Discover more from Net Interest
Plus: Upstart, Bad IPOs, Stablecoins
I actually remember where I was the day the dot-com bubble burst. It was March 2000 and I was sitting in a newly-opened branch of Starbucks near my home in London, reading a copy of the Financial Times. I was on gardening leave at the time, serving notice between jobs. Five years earlier, straight out of university, I had become a stock analyst and now the market for stock analysts was hot. There was a sense of euphoria in the air that we were at the dawn of a new economic age, funded by continually rising stock markets and a population that wanted to participate. Stock analysts sat at the centre of it all – the VCs of their day. We advised people what shares to buy and what to sell, and were rewarded handsomely for it.
I bit into my blueberry muffin and read the headline: “The NASDAQ composite index suffered its biggest loss in more than three weeks yesterday after a Japanese stock rout caused selling”. I skipped to the next story and wiped away some crumbs. No bell rang, but that was it, that was the top. The NASDAQ peaked on 10 March 2000 and went on to lose 78% of its value over the next four and a half years.
There are many differences between what is going on in markets today and what happened then. According to one analyst, the tech sector peaked at 34% of the market in 2000 yet contributed only 8% of nominal GDP; in the current cycle, it peaked at 27% and contributed 18% to nominal GDP. But there are also many similarities: the euphoria, the FOMO, the lofty valuations, the IPO frenzy.
“As with earlier technological revolutions,” wrote economists at the Bank for International Settlements in their review of the year 2000, “there was a pronounced increase in related capital accumulation… Associated with this, and reinforcing it, were sharp increases in equity prices which encouraged the abundant supply of low-cost venture capital. Moreover, with inflation generally low or falling, real interest rates stayed relatively low.”
Financial companies were not at the epicentre of the 2000 bust (that would come seven years later) but as the conduit for a lot of the financing that fuelled the boom, many were deeply impacted. In this week’s Net Interest, we take a look at three financial companies that lived through it: ETrade, Janus Capital and Silicon Valley Bank. Each of them profited from the boom in a different way, and each recovered differently, too.
History never repeats and 2022 is no 2000 rerun; this time, the financial sector has a 1970s analogy to live up to as well. But in trying to understand what happens next, it may be useful to re-examine these companies as case studies. We start with ETrade, the Robinhood of its day.
Net Interest is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.
ETrade: The Robinhood of its Day
Regular readers will be familiar with the ETrade story – it’s one we’ve discussed here before. Like Robinhood in the current cycle, it was the portal through which retail investors entered the market in the late 1990s and early 2000s. In 1992, the firm, founded ten years earlier as a provider of back-office services for discount brokerages, formed its own securities business and began to offer retail brokerage directly to investors. Rather than deploy a network of branch offices, it reached investors “by means of the CompuServe, America Online, direct modem access, touch-tone telephone and, to a lesser extent, interactive television.”
It wasn’t until the advent of the internet, though, that business really took off. In February 1996, ETrade launched internet stock trading, processing 1,300 trades in its first week; by May of that year, it was doing 11,000 trades a week.
ETrade’s online platform allowed customers to easily monitor the market and their portfolios, and its lower cost distribution enabled lower prices. It charged commission rates of ~$20 per trade. Although it was first, the barriers to setting up an online trading platform were not as high as establishing a branch network, and other players poured into the market. The number of online brokerages increased from around 12 in 1994 to over 120 in 1999. They competed on two dimensions: price and marketing.
On price, firms first started to segment the market based on customer activity but eventually dropped prices for everybody. Some, like Datek and Ameritrade, went after more active traders with lower prices of $5-10. One broker – Web Street Securities – even offered free trades, as long ago as 1999, to customers trading over 1,000 shares of NASDAQ listed companies.
A major turning point came in June 1999, when traditional broker Merrill Lynch gave into the channel conflict it had resisted until then and embraced an integrated online strategy. ETrade’s average commission rate fell from $21 per trade in 1996 to $13 in 2001.
On marketing, firms spent aggressively to solidify their position. In summer 1998, ETrade raised $400 million via a private placement with Softbank, giving it enormous firepower to invest in marketing. (Sound familiar?) That September, it launched a $100 million, 12 month advertising blitz – a budget more than three times larger than it spent on marketing the prior year. By the end of 1999, the industry was spending around $1 billion a year on advertising.
Alongside advertising there was a whole buzz associated with trading. Howard Stern conducted a popular radio show. “Datek Online. I love Datek Online,” he broadcast. “It’s the fast, easy and fun way to trade stocks online. Stuttering John [Melendez, one of Stern’s crew] uses it, and he’s making money. You get lots of free charts and you can check your account online.”
Yet the potency of marketing spend diminished; by 2001, it was costing ETrade $280 to acquire a customer, up from below $220 in 1998.
These deteriorating customer economics prompted a two-pronged strategic response. The first was consolidation, in an effort to extract the benefits of scale. ETrade bought many smaller online brokers including Web Street. There was constant speculation about a massive ETrade Ameritrade tie-up. Over time market share concentrated around fewer players (e.g. Ameritrade acquired Datek).
The second was expansion into other revenue sources. In 1998, nearly two-thirds of ETrade’s revenues came from trading commissions (around 3% came from payment for order flow) with most of the rest coming from interest income it earned on customers’ cash balances. But ETrade wanted to become more like a bank. It invested heavily in “value-added services in banking, lending, planning and advice.” In February 2002, it rebranded itself from ETrade Securities to ETrade Financial in order to convey its broader offering.
ETrade grew quickly. When it IPO’d in 1996 it had only 65,000 accounts. By the time the market peaked in March 2000, it had over 2.4 million. The market collapse in 2000 stymied its growth, and that of all the online brokers, but it took a while to sink in. Even after the market had peaked, ETrade would go on to win more than another 500,000 net new accounts over the following six months and almost a million by the end of 2001.
On a conference call in March 2001, Charles Schwab, the man, talking about the performance of Charles Schwab, the company, admitted, “We've come through a highly speculative technology bubble. Maybe I should have been more emphatic about understanding that this was a temporary phenomenon.”
ETrade implemented a number of restructuring initiatives to streamline its operations in the aftermath of the market downturn. Although brokerage revenues went into reverse as a result of lower trading activity and continued pressure on price per trade, overall revenues held up due to the diversification into banking. For a while anyway. When the financial crisis hit in 2007, ETrade found that it had inadvertently diversified into crappy mortgage assets and had to be bailed out by Citadel.
Through the 2010s, ETrade operated under the continued backdrop of falling commissions. In 2017, its average fee per trade dropped below $10. Finally, in September 2019, it acquiesced and went all the way to zero. A few months later, Morgan Stanley bought the firm for a headline price of $13 billion.
If ETrade represents any sort of precedent for Robinhood, much of the deterioration in operating performance may already be in the bag. Daily average trades fell 61% at ETrade from the peak in the first quarter of 2000 to the trough in the third quarter of 2001. At Robinhood, they are already down 64% from the peak. Perhaps Sam Bankman-Fried, founder of crypto exchange FTX, recognises this – he took a 7.6% stake in Robinhood this week. Not quite the bailout that Citadel gave ETrade – he didn’t inject fresh capital – but an endorsement nonetheless.
Robinhood’s CEO announced on his recent earnings call that he’s “challenged the team to dig deeper on cost discipline and get us to adjusted EBITDA profitability by the end of the year.” Let’s see how patient Bankman-Fried is with that. At ETrade, Citadel waited four years before agitating for change.
Janus: The ARK of its day
For those without the time or inclination to manage their own portfolios via ETrade, there was always Janus. Janus offered a suite of mutual funds and its growth focus made it especially popular. Based in Denver, the firm employed a research-driven approach, its analysts tasked with uncovering fast-growing companies before they were discovered by other investors. The firm made a big play of operating outside the confines of Wall Street. Its founder and CEO would initiate new recruits with a video, featuring himself on horseback, called “This Ain’t No Wall Street Joint”.
Fund performance was strong – the flagship Janus fund was up 39% in 1998 and 47% in 1999 – and money flowed in. By 2000, no mutual fund firm was taking in more assets. At one point, Janus was attracting 70 cents in every dollar flowing into the mutual funds industry. From $113 billion at the end of 1998, the firm grew its assets under management almost threefold to $325 billion by March 2000.
But the firm never changed the way it managed money to accommodate a bigger asset pile. Funds were highly concentrated and analysts overstretched. In 1999, the firm had just 19 stock analysts following 500 companies. In addition, a strategy of hiring analysts straight out of university meant that they lacked experience even as they were promoted. “We had a lot of folks running funds who shouldn’t have,” observed Gary Black, who came to the firm later to clean up the mess that ensued.
The market decline crushed Janus funds. Compounding their exposure to some of the worst performing stocks in the market were position sizes that had grown so large they were difficult to exit. In the two years from 2000, the flagship Janus fund fell by 64% and the more concentrated Janus Twenty Fund fell by 69%. Assets under management fell to $172 billion by September 2001.
In the aftermath, it also transpired that Janus had breached securities rules and was forced to settle charges with regulators with a $226 million penalty. By then, new management had been installed to shore up the business. They recruited a head of risk as well as more analysts – by 2005, the number of stocks covered at the firm had risen to 960. A new remuneration policy was also introduced. Fund investors would have had to have looked very carefully prior, but it turns out that portfolio managers got paid according to the volume of assets they managed rather than the performance they created. The new plan linked incentives to longer term investment performance. In addition, the firm became quicker to close funds to new money in the event they became too large relative to the investment opportunities available to them.
Janus eventually clawed its way back, while never hitting the scale of assets it managed at peak. Coming out of the bust, management conceded that while “a narrow, focused fund family served Janus well for many, many years, particularly coming into the ‘90s… the last couple of years has showed why we need offer investors more choices.” It launched new funds, ultimately offering customers 110 active investment strategies across fixed income as well as equity. In 2016, the firm merged with Henderson Group of the UK to form Janus Henderson Investors.
The parallels with ARK Invest are plain. ARK Invest was founded in 2014 by Cathie Wood and launched its flagship product, ARK Innovation ETF, the same year. For many years, the product slowly accumulated assets on the back of good underlying performance. Through to the end of 2019, it took in $1.6 billion of net inflows. But then the pandemic hit and the fund attracted marquee status. Over the next 28 months, through to April 2022, it took in a further $15.2 billion of net inflows.
The ARK Innovation ETF is now 75% off its high. Sadly, most of its inflows came just in time to capture the run of bad performance while missing out on the good. According to Morningstar Research, the average investor in the fund is down 33% on their money as at the end of April. Like Janus, the fund grew too large.
The manager of the Janus Twenty fund during the 2000 downturn had some advice for Cathie Wood going into this slump: “If I were Cathie… I would try to educate her investors to stay with it… There will be a period of time, maybe through her own doing or through some macro events in the marketplace, where the markets are going to sell off pretty dramatically and if I were encouraging people to invest in her fund, I would say make sure that you have got the mindset to – when that fund is down 30% or 40%, which it could be, that’s when you want to step back in and recommit to the fund.”
Unlike Janus, ARK Invest is not a public company. It is a private partnership, majority owned by Cathie Wood, so its financial performance is not publicly accessible. In 2016, ARK Invest granted an option to a minority shareholder, Resolute Investment Managers, to purchase a controlling stake but it bought back that option in December 2020. The option repurchase was financed via a facility provided by Eldridge Corporate Funding LLC (owned by Todd Boehly, soon to be owner of Chelsea Football Club). It’s clear that a lot of value has been sucked out of ARK Invest; whether the firm can rebuild as Janus did remains unclear.
Silicon Valley Bank: Banker to the VCs
Silicon Valley Bank provides another perspective on financing a tech boom. The bank was set up in 1983 specifically to service the burgeoning ecosystem taking root in the Valley. Its founders identified a gap in the market for financing tech startups – venture capitalists weren’t really around yet, and traditional banks were of the view that “new is not good, and growth is bad.”
Revised regulations eased the process for acquiring a bank licence, and Silicon Valley Bank became one of 72 new banks launched in California that year. It grew slowly, surviving a real estate wobble that led to a big write off in 1992, before confronting the tech boom and bust several years later.
Silicon Valley Bank offers tech companies a range of products: deposit services, loans, investment products, cash management, commercial finance and more. Because younger companies tend to have more cash on hand than debt, most of the bank’s money is traditionally made on the deposit side of the business. Going into the 2000 bust, Silicon Valley Bank carried $4.5 billion of client deposits, versus only $1.6 billion of loans.
The 2000 bust roiled Silicon Valley Bank in a number of ways. First, the bank suffered large deposit outflows as dotcom companies drained cash before their eventual demise. One of the features of that market was the dearth of new financing opportunities available to such companies. Management today argues that a key difference this time around is the amount of “dry powder” currently available – money raised by venture capital and private equity firms that has yet to be deployed. Back then, companies had to rely only on what they had on deposit – and it dwindled quickly. By the end of 2001, Silicon Valley Bank’s deposit volumes had fallen by a quarter to $3.4 billion. Its revenues shrank commensurately – net interest income fell by 20% in 2001.
The bank was also hit as the value of warrants it held in many client companies collapsed. Silicon Valley Bank has historically obtained rights to acquire stock, in the form of warrants, in certain clients as part of negotiated credit facilities. In 2001, income from the disposition of client warrants fell to a tenth of what it was in 2000, removing a substantial crutch that had supported earnings.
Loan losses also bubbled up, although these had less of an impact because of the relatively small size of the loan book. In 2000, the bank charged off 3.3% of its loans, which represents a 20+ year peak, exceeding what it wrote off in the global financial crisis. At the time, around a third of the book was early stage, so-called “investor dependent” loans – loans to higher-risk, cash-burning companies. Today, that proportion is around 2%.
However, since 2000, the bank has grown its exposure to a new type of loan: capital-call loans. These allow venture capital funds to draw down funding lines from Silicon Valley Bank rather than call on capital commitments from limited partners (LPs). The advantage for venture capital clients is faster access to capital, and higher ultimate IRRs to the extent the capital call is delayed. Risk is mitigated since LPs usually fund their commitments – in its history, Silicon Valley Bank has suffered only one loss on capital-call loans. Nevertheless, the market has attracted a lot of competition and is exposed to slow-down if VC activity stalls. These loans currently make up over a half of Silicon Valley Bank’s loan book.
The risks Silicon Valley Bank faces today are similar to those it faced in 2000: slowing deposit growth, deteriorating credit performance, volatile or negative gains on warrants and now muted capital call lending activity. But it does have a hedge: the bank is very sensitive to higher interest rates. Because its balance sheet is so skewed towards deposit-taking, Silicon Valley Bank profits more than most when rates rise. According to the bank, each 25 basis point hike in rates is worth between $100 and $130 million on net interest income. A few of those would soak up a complete collapse in warrant income.
In July 2021, Silicon Valley Bank’s CEO conceded that he’d never seen the business as strong, even dating back to the pre-dotcom period. It now appears that wasn’t sustainable. One of the lessons both ETrade and Janus took away from the downturn in 2000 was to diversify away from their narrow focus. That’s not advice Silicon Valley Bank has heeded, but it does have a different kind of hedge. Compared with Robinhood and ARK Invest, it looks best prepared for a tech downturn.
Join paid subscribers to read on about Upstart, Stablecoins and Bad IPOs. Simply sign up here: