Tagged: GME

Whacked and Lulled

D-whacked. 

That’s how some described the stock market the past few days.  Since Donald Trump has been cancelled from popular culture, you’d be excused for missing the debut of a SPAC associated with the former President called Digital World Acquisition Corp (DWAC).

You get it now, right.  D-whack.

Illustration 93383364 / Volatility © Iqoncept | Dreamstime.com

The Twittersphere was a-flurry, amusing given the ex-President’s erstwhile penchant for that platform.

The new SPAC finished Oct 20 at $9.96, and closed the next day at $94.20, an 850% increase.  It was volatility-halted a half-dozen times Thursday, five minutes at a time.  I was on Benzinga shows several times in recent days discussing it.

We’re going to talk about those halts. Hang on for a bit.

On Friday, DWAC went berserk anew amid a rash of volatility halts, and I saw a price north of $187 at one point – a gain of almost 1,800% in two days.  It’s now near $60.

Oh, and get this:  Including warrants, there are about 42 million shares outstanding.  Last Thursday alone, DWAC traded 498 million shares, twelve times what exists.  Through yesterday it’s traded 730 million shares.

It couldn’t be a short squeeze like meme stocks GME and AMC were said to experience.  There were no shares to short. I won’t relitigate that story here, save to say that short squeezes are difficult to effect because market-makers can create stock.

And that’s what happened. If there are orders to buy 100 million shares of DWAC, market-makers will create stock to fill those between the best bid to buy or offer to sell.

Traders and public companies, you best grasp this flaw.  There is theoretically no limit on the stock that brokers executing trades might create. Stock is currency. What happens when you create more currency to chase the same goods – here, prices?

Prices inflate.

The ramifications are breathtaking.  This colossal risk exists because the SEC wants a “continuous auction market” where everything is always for sale, 100 shares at a time.

It’s nobody else’s fault. It’s a choice.

Now, let’s get back to volatility halts. We’ve had hundreds the past week, most in tiny stocks. DWAC was volatility halted almost 20 times in total.

Let’s understand volatility halts. After the May 6, 2010 so-called Flash Crash, the SEC and Finra and the exchanges implemented rules to halt the broad market and individual stocks, and they’ve been updated some since.

The broad market – as we saw repeatedly in March 2020 – hits a Level 1 halt for 15 minutes if the S&P 500 drops 7% below its previous closing price.  That halt doesn’t apply after 3:25p ET, though.  Level 2 is down 13%, and we sit and wait 15 minutes again. That actually happened. Remember?

If we smash into Level 3, down 20%, the market closes for the day.

In single stocks, there are volatility halts called Limit Up/Limit Down (LULD). Nearly 10,000 triggered in the spring of 2020. LULDs stop trading in a stock when price moves outside calculated bands from the SIP, the Securities Information Processor.

That’s the consolidated tape, how we know volume and prices across a byzantine network of nodes where stocks trade.  Trading pauses for five minutes if the bid or offer moves outside the bands.

If those moves last under 15 seconds, the halt dissolves.  If not, trading stops for five minutes, and the primary listing market then re-launches trading (the CBOE says other markets can trade a stock if the listing market can’t reopen it but I can’t confirm that).

DWAC kept rising because bids and offers melted back inside the bands. Well, what’s the point of pausing then?  Why limit a stock up or down if no limit actually happens?

Traders were screaming that the price would move dramatically at resumption too.  Of course. Price bands kept revising.  And for stocks like PHUN, which also traded wildly, the bands are massive – double what they are for stocks with prices higher than $3.

These halts seem pointless. They don’t serve issuers or investors, only the parties responsible for maintaining a continuous auction. That then becomes the purpose of the market – rather than a fair place for investors and public companies to find each other.

What happened in March 2020 will happen again. We see it in the terrible challenge stocks face when they decline rather than rise.  The market shouldn’t be d-whacked.

But it is. So get ready.

Mission SEC

Gamestop wasn’t our fault. 

That’s the conclusion of the Securities Exchange Commission in a 45-page report.  I’m reminded of Gulliver.

Not tiny people with tiny ropes tying down the giant SEC while it slept.  Satire. That the SEC exonerated itself comes as no surprise. Ask the Department of Flooding if it had anything to do with the failure of anti-flooding systems and the answer will be: Nope.

I’ve got just one beef with the SEC in this report, and read on, and you’ll see.  I commend the SEC for saying things like: To understand what transpired in January 2021, it is necessary to understand the market structure within which the events occurred.

We use the phrase “market structure” all the time.  Here, the SEC did.  Coincidence?

Illustration 94594512 / Markets © Idey | Dreamstime.com

Who is the SEC’s audience?  The public, yes. But I’m a market-structure expert. That memo is for market participants, not neophytes. Those who depend on the public equity markets. Such as public companies – who should possess the capacity to understand it (like alert reader Jay D. in New York City did!).

You do it right, public companies.  File your regulatory filings. Tell your superlative story. Meet financial targets, issue your ESG reports, hold your DEI teach-ins, put “sustainable” in everything.

Then your stock soars 500% and you don’t know why. Sure, up 500% is awesome but not knowing why is a condition that should never exist in a free, fair and open market.

Or worse, your stock plunges from $40 to $10 in a day, at the open, even though not a single stockholder sold.  That happened to AVIR yesterday. Sure, bad news on the Covid-19 front. And I can’t say no holder sold.

But isn’t the market supposed to prevent dramatic, reactionary moves in prices? We have a systemwide network of volatility girders instituted by the SEC to prevent midday panic.

Did you know your stock can trade when no holder buys or sells?  We’ll come to that at the end.  It’s how the $50 trillion construct of the US stock market clings together. And why Gamestop happened.

Let’s get to what the SEC said is necessary to understand. I’ll trim it because the memo is 45 pages long – terse for a government missive but I need 800 words for this column.

The SEC says retail trades proceed from a retail broker to somebody else for execution, report to the consolidated tape, and pass on for clearing, which can take up to three days.  Meanwhile in the options market there are a million securities, vastly more than total stocks, where the prices come from brokers and trades settle in two days or less.

How can derivatives change owners before the underlying asset?  Good question.  Unanswered.

And on page 11 of the memo is a painfully discursive paragraph on what traders may do with retail orders.  I’m condensing in Cliff’s Notes fashion: Market-makers may choose to trade with money that’s less sophisticated and avoid what’s more sophisticated, to reduce the risk of loss. And apparently it’s possible to easily segment these orders.

Got that?  I’ll dumb it down one step more, and this is my beef: The SEC is fully aware that some have a huge advantage over others.  Yet the SEC says to begin Section 2.2 that its mission is to “protect investors, maintain fair, orderly, and efficient markets, and facilitate capital formation.”

Well, how can that be true if some know what others don’t? People have gone to jail for years because they bought and traded on what others don’t know. Here it appears to be happening willy-nilly.

Because the SEC needs an unfair market. The one we have won’t work if it’s fair. There is no other conclusion.

What about Payment for Order Flow, where retail brokers like Robinhood and Schwab sell trades to “wholesale” buyers like Citadel Securities and Two Sigma?

Wholesalers buy things only because they can sell what they buy at a profit.  Period.

The SEC knows it. The SEC knows some have better information than others.  And the SEC knows it exempts firms like Citadel from the short-locate rules under Reg SHO Rule 203(b)(2).

Because the mission of the SEC is to preserve the market. Which can’t hum continuously. Unless it’s gamed. So the SEC pays whomever it must with favors to see that it hums.

Sounds like politics.

As to how Gamestop happened and why no buyers or sellers are needed, see my Meme Stocks presentation.

Traders, you can’t beat the market if you don’t know how it works.  Public companies, we’re wasting time and money doing things that don’t matter.

The starting point is making sure our CEOs and boards know how the market works. We can help you begin this new mission.

Where’s It Going?

Where’s what going?

Time? Hm.

Money?  Well. Yes.

It abounds and yet it doesn’t go far.  Why that’s the case is another story (I can explain if you like but it usually clears a room at a cocktail party).

First, if you were spammed last week with the MSM, apologies! It was inadvertently set on full-auto.  And one other note, our sister company Market Structure EDGE  is up for several Benzinga Fintech Awards.  As in American politics, you may vote early and often (just kidding!). No, you can vote daily though till about Oct 22, 2021.  We hope you’ll help! Click here, and turn it into a daily calendar reminder.

Today we’re asking where the money gushing at US stocks and bonds like a ruptured fire hydrant is going. Morningstar says it’s $800 billion into US securities the last twelve months through July.

That’s minus a $300 billion drop in actively managed equity assets. Stock-pickers are getting pounded like a beach in a hurricane. Public companies, you realize it?

That’s not the point of this piece. But investor-relations professionals, realize the money you talk to isn’t buying. It’s selling.  There are exceptions and you should know them.  But don’t build your IR program around “targeting more investors.” Build it on the inflows (your characteristics), not the outflows.  If you want to know more, ask us.

So where did the $800 billion go? 

About $300 billion went to taxable bond funds.  Not for income. Appreciation. Bonds keep going up (yields down, prices up). They’re behaving like equities – buy appreciation, not income.

The rest, about $500 billion, went to US equities.  We’re going to look at that. 

$500 billion seems like a lot.  Ross Perot thought a billion here, billion there, pretty soon you’re talking real money. For you who are too young to know it, Google that.

But today $500 billion ain’t what it was. And frankly, five hundred billion deutschemarks wasn’t much in the Weimar Republic either.  The problem wasn’t inflation. The problem was what causes inflation: too much money.

Ah, but Weimar didn’t have derivatives. Silly fools.

For perspective, more than $500 billion of Exchange Traded Funds (ETF) are created and redeemed in US equities every month.  Stocks trade more than $500 billion daily in the US stock market.

And the money supply as measured by the Federal Reserve’s “M2” metric reflecting the total volume of money held by the public, increased by $5 trillion from Feb 2020 to July 2021.  That’s a 32% increase. About like stocks (SPY up 33% TTM).

Wait. The stock market is up the same as the money supply? 

Yup.

Did everybody sell stocks at higher prices?

No. Everybody bought stocks at higher prices.

Okay, so where did the stock come from to buy, if nobody sold?

Maybe enough holders sold stocks to people paying 33% more to account for the difference. Good luck with that math. You can root it out if you want.

But it’s not necessary.  We already know the answer. The money went into derivatives. 

The word “derivative” sounds fancy and opaque and mysterious. It’s not.  It’s a substitute for an asset.  You can buy a Renoir painting. You can buy a Renoir print for a lot less. You can buy a stock. You can buy an option on that stock for a lot less.

Suppose you want to buy the stocks in the S&P 500 but you don’t want the trouble and expense of buying 500 of them (a Renoir). You can buy a swap (a print, No. 347 of 3,900), pay a bank to give you the returns on the index (minus the fee).

Or you can buy SPY, the S&P 500 ETF.  You think you’re getting a Renoir.  All those stocks. No, you’re getting a print somebody ran on an inkjet printer.  It looks the same but it’s not, and it’s not worth the stocks beneath it.

Image courtesy ModernIR, Aug 25, 2021.

See this image?  There is demand.  There is supply. The former greatly exceeds the latter like we’ve seen the last year during a Covid Pandemic (chew on that one for a bit), so excess demand shunts off to a SUBSTITUTE. Derivatives. ETFs, options, futures.

That’s what’s going on. That’s where the money went. Look at GME and AMC yesterday. Explosive gains on no news. Why? Banks squared derivatives books yesterday after the August expirations period. Demand for prints (options), not paintings (stock), vastly exceeded supply.

So banks bought the underlying paintings called GME and AMC – and sold traders ten times as many prints. Options. Derivatives. It’s implied demand. The stocks shot up.

Bad? Well, not good. The point isn’t doom. The point is understanding where the money is going. Every trader, investor, public company, should understand it. 

It’s all measurable if you stop thinking about the market like it’s 1995. It’s just math. About 18% of the market is in derivatives.  But about 75% of prices are transient things with no substance.  Prints, not paintings.

Public companies, know what part of your market is Renoir, what part is just a print.  Traders, do the same.

We have that data.  Everybody should always know where the money is going.

The Missing Wall

If your house was missing a wall, would you worry whether the front door was locked?

Well, the stock market is missing a wall. 

And yes, the image here is not a missing wall but the walls you can ski at Steamboat.

To our story, investors and public companies need to understand what it means to have a market with a gaping maw in its structure.  Imagine if you went around the house before bedtime shutting off lights, closing windows, locking the front door.

And a wall was missing.

That’s funny like The Truman Show.  Yet glaring structural flaws are no laughing matter.  For investor relations, the profession linking Wall Street to public companies, the job is keeping the house in order.

The lights and windows represent things long done to promote differentiators.  Earnings calls.  Non-deal roadshows (that means visiting holders without pitching a stock or debt offering). Analyst days.  Investor-targeting. Perception studies. On it goes.

You’d be surprised at the work and effort directed at these missions, you who are not in the profession.  We have an entire body of knowledge around the discipline of investor-communication, internally and externally.  Sums spent on tools and services and regulatory compliance and listing fees and so on total billions of dollars annually.

Keeping the house up is a full-time job. But the house is missing a wall. And too often the profession acts like everything is fine.

Evidence of it splashed again through the stock market yesterday. Rocket Companies (NYSE:RKT) slammed into at least three trading-volatility halts en route to a 71% gain.

It’s a great company.  CEO Jay Farner – I heard him on Squawk Box via Sirius Radio last Friday after results as I was descending from the Eisenhower Tunnel to Silverthorne off the Continental Divide – is so impressive. The cynosure of a prepared spokesperson.

And they’re a rocket, no doubt. RKT had $17 billion of revenue and $9 billion of net income, a 53% margin. Who knew selling loans was a more profitable business than selling data?

But.

That’s not what juiced RKT.  Sure, it was up Feb 26, up again Mar 1. But Active investment – the money focused on fundamentals – was 6.8% of RKT volume Mar 1.  So 93.2% of it was something else.

What’s more, Short Volume exploded at the very time rumors surged across the media that we had another Gamestop (GME) here, another “short squeeze,” now in RKT.

I’ll explain everything, so stay with me.

On Feb 19, Short Volume, the percentage of daily trading representing borrowed stock, was 29% of RKT volume.  By Feb 26 as the stock began to jump, it was 55%.

This is the draft you feel, public companies. This is the missing wall leaving you utterly exposed to what’s outside your control, no matter how much you spend, no matter the windows and doors you lock.

Investors, we’ll get to the meaning for you too.

There is only one way for a stock to rise.  Somebody sells stock at a higher price.  Not what you expected?  You thought I’d say, “Somebody must be willing to pay more.”

Yes. That’s true.  But suppose no owners want to sell, because they believe a mortgage company growing more than 100% year over year with 53% margins is a stock to hold.  And suppose the shorts don’t react.

Where does stock come from then?

Answer: That explosion in Short Volume we saw before the stock skyrocketed.

Somebody SHORTED the stock to make it available for sale at a higher price.  Sound cognitively dissonant?  It’s not. It’s called a “market-making exemption.”

We’ve written about it often before.  Read what we wrote on GME.

Truth is, most of the time when investors buy stock, somebody shorts it first and buys it back for the buyer.  At any moment, this can happen to any of you, public companies. And it doesn’t require high short interest – that 1975 measure still in use like a Soviet relic.

No, it can happen on belief. It can happen on nothing more than a machine-driven updraft.  And it could as easily be a downdraft.  What’s to stop market-makers from creating stock to SELL instead of buy? Correct. Nothing.

No matter what your story, your fundamentals, Fast Traders sitting in the middle can create shares out of whole cloth, because regulators have given them leeway to do so, and your stock can behave in unimaginable ways.

Public companies, if you directed 10% of what you spend on story and compliance to targeting Congress for a solution, we might shore up structure.

A solution starts with understanding the problem. And we do.  We can help you (and your board and c-suite).

Investors, your turn.  Market structure is to me the most important trading factor because it trumps everything else.  If you were modeling RKT market structure, you’d have seen the entry point Feb 22.

How many more situations like these will it take before we start looking at the missing wall? 

Big Cheese Grater

Good offense beats good defense.

These five words are the heartbeat of the Saban Dynasty in football at Alabama – and the reason for the Gamestop trade in the stock market.

Promise, this piece isn’t about sports. It’s about how retail traders killed institutions.

It takes some history. Go back to 1995, and the spread charged by brokers executing your trade was about 13%, or an eighth of a dollar.  That’s on top of the commission you paid.

Along came Electronic Communications Networks (ECNs) offering to match trades at a fraction of that spread automatically using computers.  They took 50% of trading from exchanges.

The exchanges responded in three ways.

Follow me here. I’m explaining why GME went up 1,000% and why retail money is running circles around institutions.  Everybody in the markets should understand it.

Back to the exchanges.  They first cried foul to regulators and sued ECNs.  Then they bought the ECNs (ECNs Brut, Island and Archipelago became technology engines for the Nasdaq and the NYSE).

And they adopted two ECN pillars:  They began paying high-speed traders to set prices, and they invested in algorithmic technology to help big institutional customers fill trades in a stock market infested with small orders and fast trades.

Step forward.  The SEC sniffed the wind and first required stocks to trade at a penny spread and then forced all the markets together under Regulation National Market System, thinking it would address what the ECNs had highlighted: The market wanted smaller intermediary spreads.

Still with me?  GME went up 1,000% because of what we’re talking about here, and institutions have been left out in the cold (public companies, you need to know this!).  I’m getting to it. Don’t quit!

Back to our story, institutional investors and the brokers helping them execute trades invested billions of dollars in computerized trading systems that would split up big orders – like a million shares – into tiny 100-share trades that, and this is key, wouldn’t chase after deviations in price.

I liken it to seeing the market as a cheese grater and institutional orders as a block of cheese. Technologists figured that a stock market forced into tiny spreads and trades would mean constantly changing prices. So why not fashion algorithms that would position the block of cheese of buys and sells like tiny trades – all the teeth on the grater?

Genius!

And so exchanges crafted order types to help big brokers and their customers match the cheese of orders to the grater of the market.

It was a compromise, not an act of service.  After all, exchanges continued to pay traders to be the best bid or offer – the going rate is around 25 cents per hundred shares still – so the cheese graters for big institutions would sit unseen behind the displayed prices for stocks, grating away and filling trades at midpoint prices.

And then along came the Reddit mob.

The unwitting genius behind the crowd is that it doesn’t know market structure. It just followed greed, a human impulse. That is, the aim of the Reddit gang is to run prices up. The aim of the Big Cheese Grater is to fill orders as prices run up and DOWN.

The whole structure of the stock market, with blessing and help from regulators, encourages prices to move both up and down.

It all fits together.  Traders will always make 100 shares of every stock available if prices move up and down.  If prices move up and down, exchanges make billions selling data. And Fast Traders, the supply chain of the stock market, need prices to move up and down to profit going long and short.

Good offense beats good defense.

Buys are offense. Algorithms are defense. And the former flat out OWNED the latter, because the Reddit traders, without understanding what they were doing, chose only offense. They demonstrated a nervy willingness to chase prices higher that utterly demolished algorithms designed, like cheese graters, to capture up-and-down moves.

Intermediaries wanting to make $0.003 per trade didn’t care if the cheese grater malfunctioned. So they fed the mob frenzy what it wanted. Higher prices.

And this can happen every day, any day. In anything.

The problem for public companies is your big investors need to put $2 billion into the market within a price-range to get a return. A retail trader just needs 10% on 100 shares.

You see? It’s a problem 30 years in the making. Now it’s here.

What fixes it? That’s a topic for another day. But it’s coming. It’s Regulation National Market System II.  We’ll see it within two years.

Metastable

Editorial Note:  A giant thank you to Client Services Director Perry Grueber for penning last week’s Map.  He’ll be back regularly, by popular demand. This week you’re stuck again with me. -TQ

Something is worth what another is willing to pay for it.

That’s the lesson of the maelstrom in financial markets, from wood pulp futures, to bitcoin, to GME, to AMC, to wherever the next Dutch tulip bulb of the 21st century showers the shocked with inflationary sparks.

Before we get to that, we’re back from riding the trade winds around Antigua, where high season in the Caribbean looks like turnout for a vegan tour at an abattoir. Nobody is there.  And the Caribbean has no central bank doling out cash for sitting home trading in a Robinhood account.  We did our best to offer economic support.

I’ve posted some photos of our circumnavigation here.

Oh, and you’ll recall that my Jan 27 Map said, “Congrats, Tom Brady. We old folks relish your indomitable way.”  When you’re going to sea, always bet on the buccaneers. And the old guy.

We were saying a thing is worth what somebody is willing to pay.

Yesterday GME closed at $50.31.  On the Benzinga Premarket Prep Show Jan 25, right before we grabbed our flipflops and duffle bags and bolted with our Covid19 negatives for the airport, I told the audience, “Market Structure shows GME is going to go up.”

I didn’t know it would rise to $483 while we were at sea.  But somebody was willing to pay more. Until they weren’t.

Right now, AMC, BB, BBBY, NOK, and so on, are outliers.  What if the scatterplot gets crowded?

Most of the people on what Karen calls the “What Do You Think of THIS Stock?” TV shows are still talking about the PE ratio. Earnings growth. Secular trends. Economic drivers. You get the idea.

Investor-relations people, are you prepared for a market full of Gamestops – surging highs, avalanche tumbles? How about you, investors? 

In physics and electronics, “metastability” is the capacity of a system to persist in unstable equilibrium. That is, it seems solid but it’s not.  Like the stock market.

Don’t blame Reddit.  I love the flexed muscle of the masses.  I’d like to see it in society elsewhere, frankly.  A horde of people who refuse to be told what to do or told they can’t.  A mob of unruly traders is wholly American.

But all those people, and all the rest of us in the capital markets, ought to understand the metastability that makes a GME or an AMC possible.

Most retail orders are sold to intermediaries trading at extreme speeds.  Those firms aren’t calculating PE ratios. They’ll pay somebody for a trade so long as they know somebody else in the pipeline is willing to pay more.

Hordes of limit orders hit the pipeline, and intermediaries see the whole hierarchy. They race prices up, skipping swaths of limits by raising the price past them. So those traders, if they’re greedy and willing to chase, jump out of line and enter new, higher limit orders.

And mania ensues because somebody is willing to pay more. 

When the high-speed intermediaries see that limit orders to buy and sell are equalizing, they stop filling limit orders, they short stocks, and they skip limit orders on the way down, and the whole cavalcade reverses.

And it’s not just retail flow or big high-speed penny-pickers in the middle. Quant firms do it. Hedge funds do it.

Two factors make a metastable stock market possible.  First, rules require a spread between the bid to buy and offer to sell.  So somebody will always have a split-second motivation to pay more for something than somebody else.

Indeed, it’s what regulators intended. How do you foster a market that never runs out of goods? Give them a reason to always buy and sell (exchanges pay them for best prices to boot). And regulators let those Fast Traders manufacture shares – short them – that don’t exist, and persist at it for weeks.

The market is nearly riskless at any price for the raciest members moving unseen through the Reddit ranks. They bought the trades. They know what everybody is doing.

And that’s why we have a metastable market.

The alternative? You might not be able to buy FB or sell NFLX at times.

Would we rather have a false market with ever-present orders or a real market occasionally without them? Regulators chose the former. So did the Federal Reserve, by the way.

And that’s why everybody in the US stock market better know how it works. We do. Ask us for help, public companies, and investors.

Sailing Away

Sailing takes me away to where I’ve always heard it could be just a dream and the wind to carry me.

Christopher Cross said it (youngsters look it up). In this pandemic we said, “That boy might have it figured out.”

TQ and KQ sailing

So, with two negative Covid tests in hand, we’re currently near 17 degrees North, 62 degrees West readying our 70-foot catamaran for a float with friends.  Chef, bar, crew, trade winds blowing our hair around, azure waters, sunrise, sunset. We’ll catch you after, Feb 8.

And in between, let’s have a look at the market.  The big buzz is GME, Reddit now dominating chatter with WallStreetBets (y’all can look that up too), the stock streaking, a push-pull among longs and shorts, and Andrew Left from Citron cannonballing into the discourse and an pool empty.

It may be a sideshow.  GME is up because Fast Trading, the parties changing bids and offers – shill bids, I call it – and buying retail volume surged from 38% of GME trading to over 57%.

At the same time, Short Volume, daily trading that’s borrowed, plunged from 47% to 34%. The funny thing is it happened AFTER the news, not before it.

The Reddit WSB crew has the sort of solidarity I wish we’d direct at being free. Nobody says to them the words “allow,” or “mandate,” and I love that.

But.

In a free stock market, your actions as traders are known before you make them.

That is, plow millions of limit orders into the market from retail brokerage accounts, and the firms like Citadel Securities buying them know before they hit the market.  They will feed the fire, blowing on the conflagration until it runs out of fuel.

And BBBY is up 50% in two weeks.  But it’s not the same, looking at market structure (the behavior of money behind price and volume in context of rules). Quantitative money plowing into BBBY to begin the year ignited the surge.

Could the actions of machines be misunderstood by humans?  Of course. Already the pattern powering GME has reverted to the mean.  In BBBY, Short Volume is up already on surging Fast Trading, the same machines we just talked about.

All but impossible is beating trading machines. They know more, move faster.

However, they are, paradoxically, unaware of market structure beyond fractions of seconds into the future.

Humans have the advantage of knowing what’s days out.  And on Fri Jan 29, the largest futures contract in the market comes due.  It’s designed to erase tracking errors. This is a much bigger deal than GME and BBBY but not as much fun.

Tracking errors are the trouble for Passive investors, not whether they’re “beating the benchmark,” the goal for Active stock-pickers.

A tracking error occurs when the performance of a fund veers from its benchmark.  The aim is generally less than 2%.  Yet S&P 500 components are 2.5% volatile daily, the difference between highest and lowest average daily prices. For those counting, daily average exceeds monthly target).

It’s why Passives try to get the reference price at market-close. But the market would destabilize if all the money wanting that last price jammed into so fleeting a time.  It would be like all the fans in Raymond James Stadium pre-pandemic – capacity 65,618 – trying to exit at the same time.

Congrats, Tom Brady. We old folks relish your indomitable way.

Like Brady’s achievements, everybody leaving RJ Stadium at once is impossible in the real world.

So funds use accounting entries in the form of baskets of futures and options.  ModernIR sees the effects.  The standard deviation between stocks and ETFs in 2019 was about 31%.  The difference reflects the BASKET used by the ETF versus ALL the stocks. To track that ETF, investors need the same mix.

Well, it’s not possible for everyone in the market to have the same quantity of shares of the components. So investors pay banks for options and futures to compensate for those tracking errors.  The more errors, the higher the demand for true-up derivatives.

In 2020, the average weekly spread rose to 71%, effectively doubling.  In the last eight weeks since the election it’s up to 126%.

The paradoxical consequence is that increasing volatility in benchmark-tracking is creating the illusion of higher demand for stocks, because options and futures are implied DEMAND. 

And so we’re

sailing away. You guys hold the fort. Keep your heads down.  We’ll catch you after the last Antigua sunset.