The biggest financial meltdown of the past quarter century, prior to–and ex–the much bigger US housing market fiasco of 2007-09, was the popping of the internet bubble of 1999.
As I’m mentioning Henry Blodget below and thinking about Sam Bankman-Fried as another example of crookedness and excess, I’m realizing more clearly the strong parallels between then and now. The year 2000 ushered in not only recession but a two year rally in ultra-safe, bond-like sectors like Utilities and Staples (pretty much the only time in the past 30 years that such defensive sectors have been in the limelight), and the beginnings of the housing boom that would ultimately end up in epic levels of fraud. Maybe this is why today’s brokerage house strategists have been so strongly convinced, even now, that taking a defensive posture in anticipation of recession is the right thing to do.
the internet bubble of 1999
1998
Alan Greenspan, then head of the Fed, had a penchant for running money policy a bit on the loose side, even though the stated intent of the monetary authority continued to be disinflationary. i.e., relatively restrictive. Two crises in 1997-988 reinforced this tendency to err on the side of too much stimulus:
–in 1997, international commercial banks mounted a sequence of attacks on the currencies of a number of emerging markets in Asia, Thailand being the first of a number of successes, to exploit their excessive dependence on dollar-denominated debt. This culminated in a massive, but unsuccessful, attempt in 1998 to break the peg between the HK$ and the US$. Sti8ll, the bank move destabilized commerce in Asia for more than a year
–Long Term Capital Management, founded in 1994, was a hedge fund. It focused exclusively on exploiting minute price differences between the US Treasury bond issues that were used in the pricing of bond derivatives, that is, “on the run” vs. Treasuries that were virtually identical except not used in derivative pricing, i.e., “off the run.” Although both kinds pay identical amounts in interest and return of principal, the former are more liquid and trade at higher prices in the aftermarket than the latter.
LTCM’s idea was to do what commercial banks routinely do, short on-the-run bonds and use the proceeds to buy off-the-run, earning small amounts on the spread between the two. Its twist: do so on a massive scale, using boatloads of short-term (i.e., lower interest rate) loans to scale up their positions. The operation was watched over by a celebrity bond manager, John Meriwether, with rockstar finance academics Myron Scholes and Robert Merton (the functional equivalent of celebrity influencers) on the board. What could go wrong?
Russia did. In 1998 the ruble collapsed and Russia defaulted on its sovereign debt, causing a flight to safety around the world. This created immense demand for on-the-run Treasuries. “Poof!,” unable to sell its illiquid holdings to cover its short positions, LTCM was bankrupt. But, as the old joke goes, borrow $1000 you can’t repay and you’re in trouble; borrow a billion dollars and the bank’s in trouble. Here, the Treasury was the one in trouble, given the multi-billion dollar overhang of illiquid Treasuries on the LTCM books and where buying interest had completely dried up. Washington ultimately arranged a bailout that played out over a couple of years.
1999
–Y2K. A major worry here, which may seem silly now but which was a genuine worry back then, was whether computers of all types, from bank and government mainframes to PCs, would stop working on 1/1/2000. As I remember it, the issue was that most newer commercial programs were built on top of old code (think: banks) that, to save space (?), had internal calendars that only went up to 12/31/1999. So, arguably, they might all just shut down when 2000 dawned. That would mean no more working ATMs–or any personal or business access to money; no traffic lights; no air traffic control…
In the runup to 1/1/2000, lots of weird stuff happened by people preparing for the ensuing end to civilization. There was a speculative run on wooden plows, for example. Older US dollar coins, made from actual silver, sold for close to 10x face value.
Most important for this post, Y2K was a another reason for the Fed to keep money policy relatively accommodative
–internet infrastructure. There was a mad rush during 1999 to build fiber optic cable networks, including undersea connections between the US/EU and US/Asia
–cellphones and cellphone networks. Demand for cellphones and PDAs (personal digital assistants, like Blackberry) was very high, and expanding. This created shortages of semiconductors to power devices and the networks they used. FPGAs (field programmable gate arrays), big, expensive kludgy things, were particularly hot. They had nothing much going for them, other than you could erect a network with them today and install/correct network programming on the fly later on
–internet and internet commerce. AOL, the early internet leader, was being surpassed by browsers like Netscape that allowed direct access to the World Wide Web. This created potentially enormous demand for ecommerce websites. I remember going to the IPO roadshow for Amazon, for example. Management said virtually nothing about the company’s business. The entire pitch was: “look back to 1980, the dawn of the PC age. There were obscure little companies back then like Microsoft, Cisco and Oracle, trading for half-nothing that became tech giants. We’re at a similar juncture today. Who knows who the future winners will be, but buy a basket of potential beneficiaries. Include us in it.” To give this some context, MSFT was then trading at about 1000x its 1986 IPO price.
–dubious flame-fanning. The sell-side royalty of internet company analysis were Henry Blodget of Merrill and Mary Meeker of Morgan Stanley. Both wrote glowing reports about fledgling internet company IPOs that proved wildly optimistic. In his private mails, Blodget trashed some of them as total garbage. He settled subsequent SEC charges of securities fraud by agreeing to pay $4 million and accepting a lifetime ban from the securities business. Meeker left Wall Street to join venture capital firm Kleiner Perkins.
why the music stopped
Valuation was one important thing. I remember, for example, selling all my MSFT in late 1999. The company was trading at 65x earnings, despite growth slowing to about 5% and with Steve Ballmer taking over. (I bought it back 14 years later at 2/3 the price, just after Ballmer was ousted).
The global computer network didn’t collapse (I have no idea what happened to all those plows)
As with the 19th-century railroads, the mad rush to be the first with capacity resulted in much too much internet transmission infrastructure being created. Probably as important, rapid advancements in dense wave division multiplexing (sending hundreds of signals down a fiber optic strand rather than one) quickly made much of the cable laid in the late 1990s superfluous, and therefore worthless.
New chip fabs came on line, ending the shortage that had sent cellphone component prices sky-high
In other words, across the board, potential shortage turned into actual glut.
At the same time, Time Warner was agreeing to buy soon-to-be-senescent AOL for a ton of money, a bell-ringing warning of a market top.