Patrick McKenzie and Kelsey Piper sit down for a live conversation tracing a line from newsroom Twitter mobs to the near-destruction of PEPFAR, exploring how the breakdown of institutional norms in media has infected government decision-making with deadly consequences. [Patrick's enhanced transcript will be uploaded here over the weekend]

Sponsor: Mercury
This episode is brought to you by Mercury, the fintech trusted by 200K+ companies — from first milestones to running complex systems. Mercury offers banking that truly understands startups and scales with them. Start today at Mercury.com 
Mercury is a financial technology company, not a bank. Banking services provided by Choice Financial Group, Column N.A., and Evolve Bank & Trust; Members FDIC.

Timestamps:

(00:00) Intro
(00:31) The Argument
(03:19) Challenges in modern journalism
(06:42) The impact of social media on discourse
(13:37) The role of Substack and independent media
(20:13) Sponsor: Mercury
(21:30) The role of Substack and independent media (part 2)
(30:59) The PEPFAR program and its importance
(44:01) Impact of US aid cuts on global mortality
(45:25) Substitution efforts and their limitations
(47:54) PEPFAR's partial continuation and challenges
(51:21) Consequences of administrative decisions
(54:28) Elon Musk's influence and government actions
(01:00:14) Challenges in government accountability
(01:15:47) Reforming administrative processes
(01:24:45) The role of community input in development
(01:28:28) The power of constituent voices
(01:30:15) Wrap

Transcript

Patrick McKenzie: Hi everybody. My name is Patrick McKenzie, better known as patio11 on the internet. And I'm here with my friend Kelsey Piper.

Kelsey Piper: Hi everybody.

Patrick McKenzie: So Kelsey, you were on the program a bit earlier, a couple of months ago, and news has broken as I suppose it sometimes does in journalism. You've joined a new outlet called The Argument. Did you want to tell people a little bit about it?

Kelsey Piper: Yeah, so The Argument was founded by Jerusalem Demsas who was someone I previously worked with at Vox and have admired for a long time. And the premise she pitched me on this spring was, liberal values - which include classically liberal values like free speech and protection from the government and stuff like that, and also liberal values of the Democratic Party that haven't been very prominent and very argued for over the last couple decades - are really important, have a hard time in the modern information ecosystem for whatever reasons.

But one thing that liberals are really good at doing is arguing, is sort of getting into it with each other and with people we disagree with, and people enjoy reading that. And what if we built something around having in public and openly some of the arguments that I think people have been having behind the scenes as they try and figure out, where is the Democratic party, which is currently a mess, headed? And more broadly, which are our fundamental values and what mistakes were made?

Because another conversation that Jerusalem and I had this spring was, well, if you are not very happy about the current direction of the country and you are somebody who works at least somewhat tangentially in politics and policy, then it seems good to do a little bit of analysis where you're asking, what could I have done differently over the last five to 10 years that would have changed this trajectory? If I don't feel good about where our country is today, what could I have done?

And I think part of the answer for me certainly is to be louder and more emphatic in defense of liberal principles and to... everybody should be more courageous is an easy thing to say, but being courageous individually in an environment that heavily punishes that doesn't work very well. And it's in some cases not very wise. Whereas if you have a group of people who are saying, this is an ethos that we are all adhering to and we will, if you say something and you are unjustly attacked for it, then even if I don't agree with it, I will be very clear that you're being unjustly attacked for it.

You can sort of form a group of people who guarantee each other a little bit more freedom to speak without requiring as much individual courage or having quite as high costs individually. I think that's the thing I would've liked to see a lot of people do five to 10 years ago.

Patrick McKenzie: Yeah. As someone who was out of the mix for about 20 years due to being in Japan at the time, by the standards of the political culture I grew up in, I don't know if milquetoast is quite the right word to describe what you're describing or... a thing that was broadly agreed upon by every relevant section of the polity. Good to see it coming back.

So question for you, for people who might not understand the media landscape as well. The paper of record in this country is broadly described by people who like it as a crusading liberal newspaper. What do the mainstream media outlets and the media market do that makes that sort of thing insufficient for the goals you're trying to achieve? Why not just become an editor at the New York Times?

Kelsey Piper: Well, they haven't offered me a job for one thing. I don't think they would. I think one nice thing about Substack is that you only have to persuade the readers that you have something interesting to say. And if people read what we're saying, they can subscribe. And that is itself... you don't need very many subscribers to make a Substack viable. So your subscription actually substantially, on an individual margin, influences what kinds of things are viable for people to write about.

It's often much easier just to build that readership and show that you have something to say that people are interested in and that matters. And then the conventional prestige media - you see CBS buying the Free Press or whatever - will often say, oh, I guess there is something there. Whereas it's harder for them to in advance decide that there's something there.

But even aside from that, if you're the New York Times and you are conceiving of yourself as the paper of record, you're going to have a bunch of weird editorial stuff that I think is downstream of that. And I've heard very different things from different people who worked with the New York Times about what that editorial culture cashes out to. I think it can be very differently experienced in slightly different contexts. And a lot of the people who leave the New York Times are kind of disgruntled, so I take with some salt specific accounts.

But I think it is hard to both be the paper of record and be saying, we're just gonna have several of our writers really go at each other. The columnists do disagree, but there's a... once you have positioned yourself as a certain amount of importance, I think it's kind of... it would be discrediting for something that important to have been completely wrong about something. And since we're all completely wrong about something, it is in some ways stepping down the prestige ladder a little bit lets you just be a little bit more open and disagree a little bit more openly. I don't know. That would be my guess.

Patrick McKenzie: I think in many industries there's something of a culture of comity, which is why people in the tech industry, present party possibly excluded, might say privately, I don't think crypto has anything to it. However, X and Y and Z are investing billions of dollars. And so I certainly can't tweet "the emperor has no clothes." And I think in the prestige media, there was often at one point simply a culture of comity - that one did not defecate where one eats, that you don't take swings at people who are in the same office or in an office very similar to yours where intermarriage, friendships, et cetera, the casual loose definition of incestuous.

And then there had the perception that in the last 10 years it hardened into something a little more than that where criticism of a colleague was not simply something that one didn't do in one's social class, but would be a physical attack on a colleague and put them in danger. Does that align with things... I don't know that I would...

Kelsey Piper: I don't know that I would go quite that far, but I think the phenomenon of social media mobbing pretty dramatically changed people's willingness to criticize each other. And the sort of generous interpretation here is social media mobs are very bad. Even when the behavior they're mobbing about is itself very bad, and sometimes it genuinely is, the mobs are still very bad. They do a lot of splash damage. They often convey their rage through graphically violent threats. They often target unrelated but nearby parties.

It is understandable to, in response to the rise of the first social media mobs, go, we are going to be extremely cautious about behavior that sort of seems to encourage these or sort of seems to reward these. Now I don't actually think journalistic outlets did a great job of... if you'd sat down and you were saying, alright, social media mobs are extremely bad and we're going to design our policies in such a way as to not reward and not encourage social media mobs top down, I don't think you would've ended up at the policies that outlets generally did end up at. I think they did a kind of wishy-washy combination of sometimes having a backbone, sometimes giving into mob pressure and sometimes scrambling for the nearest account of why they should do the thing that they already wanted to do in a principled way, which ended up with a lot of incoherent stuff, but I think it makes...

Patrick McKenzie: Would you mind if I made this a bit more concrete, just so people...

Kelsey Piper: Sure. Go for it.

Patrick McKenzie: So, not to throw them under the bus, I frequently use the New York Times as a metonym for the entire journalism industry. Justifiably, they're wonderful. And they've done some wonderful things over the years. One thing they did was publish an op-ed by, I believe, a Republican Senator Cotton.

Kelsey Piper: Tom Cotton.

Patrick McKenzie: Tom Cotton, thank you. And in the wake of that, there was something of a mixed Twitter slash Slack event where many people in the newsroom were deeply opposed to both the things that were said within the walls of the op-ed piece itself, and also opposed to the New York Times, quote unquote, platforming it in any capacity whatsoever.

And so there was, to my understanding, a coordinated event between those employees on Slack and those employees on Twitter, posting extremely repetitive tweets to the general gist that this op-ed puts our colleagues in physical danger. My read of that was that someone with labor organizing practice realized that if you phrase it in exactly that way, you can force your employer to take action the same way you can force your employer to install physical guardrails on catwalks, because otherwise they're in OSHA violation territory.

But characterizing principled disagreement as a physical threat to the safety of people seems to have been something that happened more than a couple of times over the last 10 years.

Kelsey Piper: Yeah, I think that is an accurate concrete characterization of that event. It's also nuts, right? It is in fact not putting your colleagues in danger to platform a sitting US senator. Even to the extent that you should choose who to platform wisely, which is something I believe to some degree is true, a sitting US senator has the platform of the US Senate, and you're not substantially increasing their reach. And it is often going to be of immediate interest and relevance to readers how senators who after all make the laws think and act.

So the entire premise there was silly to me and I would certainly want to see it very clear in employment law that some categories of employer behavior are not activities that put employees in danger or create a hostile work environment, by definition, and publishing an op-ed is not to my mind such an action.

Now, obviously, my employer could publish an op-ed that was so outrageous I was saying, I quit. I will have nothing to do with you going forward. Your values are clearly not my values. And I regret having ever contributed... it is perfectly reasonable to have that line. I would have a great deal of respect for anybody who drew that line, even if they drew it in a place which is not where I would draw it.

But the specific things that played out were kind of this odd combination often of not disagreeing publicly about lots of things that were kind of foolish and ill conceived and inaccurate and should maybe have been disagreed about publicly, and then coordinated public disagreement about things that... well, the coordinated public disagreement is more marked if we're not doing disagreement in general. If you're never doing disagreement, you can at least say, well, we're never doing disagreement. I thought it was silly when that coworker said that thing that was clearly not true, but we never disagree with each other.

But if you are doing disagreement in some contexts, then I think it becomes more reasonable to say that your lack of similar disagreement in other contexts reflects maybe not a genuine value or value that you agreed with that or whatever, but certainly that you did not consider it as important to oppose as Tom Cotton's whatever.

So yeah, I think you just had this very unhealthy dynamic along a bunch of different dimensions and I don't know exactly how to change that. But what we're doing at The Argument is the column where Argument columnists argue directly with each other or directly with other people in the journalistic professional class or whatever, about topics that really matter.

And in the first one we were pretty gloves off - we called each other idiots, all of that good stuff. I think that this was intentional. Some people were asking, this is kayfabe, right? I'm saying, no, I think he thinks I'm an idiot. I don't personally call someone an idiot if I don't think it's true. But it is an intentional effort to reset norms around what you can say and make people go, oh, you can say that. And saying that is part of a culture in which when things are important to say, they get said.

And where I think you want symmetric pressures on kinds of rudeness or whatever. There should be some pressure against rudeness. Rudeness is bad. It is not, in fact, generally the best way to do truth seeking, thoughtful disagreement. But excessive politeness can also be bad and can also get in the way of thoughtful truth seeking and disagreement. And if you cannot remove those pressures, I think equalizing those pressures does something good.

Patrick McKenzie: I think there is something which you're in the media environment and ecosystem that we find ourselves in, and without necessarily being able to control all the pieces on the board. One thing that, for better or worse, has happened and we're downstream of it happening is a company with a bird logo got bought and then they changed it to a name, which is not the real name. It will always be Twitter to me.

And then through a complicated series of events, a lot of the people who are extremely vocal about the practice of journalism on Twitter have departed to other platforms. And it is no longer the case that the production function of journalism happens so tightly with the distribution function of journalism happens so tightly with the meta discourse around journalism and policing its boundaries and similar.

And I think that has caused a little bit of a change in temperature and a change in attitudes, even outside of other things that one could attribute it to like say the election results and feeling on many parts of the political spectrum. Every four years we get an update about what the American polity thinks of things and maybe we should take some consequential actions. Got some information.

Kelsey Piper: I think I would agree with that, that there was an independent effect from a bunch of people departing Twitter. I think it is reasonable to... if you say something and a couple of your coworkers are saying, Kelsey, that's not a good take, I will tend to rethink. I'll tend to be saying, okay, well I can keep in mind that if people I respect think it's not a good take, that does change my thinking on it.

But sometimes I'm saying, no, I think they're sort of missing one point for another point here. I have some takes that are strongly held that I've lost friends over, had people say they don't respect me anymore over, and I thought pretty hard about it. And I was saying, no, but I think I'm right though.

But if a hundred people have that reaction and are very vehement about that reaction and are very viscerally disgusted, and this constitutes a large share of your professional community and the places you work and might want to work and the people who might subscribe to your work and everything, the chilling effect is much larger.

And I think even if you are a very courageous person who is very willing to say what you believe is true, regardless of what the mob says, the doubt is gonna be larger. It's much easier to say, okay, I generally respect this person, but I actually think they just got this call wrong and I stand by where I'm coming from than to feel that way if everybody you know is saying, no, no, delete this, this is the most embarrassing and wrong and terrible thing you ever posted.

Patrick McKenzie: I think the echo chamber can also give the impression that everyone, you know, believes something when perhaps it is a hundred people on Twitter who are as representative as a hundred people on Twitter will ever be. Plus maybe one person internally who might even say, not that I agree with it or anything but we're being dragged on Twitter. And that's a reputational impact on us and we should avoid getting dragged on Twitter 'cause, and then they might say the following, very quietly, I would like to still attend nice parties in Brooklyn. I can't... I'm friends with you.

(If you can pause this helicopter's way too much. Oh yeah. We'll just give one second to pass. And also I'm gonna make sure everything is working. Sorry to interrupt, but it's... no worries. Thank you. Very, this is exactly why I have a person who, somebody the room pay attention to audio type stuff to peek and see we've just been gone for a while and just circling)

And these sorts of things that people will start to say about you if they have gotten mad at you about a fairly milquetoast liberal take are quite extreme and quite upsetting. Rumors will start that you are secretly a Nazi, that you secretly hold a number of abhorrent beliefs you don't hold. And it's upsetting to see people assert these things about you. And it's very unproductive usually to say no, I don't believe that, that's not true because what you really believed wasn't really an input into that process in the first place.

Patrick McKenzie: Take it from a Republican. No one who has ever said the word fascist has been convinced by arguments that they're not talking to a fascist.

Kelsey Piper: I have mixed or complicated feelings here. I have a lot of sympathy for use of even very strong language about fascism to describe some people who are influential in the current administration, the ones who have said things that they are opposed to democracy. That they think it was a mistake for the United States to seek the will of the public in expanding the franchise from white land owning men. The people who fundamentally do not in fact believe in our democracy. Many of them are fascists.

For this reason, it would be kind of nice in my opinion if we had been quite careful about reserving that word for those people. Perhaps not a Romney or a George W. Bush or...

Today I saw a proposal for how the city of Charlotte could... it was in the Atlantic, it was from Charles Fain Lehman... how the city of Charlotte could adjust its procedures to avoid a person with a long history of concerning statements suggesting not being in contact with reality as well as violent crimes would not be on public transit and would not... persona, this is, I think almost everybody wants this.

I have a lot of offline friends. I think this is very good both for them and for me because when I talk to them, I can get sort of, if somebody hasn't been through eight rounds of discourse about this already, what's their honest first reaction? And across the political spectrum it's well, yes, obviously we should adopt policies that ensure that people are not murdered on public transit. This is completely uncontroversial, even among people who are well to my left.

And I saw a bunch of responses that were saying, you can tell he wants to be a fascist in places, but the actual proposals are pretty good. And I was kind of angry to see that. I don't know what's in people's heart. I don't think what's in most people's heart is fascism. I think it's quite important that it, in fact, what's in most people's heart is not fascism. And I think that when there are some people who have said it would've been good if Trump had done a coup on January 6th successfully, it would be good if we gave up on the franchise, it would be good if America had no constitution and had state power to suppress individual liberty for the sake of the volk and the nation.

It would be good if the intense condemnation that we should lever at those people was not had nothing to do with, these are actually some pretty good policies to improve transit in Charlotte. And I got that bad vibes off the person who made them... you can just say, I got bad vibes off this. Not that you need to say that necessarily, but you could just say that and sort of reserve the term for an ideology that is in fact incredibly evil and that I think it is in fact pretty incumbent on us to notice and do something about where it's really influential.

Patrick McKenzie: I also think that combination of perfect should be the enemy of the good. And from just a consequentialist perspective, if you say that if you are for safe subways, then you're a fascist. And if you are for clean streets, then you're a fascist. And if you are for opening schools, then you're a fascist. People might say, oh, intriguing political system they have there where they get all the nice things up for nice things.

Kelsey Piper: And in fact, I think every single thing on your list is not only compatible with a liberal society, but it's obviously achieved by liberal societies all the time.

Patrick McKenzie: This is what we have for, we have the nice things, we still have the nice things in many places.

Kelsey Piper: Just that I think that is outrageously bad behavior that has very bad consequences. And I also think, yeah, no, it is totally within the power of a liberal society upholding the constitution and caring quite a lot about the rights of the accused and everything to totally solve these problems. We have the technology, we can do this. And it is good when we do so.

So yeah, I think the removal from Twitter of a very large contingent of people who had that particular approach to politics has made it easier to, on Twitter say things, wow, yeah, I feel pretty outraged that Luigi isn't being charged with first degree murder. This turns out to be about specific quirks of New York state law or whatever, but just, I think...

Patrick McKenzie: They essentially needed terrorism enhancement to get...

Kelsey Piper: So I was reading about it this morning because I was outraged that he was not charged with first degree murder, but that outrage was coming from my knowledge of California murder law, where if you plan a murder in advance, that's sufficient to make it a first degree murder. New York does not work that way, and it is if it's against a police officer or somebody else in various groups that we think it's especially bad to murder, or you are doing other particular, you have a history or you are in the middle of some other particularly heinous crime, and the one that arguably applies is an act of terrorism, which includes an effort to convince the government to change its policies or to terrorize the general populace.

And I don't know if the government has the evidence to prove in a court of law that that was the intent of the killer. It seems unfortunate to me that whether he can be sentenced to life without parole depends on their ability to prove those specific things about his intent in a court of law. So I guess I think New York should change its law.

But the point is I feel much freer than I think I once would have to just be saying, well that seems a really bad law. And expect that not that nobody will necessarily use very strong language to condemn me. It's the internet. There's always going to be some, but that this will not be the bulk of the professional community I participate in. That's nicely...

Patrick McKenzie: I feel facebound at the moment for... it is appropriate that society have moral appropriation for people who commit premeditated murder was a wildly politically controversial stance for a few years in influential industries. And that is also not something I would've said in tech. I think I would've had less risk than you folks working in journalism. But not incentive compatible, let's say. And it's always easy to just... easier to just go with the flow, particularly around things that aren't the core of the job.

Kelsey Piper: Well, and most people don't, in fact, want to be constantly getting into bruising political fights on the internet. I think this is reasonable and good. I think that in fact, most people take sanity damage from Twitter or from BlueSky and that the wisest thing in many ways is for them to check in through an occasional newsletter from a trusted source on what important things are happening in the world and not constantly try and do all of this.

But if all of the people who are saying, I am taking sanity damage from these websites because they're bad, log off those websites and proceed to not be on those websites, then there it can be, you can get a very misleading sense of whether your opinions are reasonable at all.

Patrick McKenzie: Very slightly different point of view on this one, because I think that for better or worse descriptively for the last few years, Twitter has been how the institutions that matter coordinate with each other on a sub rosa level. And so if one is simply going to one's office at it as a mid-level bank teller or similar, and just doesn't want Twitter because Twitter is bad vibes for them, that makes a lot of sense. But to the extent people were involved in policy making, involved with coordinating with the government, involved with coordinating between industries, your choices were either Twitter or the highway, or Twitter or the offline information travels at the speed of me or social network that we've used for forever.

Kelsey Piper: No, I agree with that. And even though it was socially toxic in various ways, it was the only game in town.

Patrick McKenzie: Yep. And I will give shout outs for even in the years where it was most toxic, vaccines couldn't have happened, but for Twitter, that's good. We probably wouldn't have been able to get, maybe wouldn't is a strong way. It would've been more difficult for us to get as good press as we did, as good cooperation from government as we did, but for the existence of Twitter. It would've been more difficult for us to fundraise, but for the existence of Twitter. And so I will say nice things about it in the course of saying I do not miss the battle days.

Kelsey Piper: I mean, I think Twitter is... there was a lot wrong with Twitter in 2020. There is a lot different, a lot wrong with Twitter now. It still feels to me in many ways people are being shaped by a very hostile environment that will jump on you for certain totally reasonable opinions, different totally reasonable opinions, but certain totally reasonable opinions.

But maybe having tried one particular toxic ideological rabbit hole and then another horrible, toxic, ideological radical... one thing I take some comfort in when I now see fairly outrageous and awful rightwing opinions on Twitter is that I do not actually think that Twitter becoming a toxic ideological left-wing rabbit hole served because that it was espousing at all.

And I kind of expect the same to result from it becoming similarly awful in the different ideological direction. Maybe in fact when people adopt these very echo chambery, very aggressive about disagreement, very nasty subcultures, and then they go out and do things in the world, normal people are saying, wait, you're doing what? That doesn't make any sense. It may be a Democrat, but I didn't vote for that. I may be a Republican, but I didn't vote for that.

And I think you see this sometimes you see the administration make some announcement or do something that feels extremely online to me. But then it comes into contact with the real world and everybody's saying, wait, you're doing what? That actually just seems a terrible idea though, right? Not that Twitter isn't real life, but nonetheless, if you spend too much time in a Twitter ideological rabbit hole, then when you go and try and do things in real life, reasonably often, you'll find that in fact, you are on the wrong side of an 80-20 or even more one-sided disagreement.

Patrick McKenzie: The posting to policy pipeline is not unique to Twitter and is very, very real. And has been since the internet has existed almost, at least since the blogosphere days in the late nineties, early two thousands, is when that dates to. But it certainly does, and not disagreeing with you specifically, but many people who use the phrase "Twitter is not real life," I think have failed to grapple with that.

Because Twitter is the mechanism by which people in power coordinate activity. Twitter is very much real life. The internet has always been real life. Every webpage on the internet is also a webpage that exists in the physical universe. But I think I would say the more substantive point is that I do wish that extremists on both sides will get quickly cured of their enthusiasm by the usual machinations of representative democracy. And I will also say I hold that view in a little bit of tension with the check every four years is great. But Twitter countries have very short, huge power over short terms and done very horrific things over short terms. And either of us would come up with many examples.

Kelsey Piper: Yeah, I think while I remain an incredibly enthusiastic proponent of liberal democracy, there are a bunch of ways that our system... the mismatches between our news cycles and the scale on which things happen through the institutional channels that are intended to have the appropriate safeguards is really bad.

The channels through which things happen with the appropriate institutional safety guards are very slow and very flawed, and should probably be notably faster for various kinds of actions. But at the same time, if you completely ignore them and just do whatever, then it's very easy to in fact do insanely destructive things. And perhaps this gets into our next topic.

Patrick McKenzie: Excellent transition into, and I hate to make light of it, but as one of the examples of things that one can do if one throws out all institutional safeguards, you could, I don't know, turn off a well loved program that has had bipartisan, full bipartisan support for 20 years, and you could do that in maybe about, I don't know, a day or three. What happened at PEPFAR?

Kelsey Piper: Yeah, so PEPFAR was founded under Bush. It was an effort to get AIDS medication to Africa where AIDS had reached epidemic proportions that are truly mind boggling, basically wiped out a generation of people in a lot of places.

In the US, AIDS is mostly the community of gay people, people who use intravenous drugs. It's very rare for it to be transmitted in heterosexual sex in the US. This was not true in Africa. There's a bunch of partial explanations I think we don't know yet, which was the dominant one, but it was being transmitted through heterosexual sex, and you had population prevalences that were 20%, 30%.

And when you think about that in terms of, oh, these countries are half children and oh, at least they're mostly not elderly people or whatever. You had just this absolutely awful, awful situation. And the Bush administration took this really seriously, and PEPFAR was an effort to fix it. And it was an enormously successful one, right? We basically stopped the rise of HIV infections in all of these countries. We have had a steady slide in HIV infections when it first happened, people were saying, well, it'd be mind bogglingly expensive, but the drugs at that scale are much easier to produce. You can in fact get the per person costs down extraordinarily low.

Patrick McKenzie: And the mechanism for this is essentially we spend money, we buy antiviral drugs, we ship to local partners. They distribute, right?

Kelsey Piper: Yes. And with a lot of oversight - working with local partners with a lot of money at stake is a very complicated thing in the development world. It's kind of easy to get your money stolen. But PEPFAR did a really good job. We audited these clinics. We know that for the most part, our drugs were going to people who needed them and our money was being spent in a way that we could keep track of. There are some USAID programs that I absolutely would not go to bat for. And if they had been canceled, I would've said great news. There's plenty more where I would've been saying, I think it was probably good, but I'm not... there's lots of stuff that's probably good, but PEPFAR was fantastic. It was far more cost effective than any other way I know of to save people's lives.

And disease control is, I think, a particularly appealing kind of saving people's lives because you are also not allowing a highly infectious disease to mutate and spread and make the entire world a worse place in a way that I think directly does also affect us all here.

Patrick McKenzie: So this was a point I tried to make to people who were concerned about the United States reserving the vaccine for itself back in the 2021 timeframe. If you morally weighted the lives of everyone outside the United States at zero, which I think is monstrous, but putting that aside, even if you weighted at zero, the computational capacity that the virus gets is mostly not executing on servers in the United States. And so you want to deny the virus computational capacity to do more mutations.

Kelsey Piper: And HIV mutates a ton. That is part of why it is so challenging to treat. And there being hundreds of millions of people who have HIV is very, very clearly not in our national interest. So I think PEPFAR was one of the most over-determined programs in terms of whatever framework you're using, even if you're concerned primarily about Americans, even if you're someone who just cares a lot about what are the cheapest ways to save lives in proof conditions. If you care about our programs, extremely well run and not corrupt, if you care about respect for the United States and stuff like that. This is always, I was a little reluctant at first to get into the soft power arguments just 'cause it seems very hard to quantify anything.

But I was eventually persuaded by someone in tech who was saying, look, think about it as the relationship that somebody has to one of their major investors. You might still not like the guy. You might still in various ways disagree with the guy, but the fact that that guy is one of your major investors does create a special relationship. It's gonna make you more reluctant to fight with him. It's gonna make you give him the benefit of the doubt more. It's gonna be a channel of communication. We are through PEPFAR, a major investor, basically everywhere. And this genuinely has about the effects you would expect. And that was, to me, an account of soft power that didn't feel very magical. It was just saying, okay, yeah, no, that's...

Patrick McKenzie: I will say for the other side of the argument, which, reporting not endorsing folks, that PEPFAR was situated mechanically under USAID. Right? And...

Kelsey Piper: It's complicated, sort of, yeah.

Patrick McKenzie: USAID has, let's say, a portfolio of programs. Many of them are justified by the desire for soft power. Many of them are speaking candidly, outgrowths of the United States National Security State. And yes, I think a discourse that happened about PEPFAR was, when people bring up the soft power angles 'cause they have nothing else going for it. And that I will say a reasonable person could look at the portfolio of programs and say the effectiveness of those soft power programs is extremely new.

Kelsey Piper: I would actually wholeheartedly agree with that. And I think if the only argument for PEPFAR was the soft power one, I would not find it very, very persuasive on its own. Although I do think it, I've come around to the perspective that it exists and is, in the case of PEPFAR, good. I think in the case of everything else, most American soft power is in fact a product of American media being watched and argued over everywhere in the world. That is very powerful. But I don't think we've made very many efforts to strategically leverage it in American interests. I guess selling, which is a...

Patrick McKenzie: Formal program of the government in Korea and Japan. Yeah. They both understand that and...

Kelsey Piper: They've been very successful with it. In describing success in terms of, everybody I know consumes a lot of Korean and Japanese media, but...

But yeah, so. Setting that aside for a bit, most... 99% of why I think PEPFAR is good is 'cause it saves a lot of people's lives very, very cheaply. And I will defend it on other terms because I would like the coalition for it to be big, but the place I'm coming from here is that it is super cheap and it saves tons of lives. We drove infant coffin manufacturers out of business throughout Africa, and I think that's awesome. You wanna talk about what makes America great, I actually have a constantly revised six part list of what makes America great, but PEPFAR's always been on there.

Patrick McKenzie: So, and apologies for not knowing the timeline as well as you do, in the January/February timeframe when it was very briefly on the chopping block before getting chopped... I think that you got together a coalition of willing, let's say, to write a report about PEPFAR and distributed. Can you tell people how that report came to be?

Kelsey Piper: Yeah. So, immediately after... DOGE. Began or DOGE? What, what are we saying?

Patrick McKenzie: I think they pronounce it DOGE, I'm not exactly sure.

Kelsey Piper: Immediately after this is the problem...

Patrick McKenzie: With naming government departments after internet memes that have no canonical pronunciation. I'm sorry, that's an aside.

Kelsey Piper: After DOGE started doing things, one of the things they did was the USAID shut down. PEPFAR is not technically part of USAID, but a lot of PEPFAR's programs were funded through USAID channels. And the PEPFAR stuff pretty much immediately stopped. People stopped getting money and they stopped getting things that the money was supposed to buy. And Marco Rubio said almost immediately, we are preserving the life saving programs. We don't, we're getting rid of the... and there is a lot of USAID stuff that I would say is not aimed at life saving, or at least is not saving any lives and some things that I think are really quite bad. So I was saying, great. We're saving the lifesaving stuff. And what you're hearing from people running PEPFAR clinics is we're absolutely not good. Money.

So I think we started out with this very, very mistake theory approach. Maybe they literally just haven't realized that the mechanism by which they shut the funding down also shut down all of PEPFAR's funding. And this was not a liberal initiative.

Patrick McKenzie: If I can interject there, when you say mistake theory, you're referring to something which is common in this intellectual circle. The difference between mistake theory and conflict theory. Mistake theory being that when one sees bad events happen in the world, it is likely 'cause someone fumbled a finger in a spreadsheet somewhere and accidentally turned off the wire transfers. But no one would want to turn off wire transfers to PEPFAR versus conflict theory that there are two or more sides that have implacable moral differences where the level of moral difference is sufficient that someone is saying, no, actually I want to turn off those wire transfers.

Kelsey Piper: Yeah. And I think both are sometimes descriptive of the world. I think the mistake theory is often if something is a mistake, then it is often quite easy to fix by ensuring that everybody knows about the mistake. So often it is useful as a first pass to go, alright, let's make very sure that everybody knows that their wire transfers have cut off funding to PEPFAR and that this is gonna kill a bunch of people.

Elon is going on podcasts, he is saying nobody died because of the USAID shut down. This is not true. Maybe we just need to make enough noise that people learn that this bad thing has happened and then they will fix it.

So this was not just liberals, although a ton of my friends certainly got involved. We also had a lot of pro-life conservatives. I think...

Patrick McKenzie: Which from the historical record, a major constituency for...

Kelsey Piper: Yes, this is Bush's signature program.

Patrick McKenzie: Yeah.

Kelsey Piper: Like we specifically, at the time, there were a lot of protests about things that Trump was doing from liberals, and it seemed protests of things Trump was doing was a daily occurrence. And that's not to say I didn't agree with many of those protests. I did, but that was not in fact really what was going on here. This was all of the people who think it's good when we save lives very cheaply protest a thing which Rubio and Musk both told us was not happening in order to try and get it to in fact not happen.

Patrick McKenzie: Can I ask you a very silly question? I feel sometimes in political discourse people are saying, well, A says the world is this way. B says the world is this way. That's a push. How could one possibly know the ground truth? How does one know...

Kelsey Piper: Yes. So one of the people who was involved in the PEPFAR report was a missionary doctor at a clinic in Africa where they were experiencing their funding having been cut off as a consequence of this. And another person was conveniently available to work on the PEPFAR report because of having just been fired because of the shutdown of the aid delivery organization he was part of, which was not in fact a PEPFAR one. It was one for a different disease where we also cut off all work. But as a result, she had a lot of experience reading the grant list.

So as soon as Politico released a list of all canceled grants, we were able to go through all of the canceled grants and specifically all of the PEPFAR canceled grants and specifically see, ah, this is a PEPFAR canceled grant for the delivery of treatments to pregnant women so that their babies don't get HIV in Mozambique.

And then you go to State and you're saying, this one that was canceled, did you mean to cancel this one? And State did not get back to us. Which is not surprising in not itself evidence of malicious intent or whatever because State is very busy and Marco Rubio famously has eight jobs and whatever is in his heart. I think a new Secretary of State has a lot on his plate and there is almost no chance that these decisions were made at that level or whatever.

But it quickly became clear that a lot of lifesaving programs had been cut off, that those lifesaving programs had been cut off because somebody had made the decision that those programs should not get funding, and that what was going on was not well, we just want to cut off the stuff that doesn't save any lives and is part of soft power programs that we don't believe in or don't think serve America or we're worried our funding, our enemies, which they occasionally certainly do. But we are also cutting off a lot of these lifesaving programs.

So there was a protest in front of the State Department. I was not able to fly out to DC for it, but people did an awesome job. There was, and all of this was coordinated on Discord, which you guys had used for Vaccinate California. I think that was what made it...

Patrick McKenzie: The website for gamers gets used for all sorts of political activism, civil society projects, et cetera, et cetera. Simply because it's what a number of people already have installed that in their phones. I had to install it surprisingly, not for video games, but to do the health infrastructure.

Kelsey Piper: Well, and people haven't installed already, but also it's Slack, but free. And that is extremely valuable if you're just trying to get lots of people set up on something quickly.

So yeah, we were on Discord. We wanted to write a quick summary of what PEPFAR does for people who had maybe never heard of it or had heard of it in passing, but we're not deeply invested in it. We wanted to help people be able to distinguish it from every other USAID program. You may... so that somebody could reasonably go, I think most of the USAID programs being cut off was fine and we would be saying, I think you will still agree with us. The PEPFAR being cut off was bad and we wanted it to be something where people did not believe that we would have exaggerated the numbers to keep PEPFAR going.

And we had some people with an economics background. Lots of different fields study the question of how well does this foreign aid program work? And there were in the immediate aftermath of the USAID cuts a number of efforts to estimate what we will see in terms of global mortality as a consequence of these cuts and.

To some degree these methodological disagreements, soon we will have the global mortality numbers. They aren't for the year 2025. They aren't that fudgeable. That will be pretty clarifying about which approaches were good. But definitely seems some approaches were more maximalist... they assumed that the cuts were all in place and would not be restored. And in fact many cuts were partially restored later. They assumed that there would be very little substitution as a consequence of a cut. And if a desperately poor person in an area with one clinic, that clinic shuts down and they're now supposed to travel 50 miles on a regular basis to purchase medication at approximately what they make in a week, every week they are not gonna get their medication and they're gonna die.

There's no substitution really available, but in a lot of cases, substitution is more available than that. Or the Gates Foundation has massively stepped up. Other organizations I'm aware of have stepped up now, it feels in some ways, a little frustrating or perverse or something to say, if the government immediately cuts off aid to a million people and then everybody drops, what they're doing cancels a bunch of other good programs and desperately flings themselves in spending far more money than the government would've had to spend to close half the gap, then the government only killed half those people.

But the government did only kill half those people. At some point, if you're trying to predict what the rise in global mortality as a consequence of the choices that the US made early this year is then you do have to consider all that substitution stuff. And I think you can consider it and then say also it was pretty bad behavior to...

Patrick McKenzie: Right. Even if one was of a political and moral persuasion where you had a list of United States programs that you wanted to cleanly exit. And you had PEPFAR on that list, and one was 100% convinced of the righteousness of that action. One would not have wanted to have this happen in two weeks based on yanking wires that were already approved versus getting on the phone with Bill Gates and say, Hey, Bill, here you've got a bit of money and a bit of interest in this area. We don't, we have neither the money nor the interest, and so you should be ready in six weeks from now.

Patrick McKenzie & Kelsey Piper: DOGE, PEPFAR, and Government Reform

Kelsey Piper: The difference would've been enormous between what they in fact did and cleanly exiting PEPFAR because they decided not to support it anymore. I still think that would've caused a ton of people to die, who it is very cheap to save, and I am glad for my taxpayer money to go to it. But the difference would've been enormous between what they did and any kind of clean exit. Because these other organizations did want to help and did want to step in, and it was very nearly impossible to figure out which money was going to go through at some point.

Versus having your program canceled and never receiving money again. A lot of programs lost funding for work that they had already done. And I think that will usually eventually be redressed unless the organization has gone bankrupt and is not able to pursue a lawsuit to redress it.

Patrick McKenzie: Which seems extremely unlikely to me.

Kelsey Piper: But that happens here and there. And an enormous number of organizations did go bankrupt. To be clear, the end result of all this was not the total cancellation of PEPFAR. After some initial upheavals, a lot of cancellations went out. A lot of those cancellations were rescinded. I think the state they arrived at was that the giving out of extremely cheap medication to people who would die without it would continue and is continuing at almost all clinics.

Obviously there are some cases where the disruptions in payments were long-term destructive, but the intent is that that's continuing. A bunch of the scaffolding that I think made that more effective and gave us better oversight of it is now gone. For example, we used to periodically test the people we were giving this lifesaving medication to because a small share of people with HIV will be resistant to the first round drugs and then you give them the second round drugs, and if they're resistant to the second round drugs, you give them the third round drugs. And thank God nobody seems resistant to those yet.

We are no longer, as I understand it, routinely doing that testing now. Especially from the perspective that one of the good things about PEPFAR is preventing this virus from evolving drug resistance, not doing that testing is a pretty bad thing to do. It's a little hard to calculate exactly how many people will die as a consequence of not doing the testing. But certainly some. I think we should have kept doing it, but they are still handing out the medication. That makes a big difference.

We also stopped doing pre-exposure prophylaxis—medication that prevents you from developing HIV. In some maximally sympathetic case, it's a married couple where one person is HIV positive and the other is not. And they would like to continue to live as a married couple. You can give them this medication and the person who doesn't have HIV won't get HIV and the kids won't get HIV. It's great.

The less sympathetic, or certainly less universally sympathetic version, is if sex workers are taking pre-exposure prophylaxis, they won't get HIV. This is good because the sex worker who has HIV both doesn't deserve to die because they're a human being and will not spread it to lots of people as a consequence of having a lot of sex.

Anyway, those programs seem to have been across the board broadly canceled with the exception of for pregnant women to prevent them from passing it on to their babies, which I think we as a policy are still providing in practice. When a service is widely provided, it is more reliably provided and it is a little bit hard to cleanly cancel 95% of it and keep the other 5%, but I think our intent is that we are still providing it in that case.

But then we did recently announce a deal for good prices on a different pre-exposure prophylaxis as part of PEPFAR. So the current state is not that PEPFAR is gone. I think PEPFAR is probably doing something like 60 to 90% as much good as PEPFAR was originally doing. A lot of unnecessary harm happened along the way. But we currently have most of PEPFAR. I think whether PEPFAR survives the next couple years depends substantially on whether a lot of people care whether PEPFAR survives the next couple years or not. But the end of the saga is not "it's all gone." The end of the saga is worse, but still saving a lot of lives.

Patrick McKenzie: I think that people might not appreciate how cash flow sensitive small businesses and small nonprofits are and how it is a week by week struggle to make payroll in some places. And they might not successfully model how much capacity for destruction there is if you simply delay checks by six weeks.

I think at one point we discussed how some people had the idea that if there's no money available at a particular clinic because people are so convinced of the sanctity of the mission, they will simply leave all the equipment in place, lock the door, and wait for there to be money to pay the landlord and wait for there to be money to pay the workers. And then when there's money in the future, it will reopen. Turn it back on.

I have very little experience with landlords in Africa. But the "no worries, just don't pay the rent for six months, I will absolutely not evict you, I will absolutely not sell the equipment to pay for back rent, I will absolutely wait until some definite point in the future when you can pay me" is not how I model landlords working. Anywhere.

Kelsey Piper: And in fact you get a wide range of accounts on the ground. But yeah, a lot of things go under if you don't pay for them. A lot of people are very literally living paycheck to paycheck and cannot afford to wait.

I talked to a reporter who was able to go to a clinic and talk to people on the ground who said that in fact, their whole staff had continued working without pay for the duration, partly because they were deeply decent people who didn't want to see their clients die. And partly because local unemployment had reached 50%, because almost everybody who spoke English and was qualified to work in the health services had been laid off at once. And that was a lot of people. This basically causes a local recession, right? If everybody in the health field, which is a large field, gets fired at once.

So you had a lot of people showing up to work who are like, "honestly, if I could find a job, I would no longer be showing up here to do my job for free. But the entire economy has completely crashed and there are no jobs. So here I am treating patients for free." So you get lots of oddities in individual locations as stuff happens or whatever, but for the most part, no. You can't shut something down and then turn it on later. I don't think anybody who had any understanding of what was going on could have expected that that would work.

I think a lot of decisions were made by people who were really not paying very much attention to the ground level facts of the situation.

Patrick McKenzie: I think it's been written about in many places that there is a legibility problem in government generally, and that people making decisions from spreadsheets and maps and increasingly divorced from on the ground truth will always and everywhere have a blinkered view of those decisions. When your decision is getting made by a spreadsheet that has been cropped and reposted to Twitter and you saw it for three minutes that morning, that does not improve the ability of the government to make its activities legible to itself.

Kelsey Piper: And I think that—all right. I did not actually start out wildly anti-DOGE. I know a lot of people who did, and in the local prediction markets, they all get base points. They were better than me at predicting the world. They were like, "Elon Musk is fundamentally a bad person who does not value any of the things we value. And he is not an institutionalist. And if he gets in power, he will wreck everything he doesn't understand, burn down a lot of stuff specifically because we like it and build nothing because he's too brain rotted by the internet to build anything."

And I was like, "Elon Musk has built a lot. Elon Musk has built some amazing stuff. We have a Tesla and it makes our lives better. We have a Powerwall and solar panels through Elon companies and they make our lives better."

Patrick McKenzie: It's pretty tricky to land a rocket. There's two organizations in the United States that can manage it. And one of them outdoes the other by a lot.

Kelsey Piper: Yeah, so I was on the side in that argument saying Elon Musk does real things that require being in contact with the real world. If there's anybody capable of not getting lost in a spreadsheet and just kind of treating the computer in front of him as the ground reality, I think he has an argument for it. And while I don't feel happy about much of the stuff on the Trump agenda, there genuinely is a lot of stuff that our government is doing badly and an infusion of extremely talented and motivated people who want to fix a lot of it could do a lot of good.

But the people with the most negative take possible here get a lot of base points. Because while it is still true that landing a rocket is hard and that Musk has built companies that do that, in government, they did not build anything that I can point to that I'm glad that you guys built. They mostly canceled a ton of stuff. And I think the average canceled program—this is dragged way down by PEPFAR, the President's Malaria Initiative, the tuberculosis work—but I think the average canceled initiative was above average for quality of government initiative. They disproportionately canceled stuff that was pulling its weight and then some. They canceled stuff that was really well run.

Where they departed from ordinary institutional practice, I don't feel that it let them do the "we're gonna build something better." It just made stuff worse. They fired a bunch of people and then often had to rehire those exact people because they'd made bad calls about who to fire. And if it had been that as a side effect of doing some good stuff, I think I would be somewhat willing to go, "well, you make mistakes, you learn from those mistakes and you improve."

But I don't know, are there things that you would highlight as having come out of DOGE that were good?

Patrick McKenzie: Nothing jumps immediately to mind. As someone in favor of—I'll say this as the person who wrote, stealing a line from Dan Davies, "the optimal amount of fraud is non-zero," which I think is a genuine challenge for our political system to say that out loud and mean it. But I don't feel morally attracted to any particular level of waste or fraud in government. And so there's a variety of ways that you could say, "I'm going to make waste and fraud in government the enemy, and I am going to hire some data scientists and we are gonna just whack at this piñata and save a lot of money."

And the statements that have been made by many people in the government, I'll say as a non-partisan, mostly professionally non-partisan observer of this, do not cohere with reality as to where that fraud exists, as to what the magnitude of it is, as to the mechanisms for it happening. And if you do not have a coherent map of reality, government is bad enough when it has a map that is attempting to represent reality and dropping bombs based on that map.

If your map is just whatever feels emotionally coherent to you in the moment, then your attempts to surgically remove fraud from the system will not work. They have no reasonable hope of working. And post hoc defenses of that just get increasingly removed from reality. And the current management of Twitter has made any number of allegations as to how local reality works in places like the Social Security Administration and many other places. And I often feel when listening to them, yeah, my prior that this is accurate is extremely low. Now because I keep score and people can be wrong and you've got a track record here.

And also in places where I'm capable of directly checking the ground truth or doing deep magic like reading the government report or two, they're straightforwardly wrong and failing for ways that would be straightforwardly obvious to someone who had spent a few hours reading the literature on it. And this feels very, very negative, both for the competence of the United States in accomplishing good things getting negatively impacted, and also what I was hoping for.

Kelsey Piper: An actively good outcome.

Patrick McKenzie: Yeah, I mean, why would one ever hope for not a good outcome from one's government?

Kelsey Piper: No, and being a little bit less of a non-partisan political observer, I think that the death toll of the decisions that were made is gonna be enormous. And I think that this was the consequence of a combination of lying and just literally not caring at all about the truth and prioritizing dunking on Twitter and looking and feeling cool and powerful over actually having any understanding of what was going on.

And I think there was no error correction at no point at which the people making these decisions looked at the disparity between what they were saying was happening and what was going wrong. When they got numbers off by a factor of a thousand, they would eventually correct them, but they would never reflect on what had gone wrong. To be wrong by a factor of a thousand about basically anything, you have to have a very wrong model of the world. You have to be wrong about a bunch of other things in order to let pass a wrongness of a factor of a thousand without instantly going, "that doesn't make sense because..."

And the fact that these errors would happen—

Patrick McKenzie: I'm a proud capitalist and if you misstate your quarterly revenue numbers by a thousand, even if it's somewhat in an informal fashion, you're not a very good capitalist.

Kelsey Piper: They did such a bad job. And the consequences are most obvious with the foreign aid programs because routinely delivering lifesaving medication is something where you don't have to stop routinely delivering the lifesaving medication for very long to immediately see the effects. But I think there's also probably lots and lots of other effects that are just a little bit harder to track because it is not that incompetence doesn't matter there too, it's not that destroying stuff isn't costly there, it's just that the costs are a little bit more distributed or a little bit harder to directly attribute than here it was.

Patrick McKenzie: And I will say this is a thing that has been in the discourse here. That people in some places, maybe not in this conversation, have conflated evil ends, evil means, evil intentions, et cetera. And the desire to say, "Elon Musk is a bad person and therefore he will achieve bad things." And perhaps accidentally gets at the correct answer on this one. I am frustrated that people didn't keep more of a laser focus on what was actually being done and the ways in which it was being done and attempted to make it about Twitter Slapfight 2.0 or focus on personal characteristics of people that they believed were involved in DOGE or similar.

The United States government has over the course of history accomplished many things for good, many things for ill, and taking a quick demographic survey of who was in the room when any of the ill stuff happened is not generally the most interesting thing I think that journalism could be providing to the American public.

Kelsey Piper: Well, and especially in this political moment, I do feel a very strong urge to be cautious whenever saying that a specific person is specifically as an individual because of their individual qualities responsible for a lot of harm. I think this is true sometimes, and I don't think it is an observation that is inherently not worth making.

But especially if you're making it about somebody who's a mid-level staffer who probably can't afford security, I do hope you are being really, really thoughtful about what specifically do I think happened and is there a reason that I'm mostly telling a story about the person who did it above the story about what happened. And I think often the answer is not malice or anything, but it's a kind of streetlight effect. Where to figure out what actually happened, you need to go to Uganda and talk to somebody who is holding a baby who was born with HIV and shouldn't have been.

And to figure out who did it, you can talk to his friends at the university he dropped out of, and they can say, "yeah, he used to be a nice guy, but then he got all Elon Musk pilled," and now you have a story that is easier to tell.

Patrick McKenzie: If I can go one harsher than the streetlight effect—from the first principles perspective, heavy is the head that wears the crown. Official acts of the American government are official acts of the American government. Policy priorities of the administration are policy priorities of the administration and therefore responsibility should roll uphill. However, the people at the top of the hill have resources and communications staffs and et cetera. And low level federal employees do not, and low level federal employees do not exist in a social continuum where everyone around them has been socialized "don't respond to media outlets, it never goes well for you or yours."

And so this is the streetlight effect, but a more pernicious version of the streetlight effect. You'll have a much easier time telling a story about someone who is low level than telling a story about someone who is high level, despite the fact that—I don't know how to say this—senior members of the American government exert a lot of influence on American public policy, and in a democracy, one should attempt to draw the link between senior members of the government and public policy and close the loop on accountability of that versus concentrating on who had their hands on keyboard for it.

Kelsey Piper: So I think I agree with most of that. I think we do have a partial disagreement here or something where I do want to say that a lot of how this happened is that Marco Rubio would continue making statements about what the official policy of the State Department was. And Marco Rubio would keep saying that PEPFAR should be continuing to run, and that there were waivers for these lifesaving programs.

And then on the ground, the fact of the matter would always be that there were not waivers for these lifesaving programs. And a person might at that point, reasonably go, "what is happening such that the head of the State Department is saying we are waiving the lifesaving programs and then on the ground the lifesaving programs are not being waived?"

And it seems like what is happening is that people filed these waivers and DOGE—where people who were pretty low level, but were operating with a fair amount of autonomy and making a fair number of decisions that would usually be made at a much higher level—were saying "we are declining this waiver" and in some cases saying things like, "I don't know, seems gay" or whatever.

Now I think I agree that Marco Rubio is responsible for the conduct of the State Department and that it is wildly more productive and more honorable and better for the functioning of our society to say Marco Rubio said he would issue these waivers and then ran a State Department in which these waivers were systematically not issued.

But I do think that the way DOGE marketed itself and the way that DOGE positioned itself was one where they were like, "we're gonna go in. We are gonna personally stand between all of these policies and your money getting spent badly. We are gonna personally cancel these programs." And I think that if the thing is that high level people had been saying had corresponded to what was actually happening, there would be a lot less temptation to look at these low level decision makers and say, "what are they personally doing?" Because it would be more clear that they were in fact following a policy.

Patrick McKenzie: I take your point there, and clearly to the extent that the American government is not functioning in the fashion that is described in its laws, traditions, and similar, that is certainly a reportable event. I think we run into some epistemology of media issues here. Where it is easy to get a statement from high level members of the government and digging into the ground truth of the statement is often hard. And so the statement will be reported, maybe possibly with an implication that "but who could possibly know" but very infrequently with the "so we called some people to check."

And so—heavy is the head that wears the crown. If one's organization is making statements which are systematically untrue, that should bubble up to the top. That was not the story the media told, pretty much point blank, with the possible exception of laying some of it at the feet of Elon Musk specifically, that also being an epistemology of media thing.

Granted, senior members of the American government, and I think he was ambiguously one for a bit, are responsible for the actions of the American government. It's enormously frustrating that—well, we have a tightly disciplined communicator over here who is saying the right thing. We cannot disprove the right thing. And then we have some people who are socially close to people who are not tightly disciplined communicators. I know what story I'm writing tomorrow. That achieves worse results on the level of an individual article, on the level of our understanding as Americans of how our nation operated here, and on the moral level of where should we apportion blame, praise, status, et cetera in our society.

Kelsey Piper: Yeah, and to be clear, I think that to the extent that Elon gave an enormous amount of discretion to random 22 year olds, and those random 22 year olds made a lot of really, really bad calls, but were trying to make calls in line with priorities that were communicated to them by Elon or by senior figures in the administration—obviously the senior figures in the administration and Elon are the morally culpable and the people that it makes sense to hold responsible for this and go after for this. And you shouldn't give enormous amounts of discretion to 22 year olds while also making statements that contradict the guidelines that those 22 year olds were explicitly or implicitly given to operate from. That's a very, very bad thing to do.

You find yourself as a 22-year-old in this position, I really highly recommend quitting.

Patrick McKenzie: Take very careful notes and strongly consider quitting.

Kelsey Piper: Yeah, sorry. Patrick has more experience with advising people on the wise course of action in extremely complicated administrative situations. But if the things that your extremely important boss are telling you to do or giving you discretion over are extremely not in accord with what they're telling the public the rules are, it's a very bad situation and I would not recommend participating under most circumstances in that situation where you have discretion over something while your boss is publicly claiming that it is being done a different way than that. Don't do that to yourself or your country.

But I do agree that ultimately the people who made these calls were Elon Musk and Pete Hegseth and Marco Rubio as Secretary of State. Having not apparently made any of these calls does not make him not the person who was responsible for all of these calls. And obviously under literally any circumstance in a democracy I condemn going after anybody by means other than arguing that they did bad things and should lose elections as a result.

But they did some pretty bad things. DOGE was enormously destructive. I am very angry about a lot of the calls that they made and about the thing where Elon would tweet "I could have gone to some parties, but I took apart USAID with a chainsaw." A lot of people died. It is good—there are decisions that are right to make that nonetheless cause people to die. There are decisions that are very complicated and it's hard to tell if they're right to make that cause people to die, but take it seriously, you know?

Patrick McKenzie: And we'll be dealing with the aftereffects for years. But bringing back a point about the consequentialist feeling of this kind of thing to the extent that people are worried about fraud in government and waste in government and similar—I hope this does not drain the well of enthusiasm for attempting to attract smart people into the United States government, attempting to do administrative reform, attempting to take a serious look at waste and fraud as a problem. And I feel something of a risk that the next time we have a conversation on—

Kelsey Piper: We'll be overshadowed by "you're just gonna do DOGE again."

Patrick McKenzie: If DOGE was a catastrophe, aren't you just one of those tech bros who thinks that they can control-alt-delete democracy and then put fingers straight in one's ears? And that is—

Kelsey Piper: A presumption that you think you do actually have to overcome, right? It's not wrong for people to be like, "well, last time I gave you a little bit of the benefit of the doubt, and you used it to be enormously destructive and dishonest, so now you don't get any benefit of the doubt anymore." That seems like a reasonable thing for people to—

Patrick McKenzie: And I felt like we at VaccinateCA were running up very much against the "tech bros don't know anything about the world and can't do public health administration" back in 2021. And I don't think as a moral statement that you can reasonably attribute Elon Musk's views to myself or vice versa for obvious reasons. Elon Musk does not endorse the contents of this podcast internet.

But that argument was made, and I think that argument was occasionally made cynically in the interest of people preserving their status and power. And it annoys me that people who might do the wrong thing in the future and say things in the interests of preserving status and power will get more of a sympathetic hearing, even more power and authority given to them, because there is an existence proof that, "well, the one time we brought in the tech geeks, it didn't work out very well."

I think people will remember examples of history where it actually worked out pretty well, I think, on all sides of the aisle. Okay, I won't say that. More bipartisan citizenship than exists. I think the blue side of the aisle might remember the healthcare.gov rollout and say, "I was pretty happy about that one." And I don't think anyone even much remembers VaccinateCA but selfishly I think it worked out.

Kelsey Piper: And I am really glad you guys did VaccinateCA and I think tech is where a lot of brilliant people who really have mastered a lot of ways to accomplish stuff in the world are. It makes no sense to have the government on a basically hostile footing with a large population of hardworking law-abiding young people who want to do stuff.

Patrick McKenzie: I think honestly, it's been an enormously frustrating element of the last couple of years. I think we in tech didn't do ourselves great favors because for a very long time we steadfastly denied we had power. And I think I can hold the idea in my head that we never sought to get it per se. We were just in it for the technology, just in it for the companies, et cetera. And then woke up one day in a position of substantial power. And I think we were judged harshly for that, in part because people thought "you can't possibly be as naive as you present as, or you can't possibly have not intended to get here, given your actions of the last couple of years."

But be that as it may, we should acknowledge that we have substantial power and bring that to the benefit of our nation and world. And it's frustrating when we incinerate our social capital, which was already under attack for no good reason, by doing poorly considered actions with—not merely poorly considered actions, but poorly considered actions which have the direct effect of causing people to die.

Kelsey Piper: Yeah. So I have thought a lot about rebuilding American state capacity in 2028. If people who listen to me have the opportunity to do that, I would like to have a bunch of proposals for them that are like, "here's some stuff that I think we could do that would be good." I think we should have a civil service exam. I think we should change a lot of administrative rulemaking, so my impression is that currently a lot of administrative rule procedures are just not being paid attention to. And I think that they have safeguards that are important and there were some reasons for those safeguards, and I don't think the consequences of ignoring them will be all good.

I also think that this was a very slow process. And processes being very slow is in fact quite bad and costly. And processes being completely inflexible is bad and costly.

Patrick McKenzie: And if I could get an explicit point here, previous podcast guest Dave Guarino has talked with me about some of the safeguards around administrative rulemaking, which will do things like make it impossible for the government to conduct user surveys of software. Because if you survey people about their use of software, they have become a guidance council. And there is a rule about guidance councils, which is meant to protect the American public from having their decision making captured by, say, unaccountable lobbyists in the back room, which is used to then say, "well, you can't show software to—I don't know, software about food stamps to a single mother who will apply for food stamps and see if she gets through it correctly, because that will give her too much power over the American government."

Kelsey Piper: So Dave knows a lot more than I do about all of the ways in which the system is absurd. I did not know that, but it does unfortunately not surprise me to hear it. And he has been doing heroic work for a very long time on all of the details of how this works. And I think if you were gonna go to somebody for reform proposals, he's a great person to go to.

But yeah, I would like a world where before you roll out a benefits application, the obvious thing to do is go ask a bunch of the potential beneficiaries to fill it out, watch them do so, and see where they get stuck. This is the standard that I use when checking whether my 8-year-old can use some educational software. And I don't think that the government should be constrained to not do that. And you get these forms that are egregiously badly designed that are hard to read when a lot of the people who need them are marginal readers.

Patrick McKenzie: Fifty-seven stage form for getting on the list in New York State for the COVID vaccine, which required uploading multiple attachments when the vaccine was being targeted at 75 year olds who might be less than digitally native. Or dealing with cognitive impairments and similar as a result of aging.

Kelsey Piper: It is really, really bad and I think it both needs fixing and I think it is a natural fit for liberals. I would also be excited about Republicans taking it up, but I am unlikely to be the right person to persuade them to do so, and I don't think they've got a lot of administrative reform focus right now.

But if people who are listening to me are in power, I really, really want everything that you apply for to be possible to apply for in 20 to 30 minutes. Ideally, not requiring you to reenter a ton of stuff you did elsewhere, although a lot of that is about state-federal communications, stuff like that. And I think it should just be considered a pretty egregious failure if a person who needs a program tries to apply for the program and cannot do so because it is too confusing. Either we shouldn't have the program or the program should be available to people who are not fantastic at filling out really long forms. This is—

Patrick McKenzie: I think we paper over this in a lot of places by paying for people in the non-profit sector to act as—for Obamacare, they're called navigators. And for many different benefits programs, they're called different things. But we paid people to make other people legible to the administrative state, which I think we should view that as sort of a signal that the original program is misdesigned.

Kelsey Piper: And of course, if you have somebody who's developmentally disabled and they need an assistant to interface with society, that's—we should provide that. But that being the default for 20% of the population because we made the rules too complicated for them to follow is pretty outrageous and pretty bad.

Patrick McKenzie: So if I can ask a different question, so there are many people that are having, let's say separate conversations, but in the same general realm about—how would you situate this desire and the argument for a renewable—one side of the aisle against or in tandem, more similar in the discourse with the abundance agenda, for example.

Kelsey Piper: I think the argument is super abundance aligned. I think a lot of people have sort of—you know, are trying to draw all the connections in the abundance sphere and there's plenty of yarn you can draw, but the bigger thing is I think abundance was really conceived in response to a bunch of the Biden administration's projects just literally not happening.

And I don't—I know that a common on the Republican side of the aisle, a common story is they don't even want to build high speed rail. They're only passing high speed rail programs so that they can give lots of money to their buddies who run high speed rail environmental impact surveys or whatever. I certainly cannot speak to the character or motivations of every person involved in a big process. And the more money and the less results are expected, of course the more grifters will show up. But I know a ton of people who really, really wanted to build high-speed rail and who worked on the Biden infrastructure proposal. They considered it critical to the survival of tens of millions of people that we do a climate and energy infrastructure transition and all of that.

I should note here, I think climate change is going to kill a lot of people in poor countries. I don't think climate change is going to kill a meaningful number of people in the United States of America, but I do think it's gonna kill a lot of people in poor countries. I think that's bad. So I'm pretty motivated to switch to solar as fast as is feasible and stuff like that. Just add that because people have very, very different baseline assumptions about what we mean when we say that climate change is really bad. It will kill people and I want to sort of stake out what I mean.

But people really wanted this. People cared a lot about it. They believe it to be important. They believe it will save a ton of lives. They passed the bill, nothing happened.

Patrick McKenzie: And they extremely enthusiastically ride high speed rail when they're in Japan and Europe. And they're like, "we can do this."

Kelsey Piper: "We can do this. Doing this is good. Doing this will make our country a better place. It will make it a fairer place. It is an expression of our values to have good high speed rail and good solar and good industrial policy." And they were really, I think, often confused by how hard it turned out to be to get this stuff done.

And I do think you're gonna have to fight a lot of entrenched interest groups whose interests do not align with really building this stuff. But I think there's a huge coalition that really wants to build this stuff. And I think the core motivation behind the book was, "okay, this isn't working. We've got to figure this out. Here are some changes that we can make."

And I think—okay. I also think that—sorry.

Patrick McKenzie: Oh, no worries. I think that as has been pointed out by critics of abundance, one of the parts of the book is it's quite easy for someone to wave wildly at interest groups—or "the groups" as it has become known in the discourse—and say, "oh, it is these powerful shadowy figures that we don't want having excessive influence over policy."

And it's more rough to say, "I'm brass tacks. No. This is a coalition management issue and the Sierra Club is a group." We've instituted laws like NEPA and similar that allow people to have a private right of action to stall any project anywhere. And that has made us into a vetocracy on large scale building projects. And we should roll back the vetocracy, which will necessarily mean that people stop having the ability to sue simply 'cause there is something in their county that they don't like.

Kelsey Piper: Yes. And I think we do have to roll back the vetocracy. We do have to, at minimum, very much constrain when you have a private right of action just because somebody's building something on their own land that is somewhere near you. You have to, if you're gonna have these—I'm kind of okay with there being a feedback stage, but I think it needs to be very, very time boxed. And not a veto, but an input stage. But the problem is once you have an input stage, then I think these often drag out and become these insane—

Patrick McKenzie: It becomes veto by whoever can filibuster the process the best. And that is literally in Rules for Radicals. Here's how you filibuster a process to death by using a public account.

One of my most culturally dislocating things in the brief window when I was living in San Francisco was that I got a city of San Francisco funded postcard in the mail. "Someone nearby is building an apartment. Would you like to come to lodge an objection to it?" And I'm like, my Tokyo brain is blowing up right now. Why? One, there's a pandemic on, perhaps you've heard of it. Two, why would I object to someone in the city building an apartment building?

Kelsey Piper: Well, and it's also one directional, right? I walk in my neighborhood in Oakland past some buildings that have been vacant for a long time, and a couple times I've tried bothering the city council. I've tried saying, "this vacant city owned property, what's the deal with it? Would you sell it to me?" And they're always like, "oh, there is a plan in the works for it." And I read something from East Bay for Everyone, which is our great local YIMBY group, sort of breaking down why it takes so long for these things to happen.

Part of the answer is that you have your plan and you can execute on your plan, but if you have to modify your plan in anyway, you have to go back to the city council. And this is not a "you have to loop the city council in on an email." You have to make a presentation to the full assembled city council, which takes six months to have your team prepare all of the relevant documentation for, in order to say, "we are changing which landscaper we're working with for this project," or whatever.

Now, as a consequence of this, even though the city of Oakland has all of these city owned properties that they would like to develop and frankly has plenty of money with which to develop these properties, these properties are just sitting for a very long time. And I have no real—if I wanted to delay this by showing up to these meetings and objecting to what they were doing, it would be pretty easy for me to do that. My objections would be taken very seriously. I have no symmetric way to show up to these meetings and say, "I want ground to break next week. I want you to do stuff." Even though in fact these properties being vacant is quite bad for me as a resident of Oakland who would like these properties to be in use.

It is not just that there's lots of community input. It is that the community input can kind of only move things backward and not forward. One thing the YIMBYs have sort of coordinated on, at least in our area, which I think is awesome, is when these community input meetings happen, it will now be half "I don't want the apartment building because it will make my life worse in some way" perspectives and half of these who are like, "I like it when there are apartments in our community 'cause I have friends and I want my friends to be able to afford to continue living here." Or "I want to be able to get a bigger apartment so I can have kids. Please build your apartment building." And it used to be just the noes.

Patrick McKenzie: I was considering doing this in Chicago now back in the nation where I had the franchise again and just going out to the meeting every few months and saying, "Hi, representing business owners and employed professionals here. We love people being able to live in the city. Keep doing what you're doing. Thanks much. Bye-bye."

Kelsey Piper: So I think this is maybe not the best use of your time per se, but I have heard that it is impactful in these planning and community input meetings if half of the people who show up are like, "we're excited about people being able to afford to live here." That is a different dynamic than when it was all people saying no. But you still have a process that has far more places you can stall it than places where you can move it forward.

Patrick McKenzie: Absolutely agreeing that the processes are structurally—to use the word that gets thrown around a lot—biased in the direction of not doing anything. Well, doing the very particular version of something which is—choosing not doing anything from a constructive perspective, but we have to be optimistic about the potential of democracy 'cause it's the only form of government that we have. And I think people often underrate how much a small number of voices in the right places in the political process get listened to.

I have friends and family who have worked in state congressional offices. The person who sits by the phone and takes down constituent complaints or comments on upcoming legislation—and the number of people that have to say something that sounds intelligent into a telephone to get it told to a congressman by voice later that day is like plus or minus four.

Kelsey Piper: And people constantly ask me, "what can I do about PEPFAR?" I'm like, "I do think it is impactful to keep calling your representatives and saying, 'Hey, I'm still really worked up about PEPFAR,'" because yeah, it doesn't take that many people for a Congress person to hear, "Yep, your constituents still care a lot about this."

Patrick McKenzie: And we have recent evidence of that. Both sides of the aisle in many different places, for better or worse, one of the reasons that the Republican party has swung in a very anti-immigration direction a few years ago was that the people who were quite annoyed with the bipartisan consensus on that issue called their reps frequently and made their views known on that.

Kelsey Piper: Yeah. Well, and I think while in many cases I disagree with the majority opinion in my country, I think it's straightforwardly good that people can influence their country by speaking out for what they believe in. That that is a—

Patrick McKenzie: It would be a really rough version of democracy we had if that wasn't an option, wasn't if it's literally just you can pull the lever every two to four years.

Kelsey Piper: I don't think that's enough. I think you need to be more frequently than that, and I actually wish I could get into all of the ideas I have for more democratic democracy, but we are getting a little bit close. I have a two o'clock.

Patrick McKenzie: Well, I'm sure we could keep this conversation going for a very long time, but I suppose that is what the argument is for. Where can people find the argument on the internet and you more broadly on the internet?

Kelsey Piper: The Argument at theargumentmag.com and it's on Substack. Subscriptions, including free subscriptions have a huge impact on us and we value them pretty highly. If you're interested in what we're talking about. I'm also on Twitter at Kelsey_Tuoc, and nowhere else at this time.

Patrick McKenzie: People have been trying to get me to a sky that is colored somewhere or another, and I just don't know if that will add anything to my life.

Kelsey Piper: A friend said to me earlier today, "I checked out the Blue Sky and I was already on a ton of people's block lists because on Twitter I follow Ben Dreyfuss and he's on their bad list. So everybody who follows him is on their bad list." And I was like, "I don't particularly want to know what got me on people's block lists on Blue Sky. I think I'll simply not go on Blue Sky and then never learn this and that will serve me pretty well in life."

Patrick McKenzie: Well, Kelsey, thank you very much for being on the program today and for everyone else, we will see you next week on Complex Systems.

Kelsey Piper: Thanks so much. It was always great to talk with you.