Examining Programmatic Political Advertising in the United States

Audio of this conversation is available via your favorite podcast service.

As the U.S. midterm elections approach next week, there is a renewed focus on understanding the spending on and claims made in political advertising in digital channels, particularly on social media. For instance, a report from Global Witness and the Cybersecurity for Democracy team at NYU’s Tandon School of Engineering, where I teach, found Facebook and TikTok failed to block ads with blatant misinformation about how and when to vote, with TikTok failing to catch 90 percent of such ads. YouTube fared much better.

But what is going on across the web, beyond the social media platforms? A recent report from the University of North Carolina at Chapel Hill Center on Technology Policy found that as a result of restrictions on political ads instituted by major platforms ahead of the 2020 elections, political advertisers are increasingly turning to political advertising on other platforms. Programmatic advertising accounts for a substantial and increasing share of political advertising, they say, and more attention needs to be paid to this complex and confusing ecosystem of companies- large and small- that serve up ads on websites, apps, streaming services, and other digitally connected devices. This episode features a discussion with the report's authors:

• J. Scott Babwah Brennen, head of online expression policy at the Center on Technology Policy at UNC-Chapel Hill; and
• Matt Perault, director of the Center on Technology Policy at UNC-Chapel Hill, a professor of the practice at UNC’s School of Information and Library Science, and a consultant on technology policy issues at Open Water Strategies.

Source

What follows is a lightly edited transcript of the discussion.

Justin Hendrix:

The two of you are about a month out from having published a study that looks at political advertising on the internet, but I think takes an interesting perspective on it. You point out that Meta, Google, Twitter, when most folks are talking about political advertising, they're talking about the big platforms, what happens there, what happens on social media, what happens in search, et cetera. But you've focused instead on the programmatic advertising market. What, I guess, spurred you on to look at this particular segment of the market?

J. Scott Babwah Brennen:

Yeah, it's kind of a funny story. We worked on a series of projects last year that looked more at political advertising on those major platforms and focused on the political advertising restrictions or blackouts in the 2020 election. It was really working on those projects led us to realize that there's basically this whole other world of political advertising out there that no one is really talking about, no one in our world. Of course, there are people working in that, but the policy community and the academic research side have given it far less attention.

In fact, one of the things we found in our work last year was that as a response to some of those limitations and restrictions that the major platforms were placing on political ads in the 2020 election, and now they're here in the midterms as well, political advertisers are in fact moving to these alternate platforms, and so we see them growing in prominence. And so we thought it'd be really important to understand what sort of policies these alternative platforms have as it relates to political advertising, how they treat political ads, what political ads look like on these alternative platforms.

Justin Hendrix:

I understand that you've taken an ethnographic plus secondary research approach to this. You've interviewed folks who are players in the market, people who are expert in this area. But just for the listener, give them a sense of scale. You point out that somewhere between a quarter to almost half of spending on political advertising went to platforms other than Meta, Google and Twitter. What kind of dollars are we talking about, and how important is this to the overall political spending in the US?

J. Scott Babwah Brennen:

Yeah, that's a great question, and I wish I had more concrete numbers to offer you. That range you said, a quarter to a half is huge, and that range represents billions of dollars. Unfortunately, we don't know. There's just very little data about this sort of spending, and those numbers are estimates from other organizations. The good, reliable data that we have, for example, from the FEC, which collects spending data from political committees, this is actually something we wrote a lot about last year. Thanks to this one big loophole in the data that are required to be collected, basically, we don't know how campaigns are actually spending money.

And what it is, is well, campaigns or committees have to report how they spend money. The consultancies that they hire to place ads don't have to report that. And so basically, we know that committees are maybe spending a ton of money, putting a ton of money to political ad consultancies, but we have no idea if that money is going to Facebook or to Xandr, for example. But we do know that it is billions of dollars every term, every election cycle.

Matt Perault:

It's hard to estimate with specificity, which is why we provided such a wide range here. But the point we were trying to make is there's been a lot of focus for understandable reasons about how platforms like Meta and Google and Snap and Reddit and Twitter and other companies shape political discourse, and that makes a ton of sense. What we were trying to say is, "Well, what does the rest of the market look like and how do those platforms potentially shape political discourse?" And so whether it's 25% or 50%, or even if it were as low as 10%, we think it's important to understand what the speech practices are on those platforms.

Justin Hendrix:

We're going to get into some of the specifics of your findings and also your recommendations to multiple players in this particular ecosystem. But before we do that, can we just talk a little bit about the context here? How did we end up in a situation where no one knows how much is being spent? Is this just Citizens United, American political spending, PACs run amuck? No one knows. It's the wild West. You point out in the report that there are few federal or state regulations that govern programmatic political advertising. Is it just that there's no sheriff?

J. Scott Babwah Brennen:

Yes and no. It's not really Citizens United that is the problem here, although it is a problem elsewhere, of course. Yeah, the big issue for lack of understanding is really what I said, this issue about sub-vendor reporting requirements, which is this really niche, arcane regulation. You're right that there are very few regulations that govern really any political advertising. Basically, there's two forms, disclaimers and disclosures.

Disclaimer is the required text at the bottom of an ad, for example, that says, "This ad was paid for by so and so," and/or "So and so approves this ad," like a TV commercial spin. And then disclosures are the information that's required to be submitted to the government. But I think as we mentioned in the report, the situation is basically there's existing regulations don't apply to all types of digital ads. They apply to some digital ads, but not all digital ads. And so there are some digital ads that have no requirements to have disclosures or disclaimers.

Matt Perault:

I think obviously more federal law here would be helpful. And the thing that plagues federal law, this federal law is the thing that plagues all federal lawmaking, which is Congress is unable to pass legislation on a variety of different issues that are fundamental to our discourse and democracy in life, including the regulation of elections. And so we point out a few areas here where we think congressional action would be particularly meaningful.

The second, of course, is the Federal Election Commission, which was sort of neutered on its birth by having a commission that has six commissioners. And there's a reason a Supreme Court's not eight justices. Typically, if you have an even number and you are assigning commissioners based on party with a split down the middle, that sets it up to be incredibly ineffective. So I think the Federal Election Commission has been slow to pass rules in a whole bunch of different areas, including thinking about how online platforms would shape political discourse.

Justin Hendrix:

And when I mentioned Citizens United, I just meant it in the broader context of there are billions and billions pumped into political advertising in the United States, and it would almost seem natural that you would know or be able to spot exactly what's going on in the programmatic advertising market as a result of that scale.

Matt Perault:

I think it maybe is worth just giving you a little bit of a sense of how we came to this. Scott talked about the arc of our research agenda. I guess I should be telling you that was planned in a strategic from the start, but it really has been more that we've backed into different areas where we think there's value in doing a report, because we've asked a question that seems really basic on its surface and found that there's really not an answer. As Scott is alluding to, we wrote this report on methodology in darker areas, more opaque areas of the advertising ecosystem.

We didn't set out to write a report on a methodology. We initially said, "There were political ad blackouts in the 2020 election. Did those work or did they fail?" And when we started to get into trying to answer that question, it ended up being really hard to do just because of the holes in the data. And so we ended up publishing initially a report on methodology as an initial step on this broader effort to understand efficacy. Then when we understood something about efficacy of ad blackouts, we thought in the course of writing that report, we were thinking, "Well, the relevant data isn't just what advertising spending was on Meta and Google," for instance, because those were the platforms where the blackout occurs, and Twitter, which shut down all political advertising prior to the 2020 cycle.

Then we were trying to understand, well, where do the ad dollars go? Then when you ask that question, there are all these challenges methodologically. And then also, it points to this ecosystem that we think has probably absorbed some of the advertising spend that has left these other platforms. There's a backdrop to how the research has unfolded and the dialogue between me and Scott on these issues has unfolded, that is driven in part just by where you can see data and where it's harder to see it.

Justin Hendrix:

You asked two key questions in the report, and I'll state them here. What policies do DSPs, SSPs, ad exchanges have regarding political advertising in the US, and how do these policies compare to the policies in the major tech platforms? Maybe to your point there, Matt, thinking about them in relation to the major tech platforms, but as something separate, a separate and more difficult to discern ecosystem. And then two, to what extent do programmatic political ads comply with existing laws?

Let's start with number one. When you looked under the hood of the programmatic ecosystem, which everyone has seen, I'm sure, a LUMAscape or some other confusing graphic that points out how many hundreds or thousands of different entities are involved in these landscapes, what did you find? Or is it even possible really, to sum up the quote policies of that ecosystem?

J. Scott Babwah Brennen:

Yeah, it is difficult to sum that up, but I think we can say a few things big picture. First, I should say our findings are based on, as you said, from some interviews with people in the space, but also a systematic review of the policies at 61 different companies, which we identified as the major companies in this space across those different categories; DSPs, SSPs, ad exchanges. I think Matt has said repeatedly, that likes to sum it up that the ad policies, broadly speaking, look like the major social media platforms looked maybe 10 years ago. That on the whole, the major platforms now have very detailed content policies and political ad policies or ad policies.

Most of these companies do not. They have brief and not particularly specific policies that holds across most of the big categories, whether it's about the policies about the content that is permitted, whether it's policies about the transparency requirements that they have, or about even the disclaimers that are required. Now, I think the caveat to that is a lot of these companies when they do have policies about ad content, tend to write them in ways that are super broad.

While, for example, Meta has very detailed rules about the types of false content that it doesn't allow or does allow, often if these companies have anything to say about false content, for example, they just say, "We do not allow false content," something that broad and that general, often with very little insight into how they're actually defining that, how they're actually enforcing that. And what that ultimately means is these companies probably are reserving a great deal of latitude to make those decisions on the fly. Yeah. Matt, do you want to add anything?

Matt Perault:

I guess I would just, to lift the veil a little bit on our research process, one thing that was really hard for us was figuring out, thinking through how to characterize that, so we tried not to do it in the report. But obviously, there are lots of people who would say more developed policies here would be really helpful. You'd want to see ad archives. You'd want see more specific approaches to misinformation. You'd want to see more detail on enforcement. And I think Scott and I have slightly different views on that, so we can talk about that in a little more detail.

But what we really tried to emphasize here was primarily the descriptive element, just that if you think that the direction that large platforms have moved in is really good, then you would have more concerns, I think, about the remainder of the market. If you think the direction they've moved in is really problematic and it should be a little bit more in the stage that things looked, where things looked, how things looked 10 years ago, then you might actually think the rest of the market looks pretty good.

The one thing that, I think, that we did state about why the practices of these companies that we focused on might be problematic is just from a user expectation standpoint. It's difficult to understand. I think it's difficult for an advertiser to understand what's going to be permitted and what's going to be prohibited, and how is my speech going to be treated? We think that on balance is probably negative for political speech, just because advertisers, we think should have clear expectations about what the guidance is on speech before they start advertising.

J. Scott Babwah Brennen:

Well, and I guess a couple things on that, too. There's also a way in which... The programmatic ad space is super complicated, and placing a programmatic ad involves your content being passed amongst many different companies, or at least a couple different companies. I think for me, one of the most surprising and notable things we found was we have people say, "We don't understand what the policies are at the other companies we have to work with." And so while it's important for a DSP to preserve their relationship with SSPs, they didn't know what was acceptable and what was not. And so they said, "We just have to experiment, and sometimes ads would be kicked back and we don't really know what's why that is." There's a way in which just more clarity about what the rules are could help the efficiency of the market as well.

The other thing I want to say about in response to your question, a number of the companies that we studied subscribe, ascribe to the Digital Advertising Alliance, which is a trade organization, has self-regulatory principles of transparency and accountability that are applied to political ads. What that means is basically, this one trade organization is the only trade level regulations about political ads. The interesting thing we found about this is, one, these sub-regulations are incredibly minimal. They basically are just disclaimers that a state who paid for the ad and things like an address.

But more interestingly, the way they're written is basically to further reduce the ads that are required to have them. First of all, the DAA only applies to political advertisers, which is funny, because the members of the DAA are not political advertisers. They're political advertising companies. And so that allows those companies to have the wiggle room to say, Well, none of this applies to us because we're not political advertisers or political advertising platforms." Other members might say, "Well, we're the advertising platforms, but we have a responsibility to enforce these rules for the ads that we carry."

The other thing is these rules from the DAA only apply to a very small set of political ads that contain express advocacy. Those are very clear statements about voting, that you should vote for a particular candidate or against a particular candidate, and that are only at the federal or statewide level. Statewide is governor or lieutenant governor. It does not include at the state district level where there are, of course, a lot of ads as well. So we see the one place that tries to offer some regulations, some guidance on political ads, the rules are written in a way that allows further incredible amount of latitude.

Justin Hendrix:

Let's talk about, I guess, your second question. To what extent did programmatic political ads comply with existing laws? Not their own self-regulation or industry self-regulation, but rather the laws of the land, whether they're federal or state. What did you find when you looked at this question?

J. Scott Babwah Brennen:

I should say at the start, we're really limited in our ability to assess this, mainly because of all of the companies that we looked at, only one has an archive, anything like an archive of ad creative. That company is Xandr. This is the only set of ads that we had that we could use to assess the requirements. The issue here is that Xandr's archive is not meant to be comprehensive of all of the political ads that they serve. Rather, they have this archive, as far as we understand it, to make compliance with state laws in just a couple of states that require there to be additional transparency. And so we basically only have incredibly limited examples of political ads, but for the states, for particular states. And so of course, that makes answering this question about federal laws not particularly doable.

Justin Hendrix:

So the answer is you just can't know or don't know, right? I guess, because from my point of view, listening to you describe it, you could be describing an area where there's a massive amount of illegal activity taking place and we wouldn't know. We have very little insight into that.

J. Scott Babwah Brennen:

I think that, to me, that's the stakes that we're dealing with here, is that I strongly believe that we need better transparency. We need better understanding of what the political ads in this space look like, if only to better understand if they are complying with the minimal federal rules and state rules. There are state rules on political ads as well, but if only to understand if those rules are being followed. We don't have that right now. And I know we'll talk about this in a minute, but that leads to one of the recommendations we have, is how might we gain better understanding of what political ads look like in the programmatic space.

Justin Hendrix:

Matt, based on your survey and look at this, there are various anecdotal examples in the report of different units of advertising, messages that people have seen. Do you suspect there's a lot of foul play going on? Are you concerned about that, or do you see this more as an area where more, I suppose, scrutiny, transparency may allay a concern of someone like me who's clearly concerned about it?

Matt Perault:

I think even asking the question, "Is there likely to be a lot of foul play here?", in some ways that almost is, I think, narrower than the question that Scott and I are asking, because we're looking at, what should sensible rules here be? There are lots of rules that should exist that don't. One example, I think, of why it might be hard to answer your question is there's no federal prohibition on voter suppression. And so it is permissible under federal law, or it's not illegal to run ads that suppress the vote. That's problematic in our view, and we think there should be a federal law that prohibits that.

You could run ads that suppress the vote that mislead people about the time and location of an election and not be in violation of law. That's not illegal activity, but it's problematic. I think from our standpoint, the macro issue is just there's so much that people don't know, that we don't know about what's happening in this space, and we're not on a path to learning more about it. I think that's the really distressing thing from our standpoint. And so as a result, we're rerunning over and over again a similar playbook with very limited evidence of whether it works or fails. And that, I think, is what is most distressing to us.

We wrote a report suggesting, as far as we could tell, that it's likely that ad blackouts are counterproductive, and yet Meta announced that going into this election, it's going to run the same playbook that it ran from 2020. And maybe that is the right playbook. Maybe there's evidence that those are the right solutions, but our effort to understand that question suggested that those aren't the right solutions. And yet I think because there's a high-level political conversation here that creates what's politically feasible, we are engaging in the same cycle over and over again.

I think the right question is a little bit less, "Is there nefarious activity?" and more, "Do we have a shared understanding?" This sounds too academic maybe, but "Do we have a shared understanding of what the norms are around how you should speak online, and then do we have a system for ensuring that those norms can be realized in practice?" And our finding is that we don't.

Justin Hendrix:

And I guess just to underscore this or to tie it up, on your Q2, if we were listing out your hypotheses or hypothetical questions, "To what extent do programmatic political ads comply with existing laws?", it sounds like the answer is just simply, "We don't know for the vast majority of players in the ecosystem."

Matt Perault:

Yeah. I think we don't know, and we would also suggest platforms are not making enough information public to know the answer to that. I don't want to overgeneralize, but I think typically when that's the case, it probably means that there are violations that are occurring. I think it typically is the case that specificity and transparency suggests an enforcement regime behind the scenes that's somewhat more robust. That's probably overgeneralizing, but I'd be more skeptical given what we know, rather than more optimistic.

Justin Hendrix:

Well, you're maybe not far off then from where I am, because I would assume if there's, again, not a sheriff on some level that can see the activity and understand it transparently, if civil society can't keep an eye on it, et cetera, if journalists, et cetera, that probably there are problematic behaviors going on somewhere along the way.

J. Scott Babwah Brennen:

I should just say, and I'm totally blanking on the name of it, but Joan Donovan, super well-known Harvard researcher in the misinformation space put out a report in, I want to say it was 2018, maybe partnering with that Data & Society that did find notable examples of worrying misinformation in the programmatic space. That wasn't the only report to do so, so there certainly is some indication that there is worrying content.

Justin Hendrix:

And of course, there's an activism community around some of that as well, things like Check My Ads Institute.

J. Scott Babwah Brennen:

Exactly. Yeah.

Justin Hendrix:

But let's get into your recommendations a little bit, because you point out three groups of parties that essentially could behave differently. One is this Digital Advertising Alliance that you've already mentioned. Then, of course, there are the companies themselves and of course, then policymakers at the both federal and the state level. I don't know quite where to start here, but I suppose we won't go through all dozen recommendations, but maybe we should go one by one through these different parties, and perhaps there are a couple of those recommendations that you might highlight that you think are most important. Perhaps we'll start with the companies.

J. Scott Babwah Brennen:

Most importantly, we just recommend that companies just have clear policies and make them public. And we want these policies to cover a wide range of things, including content disclosures, targeting, transparency, and accountability, the same sort of things that political ad policies at the platforms cover. Aside from what those policies should be, we just want them to have policies and make them clear. It seems honestly like a pretty low bar.

I think another one to follow up on that is that they should have clear policies about enforcement. Policies that are not enforced are not particularly meaningful. And in fact, that's been the lesson in terms of content moderation, that companies may have detailed rules, but if those rules are not enforced in a consistent and meaningful way, then they're not particularly meaningful. So we want them to have rules and we want those rules to be enforced.

And then the other big one that we say, and which actually is a similar recommendation across the three types of organizations we talk about is we recommend that, as Matt was talking about, that programmatic companies prohibit ads intended to suppress voting, such as supplying false information regarding things like voting location, date, process, or ID requirements. Our focus here was, as Matt said, a little bit more descriptive, but we ultimately concluded that here's a type of content that we think has no value, public value.

Despite there not being federal prohibitions on this type of content, we think that we feel very confident in saying that it would be better off if companies prohibited that. The same is true for the DAA. We have recommendations about the DAA's rules. We say the same thing. The DAA should recommend that its member companies also prohibit this sort of content, and then as we have done in other things that we've written, recommend that the federal government pass laws that prohibit this type of content.

Justin Hendrix:

Did the DAA or any of the 61 companies that you looked at for the report respond in any way to these recommendations, or have you got any indication that they may take them on?

J. Scott Babwah Brennen:

We have had a couple of conversations with particular companies about our recommendations. No. But as far as to take them on, I don't know.

Justin Hendrix:

Well, let's go to policymakers, because of course, barring the companies or perhaps this Advertising Alliance voluntarily doing the right thing with some of the recommendations that you make, of course, lawmakers could perhaps take some action. I laugh, only because for some reason, it seems like they haven't over the last few years. But what do you think Congress should do? What should states do?

J. Scott Babwah Brennen:

Well, we do call for the creation of a national archive of digital and really non-digital political ads for federal office, and we would recommend that it is maintained by the FEC. We can see a way that it could be piggybacked onto the existing FEC filing requirements, disclosure requirements. But the idea here is that some of the big platforms have these archives now, these Meta, Google, for political ads that they run or social issue ads that they run, which I'm really glad we have those.

But one hand, it's not a particularly workable situation to have every single company maintain their own database or their own archive of political ads. Not workable for small companies that it might be really cost-prohibitive, cost-wise, but it also doesn't allow us to get a good understanding of what the whole landscape looks like. Matt had mentioned that report we put out about ad blackouts. Our real question there was about substitution effect, which looking at an individual company's ad archives doesn't really help us look at how money moves across platforms.

So we recommend this national archive, but more because of some concerns about, actually, First Amendment concerns about compelled speech. We think that the place to put the burden here is on advertisers themselves, like candidates or campaigns. They should be required to not only disclose their spending, which they already do, and their donations, but also to disclose the ads that they run in some form. The thing is, we totally understand that there's a lot of technical and practical just challenges here to how this might work, and we don't have necessarily the answers to all of them, but we think it would definitely be in the public interest.

Matt Perault:

This is one where we actually think there are some sensible product solutions that could be helpful, too. Platforms, I think-- I'm not an engineer, so I don't know this for sure-- but I think it would be relatively easy to add into the ad the advertiser flow, some sort of checkbox or something that probably is checked as a default, where data is sent directly from the platform to the FEC or to a state election body. And I think that's the kind of thing that would likely facilitate doing this at scale, where just in the matter of course of running an advertisement, it's a default that that's going to go, the information's going to be passed along to the right regulatory authority.

Justin Hendrix:

And of course, you suggest that Congress should pass legislation prohibiting misinformation intended to suppress voting. I assume you mean by that technical information, like the polls are at the wrong place or giving the wrong time, that kind of thing, because I suppose figuring out what is misinformation intended to suppress voting could be extremely difficult on some level to interpret all the various messages that could be given.

Matt Perault:

Yeah, this is a recommendation that's been in several of our reports. Essentially, the idea here is to pass whatever would be constitutionally permissible. And so I think what you're getting at is the broader it is, the less likely it is to survive judicial scrutiny. Our sense, and neither of us are First Amendment constitutional experts, but our sense is that you can do something here.

It's very difficult to pass a law of this sort that would survive First Amendment scrutiny, but that doesn't mean that no law would pass First Amendment scrutiny. So what we have suggested is misinformation specifically related to time, place, and manner, which is the narrowness that the First Amendment requires. It's not necessarily clear that even that would survive First Amendment scrutiny, but that's the direction that we would suggest.

J. Scott Babwah Brennen:

I should just say, as a Senator, Obama actually introduced a version of this back in 2007, and I think it was included in H.R.1 and then also spun out into its own bill. I think it was the Deceptive Practices and Voter Intimidation Act in 2021. It didn't pass, but there's been some interest in this sort of bill.

Matt Perault:

Well, importantly, I guess depending on your point of view, but one of the other potential benefits of doing that is it's federal criminal law, and Section 230 provides an exemption to federal criminal law. So to the extent that any online website were itself violating this deceptive practices law, that website could be held liable. 230 would not provide a defense.

Justin Hendrix:

I guess just to finish up here, I'd be interested in a couple of things. One is if you have seen any response to your report that gives you some hope perhaps, that some of these types of changes might be under consideration with any of these parties, any lawmakers that reached out or any kind of engagement with those companies that felt positive to you. And then I guess to finish up, we're obviously right in the midst of the midterm cycle, and whether you've observed any phenomena that you think are concerning after having done this report, and what you'll continue to do even as this cycle lapses to look at this problem or related problems.

J. Scott Babwah Brennen:

Sure. Well, for your first one, honestly, I think our main goal here was to bring attention to an area that people are not really talking about. There's a lot of concern about digital political ads, though very little of that concern is directed at the ads in the programmatic space. So I am really encouraged. We've had a lot of conversations with different people across different types of organizations, policymakers in industry or other academics talking about this problem. And so I am encouraged that there's hopefully growing recognition that we need to spend more time looking at this particular problem. Yeah. Matt, do you want to add anything?

Matt Perault:

I think that that's exactly right. One, you are certainly right to say we've seen limited activity by Congress in this area, but one thing that Scott and I have increasingly tried to focus on is regulation at the state level, because states have been much more active. And so there are states, Scott, one of the states that have already passed an ad archive law.

J. Scott Babwah Brennen:

Oh, California is the big one, which they're still figuring out how it's going to be put into effect. But there are something like six, eight states that have rules that basically require companies to disclose information if asked. Not to have archives, but to make that information available.

Matt Perault:

So there is some traction at the state level and we would expect more of that. I think that that's one area that we're focused on as a locus of policymaking. And then the second thing is I think we are focused to some extent just on documenting the opportunity to learn, and then how little we learn from one election cycle to the next. And so our hope continues to be... The aim here wasn't to make some certain set of companies look bad or look good. Our goal is really to create a framework to hopefully improve how paid election speech, paid political speech is regulated from cycle to cycle. And the frustration for us is that there doesn't seem to be much improvement, but I think we're identifying areas where there could be, and hopefully trying to provide a list of potential policy solutions that could improve it over time.

Justin Hendrix:

Well, certainly that will leave lots of room for your research agenda, I'm sure, to persist well into the future, so hopefully we'll have you back the next time you've got results to report. I thank you very much for speaking to me today.

Matt Perault:

Thanks so much, Justin.

J. Scott Babwah Brennen:

Thanks.

Matt Perault:

Big fans of the podcast, so thank you for having us on.