Investing in Next Week's News

[MustacheAndaHalf]

ChpBstrd - There is a chance of a very bad outcome that wipes away the Mar 23 low point for the year.  I'll have to think about your scenario some more, but here's something to make it even worse: in Korea, numerous people were confirmed with COVID-19, then recovered and were confirmed negative, and are now sick again with COVID-19.  China reports the same thing, in even higher numbers.  So not only could we have a second wave, the same people could get re-infected and fare worse.


I also have an odd speculation about how this plays out.  Normally a company can't cover its costs, and files for bankruptcy.  It can't pay the rent.  Now, what if half of America can't pay the rent?  If a landlord doesn't get paid, and there's nobody to move in - nobody to pay rent - what happens?  If one person doesn't pay rent, the system has rules for what happens.  If nobody pays rent, the system breaks.

Maybe the least risky thing to do is invest in bank stocks (as a whole).  Most states have a new law that simply doesn't allow foreclosure.  When someone can't pay their mortgage, nothing happens - for now.  In 2008, the government didn't take aim at individual homeowners - it saved the banks.  During past crisis, something similar happened.  Right now the financial sector is down, but if everything else falls apart, I suspect the banks (as a whole) will survive.  So maybe that's the safe haven, of sorts?

---
I'm not seeing a dramatic increase in COVID-19 cases in Denmark yet - just a very small change upwards that might be nothing.  So far, opening back up the schools is going better than I expected there.

[Buffaloski Boris]

ChpBstrd - There is a chance of a very bad outcome that wipes away the Mar 23 low point for the year.  I'll have to think about your scenario some more, but here's something to make it even worse: in Korea, numerous people were confirmed with COVID-19, then recovered and were confirmed negative, and are now sick again with COVID-19.  China reports the same thing, in even higher numbers.  So not only could we have a second wave, the same people could get re-infected and fare worse.

The WHO issued something indicating that infection doesn’t necessarily provide immunity. This is just the gift that keeps on giving.

I also have an odd speculation about how this plays out.  Normally a company can't cover its costs, and files for bankruptcy.  It can't pay the rent.  Now, what if half of America can't pay the rent?  If a landlord doesn't get paid, and there's nobody to move in - nobody to pay rent - what happens?  If one person doesn't pay rent, the system has rules for what happens.  If nobody pays rent, the system breaks.

It’s not an odd speculation. Rent strikes are already being planned in the large cities. The courts aren’t really functioning that well for civil cases right now. So no foreclosures just yet because of the required forebearance on federally guaranteed loans. And no evictions. Keep in mind though that evictions are almost always a judicial matter: you have to get the courts to kick out a tenant. If they’re even willing to. And it takes time.

I’d be crapping bricks right now were I a landlord. We’ve seen this sort of apocalyptic scenario happen before, though. Back in the Great Depression. For landlords it often came down to taking vastly less in rent to keep homes occupied. I don’t know how that would work with all the leverage, probably some defaults and some forebearance.

Oh and when those bankruptcies start to hit in a few months? The courts will be backlogged for quite awhile.

Maybe the least risky thing to do is invest in bank stocks (as a whole).  Most states have a new law that simply doesn't allow foreclosure.  When someone can't pay their mortgage, nothing happens - for now.  In 2008, the government didn't take aim at individual homeowners - it saved the banks.  During past crisis, something similar happened.  Right now the financial sector is down, but if everything else falls apart, I suspect the banks (as a whole) will survive.  So maybe that's the safe haven, of sorts?

---
I'm not seeing a dramatic increase in COVID-19 cases in Denmark yet - just a very small change upwards that might be nothing.  So far, opening back up the schools is going better than I expected there.

How do you figure with banks? I agree that if the financial cartel gets the hiccups, it’s an existential crisis for the Fed who will come to the rescue. Is that your line of thought?

[Retire-Canada]

ChpBstrd - There is a chance of a very bad outcome that wipes away the Mar 23 low point for the year.  I'll have to think about your scenario some more, but here's something to make it even worse: in Korea, numerous people were confirmed with COVID-19, then recovered and were confirmed negative, and are now sick again with COVID-19.  China reports the same thing, in even higher numbers.  So not only could we have a second wave, the same people could get re-infected and fare worse.

Some people who were cleared of SARS-CoV-2 by test later tested positive again. This does not mean they were sick again through a new infection, were sick at all or were even contagious. It just mean enough viral fragments were present to trigger a positive test.

The WHO issued something indicating that infection doesn’t necessarily provide immunity. This is just the gift that keeps on giving.

Saying we don't have enough data to come to any conclusion with respect to immunity doesn't mean immunity is not likely given our experience with other similar viruses. The WHO statement just encourages people to pursue the research we need to come to a definitive conclusion on the issue.

[Body Surfer]

very interesting discourse in this thread. I agree with CstBstrd: i believe percentage of down side is more likely than upside. Market seems so out of touch with reality right now- it is shocking to me.

[ChpBstrd]

@ChpBstrd  One contrary influence to that though is the large amount of institutional money that is sitting on the sidelines that is hungry for higher yields with a longer range focus that is waiting to dive in at an opportune time. That would have the effect of muting downward prices.   

Thanks for the kind words @Buffaloski Boris. The argument about piles of institutional cash sitting on the sidelines is familiar to me. I heard it on forums just like this one in 2000 and 2008. There are cowboy fund or portfolio managers out there who will call a bottom and go in with a dramatic allocation of tens of millions of dollars, causing a bump in prices and minor momentum run, but there is a finite quantity of these folks. Institutional money such as pension funds, sovereign funds, and endowment funds are going to be run by committees who are held to account for following a pre-written plan. If anything, they'll be forced sellers as their institutions' revenues take an hit and expenses go up.

I also have an odd speculation about how this plays out.  Normally a company can't cover its costs, and files for bankruptcy.  It can't pay the rent.  Now, what if half of America can't pay the rent?  If a landlord doesn't get paid, and there's nobody to move in - nobody to pay rent - what happens?  If one person doesn't pay rent, the system has rules for what happens.  If nobody pays rent, the system breaks.

Maybe the least risky thing to do is invest in bank stocks (as a whole).  Most states have a new law that simply doesn't allow foreclosure.  When someone can't pay their mortgage, nothing happens - for now.  In 2008, the government didn't take aim at individual homeowners - it saved the banks.  During past crisis, something similar happened.  Right now the financial sector is down, but if everything else falls apart, I suspect the banks (as a whole) will survive.  So maybe that's the safe haven, of sorts?

I'm not sure I follow why financial institutions originating or holding mortgages will fare well when the government declares a payment holiday and bans on foreclosures. Maybe there's a missing assumption or reasoning step that isn't obvious to me, such as an expected government bailout to the financial institutions to cover the costs of the jubilee? Given the view out the window, I wouldn't want to own a mortgage REIT or a large bank right now. Compared to what sank them in 2008, this is a much bigger iceberg. High quality underwriting will not help them this time!

Banks tend to borrow short-term and lend long-term, harvesting the spread in interest rates. Their mortgage holdings were purchased with short-term borrowing or on-demand deposits. E.g. If banks were buying 30 year mortgage bundles yielding 3.5%, they must have been borrowing at 3% by issuing new bonds each year, holding depositors' money, CDs, etc. What happens if the effective yield on those mortgage bundles drops to 1% and they still have to pay the 3%? That was 2008 if you add 3% to both sides! Citi and AIG were demolished.

The courts aren’t really functioning that well for civil cases right now. So no foreclosures just yet because of the required forebearance on federally guaranteed loans. And no evictions. Keep in mind though that evictions are almost always a judicial matter: you have to get the courts to kick out a tenant. If they’re even willing to. And it takes time.

I’d be crapping bricks right now were I a landlord. We’ve seen this sort of apocalyptic scenario happen before, though. Back in the Great Depression. For landlords it often came down to taking vastly less in rent to keep homes occupied. I don’t know how that would work with all the leverage, probably some defaults and some forebearance.

I'm crapping cinder blocks on their behalf! On this forum, I've watched people in HCOL areas ask for opinions on rental properties, and my answer has always been hell nope. In most coastal areas / big cities the cost of rent is some fraction of what the mortgage payment would be, even at today's microscopic interest rates. You have single-family homes selling for $2.5M and only being able to pull $4k or $5k/mo rent. That's the gross, not the net! According to the 1% rule you should be getting $25k/mo for a property that expensive! The name of the game is to almost break even and harvest property appreciation. This game has created many millionaires over several decades in HCOL areas, but it's unsustainable musical chairs.

So now all these HCOL landlords and real estate corporations will not be able to make mortgage payments, because they were barely breaking even in the good times. Tenants' financial reserves in HCOL areas are probably an even smaller percentage of rent than they are in LCOL areas. Get ready for a foreclosure crisis! And worst of all, this foreclosure crisis isn't being led by deadbeats with NINJA loans, it's a more universal phenomenon.

What does it matter if you cut the rent on your $2.5M single-family home in LA from $5k/mo to $3k/mo? You still can't cover the mortgage on that, your jobless (or sick) tenants still have no way of coming up with $36k per year for rent, and now you've been banned from evicting the people who have been squatting for 6 months. You'll initially try to sell to recover some of your $625k in equity, but if prices drop below that level all you can do is walk away from the property. 

[MustacheAndaHalf]

Just for context, 30% of my portfolio is individual stock picks.  These were deeply discounted when I bought around March 25, with almost all falling more than -60% YTD when I bought.  My biggest concern is what will go bankrupt - I expected some will, but others will make up for the risk.

With that context, and ChpBstrd mentioning -70% being possible, my question is what survives bankruptcy.  Right now Congress and the Fed are bankrolling a bankrupt economy.  The U.S. economy is basically bankrupt, held together by relief packages and the Fed buying "junk" bonds on public markets.  If a critical mass in Congress reject relief packages, those stop.  At that point, we could see a market crash worse than recent crashes - if that happens.

If the funding stops, it's possible bankruptcies start spreading.  Maybe they clog up the courts and new laws are created to handle it, or maybe it just takes years and nobody knows which business is solvent - like 2008.  In that possible scenario, I think the banks, as a group, are too big to fail.  Even if some companies are allowed to go bankrupt, I think banks will survive.

I looked at the sectors of the U.S. economy to see which ones were down the most.  Naturally #1 was energy, with no demand during a price war that floods the market with out.  But in the second spot is Vanguard Financials ETF (VFH), down -30.5% YTD.  If the economy crashes, do banks fall as far as other stocks?  If a recovery occurs, do banks recover a greater amount than other stocks?

FDIC insurance was created back in 1933, owing to the great depression.  Maybe banks could go broke back then, but to prevent another great depression, the U.S. government seems willing to keep banks in business.  So maybe banks are the right balance of beaten up value stocks, while also affording some protection against worst case scenarios going forward.

[MustacheAndaHalf]

RetireCanada - While both our scenarios are possible, your simpler explanation seems more likely.
https://www.npr.org/sections/coronavirus-live-updates/2020/04/17/836747242/in-south-korea-a-growing-number-of-covid-19-patients-test-positive-after-recover
"If a patient had not developed sufficient immunity against the virus or if a patient's immune system weakens after recovery, the previously undetectable level of virus concentration could rebound."
"Another possibility is that tests are picking up dead virus particles that are no longer infectious or transmissible."

---
Normally over the week/weekend I get a sense of market sentiment, and then check data to see if that sentiment is accurate.  I flip from pessimistic to optimistic and back based on data.  Right now, I'm just confused - neither optimistic or pessimistic.  There's a famous scene from Star Wars that seems appropriate, where they "can't get a reading on the shields... up or down".  That's a bit like me looking if stocks will be up or down, and the next quote is the warning: "It's a trap!".  I don't know if markets will make a quick recovery, or crash to new depths.

Right now I'm about 80% equities, and I don't know if the future is a quick recovery (100% equities?) or a crash to new depths (60% equities? - my personal minimum).  I need to look at how I can do well under both scenarios.  Most likely, I'll go lower than 80% equities, and use margin / leveraged ETFs / options to make up for the increased cash position.

[ChpBstrd]

Banks fail all the time. The FDIC facilitates a new home for the accounts and the owners lose all equity.

For too-big-to-fail banks, we would see something like what happened to GM or AIG a dozen years ago. The company essentially declares bankruptcy so its assets can be purchased, the business continues operations, existing shareholders are wiped out, existing debt holders get a buzz cut or are wiped out, and the business resumes from a less-indebted starting point with new ownership.

[ChpBstrd]

Right now I'm about 80% equities, and I don't know if the future is a quick recovery (100% equities?) or a crash to new depths (60% equities? - my personal minimum).  I need to look at how I can do well under both scenarios.  Most likely, I'll go lower than 80% equities, and use margin / leveraged ETFs / options to make up for the increased cash position.

I'm considering options as a way to set up a position with limited loss potential and high gain potential. The recent unjustified run-up in prices sets up an opportunity to offload risk.

For example, I could buy deep-ITM calls on SPY maturing in March 2022:

SPY current price: $283.30
Call strike price: $250
Price of the call: $50.90
     Call intrinsic value: $33.30
     Call time value: $17.60
Max loss: $50.90 (or 18% of the price of SPY, and SPY would have to fall by 11.75% for that to happen)

So for $50,900, I could get close to the same exposure to SPY's upside as if I'd spent $283,300 buying the stock itself. I could leave the remainder of this portfolio in an FDIC insured account to ride out the pandemic and still be essentially 100% in equities. Meanwhile, I'd be exposed to only 18% potential downside over almost 2 years (if SPY<$250 at expiration and thus my calls lost all value).

If the S&P falls 60% from here, this "cash and calls" move would prevent $119k in losses. If the S&P goes up 20%, this strategy would miss out on 7% of the gains.If SPY rises, my underperformance would be the time value. Most importantly, there would be no chance of (1) SPY running away without me while I'm mostly hunkered in cash, or (2) a portfolio blowup that delays retirement by 5-8 years.

[MustacheAndaHalf]

Banks fail all the time. The FDIC facilitates a new home for the accounts and the owners lose all equity.

For too-big-to-fail banks, we would see something like what happened to GM or AIG a dozen years ago. The company essentially declares bankruptcy so its assets can be purchased, the business continues operations, existing shareholders are wiped out, existing debt holders get a buzz cut or are wiped out, and the business resumes from a less-indebted starting point with new ownership.

I slipped up in my post, but I meant "banks as a group" won't all fail.  Some banks will fail, making it clear the world isn't risk free.  But the U.S. will prevent the banking system from completely collapsing, which is why I mentioned Vanguard Financials, which holds 425 stocks.  I would want diversification to avoid individual banks going under - while staying exposed to a critical part of the economy that can't go under without taking the rest of the economy down with it.


It also seems to be approaching the point where I don't know any better than the market.  There's too many things to track, and a higher chance of getting it wrong.

Denmark's growth week/week growth rate moved slightly higher, than slightly lower.  Starting from Monday, April 20th, Denmark's new cases per day have been:
131   180   217   161   137   235   130

The highest count, 235, is roughly when I thought a big outbreak might occur.  And yet 100 extra people got sick, and the next day was back to normal.  There still could be some form of outbreak, but not a massive outbreak given those numbers.
(Note in Denmark's data, Tuesdays have higher counts than Mondays.  So the 131 -> 180 looks like an increase, but the Mon -> Tue from the prior week was 144 -> 193.  So both 131 and 180 are lower than the prior week)

Maybe there's enough testing, that with a decent lock down, countries can test their way back to a new normal economy.  That would suggest I don't know what will happen next in Georgia, and so I don't know what will happen as states reopen slightly.

I suspect I've lost my ability to market time - even though it hasn't cost me, yet.  I'm not aware of market sentiment, and not aware of data sources to contradict that sentiment.  If I really believe in data driving market timing, when I run out of data, I also run out of market timing.

[ChpBstrd]

Speaking of running out of data, I’ve been watching the CDC’s “Cases in the US” page for a few weeks. They post the previous day’s new infection count each evening. For weeks, this number has been flat around 30k/day. This struck me as odd because epidemics are expected to follow a bell-shaped distribution of new cases over time. How are we at such a perfect and unnatural plateau?

Maybe I’m dense, but the reason hit me only yesterday. The number of positive tests are relatively flat because the number of total tests being done is relatively flat. The number of total tests being done reflects the US’ current manufacturing and processing capacity for tests. We are using all of the tests we can produce/process each day and getting close to the same number of results each day. When the tests run out, patients who need to be tested are probably turned away and are not counted in any number.

This is terrifying because the actual number of cases could be ANY number above CDC’s count. The infection curve could still be logarithmic and we’d have no way to know. Worse yet, leaders and investors are making decisions off of “new infections” data that are really just a factor of our current testing capacity.

I need to put a spreadsheet together to check the correlation between the CDC data feed and the following day’s stock market performance. Anecdotally, it seems like the market opens up when the CDC number grows less than 35k, and down when it grows more than 35k. If there is a relationship, we can bet on it breaking down when investors realize they are operating on faulty data which implied that we’ve passed the peak.

[MustacheAndaHalf]

ChpBstrd - Now that growth is slower, I'm using the new approach I mentioned for Denmark: week over week growth in new cases.  Take the past several Sundays, starting March 29:
297%... 132%... 65%... 36%... 28% (Apr 26)

You can also divide new cases per day by new tests per day, for the past 10 days:
19.7%.. 19.4%.. 16.3%.. 16.9%.. 17.3%.. 8.8%.. 16.2%.. 14.7%.. 13.1%.. 13.0%

And since this is a deadly disease, you can also measure the week over week growth in the total number of deaths.  Nasty, but here it is for Sundays starting on March 29:
580%.. 384%.. 229%.. 163%.. 135% (Apr 26)

Those numbers are ugly, but I mention them for two reasons.  First, the growth rate is rapidly declining.  And second, the numbers of people dying is unlikely to go unnoticed.  It's more reliable, and is the whole point of social distancing and closing the economy.

https://covidtracking.com/data/us-daily
I'm using data from the above website, dumping it into a spreadsheet, and performing division.  For week over week, I divide the 4th row by the 11th row (7 days apart), etc.  If you dump that data into "Google sheets", and use some columns to the right of the data, you should see the same numbers I copied above (which are rounded).

Back when I was aggressively market timing, I even calculated New York and New Jersey separately from the rest of the U.S.  About 1/3rd of U.S. deaths occurred in New York, so it's numbers can overwhelm data from other states.  It's probably not breaking all 50 states into their own data set, but separating some of the worst hit states might be more precise that what I've done here.

[MustacheAndaHalf]

Because I don't have a good idea what happens next, I'm going to put my portfolio in neutral.  I see two possibilities: a steep decline, but I'm unwilling to go below 60% equities.  Or a quick recovery while I'm not paying attention, so I want higher equity exposure.  Sounds contradictory, but I have an idea.

I roughly plan to have a portfolio like this:
30% cash
30% active picks
25% passive index funds
15% daily 3X leveraged ETFs

Best case first: the leveraged ETFs don't suffer volatility or massive declines, and my portfolio acts like it's 100% invested (the 15% leverage acts like 45% equities, making up for 30% cash).

The average case: the 3X leverage probably drops to 2.5X short term, and can get worse with time.  So roughly 92% equities because of imperfections with leveraged ETFs.  I attempt a risk management strategy to exit leveraged ETFs after significant (but not total) losses, limiting the damage to that 15% of my portfolio.  Active picks profit more than they default.

The worst case: the stock market drops another -50%, sending most of my active picks into bankruptcy, and 20-30% of my portfolio is a total loss.  I try to sell off the leveraged ETFs, but I'm too slow and the leveraged ETFs lose -99% of their value, and then liquidate.  So another 15% of my portfolio is a total loss.  The passive index funds drop -50%, and the cash doesn't drop at all.  Overall, I have about 43% of my original portfolio after the -50% drop.


Leveraged ETFs are dangerous and require a lot more explanation than I can provide right here.  But just to reveal some assumptions I had to lose while learning  about leveraged ETFs:
* volatility causes losses.  When an asset experiences a 0.9x and then 1.1x multiplier, that fund has lost -1% (0.9 x 1.1 = .99).  As those swings get larger (0.7x then 1.3x = 0.91) the losses get larger.
* leveraged ETFs do not revert to the mean.  Their underlying index can recover, and the path of the leveraged ETF can leave it with nearly random performance.  leveraged ETFs are designed for day traders, and only work 3X during a given trading day.
* many leveraged ETFs have been liquidated recently.  After losing -99% or -96% of their value, when they can no longer continue, they can return cash to the former leveraged ETF investors.  Being left with cash at a market bottom, and no vehicle to profit on the recovery, is a risk of leveraged ETFs
* some ETFs have dropped their leverage from 3X to 2X.  The cost of leverage got so high, they could no longer continue at 3X leverage.  That means 3X leverage on the way down, for the losses, and then 2X leverage on the way back up, for the recovery.
* I have a risk management approach that can work with margin, but probably fails with leveraged ETFs (they change too quickly to act).  I'm unclear if I can manage the risks of leveraged ETFs losing most of their value, and I have time before I ramp up to the allocation shown above.

At the time of this post, I have under 1% leveraged ETFs in my portfolio.  If I cannot figure out a risk management approach, I might decide I can't do 15% leveraged ETFs.

[BicycleB]

just to reveal some assumptions I had to lose while learning  about leveraged ETFs:
* volatility causes losses.  When an asset experiences a 0.9x and then 1.1x multiplier, that fund has lost -1% (0.9 x 1.1 = .99).  As those swings get larger (0.7x then 1.3x = 0.91) the losses get larger.
* leveraged ETFs do not revert to the mean.  Their underlying index can recover, and the path of the leveraged ETF can leave it with nearly random performance.  leveraged ETFs are designed for day traders, and only work 3X during a given trading day.
* many leveraged ETFs have been liquidated recently.  After losing -99% or -96% of their value, when they can no longer continue, they can return cash to the former leveraged ETF investors.  Being left with cash at a market bottom, and no vehicle to profit on the recovery, is a risk of leveraged ETFs
* some ETFs have dropped their leverage from 3X to 2X.  The cost of leverage got so high, they could no longer continue at 3X leverage.  That means 3X leverage on the way down, for the losses, and then 2X leverage on the way back up, for the recovery.

@MustacheAndaHalf, how did you lose these assumptions? Did you mean use these assumptions?

If you are willing to continue sharing your thoughts:

Why did you intend your neutral portfolio to include 2X and 3X leveraged ETFs instead of using the type of options @ChpBstrd has been considering?

Now that you're wary of the leveraged ETFs (I suspect wisely), would the options help you achieve your goals?

***

It may be too late for mere data to give you an edge. But ICYMI, here's concrete data relating the number of known COVID deaths to the overall number of excess deaths during the COVID era, based on sources in US states per researchers at Yale School of Public Health.

https://www.washingtonpost.com/investigations/2020/04/27/covid-19-death-toll-undercounted/?arc404=true&utm_campaign=wp_post_most&utm_medium=email&utm_source=newsletter&wpisrc=nl_most

[MustacheAndaHalf]

BicycleB - That was confusing of me, but I meant I had lost certain assumptions, but had discovered that list of characteristics of leveraged ETFs.  There wasn't space to mention options.

I believe people have layers to their understanding.  My most basic understanding of the risks of options is that most of them expire worthless.  I looked online to rebut that basic understanding, and I found this:

https://www.stockoptionschannel.com/slideshows/seven-myths/most-options-expire-worthless/
"So if 10% of options contracts end up being exercised, and 55-60% get closed out before expiration, that leaves only 30-35% of contracts that actually expire worthless."

I read that as 75% of options expire worthless (10% vs 30%), so I should be cautious.  If I predict that COVID-19 will be resolved in 6 months, I might buy options for 6 months from now.  If I turn out to be wrong, and COVID-19 lasts longer, I lose -100%, just like most purchases of options (who aren't closing their position).


My take away is that the odds are against me with options, that maybe 75% of the time when I'm not creating and closing my own option, an option expires worthless (10% vs 30%).  So when I consider options, I need to have that in mind.  If the situation lasts longer than I expect, using options can result in -100% loss.  So I'm wary of using options for speculation, since I can't accurately predict the timing - and that lack of accuracy can lead to -100% loss.


But there is one area I looked into - can I get insurance for my actively picked stocks.  In "An experiment" thread, I revealed Macy's ("M", $5.50/share now) is one of my active stock picks.  What do the options look like for Macy's?  I want to buy "put" options that profit if Macy's goes to $0 (risk of bankruptcy).
https://finance.yahoo.com/quote/M/options?p=M&date=1642723200

Macy's is trading at $5.50/share.  2022-Jan put options at $5.00/share (not $5.50) for Macy's cost cost $1.91, or 35% of the current share price.  If the stock goes nowhere or goes up, the option expires worthless.  If it defaults, after the expense of this option I still lose -44%.

I could buy 2022-Jan put options at $10.00/share for $5.58, which are already worth ($10 - $5.58 = $4.42) something.  I pay roughly 21% for this, since it costs $5.58 but is already worth $4.42.  But it's a deep short against Macy's, so if Macy's doubles during a recovery the option expires worthless.  If Macy's bankrupts, this option pays $10 and cost $5.58, for a roughly +80% profit.

Finally, 2022-Jan put options at $3.00/share costing $0.90.  These options only kick in after a -45% drop!  They cost 16.4%.  In theory I could buy twice as many, or pay $1.80 to get $6/share on bankruptcy... that's 33% cost but I would get (6/5.50 = 1.09) an extra +9% on bankruptcy.


Looking at options for insurance, I view them as too expensive.  Except for people who open and close their own contract, 75% of options expire worthless.  If COVID-19 lasts 20 more months, and stock values don't change, all the 2022-Jan options expire worthless.  So to me, options are risky and expensive.

Looking back, if I predict one week at a time, options would be a great way to do that.  They effectively give far greater leverage, but at the price of going to -100% after a limited time period.  If I had an options account, I would have bought options March 19-25, when data showed testing would far surpass COVID-19 cases.  I imagine with high volatility, options cost more, but I would have looked at options vs margin vs leveraged ETFs.

[MustacheAndaHalf]

Options were a post by themselves, and so are leveraged ETFs.

UPRO is a 3X S&P 500 ETF, and VOO is Vanguard's S&P 500 ETF.  Looking at morningstar data, here's the 2017 and 2018 performance for both ETFs:
VOO : +21.77% then -4.50%
UPRO: +71.37% then -25.15%

Although UPRO uses 3x daily leverage, for 2017 it's performance was a 3.3x greater gain, and in 2018 a 5.6x greater loss.  When both ETfs swing through +0%, they have very different volatility, and that volatility has a big impact.  Although I'm viewing assets as reverting to the mean, that's not true of leveraged ETFs: they trace a 3X path determined by their underlying index.  That's why they can be 5.6x for a loss and 3.3x for a gain.

I ran an analysis by putting 2020 prices into a spreadsheet, and comparing multiples at various points in time.  I should probably expand that beyond 2020, but that helped me learn about leveraged ETFs.  I looked at how far the underlying index would move if it reverted to the mean (reverting back to it's Jan 2 price, and reversing YTD losses).  I then multiplied that performance by the recent multiple shown in my spreadsheet to get an expected recovery for leveraged ETFs.

Roughly speaking, it was half or less of what you'd expect if you think leveraged ETFs revert to the mean.  Plug in some dates and prices, and compare UPRO and VOO, and you'll see the multiples for weeks or months do not match the daily leverage of these ETFs (they are intended for day traders, who only buy and sell these ETFs during trading hours, and don't hold them overnight).

Since the performance for various leveraged ETFs over time was much closer than I naively expected, I modified my plan.  I aimed for things like the S&P 500, or developed markets.... not the specialty ETFs that have far higher risk but similar returns (in my view).  I should probably make another pass, looking at 2018 data and volatility of each index.

Many leveraged ETF have been negatively impacted by events of the past 4 weeks.  For those speculating on the price of oil, several leveraged oil ETFs closed.  To be clear, these ETFs lost -95% or so, and then handed back the remaining 5% to investors as cash.  Meaning, you could only speculate on the drop in oil prices - not on profiting when they go back up!  But it gets even better - the leveraged ETFs that short the price of oil can also go broke in the same situation!
https://www.bloomberg.com/news/articles/2020-04-22/triple-leveraged-oil-product-shutters-in-wake-of-historic-crash

Another few ETF couldn't afford leverage, and switched from 300% leverage to 200%.  Which means their "3X daily" leverage became "2X daily" without necessarily notifying everyone directly.  So it's possible to take losses of 3X the underlying index, and then recover with only 2X the gains.
https://www.bloomberg.com/news/articles/2020-03-23/ten-leveraged-direxion-funds-cut-exposure-amid-global-turmoil


Personally, I devised an approach to avoid total loss with margin loans - you sell when the losses are greater than expected, but before they reach crash levels.  Accept the loss, sell and pay down the margin loan.  For example, during my margin loan I planned what would happen if markets were going to lose -50% of their value.  As stocks were down -20% to -33%, I planned to sell and pay down the margin loan.  Because of some front loading, at -25% losses for the market I might have sold and paid off half my margin loan, and then completely sell and exit margin by -33%.  I could even have gone through great depression (-90%) level drops provided I exit my margin based on the level of losses.

Normal markets have circuit breakers, but that doesn't apply to leveraged ETFs.  I've seen a leveraged ETF fall -30% in days, suggesting I'd have to monitor it constantly.  Personally, I don't want to be "on call" for my leveraged ETFs, so I have to accept they might effectively be a total loss.  If I can implement some software triggers on Interactive Brokers, I will - but I'm doing something unusual, and I don't know if anyone has tried it.

Let's say I buy UPRO.  If UPRO's price drops -20%, I actually want to begin selling UPRO shares, and sell them all before losses are -33%.  So far, that's just standard stop-loss orders.  When markets go back up, I want to buy UPRO again... but only if I've already sold.  I need my PRO position to be sold between -20% and -33% losses, and purchased again as it recovers back from -20% to -33%.  Since later purchases depend on earlier sells, it's probably tricky for an ordinary customer to specify it.  I don't know where that will ultimately lead - hopefully I don't need to hire a high frequency trading firm to manage my leveraged ETFs!  :)

Don't forget - leveraged ETFs can lose 5.6x or even drop -95% and then liquidate, leaving you holding cash.  I'm using them as being cheaper than margin, and without the margin calls.  I prefer leveraged ETFs over options, because I don't have an expiration date that causes -100% losses.  But I may change all of this before I actually invest more than 1% of my portfolio in leveraged ETFs, so buyer beware.

[ChpBstrd]

Yes, options are expensive right now. The time to buy insurance is before one's house is on fire.

I've held puts as insurance for my stock positions for most of the past 2 years and they've done a great job of reducing portfolio volatility to bond-like levels and keeping my chickenshit self invested in a stock-heavy portfolio during the Trump administration. However, the key was buying them when volatility was low - as in VIX less than 16. Today VIX is about 35, which makes it a bit late in the game to buy insurance - but a decent time to sell it.

One way to take advantage of the high cost of puts and calls (both are high because of put/call parity) is by setting up covered call positions. For example, you could have bought prison company GEO for $12.15 today and sold the Jan 12, 2021 call at the $13 strike for $1.98. GEO would have to fall another 16.3% for you to hit break even and start losing money alongside the stock. There's a chance your shares would be called away in January, leaving you with $13 in total cash for your trouble, a 27.8% profit in 9 months on a $10.17 investment. That 27.8% profit occurs if GEO rises just 7% by Jan 12. My random example is not even that volatile a stock, with a beta of 1.08.

Edit: To be clear, this is a bullish bet, despite the hedging. I will be considering moves like this at a future time, but I'm a bit pessimistic to do it now.

[MustacheAndaHalf]

My new goal runs a bit contrary to the thread: I plan to stop investing in next week's news, but for good reason: I lack insight against the market.  Denmark has opened schools, and week/week growth in cases is still dropping.  Denmark has ~150 new cases a day (9k total cases), and plans to have 20k tests/day.  The U.S. has about 100x more cases per day and 10x more testing... so the U.S. might be able to reopen slowly, but will be in a riskier position than Denmark.


I wanted to figure out if I actually accomplished something with all this work of marking timing.  Each move was profitable, but wasn't my entire portfolio.  So I created a benchmark that's appropriate for my portfolio:
1/3 Vanguard Total Stock Market ETF (VTI)
1/3 Vanguard Total International ETF (VXUS)
1/3 Vanguard Tax-Exempt Bond ETF (VTEB)

Right now that benchmark has dropped -10.5% since the last day's market close of 2019.  Even though I moved cash out of my portfolio for a larger emergency fund, my portfolio is -3.1% right now compared to end of 2019.  I'm glad all the effort I put in lead to beating the market by over +7%, but keep in mind these are just made up numbers from someone on the internet, and I won't be providing data to verify these numbers.  "An experiment" has 3 stocks I've revealed publicly, and is where I'm tracking performance in a controlled manner in public view.

There may even be hope for risk management of daily 3X leveraged ETFs.  Apparently Interactive Brokers has a software API, which allows the scary possibility of issuing trades using software.  So selling off all ETFs and then buying them again - as part of risk management - might be possible in an automated manner.


At some point, I need to try and calculate possible negative scenarios.  What happens if COVID-19 continues for another year, with no vaccine?  What if COVID-19 is cured, but there's a big financial mess with unpaid rents and mortgages?    Could Congress open up gaps in the financial relief, and bankrupt large numbers of businesses (especially after being re-elected)?

[MustacheAndaHalf]

Apparently Dr Fauci will speak later today, possibly about FDA approval for medical trials of Remdesivir.  The news is mixed, but U.S. medical trials are probably more trustworthy than China's.
https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)31022-9/fulltext
https://www.politico.com/news/2020/04/29/gilead-coronavirus-remdesivir-220484

My lay understanding is it might prevent half of deaths in non-placebo controlled trials - might.  Today's jump of +3% so far seems desperate to me, and might go up more after Dr Fauci speaks.  The current global pandemic is being rated as a -9% problem for the U.S. stock market... we'll see.  Maybe I'm missing something else.


I dumped granular data on cities and counties into a spreadsheet, and aggregated by each state, by each day.  That allowed me to get week over week growth for each state.  Meaning I divide Monday's cases by last Monday's cases.  Here's how that calculates out the past 3 days for selected states:

New York:  18.35%   17.54%   16.40%
Michigan:  20.22%   19.41%   20.19%
New Jersey:  27.83%   25.32%   23.24%
Georgia:  28.33%   25.22%   25.36%
California:  38.58%   33.49%   30.17%
Kentucky:  38.01%   36.26%   36.55%

With President Trump and the Mayor of Atlanta both criticising Georgia's governor, I think most people will get the message that the governor is making a mistake.  With 600 cases/day now, up +25% since a week ago, it seems risky... and some people have been without income for awhile, and may be willing to risk it.


ChpBstrd - In an earlier post you mentioned AIG and 2008, and I wanted to point out AIG's situation was caused by greed, fraud and leverage.  Mortgages were very tiny compared to the financial damage of 2008 - mortgages by themselves can't cause that damage.  Greed and fraud pushed bad mortgages into the system, then greed and fraud rated those mortgages AAA quality... and finally greed and fraud leveraged it all into financial instruments (credit default swaps, synethetic CDOs, and others).  Those elements, to my knowledge, are not present in the COVID-19 crisis.

[ChpBstrd]

ChpBstrd - In an earlier post you mentioned AIG and 2008, and I wanted to point out AIG's situation was caused by greed, fraud and leverage.  Mortgages were very tiny compared to the financial damage of 2008 - mortgages by themselves can't cause that damage.  Greed and fraud pushed bad mortgages into the system, then greed and fraud rated those mortgages AAA quality... and finally greed and fraud leveraged it all into financial instruments (credit default swaps, synethetic CDOs, and others).  Those elements, to my knowledge, are not present in the COVID-19 crisis.

One person's greed, fraud, and leverage is another's reach for yield. Prior to 2007, mortgages had never defaulted at the levels seen shortly thereafter, so people's ideas of what default rates were possible were constrained to the range of historical experience. We see the same historical anchoring on this board when people point out that stock indices have never lost value in a 20 year period, implying they will never do so in the future.

Today's reach for yield is represented as historically low bond yields, low spreads between junk and IG bonds, high stock PE ratios, high corporate leverage and interest rate sensitivity, and real estate so expensive 70% of Americans cannot afford the median priced family home. Will one of these investments blow up and, with the benefit of hindsight, be described as a moral failing?

The derivatives market is bigger than ever, and a large portion of the financial economy (including mortgage holding) has moved out of publicly traded banks and into less transparent entities - entities that may not fall under the safety systems applied to big banks. The first big financial institution to blow up might be something we've never heard of - a hedge fund perhaps - and they will fail outside the protective measures used by the SEC, FDIC, etc. If an unregulated market freezes up, there is no Treasury

I agree the bad mortgages of 2007 were not big enough in themselves to tank the US economy. They tanked the US economy because the financial system was fragile and suffered secondary effects over the next 12 months.

[MustacheAndaHalf]

Why did mortgage defaults cause -50% stock price drops around the world?

https://www.thebalance.com/what-caused-2008-global-financial-crisis-3306176
"The financial crisis was primarily caused by deregulation in the financial industry. That permitted banks to engage in hedge fund trading with derivatives. Banks then demanded more mortgages to support the profitable sale of these derivatives. They created interest-only loans that became affordable to subprime borrowers."

Using rough numbers to show how it happened, a $100k mortgage won't collapse the world's financial system.  But that mortgage can appear in 50 CDOs, which means it then impacts $5 million worth of investment.  But those CDOs are then repackaged in synthetic CDOs, making that original $100k mortgage able to impact $250 million worth of financial products.  Investment banks made fees for selling the $250 million worth of products, so to them a $100k loan meant millions in fees.  Even the greed required leverage, since investment banks didn't want to sell individual mortgages.

Without leverage, there's no 2008.  Doesn't matter if people are greedy or chase yield, because the financial damage would be 50 x 50 = 2,500x less.  All that leverage meant stock markets took a -50% hit with it being unclear which banks were solvant.  Eliminate that leverage, and stock markets don't even see a -1% impact.  That's why 2008 was all about leverage.


There's a study from about 1.5 years ago asking what conditions at the end of 2018 were similar to those in 2008.  It does show some similar data points (GDP loss, unemployment), so maybe it's worth comparing the events based on damage rather than specific causes.
https://irle.berkeley.edu/what-really-caused-the-great-recession/

[BicycleB]

My new goal runs a bit contrary to the thread: I plan to stop investing in next week's news, but for good reason: I lack insight against the market. 

That's very wise! It's really important to separate the rare moment of having useful insight from the many moments of not.

(pregnant pause) But I came here to share a thought that might follow a pattern of we-can-reasonably-predict-a-surprise that got you started. It's basically that as America "reopens", there's a predictable gap between when cases rise and when the news appears, much like the beginning of your journey.

Reasons:
1) After several weeks of lockdown, the real cases in contagious mode are probably lower than at the start of lockdowns
2) Symptoms take a couple weeks to appear, often
3) Deaths lag infection by maybe 3 weeks. So:
4) Some drops in testable cases are en route to appear in the next couple of weeks
5) As America reopens and apparent cases moderate or fall, people will start thinking it's safe
6) R will now be above zero, but its effects scarcely visible during the first weeks of reopening
7) Through late May or so, hidden contagion will grow but low cases will allow enthusiasm to build
8) By the time cases and deaths are visibly growing, it will be fashionable to mock the scolds who want a return to lockdowns
9) By then (perhaps early June), actual cases will have multiplied hugely
10) By late June, enough hospitalizations and deaths to swamp many hospitals will be sweeping numerous locations
11) Big swing to pessimism once the visible disaster shatters the moment of optimism

I admit it's a detailed scenario, and relies on people's "instinctive" guesses about safety having a different timeline than the disease itself, along with the confusion caused by testing still being patchy and inadequate. Those look like reasonable assumptions to me, though of course the uneven process of opening could muddy the timelines.

Is this a reasonable scenario?
Likely enough to put money on?
If yes to both, what are good ways to apply it?

Examples:
If you sell stock for living expenses, wait until late May; sell enough for several months
Heavy equity % until late May, then sell
Wait until late May, then buy options that benefit from a falling market?

[SAR]

I'm also in the 'too uncertain to act' category right now. I got lucky with picking the recent bottom, and can't stand the though of getting reamed on short-term capital gains.

Having said that, I suspect that re-opening will be gradual, particularly in places hard hit by the virus. And I also suspect that coronaviruses don't generally like summer. It appears likely to be seasonal.

If that's true, then the U.S. and Europe at least will see a sustained improvement, but the virus will hit the southern hemisphere and if it hits hard, then there will be stock market pain then. However, it is likely limited because there are no massive economies to bring the system down.

That leads me to think that there may be something of the order of another 10% drop, but a return to upward moves.

In the interim, I'm expecting lots of +/- 2 and 3% days.

[MustacheAndaHalf]

BicycleB - I would encourage you to look for evidence that confirms or denies your assumptions.  Normally I would just speak my opinion based on information I've learned.  But for data driven market timing, I need to use high quality information when I act contrary to most investors.  Poor assumptions could lead to most people being right, and my investments being a mistake.  All of that hinges on my data (or analysis) beating the market's.... so I encourage others to view it that way, to test yourself.  Like an experiment, where ego is set aside, and you look at evidence for and against.

(2) The information I read calculated an average of 5 days for symptoms to appear.  I try to only read reputable medical journals (The Lancet, NEJM, JAMA) about this data, and I don't see any recent studies except for the earlier ones in China.
(7) can a hidden spread happen when testing is sufficient?  Over the past 3 days the U.S. performed 623k tests and confirmed 74k new cases, a ratio of about 10:1 which keeps getting better as cases fall and testing slowly increases.  Republican and Democratic governors (on Meet the Press, Sunday) both said current testing is not enough for contact tracing.
(8) Most people are scared, according to polls.  The news can give an impression that doesn't reflect actual numbers, like reporting on 200 protesters in Michigan.  I didn't see much reporting about the millions in Michigan who quietly stayed at home.

Testing is increasing (slowly) while new cases are dropping (see my per state data, above, which was calculated off John Hopkins data).  Outbreaks will get easier to spot because of those trends.

I would speculate the biggest weakness in what I know, which could provide support for your thesis, is undocumented/illegal (pick your political party!) immigrants.  They can't afford a hospital stay, and can't afford to be off work.  There's a systemic motivation for people in that situation to avoid testing and get back to work, making outbreaks in that population harder for experts to detect.


SAR - Back when I heard about viruses and summer, I didn't explore further.  I figured the outbreaks in Germany and Iran showed a wide temperature difference.  I'd welcome data that sheds light on that information - about MERS, SARS or others.


Both last week and this week, markets bounced up on news of Remdisivir.  Last week a poorly controlled study showed half expected deaths - but just 2 instead of 4.  This week, hospital stays dropped from an expected 15 days down to 11 days, but again there was no placebo controls.  Markets are down -9% year to date on this news.

Europe is farther along with lock downs and reopening.  I looked at U.S. markets, Europe, and all non-U.S. markets.  All show roughly +4% gains for the past five days.  That means whatever is moving markets upwards is not specific to Europe.  So I think that rules out markets pricing in Europe's reopening, since all markets are equally impacted.

High quality studies randomly decide who gets a drug, and who gets a sugar pill (a placebo).  A central team allocates them, and sends the drugs / placebos to doctors, who administer them.  It's "double blind" because the doctor administering the drug has no idea which drug they are giving.  The patient can't read their expression, and the doctor can't secretly tell them - they both don't know.  So I think it will be more interesting when randomized, double-blind, placebo-controlled medical studies are available for Remdesivir.

To me, a drug that lowers the odds of dying by half is welcome news, but I'm not seeing why markets should be -9% YTD.


I just updated my state by state data, and Denmark data.  Denmark continues it's week over week decline, despite being open for ~10 days so far.  The data for most states isn't updated yet in the latest download, so I'll check back later.

[ChpBstrd]

Is this a reasonable scenario?
Likely enough to put money on?
If yes to both, what are good ways to apply it?

1) Reasonable, yes.
2) Very little money, if any.
3) A calendar spread using options would be the direct way to trade this prediction, if you held the back end option after expiration of the front end option and the reversal occurred right around expiration of the front end option.

Having said that, I suspect that re-opening will be gradual, particularly in places hard hit by the virus. And I also suspect that coronaviruses don't generally like summer. It appears likely to be seasonal.

This is speculative. The virus is stable, of course, in bodies with temperatures of 98.6 Fahrenheit. However, it spreads via respiratory droplets that change size based on temperature and therefore may have different contagion profiles (as with flu). If you listen to Trump, warm temperatures will end this whole thing:
https://www.marketwatch.com/story/trump-says-coronavirus-could-be-thwarted-by-summer-heat-citing-dhs-study-2020-04-23

But if you look at the spread in relatively hot countries like Pakistan, India, Iran, Mexico, Indonesia, etc. you see a robust rate of spread. Locations with the fastest rate of spread include Liberia, Sudan, Somalia, the UAE, Kuwait, and Nigeria.
https://ourworldindata.org/coronavirus

And finally, in the US, Florida and Texas are emerging as outbreak hot spots, and they have had the sort of weather that counts as summer for most of the U.S.

I would speculate the biggest weakness in what I know, which could provide support for your thesis, is undocumented/illegal (pick your political party!) immigrants.  They can't afford a hospital stay, and can't afford to be off work.  There's a systemic motivation for people in that situation to avoid testing and get back to work, making outbreaks in that population harder for experts to detect.

The spread will be worse among low-income people without the means to go four+ weeks without a paycheck while they recover from a possible C19 infection. They have no choice but to get back to work as soon as they can pass for not having a fever. The demographics (and politics) of this virus are shifting. It initially spread among the wealthy and well-connected along airline routes and in countries with well-developed healthcare systems. Now it is spreading among the poor, who do not have the same access to healthcare. The death rates could rise as the virus reaches more people who don't have insurance or cash (primarily in the US and other developing countries without universal payment systems). This will, of course, color the political debate.


 

[BicycleB]

Good comments, @ChpBstrd (and MustacheAndAHalf). 

Hadn't thought consciously about the class difference part. That's a huge thing. Already, the pressure is mounting for laid-off lowly workers to get back to work, and societal threats are being made that they will lose benefits if they don't. I know people facing that very scenario.

[MustacheAndaHalf]

The data for most states isn't updated yet in the latest download, so I'll check back later.

I was wrong - the data for California was present, but I wasn't adding it up.  Now fixed.

ChpBstrd - Here's data from some cold states (North Dakota, Maine) and the warm states you mentioned (Florida, Texas).  This shows the last 5 week over week totals for those states:

N. Dakota...   52.08%   48.21%   50.24%   53.88%   52.14%
Texas...   29.13%   29.63%   28.20%   28.11%   27.84%
Florida...   20.98%   19.83%   18.77%   17.87%   17.25%
Maine...   13.93%   17.07%   16.91%   17.12%   16.43%

Since east-west seems to matter more than north-south, that lends more weight to warmer states not having an advantage over states with lower temperatures.

BicycleB - There's various articles on racial disparities in COVID-19 deaths, but this one packs the most data per sentence.
"The latest available COVID-19 mortality rate for Black Americans is 2.5 times higher than the rate for Latinos, 2.6 times higher than the rate for Asians, and 2.7 times higher than the rate for Whites."
https://www.apmresearchlab.org/covid/deaths-by-race


I'm keeping an eye out for negative news in the oil markets, especially related to OPEC.

"The good news is that the OPEC+ supply cuts lower the immediate risk of a disaster in the oil market"
"OPEC and its allies are notorious for "cheating" by producing more than they say they will."
"Rystad Energy recently estimated that 140 US oil producers could file for bankruptcy this year if oil stays at $20 a barrel, followed by another 400 in 2021. That would cause countless jobs to disappear."
https://edition.cnn.com/2020/04/13/business/oil-prices-opec-deal-trump-putin/index.html

Note most people expect that when COVID-19 disappears, the economy starts up quickly and people get their old jobs back.  That's not true of oil/shale companies that go bankrupt from low oil prices, which is why I think markets react more to news of job losses in the energy sector.


I'm still looking for a historical comparison to the U.S. economy being shut down.  I imagine the closest was WWII, when many factories shifted to wartime production.  Stocks rose a year after victory, then plunged a year after that.  Overall, US stocks were break even 5 years after WWII ended, and then started rising again.  An CNN business article makes a similar comparison:

"It was the largest drop since January 1946, when American industries were shutting to shift production from wartime to consumer products following the end of World War II."
https://edition.cnn.com/2020/04/15/business/factories-industrial-production-coronavirus/index.html


A hypothetical scenario, everyone has been vaccinated against COVID-19, and the U.S. economy restarts.  Most people go back to work, but unemployment is still significant.  If demand is down, why would for-profit businesses hire back all their workers?  As a group, I think businesses wait to hire people back, which is their best choice individually.  But as a group, they are all hurting demand - people without a job don't spend as much.  So it's a problem - a lower demand from so many unemployed, and businesses not ready to hire, because of lower demand.  I wonder if something similar happened right after WWII?

[SAR]

More evidence is emerging on the seasonality/temperature dependence of covid spread:

The early data is focused on China, and early on. The more recent data is doing country-level comparisons, and controlling for confounds.

I think the virus does appear to be seasonal, and I do think we will see the northern hemisphere drop off in cases, but the southern will see big increases, unfortunately.


Jan 20 to Feb 4 in China:

https://www.medrxiv.org/content/10.1101/2020.02.22.20025791v1


Jan 20 to Feb 29, China:

https://www.sciencedirect.com/science/article/pii/S0048969720317393


Cross country tests show fit for temperature and covid in northern hemisphere countries:

https://arxiv.org/abs/2003.12417


Another cross-country comparison, again northern hemisphere, with controls for wealth, population density, and population size:

https://www.medrxiv.org/content/10.1101/2020.03.23.20040501v3

This paper looks at causal relationships in 134 countries from January 22, 2020 to March 15, 2020. Hypothesis confirmed again. Their model predicts a large drop in cases in northern hemisphere, and a large uptick in the southern hemisphere:

https://www.medrxiv.org/content/10.1101/2020.03.26.20044420v1

Finally, a systematic review argues that the evidence is far from conclusive, but the findings to suggest that the virus is more likely to spread in cold/dry than warm/wet regions.

https://www.medrxiv.org/content/10.1101/2020.04.14.20064923v1

If covid is like the flu and other viruses, it will be seasonal. So far the pattern of cases for covid is the same pattern as flu, albeit at much higher rates in some areas (last I checked, 57% of all U.S. cases are NY, NJ, and MA).

Also, temperature models do not fit as well for equatorial regions for the flu. There are no seasons there, of course. But wet, humid weather predicts incidence of flu. Sorry, I don't have the reference on hand.

Anyway, assuming this is all true, I think it helps us predict something of how the future will unfold in the markets. If the U.S. gets to miss a second spike because of hotter weather, the market will respond favorably. The southern hemisphere is likely to get hit hard.

So far the Australian, South African, and New Zealand politicians are bragging about how great their response has been. It's probably not that. It's just the bloody weather.

[ChpBstrd]

Note most people expect that when COVID-19 disappears, the economy starts up quickly and people get their old jobs back.  That's not true of oil/shale companies that go bankrupt from low oil prices, which is why I think markets react more to news of job losses in the energy sector.

...

I'm still looking for a historical comparison to the U.S. economy being shut down.  I imagine the closest was WWII, when many factories shifted to wartime production.  Stocks rose a year after victory, then plunged a year after that.  Overall, US stocks were break even 5 years after WWII ended, and then started rising again.  An CNN business article makes a similar comparison:

"It was the largest drop since January 1946, when American industries were shutting to shift production from wartime to consumer products following the end of World War II."
https://edition.cnn.com/2020/04/15/business/factories-industrial-production-coronavirus/index.html

A hypothetical scenario, everyone has been vaccinated against COVID-19, and the U.S. economy restarts.  Most people go back to work, but unemployment is still significant.  If demand is down, why would for-profit businesses hire back all their workers?  As a group, I think businesses wait to hire people back, which is their best choice individually.  But as a group, they are all hurting demand - people without a job don't spend as much.  So it's a problem - a lower demand from so many unemployed, and businesses not ready to hire, because of lower demand.  I wonder if something similar happened right after WWII?

Like the Great Depression, dot-com bubble, and GFC, there is no historical precedent for a pandemic that shuts down a decent percentage of economic activity. The best we'll do is look at other, very different scenarios where it was also true that there was no precedent. It appears a large number of today's market participants view the pandemic through the lens of the 1987 flash crash or the 1998 LTCM, events that had few, if any, causes in the real economy and no predecessors, but nonetheless led to a flood of liquidity from the Federal Reserve. In other words, 1987 and 1998 were buying opportunities. I personally see very little in common with these financial market panics and a pandemic that has put 20% of us out of work, sickened hundreds of thousands, and killed 60k (basically the equivalent of a nuclear attack on a city).

When I look at a long-term chart of the unemployment rate, a couple of features stand out to me:

1) Unemployment rises rapidly. This is due to companies going out of businesses and doing mass layoffs when recessions hit. This happens all at once.
2) Unemployment falls slowly. This is because when unemployment is high, there is insufficient demand to justify more jobs. As each job is added to the economy, more demand is created, which supports yet more jobs. This takes time.

In other words, employment is a wall that must be built brick-by-brick, but comes tumbling down all at once. Economic growth builds on itself, but so does the destruction of economic capacity.

https://fred.stlouisfed.org/series/UNRATE/

I see this as a problem for the real economy, not the result of an over leveraged investor blowing up or a computer glitch. Those small businesses won't be back for a long time.

[beltim]

Cross country tests show fit for temperature and covid in northern hemisphere countries:

https://arxiv.org/abs/2003.12417

While this shows a statistically significant effect of temperature, temperature is only sufficient to explain 10-20% of the variance of the transmission rate according to this paper.

Another cross-country comparison, again northern hemisphere, with controls for wealth, population density, and population size:

https://www.medrxiv.org/content/10.1101/2020.03.23.20040501v3

This paper uses data mining to test dozens if not hundred of different models before using one that uses I think 6 parameters, in order to develop a model that explains 44% of the variance in growth rate.  Oh, and it says that population density does not play a role in determining growth rate, which is counter to everything we know about the epidemiology of infectious diseases.

This paper looks at causal relationships in 134 countries from January 22, 2020 to March 15, 2020. Hypothesis confirmed again. Their model predicts a large drop in cases in northern hemisphere, and a large uptick in the southern hemisphere:

https://www.medrxiv.org/content/10.1101/2020.03.26.20044420v1

This one doesn't even have an R² so the authors don't even give the audience a guess of how much their model explains.

COVID-19 may very well be seasonal but these drafts are in no way any sort of proof of significant seasonality.

[SAR]

I agree, the data is hardly conclusive, but there are significant associations that should be getting attention.

There is massive variance in the rates of cases and deaths, and that does correspond to temperature data. As I stated, more than 50% of all U.S. deaths are NY, NJ, and MA. All places that are cold. Likewise in Italy--the north got hit way harder than the south. Spain had unseasonably cool weather when it hit there.

Also, the seasonality measures do not apply to the tropics where there are no seasons. Flu shows increased rates in the tropics after rain, apparently. But I do not know how big the effect size is there. i suspect rather small.

Also keep in mind that these are pretty gross measures. The virus is transmission being correlated with mean regional temperatures is a broad brush, so we should not expect variance explained to be extraordinarily high, as there are clearly other variables that should contribute to the models.

Other than that, the study that "doesn't have r-square", that's because the did Poisson regression for count data, where there are no r-square measures. Instead they express the effect sizes in terms of percentage drop off- and increases in predicted cases.

My read of these studies is that there is almost certainly a temperature dependence in the data, but that we almost certainly cannot use the data to confidently predict the future course in terms of what will happen in the northern hemisphere winter. It may be that is it pretty well done this round, with a small return in the winter. Or it could be that winter turns out to be worse than what we have now.

Obviously if that happens, the economy and markets are going to tank big time.

[beltim]

I agree, the data is hardly conclusive, but there are significant associations that should be getting attention.

There is massive variance in the rates of cases and deaths, and that does correspond to temperature data. As I stated, more than 50% of all U.S. deaths are NY, NJ, and MA. All places that are cold. Likewise in Italy--the north got hit way harder than the south. Spain had unseasonably cool weather when it hit there.

Also, the seasonality measures do not apply to the tropics where there are no seasons. Flu shows increased rates in the tropics after rain, apparently. But I do not know how big the effect size is there. i suspect rather small.

Also keep in mind that these are pretty gross measures. The virus is transmission being correlated with mean regional temperatures is a broad brush, so we should not expect variance explained to be extraordinarily high, as there are clearly other variables that should contribute to the models.

Other than that, the study that "doesn't have r-square", that's because the did Poisson regression for count data, where there are no r-square measures. Instead they express the effect sizes in terms of percentage drop off- and increases in predicted cases.

My read of these studies is that there is almost certainly a temperature dependence in the data, but that we almost certainly cannot use the data to confidently predict the future course in terms of what will happen in the northern hemisphere winter. It may be that is it pretty well done this round, with a small return in the winter. Or it could be that winter turns out to be worse than what we have now.

Obviously if that happens, the economy and markets are going to tank big time.

We could quibble more about the details, but I think overall we're in agreement.  Any temperature dependence is small, and clearly outweighed by other factors.

The bigger disagreement I have is with the bolded part.  Deaths in the US are still going up, or maybe plateauing - but without the necessary measures to make the infection rate decrease.

[SAR]

Yes, I think we have the same read. I think there's likely a temperature dependence, but I wouldn't bet the farm on it on a local level, and I would expect if it is true, we should be seen trends down in the next week or two.

As for the death rates, it's very dependent on state. Plateau in many. (seems to be the modal response) Some trending up. NY appears to have the strongest trend down.

This site does a nice job animating the case and death rates by state:

https://aatishb.com/covidtrends/

[ChpBstrd]

There's also the issue of airplanes.  We know the virus spread internationally along airline routes. Quite frankly, there are a lot more people per capita flying in and out of China, the US, and Europe than there are in South America, Africa, and Southeast Asia.

The Northern Hemisphere is generally wealthier than the Southern, and therefore has a lot more air traffic. Also, the further north one goes, in general, the more wealth and ability to fly one finds. For example, New York City has a lot more air travelers than Laos, Nigeria even though Laos has 2.5x the population.

So it could be easy to mistake temperature for percentage of the population that has been on an airplane in the past couple of months.

[Michael in ABQ]

There is also the issue of demographics. Many of those colder countries in the northern hemisphere have larger cohorts of older people, the ones most likely to die. In a lot of warmer, and generally poorer countries (i.e. most of Africa, large swaths of South America, and Asia), you have the majority of the population under 25 where deaths are rare. In Africa the percentage of people over 55 is in the single-digits in almost every country. In the US 29% of the population is over 55. Places like Japan and Italy are even higher. Even China has a relatively larger cohort of older people.

[MustacheAndaHalf]

SAR - As I understand it, most of those articles are not published (they are a preview) and are not peer reviewed (nobody outside the authors has tried poking holes in study methods yet).  That said, maybe it's one more motivation for me to import worldwide COVID-19 data into a spreadsheet and aggregate by country and day.

https://www.syberscribe.com.au/blog/top-twenty-medical-journals/
I searched online to confirm my lay opinion that the following are the top (or some of the top) medical journals in the world:
1. New England Journal of Medicine ("NEJM")
2. The Lancet (based in the UK)
3. Journal of the American Medical Association ("JAMA")

A peer reviewed article from those 3 journals is the highest quality study you can find.  Note the virus (SARS-Cov-2) is distinct from the disease (COVID-19), so I searched for "SARS-cov-2 seasonal" while adding the abbreviated name of each journal.  I only got a match with The Lancet, which summarized the situation:

"Our understanding of the forces driving seasonal disappearance and recurrence of infectious diseases remains fragmentary, thus limiting any predictions about whether, or when, SARS will recur."
https://www.thelancet.com/journals/laninf/article/PIIS1473-3099(04)01177-6/fulltext

Most people won't be reading the Lancet, but they will likely listen to what Dr Fauci says about COVID-19.  In a couple articles I read, he expects a surge after summer, but I haven't seen him make a definitive prediction about summer.  He sounds more measured, like the Lancet article.


ChpBstrd - That's true of studying the outbreaks, but air traffic is currently down about 95%.  expect governors to be aware of the danger airports pose when starting to reopen.  Meaning the airports may have been a factor, and we can look to see if governors ignore them... but I expect airports will have new measures like temperature checks or wearing masks in airports.

[SAR]

I would add specialty journals from virology and epidemiology too.

The simple heuristic is to check for impact factor. As a very rough ball park (I'm a social scientist, so doesn't compare directly) I'd want to see IF > 3, and preferably a good bit higher.

The airline hub hypothesis raised is a very good one, and I'd like to see that variable controlled.

If it is, and temperature dependence is real, then we will see that other places off-the-hub get the disease later, and in good numbers, but then it goes down with the weather.

So Michigan, Illinois et al. will be interesting cases. Cool weather should mean comparatively large outbreaks, but later than places with much international travel.

[FrugalSaver]

Any data yet on air travel load factors in China?  I’ve been looking for that out of IATA but haven’t found it yet.

Are we still not flying to Europe?

[FrugalSaver]

Very much a traders market right now and everyone should be willing to experience at least 49-50% losses from here. If not you should rethink what you’re doing.

Anytime you hear fake Jews of a cure or a vaccine, that sets up a monster opportunity to short short term. Buy puts. Has worked really well twice now.

More stimulus is coming. That will only create more long term problems but is shortable when it happens

I believe we will be at or below spy 2400 by aug/sept despite all the money printing

[beltim]

Any data yet on air travel load factors in China?  I’ve been looking for that out of IATA but haven’t found it yet.

Are we still not flying to Europe?

Most of Europe is still not allowing non-residents to cross their borders through at least mid-May:
https://ec.europa.eu/info/live-work-travel-eu/health/coronavirus-response/travel-and-transportation-during-coronavirus-pandemic_en#traveladviceandbordermeasures

[MustacheAndaHalf]

Very much a traders market right now and everyone should be willing to experience at least 49-50% losses from here ... I believe we will be at or below spy 2400 by aug/sept despite all the money printing

VTI peaked at $171.32/sh and is now $141.81/sh, so you're predicting at least $72.32/sh by Aug/Sept.  What data supports your theory that markets will have "49-50% losses from here"?

[marty998]

Very much a traders market right now and everyone should be willing to experience at least 49-50% losses from here. If not you should rethink what you’re doing.

Anytime you hear fake Jews of a cure or a vaccine, that sets up a monster opportunity to short short term. Buy puts. Has worked really well twice now.

More stimulus is coming. That will only create more long term problems but is shortable when it happens

I believe we will be at or below spy 2400 by aug/sept despite all the money printing

Say what?

Autocorrect stinks sometimes ;)

[mrmoonymartian]

Very much a traders market right now and everyone should be willing to experience at least 49-50% losses from here ... I believe we will be at or below spy 2400 by aug/sept despite all the money printing

VTI peaked at $171.32/sh and is now $141.81/sh, so you're predicting at least $72.32/sh by Aug/Sept.  What data supports your theory that markets will have "49-50% losses from here"?

I'm willing to experience 49-50% losses from here. But if it's outside of that range I'm going to be pissed.

[waltworks]

I'm rooting for 99.9999999% losses so that I can purchase the entire market at once.

-W

[vand]

I'm hoping the market will go negative so they'll pay me to take it off their hands.

[Retire-Canada]

I'm hoping the market will go negative so they'll pay me to take it off their hands.

I read on the internet that this ^^ is likely to happen soon.

[waltworks]

I'm hoping the market will go negative so they'll pay me to take it off their hands.

Where are you really going to store all those hot-pants, though? Don't get burned, you'll have to figure out how to take physical delivery of, well, everything. Lots of traders end up getting caught in that bind and then suddenly they're buried in Frozen 2 figurines and bitter regret when the storage tanks run out of room.

-W

[vand]

I'm hoping the market will go negative so they'll pay me to take it off their hands.

Where are you really going to store all those hot-pants, though? Don't get burned, you'll have to figure out how to take physical delivery of, well, everything. Lots of traders end up getting caught in that bind and then suddenly they're buried in Frozen 2 figurines and bitter regret when the storage tanks run out of room.

-W

Ah, but that the beauty of paper assets. They're just digits in a computer.

[MustacheAndaHalf]

mrmoonymartian , waltworks , Retire-Canada , vand - This isn't intended as a thread for being snarky towards other posters.  Yes, an extraordinary claim does need extraordinary proof, but I think piling on isn't the point: it's to let other people test their thesis with data.  To me mocking someone like this isn't what the forum rules had in mind in stating we should "Be respectful of the site and other members."
https://forum.mrmoneymustache.com/forum-information-faqs/forum-rules/

[Retire-Canada]

mrmoonymartian , waltworks , Retire-Canada , vand - This isn't intended as a thread for being snarky towards other posters.  Yes, an extraordinary claim does need extraordinary proof, but I think piling on isn't the point: it's to let other people test their thesis with data.  To me mocking someone like this isn't what the forum rules had in mind in stating we should "Be respectful of the site and other members."
https://forum.mrmoneymustache.com/forum-information-faqs/forum-rules/

If you have read the MMM Blog humour, sarcasm and face punches for stuff that goes against MMM-ism is de rigueur. MMM is the King of Mocking in areas such as market timing, large 4x4 trucks, fancy coffees, weak humans that won't bicycle across town, etc.... the list is long. So it is both appropriate and to be expected to receive some humourous/sarcastic feedback when suggesting these practices on this forum. They do not constitute personal attacks. Nobody is being mean to anyone. They are meant to point out the silliness of the endeavour in question.

Now you might say market timing, buying a F150, buying a McMansion or enjoying a daily $12 artisanal coffee are all 100% legitimate pursuits. Sure you are free to do all those things, but here in the land of MMM they are all face punch worthy. I say that sympathetically as someone driving a F150. I wouldn't start a thread here about my truck and expect anything other than face punches and silly take downs. If I feel bad about that and want to be told how sweet my rig is I'll go find a forum with the appropriate culture like www.FORDgasGUZZLERS.com.

If this ^^^ was not true than MMM would be banned from his own site and that would just be wrong. ;-)