Money for Nothing: Thomas Levenson on the South Sea Bubble

David Durlacher

Thomas Levenson, who has written a brilliant account of the crisis, sat down with financier David Durlacher to discuss how the events echo today’s markets.
Thomas Levenson
Home » Author interviews » Money for Nothing: Thomas Levenson on the South Sea Bubble

The South Sea Bubble in 1721 was the first stock market crash of its kind, and ushered into the modern era a new way of managing government debt.

 Thomas, what parallels should we draw between the South Sea Bubble and more recent bubbles of today, like the dot-com bubble in the ‘90s or even GameStop today?

What’s most striking about the South Sea Bubble is how familiar it is; the sequence of events in 1720—the decisions, assumptions, and then escalating pursuit of increasingly outsize gains, culminating in disaster–replicate over and over again. Details change, and the complexity of the financial manoeuvres certainly grows over time, but the basic outlines of the Bubble Year do reproduce, on large scales and small.  The 2008 crash that marked the public onset of the Great Recession maps very well onto what happened in London three hundred years ago today.

GameStop, though, is a bit of a different phenomenon. There is a fine, scoundrel-laced history of attempts to game specific markets. And certainly, individual asset booms and bubbles can sometimes have contagious effects. But short-squeezing is usually a phenomenon confined to one, generally small subset of the financial exchanges. The difference this time is in the demonstration that social media changes the pool of people who can get involved in such manipulation—and that does bring it a bit closer to the popular involvement that is characteristic of larger market crashes.

The government at the time saw the risks of speculation and passed the Bubble Act – deflating a bubble caused by itself in its wake. More recently, exchanges intervened in the Reddit frenzy, resulting in a crash fuelled and led by social media. Should governments and public bodies learn from history and leave markets well alone or should they become even more interventionist?

This question doesn’t really address the circumstances either in 1720 or 2021; it relies on assumptions not in evidence, and offers a false dichotomy as possible answers.

To begin. The British government in 1720 did not “cause” the Bubble. They contributed to its rise and fall, notably by acquiescing to the South Sea Company’s demand that it be allowed to price the debt-equity swap at will throughout the exchange, rather than following prior practice and offering shares at par, or at least at a set, publicly announced level. But the Company’s actions in negotiation and then in fostering the rise of the Bubble were much more proximate drivers of events that spring than anything the ministry did.

More particularly: the government in 1720 did not see the risks of speculation as the key driver behind the Bubble Act; don’t be fooled by the name, as “bubble” had meanings then that are different than the most common usage now. Most important: to “bubble” someone meant to defraud them.  The concern in the spring of 1720 was the sudden appearance of dozens of new companies seeking funds, often without much or any clear intention of doing anything with the money they might raise. When the Bubble Act was finally passed in early summer, it had essentially no effect on the market.  Its enforcement in August was triggered by the South Sea Company itself as a way to rid itself of competition for the public’s cash, and its effects—shutting down four small companies with no visible means of support—did not regulate the market for legally chartered companies.  Instead, it posed a question: were the surviving companies priced appropriately, given their underlying business case? Simply asking that always-sensible question proved the catalyst for the drastic market moves to come.

And look at what happened afterwards: stable UK interest rates (with contained bumps during high-expenditure wars) for two centuries or more. A stock market that functioned through and well into the industrial revolution. In fact, the valid criticism of the Bubble Act is not that it triggered a crash, but that it slowed the use of some valuable forms of company organization until a new approach came in 1826.

Similarly, the GameStop crash seems hardly primarily an event triggered by injudicious regulation. Shares in a loss-making company with declining revenue shot up ten-fold in two weeks in a publicly-visible short-squeeze. The subsequent price crash (followed by a doubling in two days last week as I write this) would seem to be overdetermined, and the regulatory intervention seems to have been relatively mild and short lived. So it’s hard to see how this bit of folly makes any kind of case for either the extent or type of regulation that is appropriate in modern markets.

But the real flaw in the question above are the assumptions a) that history shows that intervention/regulation as a category is bad (it doesn’t); and b) that the only choice is between bad regulation and an utterly unfettered market. History shows all kinds of outcomes from different choices made by various actors at particular times. Among them: the long-standing success of certain kinds of intervention, particularly those that address the risks inherent in unbounded leverage in the marketplace. It also shows a pattern in crashes: financial innovation, especially in ways that multiply leverage, correlates pretty well with trouble. And that suggests an approach for what I see as essential government intervention in financial markets. One thing that history does strongly suggest is that a market free of oversight tends to fail with depressing (if not precisely time-predictable) regularity.

If even highly intelligent, questioning people like Newton and Defoe can be taken in, what hope for the rest of us in making sense of when to invest and when to run a mile?

A minor correction: Daniel Defoe, no brilliant hand with money, (there was that awkward stint in debtor’s prison, for example) did not invest in the South Sea Bubble, as far as any records I’ve been able to find would show. He did comment on it, often, and with an interesting ambivalence, suggesting a largely social or moral concern that grew as the bubble expanded. But the nub of the point is that, yes, Isaac Newton along with many other usually very clever people got sucked into the excitement of the moment and lost all or part of their shirts.

And what that means is that the most important lesson of the bubble is that all of us are likely not smarter than Newton.  We are thus similarly no more able than he to rely on our intellect to master our emotions. So that means as much as possible we need to take ourselves out of the moment in investment decisions. Some very sophisticated investors can take advantage of “bigger fool” tactics to ride the rise up and bank profits as individual stocks or whole markets shoot up—Hoare’s Bank did exactly that in 1720. But for most of us, the most important thing to remember is that if something appears to be too good to be true, it is. In practice:  take a flyer with your lunch money, not your rent money. More formally: distinguish between the investing one does as a long-term play on underlying real economic activity, and the $200 you carry for a fun and cash-limited night at the casino.

What can we learn from the bubble about who people trust? In the days of the South Sea Bubble, government backing as well as that of the big names of their day was the lynchpin – now it is the power of the people’s voice through social media and the likes of Elon Musk. Has anything really changed?

I think less has changed in this area than one might think or wish. Looking back to the roaring 20s you see the effects of celebrity (in a different but recognizable form) on playing the market.  And certainly, one of the factors in the dot-com wave of speculation was the rise of financial-entertainment TV, marked most strongly by the rise of the CNBC cable channel. That seems to me to be a pretty close analogue to contemporary social media.

But the strongest parallel, it seems to me, is the repeated juxtaposition of people who were genuinely inventing useful new financial tools (not always with the full understanding of what they were doing) and those who recognized the opening to make a dishonest or unethical profit. In some ways the South Sea Bubble may have been the least affected by this.  While the Company and its allies certainly had its schemers happy to grow rich on insider trading, most of those involved thought, in the beginning, certainly, that they were simply doing well by doing good, and that the national interest would be served by their actions. (As it was, in fact, in a development that took a few decades to play out.)

But that was at least in part because the phenomenon of market failure was itself so new: the stock exchange in London was only a few decades old. Since then, clever, unscrupulous people have much more experience and accumulating techniques of bad behaviour to draw upon, and we see deceit as a much more common element amplifying some of these events. But, again, I’d emphasize that major financial innovations carry with them inevitable unknowns and asymmetries of information that are generally much larger factors in unanticipated market crises than pure fraud.

With the benefit of hindsight, bubbles and what fuels them are clearly seen. And yet we never seem to learn from history. Why not?

I think that this is a simple fact of human experience, one that applies across a lot of domains. We don’t have very long memories. The last deep freeze in Texas was just a decade ago, and yet the fatal power outages (and power market disruptions) of February 2021 came as a surprise to Texas’s officials (and very much so to the state’s population).  The 2008 crash was not even a decade old when some of the protections put in place in its wake were weakened by a new administration.  In 2001, young traders had never seen a bear market, or a major correction that lasted any length of time.  And in the 80s and 90s, with the Great Depression fading from active recollection, reversing some of the measures put in place to prevent that kind of crash seemed just fine.

I don’t know what the solution is, except to write as much history as I can, and encourage everyone to read widely about the past. We are not smarter than our forebears, and we can certainly repeat mistakes we should long since have recognized and avoided. We’ve done so in the recent past, and we will surely do so again—though if I could only be sure exactly when, I’d be rich.

If there was one conclusion you’d like readers to take from the events of the 1700s and the South Sea Company, what would it be?

The ideas and approaches of the scientific revolution–rigorous measurement, empiricism, and the application of mathematics to experience—shaped the early financial markets, both as tools that could be used to understand market phenomena, and as rhetoric to convince the society and culture of the day that abstract ideas of money made sense. The South Sea Bubble and its aftermath demonstrated both the power of such thinking–its real ability to improve human well-being–and its limits. We would do well to remember those same two possibilities, wealth and economic catastrophe, as we explore the financial world we now inhabit.

Drifting off topic slightly, do you think the response to the 2008 financial crisis, in ensuring financial institutions are no longer as highly leveraged, means we won’t see another such crash?

No. We will see more crashes. For one thing, the response to the 2008 crisis was modest and has already been weakened at least a little, so I’m not sure just how much protection it provides against the next storm. But the record of past financial crises reminds us that history doesn’t so much as repeat itself as that it knows the chords. Financial engineering is an ongoing project; new financial instruments and new ways to trade them (combined with new media, creating a novel information ecosystem) means that the next crash can and likely will develop in ways that are meaningfully different from the securitization issues behind the last one.

That’s where thinking through market structure and market regulation becomes so important. One characteristic of the most disastrous financial failures has been the phenomenon of contagion–of the spread of the pathology beyond the original confines of the stock market and individual players who go bust.  Focusing on reducing the impact financial market fractures have beyond the initial risk-takers—checking the financialization of society as a whole–is a major task, but one that would, I think be very useful.

Thomas Levenson is Professor of Science Writing at the Massachusetts Institute of Technology, and the author of Newton and the Counterfeiter, The Hunt for Vulcan, and most recently, Money for Nothing: The South Sea Bubble and the Invention of Modern Capitalism.

David Durlacher is the UK CEO of Julius Baer, the leading Swiss wealth management group.

Aspects of History Issue 8 is out now.