SIO318: Study Says the People Have 'Near-Zero' Influence on Policy. Does it Hold Up?

Transcript

This transcript was generated automatically and may contain errors — the audio is the record.

0:00:12 Thomas: Hello, and welcome to Serious Inquiries Only. This is episode 318, but for real this time. How's it going, Lindsay?

0:00:19 Lindsey: Spiffing. I'm fantastic.

0:00:20 Thomas: Spiffing? Oh, I haven't heard that one. I've heard spiffy. What's spiffing, educate me.

0:00:21 Lindsey: Yeah. I think it's an English thing, I think.

0:00:29 Thomas: It must be related to spiffie, right?

0:00:31 Lindsey: Maybe? I don't know the etymology of spiffy.

0:00:33 Thomas: Maybe spiffi is not real and I heard that wrong or something.

0:00:37 Lindsey: I think spiff is a thing, too.

0:00:37 Thomas: Oh we're on the topic, I saw a meme that is completely unverified, but I have to share with you, which is a linguistics thing. Did you see this with the Bugs Bunny? It was hilarious.

0:00:44 Lindsey: Mm. No.

0:00:48 Thomas: It was like the whole reason people think Nimrod is a synonym for like an idiot is because Bugs Bunny called who's that hunter guy, that's what, the bald thingy, that Elmer Fudd, called I him Nimrod, which is a biblical comparison to the, like, a great hunter, like it was a sarcasm thing, like, "Oh, some Nimrod here," like as a, and because of Bugs Bunny, we all think Nimrod, I totally thought it meant like an idiot, you know, but uh, yeah, isn't that funny? I love, I love those things. I don't know if that's true, but it sounds true because I do remember Nimrod from the Bible, that actually is, a all of those facts are things that are like, I know that those things are true, so it seems like, like a perfectly plausible explanation, but I'm not going to verify it because it's a meme. And I thought about it for five seconds and thought it was funny. Uh, so anyway, uh, off of linguistics, uh, 'cause we're doing spiff, spiffing and spiffy over here.

0:00:59 Lindsey: Yeah, yeah, Elmer Fudd.

0:01:09 Lindsey: Oh, yeah. That's hilarious. Yeah?

0:01:43 Lindsey: Mm

0:01:44 Thomas: If you are not a patron, um, we're coming at you for the first time in a long time here. Uh. If you want, if you were curious about some of some amount of why we weren't recording for for however long that was. We did a little, yeah, we did a little, kind of just an informal confessional type recording over on patreon.com/seriouspod. wasn't something that I felt like was a main show thing because, you now, I just was ranting and sharing my feelings and doing that kind of stuff. So, but if you're curious about that sort of thing, you can check it out. If not, let's just get to today's episode, which is we're back with a vengeance with a science thingy.

0:01:44 Lindsey: -hmm.

0:01:57 Lindsey: Seven, Eight. Eight years, yeah.

0:02:24 Lindsey: Yeah.

0:02:24 Thomas: I can't wait. I this somewhat viral thingy, which I love it. This is like the fourth time this has happened where I'm like, Lindsay, viral science thing, or at least, you know, data thingy that you can tell us, is this real or not? And almost every time it's been like, oh, cool. Yeah. And then you look into it and you're like, well, this is from 2014, by the way. Somehow these old things, and that's that's not terribly old. It seems like it's still relevant, but somehow these dated studies or whatever, people like just find them and then do some sort of social media post about it. And then that'll go viral in, you know, the year 2022.

0:02:51 Lindsey: Breathe.

0:03:03 Thomas: It's just really interesting. But I think this one is really worth looking into. It expresses, at least the tweet, you known, expressed a lot of what people are feeling right now, which is that the viral tweet was essentially essentially the people have, you percent input on what laws get made in this country.

0:03:03 Lindsey: New life into it, yeah, yeah.

0:03:23 Lindsey: Exactly.

0:03:23 Thomas: That was the thrust of the viral thingy was, yep, the people have zero influence. It's actually, and it wasn't just like someone saying like, ah, people have zero importance. It was, you now, an actual study, a political science study reveals that the people have little to no influence on policy. And Extreme claim. I mean, it's definitely within the realm of believable, but like, I don't know, zero influence seems harsh. Like definitely I can think of like public policy that's happened that you would think is a result of people's feelings and so forth and all that. But I can't wait to get the full science breakdown on this study, see what's going on here, here. See if it holds up to the Lindsay scrutiny.

0:03:42 Lindsey: Yeah.

0:04:04 Lindsey: So this was, this was really interesting, um, and that, that claim, the way that you framed it, I think actually is consistent with what the authors said here, so that's, I don't think it's an exaggeration in the reporting. Um, this article was published in 2014 by Martin Gillans and Benjamin Page in the Journal of Perspectives on Politics, and the title is "Testing Theories of American Politics, Elites, Interest Groups, and Average Citizens." And as you said, this article made headlines for reportedly showing that America is not in fact a democracy, but an oligarchy, with policy outcomes being determined by the preferences of the quote unquote economic elites and the preferences of average Americans having essentially no policy impact. Know, you really want to to have strong evidence before you make a claim like that. Because, you know, as you say, there's a lot of sentiment floating around about, you know, why vote, nothing matters. And I think that you and I are both pretty frustrated with that. Don't actually think it was the author's intention to discourage people from voting, but it's certainly a reasonable takeaway from some of the things that they're claiming.

0:05:03 Thomas: Yeah.

0:05:04 Lindsey: Gilens and Page do make some very strong claims in this paper, including that average Americans have negligible influence on policy outcomes, comes, that any apparent influence that average Americans have is purely coincidental because their preferences happen to align with the affluent much of the time, and that America is therefore not a majoritarian electoral democracy. And yeah, yeah, it is, right? And yeah, that's part of why this prep was interesting for me because I think I largely agree with a lot of these sentiments. I do think that the ultra wealthy have an outsized influence on American politics. I think it's pretty hard to argue with that. Ultimately, I think that the connection between that belief and the actual data that they're presenting in this study is pretty weak.

In my assessment, this paper presents data that are interesting, and it's worth saying that they were extraordinarily labor-intensive to produce. It's very impressive. But their conclusions in some ways are really overstated, and I don't think that their results justify many of the inferences that they're making. I think this will be interesting to dig into. And I'll be interested to hear your perspective on some of the things that I'm going to comment on.

0:05:26 Thomas: Well, that, I mean, that last thing sounds plausible enough. It's just the 0%. Yeah. Like the little to know that degree is, is alarming.

0:06:18 Thomas: Definitely.

0:06:19 Lindsey: So just a caveat up front, you know, I'm not a political scientist. So there are some aspects of this paper that I'm not well equipped to comment on. You know, like the political theory that's motivating their research is not stuff that I'm familiar with. But I felt comfortable doing this because their methods and their analyses share a lot of common ground with psychological research. And so I feel comfortable commenting on those things.

And the things that I picked up on overlapped pretty strongly with the published responses that I saw to it in the article, right? So, so we're going to go through this article and with a particular focus on three different things. So one is how they measure quote-unquote average, average Americans versus rich elites. Second is whether they actually captured influence on policy at all. And the third is some important context for the 21-year period from which these data actually come.

These authors don't focus at all on which party had most control during the two decades that they looked at. And I think that that's actually a really important consideration when interpreting some of the results. So let's get into it. Gillens and Page describe four theories regarding how American politics work and the predictions that each theory makes in terms of who has the most influence over public policy, the average citizens, the rich elites, or organized interest groups. So I'm going to summarize these the best I can. I don't know this literature though, so I may get some of these details wrong.

So the first theoretical tradition is majoritarian electoral democracy, which suggests that policy is influenced primarily by the will of the average citizen through democratic elections. So the prediction of majoritarian electoral democracy would be that policy should tend to represent the median voter or converge at the midpoint of citizen preferences. Second theory is economic elite domination. And as the name suggests, this theory says that it's actually the wealthiest and or highest status citizens who actually have the largest influence over policy. And I mean, you know, it's been argued, for instance, that the very framing of the Constitution focused intentionally on maintaining and expanding the wealth

0:07:17 Thomas: Yeah. Mm -hmm.

0:08:18 Lindsey: Of the wealthy, right? Protecting private property and favoring merchants and plantation overs. Like, no argument here. That sounds legit.

0:08:19 Thomas: Yeah. And minor note, but, like, when the Constitution was drawn up, senators weren't even elected. They were appointed by, you know. So there's even less democracy in that in that case, even though they were appointed by elected representatives. It's still like one layer removed.

0:08:40 Lindsey: No, I didn't know that, actually. That's interesting.

0:08:42 Thomas: Yeah, it's the something amendment that I can never remember.

0:08:47 Lindsey: That one. Yeah. Okay.

0:08:48 Thomas: Yeah, you know, does this

0:08:48 Lindsey: I'm with you. Yeah, yeah, yeah! So, proponents of this theory suggest that elites maintain their power through a variety of mechanisms, including funneling resources into to foundations and think tanks and efforts to shape public opinion toward their own interests, and thereby they dominate key issues in policymaking. And of course, you know, political officials themselves are disproportionately wealthy and may shape policy toward their own benefit. So the second one. Third theory is majoritarian pluralism.

So this suggests that policy is actually the end product of struggles between organized interest groups, business firms, firms, and industrial sectors with competing interests, and that these struggles ultimately lead to policies that more or less represent the needs and interests of most citizens. So it's similar to the first theory in that it suggests that the interests of average citizens are being served, but it emphasizes the importance of these organized groups rather than just, you know, the opinions of the public, essentially.

And then the last theory is biased pluralism. So this theory also suggests that policy is driven by struggles between organized interest groups. But unlike majoritarian pluralism, it suggests that the specific groups that drive policy are not representative of all citizens, and therefore that the end product of these struggles is not policy that actually serves the needs and interests of most citizens. So instead, this theory suggests that the outcome of these conflicts between interest groups tend to favor corporations and business and professional associations.

And the mechanisms by which this is posited to happen include representatives from certain industries funding political parties in order to get their interests represented and favored, I mean we know that happens, um, classes dominate policy making and thus cause the state to serve their own interests, and that lobbyists persuade officials and spend money on election campaigns to serve the interests of business firms and professionals, because relatively few lobbyists actually serve the interest of the poor or average citizens, um,

0:10:40 Thomas: Paper talk about Citizens United? I feel like that would make a tangible difference. I mean, it suppose Citizens United would make it a lot worse.

0:10:43 Lindsey: No. They don't.

0:10:49 Thomas: Was just curious if they like mentioned that and talked about the effect it, it would have had, but yeah.

0:10:49 Lindsey: Um. They did not, um, yeah, they, they do say that all four of these political theories have garnered a lot of empirical support, um, sort of on their own, which obviously presents a kind of contradiction. I mean, all of these things could have influences on policy, but all of these actors couldn't have the most influence on policy, which I think is the main distinction between them.

What Gillens and Page point out is that no prior studies have attempted to pit these four theories against each other within the same data set to see, you know, when you've got the influence of all of these parties held up side by side, which ends up exerting the most importance on policy. And that is the gap that they are trying to fill.

Do offer the caveat up front that that their variables are imperfect, and that their test is preliminary, and that their primary hope is that their study will inspire other researchers to do more research. And that's exactly the kind of attitude that's appropriate for research like this.

But then on the heels of that, they preview their conclusions, and they say that their results suggest that, quote, "Economic elites and organized groups representing business interests have substantial impacts on U.S. government policy, while mass-based interest groups and average citizens have little or no independent influence," close quote. Um, I, I guess you kind of talked about this a little bit at the beginning, but like, a priori, how does the plausibility of this assertion strike you?

0:12:12 Thomas: Not implausible. You know, like, I, I could be convinced of it. I also wonder, maybe there may be a chicken and egg issue. I could imagine if there were like, you, things where public policy is shaped by interest groups, and then you say, ah, see, it's all the interest, interest groups, but then there could be legitimately where it's like, well, yeah, but the interest groups took up those issues because, you know, maybe people cared about them, or the people within the interest group or something like that. You know, and there may be some amount of chicken and egg of what happens first. Do the people decide something, or does the media convince people of something, or does an interest group convince people something? And how do you, how would you count that?

0:12:12 Lindsey: Yeah.

0:12:50 Lindsey: How

0:12:50 Thomas: So I don't know. Those are some of my preliminary thoughts.

0:12:53 Lindsey: Do you tease those things apart? Yeah, and they do talk about about some of this stuff, at one point, that some of the causality here can be reciprocal, and that that's that's kind of hard to tease out, and the kind of data that they have, um, maybe we can come back to that point. I was just interested Yeah, because, because a priori, like, guess my bias is toward thinking that something like this is probably true.

0:12:59 Thomas: Yeah, we'll keep it in mind. The thought process. Like, I definitely see it as plausible, yeah.

0:13:18 Lindsey: Yeah. And, and mainly, like, primarily it's, I blame Republicans for that. But

0:13:22 Thomas: And, as our system in general, electoral system in general, we have things that are, you know, obviously and blatantly undemocratic, like the Senate and like the Electoral College. So you would obviously expect those to have an effect on how the people's will is put into effect or disregarded.

0:13:31 Lindsey: Mm -hmm.

0:13:40 Lindsey: Yeah. And interestingly, that's not really the place that they're looking at influence, like electing officials. The way that they – well, we should just get into it. But like the way that they talk about this stuff is sort of like, you know, they're measuring policy preferences and then they're seeing if those stated policy preferences have influence on the outcome of the policy decisions. But it's they're not really looking at it at the level of electing officials that will implement policies. So I don't know. That's that's another set of things that I think is important in thinking about whether citizens have any have any influence over policy outcomes. But anyway. One one thing that I think it's important to point out up front is that this study focuses on data from 1981 to 2002. And in terms of like, you know, the political.

0:14:29 Thomas: Oh, really? So the data then is before Citizens United. Okay.

0:14:30 Lindsey: Yes. Yeah. I mean, thinking about who was president during that 20 year period. Right. Mean we've got to to Reagan terms.

0:14:34 Thomas: Hmm. Yeah. We've got only one Democratic president in that time, right?

0:14:44 Lindsey: Yeah, right. Right. And that's Clinton. And, you know, as Democrats go, I mean, you know. Yeah. So 1981 to 2002. And with the exception of 1993 to 1995, terms like were divided in terms of party control. Like you didn't have Clinton had a majority Democratic Senate and Democratic majority in the House for like those two years.

0:14:45 Thomas: Yeah. Hmm, for two years. Yeah, until Newt

0:15:09 Lindsey: But yeah, right. And so but like for the rest of this 21 year period, you don't have any other period in which there is there is actually pure democratic control.

0:15:10 Thomas: Gingrich took

0:15:21 Lindsey: So I think my point in prefacing it with this is that whatever the findings of the study end up being, I think that they really need to be considered in that context. Because I think, yeah, I think that these authors are trying to generalize, make this general claim about how American politics works. And they've got data sampled from like this, this period of time that like, I don't know, there's some generalizability issues here.

0:15:20 Thomas: Yeah that's true. Yeah yeah and just to be yeah to be clear it's not as though you're saying, or I would say that definitely that this is proven wrong when Democrats are in power. It's just, if you're not using that data, then you can't really make the claim scientifically either way, right?

0:15:58 Lindsey: Then you don't owe. Yeah, That is, yeah. Couldn't say it better. Yeah. So with that said, let's talk a little bit about what they did in order to test these four theories against each other. So the data set analyzed in this current paper was originally collected for another study that focused on estimating the relationship between socioeconomic status and influence on public policy. So, the data that comprises 1,779 instances from 1981 to 2002, in which a national survey of the general public asked a question regarding whether a proposed policy change was favored or opposed. And it also asked the income of the respondents. Only chose cases in which a constitutional amendment or Supreme Court ruling was not involved in the policy change. As they point out, that might entail a very different policymaking process.

And public public opinion would probably interact with that a little bit differently. So the fact that those were excluded, I think is reasonable. Anyway, so all of these 1,779 cases involved visible salient issues to the public, as evidenced by the fact there actually was a national poll on the topic. So for each case, they have estimates of the percentage of support for or against the particular policy change among citizens at these different income percentiles. And in the current current study, they use income percentiles as their proxy for average citizen versus the wealthy elite. So respondents at the 50th percentile are what they use as their group representing average citizen. And folks at the 90th income percentile, are what they use is their proxy, for the wealthy elite. Would you guess is the salary cutoff for the 90th percentile.

0:16:59 Thomas: Yeah.

0:17:49 Thomas: This is in the United States as a whole, in what year are we talking?

0:17:53 Lindsey: They adjust this for 2012.

0:17:56 Thomas: I don't know, like 250,000 maybe? I don't know.

0:17:59 Lindsey: That's a pretty good guess. So the 90th percentile adjusted in 2012 dollars was about $146,000 per year. Yeah.

0:18:07 Thomas: Oh, Oh, wow. Okay. Even lower.

0:18:09 Lindsey: In 2022 dollars, according to an inflation calculator that I found, that would be $177,000 for a year. And that's household income. That's not individual income. So, you know, that's definitely above average household income. I don't know if that's what I would call the economic elite in the sense that they were talking about it in the introduction. Mean, they're sort of talking about people who own corporations.

0:18:32 Thomas: Yeah.

0:18:32 Lindsey: And we're talking real money in terms of the ways in which they said that the wealthy were influencing politics. I do not know that folks with a household income of $20, $22 of $177,000 a year is really who they're talking

0:18:42 Thomas: Yeah, they don't have like a phone line to the president or something. If you're like, like, I'm we make 180,000 a year because both partners have good jobs. Like, yeah, they're doing well. But that doesn't really feel like they'd have any additional influence, maybe at all. You know, like you don't get, you know, for example, we raised on opening arguments, we raised, well, crap, I can't remember, but it was a lot of money for Georgia.

0:18:51 Lindsey: No. They're not the Koch brothers. Mm-hmm. Yeah. Yeah, it was a lot of money.

0:19:14 Thomas: Gosh, how much was that? I have no memory. God, my brain has been ruined by COVID. We raised, I want to say hundreds of thousands of dollars. Am I wrong?

0:19:18 Lindsey: Hmm. I think so.

0:19:21 Thomas: I think maybe over the course of the two fundraisers, we didn't even get to talk to anybody. We didn't get a phone call from Biden or anything. We got, you know, hundreds of thousands of dollars in one go. And that's entirely too, that's not like we had an income of that and we donated $5 ,000, you know, like we, that all was money going to that effort in Georgia. And we barely, we got to talk to like, you to disrespect the person at all. They were great, but we got to talk to somebody who was not very high up at all and it was just an interview. It it was not anything. So like, I totally agree with you that that's probably not a great measure of who are the like elite of elites, who

0:19:22 Lindsey: I think so, yeah.

0:20:07 Lindsey: Mm

0:20:07 Thomas: Are Koch brothers types. You know, that's not even close.

0:20:07 Lindsey: Hmm. Yeah." Which is who I imagine when I hear the word oligarchy, right? I mean, and you know, they don't use the word, oligarch in this article, to be fair, that was stuff that journalists were using To talk about their results. But that is what they're talking with this theory about the economic elites running the country. And that's not people with, yeah, $180,000 salary or household income. In fact, depending on where you live, it's pretty borderline as to whether that's even out of the middle class. I mean, in the country as a whole, it is definitely upper class. But anyway, so to their credit, they do concede that the 90th percentile are not the super rich in this country.

But they say, quote, "To the extent that their policy preferences differ from those of average income citizens, we would argue that there are likely to be similar but bigger differences between average income citizens and the truly wealthy." I think that's actually a pretty big assumption. Like there's a huge difference in the ability of people with $180,000 in household income and the Koch brothers to influence things.

But there's also some evidence that there's a big difference in terms of policy preferences between people who are just sort of like making a good living versus are incredibly wealthy, which

0:20:14 Thomas: Yeah. Yeah. Oh,

0:21:18 Lindsey: We'll get to in a little bit. But so I disagree with their assumption that the relationship between money and influence would be linear. Yeah, also disagree based on what I'm looking at that like the relationship between income and policy is linear or policy preferences are linear, right? Now, they do cite some evidence that the preferences of like the top 2% of earners are highly correlated with those of the top 10%. They say that that correlation coefficient is 0.91, which is almost a perfect correlation. But the study that they cite only examines 76 people who earn more than $350,000 a year.

And so like those folks are clearly doing great, but I still, I don't think that's the Koch brothers, right? And it's 76 people, right. So my instinct is that there's a pretty good case to be made that there's an nonlinear relationship between wealth and influence, with the caveat that I'm a newcomer to the literature that's being described here in terms of these political theories, their description, their description of economic elite domination theory seems more about the Koch brothers than it does about people who are well off Or not Struggling. So, as I said a a second ago, I did consider it more of a possibility that there was some sort of linear relationship between wealth and policy preference. But I looked at some numbers and I'm less convinced of that now. So the second author on this paper, Benjamin Page, published a paper in 2013 with two other co-authors called Democracy and the Policy Preferences of Wealthy Americans. And so this is a descriptive study examining the policy preferences of the top 1% of wealth holders.

And this is I think it was a sample that was just from Chicago. So it's not like a nationally representative sample. But nonetheless, they got a sample of some very wealthy individuals, about two thirds of the sample of, you know, these very wealthy individual had net worths of at least $5 million, and the average wealth in the sample was $14 million net worth. The average income was $1 million per year in the example, right? This is closer, at least,

0:21:27 Thomas: Absolutely. Yeah, totally. Yeah, yeah, yeah.

0:23:25 Lindsey: To the group of people that I think is relevant. And in this group, you actually do start to see pretty stark divergence in policy preferences between the wealthy and the average citizen, right. These truly wealthy folks in this 2013 study not only reported being extremely extremely politically active, but also being much more conservative than the general public on fiscal policy.

0:23:46 Thomas: Yeah. Right.

0:23:47 Lindsey: No surprises there. They And also showed that the very wealthiest in this data set, so this was folks with $40 million or more in net worth, showed even more divergence from the general pubic in the direction of fiscal conservatism.

0:24:00 Thomas: It makes perfect sense to me. I think that, you know, a question like the death, quote unquote, death tax or whatever, like inheritance tax, your average, even somebody at 200K, 300K, maybe they would think of it, but it's an issue that's really mainly only affecting people with millions, you know, tens, twenties, whatever million dollar estates they're leaving people just as an example, like that, that stuff like that. That I, I agree with, uh, with your intuition that there is going to be that cutoff that work it's really extreme. And you would think they would diverge from there more so than like, Oh, we're doing well. It's two people with good jobs who make, you know, quarter million dollars or something.

0:24:08 Lindsey: Mm -hmm. Mm -huh. Yeah. Yeah, exactly. No, I think that there's definitely an elbow.

0:24:45 Thomas: Yeah. All right. That's it. That's the image.

0:24:47 Lindsey: And yeah, yeah. And so and if you compare that to the current study, which, you know, as we'll get to, they show very, very modest differences between the policy preferences of the 50th percentile and the 90th percentile, right? Very modest differences. So you compare the study, the ultra wealthy to this one, and it really looks like the top 1 % versus everyone else. It's just, I think it's apples and oranges. I Don't. So I don't think that the 90% on income is an appropriate proxy for the economic elite in any reasonable sense. And their conclusions kind of depend on that, or at least some of the central ones do. So instead of saying that they're comparing average Americans with the economic elite, I'd say that they are comparing people with middle-class incomes to people with barely upper middle-class incomes or barely upper-class incomes.

0:24:47 Thomas: Yeah. Yeah. Yeah.

0:25:33 Lindsey: So moving on. Proxy for interest group policy preferences, and this was the really impressive part of their data collection, they compiled a list of 40 major interest groups, which I'll describe more in a moment, and they compiled information on whether each group had taken a public position on each of the policies represented in those 1,779 poll questions in their data set. And they coded these groups' positions as strongly favorable, somewhat favorable, somewhat unfavorable, or strongly unfavorble, and then they created an overall composite of interest group support for each group by subtracting the number of groups with unfavorable positions from the number of groups with favorable positions. And I really do want to give them like a hearty dose of credit for like what a time consuming process this is.

0:26:17 Thomas: A Laurel And hardy handshake. Yeah.

0:26:23 Lindsey: Yes, exactly. I mean, you know, tracking down this information for, you, know, it's nearly 2000 cases for all 40 groups like that. That's a lot of person hours. And then like doing all the content coding and stuff. I don't know. This sounded like a nightmare. I would never do this study. So it's very impressive. And there's a lot, I think, that can be done with this data. I think it's good data. So anyway, so which interest groups are included here? 29 of the 40 groups represented business interests. So they selected groups that had appeared in Fortune Magazine's Power 25 lists over some number of years. They weren't super specific about which years. I don't know. They used people from or groups from the Power 25 list, as well as the 10 industries with the highest lobbying expenditures, right?

0:26:37 Thomas: They Definitely. Didn't interview podcast moguls because I didn't get a call, so. That's

0:27:12 Lindsey: Well, you know, 2002, there was, I think it was only you that was podcasting back then.

0:27:15 Thomas: True. Yeah. No.

0:27:15 Lindsey: So yeah. Yeah, so these business oriented groups included airlines and oil companies and tobacco companies, defense contractors, American Banker Association, Association of Trial Lawyers. It's fun. Automobile companies, Health Insurance Association, you get the idea. Then there was an additional 11 groups that represented mass interests. So examples of these interest groups included the National Rifle Association, American Legion, the AARP, United Auto Workers Union, the International Brotherhood of Teamsters, which, should I have heard of that before? Because that was the new one for

0:27:16 Thomas: Teamsters.

0:27:52 Thomas: Yeah.

0:27:52 Lindsey: Me. Yeah, right?

0:27:53 Thomas: I You always hear about the Teamsters Union or whatever. What are Teamsters? Oh.

0:27:57 Lindsey: Yeah. So it's a labor union.

0:28:01 Thomas: It's A cool name. Why don't, why? Why is it called that? It's cool. Teamsters.

0:28:01 Lindsey: Yeah. Mm -hmm. I always associated the word teamsters with like gangsters or something, but that's not at all what it means, so

0:28:08 Thomas: Yeah. It's just a delightful, sort of like hamster is a delightful word. Teamster.

0:28:10 Lindsey: Yeah. Yeah, and teamster also does not have a P in it, so yeah, um, Yeah.

0:28:21 Thomas: Yeah. No, it's not teamster.

0:28:23 Lindsey: No. No, it's teamster. Yeah, yeah. So these are some of the interest group, the mass-oriented interest groups. This group, this 11 set of groups, also included the National Right to Life Committee, which is a pro-life group, The Christian Coalition, and the American Israel Public Affairs Committee. So that gives you a sense of what the, yeah, for In their initial analyses, they just used a single interest group preference composite across all 40 groups, regardless of whether they were oriented toward business or sort of mass interests. And then they did a subsequent analysis where they created two separate preference composites for the business and the the mass-oriented interest groups so that they could sort of compare those influences separately. Finally, the outcome variable that they used is whether or not the policy change proposed in each of those poll questions was adopted within four years after the poll was conducted. And

0:28:23 Thomas: Yeah.

0:28:35 Thomas: No. Yeah. Yeah. Those are some good ones to include in your study for sure.

0:29:19 Thomas: Hmm.

0:29:19 Lindsey: They note that most of the action occurred actually within two years of the poll. So, So to summarize, their goal here is to see which of these four theories best matches the pattern of results? Do the opinions of average citizens predict policy change most strongly? Or is it wealthy elites? Or are these actually interest groups rather than individuals? And if it is interest groups, is it the ones that are advocating for the masses? Or the business -oriented groups that seem to have the most influence on policy? So to evaluate these theories, they first test a model to compare the relationships between policy outcomes and the preferences of the average individuals, the wealthy individuals, and the interest groups. And then they repeat this analysis, but they replace the general interest group preference variable with those two separate variables representing the two types of interest groups. What did they find? Preference their results by noting, importantly, that policy decisions are not zero sum among the different parties that they're representing in their analyses.

0:30:13 Thomas: Right.

0:30:13 Lindsey: It's not, yeah. So it's not necessarily, yeah, absolutely. In fact, a lot of the time, right?

0:30:15 Thomas: Everybody can get their way sometimes. Yeah.

0:30:20 Lindsey: And they lead with this point because they then reveal folks at the 50th and 90th income percentiles in this data set had very highly correlated policy preferences. So yeah, so the policy preferences were correlated positive 0.78. And again, a perfect correlation is one. So 0.68 is very, very strong. And actually, this correlation was after making some adjustments to the variables to account for like correlated measurement error, which I don't think is worth getting into, but they made an adjustment here. For the original variables before they made those adjustments, the preferences of the 50th and 90th percentile voters were correlated 0.94. Nearly a perfect correlation.

0:30:29 Thomas: Yeah. Like you were saying. Saying.

0:31:02 Thomas: Guess it all depends on what the hell those adjustments are and why they did them.

0:31:05 Lindsey: The adjustments were appropriate. So what they were adjusting for was like responses to a question can be influenced by things like the phrasing of the question. And so across all these different poll questions, you're going to have some measurement error that's due to like the measurement itself. And so it was appropriate that they made the adjustments that they did.

0:31:24 Thomas: Okay.

0:31:24 Lindsey: You know, even after those adjustments, to have a correlation of 0.8 is just, it's very rare to get correlations that strong in behavioral research. That's all I'm saying. Right. Policy preferences in these groups are extraordinarily similar. And, you know, just one other note, some people might be thinking like two variables can be very strongly correlated, but still have different averages.

So like if 10% fewer folks at the 90th percentile favored policy change across these questions compared to the 50th percentiles, their preferences could still be perfectly correlated theoretically, and nonetheless be different. But that's not what was going on here. So when you look at the actual percentages of people at each income percentile that supported different policy changes, there's very, very few issues on which there were actually differences. On the vast majority of things, the 50th and 90th percentiles aligned in terms of what they wanted. Yeah.

So the similarity is important in a couple of ways, and it'll become particularly important when we talk about the regression analyses. Analyses. But before we move on, I ran, so they posted their data set with their article. And I ran a couple of analyses comparing not just the 50th and 90th percentiles, because their data set actually included policy preferences for more income percentiles. So the 10th, 30th, 50th, 70th, and 90th percentile earners were all in this data set. And all of them were highly correlated, like

0:32:48 Lindsey: The the lowest correlation was between the policy preferences of the 10th and 90th percentiles, but even that coefficient, and this is unadjusted, by the way, was 0.84. And that, yeah, right, it's an interesting question

0:32:47 Thomas: Yeah. I wonder how that has changed, if at all. Yeah, and remind me, are asking all types of policy questions, or was it fiscal? Was it focused in any way?

0:33:05 Lindsey: Um they they asked all kinds of questions um they had that coded diluted, yep, yep. Extremely highly correlated across the income brackets. That very interesting .I wasn't necessarily predicting that the correlations would be quite that high.

0:33:15 Thomas: Really? So, even social kind of politics stuff. Interesting.

0:33:30 Thomas: Here's another thing. Always have the problem, and might have mentioned it in our little bonus episode the other week, which is policy pulls way differently than actual politics happens. So you might look at something like Florida when they voted on the $15 an hour minimum minimum wage. And you might say, oh, well, everybody in Florida, regardless of income, they seem to support this ex -policy position here. And then if you had tied it to a candidate, then it would totally change. And I don't know what that data would look like.

0:33:40 Lindsey: Very true. Yeah, a good point.

0:34:05 Thomas: That may actually be the more important thing. I here's the thought. How would this whole picture of things, Dr., be affected, taking into account that like policy doesn't really matter anymore? Like policy, it's not policy preferences, however, you could make an argument that like, well, the people are still having a huge effect in the candidates they pick, but policy just doesn't matter anymore, I guess, would that still mean like, well, it's still an oligarchy then, because that means that the people are just being run around using these distraction social issues, you know, the culture war issues, to distract. And then meanwhile, what actually happens is whatever, whatever, the people in power or the elites or the oligarchs, whatever, what they want to have happen, I guess.

0:34:25 Lindsey: Yes.

0:34:49 Lindsey: Unfortunately, that tracks to me.

0:34:51 Thomas: Yeah. That could be another thing going on here. But I do think that would at least change the way you look at this a little bit, because if the data you're using is policy polling, then I don't know, I'm trying to think, what do you think about that? How do you filter filter that through this whole approach.

0:34:54 Lindsey: Mm hmm. Yes. No, I had the same thought. Like they're, they're testing this idea of, you know, the, the influence of the American electorate on politics, but they're specifically focusing on the policies they want versus the candidates that they elect. And like, I, you you know, I don't, yeah, yeah.

0:35:24 Thomas: Yeah. No, actually, now that I think about it, that probably is the right thing to do. Yeah, because if like, hypothetically, and I actually think this is pretty plausible, as I just said, if the method of, to get super tinfoil hatty right now, if the message of control of the oligarchs, of the powerful, the rich and powerful, is is to just muddy the field so much with culture or BS that that's all people care about. And then they vote their side no matter what. And then the policy is just an afterthought. I guess that would, you want to still look at the data this way.

You'd want to still say it's, I guess it would be, it would still be a relevant fact that the people have control over the politicians that they elect somewhat like that's still an important distinction because you could differentiate that with like Russia or, you know, some sort of country where it's like, yeah, you don't, there's not even really free and fair elections at all. It's like there's a difference between that and like, no, there is free and fair, quote unquote, elections more or less within our system, but it just doesn't matter because, you know, no matter who you elect, they're going to do the will of the wealthy or something.

0:36:07 Lindsey: Yeah. Yeah, this is my bias, because I'm, I am on the left, but I feel like that stuff is going on a lot more on the right. Right?

0:36:41 Thomas: Right. Good point.

0:36:42 Lindsey: So then it does matter who you elect.

0:36:42 Thomas: Like this whole time period. Yeah. Yeah, no, the time period of this data, as you described earlier, it does seem to be a major weakness here in making any broad claims.

0:36:50 Lindsey: Yeah? I agree. Mean, I would like to see this expanded. I don t want to do the things necessary to get the data for a larger time period. I would like to see this expand, or, you know, so I have their data set, maybe I'll screw around with some of the stuff and see if I can do anything with the two years in which you was more control by Democrats. I don't know, um, not, you know, two years out of 21, though, I don't, I don't think I'm gonna get very far, but anyway, yeah, I like a lot of this stuff, like, I do, I think happens so much more on the right, and so I think that, like, a totally reasonable interpretation of this study is that, yeah, when you elect Republicans, they're going to pass policy that favors the wealthy.

0:37:09 Thomas: Yeah, yeah, yeah.

0:37:29 Lindsey: Oh, wow. What a surprise.

0:37:29 Thomas: I mean, that seems just absolutely redundant to say. I agree. That

0:37:32 Lindsey: Yeah. So anyway, so the policy preferences across all income percentiles, very highly correlated. In contrast, though, the preferences of individuals were not strongly correlated with the interest group preferences, which was was interesting yeah So the 50, the 50th percentile earner's preferences were correlated positive 0.12 with preferences of mass public interest groups, which, point it was significant, but it was like nothing, like a 0.1 is nothing. And, uh, 50th percentage earner's references were correlated negative 0.01 with the preferences of business interest groups, so that's negative correlation, you know, they've got opposite preferences essentially, so those are the relationships that you would expect intuitively in terms of the direction, but like the magnitude of the correlations is incredibly weak. There's almost no relationship there, right?

0:37:48 Thomas: Is interesting.

0:38:11 Thomas: Okay, interesting.

0:38:26 Lindsey: Interestingly, the preferences of the 90th percentile earners, and again, remember, this is their proxy for the economic elites, were actually not significantly correlated with the preferences of either set of interest groups. And I think that that's very interesting, given that the theory that they're testing with this group, right? Like the interests of the true economic elites are heavily tied to the interests OF major business and industry groups. Again, it's another point that makes me wonder like, hmm, is this actually a reasonable proxy for the economic elite? I don't know.

0:38:39 Thomas: Yeah, yes. Yeah, and here's a thought. Let me know. Maybe this is nothing. Maybe it's something. What kinds of things do groups care about? You know, there's not going to be an interest group that's like, let's all breathe oxygen. You know? It's like that's something we all agree with. So why form an interest Group? It's going to be stuff. Like I would just think hypothetically, theoretically interest groups would be things that would form to try to push issues that are, it be stuff that's like close to majority, but not quite like you would feel like, okay, we're almost there. Maybe 40 % of people believe this thing. Let's get it to the point, to the tipping point where we can do something about it kind of thing.

0:39:15 Lindsey: Yeah.

0:39:31 Lindsey: That's a good point. That is a really good point, I hadn't considered that.

0:39:41 Thomas: You would never expect it to correlate with all of society. That wouldn't make sense. Right. And I guess unless truly they were part of us, they were, they realized that there's, we don't get the will of anybody passed through government. So it's like, Hey, a hundred percent of people agree with this thing, can we do something But under this theory, the interest groups are actually having an effect on policy, just not the people, right?

0:39:45 Lindsey: Yeah. Mm -hmm. Yeah, nope, that's a really good point. They examine just these correlations between the preferences of individuals and groups, then they move on to using those preferences to actually predict policy outcomes. And so first they just look at simple correlations between the references and the policy change. And what they find is that all three of these these preference predictors, are pretty strongly associated with policy change. So the preferences of the 50th percentile are correlated 0.64. The preferences of 90th percentiles are correlated 0.81. And the preferences of the interest groups are correlated 0.59.

0:40:40 Thomas: All the groups? Is that a lot different than saying the people have little or no effect on the policy? See.

0:40:45 Lindsey: They haven't gotten to the regression yet. Yeah.

0:40:47 Thomas: Okay. Gotcha.

0:40:49 Lindsey: So just the simple zero-order correlations between these things are all strong and all positive. And this indicates that policy outcomes in general are aligning with the preferences of each of these parties. But then they create this regression model that includes all three of the variables to predict the policy decisions simultaneously. And this is useful because it allows you to examine whether these variables have unique relationships with policy change. So in other words, after you account for the preferences for the 90th percentile and the interest groups, do the preferences of the 50th

0:40:49 Thomas: Okay. What's Left over. Essentially.

0:41:20 Lindsey: Exactly? Yeah, right? Does the 50 percentile have any additional unique relationship. They found that the preferences of the 90th percentile earners and the interest groups both had unique independent relationships with policy decisions, but the preferences of the 50th percentiles earners did not, after you accounted for those other preferences. So, according to their interpretation, to the extent that the, the preferences of the fiftieth percentile owners are related to policy change, that relationship is coincidental.

It's explained by the fact that the wealthy already wanted that and the interest groups already wanted that. Now, this is where we need to come back to that high correlation between the preferences of the different income percentiles. I think that there are problems with this analysis regarding multicollinearity, which other researchers have also pointed out in their responses to this article. So multicollinearity means that two of the predictors in your model are highly correlated with each other.

And this isn't necessarily a problem if all you want to do is precisely predict an outcome, right, in this case, policy decisions. But if you're trying to infer something about the relative contributions of the individual predictors, it can be a big problem because it makes the estimates of the two correlated variables unstable.

0:41:21 Thomas: Okay. Yeah, yeah. Oh.

0:42:33 Lindsey: So my contention is what they've actually done is measure the policy preferences of the middle class and then people who are like on the borderline of the Middle Class and the Upper Class, right? And that is supported by the very high correlation between these preferences. And then they're using both of those measures to predict policy outcomes.

And so they find that if you account for one of those two measures of, I think, essentially the same thing in that regression equation, the effect of the second measure goes away. That's how I'm understanding these analyses. Andso to be completely fair, Gillen at least seemed to address the multicollinearity critique in a published response to someone else's critique. But unless I'm really missing something here, I don't think he actually addressed it adequately.

So his response was that multicollinearity is not an assumption of regression and that it doesn't bias the coefficients. And those things are technically true, but it's not a response to this critique, right? If all they were trying to do was create a model that predicted policy outcomes as precisely as possible, wouldn't matter that they've got two highly correlated variables. But they're trying to interpret the individual contributions of the predictors in their model.

And the argument is not that multicollinearity biases the results, it's that it reduces the reliability of the estimates. So if you use two measures of essentially the same thing to predict an outcome, the differences between how well those two measures predict your outcome are largely going to be due to error, not because of

0:43:56 Thomas: Wow.

0:43:56 Lindsey: Some Substantial difference between how well the two things behave, right? To up this point, like, you know, they found that the policy preferences of folks who are in the middle class don't uniquely predict policy outcomes above and beyond what you can already predict with the policy preferences of folks who are barely above the middle class. And those two groups have almost identical policy preferences. So that's, I don't, yeah.

0:43:57 Thomas: I See. I see, yeah, okay, I think I understand it's it's that's a very complicated subtle point there, but I think I think you explained it very well in I think I barely get it because it's super complicated, but yeah, so it's like you can't really do that regression analysis then if your thingies are too tightly correlated, right? Is that what you're saying? You can't really tease out the difference.

0:44:40 Lindsey: You can't make the kind of inferences that they're making about, you know, like this variable, you know, being the, yeah, yeah Yeah. Um, that's, that is my contention at least.

0:44:42 Thomas: Hmm. Right. Yeah. It won't be effective, I guess is what I meant. Not that you can't do it, but yeah. Okay. Oh, it's hard as a listener to evaluate like the plausibility of that one way or another, because it's such a like expert thing that you're talking about, but it does make sense now that you say it. So like your impression is that would be, that would would be pretty fatal to the conclusions being drawn here.

0:44:53 Lindsey: Um I think so. Right. Mean, like, another, maybe this is a more intuitive example. Like if you were trying to predict, um, someone's fertility by using like the, the variables of age and remaining number of eggs and the ovaries, those two things are like incredibly highly correlated with one another. So

0:45:26 Thomas: Yeah, I just thought of an example too. What about if you had just arbitrarily divided the 50th percentile into two groups then tried to do this analysis of like, well, which group has more influence?

0:45:27 Lindsey: Yeah. Yeah. Go, mm.

0:45:43 Thomas: Then it would be like, yeah, it would just be like like you say, any difference would be due to error probably or just statistical noise, right?

0:45:37 Lindsey: -hmm. Yes, that's it that's that's a great example I like that, yeah, mm. -hm, yeah. Exactly.

0:45:53 Thomas: Okay. So that's kind of what's going on. I think I understand that.

0:45:53 Lindsey: Yeah. Right, at least that's my read on it, like, all

0:45:59 Thomas: That's super fascinating. I think it does. Like you said, the whole issue there is that 90th percent just not really being as high enough to be like the elite of elites.

0:46:08 Lindsey: Yeah. Get me the Koch brothers in this analysis and then we'll have something to talk about, I think.

0:46:15 Thomas: I already forget the numbers, but 175 -ish or based on inflation or 150, you know, like a lot of the people in the middle could be at 100 ,000 or 90 ,000, you know, like there's not going to be that much difference between those two numbers of income in terms of preferences, don't you think? They even could, based on the state, those could be the same people. You know, someone making 170 ,000 in California could honestly be at the exact same level of affluence as someone making, I don't know, half that in some other state.

0:46:22 Lindsey: Yeah, Yeah.

0:46:46 Lindsey: Absolutely. Right. I mean, I actually, I looked this up and $177,000, $176,000. $175,000 dollars a year is a middle -class income in New York City. There you go. Yeah? Yeah.

0:46:48 Thomas: Okay. Yeah, totally. Curious how that mathematically even is the 90th percentile. I'm very, are you positive on the household income thing? It's not just an individual income thing that would at least make it so.

0:47:06 Lindsey: It's household income.

0:47:07 Thomas: Wow. Okay. So it's just a lot of people around the country live in cheap places and they don't make as much, I guess. As a Californian, it is, maybe it does skew my view of things because the, you rent here is like $100,000 a year. It's just insane. That

0:47:07 Lindsey: Yep. Yeah. Right. Yeah, yeah. No. And actually that's a good point because, cause or whatever it was in this study, that was the 90th percentile in 2012 or whatever. It was, I don't know that 177 is the 90th percentile now so I was just yeah yeah I was just making that adjustment to

0:47:32 Thomas: Oh, that also may have shifted. Yeah, yeah, yeah. Gotcha. Okay. Yeah, I mean, I'm sure it's close -ish, but who knows?

0:47:43 Lindsey: Yeah I don't know but no, I mean it does speak to the point that like there is there's... there's the elite and then there's everybody else there's a massive difference between like the truly wealthy in this country and um

0:47:51 Thomas: Yeah. Absolutely. And that's borne out in a lot of different data. I remember the big old, gosh, who was it? Was it New York Times? Somebody did a huge story on the tax avoidance, and they emphasized that this really is an elite percent problem. Like, I can't remember the numbers, but it was the case of like, we're not talking, you know, people who make $200,000. We're talking elite of elites. They're the ones accounting for like all of this tax avoidance in the billions of dollars as a country. Tree.

0:47:56 Lindsey: Yeah yes Yep, pretty cool, right? So so like so this this regression analysis is is really like the core finding that was was being reported on, right? And so they report this, that you get this effect going away. And this is how they interpret this result. So quote, the chief predictions of peer theories of majoritarian electoral democracy can be decisively rejected. Not only do ordinary citizens not have uniquely substantial power over policy decisions, they have little or no independent influence on policy at all. That's a very strong statement given their actual results. Like even if the the multicollinearity thing was not an issue, this would not be an appropriate inference.

0:48:38 Thomas: Yeah. Yeah, yeah, yeah.

0:49:09 Lindsey: They've measured policy preferences and they've measured, uh, policy decisions. And then they've examined the correlation between these two variables. Using these words like influence and impact throughout the article imply that they've attempted to find some sort of causal relationship, which they haven't.

0:49:23 Thomas: And again, only for the years 1982 through 2002, right? Or whatever it was. Yeah yeah.

0:49:26 Lindsey: Yes. Let's not forget that.

0:49:29 Thomas: Important caveat.

0:49:32 Lindsey: And then after that inappropriately strong inference, they properly note, that the overall model here, the regression model predicting the policy outcomes, had really low explanatory power. So these three variables together, these three preference variables only accounted for 7.4% of the variance in policy outcomes which means we got nearly 93% of variability in policy outcome unaccounted for by their variables. That's not super impressive.

0:49:32 Thomas: Wow. Yeah huh.

0:50:00 Lindsey: So maybe the headline should have been, "Voter preferences and powerful interest groups all all had negligible relationships with policy outcomes from 1981 to 2002." Not as catchy, but yeah. So they attribute the lack of explanatory power to the imprecision of their proxy variables and therefore conclude that if anything, they're probably underestimating the effects. I think that's a pretty convenient interpretation. That's essentially saying that weak results are actually evidence of strong results.

0:50:09 Thomas: No.

0:50:25 Thomas: Yeah okay.

0:50:26 Lindsey: That's not a thing. Yeah. So so that was their main an analysis, they also did an analysis comparing the two types of interest groups and it produced very similar results. So in that new model, average citizens' preferences, once again, were not unique predictors of policy outcomes, but the elite preferences, quote unquote, mass-based interest groups, and business interest groups all did have significant unique relationships with policy outcome, right, when they looked at those two things separately. The estimate for the mass-based interest group had the lowest estimated impact of all of those three predictors. So the wealthy elites and the groups that were oriented toward business interests seem to have larger impacts in that analysis. Analysis. And attribute that strong relationship between the business-oriented interest groups and the policy outcomes with the fact that these groups tended to be more active according to, you know, how many positions or how many issues they took positions on, right? They also noted that their positions tended to be more cohesive compared to the mass-based interest group. So like, yeah, so like big oil and big tobacco.

0:51:28 Thomas: That would be a challenge for getting stuff done if you have a bunch of cross-cutting thingies

0:51:31 Lindsey: Exactly. Yeah, exactly. Yeah. So So they were more active and they agreed on more and yeah, that they attributed their greater quote unquote impact to that. Those were most of the central findings that were reported in popular press outlets, but there was one or two other interesting things that I thought were worth mentioning. So, one, they discuss several findings in this data set pointing toward a status quo bias in both policy outcomes and in the preferences of interest groups.

0:51:35 Thomas: Yeah, hmm, god. Yeah.

0:52:00 Lindsey: And I think that these are genuinely interesting. And I also think that you can't fully talk about them without acknowledging, again, that these data were collected from 1981 to 2002, which Gillen and Page do not discuss at all, right? First, they say, quote, in policy cases, narrow pro-change majorities of the public got the policy changes they wanted only about 30% of the time. More strikingly, even with overwhelmingly large pro-change majorities, with 80% of the public favoring a policy change, they only got change about 43% of the time. So I think that that is interesting. I too have noticed that Republicans are constantly putting the brakes on extremely popular policy changes.

0:52:38 Thomas: Well, and again, you talk about the systemic issues, everything in our system except for like the House of Representatives, and that's it, like every other thing makes it impossible for uh, to happen quickly without overwhelming support of public.

0:52:57 Lindsey: That's true. Is true. Yeah. The system is set up, I think, in a sort of biased where the status quo, right? But I think it is also fair to point out that Republicans are putting the breaks on things a lot more often than the Democrats are. Yeah. And yeah, so relatedly, in a section where they're discussing the evidence for the influence of the interest groups on policy, they say, quote, an important feature of interest group influence is that is often deployed against proposed policy change. Of the 1,357 proposed policy changes for which at least one interest group was coded as favoring or opposing change, only in 36% of the cases did most groups favor change, while in 55% of the cases, most groups oppose change. And again, I think this is unsurprising given the time period, right? These preferences would be aligned with the actual policy outcomes given who was controlling politics largely during this period.

But so they conclude once again by saying, well, I'm going to read this quote because again, it's really strong language and I think that's funny. So their study revealed some striking findings. And so they say, quote, one is the nearly total failure of the median voter and other majoritarian electoral democracy theories. The preferences of average Americans appear to have only a minuscule, near zero, statistically non -significant impact on public policy. So that's how

0:53:12 Thomas: Yeah, yeah. Right. Yeah,

0:54:25 Lindsey: They're, yeah.

0:54:26 Thomas: That's that strong language that we started with. That just doesn't seem substantiated at all.

0:54:31 Lindsey: It's not. And again, I'm not a cockeyed optimist about this stuff. I do believe that there's a lot to be, you know, of here and but this study just doesn't show that

0:54:35 Thomas: Yeah. Well, I won't speak for you, but I do think you could do another study that may prove that. Like you could prove that same thing maybe based on our priors. I think that, you know, if I were to just guess, you wouldn't be zero percent, you know, wouldn't be little to none, but it would be low in terms of the actual influence that the people have. But yeah, no, I think you poked a lot of holes in this and it doesn't really seem to hold up to me.

0:54:45 Lindsey: Yeah mm -hmm yeah I don't, I don't think so either, um, am interested in playing around with this data set a little bit more though because there is good stuff in it

0:55:09 Thomas: Yeah, but but it's all pre Citizens United, which I think makes a huge difference. Then again, who knows? I don't know. Like, I think Citizens United dominated like a decade of political discourse of, you know, all this. And I think it should be repealed or, you know, we do an amendment or whatever. But like then the sort of social media era of elections became more prominent and that kind of in some ways took more power back to the people in quotes, but also to the people who were able to influence people. So who knows? Story short, it's a very complicated thing that I think you would need to do another other comprehensive study to try to evaluate.

0:55:14 Lindsey: Yeah yeah absolutely, oh

0:55:51 Thomas: What a, what a fun journey there. That was, ah, that Dr. Lindsay Osterman is so smart. I, I love when you take us through stuff like that and you made it make perfect sense. So bad science thingy. What's your, what's your science thingy scale? Did we, what was the scale? Crap. We had a good one. Somebody, Roger, somebody came up with a good One. I don't remember what it was.

0:56:01 Lindsey: Good, was wondering, yeah no yeah I remember what it was, it was good, something about palps, um. Um, two palps down. We needed more nuanced scale. I need some more palp points.

0:56:16 Thomas: Two palps down. Yeah. You need more. Well, we always need more palps.

0:56:25 Lindsey: Yeah, that's true. Yeah. Um, no, I don't, I think this was terrible. You know, the, the again, I'm very impressed with the actual data that they collected and by and large, I don't know compared to some of the other things we've covered, like this isn't awful. It's just, I'm, I've, I I think that they were drawing like really, really strong conclusions given what they were actually measuring here. But

0:56:30 Thomas: Yeah. Yeah, yeah. Right. Well, thanks for the breakdown, and we'll have to look elsewhere for evidence of what we think is probably true anyway, which is we live in an oligarchy, but, but yeah, not this paper doesn't, doesn't quite do it. Well. Thanks so much, Lindsay. Thanks for listening everybody and bearing with us through months off, we're, we're trying, we're, long story short of the, the, uh, Patreon recording we did earlier, we are trying to get back to it, well we can't, we can't promise that it's going to be, um, single week, but I think we can, we can make a try of it, we'll See all Right, that was a ton of fun. Thanks for listening and do support the show, patreon.com/seriouspod, um, and especially because when we don't make anything it doesn't cost you anything. I mean, come on.

0:57:19 Lindsey: Think so.

0:57:30 Lindsey: There you go.

0:57:32 Thomas: Thanks everybody see you next time

0:57:42 Clip: Thank you so much for listening to Serious Inquiries Only. If you like the show, the absolute best thing you can do is support us at patreon.com/seriouspod. The second best thing you can do, is please share it on social media or by word of mouth. Speaking of social media, follow me on Twitter at Serious Pod. If would like to be Facebook friends, send me a request at facebook.com Thomas podcast. If not up for that level of intimacy, that's fine. Maybe you want to join the discussion at facebook.com/groups/SIO members. Thank you again for listening. We'll see you next time.