Scott Wallsten:
Welcome to Two Think Minimum, the Technology Policy Institute’s podcast. I’m Scott Wallsten, President and Senior Fellow at TPI. I’m here with my co-host Sarah Oh, who is a Senior Fellow at TPI, and today we’re delighted to be talking with Jeff Prince. Jeff is Professor and Chair of Business Economics and Public Policy at the Kelley School of Business, Indiana University. He’s also the Harold A. Poling Chair in Strategic Management and Co-Director of the Institute for Business Analytics at Kelley.
He recently served as Chief Economist at the Federal Communications Commission. His primary research focus is on technology markets and telecommunications, having published works on dynamic demand for computers, internet adoption and usage, the inception of online-offline product competition, and much, much more. Jeff, thanks for joining us.
Jeff Prince:
It’s a pleasure. Thanks for having me.
Scott Wallsten:
So, you have a fairly recent paper that’s on the economics of digital platforms, and that seems to match well with what has been going on in the Senate lately, the actual things that are supposed to go on in the Senate, not the sort of more unfortunate things.
But Senator Amy Klobuchar, just today, introduced a bill to strengthen antitrust enforcement. The bill is called the Competition and Antitrust Law Enforcement Reform Act. It would do things like change the burden of proof on mergers, give some new tools to antitrust enforcement agencies, and generally take a harder line on antitrust issues.
Now, I don’t know that anybody’s had a chance to actually read the bill yet, certainly not me, and so I wouldn’t expect, I guess, you to either, but your paper provides some framework for how to think about these kinds of antitrust issues. So, it’s very timely. So, tell us a little bit about the paper and how it should help policymakers think about these issues.
Jeff Prince:
Yeah, thanks. We really enjoyed writing that paper. I think the basic idea that we try to lay out in that paper is to kind of start with some popular notions of what might seem like threatening, anti-trust type behavior. So, thinking about some of the older notions of “big is bad” kind of issues, if a firm has a large position in the market, that you’ve kind of got this inevitable market dominance and potentially bad behavior in terms of pricing and then ultimately welfare impacts, and so we try to start from that and kind of walk through what might be popular evidence of, you know, anti-trust concern and then shift into, “Let’s think about what are the key economic concepts that one might want to think through that would lead to higher prices, use of market power, those kinds of things?” It’s bringing some of the classic economics to some of the current concerns with digital platforms.
Scott Wallsten:
And so, what do you find? I mean, like you said, I don’t think the bill explicitly says, “big tech,” but I mean, that’s what everyone’s talking about, big tech and platforms and you just alluded to that. So, what does economics have to say about it, or sort of what evidence have you seen either way?
Jeff Prince:
I think one of the main things that we try to highlight is a classic problem with a lot of the digital platform discussions is that, you know, a lot of the things that we typically worry about in other scenarios may not apply. So, the classic thing would be, “Oh, they’re going to raise prices on consumers”, but a lot of the things we’re talking about with digital platforms are zero-price goods. So, it creates another level of complication when you’re trying to think about, “What does it mean for consumers to be harmed when they’re buying something that they get for free?”
And so, I think that’s one of the highlights from the paper is to try and take that issue head-on, and one of the ways we frame that is to think in terms of shadow prices. So, thinking about, you know, other product features other than price, and then bringing in some of the economic concepts that help us to think about, you know, is it the case that market power is….
Scott Wallsten:
Actually, before you go on, define shadow prices.
Jeff Prince:
Yeah. So, a shadow price is basically, there’s a technical way of thinking about it, but let me give you kind of the, I guess, more straightforward approach, which is the way I like to think about it, is imagine a product feature and the shadow price you can think of as you know, what’s the willingness to pay for an incremental improvement in that feature? And so, you can think about then if that improvement is not being made, then that’s in a sense, the price that the consumer is paying by not enjoying that improvement.
You know, a great framework for this would be like privacy. This has come up in the Facebook case and the Google case. There’s this argument that the privacy is suboptimal in some way, and in terms of what the firms are providing, and this shadow price framework, I think, helps solidify those ideas because then you can say, “Okay, what we’re really saying is the consumers are in a sense paying in the form of a higher shadow price because they would get non-trivial value from having more privacy, but it’s not being provided.”
Scott Wallsten:
So, what do you do with that?
Jeff Prince:
So, really, it’s a framework-type paper. So, we’re not, our intention is not to say, you know, here’s what should or shouldn’t be done on the platforms or any other, you know, related entity. It’s more trying to walk the reader through, “Here’s some of the stuff you’re seeing in terms of popular arguments. Here are some of the key economic concepts that would be pertinent when you’re actually trying to do an analysis in terms of what kind of harm might we be worried about here.” So, we’ll go through the classic things like economies of scale, economies of scope, switching costs, two-sided market issues, network effects, bringing in this concept of shadow prices, and then we shift into, you know, what are the potential remedies?
So, is it divestiture? Is it price regulation, including status quo, not doing anything and trying to lay out what are the costs and the benefits that at least should be weighed when considering different approaches?
Scott Wallsten:
One thing you talk about it in the paper is the role of multi-sided markets, and so I wonder what you think about that. Explain a little bit, but also, I wonder, have we put too much emphasis on that because almost anything could be a two-sided market or multi-sided market, right? Like even a supermarket is a multi-sided market, and so tell us how you incorporated that and how important think that actually is.
Jeff Prince:
That’s a really good point. We’re actually in the midst of putting together a special issue at the Journal of Economics and Management Strategy. And so you know, Dan Spulber?
Scott Wallsten:
I do.
Jeff Prince:
So, Dan and I… Dan was at Northwestern. I studied there, and he’s put on a bunch of conferences there that I’ve been involved in, and I think he’s one of the voices out there that’s really made this point about pretty much almost anything could be labeled as a two-sided market. When you go to the market, you could say the farmer’s market is the platform, right? And I’ve got the farmers and I’ve got the customers, right? And so, it’s like, there’s two sides of the market.
So, in a lot of ways you could pretty much frame anything that way. I guess, to your question about is it overblown? I’m sympathetic to that point. I think in some ways, yes, you could say that. Because exactly to your point, if almost anything could be labeled a two-sided market, it seems like why, all of a sudden, are we so worried about these things? I think the times when it starts to be something that is warranting some attention is, I mean, I guess everyone, at least most of your audience, is going to be familiar with the American Express case.
Scott Wallsten:
I wouldn’t necessarily assume that. So, why don’t you just briefly explain?
Jeff Prince:
It basically is getting at this idea that you’ve got two sides to the market, you’ve got the vendors and then you’ve got the cardholders. When you think about prices that the credit card charges to the vendor, you know, you also have to think about the fact that they provide benefits to the cardholders, right? And so, when you think about the ultimate effects of various straps the cardholders may employ, you should be taking kind of a holistic view in terms of all of the affected parties. So, there’s this asymmetry often in situations like that, where it might be optimal to charge one side and actually subsidize the other.
So, I think there are scenarios like that, where it is quite relevant because the nature of the market is such that those kinds of strategies can make sense, but is that the norm? If you think of this broad definition of two-sided markets, probably not, but I think there’s definitely a subset where, you know, when you think about concerns about anti-competitive behavior and things like that, I think it does make sense to make sure that you’re not just focused on one slice of it, but you’re actually considering the totality of what’s going on.
Scott Wallsten:
So, what’s the right way to do that? You’ve got the consumers, the people consumers, and then the advertisers. And so, I mean, how do you put those two together when you’re thinking about cases, remedies, analysis, anything?
Jeff Prince:
Yeah. I mean, when you start talking about advertisers, I guess now we’re thinking more in a world like a Facebook. In that case, I don’t want to dig deeper into the actual cases that were filed. I mean, I don’t want to presume too much knowledge on the Facebook and Google cases, but I guess when I think about just popular discussion on these matters, a lot of the conversation, at least in those cases, seems to center on consumers. Whereas when I think about the American Express case, you know, you might be saying that it was the vendors that were being charged perhaps too high a fee or whatever the charge might’ve been.
I don’t necessarily see the analogy here because, and this is just my personal take on this, but I don’t feel like a lot of the conversation is centering on all these poor advertisers having to pay these high rates. Now, I’m sure some of the advertisers would say that, and there’s certainly a subset that would make that claim. But it seems to me, a lot of discussion is centered on the consumer side where I think that’s where some of the stuff like Mike Baye and I, in our paper, I think has some relevant points because then it’s not so much about the two-sided market and one side that’s taking the raw end of the deal and the other side’s getting subsidized. It’s more that the side that presumably looks like they’re getting the good end of the deal, they’re getting the zero-price good, seems to be unsatisfied. Or at least the claim is that they’re somehow getting a raw deal, that has to then be on non-price dimensions.
And so two-sided market arguments can be at play here, but I think in this particular case, there seems to be a lot of emphasis on the one side and interestingly, the side that at least ostensibly, in terms of price, is getting the better deal.
Scott Wallsten:
That’s a really good point. So, the examples that policymakers bring up, you’re right, it’s always on the consumer side, whereas it does seem like the other side is where one might be able to make a more coherent case. Not saying whether or not it’s a winnable case or a true case, but at least you can think of how to define it. This one seems very strange.
Jeff Prince:
I completely agree. Mike and I discussed this many times when we were writing the paper, that kind of classic thing that economists and others would typically think of would be a concern for the advertisers, right? Because you’d say that’s where the price mechanism seems to be at play, and if these guys are exercising significant market power, then you would think the advertisers are the ones that are going to be the ones harmed in the most obvious way. But absolutely, this one takes on a different flavor because so much of the emphasis is about consumers.
Now, you know, there’s various reasons why that might be true. Of course, you could say the consumer base is much larger. You know, we’re talking about millions, even billions of people, when you talk about Facebook, if you go global. So, sometimes there’s a temptation to just focus on the larger group, and I understand that, and if you think they’re being harmed then you’re multiplying by a bigger end perhaps, but I’m not even sure that’s it. Again, that’s just my opinion, but I think there clearly is an emphasis on these non-price components, and I feel like that’s been taking center stage and these types of cases and several others where there’s been this increased emphasis on say quality dimensions or other kinds of non-price features as the area of concern.
Scott Wallsten:
I mean, cynically speaking, I would imagine it’s mostly to do with what sounds better politically because that’s the way it can be consumers against big business. Otherwise, it’s just business against business, and people care less about that.
Jeff Prince:
It’s hard to ignore that as a potential angle as well, for sure.
Sarah Oh:
Yeah, it reminds me of the Apple v. Pepper case, which is where the developers were at issue in the App Store model, and there isn’t really a popular uprising for the developers. The idea was like all the costs were being passed on to the consumers, or that was the argument, but they didn’t find it, but I agree. I mean, business against business doesn’t get headlines.
Jeff Prince:
Well, I think that’s right. I think that plays a role here for sure.
Scott Wallsten:
So, you’ve recently finished your time as the Chief Economist at the FCC. It would be good to talk about that a little bit. So, you recently wrote a paper with others at the FCC about economics, a review of economics that year at the FCC. What were some of the most interesting things you worked on while you were there?
Jeff Prince:
Yeah, so there was a range of projects that before I got there, they had just kind of closed out their commentary on the T-Mobile-Sprint merger, and so that had taken up a lot of the time and effort of the Commission to deal with that.
But then on the flip side, that meant there were a lot of projects to work on that kind of had been building up need, and so one of the things that I had focused on was, you know, some of the auctions that the FCC has been running.
One of them is for the 5G Fund. So, trying to get 5G connectivity in rural areas, and the other one also worked on there, the RDOF, the Rural Digital Opportunity Fund. So that, it’s really kind of the mirror image of the 5G Fund in some ways, although they’re structured differently. The RDOF is about broadband deployment.
Sarah Oh:
So, tell us about the 5G Fund and what that auction is supposed to look like.
Jeff Prince:
Yeah. So that one, what they were trying to do… So, they want to get funds out into kind of rural areas that generally the expectation is, it’s either they’re not going to get wireless service or it’s going to take a while for them to get, you know, advanced wireless service.
And so they have a set-aside amount of funds that they want to allocate to try and subsidize wireless firms to provide service, you know, in underserved areas. And so then the challenge is trying to think of, you know, how do you play the balancing act between having a significant impact from the funds? So, you know, there’s people in business that are going to benefit from it, from the fact that, you know, in some cases, there already would be the business case and so subsidies may not be needed.
And so they were developing and I was involved with this you know, weighting scheme, to try and balance out areas, regions for subsidy funds, trying to take that into consideration.
Scott Wallsten::
And so, I mean, they came up with this, with the weighting scheme, but there are still issues with the auction, right? I mean, there’s still some controversy about it.
Jeff Prince:
Yeah. So while I was there, you know, one of the pushback points was, you know, the quality of the data. So, you know, challenges in identifying, you know, where service currently was with 4G. So, trying to have a sense of, you know, what does the wireless map look like? And that’s an understandable concern, no doubt. It’s a challenge to get, you know, extremely accurate maps of these things, and wireless can be kind of tricky because it’s, it’s not as simple as, you know, like with broadband, you could say, you know, is there a connection or not, right?
With wireless, it’s kind of like, you know, you have to think about like, how often do you have a good signal, right? So what, you know, what is the strength of the signal? Those kinds of things. So, it’s a little bit of a grayer area when you think about what does connectivity mean, and so that was a pushback point, and my understanding is they’ve now decided to delay that to try and get, you know, more accurate data.
Scott Wallsten:
I mean, I’m also sympathetic to the data issue. When you’re talking about wireless, it’s sort of, it’s a little bit, you know, it’s a spectrum, not to make a pun there, but of, of how good your connection is and it’s sort of probabilistic in a sense.
But you know, I also wonder whether the kind of data that they’re looking for is, one, if whether it’s possible. And two, whether we didn’t already have in place ways to generate pretty accurate maps through the challenge process, they, you know, they use the best data that they have come up with geographic areas that they defined geographic areas and then companies can say, “No, I’m providing service there”, and challenge it.
Now, of course, you don’t want incumbents just to be able to say, “No, no, you know, we’re providing service there,” in order to block competition, but it is a way to know better, help identify where there is and where there is not coverage. And, you know, given that we’re talking, you know, inherently talking about places with scarce coverage, it’s hard to see how much better the data will be ultimately.
Jeff Prince:
It’s a really good question. I mean, I definitely got to see the inside of a lot of the logistical challenges that the Commission had to deal with in terms of navigating the challenge process. So, you know, they had done prior attempts to get information about mobile coverage. And you know, there were some really challenging logistics in navigating, you know, here’s what the firms say. Here’s what crowdsourced information is countering with, and you know, sorting through all of that was a non-trivial task.
So, it, you know, in some ways, I mean, it really actually calls for some careful economic analysis, because you could say, you know, there’s a real cost to, you know, trying to clarify the data, get more precision in the data. And then there’s also the time cost and delay to be able to do that versus, you know, the cost of not getting the funds out there and getting the deployment. So, I don’t want to pretend like I have the magic answer to that, but both sides of those things should be considered.
Scott Wallsten:
And so, another one of these, you pointed out a similar auction, although for fixed broadband, is the RDOF auction, which just recently concluded, but did you work on that at all?
Jeff Prince:
I did. I didn’t get as deep into some of the weighting schemes. So, as you know very well, you and I have done some work looking at the weighting scheme for RDOF in a prior iteration. And with RDOF, the difference there is the weighting is about trying to level the technology field, or I guess, account for technology differences. So, different technologies are going to have, you know, different bandwidth offerings, different latencies, different data allowances. And so the RDOF tries to weight those different offerings to try and kind of account for the differences in quality. And so, you know, they, they didn’t deviate too much from their prior iteration this last time around. So, I definitely was in conversations on that, but it didn’t have… it wasn’t as kind of novel as the 5G Fund was in terms of its weighting scheme.
Scott Wallsten:
Not to confuse issues by mixing up the two auctions, but maybe actually explain a little bit of how with the 5G fund, the weighting mechanism was different in the two auctions?
Jeff Prince:
Yeah. So, the 5G Fund, it’s not about weighting technologies, it’s about weighting regions. So, it’s saying, you know, I’m going to think about different regions that differ in terms of their demand condition.
So, the idea was recognizing different regions that are going to have different levels of attractiveness to firms. And that’s going to be based on two core ideas that on the one hand it’s, “What does demand look like? So, is this a place where you’d expect to have decent revenues?” You know, you expect there to be a demand for the services, and then, on the other hand, there’s the cost side. So, “Is the terrain costly to build out on,” those kinds of things.
And so, the combination of those two things would contribute to how attractive a region is, and so there was a weighting scheme that was devised to try and level out some of the inherent attractiveness differences across regions to try and make it so that it avoided the outcome, for example, that all the money just goes to the absolute most attractive areas that were much more likely to already get a business case for deployment anyway.
So, to contrast that the RDOF for the fixed broadband, that the weighting scheme there is about technology differentials. So, you know, allowing for the case that if you’re offering a higher quality product than somebody else, then your bid gets heavier weight in a sense.
Scott Wallsten:
So, I mean, I guess I don’t completely understand the reason for the 5G weighting. I mean, I can understand why there might be political pressure to weight regions differently, but it seems like adding weights to the regions is undoing some of the benefits of having an auction.
Jeff Prince:
That’s a very fair point. And it’s, you know, I guess I tread that one delicately just because there is, it’s hard to say, you know, I don’t, I can’t speak for all the forces at play that we’re driving, you know, why there was a weighting scheme in the first place? I think my fundamental, you know, the argument I had in my head was what I just described.
So, I mean, one could make the case that when they determine what the eligible areas are, that’s a very coarse decision. So, usually it’s like on some pretty basic like census cut or something like that. And then from there, I think it’s reasonable to at least say, well, there’s always a risk then that after you’ve made that very course cut when you allocate the funds, they’re all just going to go to places that are likely to get served anyway, and so we want to try and flatten that a little bit.
But I absolutely believe, you know, I can see that there were other forces that were at play in terms of what drove that decision, but I certainly was not in all the rooms that determined those outcomes.
Scott Wallsten:
Right, and I certainly don’t mean to criticize the FCC staff. They work under lots of pressures from all kinds of directions.
Jeff Prince:
They were great! They were great!
Scott Wallsten:
So, but, and then this question might not, I’m not sure if it makes sense or not, but why did the 5G Fund not wait for technology types?
I mean, it would’ve made it hard to call it the 5G Fund if you could use other technologies. But you know, in principle you could imagine a weighting scheme that, you know, weights a high end 5G service more heavily than an LTE service.
Or is it that, you know, at this point, that wouldn’t make any sense because the firms just aren’t going to build an LTE network, it would just be a waste. You know, I don’t know.
Jeff Prince:
I think there’s something to that last point, and I mean, maybe to broaden that a little bit, I think you kind of just alluded it, like even within 5G, you can make the case that not all 5G is created equal, right? And so, in principle, you could have at least considered, you know, some weighting scheme, according to the type of 5G technology that’s being rolled out.
You know, there, wasn’t a lot of conversation about that that I can recall. I think, you know, my rough memory on this is that, you know, a lot of it was driven by speed. I know the prior administration, you know, there was a lot of focus on we want to get this done quickly, because we need to get 5G out there quickly. So, I think that was a driving force behind that, but I absolutely agree that, you know, there still could be, and actually now that there’s been a delay, I mean, that, that is something that could be considered as you know, factoring in differences in the type of 5G technology that’s utilized.
Scott Wallsten:
So, on RDOF we don’t know a lot because the long forms haven’t been sent in yet, but what do you think of the outcome?
There is this, you know, sort of surprise that this fixed wireless company got the largest amount of money, and there’s some controversy about whether that firm can do what it promised. And so, you know, in the early days of spectrum auctions, there were plenty that didn’t go as planned, and it was a long learning process. I mean, is that possibly what might happen here or did this show that certain technologies can be used for things we didn’t expect?
Jeff Prince:
To be honest, Scott, I think it might be too early to tell, I guess I’m trying to take a holistic view. I mean, the FCC has been doing these auctions for what, over 20 years now? And I think there’s a lot of reasons to point to success, but I think, you know, in any given auction, you’re going to be able to point to, well for losses, you know, perhaps inefficiencies in terms of who won and the reliability of what they deliver and things like that. And, you know, potentially some gaming of the system and those kinds of things. But I guess to take a fair view, I guess I’m thinking about historically, I think it’s definitely been a big step forward that they’ve taken this approach.
So, I guess for me, the jury’s still out on these most recent auctions in terms of, you know, how well did they ultimately do, but I think if I’m putting them in the broader context of these auctions, I think, by and large, they’ve had a lot of pluses in terms of increasing efficiency and allocation.
Scott Wallsten:
I’m a big fan of reverse auctions. I started writing about them a long time ago. I’m a little concerned that if this auction turns out not to have gone well, groups that oppose the auctions will use it as a reason to try to push back against it when it’s a learning process.
Jeff Prince:
Yes.
Scott Wallsten:
Early spectrum auctions, like I said, not all of them went well. We learned from them and changed.
Jeff Prince:
And I think it needs to be highlighted that these are really complicated. I mean, it’s not like the models that we start with in undergrad and grad school. There are a lot of complicated details that had to be worked out to make these go, and so it’s unfair to point to any single auction and say, “Oh, here’s this thing that went wrong and that thing went wrong.” That’s always going to happen. I understand that. But I think the only fair way to assess them is in the broader picture, and like you said, I think it’s going to be a constant learning curve. Because there’s so much complexity to these large auctions, there needs to be learning, right? There’s just some things that the model’s not going to be able to tell you in advance. I think sometimes just have to learn from doing,
Scott Wallsten:
Did you work on the C-band auction at all?
Jeff Prince:
I definitely talked a lot about that around the office. I did get involved a little bit in designing the incentives for clearing. I was involved, but it was, at least from my perspective, that a lot of the issues were dominated by like legal concerns. The economics started to come into play when we were thinking about things like, you know, incentives for clearing and things like that. And just in general, you know, economics gives a nice framework to think about some of these “problems of the commons” kinds of things. But I think, you know, a lot of it was just, there was a lot of legal challenges as well. That seemed to be a dominant concern. What’s kind of in the legal “okay zone,” to make sure that we can get this done in a way that, you know, everybody’s going to be okay with it from the legal side of things, and I came to appreciate how challenging that can be in these complicated situations.
Scott Wallsten:
You were also there for the start of the Office of Economics and Analytics, or at least for the standing up process of it, and we spoke with Guilia and Wayne a couple of weeks ago about it. And so, you know, it seems like you’ve got this group of people really committed to making this work, to make economics have a bigger voice in it. Do you think it’s a sustainable idea? I mean, have there been times in the past, Well, even the FCC has tried this exact same thing. The office of… I guess it was the Office of Planning and Policy at the time was set up for similar reasons, or is this something that we kind of have to redo every some number of years?
Jeff Prince:
I mean, for their sake, I hope not.
Scott Wallsten:
I mean, they’re serious about it and they want it to work.
Jeff Prince:
Yeah, absolutely, and I don’t have kind of a before/after in terms of what might I be able to say in terms of the change in the culture and other administrative logistics that might’ve been clear to have changed. I can say, I thought it was going very well. I was very happy with the structure. I think it makes a lot of sense. I understand some of the counterarguments, right? You might worry that now we’ve got all the economists in one place, we can just ignore them all at once, that kind of thing. But I guess I feel like that was outweighed by, I really enjoyed the culture there.
It was great to have all the economists accessible, we’d have conversations all the time launch, just dropping into offices, talking about all kinds of different issues that were going on, and I felt like one of the things that was continually going on was just really trying to make sure that our voices were carrying up to the Eighth Floor, right? Where the chairman’s office was. I felt comfortable with that. I felt like we were being listened to. I feel like we were getting a seat at the table. I felt like what we were advising was often being even implemented. So, I think it’s great. Yeah.
And I think they were, to Guilia and Wayne’s great credit, they’ve really been working hard to put good processes in place, good standards in place to kind of solidify, you know, good product, good quality outcomes from the group. I just really commend them for that. I feel good, you know, with where it was going, and I really, really hope that it is something that sticks.
Scott Wallsten:
So, there’s always a risk that the incoming Chair will think of it as a creature of the former Chair and then therefore not like it. I mean, that’s always true, but internally, was there a sense that this was associated with Pai, or did people sort of see it as a structural change, a reform that was just inherently a good thing, or at least there were arguments for and arguments against that were unrelated to politics?
Jeff Prince:
I mean, obviously this is just my ad hoc take on it. I guess I never got the notion from my interactions that people saw this as the Chairman Pai experiment, right, and that, we’ll just see how this goes. I got more of the sense that this seems sensible and we’re going to really buy in and make this work. I’ve got to say, I mean, on the one hand being there for just a year, I think that that guy’s going to be kind of on the outside the whole time, but on the other hand, you can probably bring the, you know, “Hey, this guy is only here for a year. I can really air what I think,” and I guess, you know, to the extent that the latter has any truth to it, I didn’t really get any sense of anybody feeling like, “Oh, this is a mistake, or this really isn’t working.”
My general sentiment seemed to be pretty positive in terms of the vibes I was getting from the staff there. And I guess, you know, a related point to make is one of my big pushes while I was there was to try and really get the White Papers up and running. So, there was a tradition of that, and it had kind of fallen off in recent years, and I really saw that as being part and parcel to the office. One of the potential benefits of having the economists together more than they had been is that hopefully you get some collaboration, you get research ideas being discussed, and they have great people there. And they have a lot to offer in terms of the research that they can put out, and I’m really excited with where that’s going. I think they’ve got four papers that have come out in just the last couple of months, and so I think that could really also be a positive outcome from the office and, you know, something that could persist.
Scott Wallsten:
I agree. Yeah. The White Paper Series, I think historically has been really important. Let people think creatively. It took a long time, but it was incubated right there.
Jeff Prince:
Yeah. That’s how things get started. You gotta be willing to let people kind of let their ideas flow a bit. I think that’s really great. I hope that persists. In fact, like the first conversation I had with the Chairman, my number one thing was just to really kind of make sure we were on the same page that, “Hey, we want quality economics research coming out, and we don’t want the economists to feel like it always has to pass a litmus test. It’s basically we want quality work. As long as it’s relevant to what we do, if it’s quality work, it is going to see the light of day.” I wanted to do everything I could to foster that attitude because I think that creates a good culture amongst the economists. That’s what we generally want.
Scott Wallsten:
But it’s still, in some ways it’s kind of a strange institution anywhere because the Chair has his or her priorities and doesn’t like, [inaudible] people questioning them. I’m not saying this about Pai, Rosenworcel, or anybody. It’s just the nature of people. And you know, you have a working paper series and there’s always that risk. And then people come back and say, “the FCC said…,” rather than saying, “it’s a working paper.”
We did know that it went away for a while and now it’s come back. I’m so surprised it’s as stable as it is. I would think that there’s often pressure for it to go a way because it’s inconvenient.
Jeff Prince:
No, I hear you. I think that’s one of the hopes is the Chairman can really set the tone on that. I think that’s why that was so important to me to make sure that, at least in the case of Chairman Pai, that we were on the same page, and I suspected we were. Because I mean, he had really sent some positive signals about that, you know, and that’s one of the things working with Guilia and Wayne. One of the main things I was working on was to put procedures in place with regard to White Papers, to try and set the precedent. So, the hope is whoever the next chairperson is, if it’s Rosenworcel continuing or someone else, that the way we structured that is that that person can kind of buy into it in principle. Because I think the economists need to feel like there’s a reasonable process in place.
I think everyone understands there’s always going to be challenges with what the Commission has to deal with. The Chairman always has to have a range of things in mind in terms of what’s going on at the Commission, but at the same time, have a reasonable process so that say, I’ve got a research idea. I want some feedback on that. I want to know that it’s going to be able to be done. I want to know that the Commission is willing to support this kind of work. I’m very hopeful that we’ve put something in place that can last.
Scott Wallsten:
I’ll be sure I’ll probably move on from the Chief Economist’s discussion, except I want to ask one more thing. Probably a lot of our listeners know what the position is, but it’s a bit of a weird position, right? Because it’s somebody… you kind of parachute in for a year and then you leave. So, what’s that like?
How does the staff feel about that? Because people who haven’t worked with the FCC probably don’t know how great they really are. They know their stuff, and so how did they feel about somebody just coming in right in the middle of all of their work, and what are the benefits and costs of this setup?
Jeff Prince:
I can’t speak in general about how they feel about it. I guess I will stick my neck out and say, you know, one of the vibes I got is a lot of how they feel about it kind of depends on the attitude that the Chief Economist brings in arriving, right?
You know, the way I’ve described it to other people is it’s kind of like a free safety position, which I love. You’re not really burdened with a lot of administrative tasks. It’s more like we’ve got a lot going on here. We know that you have expertise in a lot of these areas. Let’s talk and find ways to get you involved in what we’re doing, and they really do listen to the Chief Economist. I mean, I felt like a lot of the things I brought got implemented or heard, I was really grateful for that. I thought that was fantastic.
A lot of it’s the attitude of the Chief Economists. I think if you walk in thinking, everyone’s going to listen to me and all the policies… the chairman’s going to do what I’d say and all that, then you’re in for a disappointing year. But I think if you come in and say, you know, I want to contribute to the culture of economics in this place, bring high-quality assessment and analysis and really get involved and be supportive of what’s going on there, I think they will receive you very well, and I just could not have been happier with the experience.
Scott Wallsten:
So, I guess the last topic, another new paper that you have, which I was surprised to find online was this paper on COVID and travel restrictions, which I guess just came out, really just totally fascinating. Tell us a little bit about that.
Jeff Prince:
So, my colleague, Daniel Simon, and I, we’ve done quite a bit of research on airlines over the years, and full credit to Daniel. He’s the one that kind of brought this up originally. It was basically this idea that, “Hey, you know, we’ve done a lot of analysis on airline demand and competition and these kinds of things.” In the context of COVID, right, like early on, pretty much anywhere that where the virus ultimately took hold, it’s pretty safe to say it was largely from international travel that it began, and so, you know, our original idea was to basically try and look at various factors that likely would influence cases and deaths and try to measure their relative impacts. We kind of broke things into like exposure versus environment. Whereas exposure is clearly international travel, environment is, you know, population density, those kinds of things, health demographics.
But then in conducting our analysis, we recognized early on that the primary likely contributors in terms of exposure would have been travel from China and from Italy. That was just obvious in the early months, and so we were able to then look at, you know, what were the flows of people originating from China into the United States. People originating from Italy into the United States, across various counties across the country, and the interesting thing we found, and it was just robust, as you can imagine, that the influx of people from China didn’t seem to really move the needle. Whereas the influx of people from Italy notably did, and that’s perhaps not that inconsistent with what we saw.
We saw at least a significant shutdown in travel from China in late January, whereas Europe continued through the middle of March, but despite that, we kind of knew there was something going on in Europe well before March, and so one thing you could take away from that is perhaps China’s shutdown may have had some impacts early on. And unfortunately, we probably would have benefited from a European shutdown sooner, because it seems like the passenger flow from Italy did have a significant impact.
Scott Wallsten:
So, what’s your advice to the Biden people, when thinking about new mutations of the virus when they’re identified in particular places.
Jeff Prince:
We’re economists. We’re always reluctant to give hard advice. We’re more like, you know, think about pluses and minuses, but I guess the takeaways we’ve kind of highlighted are, you know, to the extent that you think travel shutdowns might make sense, timeliness is of the essence. So, if you’re going to do it, do it quickly, but then also think broadly, if you think a shutdown is worth it, and then you have reason to believe that a strain is now non-trivially in a certain country, then that should be on your radar as a place that you might want to shut down as well, and so I think contrasting the current situation with the earlier one, the case would be that there was reason to strongly believe that this thing was non-trivial in Italy well before mid-March, and so thinking in these terms might have saved us some of the earlier pain in the virus, but you know, again, hindsight is 20/20.
So, I don’t, I’m not interested in beating up on everybody after the fact, I know these were all very difficult decisions, but you know, our job is to try and learn and do better next time.
Scott Wallsten:
Right, yeah, no, hopefully there were lessons. Clearly, we had a lot to learn at the time.
Sarah Oh:
Well, one more paper, maybe Scott doesn’t want to bring it up because he helped co-author it. But Jeff, you co-authored an op-ed on polling and suggested using discrete choice methods. Can you talk a little bit about that?
Jeff Prince:
So, Scott and I have a recent paper looking at using these discrete choice experiments to measure how people value different types of privacy. I’ve got a range of work, Scott and I also have prior work, trying to measure how people value bandwidth versus latency in their broadband connections. So, there’s a lot of applications for these types of analysis. A lot of times people associate them with marketing, but it goes well beyond that. So, one thing Scott and I have been talking about and put out in this recent blog is thinking about their potential use in trying to assess people’s political preferences. To give you a basic idea of what we’re talking about, it’s in a sense, trying to replicate how people might shop for different products or services where you would describe here’s “Product A.” It costs $5, it’s red and it weighs five pounds, right?
And then here’s another one that’s $10, it’s blue and it weighs eight pounds, right? And so, by seeing how people make choices across products described along those dimensions, you can assess how they value different dimensions. How do they make trade-offs when they decide which product they like over another?
That doesn’t have to be limited to the products. You could be thinking about that in terms of political policies or candidates, and so one of the things that we think there’s potential in these types of analysis is when you frame it along multiple dimensions like that, you can kind of dilute people’s attention on a particular feature. So, if you think about, if I just say, who are you going to vote for? Then obviously I’m focusing you on a particular election, right? On a particular potential set of candidates. Whereas if I frame it in the context of, “Here’s some policies you might think about. You know, here’s a combination of policies and candidates, and amongst those combinations, which ones do you like the most?” Then, I’m not putting particular emphasis on any given election or any given policy issue, but you are still making trade-offs and you ultimately then have to think about, “what are your preferences,” in order to make those decisions.
And so, the hope is that framing it that way you might be able to elicit people’s preferences on maybe more sensitive topics in a better way than if you just directly ask them about it. So, that’s something we’re thinking about pursuing.
Sarah Oh:
Do polling firms not use those types of methods right now?
Jeff Prince:
It’s not common. So, to Scott’s credit, he found one application, I believe it was in Scandinavia where they did something like this. But my recollection is it wasn’t necessarily tailored for trying to get preferences on say, ostensibly sensitive issues. It was just more kind of applying it to the political sphere to see if it can give us reliable information about what we might think people would vote for say, in a given election, but I think it’s a nice starting point for us because at least, you know, something like this has been tried in the field, it seemed to give reasonable results.
But I think our hope is to try and push it to specific applications where you might say, this is a sensitive issue or this polarizing candidate, you know, those kinds of applications, and see if this type of approach might be able to address some of the concerns that people have had in some recent elections where at least part of the concern was that people might be a little more reluctant to voice their true opinion on say, a polarizing candidate or perhaps a controversial issue.
Scott Wallsten:
You raised another point that we were actually talking about this morning. Why do polling firms continue to use methods that we know are not likely to elicit truthful responses? You think they all have an incentive to try to be the one that’s the best, you know, whether you’re an external polling firm or polling for a candidate internally? So that’s another, it’s an interesting question. We don’t know, but we are looking to start this project. Our donation lines are open. *Laughs*
Jeff Prince:
*Laughs* We’ll post that number! *Laughs*
Yeah. But it should be fun. I think it’s a worthwhile project. I mean, obviously, in recent applications, obvious things come to mind, but I think this has wide-ranging implications. There’s lots of controversial things, you know, there’s lots of things that people don’t necessarily want to be truthful about in terms of their preferences, where deeper understanding of what their preferences truly are could be very beneficial towards policy and business strategy and all these kinds of things. So, anything we can do to move the ball forward in that I think could have some real value.
Scott Wallsten:
Well, that was a pretty good cross-section of your work. We’ve gone way over time. So, we should probably wrap up. So, Jeff, thanks so much for speaking with us today. It was really interesting.
Jeff Prince:
Thank you both. I really enjoyed it.
Scott Wallsten:
Thanks.
