A Brief History of Financial Bubbles

have mine. You can go ahead and click that link and take care of it. And we're going to talk about a lot of stuff, but really what the conversation is going to focus around is going to be bubbles and having a conversation about how how financial bubbles happen.

After a couple of quick messages, here's the interview. All right, welcome back, And as I said before in the introduction, I'm sitting down with somebody a little.

stage for us. You graduated from Harvard Law School, you're working at Lehman Brothers, and what position were you in when two thousand and eight starts to hit, Like, what kind.

we ran in a lot of the same circles. So when he started PayPal, he offered me a job, and eventually I ended up at PayPal in two thousand and

two and beyond. But I did stay very close to a lot of my colleagues and friends at Lehman Brothers who had worked there during those late during that period in the late nineteen nineties when I was there, and so I got a pretty good glimpse of what was going on between two thousand and two and two thousand and eight, And was also a finance lead at PayPal, so we were working in the capital markets and working

with investment bankers all the time. PayPal went public in two thousand and two, got acquired by eBay, and I was on the finance team and was chief financial officer at eBay in the North America business when it all went down in two thousand and seven. In two thousand and eight, so I'd say I got a pretty good front row seat to the show and then a little bit of backchannel with just colleagues and friends of mine on the street.

But as someone who was more on the inside, did you have a sense that things were moving in a rapidly negative direction before maybe it was starting to hit the broad media or did it come as just as much of a surprise to you as to everybody else the way things started to all of a sudden, very quickly unravel.

And I loved financial derivatives. I was actually a trader of financial derivatives and between nineteen ninety six and nineteen ninety eight, so that's where I got my training. And is I think as early as two thousand and six, you began to see things like spreads on some of the securities that were securitized by real estate bonds, by commercialized debt obligations, so very very difficult, very hard to understand things that you know, most people don't really pay

attention to. But if you do look at securitized bonds and those trading spreads and how those how those CDO spreads were moving around, you could see as early as two thousand and six that there was a bit of a credit risk. And then it just accelerated and became it hit the popular imagination probably in two thousand and seven, when it began to defect equities and real estate prices,

and it began to see the defaults coming. So I was probably about a year I had just in terms of my understanding of the market.

have should it would have happened. And I covered this in great detail in a chapter on my new book, which is called The History of Annational Bubbles, and it comes out in early next year, and one of the chapters was on the subprime bubble. I think it really started back in nineteen ninety two. There was a law

passed by the federal government. This was passed by a Democratic Congress but signed by President Bush Senior, and it was called the Federal Housing Enterprises Financial Safety and Sound Is Act, and it began to for the first time, direct Fanny May and Freddie Mack to deploy money that they were spending to stabilize the mortgage market into LMIs or low and middle income households, which are basically households that are at or below the median income in the

neighborhood that they're in. And these quotas began at thirty percent, and then Bill Clinton takes over, and then the quotas bidon't expand from thirty to forty to fifty percent. By two thousand and seven eight it was fifty six percent. So it was really the Clinton presidency the Bush presidency, both parties of Congress that were pushing these quotas, and Fanny and Freddie what they were doing was deploying capital into it became the subprime bond market. And that's I

think something that took both parties by surprise. They didn't realize how quickly the subprime bond market was expanding. Between two thousand and one and two thousand and seven, there were there were not a lot of defaults in the subprime bond market, even though that that market was a little in niche industry when I was a bond trader in the nineteen nineties. It multiplies by ten x by the time it get to the middle of the two thousands, but it didn't. It didn't default because the quality the

mortgages weren't high, but that wasn't immediately apparent. And then interest rates were held down by the FED between two thousand and two and two thousand and six at historically very low interest rates, so they kind of kicked the can down, you know, down the street a little bit, and eventually it metastasized in two thousand and seven, and I think it's fair to say took everybody by surprise.

almost a quintessential moment where a policy clearly just missed. Like, I know, I can see the goal, I understand the direction that it's supposed to go in, but it doesn't go that direction.

What they tended to do was either they built housing in Europe to something called social credit housing, where you basically build build housing and make it available to low LMI individuals or families in certain neighborhoods. You could also subsidize those individuals directly and give them a credit towards a purchase of a home. But only in really four countries did the policy become subsidized banks or direct financial institutions to give loans to people who wouldn't otherwise qualify

for them very easily. And those four countries were the US, the UK, Spain, and Ireland. Canada didn't do it, Japan didn't do it, Germany didn't do it, France didn't do it. And so look at where the two thousand and eight bubble hit disproportionably, It was those four countries where the subprime bubble happened. You had a proliferation of lower quality loans, a massive increase in housing prices, followed by the collapse

in housing prices and the compromised collateral that resulted. I was living in Canada, grew up in Canada, so I had a pretty good you know, my mom and dad own a house in Canada and they were looking at the US. I remember watching them and talking to them about the US housing bubble and Canada the housing prices went up a little bit, but not a lot, and it never had the same shakeout that the US did. So that's that's a pretty good clue as to how

to do it. Like maybe Canada and Germany and France and Japan were onto something and in the US there were not. And I think what happened was in the US by the late nineteen nineties, Fanny and Freddy they became market leaders in mortgage lending, and they were telling originators that they wanted more of these loans from lower

income households. And if you're a bank and you've got Fanny and Freddy that are your two biggest clients saying I want more of these loans from low income households, Well, what do you do? The banks had to lower their underwriting standards, and Fanny and Freddie were buying loans with a loan to value of ninety percent, ninety five percent,

ninety seven. By the time you got to two thousand and seven, they were buying mortgages with zero down and all these unusual payment features like no down payment or teaser rates, negative amorization features that stuff just never happened in Canada where you didn't have players like Fanny and Freddie moving down credit. So it is it is, I think, just a matter of unintended consequences and not thinking through

all the implications. And then once the policy gets rolling, it's really really difficult to roll it back because you have politically motivated interests that want to maintain the status quo.

market difference post two thousand and eight. You know, as you're talking, like part of me starts to think, you know, we're looking at these countries, looking at Canada, We're looking at you excuse me, the United Kingdom of Ireland, Spain, the United States, And then I'm thinking about this policy, and I'm thinking about things that work and part of me wonders, like, is it just that the government isn't

willing to assume enough responsibility? You know, we have this weird partnership between you know, capitalism and private equably, you know, the private side and the government side, you know, and here we're trying to accomplish a policy objective which is laudable that maybe should have just been handled by the government itself, and instead what we're doing is we're kind

of outsourcing it to private banks. And as you say, the clients are Freddie and Fanny saying, well, this is what we want, give us this, and it's just the outcome is adverse to what we're expecting. You know, what do you think is the problem that you know, the government just needed to be more involved. We just needed to build houses, We just needed to do what Canada did. I mean, why isn't that an appropriate answer to the problem we still find ourselves in.

recurring thing is easy money. Sometimes it's it's very obvious, like in twenty twenty twenty one, the Federal Reserve just cut interest rates to zero and held them there for two years, and that fiscal surplus, and that that fiscal deficit rather and that monetary stimulus created an excess of demand over supply. Sometimes it's a little trickier to understand, like the incentives created and something like Fanny May and

Freddiemac and government direct lending. So you know, it's a little bit of a tricky balancing act sometimes that these governments try to try to try to try to manipulate. But ultimately the something like housing, the market works I think quite well to develop to do things like build housing where it's needed, to figure out how to underwrite

loans and provide high quality loans. That the downside is if you're subsidizing or creating a negative incentive or incentive for banks to lend to people don't otherwise qualify, then then you start to distort the market. I think that's actually the worst. The worst outcome. The better would be to identified populations at need housing you can help them directly.

You can, you know, you can sponsor projects or make it easier to build in certain areas where we have a need for low income housing, And that to me seems like a much more constructive role for government in housing, the subsidization of rents, try to medipulate prices, holding rents down, rent controls, that these things always have unintended consequences and never quite achieve the outcomes that you hope.

hugely negative consequences. The other ones that are interesting are the positive bubble, so the ones that actually do bring investment into a new acid class and then that finance is a new technology, the ones that I covered in the book, or the nineteen ninety seven to two thousand telecom dot com bubble, which did create a lot of wealth, and then a lot of wealth was rapidly destroyed from

top to bottom. The NASTAK went down by seventy eight percent, so there were a lot of people who were burned by that. But if you consider the lasting impact on the US, you had the emergence of many the tech companies that now dominate or landscape. You had Nvidia was founded at the start of that bubble period. You had Amazon, which was in ninety five. You had PayPal where I worked in ninety eight. eBay was ninety five. Yahoo was founded in nineteen ninety five. Google was founded in nineteen

ninety eight. You also had Salesforce in nineteen ninety nine, and Broadcom came at the tail end of that and then even out of the detritus of that bubble. So the two founders of PayPal were Peter Teel and Elon Musk. Peter took his profits and then funded Facebook, which was founded a little bit later, but primarily with Peter's money, and Elon Musk of Force went on to fund SpaceX and Tesla, and you bought x, dot Com and Twitter

and the Boring Company and everything else. So very in the long run, a very constructive, productive bubble which has

created a lot more value than was destroyed. And the one that was interesting was the eighteen forty five eighteen forty nine UK railway boom, where again you had a lot of investment in a short amount of period and overbuild of a technology but ultimately the UK came out of that stronger then it went in, and then went on about a twenty five year run where their economy was expanding and they became the wealthiest country in the world and really led us into the Industrial Revolution that

also came out of that bubble. So those to me were really interesting.

mini recession in two thousand and one. A lot of that was the result of the telecom part of that bubble, not the Internet bubble, and then became out of two thousand and one, growing and with market leadership and you know, new technologies that have made in America among the richest countries in the world today. So I'd say that's a net positive.

They went up by twenty times and at the peak of the bubble you could basically buy a house on the Amstell River with a single tulip, and that tulips were commanding prices that were three to six times the average salary of a merchant at the time. And then just as quickly as the excitement had built, the excitement went away, like the brittle fog in Amsterdam that sometimes rolls in over the river. If you ever experience the

weather there, you'll see what February is like. And it was just like that, a chill rolled in and all of a sudden the market died, and the result of it was no economic value. I found. Actually, maybe that

was interesting. One was in seventeen twenty there were two massive companies that were built side by side in London and in Paris, the south Sea Company in the Mississippi Company, and they were built as these massive trading enterprises with financial engineering, and they both got obliterated in seventeen twenty and seventeen twenty one, and at their peak they were both worth more than Nvidia is today, and like today's dollars, the south Sea Company was a maybe a six trillion

dollar market cap company. The Mississippi Company was eight or nine trillion dollars in today's dollars. There were the second and third most valuable companies of all time, behind only the Dutch, East India company around the same time, and then you know, within within a year there were no more and it left a lot of damage in their wake.

I mean, it was, but it was basically a bunch of networks built by a bunch of academics and government officials, and you couldn't make money on it, and it was sort of, I don't know, you know, kind of useless, frankly. But Netscape changed everything and there was a browser that was a consumer facing browser. People could now use the Internet.

There was massive adoption of technology TDMA, and it rolled out a telecom network and fiber optic cables that have just connected the world in a way that hadn't been done before. And you could even argue that you needed the bubble, you needed the excitement of that mania to really make that investment happen, to catalyze investment in a new category. And without a little bit of irrational exuberance,

it might not happened. And so there are losers along the way and people who jumped in and then lost money. But then the result was a technology that you and I and others are using and that were utility from even today.

think we're in right now, the AI bubble. Okay, there certainly there's been a lot of discussion on this topic, and I guess, well, let's just talk about the hallmarks of a bubble, and do you think we're in what at this point?

and some people would define it that way. I don't really know what that means. I don't really understand how you you know your opinion about fundamentals and valuations aside, I don't know how you decide based on that definition.

What's a bubble? And what are the fundamentals with cryptocurrencies or you know, tulips I guess, like you know, it's hard to then it's hard to say that that is a bubble if you don't accept the premise that fundamentals and valuations and the relationship between them are something we can all agree on as a bond trader and as

an investor, I don't think we can. So the hallmarks of the AI bubble, if you want to argue we're in one would be one a very rapid run up in valuations and a handful of companies that are AI native AI adjacent. I would say, you see a little bit of that in the public markets. I think in here in Silicon Valley you see a lot of it.

There are a lot of young kids pitching companies that are AI native and getting twenty five to fifty million dollars valuations, whereas when we invest in companies with real businesses, we try to stick to five to ten million dollar valuations. The business models, you're not very clear. There's a lot of you know, which you might call what John Maynard Kains might call animal spirits, what Ellen Greenspan might call irrational exuberance. What my mother might just say is, these

kids don't know what they're talking about. So three, compending on how you define this period of excitement, you know, it's some combination of excitement that just seems to exceed the fundamentals of the business. There's a lot of capital intensity. That's another hallmark of bubbles. The amount of money we're spending in an AI capex. If you look at the you know, the especially in the public companies and the hyperscalers, we're probably at about four to five hundred billion dollars

in AI capex this year. That's one and a half to two percent of US GDP. At the peak of the telecom bubble in nine to nine two thousand, we're probably you know, five or six or maybe seven or eight percent of US GDP was in those investments. So or we're beginning to approach nineteen ninety nine levels of investment. The other side of the argument, though, which is kind of where I come out, as I think the businesses in the fundamentals around AI are a lot better than

they were in say, nineteen ninety nine. In nineteen ninety nine, the NASTAQ one hundred peaked at seventy three times priced earnings. Today the mag Sabin is like twenty five to twenty nine times price earnings. In Vidias, the company at the middle of everything, they're trading at twenty four times fully taxed earnings for next year. In ninety nine, the company at the center of everything was Cisco, and they peaked

at two hundred times earnings. And they were the ones that were like building routers and building the technology and the picks and shovels of the Internet. But they didn't make a lot of money. Their best year, they made money. They made two point seven billion in profit in two thousand. They lost money in two thousand and one. And Vidia is a very profitable company. Not only are they cash

flow positive, but they're buying backstock. They spend one hundred and twelve billion dollars buying backstock over the past five years. Cisco didn't do that. All the customers that Cisco had, well not all of them, but half of them were telecoms, and some of these were Worldcomm and Global Crossing. Those were not profitable companies. They a lot of them. Those two bankruptcy in fact, and had accounting scandals and massive restatements. In two thousand and one and two thousand and two,

a couple of their executives went to jail. The customers for today vidious customers are Meta and Microsoft and Alphabet and Amazon. They're all profitable companies, paying dividends, buying backstock, and cashlow positive. They don't have more cash than debt. Those the mag seven has three hundred billion dollars in cash more than debt right now, and that was the inverse.

The telephon companies were massively indebted back in two thousand and Maybe the most important thing is all those companies back in two thousand were laying dark fiber like they were building fiber optic cable networks in advance of demand, but the supply where they'd built so much supply and capacity that the demand never caught up. So at the peak of the bubble, ninety seven percent of the global

fiber optic cables were dark. They weren't being used. Every GPU that Nvidia rolls off of its assembly line lights up. There's demand for it right now, So the hyperscalers are buying all the gps that can yet they're they're letting them up, and they're actually not constrained by supply. It's more of a it's more of a demand problem, if so, meaning, and they could probably double the number of GPUs are

selling tomorrow and still be sold out other inventory. So those are the reasons I don't think the AI bubble quite qualifies as you know, what you saw in nineteen ninety nine or eighteen forty five.

It'll become a super intelligence that can do pretty much whatever whatever it wants to, and have initiative and the ability to, you know, to take control of different processes. I don't know that we'll frankly ever get there, or maybe that that is much further away than than people think.

But at the same time, there are a lot of AI applications that will emerge over the next five years that I think will make life significantly different for most people without I'll get without getting all the way to you know, sky in It or Cybernigne systems and having something to run your run your defense and plan your day and tell you what to do and cook your food and everything else, and you know, maybe order you around or make to the human battery if you go

down the matrix, the matrix, you know, pessimistic view of the world, I guess there are optimistic versions of a utopia where that where it could all be run by

computers and happy. But it does seem like AI is going to, you know, within five years, they will be legally they'll be they'll be drafting legal agreements and creating contracts, and they will be driving cars, and they will displace a lot of human activity right now that we are doing that you know, we don't really need to be doing and so that will have a life changing set of impacts on what we do and how we work

workflows for companies. I think it'll remake education. I've got two daughters from school who are eleven and fifteen, so that's top of mind right now. I think that's another area where we are still teaching people, you know, how to work and how to live in a way that reflects in a lot of ways in industrial society from the nineteen twenties and having education hasn't really caught up, but

it will. And then healthcare, the diagnostic outcomes from you know, better analysis of medical data will be quite literally life changing. And so I think those are all things that you can get to without getting to AGI.

What are they all going to do? You know what? And how do we find jobs for them? This has been the same argument that effort about every technology. I remember hearing it back in the nineteen nineties about the Internet and computing and that was going to, you know, put a lot of people out of work, and it was going to have this big productivity bump and then people will wouldn't have work to do. Accountants and lawyer were kind of, you know, front and center at the time.

I remember thinking about what profession do I want to go into? And I was the bond traders, hosting a lot of you know, manual work like you settle trades at Lehman Brothers. You actually have to write out what your purchase of, what your sale was, and your file paperwork at the end of the day. And now a lot of that stuff is just automated, right The Nastak and all these exchanges could entirely run on computers. The trading floors if you go to a big investment of

make today it's comparatively quiet. It's not the trading floor that I grew up on, or it's not the trading floor of trading places the Eddie Murphy dan Akroyd movie where there's like the yelling buy and sell orders. So what do all those you know, what are all those couriers do? Well? They all found new jobs. We have a four point six percent unemployment rate today in the US.

If you go back even further, at the end of World War One, World War in the in the nineteen twenties, forty percent of Americans lived on farms worked on farms. And if you go back to eighteen hundred, eighty percent of Americans lived on farms. But then the agricultural tech logical revolution happened. You had the spinning, the spinning Jenny, and the textile you know, manufacturing technology that just automated

all that stuff. And now we have two percent people living on farms, maybe down to one percent now, and they're producing all the food that we need and then some and we don't have an eighty percent unemployment rate. So all these technologies, you know, created as much work as they displaced and productivity has gone up and people have found new things to do in the Internet. We didn't know it at the time, but it's created a

whole new digital economy. So for every job that the Internet obliterated, you've now got YouTube influencers and people making money in media, and people are in the computer science field, you know, generating incomes off of these off of these new business models. They just just weren't therefore before PayPal and before eBay, and before Amazon and Google and YouTube and everything else. So I don't really know where these

jobs get created. I would probably tell my daughters to study AI and get comfortable with robotics because I think that will have something to do with the future, and I might I might guide them away from being a driver as a profession or a you know, a call

center operator. So a little bit of this is free career advice, but you know, this is also the belief that these technologies are net positive and if you embrace them and retrain with the right mindset and skill set, then that's what you have to do to benefit from the new world.

I'm just going to tell you that right now. But having done both at various times in my life, you know, sometimes I wonder about this, and as someone in education, you know, I sometimes look at the way we do things and think, like, I mean, it's not nineteen twenty, Like, why are we.

Look at Hollywood today. There was a big writers and actors strike earlier this year and the number one issue was we don't want not Maybe not the number one issue, it was one of several issues how AI works with artists. And you know, I can point back the nineteen twenties as being another example of a technology that that disrupted the film industry, the talkie. If you look at the early nineteen twenties, you had black and white films and there were there were I was gonna say silent, They

weren't really silent people. Did you know? There was music but it was like a live organists right in a theater who gave it a sound. And you had wonderful actors like Charlie Chaplin and Clara Bow who could express more with her eyes to beautiful actress, you can express. She could express more with her eyes than most actresses can with the you know, the full benefit of visuals and sound and voice. Douglas Fairbanks and they didn't, they

didn't embrace the talkie. And if you look at the early talkies, the Jazz Singer's the first one in nineteen twenty seven, you look at it now, it's like, huh, that's not great. They're they're out of sync. It's awkward. It just feels like a little bit underwhelming. But that's that was technology in nineteen twenty seven. That was life changing. That was that was like special effects. It was like

a light bulb went off, you know, for viewers. And after nineteen twenty seven the sound gets better and then, you know, slowly and surely a few of those actors never made it right. Douglas Fairbanks and Mary Pickford and Clara Bow never made a talkie and their careers ended in the nineteen thirties. Charlie Chaplin held out. He did his first hoke in nineteen thirty nine and was wonderful

in it. But some you know, some actors had weird accents, I guess and speech impediments and thought it would compromise the art, and they thought that this technology will be destructive, and that people like Walt Disney with this animation and the talkies and these musicals, you know that there would be the death of Hollywood. And you get to the nineteen forties and fifties and Hollywood's never never been stronger. Right, there's the golden ear of Hollywood because they ended up

embracing the technology. And I guess it is leaders like Walt Disney and others who who said, you know, here's how to make this work. And then when you actually show people the impact of animation and sound and music and you can see Fred Astaire and Ginger Rogers dancing in the movies like he couldn't before, then you know, then people feel like, oh, maybe it will be better. But it takes that, It takes those examples of innovation, and I think that's success to inspire people to be

happy with the technology and then embrace it. And until then, you know, fear, fear and fear mongering often wins out.

I wanted you to talk about the book a little bit. First of all, when can we get it? What's the exact title. Let's get the folks the information that they need. And then what are you going to talk We've talked about some of it, but you know, what's the thesis of the book, Like, what are we going to understand by looking at it.

Amsterdam in sixteen thirty six, the Tulip Bubble. It then goes to the seventeen twenties and the south Sea, Mississippi Company, and then it ends in twenty twenty twenty one, which I call the everything Everywhere, all at Once bubble, which was a very unique, multi asset, multi country bubble that was created. And then I kind of do the epilogue on AI and is AI a bubble? And you know you've already hehard what I think about that. When the reason I really did the book was, you know, one

part it was just education. I lecture guest lecture economics at Harvard Business School, and I found those students who are in their mid twenties and even a lot of entrepreneurs who in the valley. They just don't. They don't have a very short memory. The average lifespan, I guess of a money manager, even on Law Street today is about eight or nine years, like the only managed money have been managed money for the last nine eight or nine years. So they don't remember two thousand and eight.

They don't remember two thousand and one. They haven't read about Japan in nineteen eighty four. They always think these new investments and you know, these technologies are different, the opportunities we should be investing in certain technologies, you know, at ridiculous evaluations, and they don't know or remember what happened in the last bubble or how to navigate them.

So one part of this was just to do a research project on historical bubbles, how they come together, what happens, and then how do you navigate them once they're here. And there's some lessons for policymakers on you know, recurring patterns with bubbles that that that that repeat, and how not to make those repetitive mistakes. And then in part it's it's part economics, but there's also a lot of

history and social context behind each of this bubbles. So if you're interested in history if you're interested in in politics, or why did the bubble in Amsterdam happen in sixteen thirty six and you know not sixteen twenty six? How come it didn't happen in Paris? How come it didn't happen in London? How come it was tulips? Why wasn't it I don't know, sugar or tobacco or land. I go into all these different you know, backgrounds and aspects

just to be just to be educational. So I hope it's going to be fun, entertaining and educational read for for your listeners and others.

was fun. It was a narrative. You know, you talked about things like I talk about my class evaluations in nineteen twenty nine versus today? Are we higher than in nineteen twenty nine? And what happened historically? When valuations are training at mid twenties pe multiples? What is their forward looking returns just based on historical data and analysis. When my daughter learns history and her last school. There was a lot of facts and figures and you know, memorize

this guy and this battle on this date. And I think she, you know, she didn't get as much out of it, probably as as she might have, because I don't think it was taught in a way that was very compelling or that speaks to the student. And unfortunately that you know that a lot of that's just on the teacher, right making it, making something that's going to be relevant to your student and to capture their imagination. Is he says, you got to be you got to

be a good teacher. And maybe that's where you know, maybe that's where the opportunity is. Maybe that's where AI or just you know, these technologies can help to make in podcasts like yours, they make history irrelevant and entertaining and something you can really get into because talk, right, it is fascinating stuff and very valuable to an investor.

the editor. It should be end of January. If you go to Big Bubble Trouble dot com, you can do you can get in the pre order list right now. It should hit Amazon and all the fine bookstores in January of twenty twenty six.