Full Q&A: ‘Temp’ author Louis Hyman on Recode Decode

On this episode of Recode Decode, economic historian Louis Hyman talks with Recode’s Kara Swisher and Rani Molla about his new book, “Temp: How American Work, American Business, and the American Dream Became Temporary.”

You can listen to the entire conversation right now in the audio player below. If you prefer to listen on your phone, Recode Decode is available wherever you listen to podcasts — including , Spotify, Google Podcasts, Pocket Casts and Overcast.

Below is the full transcript of the conversation. You can read a more condensed, lightly edited version here.

Kara Swisher: Louis, welcome to Recode Decode.

Louis Hyman: Thanks for having me here today.

KS: Also joining us today, who is going to do most of this interview because I’m feeling lazy today, is Recode’s data editor, who is much smarter than I am, Rani Molla. Hi Rani.

KS: Good. Rani deals in a lot of data. I’m going to start off. We’re going to talk in three sections and talk about various things. This is a big interest of mine, as you know. I’ve been doing these MSNBC shows on the future of work. What I’d love to know, just to start off, and then Rani can jump in any time, is how did you get started on this? Give us a little bit of your background. I think people want to understand how you got to this topic.

Well, my first two books were about equally unhappy topics, the history of personal debt in America. As I was writing those books I noticed that really the story of finance in America is also the story of work. I decided for my next project, I wanted to write about the history of how, not just our finances became insecure but also our work became insecure.

KS: In terms that it wasn’t before? Because this is the sort, what is it called, the economy, the gig economy, right?

KS: Which isn’t new, from …

So, yeah. What I wrote about in the book was how there was this creation after World War II of secure work, secure investment, big corporations, stable jobs, and how that all fell apart starting around 1970. To understand not just the fall of that older model of the workplace, but also the rise of what came next — consultants and temp workers and undocumented migrants — and how those were all pretty central to the remaking of capitalism since 1970.

KS: All right. So why don’t we start with, though, the background of work in America. It was a farm economy that moved … Why don’t you go through that, because that’s an important way to how we get to where …

I feel like I’m taking my orals again. This is exciting!

KS: I know. We need to know. We need background!

“Professor, answer my questions.”

KS: We can’t just jump into “Temp.”

Yeah, so I think the big thing to realize is that nearly everybody worked in the agricultural business in the 19th century, free or enslaved. This was the way our economy functioned. We were an export-based economy. We made cotton. A little over half of our GDP before the Civil War was either products made by enslaved people or related to that, like textiles.

What’s kind of interesting to me is how that changes. We think of capitalism and corporations as these very static things. But after the end of slavery, we reinvent our capitalism to be on the basis of free labor and oil and all these other things. We see in the history of capitalism, its constant reinvention. This happens again after …

KS: Right, people don’t think of it, that’s why I wanted to get to it. It changes really quickly.

It changes really quickly. We think of the corporation …

KS: And they’re difficult changes, by the way. I was arguing this with Marc Andreessen, and he was saying what’s happening now in the new job economy was that it’s like farm to manufacturing. I’m like, “Well, that took 70 years and was super socially problematic, that you aren’t remembering because you’re not a student of history, because you people in Silicon Valley remember nothing.”

It’s what computer scientists called a “non-trivial problem.”

KS: Yeah, exactly. I was like, “There’s some issues that go along with it.” My feeling was that the work change that’s happening now was really severe and fast, and stuff like that.

RM: Right, since the Industrial Revolution, incredibly painful. Also, mechanized farming, super painful. Yes, we all ended up with jobs, how many-ever years later, but …

In the long run, as Keynes said, we’re all dead. I think, yeah, the mechanical thresher is the best analogy to this in the sense that it put millions of people out of work. When I talk about AI, I often talk about mechanical threshers. Now, that didn’t naturally mean that we ended up in a straight line from mechanical thresher to the nice suburban house with two kids and a garage. It took 100 years of bloody warfare and near-insurrection.

RM: Right, and you write about a number of intermediary things that showed us having temporary work. Could you give us some examples?

Sure. So one of the things I write about, it’s important to realize, is that people have always worked in a sort of gig way. There would always be work to sell. A man with a strong back could always sell it, you know, find something to do.

But what’s different after World War II is that it gets mediated through agencies, whether those agencies are temp agencies like that at Manpower Incorporated, or consultancies like McKinsey and Company. These are new ways of selling people for the short term. This emerges in the midst of an economy based around long-term investment and secure work, so there’s a little bit of an irony there in how it comes about. Is that what you want me to talk about?

RM: Yeah, yeah. iPhones and Uber didn’t create the gig economy, this is something that had social underpinnings, economic underpinnings, for a long time now. Do you want to talk a little bit about Silicon Valley’s history of temporary work? Because that didn’t start 10 years ago, either.

Yeah, it’s very important to realize that Uber is the waste product of the service economy.

KS: Oh, I like that, what do you mean by that? Waste, product. Shitty, in other words.

Ehh, I don’t use these kinds of words, but yes.

KS: You may use them on this program. You may curse at will.

Okay, good. Awful. So the why Uber is the waste product is that it relies on a bunch of people who don’t have an alternative. This is the thing you have to realize, that the alternative between driving for Uber is not a good job in a factory with a union wage or working in a stable office job, it’s slinging coffee at a Starbucks where you may or may not get the hours you need.

That is what people are shoring up. They’re shoring up getting enough hours, trying to make ends meet. Oftentimes, people talk about the gig economy as “supplementary income.” It’s only as supplementary … It’s not like …

KS: Well, the companies talk about it like that. “Freedom, the supplementary income, you can do it on your own time.”

RM: Yeah, “make your own schedule,” yeah.

Yeah. It’s not supplemental if you need it to pay for your kids’ braces, or food, or rent. I think when we talk about the gig economy, it’s very easy to say, “This is awful,” and people point to it. But really, those kinds of awful problems are already present and have been since the 1970s.

The working Americans have faced increasing income volatility, income inequality. This is just reified in the so-called app-based, digital …

RM: So, what’s causing the income volatility? What’s causing people to actually take these gig jobs?

One of the things I argue in the book is that there was a wholesale move away from how corporations thought they should be organized, from both business leaders and policymakers and investors, right after the so-called “conglomerate craze” at the end of the 1960s. In the 1960s, corporations were making tons of money, just like now. Corporations were making tons of money and they were buying up lots of other companies.

KS: So employees, employees, employees.

Employees, employees, employees, but …

Scaling, scale, scale. But because of anti-monopoly laws, they were buying these unrelated companies and then creating more profits.

KS: Gulf and Western, like that.

Gulf and Western, yeah. Litton Industries. LTV. So a small electrician buys his way to being the 25th largest company in America. It turns out, they were all terribly run.

KS: How could you run them?

How could you run them? This engulfs about 95 percent of the Fortune 500. They all fall apart. Then people begin to cast about for alternative models. They blame the corporation, the postwar corporation, the postwar world of work for this. Into that intellectual void come consultants and business gurus who sell a new idea of how to run the corporation — leanly with only limited commitment to their workers and employees.

Et voila, you have the origins of today’s firms. To understand the history of today, it’s not just about technology. The real technology that changes is not the phone, it’s the corporation, how we organize people.

KS: But that’s different in Silicon Valley, though, how they were organizing people. Because this would be teams that grow up for a single purpose, essentially.

So, the history of Silicon Valley is really important. I write a lot about Silicon Valley in this book because Silicon Valley emerges as the leading sector of the economy in the 1970s. By 1980, it’s the most profitable part of the economy.

KS: That’s even before the boom, the really big boom?

It’s before the first Silicon Valley semiconductor and manufacturing boom. It is really, really reliant on a very different kind of manufacturing. Silicon Valley is never unionized the way that Detroit is unionized. Silicon Valley never provides good pay for its frontline manufacturing workers. Silicon Valley, actually, was reliant on hundreds of thousands of undocumented migrants, who are outside of those new laws that emerge in the early ’70s called OSHA. All the environmental standards, all the environmental standards …

KS: This is to create the chips.

… to create the chips. They were born subcontracted in a way that really portends the way that corporations are organized today. It was the rehearsal for what was to come.

KS: Then they moved them overseas.

Then they moved them overseas. In fact, they trained their own replacements. In the book, I write about the first Macintosh factory.

It was right across the street from the Nummi plant, where Toyota first built cars, and of course, where today Tesla has its own operations. It was in this … So I traced the history. The histories of Silicon Valley, as you know, are largely about Steve Jobs, or the Woz, or these other kinds of …

KS: Right, in their garages.

The garage men and their innovations. But they’re reliant on hundreds of thousands of mostly immigrant women. So every time somebody says “robot” in Silicon Valley, they usually mean woman, generally woman of color. In the book, I trace how this idea of automation is used to justify … this idea of progress is used to justify treating people miserably, workers miserably.

RM: Yeah, you make this really interesting point about how people right now perceive of Uber and you compare it to Etsy. They both are similar things, they’re platforms that are selling other people’s work that aren’t necessarily indebted to the people who are working for them. Everyone gets mad at Uber and not so much for Etsy. You said, “The reason for it is because we don’t value women’s labor.”

Absolutely. I think the fundamental question is, who counts? In the postwar period who counted were white men and everybody else didn’t. If you were a woman, if you were a person of color, if you were living here but not an American citizen, your rights didn’t matter as much to the people who wrote the rules and ran the companies.

That’s certainly true today, as well. I think that this is how we get upset about Uber, because men were taxi drivers and women were not taxi drivers. It’s okay for Etsy.

RM: It’s a downgrade for men, but it’s…

Just like we don’t really care about the cars people drive to deliver our pizzas. No one ever is after Domino’s for the depreciation on their cars, because we imagine, incorrectly, that they’re all teenagers. So this question of who counts is so important to understanding why the rules are set the way they are.

KS: Right, so let’s go back to that idea. Here we have the companies in Silicon Valley growing up in a way people, they have this image of the garage, the garage company of a group, a team of small, usually men, together, making one thing, or making one idea.

Sure, the garage counts as a place of innovation and certainly their ideas that are important. Shockley and all those other guys started inventing new technologies. Of course, those new technologies were reliant on government funds, in a lot of ways, that are just now written out of that story.

But they’re also reliant on Quonset huts, which aren’t as much a part of the story. The way in which Quonset huts also filled with men, just men who didn’t speak English, were dipping those different kinds of chips into big, boiling vats of material, without the rest of Silicon Valley couldn’t happen. It’s important to realize that those men at the top matter, certainly, in terms of their ideas. Often, they had the best impulses.

I read a lot about the history of Hewlett Packard and how Hewlett Packard goes from being this shining example of commitment to workers to falling apart and laying people off and losing its taste for innovation. That story takes places in the ’80s and ’90s as they move away from focusing on innovation towards focusing on short-term finance and profits. The story of that is deeply embedded with McKinsey’s reorganization, beginning in the early 1980s. These consultants selling a certain vision of how the corporation ought to be run.

KS: We tend to focus on the idea that this phenomena is new, what you’ve been talking about. It’s not. Let’s talk about how it morphed, this idea. Because Silicon Valley companies, the way people think, “Google did business this way. It’s the loose campus. It’s the weird food.” It’s the this, the that, then that has morphed into something else?

RM: Yeah, what’s the gig economy now? Could you just give, maybe, a blanket definition?

Sure. The gig economy, the easiest way to think about it is this very dissatisfying negative definition. It’s all the work that doesn’t look like what you expected. It’s not nine to five, it’s not secure. It’s ad hoc. It’s what economists and sociologists call “alternative work arrangements,” and so it can encompass a wide variety of things.

In the book, I call them all temp jobs, and this sort of lumps together independent contractors and actual people working for temp agencies, as well as freelancers and gig workers. The “gig economy” certainly has this exciting overtone of futurism and technology and being mediated through apps. But of course most freelancers today are not working through apps. Less than 1 percent of the workforce works through these kinds of apps. Most freelancers are just working, finding jobs, scrounging for work on their own, maybe assisted by the internet, but not directed by their phones to act.

RM: Which is kind of how we started talking. We were discussing how big the gig economy is, and you actually have a project called the gig economy data hub, because … I was trying to find out how many people, or what percentage of the population is actually working in these gig jobs. It really depends a lot on your definition. The Bureau of Labor Statistics recently came out with a report that the last time they did it was before the iPhone, so you expect the number of temporary workers — or contingent laborers, as they call it — to have boomed. It actually went down slightly, so they’re not measuring it. How does that work?

Yeah, so it’s really easy to find out if somebody has a normal job. I say, “Did you work last week?” You say, “Yes.” “Are you employed?” You say, “Yes.”

If you ask somebody who drives for Uber, you know they drive for Uber, “Are you full-time, are you part time, are you unemployed, are you temporary?” Different people answer “yes” to all those questions. So the way we even conceptualize work itself doesn’t fit neatly into our service, which the Bureau of Labor Statistics is fully aware of, and they’re trying to figure out how to ask these questions.

RM: Right. They created this for a different time period, these questions.

Yeah, and so they want to have sort of time-series data. They would make sure that the data lines up over time, and they’re fully aware that it doesn’t. And this is a real tricky thing because if you ask somebody if they worked in these kinds of gig economy days, no matter how clearly you ask it, if you ask in the last two weeks, a lot of people say no. If you ask, you know, it’s 10 percent. If it’s in the last few weeks, then that’s what they say. 10 percent of the people are working in this gig economy or freelance contingent work economy. If you ask over the last 12 months if they’ve done this, it’s about 40 percent. And certainly for young people, it’s about half.

RM: So we have anywhere from 10 percent to 50 percent of the population is working in the gig economy.

Exactly. It really depends on who you are and when you ask. And to say that it has to be done every two weeks for it to count is ludicrous, right? Especially when your hours shift, right? This is why it’s very hard for people who have salaries to understand the constraints of working people. So that the third of Americans that only have shift work, right, and may not get the hours they need from week to week, they need to make that up.

RM: Right, which makes me think if you talk about how people who are doing the gig economy, they’re doing that to supplement their income. They’re not making as much anymore, and you had some thoughts as to why they’re not making a stable amount anymore.

Sure. So if you ask where technology really intervenes in the workplace, it’s not the iPhone, it’s the Chronos system, that manages time cards for most major chain restaurants, that minimizes the number of hours people work from week to week to make sure that they don’t go over the minimums that kick them from part-time to full-time, where they have to start earning benefits. And this kind of just-in-time workforce is right at the center of this.

KS: Right. Talk about that just-in-time workforce because, again, it’s something that I think technology’s done, that’s created suddenly. The workforce going forward is going to be a largely just-in-time workforce. Correct?

Correct. And you know the just-in-time manufacturing came to America right across the street from that very first Macintosh plant at the Nummi plant, now Tesla, where Toyota had invented this as a way to minimize their inventory stocks. And it makes a lot of sense, and you can do the math.

KS: Sure. From a company perspective, for sure.

Totally. And this just-in-time workforce then becomes a model for a way to think about supply, whether that’s labor supply or components. So this is what these algorithms are all about, and it’s a very one-sided algorithm that works for the employer and not for the employee. And so working Americans are left scrambling to fill out their time.

RM: So in a way, Uber is a godsend for some people.

And it absolutely is. So JPMorgan did this really amazing study a few years ago and found that it becomes a direct … these kinds of on-demand jobs is a direct substitute for debt. So when people have access to them, they borrow less. Full stop. And so yeah, for some people it’s a great way to fill out their time card, but it’s not as good as a real job.

RM: But it still sucks. It’s like it’s a great way because they don’t have enough hours to make the amount of money that would make their living sort of regular and sustainable.

It sucks to be a working person in America if you’re in this. And I think that’s the point, that the problem isn’t Uber, the problem is working in this kind of capitalism.

RM: But it’s not like Uber isn’t part of the problem or it doesn’t contribute to the problem thereafter.

Sure. I mean Uber could be, it could give them a bigger share of the money. I’m sure their venture capital would dry up.

KS: Trust me, I get emails every day from drivers who talk about that, the tricks Uber plays with them.

Yeah, they do tricks. Lyft plays tricks, everybody plays tricks. This is part … They use this language of entrepreneurship and independence and flexibility.

KS: They want to get the most efficient use of these drivers, right.

And earlier in the program, you asked about how this is similar to the agricultural revolutions of the 19th century mechanical thresher. Well, people bought these mechanical threshers and then they borrowed a lot of money to buy them. They borrowed money for land. “Oh, watch out!” When everybody has makes a lot of wheat, suddenly the prices crash and you can’t pay back your debts. And that leads us to the populist revolts of the 19th century.

RM: So what’s the corollary? That the cars lose their value?

The corollary is that people think they are using this as a path to autonomy and independence just like on the farms. And in fact, what’s happening is that people in the east or in the west, in this case now in San Francisco, are making all the money and control the railroad. Today, is the equivalent of controlling the app. And those are the people that make the money.

KS: I absolutely think that workers are utterly fungible to them. As much as they go on and on about it, I think it’s just … And by the way, it was really interesting a couple of years ago, I had an interview with Travis Kalanick, one of the co-founders, the infamous Travis Kalanick. And he actually managed to tell the truth in an interview once. And I said, “What do you think about …” you know, we were talking about various things and his reputation.

And I said, “What about” — this is early, early in the self-driving phase, you know, before it was really gotten as big as it has, and it’s still going to be slow, by the way. And I said, “What’s going to happen with self-driving and all your drivers?” You know, what’s going on? He goes, “Well, you know what the problem with my business is, is the guy in the front seat. He costs a lot of money, and if I can get rid of him, that would be a business.” Literally, there was a sharp intake of breath by everybody in the room because he actually told the truth and I went, “Thank you. Thank you very much.”

And I was like, yay, he said it, like what was the truth. Which was because he has no ability to control himself. And it was really a fascinating moment because it was really the truth. It was, just as soon as they can replace you or if you go to, say, coal mining, you know, because the president is always going on and on about coal mining. They’re not doing the coal anymore, and if they do it, it’ll be by machine. That kind of thing. And so it’s an interesting time when these workers are fungible.

So part of that, yeah. And it’s important to realize that the sort of driving ethos of tech people is the virtue of abstraction.

KS: Oh, hello. Hello, Alex Jones.

Yeah, now that we’re talking …

KS: Alex Jones, meet the workforce of America. Go ahead.

Yeah. So this idea of abstraction, this is a tech podcast, so I can talk about computer programming, right?

This idea of abstraction encapsulation, the idea of not knowing how it works, this is what computer scientists are taught in schools as the best thing going. And that’s exactly what this just-in-time fungible workforce is. So it’s not surprising me that they hold it up as virtuous and inevitable. So the flip side of that is, I don’t think anybody — or rather, very few people — voluntarily like to mine coal. Very few people voluntarily want to drive me around Manhattan. A lot of people do it because they need the money.

So you know — and this is the other side of the mechanical thresher — that it’s nice that we don’t all have to go out into the farm every fall and bring in the wheat, right? And so it can liberate us. So this question of how that productivity is distributed is not a question of economy, it’s a question of politics. It’s a question of organization, institutions, social norms. And so technology is easy. What’s hard is shifting a culture. What’s hard is shifting of politics.

KS: Right, but it inevitably leads the fact that less people are needed. Correct? I mean, we’re going to get into that in the next section and talk about where it’s going, where work is going, but it inevitably leads to the idea is that you may not have a job.

As, as you have now. Yeah, absolutely.

KS: Oh, 100 percent. No, I always say anything that can be digitized will be digitized, and everyone’s like, “oh, that’s interesting.” Like, no, no. Think about that. Like think about it very carefully and walk it out. Not just the fact that we’ll have self-driving cars, but we won’t have mechanics. You won’t need a mechanic, you won’t need an insurance company. You won’t need … The way retail is done. It will not be at malls. What about the mall people?

KS: No, that’s hard. It’s hard! Rani, I’m in jobs where you … So creativity is the only thing that matters, really.

I mean, I think being curious and being creative, in short being human, is what matters. So things that aren’t human are things that a machine should do. I write in the book that no human should do the work of a machine. And to me, that’s not a bad thing. The optimistic futurist in me loves the fact, loves this, but I think the question is, well, what do we do with people? And we have a system set up, an educational system and economy that has treated people like machines for 100 years. That’s what industrialization is.

KS: Right. And in fact, the original computers were people, were women.

Were women, which is not to be ignored. Right? So this is, you know, why are we surprised when the robots finally come to take our jobs? And you know, it’s a godsend. It could be the end of paperwork, which I think all of us hate.

Or mining or all the other things, but it does mean, what do we do politically and socially to make sure that our societies aren’t ripped asunder between a new kind of digital ruling class? And that’s where the dystopian and utopian visions, and it’s why history is so important. Because when you talk to people in Silicon Valley, as much as I love them, they are people who have largely just read sci-fi novels.

KS: That have no historical underpinnings whatsoever. No humanities, no …

Yeah. And they think of the transition from the agricultural economy to the industrial economy as a smooth line rather than, in 1877, railroad cars with machine guns reconquering parts of Pennsylvania when the workers rose up and destroyed the tracks and overthrew the bosses.

KS: Exactly. The Whiskey Rebellion even further back.

RM: And then later, unions.

And unions that, you know, shut down supply chains, beginning in Flint. And so I think that I personally would like to come to an equitable solution to this before we have to put machine guns in the back of Ubers.

KS: I think that would require an intelligent political atmosphere.

RM: So sometime in the near future, or depending on how long you think it’s going to be, everything’s going to be automated, a lot of us are going to lose our jobs. What are the jobs of the future?

Yeah, I think it’s really hard to guess in advance. I mean, I don’t think “podcaster” was a job that you could have predicted 10 years ago, but we do know what those will be. They will be these things that stake out what it is to be human, to be curious and creative and also caring.

So I think that the jobs of the future will be us sort of caring for — liberate us to do the things that humans like to do naturally that we don’t have to be corralled into doing. You know, you have to corral someone into going into a dark mine and getting black lung. You don’t have to corral someone into taking care of their children.

RM: Can you give me some examples of what those jobs might be?

RM: What’s very human that we’d want to be doing?

What’s very human? I think caring for children or the elderly, I think doing scientific research, I think making art, I think explaining and meaning-making in various ways is very human. I think learning new things and doing things that are not repeatable. So this podcast, most likely, will not be the exact same podcast as the next one you have. And those kinds of novel things will be very appealing to people in the future.

And if this seems impossible, it also should seem impossible in 1850 that most of us would be sitting around moving paper, or selling things, or typing, or all the things that most people do in the service economy and that less than 2 percent of us would work on farms, and even the people who work on farms are basically managing robots. So I think that those are the kinds of things I think about in a positive way. Now, it could easily go in a very different way as well, in terms of that …

KS: Okay. we’re not going to be selling because, why? Stores are terrible. They just don’t make sense to run around a store and go collect things like you’re the hunter, like in a forest or something. You know what I mean? That’s what it is. It’s a hunt-and-grab economy, essentially. You go and find your things. I find walking around the supermarket idiotic at this point.

I think it’s very depressing to imagine all those giant cavernous air-conditioned warehouses that we’re burning coal for so that some people can drive up in their oil-burning cars …

KS: It’s not gonna happen. And then all those jobs go. All those jobs go.

Yeah, all those retail jobs are gone within the next 15 to 20 years.

KS: Yeah. We had the head of Walmart at Code a couple of years ago, and I think I’m the only one who heard him say, “We probably will only have 10,000-square-foot Walmart stores,” because just the people would go in …

RM: So no more Super Walmart?

KS: And everyone’s like, “eh.” I’m like, “What did he just say?!” The repercussions are massive. The same thing, so we won’t have that. We will probably have self-driving cars and we won’t have all the jobs that go with that. We might not even own a car. Right? You might not even have it. So when you think about doctors, there’s all kinds of jobs that are diagnostic that …

Radiologist. I feel bad for the radiologists.

KS: Me too, I say that. Someone’s mother was like, “My son’s in …” I’m like, “No, no. Don’t be a radiologist,” because there isn’t going to be any radiology. So when you think about that, talk about more of the jobs that are going. What are the areas? Retail?

RM: Which ones are going first?

I think the most important one is retail. Certainly for working people, the people who are already skilled in ways that they can’t access higher-paying jobs, retail. The places where people go as entry-level positions. Retail’s gone. Anything that can be done three times, has to be done three times in a row, will be automated.

And I think part of this acceleration I wrote about in the book is this idea of digital migrants. So sometime in the next few years, we will see robots that are tele-operated by somebody else, and I think people aren’t as attentive to this as they need to be. The intersection of machine learning, virtual reality and robotics. I’ve already seen robots that you can …

RM: Can you explain that a little bit?

Yeah, unpack it a little. It’s a little … It’s really interesting to me. I went to a lab a couple of years ago at Berkeley, and you could put on virtual goggles. Like we all now have these — well, I guess six people have the Oculus Rift or whatever. And you can run a robot body through that. And people there were very excited about this towel-folding robot that could see a towel and fold it. And I sat there for an hour waiting for this towel to be folded and it never could. I hate folding so I was super excited to see this. And I put the goggles on and I could fold the towel almost instantaneously, even though I’ve never …

RM: So using the robot, you could fold the towel.

I could reach the robot’s arms and fold the towel. And I realized when I did this it was like, oh wow, I could do this anywhere. And so I can easily imagine the next couple years, some entrepreneur offering very cheap house-space robots the same way that Tesla used its own drivers to train its Autopilot, to use just hundreds of thousands of people around the world through some kind of online labor program in putting on virtual reality goggles somewhere in Bangladesh or Mexico. And then operating these robots.

And then because of machine learning, the robots would learn how to do all kinds of manual tasks. So all that physical labor that we now think can’t be put oversees, all those migrants that people …

RM: So even the digital migrants would lose their jobs.

So even though the physical migrants would lose their jobs, but then the digital migrants would then be replacing themselves. And then there’d be a next step where you have one or two people orchestrate taking over when …

KS: It’s a little bit like “Ready Player One” when you were watching them do that. You know, they were somewhere else but they weren’t in the place, you know? It’s really interesting. I just took a virtual reality roller coaster ride. I’m not even going to go into it.

KS: No, not at all. I’ll explain it in a column soon to come.

KS: But it was really interesting and I was thinking, how do you replace … It made me think of jobs. It was a really interesting way to think about it. What about the caring professions? Because a lot of people think there’s going to be robot caretakers. There was just a story in Atlanta… One of the things about robots, sex robots, and that’s been in sci-fi many, many times. That’s the first place those dudes go.

Yeah, and people love sex, I hear. So there’s a lot of money in that.

KS: Those dudes always go there, with the sex robots. So talk about that, like, what happens? Let’s stick with the caring ones because I don’t feel like talking about sex robots. But what about the caring ones? Because one of the issues when I was at MIT many years ago — and I think I’ve talked about this before — is they have problems with the eyes. They can’t get the eyes right and so people are hard to replace.

“The uncanny valley.” Yeah, I think visual tasks are super hard to program and fix and this is why people, virtual reality is so key here. It’s that intersection where virtual reality can train people to change bedpans, train the robots as they observe, machine learning is all inductive, right? It’s pattern recognition. It’s doing something a million times. And those kinds of things will be initially done by people overseas or in poor rural communities in the U.S., probably poor rural communities first. And then will be automated over time.

We’re going to have a lot of old people pretty soon, especially in Europe and in East Asia. So the robots, this is the argument that some people make, that the robots are coming just in time to save us from our aging.

KS: Right. We’re not going to get into robot rights but, so …

None of this is artificial. None of this is like “Westworld”-style artificial intelligence, this is all narrow AI where …

RM: Right, so everything that can be digitized, will be digitized. A lot of things will be automated.

And it will be digitized by cheap people.

This is the important part.

To realize that these things are always based … In the book I wrote a lot about transitional labor forces. And the people who don’t count, in the sense that, people of color, women, migrants, people who are left out of the social compact. And where we draw those lines will matter just as much going forward as it did in the past.

RM: So if we don’t have jobs in the future, if we don’t get these caregiver jobs, there might be things that are automated out so that we have a lot more free time. Which could be its own thing, but how do we deal when we don’t have a job? What do we do?

KS: So talk about UBI, I’ve had a lot of UBI. This is Universal Basic Income. How do we pay for this?

Yes, so the Universal Basic Income is very exciting to people who think that automation is going to get rid of all humans. All current human work.

KS: No, there’s some interesting views, Chris Hughes is interesting. I just did … Annie Lowrey wrote a book about it. It’s a big … Sam Altman is all over the place with it.

Yeah, there’s a lot … and whether or not there’s a space for people going forward. I think people will always be valuable. We’re so versatile. We’re creative in a way that machines are not.

There’s tricks to UBI, right? And there’s different ways of doing it. You can either have a tax and redistribute, you can have something where everyone … The model I favor hearkens back to the early 19th-century model, the corporation, where you couldn’t have a corporate charter unless it fulfilled a social good.

And right now, as people often say, we have socialized risk and privatized profit. I think that people should get a cut of that profit. So every time you issue a stock, some public holding company gets a share of that stock and we all get a cut of that, rather than a direct payment. Because I think it’s important to not feel like we’re just giving money away to people. I think that purpose and autonomy matter a lot to people.

And I do worry what happens if we just start giving people money out. You see the kind of hopelessness that comes with that kind of direct payments. Whereas if we had a sense of shared ownership it would tap into older American values.

RM: You had a different term for Universal Basic Income or a different idea, you called it something like an “investment.”

Yeah, I think it’s an investment in each other. I think that’s a better way to think about it instead of as Universal Basic Income. I think it should be, because you go talk about “Ready Player One,” again, the sci-fi vision where people are living in The Stacks, stacked houses, and it’s all very bleak. Well, I don’t want to live in that world. And the danger of automation is not just losing your job but losing purpose.

So figuring out how we can liberate people from tedium and drudgery, which of course is what most people have at their job every day, still. But figuring out how to enable them, not everybody will be able to be a research scientist, right? Full stop. That’s just not how humans work.

KS: Well, you’re removing tedium and drudgery but your not replacing it with anything else. So tedium and drudgery is better than nothing.

Tedium and drudgery is better than starving but it’s not better than taking care of elders or children or making art.

RM: Okay, so we’re making art?

I am going to comfortably assert that …

KS: I’m saying we train people not to think like that, of course.

Yeah, and I think what we should be solving for is not giving everybody terrible jobs but trying to figure out how do we maximize growth and unleashing potential. We have a lot of challenges in the 21st century, like global warming, and we need a lot of people to deal, and so how do we deal? And this could be one of the ways we deal.

KS: What about the higher-level jobs? Because I think people talk mostly about the lower-level jobs. I think higher-level jobs are very much at risk. And I think what has happened, and that’s our listeners in a lot of ways, but that it’s not just lawyers. There’s a lot of legal stuff that can be digitized. There’s a lot. You can go through every profession of the white-collar jobs, essentially, that are all at risk, as far as I can tell.

Certainly, contract review … and again, you see the same process. Contract review was first outsourced to India to very smart, well-educated Indians who cost less. And it is quickly being automated. And you’ll see that in a lot of things. Certainly all the talk a few years ago was how do we automate or digitize higher ed, a field that I’m more familiar with than law.

So I think that one of the things that we can do is to think about, how do we use this as an opportunity to train billions of people and educate billions of people? But we’re also learning that in fact MOOCs did not destroy the university.

And it’s nice to create different levels of education. Levels of YouTube … like YouTube, I learned how to play guitar this summer watching YouTube videos. I can make cheese now. It’s pretty amazing, at least for me. I love cheese. And I think it’s going to be like that with law too. That part of being a lawyer will be, “How do I learn what I need to learn?” Just-in-time education. “How do I manage the AIs to do contract review to become more productive?”

So I think that’s part of what’s going to happen is that people that are higher up the ladder are going to be trying to figure out the right balance between handing off work to an AI, also receiving work from an AI, and that sort of layer cake of human-AI, human-AI. It’s not going to be one-directional like it is now, where a few algorithms are telling us what to do. We’re going to be working in both directions at once.

RM: Okay, so you, not a journalist, have a rosy idea of the gig economy of automatization.

RM: We’re very bleak, yeah. Everything’s dark and dystopian. In your book you describe two ways forward. A conservative and a radical path that would make automation, that would make gig jobs — or whatever the future of our jobs look like — sustainable. Do you want to describe those two?

Sure. I think we already kind of touched on them. One is we sort of figure out how to transition away … So one of the things I think about a lot is how the workplace is like a marriage. In the sense that marriage is disappearing. And the only people who get to be married anymore are fantastically wealthy and educated. Same thing with secure work. So a few elite people get to have secure work and everybody else is in this world of insecurity.

And just the same way we are redefining how our society supports the children that come out of marriage, we need to redefine how we think about the benefits and rights and obligations that float out of a workplace. And one way is to just look to the 401k. The 401k is a pretty conservative solution for retirement. And what you do there is every time you get paid, you put a little money into a retirement account. And now it’s tied to a job, but you can easily imagine just creating a more universal system of portable benefits.

RM: So you get these benefits no matter which job you’re at?

Yeah, so as you move from your gig to gig, your 1099 to your 1099, you get sort of … A slice of that money goes into an account for your health or for your retirement or your childcare. And right now, it’s really onerous because we assume everybody has these W-2 stable jobs. So that’s a conservative, just fix that, make it transactional, make it easier and businesses can create a more fluid workforce.

Now, a more radical solution is what we just talked about. How do we think about — I don’t want to say “socialism,” but in this case — how do we make capitalism work for everybody? So that our corporations aren’t just accruing money to the top .1 percent. And so that we socialize the benefits as well as the risks. Without getting rid of all the wonderful things about markets that make our lives easier and more efficient. So I think that’s a more radical path, sort of rethinking how corporations work as well as …

RM: What would it take for any of this to happen?

Unfortunately, I think it’s going to take a giant economic crisis. I think it’s going to take people in the streets realizing that things have gone horribly awry. It would take political parties realizing they can’t just do quick fixes, that we’re actually at an epochal shift. And that epochal shift is not the smartphone. The epochal shift is that individuals can sell, consume and work globally, maybe through a corporation, maybe not.

And that’s actually what’s truly different about right now, that the corporation is no longer necessary to organize workers. And I think that is something that we have not wrestled with and we keep trying to put everything back in Pandora’s box, but the corporation as it existed is no longer there.

KS: Yeah, so what are we going to replace it with?

We can’t put band aids over our employment law, yeah.

KS: So, let me ask you this then to finish up, and then Rani may have a final question, is what can tech do to help assuage this? Because they’re causing and creating … A lot of the things they’re making are moving to this.

Certainly, when I talk to my fellow computer scientists at Cornell, they’re very afraid. They are building AI and they’re afraid that’s the equivalent of the atomic bomb. So not just technologists but also computer scientists and academics. And I think what’s important to do is have conversations so everybody knows what they do but they don’t always know what it means.

KS: Really? We hadn’t noticed. Hello Cambridge Analytica, nice to meet you.

Exactly. So I think it’s important to have these conversations across disciplines, outside of our own narrow fields. I think it’s important for technologists to be a bit more humble about the kinds of problems that they’re solving.

So one of the things I write about in the book is that technology tends to solve for social problems. And the problems are the people who can afford to fix them. So, the problems of 23 year old bro in Silicon Valley are pretty much dialed.

KS: Yeah, you know my joke right?

No, I don’t know your joke.

KS: “San Francisco is assisted living for millennials.”

KS: Thank you very much! I’m here nightly. It’s true. It’s so true through. They’re just lazy. Whatever.

I used Grubhub last night, it was delicious. So, I think yeah, that we need to have these conversations. It’s wonderful to see that so many people are excited about the UBI. I think it’s kind of condescending to think that people have no value unless they’re programmers, which is often the vibe I get when I’m in Silicon Valley.

And it’s also ridiculous to think that everyone is going to become a programmer and only they will have jobs in the future. So, I think sort of being open about what is valuable, going forward about being human. For me this is a wonderful opportunity. It could also go very badly if we don’t make the right choices. And it’s important to emphasize that we do have choices, that it’s not determinant.

RM: So I guess some of the people that are really pushing this UBI are Mark Zuckerberg, a lot of the CEOs of the world. What do we have to do to have these — not necessarily UBI but any of these social, I guess you would call them nets — happen?

I think we need to realize what has changed the workplace and it’s not just enough trying to … When I talk to a lot of people, they think that we can cram ourselves back into the regulatory state of the postwar. And without having the attendant social environment — and the social environment is not just the “restoration of unions qua unions.” It’s about the voice for workers in a distributed, just-in-time economy, in a logistically based economy that we don’t have.

The AFL doesn’t do it, the CIO doesn’t do it, and we need to have a new kind of workers’ voice just they way the CIO sort of organized industrial workers. And that workers’ voice is essential because laws don’t fix the economy. Laws sort of address little problems at the edges, but they don’t address this question of power.

And it’s power that really makes the economy go and so I think that’s what we need to do. And also this question of who counts, who deserves to have, not a secure job, but a secure life. And I think for me that’s the difference. We thought that a secured job lead to a secure life and that’s what we were solving for, but we really need to make sure that everybody gets that going forward.

KS: Or, welcome to populism, right? That’s what we’re seeing now.

Or, welcome to populism and Mark Zuckerberg won’t have to worry about the UBI, he’ll have to worry about all of us coming to his house, whether the one he lives in or the one he’s bought next door.

KS: That’s the thing, they’re moving to New Zealand in case you’re interested. In their giant compounds.

KS: You know about that. They have all these plans. They want to go to Mars.

RM: Oh, right, right, right.

KS: They want to get off the Earth and they better move fast in a lot of ways. You’re right, 100 percent. This is really an important issue and there’s all kinds of issues that sort of hang off of this over the future, so we’d love to have you back at some point.

KS: And it was great talking to you. Thanks for coming on the show and thanks to Rani for joining us as well.

This article originally appeared on Recode.net.