- “Changing the weighting of a stock in an index should not change the price, since nothing fundamental is occurring. However, this would be the biggest rebalancing in history, and nobody is quite sure how it could play out.” – Bob Pisani CNBC
- “Index funds tracking the S&P 500 have a small window today to by nearly $80 Bn worth of Tesla…” – A Posting on Seeking Alpha, at 4:37 am in the morning on Dec 18, 2020 [the last trading day before Tesla’s addition to the index]
- “That kind of volume is just sort of insane.” – Charles Schwab executive
The addition of Tesla to the S&P 500 Index triggered an extraordinary upheaval in the market.
First, and predictably, there was a huge surge in trading volume in Tesla’s shares, which also affected the shares of many other companies. Tesla’s enormous market capitalization magnified all the (typically minor) disruptive effects of an index change, as portfolio managers around the world reshuffled their holdings to make room for Tesla shares they now had to buy. The index managers at S&P knew generally what could happen, and took various measures to counteract the potential for turbulence – described below – but the impact was far greater than their initial forecast.
A market breakdown was avoided, but the aftereffects are worrisome. The momentum that was created by the index change event has continued to build. Tesla’s share price has inflated tremendously. The stock has traded above $800 since early January (up from $400 in November) – the company is now worth more than the annual economic output of, say, Sweden, or Switzerland.
The share price is now clearly disconnected from anything resembling “true value.” Its price/earnings ratio is over 1500, which is 40 times greater than the market average P/E (which is itself higher than any any point in the last decade).
It is a bubble. Regardless of how we assess the company’s business prospects.
The Bubble Is Not About Tesla As A Business
Asset bubbles are often driven by investors’ (over)excitement about a company’s business potential. But this particular bubble has very little to do with the Tesla’s fundamentals. The company is barely profitable. There were no game-changing fundamental announcements through the whole period of the surge. The company has an intriguing storyline, but the market has known about it already for quite some time.
This bubble is “technical,” triggered not by the company, but by S&P’s announcement.
But the standard “S&P effect” – the above-market returns associated with joining the index – has been extensively studied, and it is typically on the order of a 3-6% gain prior to the index change. Tesla is up over 100%. What else is going on?
price Is A Side-Effect – This Was A Volume Crisis
A bubble is usually seen as a pricing anomaly. But the Tesla bubble is primarily a volume anomaly. Yes, the shares are overpriced. Elon Musk told us that in May, when the share price was 85% lower than today. But in a bubble, price can lose its normal significance as a measure of business value. (That is why the comparison of Tesla to Switzerland’s GDP is not really meaningful. The “price” of Tesla’s stock is no longer being measured in the same units as the economic output of 8.5 million hardworking Swiss citizens.)
What happened here is that the S&P announcement created an opening for a particular trading pattern to develop which created a tidal wave in transaction volume, which tore the share price loose from the underlying fundamentals.
Where did the wave come from? The run-up period between the announcement of the change and its execution lasted 32 days, during which the tension in the stock accumulated like a coiled spring being compressed. The release occurred on December 18, during the last minutes of the the last trading day before the index change was to take effect. It was enormous.
- On December 18, Tesla traded 222 million shares at an average price of $665 a share. This works out to almost $150 Bn – by far the largest dollar amount for a single company, in a single day, ever. It is, for example, almost 4 times the dollar volume of Apple’s heaviest day.
- Almost 30% of Tesla’s “float” changed hands – far more that what any other large tech company has ever experienced. This is a sign that supply and demand were severely out of balance.
- The volume surge was compressed into a very short time window:
- “About 1.7 billion shares worth more than $150 billion traded within seconds in Nasdaq’s closing auction [the day’s last official trade], the highest dollar amount ever.”
- Tesla’s volume surge was sustained through December and January. Tesla traded much higher than Apple, and higher even than the S&P 500 ETF (SPY) – which is normally the most actively traded security in the market.
All in all, the volume wave was (and continues to be) extraordinary, record-breaking.
This Was Not A Surprise
Usually, volume surges are caused by news – especially news that catches the market by surprise.
That was not the case here. The most extraordinary aspect of this Tesla event is that the market knew full well what was going to happen, when it was going happen, and how big it would be. In fact, the event was essentially designed and engineered, in plain sight of all, by S&P itself.
The challenge for an index manager is to manage the volume of trading created by index changes. Volume can be dangerous. High trading volumes in general correlate with higher volatility. Imbalanced trading volume that can create stress in the market, especially if it is compressed in time. The price bubble is merely the side-effect. Index managers don’t try to control prices, but they construct explicit strategies to manage the volume.
S&P’s Challenge: To Manage a Massive “Rebalancing”
The volume crisis in this case arose because Tesla’s huge 1.7% weighting for a new component (the largest ever) forced index-tracking and index-benchmarking funds with $10-12 trillion under management to rebalance their portfolios, and to do it in a very short window.
- “S&P estimates that approximately 129.9 million shares of Tesla will need to be purchased [today]… At the current market price of $655, indexers need to buy $85.2 billion in Tesla stock at the close on Friday. However, there are billions more that will need to be bought by “closet indexers” that do not officially pay S&P, but nonetheless track the index. No one knows how much these “closet indexers” will buy, but it could be 50%-100% above the “official” $85.2 billion estimate.”
In the process of rebalancing, these buyers are indifferent to price. The index funds are compelled to buy the shares, to adjust their portfolios in accordance with the new index weights. price becomes irrelevant. Value is irrelevant. The only thing that matters is alignment with the index.
Tesla’s size was the root of the problem, of course. It became “the biggest rebalancing in history” and it led to the bubble conditions we now see. But it was exacerbated by the long delay of more than a month between the announcement (Nov 16) and effective date (Dec 21), discussed in my previous column. The rationale for this delay was to give the market participants time to prepare and adjust. S&P’s lead analyst explained:
- “We’ve given a month’s notice, so everyone has had time to plan out their strategies.”
What it gave time for, as it turned out, was the opportunity for hedge funds and other active investment funds to front-run the index funds, virtually without risk. The hedge funds could buy ahead of the effective date because they knew that by December 18 the index-tracking funds would have to buy from them, regardless of the cost. This set in motion a self-reinforcing cycle of spiraling upward prices, and a build-up of the disequilibrium conditions that finally exploded on Dec 18.
S&P knew all this could happen, and evidently considered several options for keeping things under control. They had floated the idea of introducing Tesla into the S&P 500 in two separate steps, perhaps a week apart (or longer). Important customers were polled, and apparently supported it.
- “Traders and investors had said that adding Tesla on two separate dates rather than at once would have made for a smoother transition. Some even advocated for adding Tesla in two separate tranches, one in December and one in March. “There will be less aggregate market impact if it’s broken down into tranches by S&P,” said Patrick Nichols, a partner at trading firm Old Mission Holdings. That problem is compounded by the fact that index funds typically buy and sell shares near the close of the trading session, ensuring they get the last price of the day and to avoid deviations from their indexes, added Greg Sutton, head of portfolio trading at Citadel Securities LLC.”
Nevertheless, S&P eventually decided to go ahead with an all-at-once index change. The multi-tranche approach had never been tried before, and the index managers may have concluded that it was not the moment to introduce an untested innovation.
Still, by Nov 30 (when the decision to add Tesla in one shot was announced), a disturbing trend was clear. The price of Tesla’s shares had already run up almost 40% in just two weeks following the Nov 16 announcement. As Tesla’s market cap grew, its prospective index weight grew. Which increased the projected size and cost of the rebalancing. S&P’s own estimate of the rebalancing volume had grown by 66% in two weeks, and would keep growing as as more investors piled into the front-running trade and Tesla’s price kept rising. S&P’s estimate of the forced demand for Tesla’s shares reached $85 Bn in early December. The Wall Street Journal projected higher numbers. In the event, all the forecasts were significant under-estimates.
Cushioning the Impact? The “Quad Witch”
It was clear by early December that things were not quite under control. The trading window of 35 days was too long. The feedback loop was relentless.
Fortunately, S&P had a contingency plan built into their schedule. Friday Dec 18 was the last trading day before the change took effect. That date had been deliberately set to coincide with three other major market events which all drive high trading volumes – all also set to occur on Dec 18. These were:
- S&P’s own quarterly rebalancing of all three of its “headline equity indices” – the S&P 500 (“large cap”), the S&P 400 (“mid cap”) and the S&P 600 (“small cap”). In other words, all the stocks in all three of S&P’s flagship indexes – comprising over 90% of the total U.S. stock market value – would be rebalanced at the same time, to reflect changes in their float-weighted share counts, as well as additions and deletions. This programmed rebalancing regularly drives significantly above-average trading volume.
- The Nasdaq 100 – the world’s most important tech-company index – was also scheduled to perform its once-a-year rebalancing on Dec 18 (Tesla was already the 6th heaviest-weighted component). This would add volume to the mix.
- Strangest of all – Dec 18 was the “quadruple witching” day – the third Friday of each quarter when all stock options, index options, and futures expire at the same time. This always makes for one of the busiest trading days of the year. On this particular witching day, in fact, trading volume was 53% higher than normal.
Of these three factors, the Quad Witch probably contributed the largest volume boost – 4 or 5 billion shares in the broader market, as option traders had to close out their positions, buy or sell, amounting to perhaps several hundred billion dollars.
Interestingly, the average share price across the market traded on the Quad Witch was also about 25% higher than normal, suggesting that the trading was skewed towards large caps – exactly what one would expect in the Tesla addition scenario.
Some observers found it puzzling that the Tesla change would be scheduled right on top of this cluster of volume amplifiers. It seems like throwing a turbulent stock into an even more turbulent market. But S&P’s Nov 16 press release was clear.
- “Adding the stock at the upcoming December quarterly rebalancing coincides with the expiration of stock options, stock futures, stock-index options, and stock-index futures, which may help facilitate the funding trade.”
A surge of unusual trading volume – like the Tesla event – would have a bigger impact in a thin market. S&P purposely scheduled the Tesla index-buying to occur on one of the heaviest trading days of the year, so that the Tesla increment could be more easily absorbed. Toss a big stone, and the splash is higher in a calm pond than in a large windy lake.
- “Volume is usually heavy on those days and would help boost liquidity on the day of Tesla’s inclusion, several investors said. “We’re expecting to see a day where volume is much higher,” said Matthew Bartolini, head of SPDR Americas Research at State Street.”
It was a smart move, and it seems to have worked. As shown, the Tesla trade generated extraordinarily high volume – the highest ever for a single company in a single trading day, and at least 2 to 3 times larger than it seems S&P had expected. Yet the index change was carried out without causing an obvious market break.
What did change was the psychology of the market. The volume surge smashed the normal price/value relationship, and set the stage for the bubble that ensued. In the next installment, we’ll consider this aspect of the event more closely.