Terra-Luna was unprecedented? The exact same thing played out ONE YEAR EARLIER when the algorithmic stablecoin created by Iron Finance collapsed over the course of a few days. The collapse of Terra-Luna was simply a repeat of Iron Finance on a larger scale — all of one year later.
“The longer you can look back, the farther you can look forward.”
— Winston Churchill
Shocking? Yes. Surprising? No.
The oversimplified explanation of the Great Financial Crisis is that Wall Street made a levered bet that something that had never happened would continue to never happen.
Prior to 2008, the US housing market had never suffered a synchronized, nationwide downturn: Historic data suggested that Phoenix, Arizona real estate was uncorrelated to, say, Miami, Florida real estate.
Plugging that data into their risk models, Wall Street’s financial wizards concluded that by pooling mortgages from all over the country, high-risk subprime loans could be magically turned into low-risk, AAA credits.
It wasn’t a totally crazy idea: This is, after all, how Michael Milken created the junk bond market in the 1970s.
Milken’s basic insight was that a diverse portfolio of high-yield junk bonds was actually a better credit risk than lower-yielding investment-grade bonds.
Some junk bonds would default, but the defaults were mostly uncorrelated — they wouldn’t happen all at once. The uncorrelated defaults smoothed out the risks, allowing the large majority of winners to more than pay for the small minority of losers.
That is how the junk bond market transformed from a fringe pursuit for only the most risk-tolerant investors into a standard asset allocation for even Vanguard types.
(And in the process radically lowered the cost of capital for entrepreneurs, to everyone’s benefit.)
That is not how things played out with the pooling of subprime mortgages, however, which turned out to be far more correlated than junk bonds: They were all losers in 2008/9.
Investment banks are rightly derided for getting so leveraged into subprime mortgages that the failure of their risk models caused the entire global economy to implode.
It was an expensive error for everyone. But maybe an understandable one? Their spreadsheets failed to foresee something that had never happened.
(Side note: The spreadsheets may have been right, even. You could argue that subprime only collapsed because Wall Street artificially created so much of it.)
What is much less understandable to me, however, is how professional investors could get blown up by the recurrence of something that happens every three or four years — because that’s what we’re currently witnessing in crypto.
We’ve Been Here Before
Bitcoin is now 70% off its highs. No financial modeling is required to recognize that this is not an unusual occurrence.
The current drawdown is smaller than the one we had just three years ago and smaller, too, than the one we had four years before that.
How, then, can it have caused so much trouble? How have the likes of Celsius, Three Arrows Capital (3AC) and likely others been pushed to the brink by such a common occurrence?
It appears at least partly to have been a failure to look back at recent history.
"The Terra-Luna situation caught us very much off guard,” Mr. Davies said, adding that the massive selloff was unprecedented.
Terra-Luna was unprecedented? The exact same thing played out ONE YEAR EARLIER when the algorithmic stablecoin created by Iron Finance collapsed over the course of a few days.
The collapse of Terra-Luna was simply a repeat of Iron Finance on a larger scale — all of one year later.
Which is not to say it was inevitable. Maybe Terra could have succeeded where Iron Finance failed.
But how you can be surprised that it didn’t is…well — I don’t get it!
I liken the collapse of Terra-Luna to the death of an elderly grandparent: It was shocking, but not surprising.
And I think you could say the same for the whole of the current sell-off in crypto — it was a hard-to-predict but nonetheless unsurprising event.
Unsurprising events should not pose existential threats to professional investors like 3AC and — even worse — retail-facing would-be banks such as Celsius.
The fact that they have is a baffling oversight — far worse than investment banks being duped by subprime in 2008, in my opinion.
That is, of course, easy for me to say as an outside observer (with a perfect record of predicting everything shortly after it’s happened).
But with the benefit of hindsight, I think it’s worth noting that crypto has proven more interconnected than we thought it was — transparent DeFi is looking disturbingly similar to opaque TradFi at the moment.
That is perhaps inevitable, as centralized actors such as 3AC and Celsius were always going to become large players in crypto.
But for crypto to succeed long term, it needs to become less correlated to itself and more resistant to centralized leverage.
Which is to say, more like the TradFi success of junk bonds and less like the TradFi fail of subprime mortgages.