Why is there a 100% spread between SPY and stocks?
SPY is the State Street S&P Depository Receipt, called a “spider” for the acronym SPDR, tracking S&P 500 stocks. It’s an exchange-traded fund (ETF).
The term depository receipt reflects the roots of ETFs at State Street, where commodity traders invented the instrument. They said, “What if we could trade the deposit receipts on what’s in the commodity warehouse rather than grabbing the wheelbarrow and loading up a bunch of wheat or silver, or stocks?”
Photo 187656017 | Etfs © Jannygto | Dreamstime.com
And ETFs were born. Substitutes for real assets. The first one, SPY, commenced trading in January 1993, two years before I joined NIRI, the investor-relations professional association.
Back to the spread. SPY is averaging 1% volatility, the intraday spread between highest and lowest prices.
Compare to NVDA. It moves 3.2% from highest to lowest intraday price daily. At a 20-day average price of $455, that means the price varies by nearly $14 every day. For a small retail trader owning a hundred shares, it’s fourteen-hundred-dollar daily swing.
We haven’t gotten to the rest of the spread yet. The average S&P 500 stock has intraday volatility of 2%. That’s 100% more than SPY, which ostensibly tracks S&P 500 stocks.
There’s your 100% spread.
What would you do if you had servers parked at stock exchanges running high-performance trading systems feeding on realtime volatility data?
You would trade the spread.
And it would become the purpose of your investment thesis. You would buy or short a basket of securities on volatility. That’s statistical arbitrage.
All ETFs depend for prices on statistical arbitrage – variance versus an underlying basket.
SPY is routinely the most actively traded stock in the market, averaging nearly $35 billion of dollar-volume daily. Picture two continuously intersecting lines on a graph, one vacillating twice as much as the other.
As a trader, you’d shift from long to short in them whenever they intersected, and you’d reverse the trade once half the volatility elapsed.
Not coincidentally, index funds want tracking errors of no more than 2%. Vary that much versus the benchmark you’re tracking, and your distributors stop selling your funds and the regulators come knocking with fines for failing to deliver to investors what you promise in your prospectus.
Thes are observable, mathematical facts. And consider: Passive money and closet-indexing (buying what indexes buy) is now 70% of institutional equity assets.
If the market moves more than 1%, there’s trouble. Why? Because the arbitragers keeping SPY and the underlying basket of stocks tracking each other can no longer calculate reliable profit probabilities.
If those firms quit trading, the market tanks. Like yesterday.
It’s part of the reason why the market can turn sharply more turbulent once the move is more than 1%. And why the market can go for long periods without a 1% move when the machines have got it dialed in.
Roughly 96% of the trading volume in SPY is a form of arbitrage (heck, I do it on occasion with leveraged versions like SPXU and SPXL when the Supply/Demand balance gives me an actionable probability).
And arbitragers use an evolving basket as the other side of the volatility trade. It shows up in the data. About 15% of the S&P 500 is down 20% or more this year. Half the index’s components are negative year-to-date.
Yet the broad index itself, comprised of futures contracts, is up 13%. How? TSLA, NVDA and META are up more than 100%.
Further, trading machines – Citadel, Hudson River Trading, Susquehanna, Jane Street, Two Sigma, Infinium, Quantlab, GTS, Tower Research, Optiver and other names you never hear – sift the data for baskets that drive a predictable arbitrage trade.
Sure, they trade everything and drive 50% of market volume. But what produces gains is a predictable variance model.
These firms are making billions of dollars. Your average stock-picking institutional investor isn’t matching the benchmark.
There’s a disconnect between SPY and underlying stocks, and a disconnect between what most think drives the stock market and what’s observable in the volume data.
What to do, investors and public companies? For one, stop wasting time and money pursuing a thesis that doesn’t match the market’s mechanics. Try to get in step with it to the degree you can.
Public companies, earnings season looms. Don’t do what you’ve always done with earnings. Shorten your release. Keyword it. Emphasize your characteristics not your quarterly metrics.
And investors, learn the ebb and flow of money. You can’t do much about arbitrage. But you can see probabilities for getting on the right side of the wave that continuously waxes and wanes.
