Latest / Statecraft / What Is America’s Infrastructure Cost Problem?
Transcript
- 0:04SPEAKER_00: Hi, guys.
- 0:05SPEAKER_00: I'm Santi Ruiz.
- 0:06SPEAKER_00: I'm the senior editor at IFP,
- 0:07SPEAKER_00: the Institute for Progress,
- 0:09SPEAKER_00: a nonpartisan think tank that I think is abundance-coded,
- 0:12SPEAKER_00: if not formally an abundance think tank.
- 0:14SPEAKER_00: Today, I'm interviewing Zach Liskow.
- 0:15SPEAKER_00: Please give him a little round of applause.
- 0:20SPEAKER_00: Zach, my guest today is a professor of law at Yale Law School.
- 0:24SPEAKER_00: In 2022 and 23,
- 0:26he was the chief economist at the Office of Management and
- 0:28Budget at the White
- 0:29SPEAKER_00: House.
- 0:30SPEAKER_00: He's also now my colleague at IFP, is a non-resident senior fellow.
- 0:35SPEAKER_00: And I have a little bit of a problem today,
- 0:37SPEAKER_00: which is that,
- 0:39SPEAKER_00: Zach,
- 0:39SPEAKER_00: you may not be a national household name,
- 0:42SPEAKER_00: but you might as well be in this event.
- 0:46I think most of you will be aware Zach's done a lot of work on
- 0:51many interesting
- 0:51SPEAKER_00: economic topics,
- 0:52SPEAKER_00: but especially on infrastructure costs,
- 0:55SPEAKER_00: on why it costs so much to build in the U.S.,
- 0:57SPEAKER_00: what the inputs are there, the kind of cross-cutting comparisons.
- 1:02And the challenge for your interviewer today is that I think
- 1:05everybody generally
- 1:06SPEAKER_00: gets it now,
- 1:07SPEAKER_00: in part because of your work,
- 1:08SPEAKER_00: that infrastructure in the U.S.
- 1:10SPEAKER_00: costs a huge amount to build.
- 1:11SPEAKER_00: I reviewed a project,
- 1:13SPEAKER_00: some submissions for a project at IFP recently on transit,
- 1:16SPEAKER_00: in which almost every submission started with the fact that
- 1:19Cost per mile to build subway in New York is like eight times
- 1:23more than the
- 1:23SPEAKER_00: equivalent project in Paris.
- 1:25SPEAKER_00: It's just like a fact that people understand now.
- 1:27SPEAKER_00: These stylized facts are just out there in the discourse.
- 1:31So that makes it a little bit hard to figure out how to have a
- 1:34conversation that
- 1:35SPEAKER_00: isn't just all of us nodding in agreement.
- 1:37SPEAKER_00: It really is a problem, huh?
- 1:39SPEAKER_00: I'm going to try and tackle that problem.
- 1:40SPEAKER_00: I just want to lay my cards on the table.
- 1:42SPEAKER_00: This is my fear.
- 1:43SPEAKER_00: I'm going to try to avoid it.
- 1:45SPEAKER_00: Let me start here.
- 1:46SPEAKER_00: So we all know infrastructure costs too much in the U.S.
- 1:48SPEAKER_00: and there are a lot of plausible stories you can tell about why that is.
- 1:53One of the ones I think people would most associate with the
- 1:55kind of abundance
- 1:57narrative is we've built up a ton of regulation and the
- 2:01regulation creates soft
- 2:02costs and it increases the time that it takes to build
- 2:06infrastructure and that
- 2:07SPEAKER_00: itself is
- 2:08SPEAKER_00: creates more cost because the longer you work on a project,
- 2:11SPEAKER_00: the more all kinds of costs accrue,
- 2:12SPEAKER_00: and complying with the regulations themselves creates a cost.
- 2:16SPEAKER_00: That's one stylized narrative.
- 2:19SPEAKER_00: Before we get into what we actually know,
- 2:21SPEAKER_00: just from the fact that it costs too much,
- 2:23SPEAKER_00: what are some possible theories that get thrown around?
- 2:27SPEAKER_00: Then we'll get to what you actually think is a problem,
- 2:29but what's the world of potential explanations for why it costs
- 2:33so much to build in
- 2:34SPEAKER_00: the U.S.?
- 2:34SPEAKER_00: ?
- 2:35SPEAKER_01: Yeah, first, delighted to be here.
- 2:37SPEAKER_01: Thanks for the very, very kind introduction.
- 2:40SPEAKER_01: I divide the things that,
- 2:44SPEAKER_01: at least I think,
- 2:45SPEAKER_01: drive up transportation infrastructure costs into four buckets.
- 2:50The first is the one that you mentioned,
- 2:54kind of permitting and litigiousness.
- 2:57SPEAKER_01: The second are lots of procedural rules,
- 3:02SPEAKER_01: including especially on procurement for purchasing things,
- 3:05which often can reduce competition because it's hard to get
- 3:08competitors to go
- 3:10SPEAKER_01: through the complex process.
- 3:12SPEAKER_01: The third is personnel.
- 3:15SPEAKER_01: We have state DOTs,
- 3:18SPEAKER_01: also federal DOT that's maybe understaffed,
- 3:22SPEAKER_01: especially for more highly skilled positions.
- 3:25SPEAKER_01: And then fourth, we just really weak data.
- 3:29SPEAKER_01: It's hard to know actually what's going on out there in the world,
- 3:32and it's hard to fix the problem if you kind of don't know all
- 3:34the facts on the
- 3:35SPEAKER_01: ground.
- 3:36SPEAKER_01: So I'm not going to be able to offer exact percentages to those,
- 3:39SPEAKER_01: and it's going to vary depending on the exact project,
- 3:41but I guess those are kind of the four things that I would
- 3:43divide the issues into.
- 3:45And just give us one more time so that it's in our memory,
- 3:47the four buckets.
- 3:48SPEAKER_00: Permitting?
- 3:49SPEAKER_01: Yeah, three Ps.
- 3:50SPEAKER_01: Permitting, procedure, personnel, and data.
- 3:54SPEAKER_00: Super.
- 3:55SPEAKER_00: Let's start with permitting, which I think
- 3:57It's definitely the one I have spent in my professional career
- 3:59the most time thinking about.
- 4:01SPEAKER_00: People in this room are familiar with NEPA.
- 4:03SPEAKER_00: There are many other pieces of the permitting puzzle,
- 4:07SPEAKER_00: but I think NEPA,
- 4:08SPEAKER_00: the national level one,
- 4:09SPEAKER_00: gets the most attention.
- 4:11SPEAKER_00: Maybe just at a high level,
- 4:12SPEAKER_00: you can tell me,
- 4:13am I or are people like me spending too much time on the
- 4:18permitting piece,
- 4:18SPEAKER_00: on NEPA specifically,
- 4:20SPEAKER_00: relative to all the other blockers that exist?
- 4:22SPEAKER_01: Yes.
- 4:24SPEAKER_01: But I'm part of the problem.
- 4:27SPEAKER_01: I've written a lot on NEPA.
- 4:28SPEAKER_01: I think it should be reformed.
- 4:30SPEAKER_01: I think it's problematic.
- 4:32SPEAKER_01: But I do think it gets,
- 4:34SPEAKER_01: at least for transportation infrastructure,
- 4:36SPEAKER_01: I think more time than it deserves relative to the other three buckets.
- 4:43SPEAKER_01: So I do think it's an issue.
- 4:45SPEAKER_01: There are lots of reasons to think it is.
- 4:47SPEAKER_01: It takes four years to do the average environmental impact statement.
- 4:53SPEAKER_01: And then there's often litigation.
- 4:56SPEAKER_01: Even if the project opponents lose on everything,
- 4:59SPEAKER_01: it takes a year and a half for the media and lawsuit.
- 5:02If they went on one thing,
- 5:03it takes two and a half years for the media and lawsuit.
- 5:05SPEAKER_01: And then you can appeal.
- 5:08SPEAKER_01: So there's the time issue, but I think there's also the cost issue.
- 5:12And it's partly,
- 5:14it takes time and money to write the report for four years.
- 5:18But I think the issue is more the things that you do either
- 5:22defensively or
- 5:23SPEAKER_01: reactively to the project itself that make it a lot more expensive.
- 5:27SPEAKER_01: So my paper,
- 5:29SPEAKER_01: Infrastructure Cost with Leah Brooks,
- 5:31SPEAKER_01: we talk about this section of interstates outside of Detroit.
- 5:36There were some earlier sections that were built pretty
- 5:39cheaply,
- 5:39pretty quickly in the 1960s.
- 5:42SPEAKER_01: And then various things happened.
- 5:46SPEAKER_01: For the last segment, there was a lawsuit.
- 5:49SPEAKER_01: It was challenged under NEPA.
- 5:51SPEAKER_01: And it costs several times as much.
- 5:53SPEAKER_01: But what happened, well, they built sound walls.
- 5:56SPEAKER_01: It was built below ground.
- 5:58SPEAKER_01: There were parks built on top of the highway next to fields.
- 6:03SPEAKER_01: Incidentally, it's not an area where lots of open space was needed.
- 6:07But these things were all done to get the lawsuits to stop so
- 6:10that the thing could
- 6:10SPEAKER_01: actually be built.
- 6:11It was really important this highway was built because it
- 6:13connected the other two
- 6:13SPEAKER_01: pieces of the highway.
- 6:15SPEAKER_01: So...
- 6:16really drives up costs just to try to get things done in terms
- 6:19of what's actually
- 6:20SPEAKER_01: done to the projects themselves.
- 6:23So given that kind of backdrop,
- 6:26why should we think DEEP is not the core issues?
- 6:30SPEAKER_00: This all sounds very compelling,
- 6:31SPEAKER_00: and this is a story,
- 6:32it's a version of a story that I've told before to my parents
- 6:36about what's the
- 6:37SPEAKER_00: permitting thing.
- 6:38SPEAKER_00: So why,
- 6:39SPEAKER_00: I mean,
- 6:40SPEAKER_00: it might be under-discussed out there in the world,
- 6:42SPEAKER_00: but maybe it's over-discussed by people in this room.
- 6:44SPEAKER_00: If so, why?
- 6:46SPEAKER_01: So I think there are two pieces to it.
- 6:48SPEAKER_01: One is there are lots of issues of permitting,
- 6:50SPEAKER_01: especially at the state and local level,
- 6:51SPEAKER_01: that are not NEPA.
- 6:53SPEAKER_01: So there are all sorts of water permitting rules and local mini-NEPAs.
- 7:01SPEAKER_01: And these things all matter as well, just in terms of permitting.
- 7:05SPEAKER_01: But then there's this whole other half as well.
- 7:08And it's going to vary depending upon the kinds of project that
- 7:10you're talking about.
- 7:11SPEAKER_01: So especially if you're building a highway,
- 7:13a linear project that's above ground where you're going to have
- 7:16to knock stuff
- 7:16SPEAKER_01: down,
- 7:17SPEAKER_01: that's going to have significant environmental impacts.
- 7:19SPEAKER_01: Some people are really going to be ticked about it.
- 7:22SPEAKER_01: There's a good chance of a lawsuit.
- 7:25SPEAKER_01: Permitting is really, really important there.
- 7:27Other things,
- 7:29you know,
- 7:29a tunnel underground in New York that's below ground.
- 7:34SPEAKER_01: It's like the environmental impacts on people,
- 7:37SPEAKER_01: the litigation because of the environmental impacts,
- 7:39SPEAKER_01: that's not the main thing that's going on.
- 7:42SPEAKER_01: And yet...
- 7:43SPEAKER_01: you know,
- 7:44the miles in New York City are the most expensive ever built in
- 7:47the world for no
- 7:49SPEAKER_01: reason having to do with geography.
- 7:51SPEAKER_01: And for that,
- 7:52SPEAKER_01: I think you want to look to procedures,
- 7:55SPEAKER_01: personnel,
- 7:56SPEAKER_01: and just at a high level,
- 7:58SPEAKER_01: the data that we lack for understanding what's really going on.
- 8:00SPEAKER_00: Let's talk about personnel, which is a piece of the puzzle.
- 8:03SPEAKER_00: I keep alliterating.
- 8:04SPEAKER_00: This is terrible.
- 8:06SPEAKER_00: The three P's is really horrible,
- 8:07which is a piece of the puzzle that you've been spending a lot
- 8:08of time on recently.
- 8:11SPEAKER_00: You just had a paper out.
- 8:12SPEAKER_00: Was it
- 8:12SPEAKER_00: Last night or this morning that I just want you to talk about,
- 8:16SPEAKER_00: about what's the role that personnel play in costs.
- 8:21SPEAKER_01: So a paper with Caitlin Slattery and Will Nober studying,
- 8:25SPEAKER_01: in part,
- 8:26SPEAKER_01: transportation and construction in California.
- 8:30And we assemble data on the individual engineers who are
- 8:33involved in building
- 8:35SPEAKER_01: transportation projects in the state.
- 8:38And what we found really surprised me and shifted my views
- 8:42about the basic issues
- 8:45SPEAKER_01: that drive up costs.
- 8:47SPEAKER_01: So we found a couple things.
- 8:49SPEAKER_01: First, we found that when an extra 1% of state engineers retire,
- 8:56SPEAKER_01: costs go up by 4.5%, which is like a really, really big effect.
- 9:00SPEAKER_01: Furthermore,
- 9:01we measured the quality of the engineers who work for the
- 9:06government in the state,
- 9:07and we found that if you increase the quality as we measure it
- 9:11from the 25th to the
- 9:12SPEAKER_01: 75th percentile,
- 9:14SPEAKER_01: that results in huge cost savings.
- 9:17Cost savings that amount to three times the amount that it
- 9:20costs to hire one of these engineers.
- 9:23So you should be willing to pay just a gigantic amount to
- 9:26increase quality because
- 9:27SPEAKER_01: it would pay for itself.
- 9:29SPEAKER_01: So it would pay for itself to reduce the retirements,
- 9:30SPEAKER_01: it would pay for itself to increase quality.
- 9:33SPEAKER_00: Just by reduce their retirement,
- 9:35you mean just to keep these people on for a few more years is
- 9:38hugely valuable?
- 9:39SPEAKER_01: Yeah.
- 9:40SPEAKER_01: I mean,
- 9:40SPEAKER_01: a lot of these highly skilled people,
- 9:43SPEAKER_01: because of a phenomenon called wage compression in the public sector,
- 9:47SPEAKER_01: state governments just don't pay that much for highly skilled workers.
- 9:52SPEAKER_01: So they maybe work in state government for a handful of years,
- 9:55SPEAKER_01: and then they go work in the private sector and have a much,
- 9:58SPEAKER_01: much higher salary.
- 10:00SPEAKER_01: And if you could retain them, you could benefit from their expertise.
- 10:04SPEAKER_00: One thing that was really striking to me about that paper,
- 10:07SPEAKER_00: just reading it last night,
- 10:08SPEAKER_00: was you talk about not just these kind of individual-level effects,
- 10:13SPEAKER_00: but about the kind of broader issue of talent across state DOTs that,
- 10:18SPEAKER_00: like,
- 10:18SPEAKER_00: for the last 20 years,
- 10:19both the total number of employees at most departments of
- 10:22transportation at the
- 10:23state level has been dropping and that the relative size of DOT
- 10:29relative to the
- 10:29SPEAKER_00: rest of the state government is shrinking.
- 10:32So you're just a smaller piece of the government and you're
- 10:34smaller numerically.
- 10:35That's a 20 year phenomenon and my impression is it's pretty
- 10:39consistent across the
- 10:40SPEAKER_00: US that this is happening maybe with some variation.
- 10:43SPEAKER_00: Why is that happening?
- 10:44SPEAKER_00: Why are we just generally bleeding talent over this period?
- 10:47SPEAKER_01: Yeah, it's a huge issue.
- 10:50State departments in transportation are doing more,
- 10:53yet they have fewer people there.
- 10:55SPEAKER_01: So I think that there are a few theories.
- 10:58One is we had a financial crisis that imposed huge pressure on
- 11:03the states.
- 11:05SPEAKER_01: And what you often see,
- 11:07SPEAKER_01: this is definitely the case in Connecticut even recently,
- 11:11SPEAKER_01: is like the thing that you cut is not Medicaid.
- 11:16You instead cut the workers who are the bureaucracy that
- 11:21implement the things across government.
- 11:24SPEAKER_01: And you could tell stories where that's the case, you know,
- 11:28SPEAKER_01: The impacts of these things are often kind of diffuse and long-term,
- 11:32whereas a salient cut to some benefits program or paving fewer
- 11:38highways would be
- 11:39SPEAKER_01: kind of salient and opposed.
- 11:42And states have balanced budget requirements,
- 11:45so the fiscal constraints really bite.
- 11:49SPEAKER_01: States are under more fiscal stress,
- 11:51SPEAKER_01: and this appears to be the way that the states have dealt with it.
- 11:55SPEAKER_01: And what my paper suggests is that they've dealt with it in a way that's
- 11:58SPEAKER_01: counterproductive and self-defeating.
- 12:00SPEAKER_00: Do they end up paying more for those short-term decisions?
- 12:03Because I think part of the stylized story that I've heard and
- 12:06I think I'll repeat
- 12:07is that you cut headcount in a downturn because you've got to
- 12:11get rid of costs,
- 12:12SPEAKER_00: and then later you pay contractors,
- 12:13SPEAKER_00: maybe those same people on the private sector,
- 12:16SPEAKER_00: a premium to come back and do the same work.
- 12:18SPEAKER_00: How correct is that toy model?
- 12:21SPEAKER_01: So if you look at civil engineers,
- 12:25SPEAKER_01: who are the people who build transportation as well as other things,
- 12:27SPEAKER_01: if you look at that over the last quarter century,
- 12:30SPEAKER_01: the number in government has been basically flat.
- 12:34SPEAKER_01: There's just been this explosion of those outside of government.
- 12:38So what that says to me is we're just outsourcing more and more
- 12:41and more.
- 12:42SPEAKER_01: And to be clear, we need to outsource.
- 12:46SPEAKER_01: The people who actually build the transportation infrastructure,
- 12:50SPEAKER_01: they're 100% private contractors,
- 12:52SPEAKER_01: basically.
- 12:53But you need to have the capacity inside of government to
- 12:56manage this very,
- 12:56SPEAKER_01: very complex process,
- 12:57SPEAKER_01: to choose what to build,
- 13:00and to permit it and to put the thing out to bid and then to
- 13:04manage this often
- 13:06SPEAKER_01: pretty complex thing.
- 13:07SPEAKER_01: If you don't have enough people inside of government,
- 13:09you're maybe not going to have enough contractors who actually
- 13:13bid on the thing.
- 13:14SPEAKER_01: You're not going to manage them well.
- 13:15SPEAKER_01: You're going to change your mind, which is very, very expensive.
- 13:18SPEAKER_01: You're going to let them do things that kind of take you for a ride.
- 13:22So there's lots and lots of reasons to think that if you have
- 13:25just too few people
- 13:26SPEAKER_01: in government,
- 13:27SPEAKER_01: you're going to outsource too much,
- 13:28SPEAKER_01: which ends up being very,
- 13:29SPEAKER_01: very expensive.
- 13:30why is it the case that if you don't have really good folks in
- 13:33your department of
- 13:33transportation you get fewer contractor bids is it because
- 13:36you're writing bad
- 13:37contracts and nobody wants to take you up on those contracts
- 13:40just flesh that out
- 13:41for me yeah so one finding in the study that we have is that
- 13:45the quality engineers
- 13:47SPEAKER_01: matter most in places with less competition
- 13:50which to us suggests that one of the ways in which state
- 13:54engineers really matter is
- 13:56SPEAKER_01: that they call up some people and bring in more bidders.
- 14:01SPEAKER_01: They have simpler procedures that are easier for bidders to understand.
- 14:05So one possible mechanism of what's going on is they bring in
- 14:10more competition.
- 14:11SPEAKER_01: The level of competition in these bids is kind of astonishingly low.
- 14:15SPEAKER_01: The typical bid for one of these things has like three bidders,
- 14:20and yet states basically do almost no outreach to try to
- 14:23increase the number of
- 14:24SPEAKER_01: bidders.
- 14:25And if you have like the same bidders coming over and over
- 14:28again,
- 14:28and there aren't many of them,
- 14:30SPEAKER_01: you know, competition, reduce competition sends drive-up prices.
- 14:33And sophisticated bureaucrats are going to try to increase
- 14:38competition to reduce costs.
- 14:39SPEAKER_00: If you're talking to a policymaker who's not a transit wonk,
- 14:44but you want to explain to them which kinds of capacity at the
- 14:48state level you want
- 14:48SPEAKER_00: to build in your state department of transportation,
- 14:51SPEAKER_00: what's the quick explanation?
- 14:53SPEAKER_00: Because the answer isn't hire more people, period.
- 14:56SPEAKER_00: Presumably you want
- 14:57SPEAKER_00: I think, more engineers.
- 14:59How do you represent that to somebody who's not really familiar
- 15:01with the transit world?
- 15:03SPEAKER_01: Yeah.
- 15:03In the context of transportation,
- 15:04the real choke point are these highly skilled engineers.
- 15:08SPEAKER_01: And so that's where the pay compression in the state really hits.
- 15:11SPEAKER_01: These people, they're highly educated.
- 15:14SPEAKER_01: If they're good, they're talented people.
- 15:17SPEAKER_01: And they command high salaries.
- 15:19SPEAKER_01: So my view is we need...
- 15:21more of them and we need to retain those that we have and we
- 15:26need higher quality
- 15:27ones and this amounts to creating more slots and paying the
- 15:32ones that we have more
- 15:33SPEAKER_01: to increase the talent of those who are working for the government.
- 15:37I have like a political economy follow-up question here which
- 15:40is I can imagine if
- 15:42I'm the head of my state department of transportation and this
- 15:45is pure speculation
- 15:46SPEAKER_00: but I want you to tell me if this is right or wrong
- 15:48SPEAKER_00: I can imagine totally understanding this problem,
- 15:50SPEAKER_00: understanding,
- 15:50SPEAKER_00: yeah,
- 15:51we pay the same people three times more in the private sector
- 15:53to do this work.
- 15:55SPEAKER_00: I would be able to keep down costs better.
- 15:56SPEAKER_00: I would be rewarded politically if I would bring these people in-house.
- 16:01SPEAKER_00: Is there pressure from politicians,
- 16:04from the kind of civil servants who run these organizations to
- 16:08bring this stuff
- 16:08SPEAKER_00: in-house to improve their own state capacity?
- 16:11SPEAKER_00: And if not,
- 16:12why don't we see huge pressures from people who want to succeed
- 16:15and be seen as
- 16:16SPEAKER_00: succeeding and keeping costs down
- 16:19SPEAKER_00: to solve all these problems?
- 16:20SPEAKER_00: Or do we, and it's just, I don't know, help me understand that.
- 16:24I think it's a very interesting set of questions that the folks
- 16:27should look into more.
- 16:29SPEAKER_01: I'm going to give you one hypothesis about part of what's going on.
- 16:34So if you think about the incentives of the engineers who work
- 16:39in government,
- 16:41many of them are going to leave and then go make a lot more
- 16:45money in the private
- 16:46SPEAKER_01: sector with private contractors.
- 16:48maybe they don't want to really rock the boat with the private
- 16:52contractors who they
- 16:53SPEAKER_01: want to be their future employers.
- 16:55Is the last 20 years of your career going to be spent making
- 16:59all that money on the back end?
- 17:00SPEAKER_01: Yeah.
- 17:01SPEAKER_01: So,
- 17:02SPEAKER_01: listen,
- 17:03SPEAKER_01: I've not worked for a State Department of Transportation,
- 17:06SPEAKER_01: but from the outside,
- 17:07SPEAKER_01: as an economist,
- 17:08SPEAKER_01: just looking at the raw incentive structure,
- 17:12SPEAKER_01: there's reason for concern.
- 17:14SPEAKER_00: Another political economy question, I guess, is...
- 17:17Why is this a pretty consistent pattern across states,
- 17:21including liberal states?
- 17:23SPEAKER_00: I think you can tell a very natural story where red states,
- 17:26SPEAKER_00: they're happy to cut headcount at their DOTs.
- 17:30SPEAKER_00: In fiscal trouble, you do that.
- 17:31SPEAKER_00: That's your first instinct is to cut people, cut services.
- 17:35SPEAKER_00: I guess it's kind of curious.
- 17:36SPEAKER_00: It seems to have happened pretty broadly across the U.S.
- 17:39SPEAKER_00: that DOTs have less capacity, fewer people than they used to.
- 17:44SPEAKER_00: Why is it such an even distribution?
- 17:47SPEAKER_01: Yeah, I think it's a great question.
- 17:49Your underlying question is,
- 17:51shouldn't blue states be kind of pro-state employee?
- 17:55So one hypothesis is that it's the blue states that are often
- 18:00under more fiscal pressure.
- 18:02SPEAKER_01: So blue states often spend more on Medicaid.
- 18:07SPEAKER_00: But don't they tax more to make up for it?
- 18:09SPEAKER_01: So if you look at Connecticut,
- 18:11SPEAKER_01: for example,
- 18:12SPEAKER_01: a blue state where I live,
- 18:13SPEAKER_01: we have the highest,
- 18:14SPEAKER_01: one of the couple of highest debt loads in the country.
- 18:16So we're under like,
- 18:18I think we have like three times the debt per person as Texas
- 18:21does.
- 18:22SPEAKER_01: So we're just under much, much more fiscal stress than Texas does.
- 18:26Texas also is growing and kind of grow out of the debt that
- 18:29they do have in a way
- 18:30SPEAKER_01: that blue states typically are not growing.
- 18:32SPEAKER_01: So I think that the fiscal stress really hits in blue states on average,
- 18:38SPEAKER_01: substantially more than in the red states.
- 18:40SPEAKER_00: I'll just flag,
- 18:40SPEAKER_00: I talked a few months ago to Randy Clark,
- 18:43SPEAKER_00: who's the head of the DC metro system,
- 18:45SPEAKER_00: seeing some cheers in the audience.
- 18:47SPEAKER_00: Great podcast.
- 18:51SPEAKER_00: In an earlier conversation,
- 18:52SPEAKER_00: I think with Axios,
- 18:53SPEAKER_00: he talked about the consultant industrial complex.
- 18:55SPEAKER_00: That's the term he'd use.
- 18:56And when I asked him about it,
- 18:57he said,
- 18:57I think what I mean is the transit industrial complex.
- 19:00SPEAKER_00: And he kind of pulled back from saying the consultants are the problem.
- 19:04SPEAKER_00: I kind of poked him on it.
- 19:05But I think he had a point,
- 19:06which is that people like him need these consultants to
- 19:08SPEAKER_00: because they don't have the capacity in house to do all this stuff.
- 19:11And even when they do,
- 19:13sometimes you need consultants for specific things.
- 19:15You don't want to,
- 19:15like you said,
- 19:16you don't want to bring everything in house.
- 19:18SPEAKER_00: So I guess to go a little deeper,
- 19:20SPEAKER_00: like in an ideal Zach Liskow,
- 19:23SPEAKER_00: like directed and advised department of transportation,
- 19:26SPEAKER_00: you're adding a bunch more engineers.
- 19:28Like what's the role of the consultants in the private sector
- 19:31besides literally
- 19:32SPEAKER_00: pouring the cement?
- 19:33SPEAKER_01: Yeah,
- 19:34totally agree that you definitely need consultants not just for
- 19:38actually building
- 19:38SPEAKER_01: the things,
- 19:39SPEAKER_01: but also for planning them.
- 19:40SPEAKER_01: So take the example of the Second Avenue subway, big tunneled project.
- 19:45It does not make sense for New York City to hire for 30 years
- 19:50people who are
- 19:50tunneling experts because they're only tunneling a small share
- 19:54of the time.
- 19:55SPEAKER_01: You need to hire large international firms who are experts in this to...
- 19:58SPEAKER_01: to do it.
- 19:59SPEAKER_01: So you definitely have to,
- 20:01SPEAKER_01: you need enough to manage the project well,
- 20:03SPEAKER_01: and that's going to depend upon the context.
- 20:06I've shifted my views about the importance of personnel in part
- 20:09from my own work,
- 20:10SPEAKER_01: but also in part from the really,
- 20:12SPEAKER_01: really excellent work from the transit cost project.
- 20:16They talk about the Green Line in California,
- 20:18in California high-speed rail.
- 20:20SPEAKER_01: These are just like billions-dollar projects.
- 20:23SPEAKER_01: And at one point in the Green Line,
- 20:25SPEAKER_01: they had five people managing this couple billion-dollar project.
- 20:28SPEAKER_01: And it's like you need enough people at least to manage the contractors.
- 20:32SPEAKER_01: And they just did not have that.
- 20:34That's really interesting because California is not a state
- 20:37that I think of as
- 20:38SPEAKER_00: having insufficient access to good civil servants.
- 20:40SPEAKER_00: Right.
- 20:41SPEAKER_01: Well,
- 20:42so the structure that was done for the California high-speed
- 20:47rail was to create a
- 20:48SPEAKER_01: whole new organization.
- 20:49SPEAKER_01: And, you know, it takes time to staff it up.
- 20:52SPEAKER_01: And they were managing, like, huge amounts of money with 10 or 12 people.
- 20:57SPEAKER_01: And they...
- 20:59SPEAKER_01: started things,
- 21:00SPEAKER_01: and they realized,
- 21:00SPEAKER_01: oh,
- 21:00SPEAKER_01: yeah,
- 21:01SPEAKER_01: that doesn't work,
- 21:01SPEAKER_01: and they had to go back and do it.
- 21:02SPEAKER_01: It's really,
- 21:02really expensive to start something and then stop it and undo
- 21:05it when you're
- 21:06SPEAKER_01: talking about concrete on the ground.
- 21:08SPEAKER_01: So there are questions about the right institutional structure,
- 21:11but their own audit suggested it did not work out well for
- 21:14California High-Speed
- 21:15SPEAKER_01: Rail.
- 21:15SPEAKER_01: There's lots of issues, but this is one of them.
- 21:18SPEAKER_00: Sure.
- 21:18SPEAKER_00: Let me ask you about other big blockers.
- 21:20SPEAKER_00: We've talked about personnel and permitting.
- 21:23SPEAKER_00: I'm trying to remember my P's.
- 21:24There's a lot of other things that I think you could and the
- 21:27people in this room
- 21:27SPEAKER_00: would point to as blockers.
- 21:29SPEAKER_00: Some people like me would say we don't talk enough about the unions here,
- 21:32but you could also talk about maybe it's about like Anglo
- 21:35common law practices,
- 21:36and that's why it always costs more in English-speaking
- 21:40countries than in France or
- 21:41SPEAKER_00: in Morocco.
- 21:42SPEAKER_00: You could talk about like citizen voice.
- 21:44SPEAKER_00: The problem is too many people, citizens get a say in projects generally.
- 21:48SPEAKER_00: How should we think about that kind of,
- 21:49SPEAKER_00: there's like a whole bunch of other things that you could point to,
- 21:51SPEAKER_00: and presumably it's not one of these things,
- 21:53SPEAKER_00: but like you said,
- 21:55SPEAKER_00: not all of them are the same level of importance here.
- 21:57SPEAKER_00: How should we prioritize?
- 21:59SPEAKER_01: Yeah,
- 21:59I think that part of what makes the infrastructure costs at
- 22:03least so interesting to
- 22:04SPEAKER_01: me as an academic is that there's not one answer.
- 22:08SPEAKER_01: There are several things that are going on simultaneously,
- 22:10SPEAKER_01: and sometimes they're related,
- 22:12SPEAKER_01: and sometimes there are just multiple issues going on.
- 22:15SPEAKER_01: So you gave a great list.
- 22:17SPEAKER_01: Something that I think really does not get enough attention,
- 22:21SPEAKER_01: maybe because it's the most boring word in the English dictionary,
- 22:24SPEAKER_01: is procurement.
- 22:26SPEAKER_01: And super important.
- 22:29I think the ratio of discussion of permitting to procurement is
- 22:33not where it should
- 22:34SPEAKER_01: be,
- 22:34SPEAKER_01: and I'm part of the problem.
- 22:35SPEAKER_00: Surely that can't just be because of boringness,
- 22:37because permitting is not the most interesting topic in the
- 22:43world to lay people.
- 22:45SPEAKER_00: We'll have to dig into this.
- 22:46SPEAKER_00: Why is procurement really got the short end of this day?
- 22:49SPEAKER_01: Fair enough.
- 22:50SPEAKER_01: It does involve really arcane rules.
- 22:52SPEAKER_01: Let me give you one example.
- 22:54SPEAKER_01: For a large number of infrastructure projects in the U.S.,
- 22:59SPEAKER_01: you have to take the lowest cost bidder.
- 23:01SPEAKER_01: Which kind of makes sense.
- 23:02Like you put out a bid and then there are a variety of bidders
- 23:07and like makes sense.
- 23:08SPEAKER_01: Maybe you take the cheapest one.
- 23:10Except maybe that firm that had the lowest bid has a history of
- 23:15going way over
- 23:16SPEAKER_01: budget or going way over time.
- 23:19SPEAKER_01: or having low quality, or having a proposal that's worse in some way.
- 23:25And it's often the case that you actually have to take the
- 23:28lowest cost bid.
- 23:30SPEAKER_01: And in European countries,
- 23:31SPEAKER_01: to turn to international comparisons that spend much,
- 23:34SPEAKER_01: much less on transit,
- 23:35SPEAKER_01: they give like 25% weight to the bid.
- 23:40SPEAKER_00: In their decision-making criteria?
- 23:42SPEAKER_01: In their decision-making criteria.
- 23:43And like 75% to these other quality-related things that we
- 23:47often give little or zero way to.
- 23:50SPEAKER_01: There's been some progress here.
- 23:51SPEAKER_01: Some states have reformed, but many have not.
- 23:54SPEAKER_01: And it's an arcane thing that is not working well.
- 23:58SPEAKER_01: Let me give you another example.
- 23:59SPEAKER_01: So from our paper, this is a real surprise to us.
- 24:02SPEAKER_01: So part of the paper just focuses on repaving.
- 24:05SPEAKER_01: So it's like kind of the simplest thing that is done in transportation.
- 24:10So I would have thought that when a state DOT puts out kind of
- 24:17a description of the
- 24:17SPEAKER_01: project with kind of the rules and regulations that need to be followed,
- 24:22SPEAKER_01: I would have thought it would be like a dozen pages,
- 24:25SPEAKER_01: something like that,
- 24:25SPEAKER_01: just like something not very long.
- 24:26SPEAKER_01: It's like you repave and like here's where you need to repave.
- 24:30SPEAKER_01: Here are a few rules.
- 24:32SPEAKER_01: These things are often hundreds of pages,
- 24:34SPEAKER_01: and it is the case that states that have longer,
- 24:38SPEAKER_01: more complex documents end up costing substantially more.
- 24:43So this suggests to me that states that have more complex
- 24:46procurement rules,
- 24:47SPEAKER_01: because there is variability across states,
- 24:49SPEAKER_01: end up having more expensive projects,
- 24:52SPEAKER_01: in part maybe by driving out competition,
- 24:55SPEAKER_01: in part by...
- 24:57just having more compliance that you need to do to actually
- 25:00figure out what to do.
- 25:00SPEAKER_01: So that's one set of things.
- 25:03SPEAKER_01: Unions, listen, I'm a liberal, support unions.
- 25:07SPEAKER_01: Unions are sometimes problematic.
- 25:10SPEAKER_01: So in the state of Connecticut,
- 25:12SPEAKER_01: if you want to hire someone and you put up a job posting,
- 25:16SPEAKER_01: you need to hire someone internally if that person meets the minimum job
- 25:22SPEAKER_01: qualifications,
- 25:23SPEAKER_01: even if there's someone outside of government who is way,
- 25:26SPEAKER_01: way,
- 25:26SPEAKER_01: way better.
- 25:27SPEAKER_01: and this is part of a union contract.
- 25:30SPEAKER_01: This is something that does not get much attention to the state,
- 25:32SPEAKER_01: and that's not great practice for having a high-quality workforce.
- 25:38SPEAKER_00: Now I'm just wondering,
- 25:39are there places in the private sector where that's a feature
- 25:41of the contract?
- 25:43SPEAKER_00: Because that's something I've heard about.
- 25:45SPEAKER_00: I know there was a GSA Inspector General report recently on exactly this,
- 25:50SPEAKER_00: where the Inspector General slammed the GSA for...
- 25:53trying to hire people better than the people internally who met
- 25:56that bar for trying
- 25:57SPEAKER_00: to look for better talent.
- 25:58And it was a,
- 25:59the IG came down hard and said,
- 26:00you,
- 26:01you should not be doing that.
- 26:02SPEAKER_00: You should not be trying to find the best people for the job.
- 26:04SPEAKER_00: It was like explicit from the inspector general,
- 26:06SPEAKER_00: like,
- 26:06SPEAKER_00: how dare you try and find people even better for this role?
- 26:09SPEAKER_00: It just seems like a,
- 26:11SPEAKER_00: like a government thing that I've never encountered elsewhere.
- 26:15SPEAKER_01: Like,
- 26:15SPEAKER_01: you know,
- 26:16sometimes you want government to operate differently in the
- 26:18private sector,
- 26:19SPEAKER_01: but I have never heard of this in the private sector,
- 26:21and it strikes me as good for neither the private sector nor
- 26:25the public sector.
- 26:26SPEAKER_00: Are there places where bureaucratic bloat is the problem?
- 26:30SPEAKER_00: We've talked a lot about undercapacity,
- 26:32SPEAKER_00: and I've been really,
- 26:33SPEAKER_00: I think I've been compelled by your work on this,
- 26:35SPEAKER_00: by others,
- 26:35SPEAKER_00: that in the specific realm of transportation,
- 26:37SPEAKER_00: there's expertise that
- 26:39people in government should have that they don't have for all
- 26:42these kind of
- 26:42SPEAKER_00: complex,
- 26:42SPEAKER_00: like long duration reasons.
- 26:44Are there places where you can say the problem is there are too
- 26:47many cooks in the
- 26:47SPEAKER_00: kitchen at the state DOT?
- 26:50SPEAKER_00: And if so,
- 26:50SPEAKER_00: I guess my follow up would be like,
- 26:52how can we distinguish between places where there's like too
- 26:55many people in this
- 26:56SPEAKER_00: process and not enough people in the process?
- 27:00SPEAKER_01: So I think two answers.
- 27:03SPEAKER_01: One is if bureaucratic bloat means rules,
- 27:06SPEAKER_01: too many rules,
- 27:07SPEAKER_01: like for sure,
- 27:08SPEAKER_01: too many state rules,
- 27:09SPEAKER_01: too many federal rules,
- 27:10SPEAKER_01: too much interaction,
- 27:11SPEAKER_01: or too much conflict between the rules.
- 27:14SPEAKER_01: Second,
- 27:15I do think that there is sometimes bureaucratic bloat just in
- 27:18terms of too many
- 27:18SPEAKER_01: cooks in the kitchen.
- 27:20SPEAKER_01: This takes a couple of forms.
- 27:22SPEAKER_01: One is when you have two different state agencies involved,
- 27:25SPEAKER_01: like you have the state DOT,
- 27:26SPEAKER_01: and you have the state environment department,
- 27:28and they don't agree because you're going to fill in a ditch to
- 27:33repair or build a
- 27:34SPEAKER_01: highway,
- 27:35SPEAKER_01: and that can be problematic.
- 27:38I want to protect the environment,
- 27:40but there are also needs to build things.
- 27:43Sometimes there's also bureaucratic bloat in the sense that you
- 27:47can have the state
- 27:47SPEAKER_01: layer and the federal layer.
- 27:50SPEAKER_01: And just to give one example of this,
- 27:53SPEAKER_01: there is a bridge that we're rebuilding,
- 27:57SPEAKER_01: a rail bridge in Connecticut.
- 27:59SPEAKER_01: It's going to cost about a billion dollars,
- 28:02SPEAKER_01: and it also costs a lot to operate because it is a movable bridge.
- 28:05SPEAKER_01: It goes up and down.
- 28:07SPEAKER_01: And remarkably,
- 28:09SPEAKER_01: this is a movable bridge because one boat,
- 28:12SPEAKER_01: one boat,
- 28:13SPEAKER_01: goes out once a year and goes in once a year,
- 28:17SPEAKER_01: so I am told.
- 28:18SPEAKER_00: Once a year.
- 28:19SPEAKER_01: One boat out once a year and one boat in once a year.
- 28:22SPEAKER_00: I'm really having trouble falling.
- 28:25SPEAKER_01: Yeah, so it sounds like I'm pulling your leg or lying or something.
- 28:31SPEAKER_01: As I understand it, this is the truth.
- 28:33SPEAKER_01: And the...
- 28:34SPEAKER_01: Army Corps of Engineers,
- 28:35SPEAKER_01: sorry,
- 28:36SPEAKER_01: the US Coast Guard,
- 28:38SPEAKER_01: federal level,
- 28:39SPEAKER_01: has apparently insisted that because of the one boat,
- 28:42SPEAKER_01: it needs to stay a navigable waterway,
- 28:44SPEAKER_01: and you need to have a movable bridge,
- 28:46SPEAKER_01: more expensive to build because it goes up and down,
- 28:49SPEAKER_01: trains have to slow down because the rail is no longer continuous,
- 28:51SPEAKER_01: you have to have someone operating it,
- 28:53to move it up and down much more expensive to operate that's
- 28:56going to break more
- 28:56because it's movable because of the one boat so some this is i
- 29:01guess political
- 29:02SPEAKER_01: scientists call this marble cake federalism where it's not clear who has
- 29:05responsibility and you have like multiple layers of bureaucracy
- 29:08and rules in this
- 29:08case it is not working out well we've got a couple grab bag
- 29:12questions one question
- 29:13SPEAKER_00: is i want to
- 29:14SPEAKER_00: take everything we've talked about here and then ask,
- 29:17SPEAKER_00: you were in the Biden White House,
- 29:18SPEAKER_00: you were at the OMB,
- 29:19SPEAKER_00: kind of a critical nerve center of the federal government,
- 29:22SPEAKER_00: of the executive branch specifically.
- 29:24Was there anything that you've presented here this afternoon or
- 29:28any of these themes
- 29:29that we've talked about that you had trouble convincing
- 29:31colleagues of or lenses
- 29:33that you felt like you had trouble getting across to other
- 29:36people in that building?
- 29:39SPEAKER_01: Yeah.
- 29:39SPEAKER_01: So there are lots of very,
- 29:42SPEAKER_01: very talented people,
- 29:43SPEAKER_01: a lot of people I really admire from the Biden White House.
- 29:48SPEAKER_01: Obviously,
- 29:49SPEAKER_01: we were not successful at bringing costs down,
- 29:53SPEAKER_01: cutting costs,
- 29:54SPEAKER_01: and I wish that had been different.
- 29:58SPEAKER_00: I'll leave it at that.
- 30:01SPEAKER_00: Well, let me ask you to give us all some directives.
- 30:04SPEAKER_00: We're both now at IFP.
- 30:05SPEAKER_00: I'm there full-time.
- 30:06SPEAKER_00: We
- 30:07SPEAKER_00: think a lot about tractability of policy ideas.
- 30:09We want to work on stuff that has a chance of happening and we
- 30:11don't want to work
- 30:12SPEAKER_00: on the other stuff.
- 30:13SPEAKER_00: If the goal of people in this room is to get transit costs down,
- 30:16SPEAKER_00: just think about the cost piece,
- 30:18SPEAKER_00: of everything we talk about,
- 30:19SPEAKER_00: what's tractable here?
- 30:20SPEAKER_00: What could we absolutely take a run at and make big progress on?
- 30:23SPEAKER_00: Because I think obviously some of these problems are highly conflictual.
- 30:28There are people who would like the system to exist as it is
- 30:31and it will just be harder.
- 30:32SPEAKER_00: Are there easy wins available that haven't been taken yet?
- 30:36SPEAKER_01: So I think that there are.
- 30:38SPEAKER_01: So for personnel,
- 30:41SPEAKER_01: especially in blue states,
- 30:43SPEAKER_01: which tend to have the big transit mega projects,
- 30:46we're talking about spending more money on state employees who
- 30:49are powerful
- 30:50SPEAKER_01: constituencies.
- 30:52SPEAKER_01: And if you believe my...
- 30:54SPEAKER_01: Highway results, they will often pay for themselves.
- 30:58SPEAKER_01: So it's a win-win,
- 30:59and one of the most powerful constituencies should support it
- 31:02because they get more
- 31:04SPEAKER_01: money.
- 31:05SPEAKER_01: In terms of procedures like procurement, these are kind of arcane things.
- 31:09SPEAKER_01: Sometimes the contractors may not like it.
- 31:11SPEAKER_01: I'm talking about
- 31:13SPEAKER_01: bringing in more competition,
- 31:14SPEAKER_01: driving down costs,
- 31:15SPEAKER_01: but you can reduce the rules they have to go through.
- 31:18This seems like the kind of thing that should not generate huge
- 31:20opposition.
- 31:20SPEAKER_01: It will not generate protests in the street to change procurement rules.
- 31:24SPEAKER_01: And something we have not talked about tons, just better data.
- 31:28SPEAKER_01: I think federal mandates,
- 31:30federal unfunded mandates would generate lots of opposition in
- 31:34state houses,
- 31:34SPEAKER_01: but...
- 31:35Just the tiny amount of funding that this would take to have
- 31:39much better data,
- 31:40to learn just huge amounts about the hundreds of billions of
- 31:45dollars a year that we
- 31:46SPEAKER_01: spend on this stuff.
- 31:47It's teeny tiny with,
- 31:50I think,
- 31:50upsides that are many,
- 31:51many,
- 31:51many times its costs.
- 31:54SPEAKER_00: Well,
- 31:54SPEAKER_00: that's my last question for you,
- 31:55SPEAKER_00: which is to the researchers,
- 31:57SPEAKER_00: to the people trying to better understand these problems,
- 32:00SPEAKER_00: what data could they be digging up that doesn't exist cleanly right now
- 32:05And what are the unanswered questions that people like you are
- 32:07working on that we
- 32:08SPEAKER_00: could just use more bodies and more minds on?
- 32:10SPEAKER_01: I have just the basic data, like what's every project cost?
- 32:14SPEAKER_01: What were the components of what it cost?
- 32:18SPEAKER_01: What was the timeline for those projects?
- 32:20Those are actually not,
- 32:22to my surprise,
- 32:24took much longer to write the project than I expected.
- 32:26SPEAKER_01: Not easily out there.
- 32:27SPEAKER_01: So those should be out there.
- 32:28SPEAKER_01: It's like public spending the money, the public should have the data.
- 32:32And then just kind of all sorts of things just have not been
- 32:35super well studied.
- 32:37SPEAKER_01: There are many, many kinds of ways you can procure.
- 32:39SPEAKER_01: There's no awesome paper on lowest cost bidding.
- 32:44SPEAKER_01: There's lots of case studies.
- 32:46SPEAKER_01: There's intuition that this just makes no sense.
- 32:48But studies where you can compare places that have one process
- 32:53versus another,
- 32:53SPEAKER_01: that would be super useful.
- 32:56SPEAKER_01: Be wonderful to have much, much more on personnel.
- 33:00SPEAKER_01: Permitting, I've written papers on permitting.
- 33:03SPEAKER_01: There's lots of good descriptive data out there.
- 33:04But in terms of the exact causal effects of which particular
- 33:08permitting regimes
- 33:09SPEAKER_01: cause what,
- 33:10SPEAKER_01: this paper does not exist.
- 33:12SPEAKER_01: Heidi Williams and I are trying to write it.
- 33:14Maybe it will exist at some point,
- 33:15but others should get to it before we do.
- 33:18SPEAKER_01: So I think this whole area is a hugely important way understudied.
- 33:23SPEAKER_01: I think that there is a huge amount of potential.
- 33:26SPEAKER_00: Fantastic.
- 33:27SPEAKER_00: On that encouraging note, Zach, thank you very much.
- 33:29SPEAKER_00: Thanks much.
- 33:29SPEAKER_00: Real pleasure.