Acceleration of market trends - due to COVID or new normal?

[Lucky C]

It seems like trends in the markets occur over faster time frames than in the past. For example, the recent record fast bear market followed quickly by new highs. Also, oil hit negative prices and gold hit new record highs this year. But beyond just record numbers, it seems like sentiment has been changing very quickly (in a matter of weeks or months) in a variety of assets.

I'm not sure if this seems "new" to me because this is the first market cycle that I'm paying attention to sentiment and these rapid shifts in trends (it occurs during every cycle?), or if these rapid shifts make sense (due to the sudden COVID-related changes?). Or is this a "new normal" to be expected in future market cycles as "news" travels faster (internet and trading algos), investment style changes play out (everyone doing passive), and just general differences in "kids these days" - instant gratification, nobody long-term investing, /r/wallstreetbets...

Has anyone studied this, e.g. published a blog post or paper on acceleration of trends within market cycles? Or is this just how things play out during a market-top mania?

[jacob]

https://www.amazon.com/Forecast-Physics ... 608198537/

My "intuition" says increasing lack of strategy-diversification dominated by the futures market (limited lack of levers) leading to lots of beta-ripping. Combine that with more "stratospheric" hot-money that has nowhere to go but equity because of ZIRP/NIRP.

Or in pedestrian terms ... In the short term you have a bunch of futures traders playing the hourly psychology with a "green lumber" mentality and block-arbitrage moving individual issues without much concern for any connection to the underlying economy. In the long term you have an equity market that's too big to fail (for political reasons and because all the retail/pension savings have been plowed into it) and which has largely been detached from its role in corporate financing.

[unemployable]

I see it pretty simply: that markets are schizophrenically reacting to an uncertain environment. Things are getting worse, no they're getting better, we have a vaccine, now we have another one, but wait it'll still be a few more months, no one's traveling, but they will be traveling soon, what do you mean by soon, Biden won, no Trump doesn't think so, now he's conceding, no he isn't conceding.

It's not really Dow versus Nasdaq but large-caps versus small-caps. VV versus VB. Their day-to-day moves must have perfect negative correlation. I'm on Team VB in case you care. Smallcaps are down today but somehow my "recovery portfolio" is slightly up thanks to oil.

So as long as this environment persists, I don't see why markets won't keep reacting the way they do.

[Lucky C]

To answer my own question with a little data, I looked at volatility in equities and found that the past year's volatility* is not exceeding any historical norms and is right where you'd expect it to be based on a bad crash/shock over the past year. It's close to the peaks seen in 1974-75, 1987, & 1991, and below the 2008-2009 spike. It's above the 1998-2003 peaks from LTCM, the tech bubble, and the bust/recession.

So that's 12-month volatility which shows that if you take the whole past year into consideration, volatility hasn't been that crazy. But we did have an especially crazy three months around Feb-Apr. Calculating 3-month volatility instead, the vol spike from earlier this year was only exceeded by the 1974 and 1987 crashes. Now I realize that without modern monetary policy and circuit breakers we could have had more volatility than those events, but it seems reasonable that we have been seeing this amount of volatility (3x in 50 years occurrence) given the events that have transpired.

I think I've been experiencing some recency bias (pretty smooth bull cycle before this year) combined with lack of experience with other stock crashes and manias, and that there may not be much evidence that we are in a new era of faster market cycles.

It would probably be wrong to think that the next bear market, whenever it comes, would be another rapid 1 month drop rather than the historical norm of taking several months/years to play out.

*Monthly MSCI total return data for both USA and WORLD indices. Calculated standard deviation of month-month changes using month-end total return data, and scaled by *sqrt(12 months).

[Mister Imperceptible]

@unemployable

I would say the markets more likely behave because of the dynamics Dr. Fisker described and the idea that markets are reacting to the headlines are post hoc rationalizations because no one wants to admit the truth.

[unemployable]

I agree with the "equity market is too big to fail" thesis, but not that suddenly it's more futures traders' fault than it usually is. We've had 24-hour electronic futures trading since the early 1990s and speculation in contracts for future delivery in some form is a centuries-old idea (see 1637 Holland).

Derivatives on the whole serve to make price discovery easier, not harder.

Anyway, why not Robinhooders? I believe zero commissions have had a big impact on market behavior -- little guys, which futures traders are generally NOT, can trade much more frequently and efficiently. That's not a complaint, as I have benefitted as well.

[7Wannabe5]

Relatively less friction for the littlest guys ( like me), because not as much need to fret about taxation as the biggest guys. Seems to me that it must totally change a good deal of conventional wisdom.

[Lucky C]

Plus the elimination of trading fees over the past couple years with the biggest benefit being amateur day traders.
And nudges toward getting into stocks / it's not just for old people, through the gamey Robinhood interface, teenage youtube investing channels, etc.

[IlliniDave]

The March recession was different than the several prior I've witnessed in that it was essentially voluntary and has a reasonably known endpoint, or so most people believe. Low interest rates mean TiNA (there is no alternative) and the 2017 tax changes (watch the drop as soon as serious talk about jacking corporate/business taxes up start) seem like upward pressure. I've been selling small chunks over the last 3 years, as have the few people I talk investing with in person.

One time I added up how many All Time High(!)s I have lived through looking at month-end data. Dozens. That Shiller PE is pretty high is more interesting and why I've been keeping my exposure limited rather than letting it run. If I didn't expect the SP500 or Dow to routinely break their prior records I probably wouldn't invest in stocks.

[7Wannabe5]

This might be a stupid question, but how out of whack is Schiller PE compared to something like 10 Year Treasury Rate? Might be time to start buying solar acreage and meat rabbits and/or building your own robots.

[Lucky C]

I looked at my spreadsheet of daily S&P500 price history 1928-present to get a finer grain look at the extreme moves this cycle instead of only end-of-month data I mentioned in my earlier comment.

I computed 1-month returns as the price change over 21 trading days (the average number of business days per month). Then I computed the magnitude of a 2-month whipsaw (1-month down followed by one month up) as the most recent 21-day price change minus the previous 21-day price change. This one month down / one month up time scale is what I am concerned with when I think the market is moving "too fast" - much faster than a typical months- or years-long bear market and much harder to determine how to react, if at all. Of course it is also the length of time of this past Feb-March bear market, so we can compare previous whipsaws to that recent one.

Ranking these whipsaws, the biggest over all this time based on my definition above was in 1932 after the market finally bottomed years after the 1929 crash. There was about a -6.5% drop before the bottom, followed by a gain of 61% in a month.

The second biggest whipsaw was a few years earlier during the initial 1929 drop. Obviously the most famous historical crash, but it was hardly a straight path down! The S&P lost 39% in a month followed by a 23% gain the next month.

The third biggest 1 month down / 1 month up whipsaw was the one this year. About a 31% drop followed by a 22% gain the next month.

After that, there were a couple more spikes in the early 1930s which were greater in magnitude compared to more modern ones you'd expect: the 2009 bottom, the 1987 crash, the 2002-2003 bottom.

I'm not sure if there is a big lesson here, but at the very least this shows that the extreme market moves in the spring fall somewhere between once-in-a-cycle (about once a decade) moves and once-in-a-century (Great Depression) moves. Sounds about right. Also seems like there isn't much reason (yet) to fear that future market cycles/crashes will keep getting faster and faster. Looking at these whipsaw events and volatility over decades doesn't point to any recent upward trend.

But what made this year extra difficult for me to understand and react to correctly is that these type of quick notch-like moves are typical of major bottoms that occur after long bear markets at much lower valuations. We've never seen this type of quick rebound to new highs while remaining in high valuation territory. The initial 1929 drop and bounce was soon followed by lower highs and then lower lows.

[unemployable]

This might be a stupid question, but how out of whack is Schiller PE compared to something like 10 Year Treasury Rate? Might be time to start buying solar acreage and meat rabbits and/or building your own robots.

It's not a stupid question.

https://www.yardeni.com/pub/valuationfed.pdf (live link to PDF updated weekly)

If you hold the Fed Model as gospel, then US stocks are undervalued by a factor of about five. Fair value of the SPX would be around eighteen thousand (18,000). So out of whack I feel the need to invoke the lawyer convention of using both digits and words. All else you need to know is how many different grades of bonds Yardeni compares it to, just to find some reason to justify the model as relevant.

I know you asked about Case-Schiller. What's a typical cap rate on a house now, maybe 5-6%? With tennies yielding in range of northeastern Interstate highway numbers, houses are "cheap" by a factor of what, around six or seven?

[7Wannabe5]

@unemployable:

Yeah, that’s what makes gut sense to me. I’m about 40% in cash, but only because I might buy some real estate. But maybe that’s because I’m used to thinking more in terms of trading rather than investing. If it costs next to nothing to turn over your inventory, why wouldn’t you do it as frequently as seems sensible?

[jacob]

Short answer: "next to nothing" * "frequently" >> 0

Medium answer: https://en.wikipedia.org/wiki/Bid%E2%80%93ask_spread and https://en.wikipedia.org/wiki/Slippage_(finance) + you're up against both market making and ignition algos which aim to profit from exactly that behavior.

Long answer: TANSTAAFL. Fees are quite complicated under the hood. See e.g. https://markets.cboe.com/us/equities/me ... edule/bzx/ (random but representative of how your "free" broker sees the world.) You can actually have gross losing trading strategy that's a net winner after rebates. Also see https://en.wikipedia.org/wiki/Broker-dealer ... some of which have been known to sell their customers' order data

[unemployable]

Pretty sure your average Robinhooder has never heard of the phrase implementation shortfall. Or thinks it's stupid, because scoreboard, my 8-lot of TSLA's a winner! With zero commissions by definition all the costs of trading are invisible.

[7Wannabe5]

@jacob:

Gotcha, but I meant more like selling if you’re already up 20% and you have some other prospects or clearing out the deadwood. Like with trading rare books or doing retail arbitrage. I don’t really know what I’m doing but I’m up 82% for the year in my little Robinhood account. Probably next year I’ll be broke

[Flurry]

Historically speaking, low interest rates have the tendency to increase speculation much more than of real economic growth.
The flash crash in march was the best example. Everything (even highest quality government bonds and gold) were sold and there was a high correlation of all asset classes. Why? Because everyone's so high on margin and when stocks crash you need cash to offset that.
But now we are back and there is record high leverage in the stock markets again. This really has potential for the worst financial crisis imaginable, let's see how the central banks will handle it this time.

[7Wannabe5]

Yes, but even if you believe the market is bloated and due for a crash or harsh correction, isn’t it still rational to apply some probability to the interim? The odds that the market is going to crash on December 2nd do not equal the odds that it is going to crash some time in the next year. Also, why not have different accounts and/or realms in which you practice at saving, investing, trading, and speculating? My little Robinhood account is my fun trading account.

[Lucky C]

record high leverage in the stock markets again.

I'm sure there are many ways to measure this, but I like to look at this page on FINRA margin debt, which up until recently showed NYSE margin debt (NYSE suspended its publication at the end of 2017): https://www.advisorperspectives.com/dsh ... -september

Anyway by this measure, yes margin debt is well above where it was pre-Covid. But we are still setting lower highs than in January 2018. I interpret this as less confidence that this is a sustainable multi-year bull market. Also that while some "dumb money" day traders may be leveraging as much as they can, the big guys are probably more cautious vs. when "everything was awesome" a couple years back - post-Trump tax cuts, pre-manufacturing slowdown, and of course no pandemics.

There are not a lot of cycles for a good sample size, but based on this limited data there is a clear pattern of margin debt falling before the market falls, and bottoming in the same month that the market bottoms.

September's margin debt (October not published yet) showed weaker margin debt growth which makes sense with September being bad for equities. October probably won't see a big increase either. If margin debt starts to contract consistently over the next few months, I'll take that as evidence that we're in another bull market top where the start of a new bear market in the next year or so shouldn't be a surprise.

[Lucky C]

https://www.amazon.com/Forecast-Physics ... 608198537/

My "intuition" says increasing lack of strategy-diversification dominated by the futures market (limited lack of levers) leading to lots of beta-ripping...

Thank you for the book recommendation. I have bookmarked it for some point in the future after addressing my tsundoku (I knew there had to be a word in some language for acquiring lots of books but not getting around to reading them).

By "beta-ripping" do you mean anything more/deeper than my basic interpretation of "high beta stocks are going up bigly"?