He finds though, that in the suburban context, this is never actually the case. When the life-cycle of the infrastructure is over, the cost of reparation is covered by government money infused expansion of the suburbs, which brings a whole new set of debt obligations.

Secondly, our land-development patterns have largely tilted the balance away from public transit and towards the private automobile. It would be one thing if these land-development were purely free market forces at work, however they two had government policies thumbing the scales. For instance, as you’ve already noted with Shoup’s work, the requirements for free, unlimited parking have warped our landscape and spread our cities thin, further discouraging walking and transit. This pattern encourages more driving, which in turn encourages more parking, and so continues the vicious circle. If drivers were required to pay the full cost of parking, I imagine our landscape would look drastically different, favoring the far more efficient systems of public transportation and walkable neighborhoods. Furthermore, government zoning which separates residential from commercial activities further incentivizes car ownership and decreases the feasibility of transit operating effectively. Much more can be said on this, but the point is government requirements have forced our entire built environment into a configuration that heavily favors the automobile and makes operating transit much more difficult than it would be had these places developed more organically.

Thirdly, the economic feasibility of driving everywhere is largely predicated on the idea of cheap oil. I think that in the U.S., this cost is kept artificially low by government action. I don’t know as much about this topic, though I expect to find more information about it as I continue in my research.

I’ll conclude by saying that although transit may appear to be less economically efficient than driving when looking at a snapshot of our current system, looking at the system as a whole reveals a number of examples where the government has tilted development to favor automobile use.

Jason Scheppers
Oct 20 2013 at 3:31pm

@ William F

Thank you for your respectful and on point reply. My source for the 65% was Dr. Winston’s paper “On the Performance of the US Transportation System: Caution Ahead” page 780. Dr. Winston States “total user fee revenue accounted for only 65 percent of all funds for highways—down from 84 percent ten years earlier—with nonuser fees and bonds providing the rest of the funding.” Most gas taxes are at the $.40 to $.50 per gallon which given a 20 to 25 mpg vehicle the user fees collected per VMT $0.02 per mile. Therefore per my assumptions $0.02 is not paid for by user fees.

As for vehicle operating costs, there is AAA high estimate, but I prefer Steve Polzin from the University of South Florida esitmate. He comes up with a $0.36 per vehicle mile cost, which does account for insurance and other pecuniary costs. His Plantizen blog entry “The true cost of driving and Travel behavior” made on April 30, 2012 details his methods and sources. I presume that these costs are all personal expenditures.

So if we combine the infrastructure costs we get total user expenditures of $0.36. In addition $0.02 of bond or tax money is spent on roadways. I also grant that there are pollution externalities and I assign a value of $0.02 per VMT, based on Ian WH Parry’s March 2006 Discussion Paper: “On the Costs of Policies to Reduce Greenhouse Gases from Passenger Vehicles” prepared for Resources for the Future.

Using the sources and assumptions above you get $0.36 of $0.40 spent by the user for 90% roadway user costs being covered by roadway users.

The transit numbers are from the American Public Transit Association annual report that details $55 billion in capital and operating expenses and $13 billion in operating revenue.

I am very familiar with Chuck Mahron’s work. We agree in principle on many issues, but my solution is where we identify subsidized development it should be addressed by eliminating the subsidy in any form of transportation.

Development regulations do favor auto oriented development, but what came first auto-oriented codes or autos that drove auto-oriented codes? Free parking is not required in most codes, just a large amount of parking to protect every development’s neighbor so no one parks in the others property. It is perfectly legal to charge for parking. It is also legal to use the extra space for other purposes when it is not being used. I believe that the minimum parking requirements are inefficient, but they are responding to a political need.

I believe that energy prices are low in the US because we have no coherent overall energy policy and markets are responding to the demand. I concede I have no one specific source for this belief.

Let me end by agreeing with your concluding statement. I think we differ in the degree that we believe this is material. Thank you for your thoughtful exchange. I hope that I have provided some information that is useful to you, and I will be sure to check the comments to see if you add the final comment to our exchange.

William F.
Oct 20 2013 at 5:09pm

@Jason Scheppers

Thanks for your thorough reply. I will hopefully get a chance to take a closer look at the sources you’ve provided. The one thing I will say though is while you are correct in that the government does not mandate free parking per say, it *does* mandate a vast oversupply, to the point where charging for it would be like charging for air. By mandating an oversupply, they are effectively enforcing a price at or close to zero by way of classical supply and demand.

Furthermore, they do not usually allow land uses with parking demands that peak at different parts of the day to combine their requirements in order to meet the minimum requirements. I think these practices and the many others you have surely seen in Shoup’s work are present in the overwhelming majority of American towns and cities, and have drastically transformed our road networks and built environment for the benefit of convenient automobile use.

Nice chatting with you, and have a great day.

Dr. Duru
Oct 24 2013 at 9:45pm

The benefit of this podcast to me was diminished by some of the surprisingly broad generalizations made by Winston. I am hoping that in reading the paper I will find a more careful and measured assessment of transportation issues in this country that is not completely blindsided by the strong anti-government bias. I mean, really? America has done nothing since roadways were first built to improve how they function and how they are built? I am about to drive home on a highway that features congestion pricing. Such roadways are also in Atlanta. My neighborhood is full of “smart” traffic lights. I have used advanced parking meters in SF and Berkeley. Traffic cameras are sprouting like weeds in order to reduce the number of drivers who zip through red lights.

I could go on and on and list a bunch of things that seemed very wrong in this podcast given my own experiences traveling and driving in many of the large metro areas of the country, but the one big glaring omission in this podcast was any mention of the traffic planning and innovation done in Los Angeles. NPR did a GREAT story in April on LA’s ambitious project to synchronize lights throughout the city in order to reduce congestion. No mention of whether the innovations originated from the public or the private sector, but I do not think it matters much as long as the provider of the system was paid a market price for its services and products (and I am surprised Professor Roberts did not challenge Winston more on the broad premise that ALL transportation innovations come from the private sector and the public sector is COMPLETELY incapable of doing anything well).

I think less of a focus on anti-government rhetoric would have allowed such an example to shine through in this discussion. For those interested, you can check out the story here:

I am also surprised Professor Roberts did not press more on what exactly a completely privatized roadway system would really look like and how it would work: from large metro areas to sparsely populated rural towns.

Oct 25 2013 at 5:34am

Recent article about the Hong Kong MTR which should interest folks here.

Oct 28 2013 at 8:05pm

First off, thanks to all involved for EconTalk. I think I have listened to all of them.

I noticed a pattern very much on display. Russ and guests frequently pooh-pooh all proof or logical extrapolation of positive consequence to any policy, regulation, rule, law, or approach violating their version of free-market libertarian purity. All consequence, that is, EXCEPT UNINTENDED! EconTalk is simply chock-a-block with discussions of unintended consequences with an unquestioning conviction of causality withheld from even the most statistically studied aspects of INTENDED consequence.

I am not puzzled as to why. Unintended consequence with its attendant causal relationship to negative effects help bolster the Hayeckian underpinnings of most, though not all, EconTalk podcasts. Intended consequence, even backed by, admittedly weak, studies and statistics, encourages centralized (rather than stochastic…thanks Taleb) economic tinkering.

Effective centralized decision-making violates the moderator’s foundational belief…therefor offered results must be discounted, ignored, ridiculed, and/or tossed in the waste bin of multivariate inscrutability.

But Mr. (Dr.?) Roberts rarely meets an UNINTENDED consequence he cannot summarily adhered to a disliked decision, regulation, law, program, or tax. Winston is far the worse for this…making broad statements about road availability, the growth of technologies, energy supplies and the like without a single shred of evidence linking the consequence unintended to the goal intended…other than of the correlative sort. And we all know what Russ thinks of using correlation as an indicator of causation;)

Incidentally, as a tax-paying funder of the vast national framework that feeds into Mr. Winston’s private infrastructure utopia I would like to discuss

1) Repayment for his reliance and attachment to my original outlay. His roads are going to be attached to my highways are they not?

2) Payment for support expenses beyond base costs (fire, police, armed forces to keep the shores safe so his roads are not used to transport foreign invaders to my front door, etc.)

3) A willingness to sign binding pledges to NEVER resort to eminent domain law to acquire necessary land and access. This one alone should pretty much end any discussion of private road ownership…at least in any of the population dense areas attractive to private road developers.

4) A legally binding contractual pledge to abstain from rent-seeking behavior of all kinds…lobbying being first and foremost.

Of course, such a prohibition could never occur. Virtually, the only way private roads (to name one example) can come about is through effective purchasing through legal lobbying means of the local and state legislatures responsible for making infrastructure decisions and allocations.

I could go on but my time and your patience are not without limit.

To recap: If you cannot trust purported consequences of “altruistic” acts then I am equally unwilling to believe in the causal link with the unintended variety…or…In the battle of the unprovable…all causal links are religion.

Oct 29 2013 at 8:46am

Privatizing large scale activities or needs, like transportation and health care, pose as many problems as total government regulation of either system.

The examples both speakers used concerning airports largely centered around large metro areas. While they lamented that smaller metro facilities were underused to the point of being vacated, my concern for total privatization is loss of access.

For years Delta held a huge advantage at the Covington Field airport. Since 9/11 and its merger with Northwest it has severely reduced its operations at the airport in favor of larger metro areas where assets (gates/buildings/etc) already exist. As a business traveler my business has been significantly impacted. No new carriers have entered the market here, in part due to the attractiveness of larger population centers.

The idea that some market player will want to serve marginal areas as easily as they will high volume areas, when capital outlays are huge, doesn’t play out historically.

Trent Whitney
Oct 29 2013 at 10:41am

This was a very interesting podcast that still has me thinking about 2+ weeks after listening to it.

While I’m sympathetic to Mr. Winston’s idea of privatizing roads, I have trouble seeing how that will play out in reality. As toll roads have been built, new bureaucracies have been created – Dallas North Tollway Administration in Dallas, IDOT (Illinois Department of Transportation) in Chicagoland, etc. They continue to expand, building large, shiny office complexes and hiring more and more people to enable them to do their ‘complicated’ task of collecting tolls. How, in practical terms, can these be dismantled?

In Chicagoland, the Chicago Skyway was “privatized” a few years ago – I use quotes because I think the deal wasn’t really privatization, but the local government selling the rights to collect tolls for X years in return for a lump sum today….which, of course, they’ve already spent and you hear politicians today railing against the deal, claiming they should get more money out of it.

With the news stories resurfacing about the potential for a black box that records our car’s annual mileage for which we’ll be taxed, I fear we’re soon going to be triple-taxed: Gasoline, tolls, and mileage taxes. While I’m again sympathetic to Mr. Winston’s argument that driving/parking are underpriced today, I see the potential for this to swing too far in the other direction rather quickly. Tax the activity once, and make the tax as transparent as possible.

And now that I’ve listened to the 10/28 podcast about Prof. Coase’s enormous contribution to economics, that’s caused me to rethink this…what if property rights to roads were alloted/sold? I know I’m falling victim to the problem of the unseen because I’m wary of how it would all play out…but if it were allowed to play out in an open market, I know the end result would be beneficial to drivers.

Thanks for this podcast in particular – rich in ideas…great discussion…one that I know I’m going to end up listening to again.

Comments are closed.


About this week's guest:

About ideas and people mentioned in this podcast:Articles:

Web Pages:

Podcasts, Videos, and Blogs:



Podcast Episode Highlights
0:33Intro. [Recording date: October 3, 2013.] Russ: We're going to talk today about an article you've written for the Journal of Economic Literature on the transportation system in the United States. And we're going to put a link up to the paper itself and a video that Brookings has put together summarizing some of the points. Transportation is a pretty big part of our economy, right? Guest: Yeah, a lot bigger than people realize. If you think about it, it should be a lot. But no one typically sort of thinks of the annual expenditures on transportation. But, when you sum these things up in terms of what consumers spend in terms of getting to work, various pleasure trips and non-work trips, and then you look at what shippers spend on shipping freight and so on and so forth, and then what the government spends building infrastructure, we're talking shares of GDP (Gross Domestic Product) that approach the amount that we are spending on health care. But people obviously don't seem to pay as much transportation as they do about health. And that's just out of pocket expenditures. It doesn't include the time expenditures, which it turns out are about as large as the monetary expenditures. Russ: So, you look at a whole range of policy issues related to transportation, including some of the history. I want to start with the history. How has the role of government in the transportation sector evolved in the history of the United States? Because it would surprise me--I probably knew this but I'd forgotten--how recent some of the government involvement has been. Of course some of it goes back a long way. So, give us a quick thumbnail sketch of the sector and government's role. Guest: Sure. Actually, the interesting thing about transportation is how, although certainly pervasive government involvement now, how virtually all components of transportation, whether they be the modes or the infrastructure were initially started by the private sector, not the government. That is, the first roads in America actually were built by private toll or road companies. The first airports in America were private airport companies. Even air traffic control, which you think has got to be a government function, was actually set up by a private consortium so to speak. So, basically, the system started out in its infancy if you will with the private sector. And gradually government got involved in roads, for example, regulating the roads. And then we get a familiar story. As any part of the system experienced financial distress, and that's certainly was going to happen, then government tended to come in. So that was sort of it's entrée into the road system. Certainly by the early 1900s. And then other things happened. During the Depression we had airports that were private but that had serious financial problems; again, government came in and wound up taking them over; subsequently set up an air traffic control system. In the 1950s, by the time it was the 1950s, there were public transit companies--as the auto came in, it was competing with transit and causing it to have financial distress, and the then-private companies collapsed. Again, there were regulatory issues that may have contributed to this. And again government took over the mode. So you have sort of a familiar pattern that things started out private. There tended to be some sort of crisis. People debate to this day whether the private entities could have pulled through with government assistance. In any case, that was not forthcoming. Government wound up getting involved and taking over the operations. Those things it didn't take over, mainly on the inner-city side--railroads, airlines, trucks--it stepped in and regulated them. And then we saw the withdrawal of economic regulation in the late 1970s as we generally started moving away from government intervention. Russ: Now, you don't talk about it--I don't think you do--in the paper, but one area that government got involved in, in an unusual way, would be the taxicab market. So in many cities it's highly regulated, but it's still run and owned. It's fascinating to me that we are living in a time now where private competitors are starting to bite into that government-run cartel, or whatever you want to call it--sometimes, it depends on the city obviously, but companies like Uber and others are offering private taxicab--I think. Correct me if I'm wrong. Guest: Some areas of transportation are known [?] a lot more than others. Taxis I just briefly mention because there isn't as much knowledge as we'd like. Generally, the stories vary from city to city. Some cities, taxicabs are largely deregulated. Others, they are highly regulated. And certainly the hallmarks of the highly regulated taxi areas are no new people can get in. And then you hear these stories, like New York City, where taxicab medallions are worth millions of dollars. Now, Uber offers a technology that enables a user with an application to get access to some form of what we would call a taxi service. There's nothing preventing somebody from getting a traditional taxi. But in terms of getting new entry, other forms of car service are also trying to be hooked up on Uber. This is the thing that the incumbent taxis want to fight. They don't want this kind of competition. And we're seeing how this is playing out in certain cities, with this kind of entry experiment through a new technology, and we'll just see what it does. Russ: Yeah; I'm sorry. I think the distinction--I didn't make it clear. Taxis are allowed to cruise the streets and look for customers. Car services, you have to call and make a reservation. What Uber does essentially turns a car service into more of a cruising opportunity for pickup of a random passenger. Guest: Yeah, a real-time type of service that's very responsive. Which obviously is something that you can do with Uber at the same time, compensate for cases when it's really just difficult to get a taxi. And you let somebody know you want service, and somebody watching this kind of thing can provide it.
7:55Russ: Let's start, though, with airports. Airports are kind of mysterious to most of us. We don't spend a lot of time talking about it, and we don't really understand what's going on. Airports typically now, I think almost all major airports in major cities, are run by the city or the county that they are in. But that, you say, wasn't the case. When did that change, roughly, and what were the consequences of that? Guest: Well, roughly it was during the Great Depression. The airports were private as air service began. And during the tremendous drop in air travel--which never was very big at that time, as it was just getting going--airports had problems. And governments mostly came in. For the most part, municipalities have taken them over. And they are the ones that now are providing the infrastructure. And, to a certain extent, it appeared to be a successful way of doing things, in the same way that public highways seemed to be a successful way of doing things. Because it really wasn't congested. The system wasn't really falling apart from excessive use. And so on and so forth. And so we thought, okay, this is a very reasonable way to do things. But as expenses have grown, as delays and congestion has gotten worse, and as airports--as well as highways--are not terribly responsive in dealing with this, we are beginning to see lots of problems with public ownership. Russ: One of the ones that fascinates me is access to the gates. You'd think, well, if I want to offer a flight between Washington, D.C., and Chicago, I'll just go fly my plane, fly my airline to one of those places and land my plane; pick up some people; take them to the other city. And it seems pretty straightforward. But the gates problem is a very severe constraint. Correct? Guest: Correct. One thing that is part of the story is yes, the airports were owned and operated by the municipalities, but, for expansion to the airport terminals and the like, runways and the like, they needed somebody to guarantee the investment that they were going to make. And that came from bond purchases or bond holdings by the airline. So the airlines actually had a very, very important relationship to the public owners of the airports--they were basically funding a fair amount of the capital improvements and enhancements, if you will. But in return, they had say on the types of investments that could be made and effectively were able to block entry to people, who weren't going be able to get gates. The airports certainly weren't going to give them to people and if you wanted gates, in some cases you would have to have a whole new wing built. So, with gates sort of locked up, in certain places that made it very difficult for new airlines to enter. Russ: And they weren't all being used. Right? Guest: They were not all being used. And there were different classifications of gates. There were what we call exclusively used gates where certain airlines had the right to use these things. And then there were things where you did have some flexibility but oftentimes the new airline couldn't get the use of the gate at the preferred time , and these kind of things. So it has become a form of entry barrier. That is, ideally you might want to serve an airport, but if the gate space isn't there for you and you are not going to shell out the money to go out and build additional terminal capacity, you can be out of luck. And of course restrictions on competition are reflected in higher prices. And that's what the evidence has indicated. Russ: How would that work in a private system, though? A private system, say, a consortium of airlines that currently serve a city that own an airport, they are not going to like entry, either. Guest: Well, they are going to have, obviously, a very different financial situation in that they are not going to be beholden on any particular airlines as dating back from when the airports were built, when regulation was set up. Which, by the way, was kind of a convenient situation in restricting entry, because that was done under regulation. So the question about getting gate space never came up during regulation because virtually all their airlines had trouble entering routes anyway. It was only deregulation where the problems started to arise. But in any case, private airports, where in many cities throughout this country competition seems quite feasible, right here, obviously--Reagan National, Baltimore Washington International (BWI)--could compete. The owner is presumably going to try to get entrants who are willing to pay the cost of the services. They are not going to immediately say, You're out of luck, you can't come in. They are going to try to work with the airlines and if they can find ways that they are going to cover their costs, they'll accommodate them. And if demand is growing they'll say, okay, we're going to build new terminals and handle new traffic and we're expecting you to pay for it. But we're not going to call on other airlines to bankroll these investments. We'll do it--because we're going to be able to cover the cost by charging you. So there's obviously going to be a lot more interaction between both the airports and the users of the airport, which is one of the problems we have now. You have customers that use airports but their relationship--the landlords, so to speak--is virtually nonexistent. Russ: I'm just trying to think, imagine, what the owner of a private airport, what kind of entity it might be. So, for example, if I wanted to acquire--I'm thinking back to a city I lived in that had a little more vacant land near it, St. Louis--if I wanted to acquire a large plot of land near the city of St. Louis, maybe in Illinois across the river, and build a private airport and just offer the opportunity for airplanes to fly there, that can't happen right now. Correct? And as a result, if I'm Southwest Airlines and wanted to go into St. Louis, it had to go to the St. Louis airport and build its own terminal. Which is nuts. Unless the other part of the airport is totally full; and I suspect it wasn't. So what is strange about the current world is how difficult entry is and who controls it. Right? Guest: Right. The difficulty--again, let me be clear. We do not need many, if any, new airports really being built. In other words, when I envision privatization, all the existing airports will be sold off. Which I might add will do quite a bit to help a lot of cities' financial situations. They are going to sell off these assets, and infrastructure firms are going to buy them. For budgetary purposes that could be helpful--assuming away what's done with the money. Russ: Good point, but who do you think-- Guest: Most of these things will be sold off, and there will be the existing airports. But then there's also something else, part of this: there are thousands of smaller airports that could be used--for commercial service--but they are not being used now. Russ: Why not? Guest: Basically, these are not airports that are going to get public money. So, to the extent they want to compete they are going to be at a disadvantage. Russ: There's no mechanism for them to use private money right now. Guest: Not right now. Not for that purpose. So, it could be very, very different from what we've got now.
16:45Russ: Let's think creatively for a minute. So, if every city in America either out of financial desperation or economic wisdom decided to auction off its airport to the highest bidder, how would you imagine that situation settling down after a couple of years? Would there be sufficient competition among airports to prevent them from extracting large sums of money out of customers and airlines to have access to them? Would you let them freely buy the airport and do whatever they want with it? Guest: Well, okay. First of all, I would allow for one requirement. And we've actually learned this from an experiment in London, which, by the way, has privatized its airport. So we're behind the curve here. Interestingly the leader of capitalism in the free world tends to be really slow on the uptake in terms of privatization. Many places have actually been exploring privatization of the airports. But in any case, London made a mistake and initially sold of their airports to one infrastructure company, a foreign one. And subsequently realized that was not the right way to do it. And forced that company to effectively sell two of the airports. Russ: What should they have done. Guest: They should have sold them to independent buyers. So, that's the approach we should take in the United States--at least initially say, yes, we are going to sell off airports but we are not going to have the same company buy all three of them, at least in this area, and give them an outright monopoly to start. We would like independent firms to purchase these things. And so then they will go head to head. And like any fundamental institutional change, like deregulation, there's obviously going to be enormous learning. And I can envision, on the one hand we will reveal inefficiencies that we had no idea existed in terms of operations that could be vastly improved, in terms of interactions amongst airlines and airports that say, look, we could provide better service if you let us do this and get rid of crazy things like tarmac rules. At the same time, there will be shocks, as there have been following deregulations. And undoubtedly financial crises where it will be difficult for certain airports that may have much less traffic. I don't want to oversell how smooth a ride privatization will be. If people thought deregulation was tough, privatization is going to be even worse--it's now going to have people who have no experience for the operating of the private sector. But I think when all is said and done, it will be clear this is a much more efficient way to do things. That, airports--who are already competing already, let's not kid ourselves--public sector airports do try to get passengers and do implicitly compete with each other. We're just going to try to make this explicit. And I think the results could be quite positive. Russ: Yeah. They do a little innovation. Here in the D.C. area there are parking garages that tell you how many spaces are available on each floor and which rows even have available spaces. I find that a very pleasant experience in parking life. But they are not so good at minimizing the distance from where you park your car to where you get on the plane. Dulles in particular I find frustrating. It used to be you had to ride a special bus. Now you've just got to ride a special internal metro system that seems to be designed to go up and down a lot of escalators as much as possible. So it would be interesting to see how much more pleasant that would be if someone was actually trying to make more money off of it. Guest: Yeah. Absolutely. And again, start exploring ways in which passengers, who aren't happy, could be made more happy. What is it that you want at your airport? Right now, the public airports just don't have that big an incentive, since they are getting their funds and are protected from competition. At the same time as we are now seeing in a lot of public transportation services as well as other services, they are starting to run deficits, and that's raising problems with trying to improve services and even trying to maintain current operations.
21:35Russ: Let's talk about commuting, some commuting issues. I was interested to see in your paper that, despite the passion for public transportation that has engulfed many American cities--the excitement of light rail, the remaining bus systems that are still in place--commuting by car remains not only the dominant form of commuting, but it's grown steadily over the last 40 years. 1. Why do you think that's happened; and 2. What do you think of the claims that we need more public transportation to reduce dependence on the car? Guest: Well, people always look to Europe, high density areas, and say how great their transit systems are. And what they fail to realize is the dynamic of the urban and suburban area. A lot of these older communities, people who have been there for many hundreds of years, and the density is established. America is still a place on the move. And when I came to D.C., no doubt you were--there wasn't the expansion of Tyson's Corner, and out to suburban Virginia and Maryland and so on and so forth. And now, just over the last 10, 20 years, there's been a huge change in this area. Well, if you have a metro system that stops at Vienna and is not going to be able to cater to those people, they are not going to ride it. What are they going to do? They are going to use cars. Well, you multiply that by many experiences throughout the country and you are talking about the inability of transit to keep up with fundamental changes in demographics in our sort of dynamic metro areas. So, that's a big problem, physical problem, with urban rail. Bus, you'd think, well, they are like cars; they ought to be more flexible. Yes, and they run into problems, one big problem--regulations. Trying to establish a new bus route, or changing an existing one, can be quite difficult, time-consuming, the usual kinds of regulatory problems. And bus just isn't all that good at responding quickly. And again, what's the incentive? They are still getting their subsidy. So, I think what you have with transit--and I've been hinting at this all along--is again, no real fundamental connection between the suppliers of the service and the customers, trying to respond to the changes in the customers' needs and preferences, trying to respond to just exogenous changes in society, what have you. You have a system that is in place; it gets subsidies; things are good; why change? And the public, who don't really know how things could be so much better, seems to go along with it. And then also shares what I would call this engineering mentality. What I mean by engineering mentality, that all the solutions to transportation are thought of in terms of more spending. If we could just put more money into rail and expand its network. If we could just put more money into a new rail system, where we don't have any--all this would solve our problems. But it wouldn't, because it's not dealing with the fundamental problem of being able to respond effectively to consumer preferences. These are very expensive solutions. And of course they are not dealing again with the fundamental problems with autos about mispricing congestion and the like. So the real tragedy in transportation is it seems to be more dominated by engineering than economics, where in fact so many of the solutions really involve just basic improvement in pricing investment.
25:49Russ: So do you think we spend enough on infrastructure? We've had an interesting debate with Robert Frank on this topic, centering around this claim you often hear that we don't spend nearly enough; our infrastructure is crumbling; it's dangerous. How would you respond to that argument, given the point you just made? Guest: Okay. So, this is just a basic economic point that's obviously very simple for economists. But non-economists often don't recognize this. Before you make an investment decision, one, whether you should do it or not or make an overall assessment are we doing enough of it or too little of it, you've got to first ask the question: Is the facility properly priced? Forget about transportation. Just think of any commodity. Talking about oranges: and if oranges were priced at, oh, a nickel, and we are constantly running out of oranges, then the instinct of most people is, well, we need to increase production of oranges and build more orange factories or groves or what have you. Whereas in fact someone else might say, en economist, hopefully: Wait a minute--you know, oranges are greatly underpriced. Russ: Underpriced--when you say underpriced, you mean legally there is some constraint on them. That someone has set the price artificially low. Guest: Yeah. There was an artificially low price; this is not a market price. And you say, look, let's get market pricing here. And prices go up, and all of a sudden you realize supply is equal to demand, and the market is clearing and we really don't need to make any investments in the production of oranges. Well, that's the same idea with roads. Roads have an artificially low price. Cars are not charged for congestion, so they put pressure on peak capacity. Trucks are not charged efficiently for the damage they do to roads; they pay a gas tax when they really need to pay an explicit charge that reflects the damages they do to roads. This underpricing causes road capacity to fill up, causes the roads to wear out a lot sooner. And it generates a demand for: We need more spending and our infrastructure is underfunded. I say, let's get the prices right. Now, the same thing is true for airports, same kind of thing; even ports. Same thing. My guess would be after getting the efficient prices, yeah, there probably is room for some efficient--and I mean efficient that would satisfy cost-benefit criteria--investments in highways in some high density areas. Certainly additional runways in high density airports. But not the trillions of dollars that people talk about and not nearly as much as people are led to believe. So, I think that the claims of the infrastructure crisis are grossly overstated, and what we really have is a pricing crisis. And if we can get the prices right, that will do an awful lot to improve the condition and service of our infrastructure. Russ: I want to come back to that in a second, but I want to make a general observation first. Which is: not only is transportation about the size of the health care industry, it's kind of structured the same way, where the customer doesn't do the real paying and the supplier doesn't receive the money from the customer. And that messes up the incentives. One view would be that it works surprisingly well given how badly designed it is. I get on my airplane; they are incredibly safe. I don't think that's due to government regulation, but they are incredibly safe. Similarly, I get in my car, they are incredibly safe. All these methods by the way are much safer than they were, and they were safer long before government got involved on safety; so the trends are basically unchanged for very long periods of time and it's mainly, I think, driven by our increasing wealth and desire for safety. But my point is that it works pretty well. I agree that the urban congestion thing, not so well. But the rest of it--I think we've built too many roads, we repair them too often--I assume that's because of the political power of contractors. Right? That creates lots of delays and wastes money. And it sucks money out of my pocket into their pocket more than is probably necessary. But you could argue it works pretty well. Guest: The parallel to health care is a nice one, because--yeah, I think most Americans think health care works pretty well. And when you think about it, there is a reason for it. We spend an awful lot of money. It should work well, given how much money we spend on it. But what also is interesting is people really have no idea. Like, most people will think that transit covers its costs. Without even cost being defined. Like, transit breaks even. Russ: Right, like, I get on the metro, I pay money-- Guest: The metro is full, people are on it, I'm in NYC, the subway, there are tons of people on it. No one is going to realize that these transit systems cover none of their capital costs, and generally a modest fraction at best of their operating costs. And the rest of this is coming from the public. Highways--it used to be that the systems were designed to break even and were paying out of a trust fund from a gas tax. So that's the way people pay. They just pay for gasoline; they don't realize that a chunk of this is going to the roads. But now we are running a deficit there. And in a sense it's been a deficit building for a long while, because there is deferred maintenance. So, what we have is we really have a multi-trillion dollar system in terms of the value of the assets. If you actually look at the value of highways, airports, our rail stock, ports, so on and so forth, these are in the trillions of dollars. We've just built this enormous system. So, yes, it should be pretty good given the money we've spent on it in this country. At the same time, I think the tragedy is the counterfactual--people just don't realize how much waste there is-- Russ: And compared to what? Guest: Compared to that we get a more efficient system. And I'm not even getting into yet the technology. So that, forget about also the waste, things could just be infinitely more efficient. Let me just give you one simple thing that I think anybody can relate to: You are driving home late at night, 12, 1 in the morning, and you hit a cross street, and it's got a light. And it's red. And you are waiting and you are waiting, it's green on the other direction, and no one is coming. And obviously people are often tempted to run the thing. Russ: Well, they used to run it, Cliff; the cameras now ruin that common-sense strategy. Guest: Right. But how about this, which technology could do: you have stoplights in these off-peak times that are timed to traffic flows? And if no one's there, boom, it's automatically green. Or, it merely switches to blinking red, so you don't have to wait. This is just a simple thing, but it's just an example, again in terms of operations and improving technology, how it could be done. So I think the bottom line is: We have this enormously expensive and valuable transportation system that's the envy of the world. There's no question it's better than other countries' systems when you look at the thing in toto. But there's an enormous amount of waste, enormous amount of inefficiencies in terms of operations, and for what we spend, we could have something that's even better, and in ways that people find hard to imagine.
34:36Russ: I think a lot of people would disagree with your claim that it's the envy of the world, right? Guest: Well, I'm talking about the whole system. Russ: Europe has better trains. Their airports are nicer. Their roads--the Autobahn--they are in better shape. I agree with you; I think our system works pretty well; I love your point, yeah, at what price. But I don't know if we're doing so well relative to everybody else. Guest: Our rail freight system is the envy of the world. We have a really excellent rail freight system. Other countries go in passenger systems and if you think we're subsidized, wait till you see what theirs is. Extremely, extremely inefficient and very expensive. Our road systems tend to be better. Again, other people tend to have urban rail systems and so on and so forth, but those too, even with the density, tend to be very highly subsidized. Our air service tends to be much better, far more competitive. I mean, look how long it was before we'd get affordable air travel in Europe. So, I'm just saying, across the board, I think overall our system is better. Russ: Let me ask you a question about the efficiency issue. I agree with you that roads are underpriced. They are zero, the money price; so time is a way to ration the limited space in urban areas. And when we expand the roads it just takes them a while to fill up again because people move closer to the city until they find their commute unpleasant and then they move back--they move closer, back out, depending on the, as people adjust. But the question is, if you improve the efficiency of this system--and you could say this about a lot of different parts of the transportation system--who is going to capture those gains? If you put a tax on commuting, you are going to make drivers worse off. It's true they are going to save time. But to get them to save enough time, you can't save them enough time to make up for the money because then the tax is too low. You've got to, by definition, discourage people from driving on the roads. So those gains aren't going to go to drivers. So the political mechanisms for these kind of changes seem to me to be very low. Having said that, here in D.C. and elsewhere we are trying toll roads, for the first time. And I assume--I don't know the politics of it--it has partly to do with the fact that they just like the money. I don't think it's just--and maybe it's from the demand from some people who want to be able to get to work quicker and not spend so much time in their car. How do you respond to that challenge on efficiency grounds? Guest: Okay. So, the traditional argument against a lot of the calls for efficiency improvements has been that, yes, these things would improve efficiency but the distributional effects, and in particular, money going from consumers to the government, is what's driving the benefit. That, I would say, is the conventional view. And I think this is the beauty of economic research and greater understanding of technology--it's pretty much out of date. First of all, the major improvement that we've made in microeconomics research is dealing with heterogeneity. And what that basically means is that people are different. So, whereas before the conventional thinking was about the average user and so on and so forth, now we can analyze in more detail specific users who have different values of time. And also, users whose value of time may differ on different days. So, it may be the case that on any given day someone says, I've really got to get to work and I'm willing to pay more to do it, a lot more, and I want the option to do it. And I'm better off by doing this. Whereas other people would not feel that way. So, it's this heterogeneity and now the ability through technology to actually charge different [?] prices, that has opened people's eyes up to say: Wait a minute; you know, there can be sort of a better matching here, where yes, the prices are pretty high but people really have a high value of time, and it's not just rich people. So this kind of pricing thing, this could be a good thing, leaving those people who self-select to pay a lot, because they really value it, better off; and those who don't want to do it don't have to. And they obviously pay much less and get worse travel time. But it doesn't even end there. Now we are starting to realize: Once we start introducing these fundamental changes in pricing and we then realize that transportation interacts with an awful lot of economic activities. So, by putting in the pricing that we talked about, we realize, huh, there also might be, in the long run, changes in land use. Greater density. Russ: Absolutely. Guest: And so instead of having people spread out, things are closer. Well, that means government services are going to be less costly. You don't have to string wires from miles apart or whatever to provide some sort of public services. And now even, you know, modern research is talking about the benefits of economies of agglomeration. You hear from people to [?], get around, so on and so forth. Russ: Less carbon. Guest: Environmental effects. Now here's something else. People are realizing: I'm traveling during these congested conditions and I'm going out there and it's almost sort of a race to the top. I need a bigger car; I'm just around other cars and I don't like it, I'm uncomfortable in this kind of traffic and I want to be able to accelerate, so I'm going to drive a large car. Indeed, we have found, just as a stylized fact, that virtually all of the improvement in engine technology since the 1980s have been reflected in horsepower. Virtually none in fuel economy. Well, what might be at work since the 1980s? One thing is congestion. That just continues to go up and up and up. So it may very well be that a big part of our energy consumption is simply related to congestion. And if we smoothed out traffic flow and gave people less incentive to go out and buy big cars to try to feel more comfortable in traffic, you wouldn't have these what we call 'peer effects'--that could be another benefit of congestion pricing that falls on users. So, I think between heterogeneity--is the buzz word; people are different--and then thinking more broadly about how transportation interacts with so many other economic activities, slowly we are trying to build a case to say: Wait a minute; this is not simply a transfer of revenue from consumers to the government. This is something that could lead to fundamental changes that do benefit consumers. Russ: These are all excellent points. Although I do have to say that as someone who has recently spent a great deal of time commuting in the D.C. area, there are a lot of times I wanted a smaller car. Because traffic is moving so slowly, I'd feel perfectly safe in a very small car.
42:32Russ: But I want to ask you a different question. We recently had Tyler Cowen on EconTalk, and we were talking about technology and how it might change our lives, smart machines, etc. And he mentioned the driverless car. I'm curious what you think is going to be the barrier to--obviously there are legal issues; there are going to be issues of insurance that are going to make it interesting. But is it imaginable? When I think about a driverless car, it opens up the possibility of going 80 or 90 miles an hour, basically being in the equivalent of a high-speed train that would not crash into other cars--because everyone is going 80-90 mph and you wouldn't have the issues of bumper tailgating that slow down, which are the major causes of traffic. And as a result, you'd be on the equivalent of a high-speed train, except when you got off you'd still have your car, if you wanted to go somewhere else that wasn't connected. So to me the potential is very high. Do you think we're going to allow them on regular roads? Or could we imagine a world where roads are going to be created that will just have driverless cars and no other cars? Guest: My view, and I think transportation illustrates this, is that there is always a sort of battle between the private and public sector to innovation and introduce technological change. I've already mentioned just one example about stoplights. People know: If the roads were private, I assure you that private owners of roads would have traffic lights that respond in real time to traffic flows. And I think what we often have is the private sector trying to find ways to innovate, despite constraints of the public sector. We learned this again in deregulation. Before we had deregulation we had private entrepreneurs trying to show, yes, we could get competition in airlines. We certainly could find ways that we could compete between rail and truck. That kind of thing. Well, I think this is a textbook case, that, here we have a highway system that basically hasn't changed since we started building roads in this country. No major innovations in performing. Russ: EasyPass. That's our biggest innovation, E-ZPass. I don't have to slow down. Guest: Which, at least around here, came from a private innovator. The regional owner of the Greenway I think rode--was the one who put that in. I seriously doubt that it was anybody in the public sector who was the first to do this. My point being the driverless car is effectively going to be able to leapfrog technology to get over the fact that roads have not seen any innovation, where there could have been countless areas where we could have innovated, from the mundane to improvements in asphalt design, resurfacing, lane width that's adjusted in real time to traffic flows. Even realizing that cars don't break down any more; why are we wasting a lane on a breakdown lane? All sorts of things that could have been changed. And even ones that are more technologically sophisticated than that. So, the driverless car is coming. And it's a way that the private sector--in this case, the automobile companies--have been trying in a sense to compensate for inefficiencies in the road system. So, technologically we are going to have it. And I think as people get new cars they'll realize: Wait, the technology that I have in my new car that I can see in this, you know, screen on your dashboard, what's behind me--that's the same kind of thing that they are using in a driverless car. There are just cameras everywhere and cars will effectively be able to communicate with each other. So, it's an exciting development, and in principle could obviously improve traffic flows. Driverless cars do not rubberneck when there are accidents. You are not going to have that problem. And countless other things. But you are right: the problem is that driverless cars are going to have to operate on the public infrastructure. What's one obvious problem? Potholes. If the roads are not properly maintained, your nice driverless car is going to start slowing up and probably even going to be sensitive to the roads. Russ: They'll figure that out. Guest: But it's going to be a problem, in the sense that you'll start getting stop-and-go traffic. Again, signaling. If you have signals that are not really responding to real time traffic flows, but just continue to go on this here's-red, here's-green, so on and so forth, again, that's going to sort of undercut what the driverless cars are trying to do. So I think the bottom line is, yes, we're going to be able to get driverless cars. I think it's unfortunate, though, that we'll probably still have to get them on public roads. It's again, it's sort of the same problem that airlines have, that airplanes have, operating on the publicly-managed air traffic control system. That's something that's at least more visible. People know or at least have heard that we are trying to build a new air traffic control system based on satellite technology instead of radar. But I think they've also heard the horror stories--wait, this thing is year behind in being implemented, it's having cost overruns, and who knows when we are going to get this. The same kind of thing. You have the private mode that's trying to operate on this public facility, and you have people in the public sector that basically have to manage new technology, and they are not up to doing it. Russ: You note in your paper that airplane travel, the delays, have gotten longer. I was struck by two things when I saw that. One is how small the increase has been. It has gotten longer, the ground time versus flying time. Ground time is longer. But it hasn't gotten a lot longer. Having said that, it seems to me that there's a lot more caution about traveling in bad weather and there's a lot of lengthy delays. Not just turnaround being slower, but the canceling and delaying of flights. Am I wrong on those two points? Guest: Well, in terms of technology, we've gotten better in some ways, in that planes can certainly see around wind shear--I wouldn't say 'see' but they can navigate around wind shear. Remember this is all consistent with our safety data, where it is now so, so rare, fortunately, that we don't experience airplane crashes any more. So I think we are certainly doing better there. But I think that's a lot of learning on the part of pilots, better training, and improvements in uses of technology. So I don't think that's our problem. Our problem is just there are a lot of planes in the sky, and interacting with weather you are continue to have congestion until you can change technology. Which is what the satellite-based system is going to do. In the same way that driverless cars effectively will increase more capacity--cars can be closer together--the new satellite-based system will increase capacity. Planes can be closer together. And that's really going to be a major solution to the delays.
50:20Russ: I want to get into some big picture questions in a minute, but before I do, though, talk about what's going on in the parking world. I was surprised to see there's been some innovation, at least in some cities. I've never experienced personally, but the amount of time that people spend searching for parking is a big part of urban congestion you note in the paper. What's one of the technological things that's changing there? Guest: Well, again, you touched on this. You go to an airport, you have a sense where a parking space is. We have technology that can enable us to do that. You can embed sensors in parking spaces and have applications that say this one's available. You can then even set real-time prices. And various cities are exploring that, those kinds of innovations. Again, this is an example of trying to use improvements in information technology, interact them with transportation facilities, and make much more efficient use of the available capacity. And that could easily be done, in parking, as I said, in terms of reducing search so that people could know in real time what spaces are available. Obviously it would be more efficient to price the ones that are in the more congested areas or used in more congested times. But again, in the public sector these things are difficult. What I've been saying, we have no idea how much better the system could be, these are examples. The system could be so much better not so much because we could be building more roads or investing in more highways. It's because we could be introducing technological innovation and its advances much quicker than we've been doing so before. And again, if I can point to a parallel with deregulation, that was one of the lessons that we learned from deregulation, is how much innovation and technological advance have been held back because of regulation. Russ: What are some examples of that? So, you are talking about the late 1970s, when trucking and air travel-- Guest: railroads-- Russ: railroads, and trucking were all relatively deregulated, or at least regulated less. What were some of the innovations that happened in the subsequent period that might not have happened otherwise? Guest: People simply don't realize that using information technology has vastly improved railroads. If you wanted to know where a freight shipment was in any of the regulated periods, you'd call them up and they'd say: We have to check our camera. And you'd say, What? What would happen is railroads would bring their cars into a yard, and then there would be a camera taking pictures of the cars that came in. And they would look at those and say, There we go. That's where your shipment is. Or: that's how we know where it is. Now, we have real time identification and people know exactly where their shipment is. And people say, well, we could have done that during regulation. Sure. But what was the incentive? Where were innovations going to be generating more profit for these companies? They still face regulations and constraints. So what deregulation did was just open up the incentives and in some cases opportunities. Like, airlines could then dramatically change both their networks and their pricing algorithms, and again introduce new technologies to do yield management. So, these were sort of basic things, but they did greatly improve the provision and operations of inner city transportation. The potential in so much of the infrastructure to do these things and apply innovations in technology is even greater. And that's why my belief is the payoff will be even bigger.
54:22Russ: So, let's have a thought experiment here. I want to make Cliff Winston the Transportation Czar. But he's still Cliff Winston; he's not going to be subject to whatever awful political incentives the current Transportation Czars are under. What do you think are the two or three biggest things we could do that would have the biggest impact? What are the two or three policy changes? I have a feeling I know one or two of them already. But, what are the two or three you would list that are the biggest mistakes we are making, that we could do better if we had more economic wisdom and less political oversight? Guest: I think there's one general mistake, and that's just the notion of experimentation. What we learned that worked in regulatory reform were experiments. In other words, the fact that we were able to point to California and Texas as having lower airline fares for flights of comparable distance on interstate routes opened policy-makers' eyes. And I think that's the kind of thing that the Department of Transportation under Secretary Mary Peters and Assistant Secretary Tyler Duvall were trying to do: have experiment. My general thing would be, look, let's start trying experiments in ways to get things to be more efficient. People know what these things are. Most of what I'm saying for anybody who is in the public, or even the private sector, is no surprise. They know about these things. The question is finding places where you can conduct experiments. So, to try to put in road pricing based on real time traffic flows. Allowing competition with no constraints from other kinds of transit companies, not just the traditional bus and rail. These kinds of things. And let private entrepreneurs think of these things. They are going to do a better job than I can. That's the kind of thing we've learned from deregulation--people who are actually in this line of work are going to come up with important ideas and innovative ideas. So I would say that's the kind of general mindset I'm trying to push for, is experiments that start showing the public: Look, this can be much better. And the advantage of America is we've got a lot of cities and a lot of states, and we can start trying to do these kinds of things. That was really the direction it was said the the previous DOT (Department of Transportation) people wanted to go, and unfortunately they were constrained in doing so. Russ: In terms of technology, we've talked about the driverless car. Elon Musk has proposed the Hyperloop, which is this wild concept of--I don't know, how would you describe it?--a slingshot of--Google it out there. Guest: The time tunnel. Russ: We'll put a link up to it. You've got to see it to really understand it. Which of these things or others that I haven't mentioned, are really about to happen and could really change--we have other proposals, high speed rail, which I think people are starting to realize is not really the best idea: way too expensive, not likely to be a useful use of public funds. Close with what technologies might transform the transportation sector. Guest: Driverless car, absolutely. That will have a huge impact, both in reducing delays and in safety in this country. And also, then, give people a sense that transportation is related to so many other activities, and free up people to do other things. And I think also the satellite-based air traffic control system will also be excellent. It will make safety that much better for air transportation. It will improve routings; it will improve speed; it will reduce delays; it will effectively expand the sky in congested areas and help there. But beyond that, there's still other things that we just don't know. Transportation used to be the place that was the source of all the major innovations, among the major innovations of mankind. The jet plane, the car, so on and so forth. It's been a long time, many decades now, since we've had these kinds of things. I think part of the problem really is the government's pervasive role and the public sector provision of management and operation. I would like to see much more involvement of the private sector, and I think despite the constraints of the public sector, the private sector is still trying to do things like the driverless car. Private sector could certainly get involved in provision of air traffic control systems that are private. It's going on in other parts of the world; another thing that most Americans don't realize is that there are other places where private people are much more involved. So, I think that's the kind of thing we want to look for, is major innovations of the sort that we have. But just also a great realization of how transportation intersects with so many other areas of our life that is just so important to greatly improve the system and make it as innovative as possible.

More EconTalk Episodes