« The Weeds

Who models the modelers?


Ezra and Matt on the failures of some popular epidemiological forecasts and a viral Silicon Valley essay urging America to build again.


"The US has a national service for predicting the weather. It needs one for predicting disease." by Brian Resnick, Vox

"What happens next in the coronavirus outbreak? We mapped 8 scenarios." by Julia Belluz, Vox

"It's time to build" by Marc Andreessen, a16z

"Why we can’t build" by Ezra Klein, Vox


Matthew Yglesias (@mattyglesias), Senior correspondent, Vox

Ezra Klein (@ezraklein), Editor-at-large, Vox


Jeff Geld, (@jeff_geld), Editor and Producer

Open courtesy of StatQuest

More to explore:

Subscribe to Impeachment, Explained on Apple Podcasts, Spotify, Stitcher, Overcast, Pocket Casts, or your favorite podcast app to get stay updated on this story every week.

About Vox

Vox is a news network that helps you cut through the noise and understand what's really driving the events in the headlines.

Follow Us: Vox.com

Facebook group: The Weeds

Learn more about your ad choices. Visit podcastchoices.com/adchoices

This is an unofficial transcript meant for reference. Accuracy is not guaranteed.
Support for this episode comes from click up. We lose an average of three hours every day, switching between all our work apps, but you can get them back with click up a flexible platform that brings all your essential tools in one place. We can prioritize tasks, collaborated, docks check with your team and track goals, so companies like Oberon web flow use click up as their mission Control Centre replaced every other. After we're using before cook up even guarantees to help you save one day week and get more done, it's completely customizable. It's free forever. So try click up today at click up dot com, slash the weeds support for those others. Outcomes from INDIGO at INDIGO brings together leading companies committed to activating the full potential of agriculture to address the climate crisis in partnership with innovative businesses, farmer, scientific partners and informed advocates. We can re. Imagine agriculture for the benefit of people and the planet. Farmer, Casey, Bryant Bamberger says our industry hives biggest platform,
change our environment and working towards a better future, learn more it indigo Ac Com Recode. What is a model when I was a kid? A model was a toy that I glued together when I was a little older model was someone who wore fancy close now that I'm an adult the term model has more to do with math and statistics hello, welcome to another episode of the weeds on the box media podcast network. I met who Iglesias here with as recline put up by building things at the end of the show and that is again a good discussion online about that. But but first I think, as you and I are both them- we're not epidemiologists or quantitative models of anything
but have been interested in some of these efforts to make projections about what is happening with colonel virus and the way they have captivated are hurting. Him in the sort of policy world and then I think in many cases proven to not be not be very good guides to anything and it's it's a tough price What right I mean I this is like I'm a journalist. I, like I, try to talk to experts and figure out what's going on, and you know a sort of frustrating reality This is that it seems like the top experts. Don't really know how to tell us. What's gonna and they don't know what's going on so I want to frame this conversation in two ways. One there's a line out there that I like- and I think going, could be very important here, which is all models are wrong, some are useful and to the quest dealing with with a model is not. Is it going to be right? I mean a model by its nature. Is a simplification of reality and when you're modeling, something like krona virus
Dont know almost anything about it. Early on in an epidemic, you're gonna be making assumptions going to turn out to be wrong. One of the questions here for our ability to learn from things are we able to structure this conversation things can be wrong and then people can update and people can listen and an end of all, because we going to have to do that and in all directions. But I don't even think this is really just a car. The models? The models are reflecting something broader, which is an when state of knowledge, when people have to make very certain very consequential decisions, very fast right? We don't know nearly enough to say How should we shut down a reopen economies? How should we treat people best like what should the exact social distancing guidance be in a city versus in a rural area versus in a suburb? But we have to make those recommendations anyway,
and so then, when the models have to be updated, there is understandably a lot of frustration. But let's talk about, I think what has been the central model here: the Ihme model, which has been revised down from predicting something like two hundred and forty five zero deaths to predicting something like sixty zero and in both of those cases of predictions included. Social distancing effort so that is predicting a much less lethal outcome for the virus in in America and that revision, that's a big deal. Yeah It is a big deal in its it's become a big deal politically because they, the Trump administration, is citing it as evidence of their own successes. Other people sort early reopen skeptics of social distancing, are citing it to say that. Well, we got spooked into taking these dramatic measures that were unnecessary by people who basically didn't know what they were talking about, and the interesting thing is, as I've tried to dig into this model
about about the Ijm model. Is they are attempting to cope with the lack of information about the underlying virus in an interesting way, which is, you might think, ok to build a model. Disease. I need to know a lot about its properties and I then I need to know a lot about the specific situation and out of this kind of Micro founded model- I can say, what's gonna happen agent, He says we don't need to do that, which is lucky because to an extent we we can try, and so they have this kind of pure mathematical curve, fitting exercise, which is that they looked at early outbreaks in China and Northern ITALY and
they plotted the right. So did they just sort of model? Is a mathematical system like this is what happens and then you look at? U S, states and you fit the data points from the? U S states on to a purely mathematical model from the earlier countries, and you say: ok, here's what I think is gonna happen here if this wasn't a crisis, I think we would say is an abstract intellectual exercise. This is worth a shot. It's a promising ideas, not obviously absurd, and it gets you around a lot of practical difficulties with the kind of trying to build
a bottom up model, there are a lot of systems in the world that you can model this way sort of on a high level and ignoring the details. I think what we've learned from Ihme as it's been revised several times as more american data come in. It's just that it doesn't work separate from, like all models, have uncertainty associated with them, but they have confidence intervals on this model and they modeled each of fifty states separately. So you can say: okay, well, did ninety five percent of the states fall within that confidence interval, like is this model been specified correctly and in seventy percent of cases it was outside the ninety five percent confidence interval, which is that's a good big mistake. So in seventy percent of cases the outcome is proving differently than the outcome. The models were ninety five percent certain would have get outside that year. Soldiers, it just
the bottle is wrong right. So that's not just like in precision in your modeling. It means that on some higher order level, like you didn't know what you are doing here- and I mean I don't wanna be too harsh on this, because it seems like a promising approach like I don't know that you would say: exceeded be that as this wouldn't work, but I think it clearly hasn't worked and that, what's going on now, where they just sort of keep tweaking it with new data, I mean, maybe it will eventually start to make more reliable forecasts, but the real thing we we learned there is that this approach is, is unsound, around, and that you actually need to try to no more honour on a micro level about about how this works. But let me try to take her, not differ interpretation of the model, but maybe offer like a theory. What happened in that model. So
mention the model is built on what happens in who hunt China and then what happens in northern ITALY and the model predicts New York City pretty well actually of the things that are predicts effectively predicts New York City, and then I think this gets to what is to me. The single biggest question right now, which is why has there not in America, been another New York city I mean even the place, have gotten really bad New Orleans are not New York City, bad and definitely other places. It look in many ways like New York City. They had early, identified, spread they're pretty dense? They have a pretty significant usage of mass transit like San Francisco. Have not it now Is bad is New York City and, they might say, say s SF closed down early, but then Florida closed on April third, and has this very senior heavy population wise in Florida, told disaster and there actually I've been reporting on the source? Not a great answer
this question, but what seems to me to be happening in our colleague, Brian Resnick, we're a very good piece on on modeling and modeling uncertainty and his big. Between that piece, peace, that the thing models have trouble with this human behavior and practically what happens if human behavior changes in a systematic or systemic way, and one theory I have about what's going on, but GINO comes from talking to people, than I do about it. Is that what was happening hang when societies got hit by this hard before they knew what was coming and before they had a human behavior had begun to change in reaction to it is very different than what is happening now in societies get hit by it or city states whatever and social differences is either statutorily.
In place or people are doing it themselves, which was happening in Florida to some degree, even before the governed, lock, the state down and health systems know better what to look for excetera, and so there may just be something here that what happened in the ig model is that it is true that, if systems acted, the way, ITALY and whew Han and New York City did that you would have those bad outcomes, but that, in this way, where you now had this like move of human hey into into this very different space for people like washing, Hansel, odd and staying away from people and quarantining if they feel sick or not going near friends who have had a cough, etc, etc. Like maybe you stop the? go back to spread from happening. More easily may be. Super spiders are not getting around in the way they were, and so like we ve, like can actually done like a hard shift into a new equilibrium that a model built on the sort of legal places it got. The first heads of it,
had predict for you. I mean, I think that's true. I mean, I think, there's also other extrinsic factors I mean, I think there are clear, genes about how the weather and ambient temperature impacts different kinds of things. You know their questions about the role of of uv light and how that impacts different things and on some level, though I just think part of what seeing is that we don't. We don't know all of the relevant issues. I mean who, at the very beginning of this, something that somebody said to me was that they thought. We were seeing someone spread in ITALY, because people kissed each other as a customary greeting there, which is in the case in every country, that seems to me to have completely dropped out of the dialogue. Still, when I look at a map of Europe, I see the kiss greeting countries as having much worse outbreaks. Then the handshake countries- I don't know like I'm, I'm not negative allergist, but like what one thing
that modeling is that, like you choose a model right to simplify, and so you have to make choices about what things you want to put in and what things you dont want to put in and usually over time. As we understand systems better were able to make reasonable judgments.
About what kinds of factors to include or neglect, but we don't know that the science of the virus is just not good right now we have not had people working on this for decades and we don't know very much about it and I think it just turns out that extremely high levels of abstraction are not capturing something. That's that's really relevant here. I was struck by the fact that you know like like a lot of people. I thought I heard a lot about a bowler in twenty fourteen, but by the time that outbreak occurred, scientists had been working on a bowler for thirty five years, but something recently there eventually had really
large apple outbreaks. So it's become a more pressing problem, but it was a subjective scientific curiosity for like decades, and so fortunately they were able to attack those large outbreaks with a complicated policy problem, but a really rock solid, like. Don't ask me: ask of ivory allegest perhaps suggest, will tell you the name of a viral just who studies of Allah and then that Abdullah expert will deliver you like the truth from science and what's frustrating about that. The current very situation from a journalistic and citizen- and I hear it from politicians do standpoint. Is that like do I really experts on this subject because it's a brand new subjects, but nobody did a dissertation on this. Nobody like has their book or has taught semi cars are on this subject, so everyone whose whose working on everything there the sort of making stuff up, I wouldn't say making stuff, but there,
dimensions or levels of witches. Communication happens, and one is the sites communication he'll be Talkin to infectious disease. Expert in and then you'll be having this conversation and nobody's really careful, really cautious about what we don't know really like emphasised how much uncertainty there is in in the state of our research here and then there's level of like science, communication and these same experts will get on twitter or alike, will be in an article and what their responding to is a stupid or offensive Chris right or like one of these protests outside of governor's mansion where nobody's wearing masks and everybody's demanding. I think it's opened up all at once and they'll get a lot more sure they hit that back much harder. So if you ask them like what do we know, they'll say: oh, we probably like thirty percent of what we know right now is going to be wrong,
but then, when like faced with like a very wrong or what feels like crudely motivated critique, the public communication gets a lot less nuanced and I think it's hard in there to keep These conversations separate I'm not even saying it's wrong to have that the two levels of of communication, but I do think there is a lot of comments from scientists being reported as if those comments were the science and their different right. The scientist I could be us are human beings who have opinions and believes in earth are trying based on uncertain information, echoed the best recommendation they can and their fraid of what will happen worst write a force. Recommendations are followed, but at the same time- and he says all the time of the climate debate to fully communicate, your uncertainty makes it very hard for, like a journalist to quote you or makes it makes it hard for like people to listen to you Like your interviews, don't go viral, you don't become a big name. The most cautious people in this are not the
the names are hearing in all the articles, and so there's just like by its nature. A problem here: will become a particular problem as we move forward. So I think it had like one dimension back early. So if you go back a couple of months ago, Anthony found she was saying in February. The risk of coronavirus is minuscule and you should worry about the flu and a the journalists like including people of oxen. Like heard this sort of message like this is probably going to be. Control like this looks bad, but don't worry that much about it. Specific what your hearing in January and early February and then they are all saying this kind of worry about the flu thing, because a flew like they're always angry about. Is people don't take flu seriously enough? A lot of people die from it. It's like a bad disease and it turned out. The coronavirus was for all kinds of reasons, a lot worse than their state of knowledge- understood that time that asymptomatic transfer is more of an issue that China did not have it under control. China, with like obscuring,
different things, and so then it gets really bad and I think now there's like this opposite thing, where people who got burnt by not taking this seriously enough. I really want to make sure they don't make that mistake again, but that makes it hard then to discuss like. Why is it that the worst predictions, and also the worst places don't seem to be happening in in nearly many spaces? I mean why is it Brazil, a total catastrophe right now but also wise glycol Ecuador, Toller catastrophe right now like get wise in South Africa, with its very large amino compromised population as big a disastrous people were telling me was gonna, be three or four weeks ago, but also like you can't look at New York City, and I think this can get really bad. So there's something in this like tension between the really bad outcomes that have happened and the really bad outcomes haven't happened that we really do need to be discussing in working out
but it's a very people are afraid to have that discussion. Just like nobody wants to be seen now is under estimating it and it just is it difficult like you, don't know enough to have it so everybody kind of wants to air. I think correctly on the side of caution. But it's a politically difficult space to be in for the politicians who have like hold these very difficult policies and in trying to make them sustainable. Just a word saying you know you were talking about like their science and then their size communication right. But this isn't this other thing, which is public health communication and what you seek frequently right is you know somebody like the surgeon, general of the United States, has a real medical doctor. His expertise is not in infectious diseases or viruses, but like that's, okay, because he's a general because before he had a public health role in Indiana and
I mean- I don't know if he's good at that role or not, but it becomes a distinct role you occupy, which is not doing science or communicating science, it's trying to shape public behavior, but then there's always a question of like what level of authority is that, like, I think it's crystal clear in retrospect that the message that the public health experts in the United States were giving about masks. Was not based on like bad science or not and like what they said. All along has stayed the same, which is that we need surgical masks for medical professionals and that and ninety five respirators.
Prevent you from inhaling virus, but they need to be fitted right. You need to be a real expert and we need them for healthcare prose, but that, if you're, sick wearing a mask will stop it from spreading elsewhere. They just flipped from saying, because of all those things don't get a mask too because of authors. Things do get a mask and the reason they flipped is because they had a shift in understanding of what the amount of symptomatic transfer, but also, I think they had a shift in the understanding of how easy it was just so. Cloth masks right that, like they are not experts in sewing or apparel manufacture or like how it sees supply chain works, and things like that so they're trying to give people advice that they can take right in the wall accomplished something useful. So what they were thinking was tell people, stop hoarding surgical masks, you something they can do and that will help. It turns out TAT, What is so a mask at home is also
and you can do, but I think they didn't know that now because, like their science is bad is their understanding of, so we or human psychology was not exactly what needed to be, and you have this A lot. I mean I've. I've been I've been working on this question up with: what's up with parks right and one person works was city government was telling look. Look we don't want to tell people that outside transmission is likely, because then they're gonna make plans to meet up with their friends outside an I don't know I've ever glistening. I see what he is saying, but also like that's not a scientific judgment. You know it's like it. It just a guess about mass communication. It is, as far as I can tell back
Willie. True that you are unlikely to inhale corona virus particles, while sitting outside the idea that we should obscure that fact to create a better outcome. I get might be sure it might not be true, but I honestly I'm not sure what kind of evidence you would even get to sort of make that judgment. But the public health community seems very comforting. What sort of reaching a consensus about what direction people need to be steered and then delivering a message. That is not a hundred per cent factual to try to produce that outcome. I think something you see in a lot of forms of communication is that they they like have their message. Then they have their direction reality and peace often know the direction they want to take better than they know the specific message they want to take raven. How to get to that. Like one thing, I'll often happen in stories are different kinds. Is that there's a lot of information in the story,
like. Whatever this possible thing is, but there is a a fundamental action. The of is the author of the story. Trinity Unum worry more about it or less about it like do they want We bore about krona virus or this possible financial crisis, or this candidate getting elected or this bill passing or not passing or whatever I'm super volcanoes, or did they waiting bore you less about it, and you can write the story actually with the same information as you both ways right like you, can read the same information into ways, but that you get you end but like the headline Emmy Atmospheric of it, a story that, like might in many ways, be quite Kaweah added, like you think, there's an eighty twenty chance of this really bad thing. What happen so is what you want to do to get people to like chill out a little bit, because I eighty twenty, it's gonna, be fine or is what you want to do to get them to really.
Freak out has twenty percent. It won't be. I mean a version of this. He saw a lot was at the end of the twenty six new election. You re of bottles but basically than time's up shop model and then a silver model, where boasting something between like eighty twenty four content for the shot and seventy thirty four Clinton infer silver, but in their public communication, like needs of a really was emphasising that thirty percent is a real chance. Donald Trump gets elected, unlike like you should think this is possible and I think that times did a good job warrant out there like trying like hit that message in the same way, and so people who reading the Times coverage we're feeling really good about Clinton and then, of course, she loses, and so often times. I think
journalists, scientists, politicians, all kinds of people, sort of mistake that direction quality for the thing itself and, in part, like also stories with a heavier direction, already get shared more like we, this great story from Julia blues in early February, and it was called um. I forgot what is called, but it was scenarios for krona virus right. This is like before really knew what was going on and she had forced scenarios in which gets really bad and foreign which it doesn't like. I thought that store was so good because, like we didn't know a lot and is a really good way of showing what we didn't know, but also that story did not do that. Well, happy like it's, not the story that like got shared or even in retrospect as good as I think it is the story that, like people want to talk about because it could, in fact, I think that the very strong direction out to them. A clear message, and so just like also in the generalized content world. The things with a strong directionality rise to the top and
I don't know. I think this is making all conversations of an uncertain issue, much harder, because it's the models with a strong direction, amity and their message. It are getting a lot of attention and then it's like messaging with a stronger action. Other gets a lot of attention. It's their allotted epidemiologists out there, but some of them are getting more famous and others right now and like they're, the ones with, I would say, spicy or public presences, unlike at all kind of like stocks, on top of itself, to take something where there's a lot of uncertainty at the base of it and Crete. At the end, much more certain feeling messaging. But then, as things change- and I think a hard part about this- I think we're and can end up here in an equilibrium were like. On the one hand, it is gets out of control in the wrong population. It's unbelievable trophically bad, and on the other hand, I think it is actually somewhat more reliably controllable than we feared the beginning, and that's going to make people feel that it's not that bad
and then they will get. You know it's a really like bounced back and forth between like doing too much doing too little more people feeling like we're doing too. You're doing too little in a way. That is not how could outcomes for anybody? That's a break in and always talk about that in the difficulty of sort of problem, probabilistic forecasting, if you're a gig worker or self employed. There's some good news about PPP loans. You might want to consider millions of self employed workers may qualify for up to fifty thousand dollars in one hundred percent forgivable loan. You might be one of those millions as the leader ppp allowance wobbly can help you find out. They have helped over three hundred thousand small, businesses across America get a ppp loan. Funds are limited, so apply now at Womply, dot, com, VOX and see. If you qualify for a ppp loan That's w, o M P, L Y dot com, VOX Not a lender terms and programme rules apply.
You know of hours in the day to get everything done. It might be because you're missing out on three of them. Where does through our scope? They probably fell into a deep dark. A bits opens up when we switch between work. Apps, add those three hours to all the productive family, miss out thanks to at home, distractions, disorganization fatigue. It's no wonder the days feel too short. Work should work and with click up it does put up as a flexible productivity platform. That brings all your work into one place. That's all you're, chats, apps docks and tasks. One central S Place Mission control companies like Hoover and Google use cook up to
their days were productive, managed projects, people and calls for effectively both for teams of all sizes and industries. Hookups blazing we fast features of one thousand plus integrations beginning must have for anyone wanting to track, manage and tackle the work in one place and showers back with clear up tried for free today had cleared up the palm slash. The weeds I've been looking it at the state of Georgia. It'll governor brain Camp is moving tat to open things. Up relatively fast and like a lot of people, you know on the internet. I have a kind of as a sinking feeling about it, but when it, when I look at what, when he is putting forward, I feel frightened it seems unwise. Then I also see a lot of people. Sort of you know dumping on him and saying this is gonna, be a disaster, and that gives me it's
on sinking feeling, because if you look at it and if you look at this situation, you say: ok, if Georgia does, what camp is saying, there is a thirty five percent chance. That metro. Atlanta gets here, hard as Metro New York. That strikes me as an credibly unwise risk to run right if you like, do the population with land you bought apply by point three five, you say: look in expectation. He is costing thousands of people their lives here and that's a really bad idea, but also I mean I'd making up the thirty five percent chance, but I'm trying to convey at Sgi here if you run a thirty five percent chance of catastrophe. Catastrophe probably won't happen. If everybody is like holy shit, this is gonna, be a catastrophe and then the most likely scenario rules out and it's not a catastrophe. Then you ve lost your credibility.
When seven more states are like aha look at the naysayers, but if ten states all do something that has a thirty five percent chance of ending it, catastrophe, several states are going to end up in catastrophe and that's a real bad idea, but I don't know how to write a pity tweet that it's like most likely. This will be ok, but it's a very unwise risk anyway, like that. Doesn't this on a good prediction Conversely, you know when people look back at people They say like: oh, these of this is the guy who got corona virus right. It's never that somebody wrote on January. Twenty seventh, I think, there's a five percent chance of this becomes a global pandemic and that's a really bad.
And we should take that seriously it somebody who, on January twenty seventh, was like. I am absolutely certain that this will be a global pandemic and, of course it retrospect, that's court, unquote getting it right, but like there was no way to know their on January twenty seven three. I wish I was more alarmed. I was kind of reading these things are, we were like it's probably gonna be fine and I was like Laddie dad. I'll, probably be fine, but you know like small chances of bad outcomes are a big deal, but it's hard to convey them without being hysterical and then randomly somebody who is just guessing turns out. Do the person who was right as a result there. I wanna go back to the Georgia thing, though, for a minute, because I think there's like some interesting things tucked in their wine is that, like the list of things that make me very nervous, there's a prepare, an article that came out and what it did was it looked at like everywhere in the U S by a bunch of the factors that make
population more vulnerable to covet nineteen, so all kinds of co like hypertension and cardiovascular issues and among issues we kill systems but other things age. Obviously, and if you look at that heat map and like you just look at where looks worst on the heat, up. It's Georgia, so you're opening this up. I mean it's Georgia and Florida and a couple areas in that southeastern belt, so you're opening this up in one of the most vulnerable populations, which is really scary at the same time, one thing I don't think we know in areas that are not as Density York. City is how much social distancing and sure do you need to get most of the effect and Brian Camp saying that he's gonna open up Nell salons bowling alleys, as you already know, I think a really good piece on it. Can't you
the economy by reopening the economy, a lot of people aren't going to go. A lot of people were social distancing in Florida before Governor De Santis decided to actually close the state on April three rd, and so one thing that may happen here is Kemp reopens Georgia, Georgia, like does not really reopen in the sense that a lot of Georgians. I do not feel like taken this risk and there's like social pressure, not too, and so it sir. Like he abasing moves and you'd like twenty percent towards reopening, but that actually isn't enough to create a catastrophe: in this sort of new equilibrium or again like the system knows, the health system knows what looking for, like people or washing their hands are wearing face, masks their eccentric such and such a now. Maybe it does create a total disaster, but I think is also going to be this potential for an outcome where, like camp has officially reopened Georgia, but people in Georgia haven't followed the reopening path, and so it looks like he right to reopen. But what actually happened so soliciting continued
gonna, be very hard to track and understand, subleasing correctly, in a way that it is gonna, make it right really running experiments here right? It's not like or doing is like. We ve got like a control place in social distancing and then we're like paying Georgians to go out and like we renew their normalize, readers can to see what happens like people also make an own individual calls here, and I do think something that has been is that social distancing is a powerful individual phenomenon as well, and I think it's working on that level pretty effectively and that's part of why states close later did not actually get as bad outbreaks as we feared, and I think it's also potentially why this Kemp thing as bad as it looks to me, may not be as bad as I worry. It will be because he doesn't actually have the Our till I reopen Georgia Armenia reopen epidemic any he can't make people go to things in the numbers they were before. Why? Then you get really into, watching of how do we model the human behavior right? If you say, ok, things are opened up, but for the first two weeks, everybody's incredibly cos,
Yes, but nobody knows how cautious everybody is being exactly because it's not that well indexed and then it goes pretty well. So then everybody gets less cautious and it could get much worse again and you don't we just don't know right like I, dont have a gas as too like how will check she's in Georgia operate in a post, real, bidding environment and how parishioner sort of respond, because that strikes me as the in like Exurban Sunbelt America. The sort of most plausible avenue for super spreading is in church services because there's no subway car to be packed Are there no ever say? No, but most people are living in apartment buildings. Our budgets, because you have a low residential density, doesn't mean you never have crowds, and so it's like will what people actually do in those kind of cigarettes
since and are we even gonna, be paying attention to what they do so that we can learn something for for the future right, because this is like one of the things is, in general. You would never try to model. What's gonna happen in Dallas, based on data from MILAN. Right does a very different cities, but it's like you do in the best. You and with the data available. But if we really track like what everybody does in Atlanta, that could be very informative, but as far as I can tell were not does not plan to do that nobody is volunteering themselves? We did have the mayor of LAS Vegas, like use the phrase control group on a television. True, but all she seemed to be saying was what she wanted to get her casinos back open because the city needs that tax revenue, but I think I mean by overall
My guess is that honey is gonna book, a flight to LAS Vegas to go gambling. No matter what the mayor says there, either you shouldn't overrate the sort of public officials ability to drive human behavior. I got right, you wanna, take a break speaking of public officials, the ability to draw behaviour and talk about building. Let's do it s built if the last year's hottest anything it's that we don't know what will happen next, but there's one thing we can all be sure of the only future is one we can all share and leading the charge in building that future mercy core with over forty years of humanitarian work under its belt building together mercy course. Dna and, as the climate crisis increases their partnering with those on the front lines, making resources more accessible to farmers across the globe. Strengthening community is again escalating natural disasters and ensuring people have the tools they need to thrive. Mercy is doing the work that matters, but they can't do alone. That's where you and I commend together. We all have the power to reshape the world
when it seems like every day, brings a new crisis when every news alert makes you want to throw your phone across the room. We may say feel a little powerless mercy. Court here to remind us, we don't need to Turkey community based action, we can make change. We are nothing if not in this together. What's next is up to all of us learn how you can be a part of is possible at mercy, core dot, Org, that's our see Y, see
p s, dot, org support for this episode comes from America's leading beverage companies who are working together to reduce plastic waste in our environment. Not all plastic is the same. America's beverage companies are carefully designing. One hundred percent recyclable plastic bottles, including the cap's their bottles, are made to be remained and their investing in community recycling programmes to help get more bottles back, so they can be turned into materials used to make new bottles that completes the circle and reduces plastic waste. Please help get every bottle back, learn more Every bottle back dot, Org Mark Andreasen, who developed the Mosaic web browser like a budget years ago, which was like TAT, the pioneering web browser for the world and He made some money off that, even though, was never actually a commercially viable product, and then
came the sort of co founder and head of, I think it's the biggest venture capital firm in America in the world in recent Horowitz, and that means he's a person with incredible sort of influence and reputation. In Silicon Valley he's invested in a million important companies, he invented one of the most important underlying technologies and for a while he sort of dabbled in a public intellectual space, a lot on twitter and elsewhere. He really kind of step back from that that recently and then re emerged from the ether with a blog post called it's time to build, saying that you that the lesson of corona virus was that it's kind of along lines of like this once great nation like can get things done any more like we don't have swabs there's no houses
the Bay area. Oliver infrastructure is garbage, and it's time for us too, to fix everything- and I saw a lot of people really excited about this. Ranging from heat, a libertarian ash kind of intellectuals, but also more than a centre left people? I think a lot of folks who, like to think of themselves as forward thinking non, ideologically dogmatic sort of we used to call radical centre sorts of light of thought. I think we're we're really to this essay and it became a sort of a selling point for people who wanted to express their enthusiasm for the idea that we've we got to start getting things done you wrote a well trafficked and received sort of response as they do this about, like what is actually stopping us from. Doing so will always I liked the andreasen essay. A lot like I want to build
or two, and I think a lot of people want to build more and in some ways, that's like the knot of my response to it, which is in a couple key moments in his essay. He frames of the problem, as people don't seem to want to build or people don't seem to like heavy yen for building. So like up top he's, expecting a bunch of criticism, and he says they. Here's a modest proposal to my critics. Psych instead occurred asking me. Why did you say what you want to build like chancellor? I will agree with it and then at the end, his solution on building and- and I quote, this building- is an easy or we'd already be doing all this. We need to demand more of our political leaders of our ceos. Our entrepreneurs are investors. We need to demand more of our culture of our society and we to demand more one. Another were all necessary and we can all contribute to building every step of the way to everyone around us. We should be asking the question: what are you building? What do you building or helping helping other people to build or teaching other people to build or taking care of people who are building, and this is where I don't to say, get off
train, because I agree with all that like we need to demand more people and we should encourage people to build, but this is like the Classic Silicon Valley, political, like disagreement, which is you have to take the institutional problems here seriously like I, like people want to build things like entrepreneurs want to buildings developers want to build houses, including in the Bay area. That's why they're these big fights it planning meetings and even unlikely. The centre of supposedly stagnation, Washington DC, if you cover Congress, the amount of time and like you and I have covered for twenty years. It is full of bills like people want to build all kinds of things. New health care systems need to new social insurance systems,
de Carbonizing, the: U S, economy, new education systems like everything like infrastructure, it's always gonna, be infrastructure, weak and it doesnt work. And so I guess like what I would say. What we need to do is rebuild institutions and, like I make the argument in this piece, it would become a guitar secrecy, which is a term from France. Fukuyama were too many different actors can veto things at all levels with the federal government people heard me talkin, this forever. You have things like the filibuster and polarisation and the way we descent as power in assisting the requires very levels of compromise and consensus to function, makes it impossible for really anybody to do anything and that state and local level, which is more a space. You focus on that. I should talk about some of these arguments and essays that progressive in their scepticism of power have created a like a form of contact.
Representation that ACT Sheikh ends up only representing like the status quo in India. Stick interests because, like the people who show up to these things, are there to stop things and we ve made it like. We ve built all these things to stop power from being wielded unwisely, but we have made it impossible levels to even willed it wisely, and in addition to that, you have conservatives who actively are trying to make state government and local government work what I would call poorly andreasen talks about how bad. We ve been getting say: unemployment shirts out the door like in Florida. That is because a republican governor Florida built an unemployment insurance system meant to make it hard to get unpleasant insurance. It gives a successful effort at building that leads to an unsuccessful country. And then in the capital markets and in business more generally, not space Andreasen obviously knows better than I do like you have this short firm, shareholder capitalism that leads to like the market, constantly vetoing anything. It's gonna fuck with quarterly profits andreasen to his credit, has joined on,
Greece in this effort create like a long term stock exchange, but I think, like taking were seriously the ways in which kind of advanced capitalism and hyper efficiency creed- the kinds of like just in time, very fragile supply chains that he's like acting on cotton, swabs and reagents like it needs to be taken seriously and the promised people want to build all kinds of things, but they don't want to do. Like patiently engage in reforming, very frustrating an annoying institutions filled with people who want things to be the way they ve been for a long time. We think I would say about addresses essay is that you know like a lot of things that have written on the spur of the moment this differently, to kind of read it and one way I've seen a lot of people read. It is a kind of like a tech godfather, throwing a bomb from the West Coast at politics and
think. If you read it that way, it's not that persuasive, although you know he, he cites continuities book about housing and in the Bay area we Tad Conner, and the weeds, and you know that the it suggests a grounding in some understanding of the specifics of the political situation. But I think, as I got a macro diagnosis, as you say, like people don't want to build enough, does not really explain anything in the federal government the way of looking at it is as a kind of tech industry, Godfather saying something to his peers or to his his followers and those sort of come under him and in that view I give it more credit actually because it's true that there's a question about like what to capital markets want to allocate capital to, but is also a question as to what do human agents want to educate their
personal talents to, but you really people who are smart and hard working and are in sort of high demand as people command, good salaries in a variety of different locations and remember. I was at a social event last year and you know it's one of these things where you get seated at the table, the people. There know each other and you don't know them, and it was. It was to people who you not working in machine learning on the West Coast somewhere and one of them was working on basically like speech to text transcription and one of them was working on algorithmic lending and a pain in the ass. I was talking to these people and the upshot of it's at algorithmic landing. Is that banks and find a way to violate civil rights law, in a way that its legal is like what it comes down to like if you can successfully identify,
like strong correlates of being African American and then steer loans way from those people you will gain a market edge over tricks. Underwriting where that's gonna be illegal. There is absolutely a valid business opportunity there, but it's like strikes me as a shitty thing to do with your life and, like I've, really wanted this woman, who was working on speech to text transcription two like ruthlessly dunk on this other machine learning guy, because I would really like automated transcription of interviews like much easier and cheaper, like I could go on and on and on about the benefits to the world of improving that kind of technology.
And I could go on and on about the like likely harms to the world of improving the algorithmic lending technology from a computer science perspective. These are both hard problems to crack and, like investors are going to try to make money, but, like you can't force people to like do something worthless their lives and in that Spirit, like I do think, having more people ask themselves the question: what am I building here not like this operate at scale and have a moat which are like the business buzzwords but like what am I,
Creating here- and am I gonna be proud of It- are people gonna be like, were really glad. This thing happened, which is not to say like it's, not that thing, don't like lots of things in the Tec world. I think, do have that quality like they're, really good and there really useful, but there's a tendency of both like critics and like defensive people too, like lump tat together when, like any other sector of society, people actually doing very different things there, some of which have incredible value and some of which are like Canada, dumb and getting like smart hard working people to try to orient themselves to building things that are worthwhile seems like like a really valuable endeavor to me. I agree. All that and and being good is better than being bad. That is true and, to the extent that's the point like I'm there, think in a way that was really interesting.
Yes, a, although was not something Andreasen said explicitly. It seemed to me. The essay with fitting into a debate inside Silicon Valley that Andreasen has often been understood is on the other side. So Andreasen is like very famous, I softwares eating the world and the idea, basically, is that huge amounts of like Egon, activity are going to now be like built on top of a software overlay. You know so like you're, not going to like Hail York at Beacon, like user uber, app and things, and if you can figure out like how to build the software platforms, you have these almost infinitely scale. All super multi billion dollar industries that you can like profit from end, Jason and Alien. Sixty a general have done that really. Well I mean he's early Facebook investor who is on Facebook sport even today, but then there's in recent years this critique coming from people like Peter Tee. All that
Golly has over invested in bits and under invested in Adam Sir, making too much software and funding too much software and not funding like actual building of things in the real world, and so like the person who started the that. The counter example is like Elon Musk, who, for all of his like color, is a character like actually mean spacecraft an automobile Sars solar panels. Aren't you can you can death like you? I'm has taught us You can drive like it's quite amazing, I'm technology, and it means that he is constantly dealing with federal regulators and convincing them to let him do things, and he took a big loan from the Department of Energy on the same programme. The glad to cylindrical help save Tesla actually, which people do not know in the way which they did and so like musk is, like he's really he's been using the softer somebody's been making Adams, where I d like trying to sell things like you know, he cares about electric cars because of climate change. I can
cares about about colonizing Mars because you know he believes we're earth eventually, and so some of it can be a little dystopic, but I think it's fun the very admirable admirable project and I don't know range, what a one sixty is funded but but they ve been really strong at funding software companies. And if you look at this mark Andreasen essay, it is very heavily about things. You would build physically, like cotton swabs, the latest different approaches to schooling, where you need to deal with huge numbers. He talks about like tutoring as a like, a like the next space of schooling, which has great, like that's a huge super labour intensive thing. He was, he didn't talk about Mewks as he answered to the future of schooling, building more homes, and I think something that has been true like in Silicon Valley, and I know it is a criticism at all. I think it is a very natural thing to have happened, but now it it's like that, like the low hang forgot pact,
is that if you could create software and including, like it you're lending example, hear your algorithmic lending example. Is it a good example of a very bad way of implementing this? If you can try to disrupt things to create a new software platforms, you can often evade the old systems of like regulation and accountability that made to make new things right like you didn't have to go this, maybe like OSHA regulations, if you like automated your factory, for instance, that kind of thing and the more that, like we, have to build real things in the real world. What using real people. The more you end up having to deal with these institutions at I'm talking about like much bigger, like infusions of capital from capital markets like if you want to do a huge manufacturing play, you can take three hundred thousand right investment right, you need millions and millions of dollars to do something at that level, and so you end up having to go through all these institutions like stock markets and local regulators and
that the federal government and laws and is its legal and will the FDA approve it, and so I dont think of like what I'm saying really is actually even enough. Listen what Andreasen saying, but if what he is doing is trying to push his own pure network, into investing in the real world like into pressing more into Adams, building more stuff, more things without like pushes you back towards our regulatory structure, which increasing suffers gotta, be under also because, like regulated efficient, mister out, and so you, in my view, like you, do need to build institutions, so they can be more action oriented, as opposed to being quite so, veto. A dead and obviously I'm a fan of a lot of forms of effective regulation. So my point: isn't that should be like free ceasing to build whatever you want to get people SEC, but things should be quick. They should be fast, they should be well done and, like Congress should be able to say appropriate, the money to De Carbonizing Economy, and so like I
I think that this is a way in which you can't use the thinking of software exactly four like what he wants to do. You catch us exhort people are going to tackle that you actually do like have to engage with the institutions, which I think this genuine thing where the entrepreneurial personality is often very impatient. I'd like live out here, I know a lot of these people and it makes dealing with the regulators super frustrating and like engaging, would like long term efforts like filibuster reform like it just you want to build something and have it happen, and so in some ways I think it's actually a bit of a tension between the personality that builds. Unlike the personality that reforms, but somehow like Unita, you need to combine those living anything way to serve further thoughts and that's like one. Instead, everybody finds it frustrating when the government doesn't work well on something they think is important, but also way a lurking threat. You know that people are aware of. Is that a wealth
sitting democratic government will exploit the fact that the median income is much. Lower than the mean income and we'll ten redistribute wealth downward, so it's was a question in my mind, have like do it players in the american business community? Want the government to work well or don't? You know, and you can find various efforts to thread the needle, like Tyler Cohen, did his post about state capacity libertarianism, which is to say like what? If the government did work well but didn't do the things I don't want it to do, but like a well functioning democratic government, is going to do a lot of income redistribution. There's just like no way around that lady, like particularly because the public services will be really high quality like by hypothesis. People are going to want to fund them by taxing the rich, and you won't have all these veto point so, like people won't be able to judge
kind of hold that up. I hope that people who are looking at corona virus and are frustrated by the lack of state efficacy are gonna like bite. That bullet and say like like. If this means social security benefits go up and taxes go up like I'm off We need a function in government, but they're, not unrelated issues. It seems to me the other thing, that's interesting. You know your turn. My Adams versus bets right, and what is its relevant here, is that the financing model, because the whole reason we have venture cap so closely associated with eighty of software right is you can get a bank loan for the proper, The Shin I'm gonna like hire some guys in hoodies and we're gonna write some code. That will be good because the banks could a saleable. What do I repossess if this doesn't work out? And the answer is nothing like code, that doesnt work or like old, so
like to eat. You know it. It doesn't make sense right. So you you need an investor who is going we run the risk that is, investment goes to zero and in exchange that investor wants to build a ride. The elevator, like all the way to the top and that's venture capital right when you doing Adams, if you say like, I want open a restaurant, you go to a bank and you get alone, because if you restaurant fails, when you ve got a built out kitchen right, it's got all the equipment. You got tables dislike, actually something for them too to take, and then the bank does it get nearly as much the upside it is as a vc word and you can have like venture capitalists. You know investing in physical things and you can have banks lending to software companies and their companies like Apple, that sort of straddle the divide between between bits in items, but there is a real sort of distinction, their conceptually and it's like if you're a venture, capitalist and you get frustrated
with everything being these kind of like wait, software company is, I think, you're gonna actually have to get out of venture capital as a as a line of endeavour that makes sense to make VC investments in building hospitals, which is one of the things he says. He's like. Let's build, the hospital of the future like, I think, would be a good idea to, but like the the financing model, for that like wouldn't make sense, because even if your hospital was incredible, you're not gonna, get like a hundred ex exit out of that, like it's still a freak in hospital and it needs a whole sort of mentality- and so I don't know like I couldn't tell like the ESA was meant to be reflective in that sense or was just like brushing past. It to go back to something you're tagamet earlier and that which is this issue of, like any logical disagreement over what to build right. If you have a democratic government that works well, it might do more on redistribution.
And obviously Democrats are worried if youve republican government, that is not capture or stopped by the filibuster and other kinds of veto points it will you no further restrict reproductive rights, are, will cut taxes on rich people, do all kinds of things deregulate. The Democrats don't like and that's why and by I say one of the arguments I make is that we actually need to try to make institution to action oriented without that being secondary to what the action they take is like. We need to be more comfortable with the problems of like accountability than the problems of paralysis right now we prefer like, like. I can't get what I want done, but at least you can't get what you want done either and I would prefer to say like if I animal action. I can get what I want done and if you win the election, you can get what you want done and in both those will work like. I think it's recent who says in his peace like a lot of he says energy people think we could just
of the clean energy problem with a few thousand of these next generation, zero mission, nuclear reactors and like email some people, I know on clean energy and, as is that possible- and they didn't think it was. But there are a lot of people who care a lot about clean energy and partially from a technology perspective who are furious at the degree to which Democrats and particular the more like left, side of the Party like Bernie Sanders Animals with Warren Artistic, they taken nuclear, basically off the table, and I think the right to be upset about that. Like that's, really the point that if you like right now, nobody can do their thing, so it's like Bernie Sanders or for that matter Joe Biden, probably can't do like their climate change plan, but nor can the nor Andreasen would prefer do they're like nuclear energy reactor plan, and so what I would like to see as institutions oriented towards action as opposed to institutions or ain't. It hurts paralysis. I mean some
that this is all coming in. The context of is after a long time or Silicon Valley was like the God that hadn't yet failed in America in the past couple years has been this big tech backlash, lot of attention on like move fast and break things now, not as an awesome model on which to found a company, but all the dangers and irresponsibility is encoded in there and then I think this krona virus period. This been the sort of a little bit move back to saying you know man, would you prefer the problems of disruption to the problems of stagnation and and and paralysis, and like? I think that that basic idea is correct and that in like like, yes, I think so candlelight could be pulled back a little bed has been for that matter towards, like thinking more seriously, about like downstream social impacts of what they do. But a lot of other institutions need to be pushed way forward towards being willing for things to go a little bit more wrong. Other big things can go
more right, but it's hard to get people to focus on that. Like it, I mean this has been kings all the way out of the context of increasing, like my thing throughout this whole democratic, prime, He's been stop telling me like what your imaginary plan is like. What are you gonna? Do the institutions to make it possible to pass any plan whatsoever, and I got the asset in every candidate interview. I've done and I don't at all believe Joe Biden is going to come in, and the first thing he's going to do is like a massive package of structural reforms. Maybe I'm wrong. I would like to believe that I am like. I dont see evidence of it, and so I dont think his agenda will be successful as I wanted to be, although who knows, maybe we have twenty three percent unemployment and not totally wipes out Republicans in the Senate and, like created the preacher unequal IBM. I wasn't expecting but yeah like People can argue about what to build. Like that's, ok, I would
for a world where all kinds of people get to try building things within illicit within like some reasonable limits, as opposed to world, which almost nobody can build things, because everybody can veto them and like the sucker we all take? Is it well? If we didn't Crete, the things we had hoped to do to solve their own problems release, nor did those other guys. Yeah me. There is not just like a question about about risk thou because it's like one place where I do think that, like right of centre, business, people and left of centre writers, Kind of like meat as have the melting of the mines is in having like a lot more openness to experience, risk tolerance and am like comfort ambiguity that the typical person and I'd ever know exactly what to what to make of that, because I I simultaneously think that I am act that, like american society is too small, see conservative flick about things sort of hunger.
The that you were saying that it would be better to let like ten zero, go forward, but also have fifteen triumphs and the question of like well Is that just like my aesthetics that people authentically disagree with that like this is the like Buzz prosperous society. You know ever built most people have it have good in America, something like we ve seen with current virus is like we were talking about this in the models it seems like
typical American really does not want to get corona virus and get sick. People are being quite conservative with their actual personal behaviour in a way that has surprised public health researchers who thought they would be more reckless and seems to be frustrating supply side. Economists who, like really wants us to like not just like they want to lift form of restrictions like they want the economy to get going again and like we had a mood of the lieutenant governor of Texas, saying, like there's more important things than living which, like unsettled True right I mean it's like people do things that carry some risk of death all the time, but, like most people just seem fairly cautious in their lives like separate from institutional dynamics and
had this nuclear suffers. A great example you I did. It did an interview with one of these like new nuclear enthusiasts, and you know I posted at. I wrote an article and all kinds of people wrote into me like they were like oh, you can put nuclear plant in your basement and it's like, I don't know like probably where he like, they seemed safe to me, but it was just like not the mass reaction to like next generation nuclear reactors that I was hoping to be the case that it's like just like seven hippies and like a bad institutional structure, the Nuclear Regulatory Commission that was blocking these next generation nuclear devices, but just like all kinds of people are coming out of the woodwork you tell me that was a terrible idea and hadn't. I seen the Chernobyl Mineral Series and like what are we gonna do I have not seen the Chernobyl many cities, I haven't either. Does it there's never been a night? I know I need to, but there's never been a night right
even what's going on the world, where I feel like like emotionally able to handle that yes exactly, but so maybe outrage Mojave around ok, so states say that they have run just by makes rotations. Tell me if the Chernobyl many series is good happen to the weeds Facebook group. Ass. Let me now, whatever you got, thanks to our producer, Jeffrey GUILT and that will be back on Tuesday
Transcript generated on 2021-05-19.