Peter Thiel on Financial Markets and the Singularity

All right, there we go. I thought I would dispense with the PowerPoint presentations and sort of the detailed slides and just share a few different thoughts I've had on investing in a world where the possibility of a singularity exists. It's sort of a very different perspective from some of the discussions people have had. I guess there's sort of two different levels one can think about investing. One is as a venture capitalist, investing in early-stage companies. And, you know, the basic rule there is that, you know, the basic rule for investing in general is you want to do something that's fundamentally true and that nobody else sees. And that's how you do really well. And artificial intelligence or near-artificial intelligence or quasi-artificial intelligence.

This is so out of fashion today that it's probably the only thing thematically that I think it makes sense on a venture capital side to probably do, is to try to identify things that are promising in this sort of direction. So I think the venture capital end of it is fairly straightforward and, you know, there are all sorts of details you can discuss, but I want to focus instead on a very different element of it, which is the big picture, in terms of how do you invest in the world as a whole, and what's going to happen to the world's stock market or the world's larger financial markets in a world where this possibility of a singularity or something like this exists. And sort of how will the world's markets be different from a singularity-type world, from a world where such a thing would never happen or not even be possible.

And I suppose the basic intuition that I have about it is very simply this, that in a world in which there is a possibility of things going extraordinarily well or extraordinarily badly, the good things and the bad things are bigger than people think, in a sense where if you have a bell-curve distribution of possible futures for the world, the tails on that bell curve are much fatter than people think, and there's far more that can happen at the far edges. And this would lead to sort of a very different behavior in markets from sort of a normal bell-curve of distributions where nothing really that interesting or extraordinary is going to happen. And in particular, singularity will either be very successful, in which case we're going to have the biggest boom ever, or it is probably going to blow up the whole world and there will be nothing left to invest in whatsoever.

And so this leads to kind of an interesting investment dynamic. Now, I will start parenthetically by saying that the second category is one that's rather difficult to invest in. And so, you know, if you are—if you are somebody who is predicting the end of the world, even if you're right, I think you will still not make a lot of money. Even if you put all your money into gold coins and a silver chest and hide it in some forgotten corner of the planet, when the world does come to an end, there will be nothing left to buy or to sell, and probably well before then, some humans or robots or something else will have come along and taken your gold away from you. I think that's very difficult. The bad versions of the singularity are things that one cannot invest in at all.

They are not investable. There are all these possibilities that one can't even really think about. And so in some sense, if you believe in something like this, you have no choice but to bet on it as an investor. And the best investments will be the ones that represent the most aggressive bets on it. You have no choice at all, ultimately. And of course, the problem is, and the challenge is, that maybe the things you will choose aren't exactly right, and they turn out to be off, and you thought the singularity was going to happen here, and it turns out you made a small mistake, and these catastrophic approximations lead to things that will be very, very disappointing. And so, what would you expect to see happen in a world where the singularity is near,

the alternative to a good singularity is the apocalypse, and we don't really know where it's going to happen—well, you'd expect the world to be full of massive manias, booms, and busts on a scale unprecedented in all of history. And interestingly, if you actually look at the world's financial markets over the last 25, 30 years, that is exactly what they have manifested. And it's one of the things that I think is very striking, that all of the conventional theories say that, you know, markets should be getting more smooth and efficient as there's more and more information out there. Somehow everything gets smoothed out, the volatility gets suppressed, and everything—stocks should move up like 6%, 7% a year in a smooth, monotonic function.

And instead, we've seen bigger booms, busts, bubbles than ever before. And I want to suggest to you that one of the ways to think about a lot of these incredible booms that we've seen are that they represent different bets on the singularity or on various things that are sort of proxies for it, like globalization. Or, you know, if you have a singularity, there's presumably going to be one entity or one company or one technology that's going to basically take over the whole world and will basically be worth more than everything else put together. And so you've had these series of bets on what that is going to be over the last 25, 30 years. And if you actually were to do a chart of the booms and busts in the history of the modern world,

the amplitudes of the booms and busts have gotten much bigger. So, you were like a doctor looking at this graph, and you said it starts going like this and oscillating really wildly in the last quarter century, and you said, 'Well, nothing interesting is going on in this world'—you're probably not the person I'd want to have treating me or trying to make sense of what's going on. Now let me just run through a few of these by way of illustration. I think the first of these hyperbooms took place in Japan in the late 1980s. And I think it is no coincidence that Japan in the '80s was perceived as being on the cutting edge of technology. There was all the fifth-generation AI stuff that Japan was pushing.

And there was sort of a sense that you had all these textbooks about how Japan, Inc. was going to run the whole world. By the peak in 1989, the Japanese stock market was worth more than all the other stock markets in the world put together. The Emperor's Palace in downtown Tokyo was assessed as being worth more than the entire state of California. And people said, 'You know, the entire future of the world is going to be based in Tokyo, and this is where it's going to happen.' And people took out 100-year-long mortgages that their great-grandchildren would pay off for a 400-square-foot, sort of hole-in-the-wall-type condominium. And, of course, it then turned out it was not quite like that, and you had just a massive, extraordinary bust.

And then we saw the same thing in the mid-90s. There was sort of an emerging markets boom. And some of the busts have also been extraordinarily big. In the financial derivative context, the Long-Term Capital blow-up in 1998—the mathematicians who had priced these financial derivatives estimated that something like the scale of the move on these markets was something that would only happen once in a trillion times the history of the universe. And, of course, the conventional post-mortem on Long-Term Capital is that the mathematicians were just wrong and they hadn't done a very good job programming their computers, since, obviously, the math must be wrong. But I think there's an interesting question to be asked: What if the math was actually right, and, in fact, we just had an extraordinary bust?

and then you know in 98 when long-term capital blew up and russia went bankrupt where basically you had um you had a country with 10 000 nuclear bombs um everybody's bank account went to zero and in case you still have any money left in your bank account the currency also went to zero and normally uh normally emerging market banking crisis either the banks go bust or the currency goes bust in 98 in russia they did both so zero times zero definitely equals zero and And, of course, in that context, you bring to bear every kind of resource you can because, you know, a country where everyone's going to starve to death with 10,000 nuclear weapons doesn't seem like a good outcome for the singularity.

And then, all of a sudden, within 18 months, we had the hyper-boom in Silicon Valley where it was going to be the Internet. And I can certainly, having lived through the boom, the hyper-boom, and the last few months of insanity, I've often thought I could write a whole book on it, and it still seems like yesterday. There are aspects of it, of course, that seem like a strangely different time. And again, there was obviously something real about it. And the conventional post-mortem on the Internet boom was that in March 2000, the delusion and the insanity had reached its peak, and people were completely crazy. And, of course, I can tell you all sorts of anecdotes that one could see as confirming that thesis.

However, let's ask the opposite question. What if, actually, March 2000, in some sense, represented not a peak of insanity, but a peak of clarity? And that, at the peak of the boom, people could actually see the furthest. And what they could see was that, in the long run, in the next 20 years, 30 years, the entire old economy was going to be doomed. And that all sorts of businesses and ways of doing business were no longer going to work. And that basically, therefore, the tricky part was you had to bet on this one way that was going to be the way out, and that was the Internet. Now, it turned out that perhaps online companies selling pet food by buying Super Bowl advertisements offline were not really the technology that was going to be the decisive set of breakthroughs that was going to lead to the Singularity.

And so, again, we had this pattern of catastrophic approximations, but basically a picture that I think was very much correct. The bust in 2002–2003 was, again, quite extraordinary. There's a number of different ways to describe it financially. An interesting one is Japanese government bond yields got to 0.43% on a 10-year bond yield in 2003, which was the lowest bond yield in the history of the world. And so, again, we have multi-trillion-dollar markets that go to massive extremes one way or the other. And then, basically, in the last three, four years, we have a whole series of new booms. And you have sort of a choice: Which one do you believe? Which one is going to be the Singularity? Is it going to happen in China?

which is basically, at this point, looks like it's accelerating on every front, and it has probably the most extraordinary valuations of any emerging market in the history of the world. Is it going to happen through finance, where there's obviously been these enormous amounts of financial engineering? A lot of very smart computer people have tried to apply computer technology to Wall Street, to repricing things, doing financial engineering to fundamentally change the valuations of things, and so you have an enormous boom in the hedge funds, all these sorts of quantitative trading strategies. And are the computers actually going to figure out how to allocate all the money perfectly? And if you can get the computers to figure out how to allocate money better than humans, maybe the place to bet on the singularity is on finance itself, which is Wall Street.

Even something like the real estate bubble, which seems to have nothing whatsoever to do with the singularity, I think has been centered in these places that are somehow going to be at the center of the new future world: London and Manhattan, if you believe in the finance version; Shanghai, if you believe in the China version; and so on down the line. And of course we have, you know, we have a Web 2.0 boom, which again—and the question whether these things are booms or bubbles to some extent comes down to: is this how the singularity is going to happen? Now, the macro thesis I want to suggest is that even though it is likely that most of these massive moves are going to be fake, and there's something wrong with them—and Japan, in fact, was fake in the late '80s, and there was a lot that was not quite right with the Internet boom in the late '90s—

it can't be the case that every single one of them will be fake. One of them is going to be real, or the world is going to come to an end. And so I come back to this point that, as an investor, you probably have to try to figure out which ones are real and try to invest in those. You don't want to fade these things. And this has certainly not been a good environment for investors who are merely sane. I think this is sort of a recurrent motif of the last 20 years. All the classic funds that were trying to figure out what does a normally overvalued market look like, they all blew up in 1999–2000 because what looked really overvalued, it went much further. And the reason these bubbles go much further

is because people see no alternative, and they're implicitly betting on something like the singularity, whether they know it or not. There may be some very indirect ways to do this. I think one interesting one is that perhaps the smartest investor in the world, or thought to be one of the greatest investors of all time, is Warren Buffett. You can ask, how is he betting on the singularity, since obviously that's what you should do if you're a smart investor? And he's doing it, I think, in a very interesting way. He used to be always focused on value stocks like Dairy Queen, or selling things that were like regional companies in long-term, gradual decline that were still really cheap—so, nothing that seemed to have much to do with the singularity.

But the entire Buffett portfolio in the last 10, 15 years has been basically shifted towards selling insurance and catastrophic sort of reinsurance products. And so, if you think about it, we're going to insure against a nuclear bomb going off. And I think there are basically four different things that can happen. The first scenario—we're going to insure against some sort of catastrophe—scenario number one is nothing happens, you just get the premium. That's pretty good. Second scenario is you have a mild disaster like 9/11, but it actually helps the insurance companies because people expect the future premium to go up. And in fact, perversely, after September 11th, all the insurance stocks went up dramatically in value.

So that's scenario two. Scenario three is you have a... even bigger catastrophe, like, say, a single nuclear bomb going off, in which case, maybe the rules get changed, the government bails you out, something like that happens. And scenario number four, you have maybe 10,000 nuclear bombs go off, in which case you're in trouble, but there may not be anybody around to collect. And it's very, very striking that something like this, I think... It's hard to know whether this has been thought through precisely, but I would definitely not underestimate someone like Buffett, and I think he has probably thought this through at least to the level that I've described just now. One thing that's, of course, very tricky is we don't really know what the probabilities for these different things are.

So if you have a situation where in a successful singularity, a stock's worth $100 but there's only a 1 in 10 chance of it happening, then you might say, well, the fair market value is $10. But then the alternative is one where everybody's dead. So how much do you pay for the stock? Do you pay $10? Do you pay $20? I would say you pay anywhere. It's sort of a somewhat indeterminate amount, somewhere between probably about $10 and $100. You can make an argument you should pay $50 or something like that. And I think the inverse, of course, is that these insurance policies, maybe they're priced correctly, maybe they're priced incorrectly. We don't have any idea what the probabilities really are.

But we are living in this very strange world where this seems to be sort of a very big part of the fabric of what is going on. You know, it's very difficult to know where one necessarily goes from there, but I think the point that I would stress is probably the best things to try to focus on are ones that are sort of these incipient booms that people have not yet realized. My guess is we will see probably a whole series of booms and bubbles and busts for the next 10, 20 years as we're getting closer to this. And it's not at all clear that it'll be any of the ones we've seen. If you look at the historical ones, we had Japan, we had the Internet, we now have financial engineering on Wall Street, we have emerging markets.

And probably even the boom in the late 1960s, it was outer space. It was basically thought that the singularity was going to be driven by whoever controlled outer space. And maybe we'll have some combination of all of those. If I had a bet, I think it'll probably be something completely different that people are not expecting at all. And that makes it quite difficult to figure it out. But I would say that as a baseline, I think it's going to happen, and it'll be none of the above. Anyway, I do think that there is sort of this very big frame out there that there is something about the world we're living in. It is too easy to dismiss people who are sort of screaming in Hyde Park, predicting the end of the world and stuff like that.

Obviously, the people who've predicted the end of the world have been wrong for too long, and we cannot measure whether they're right or wrong, because they only have to be right once. There's sort of a cartoon that we've given to some of our investors. It basically shows somebody saying, 'At the end of the world, we predict disasters and catastrophes, but in the run-up to the end of the world, we predict extraordinary opportunities to make lots of money.' But, you know, it's not clear which of the two is going to happen, and I think it'll be a very exciting thing to try to sort through this in the next 20 years. It's going to be a very exciting time. Buffett may be doing it best, but there probably are some things that you could do that may be a little bit riskier but somewhat better than the Buffett selling catastrophic insurance strategy.

So, anyway, I guess I'll answer questions when we get to the panel. Thanks a lot.