No, the point isn't that it gets made up, but that the loss is accounted for in the period it occurs. So if COVID-19 zeros out earnings for an entire quarter (that's roughly the impact expected), and stocks are valued on the next 10 years of earnings (in reality, this is much longer), that's 2.5% of earnings.

Likewise, even if GDP was down by 10% this year, if this doesn't change the world's productive capacity, over the course of 10 years, that's just 1%.

If you remember the context for this conversation, we're talking about this because Tooth took exception me comparing some preliminary estimates of the overall loss in wealth to the total amount of wealth in the world. He went on about how even over short periods of time, huge amounts of wealth are lost permanently. But most of this is not permanent and if you're insisting on talking about how we've permanently reduced consumption in aggregate over long periods of time, we're talking about numbers that are far larger than a given year's GDP as a baseline. What's the expected GDP over the next 10 years? Something like over 1 quadrillion right? What's a few trillion compared to that?

No one knows how big the loss is, but there is a loss, however small, and you said there wasn't. There is.

Also long term forecasts of 'growth companies' are notoriously unreliable and you give them way more influence than you should. No one has any clue how Amazon will be doing 25 years from now. But we can be more sure of how Coke or Heinz will be doing. TS is quite correct in his assessment of long term forecasts as being speculative. That's exactly what they are.

The point is that these shocks didn't have much to do with the real economy being bad. The dot com bubble crashed before the economy was bad. NASDAQ was at 5048 in March 2000 and by the beginning of April 2001, it was already at 1720 and had lost two thirds of its value. The recession hadn't started yet. The unemployment rate was 4.3%.

This supports my case, not yours. Amazon wasn't being valued on short-term earnings (what earnings) before or after the tech bubble. Their valuation didn't crater because they performed badly due to the recession, but because related assets all crashed in value.

Amazon Net Profit

98: -125M
99: -720M
00: -1.411B
01: -567M
02: -149M

Amazon Revenue

98: 610M
99: 1.64B
00: 2.76B
01: 3.12B
02: 3.93B

This is just a pure semantic argument. Your subjective enjoyment in the past is irrelevant for the purposes of predicting economic output in the future. Your past enjoyment isn't the kind of wealth anyone can use to produce more wealth in the future. Also, you're ignoring the subjective value of time as well - if we're neutrally accounting for value, a dramatic decline in hours worked should be added to the subjective enjoyment.

You're missing the point. It's not about me giving them more influence - that's how the market works. And this being unreliable doesn't mean the market estimate will change in the direction you expect. Is your thesis that if short-term earnings are bad, that will cause people to think long-term prospects for these companies also bad? In that case, why hasn't this happened already? Earnings are already expected to drop by over 40% this quarter.

So how much are GOOG/FB earnings going down by in your estimate?

Future output isn't tied to past consumption?

Please elaborate if you have a point.

But that wasn't the point you're making. The reason you said the S&P 500 will stay up was "moat" and "somewhat insulated from economic damage" and "they're valued on 10+ year earnings"

If these are true, why did these indexes drop 50% in the last two bull contractions? Your theory doesn't account for reality. Why did Amazon's valuation fluctuate 14x from highs to lows in two years?
Right. And why did all assets crash in value in the tech bubble? Why was your "market values on 10 year horizons" valuing Amazon at 2000 at 14x the value it was valuing it in 2002? For that matter, why did it do it with Microsoft, a moat company which lost over 2/3 its value? Why did this happen?

Nearly 80% drop in value. 1/5 of what it was worth just two years early. This is weighted to the biggest and moatiest companies that "the market is valuing on 10 year valuations", the same arguments you're making now.

Your theory doesn't work. It can't predict the past. My theory is that asset prices crash after shocks as the economic damage propagates and risk taking behavior runs away with its tale between its legs in a new lower energy environment (usually after a rip back to near highs after the first shock in which everyone rationalizes which this time will be different). Your theory is that this doesn't happen and that moats and 10 year valuations protect from crashing even at 25, 30, 40 P/Es and that -20% is the worst we'll see on the S&P 500.

And before you scream rates, that doesn't work. We're in a somewhat different environment, but even in 2000 the market bottom happened almost a full year after rates were aggressively cut to below 2% (P/E of 50),



So I don't think your theories of "moats" and "sector weight" and "10+ year valuations" have much bearing on what actually happens to asset values during and after recessions, especially in bubbles. And we are in a bubble. Tech is up 100% in a 1.5 years, many 500+% off lows and high P/Es even on pre-corona earnings. Did tech companies suddenly become twice as valuable over the next 10 years in the last 1.5 years, or are we in a bubble? Did they become 2x more moaty? Did they suddenly get 2x higher future earnings in the last 1.5 years (pre-corona), let alone post corona?

You have two forces working here - P/E contraction as risk appetite floats away as a recession takes hold after a shock (this happened in 2008 and 2009 and 1929) and E contraction as a recession takes hold (this one will be worse than most for that). This happens after every big shock as far as I can see. They multiply each other and at 30% each you're at -50% for the S&P 500.

The only out I can see from business as normal -50% crash here is that vast sums of fed money will do some magic in keeping risk appetite going and systemic despair from setting in as businesses close and economic damage spreads and defaults spike. Maybe it will. I'm really not buying it (90+% this isn't going to save the market/economy not buying it) and that's not what you seem to be arguing.

Also, are you going to retract this in light of my analysis showing this to be ridiculously wrong? I don't know where you're getting 4.8T bu and my analysis (it adds up to more than 4.8T so excluding non-OECD didn't make much of a difference) clearly shows that less than half of it was in Q2, which means the weekly average was likely less than 200B. This is also corroborated by the 287B output gap in Q1, which included 1-2 weeks of a bunch of major US states being shut down, several european countries being shut down, and lots of COVID-19 related lockdowns and slowdowns everywhere in the world (South Korea, Japan, etc).

And this should've been a sign for you to think more carefully. How are future lockdowns going to retroactively reduce output in Q2?

How are you accounting for the lag time between economic shock and economic damage and healing/retooling?

I'll take the global conference/trade show/business getaway industry as an example since it's dead right now (no one is getting these haircuts) and I know a little bit about it.

These vast conferences need to be planned at least a year in advance, and are huge capital moving engines around the world - they bring together suppliers and customers, forge business relationships, allow lowest-cost shopping/supply chains to be put together, prop up tourism industries and boost local economies, and enable global trade to work well.

They're dead right now:
So we have an entire industry worth over a trillion in GDP just gone, and it's also the grease to lower the friction in the global economic engine and make the world more efficient. And it's just gone. Wiped out. And because of the uncertainty, few are planning new ones, which means there's likely a year lag before this gets to even a shadow of its former self. Business as usual just can't flow. Parts can't get sourced. Suppliers and customers have a harder time building relationships. Ideas can't spread. Risky bets can't be made on capital and new products.

We will be in a far more friction filled world economy because of this alone, and friction equals economic loss, which ultimately flows somewhere - usually into corporate bottom lines.

And this trillion dollar hit is just one part of the impact of covid. We truly have no idea how ugly this will get as bankruptcies and closures mess up global supply chains, credit risk-taking, capital building risk taking. Every factory not built because of the uncertainty now will hit global output 1-3 years from now. Every relationship not developed will do the same. Every supply chain not put together with previous greatest-efficiency. It adds up fast to a lot of money and a lot of economic loss.

How is that you still have no idea what you're even arguing against? I didn't say S&P 500 will stay up. My point is that big crashes are generally not about earnings. The moat argument wasn't even about whether S&P 500 should go up or down, but that monopoly profits protected by moats should be more interest-rate sensitive than commodity profits.

My general point has always been that crashes don't happen as a result of the economy being bad, then the bad economy dragging corporate earnings down, then investors projecting bad earnings perpetually forward. It's about a change in risk premium and flight to other asset classes. This is my original post from way back, which still looks good.


That's my whole point - relative performance is driven by earnings but large stock market crashes are typically not about the real economy and the earnings. Your point has been crashes are about the real economy and the earnings. What economic reality was Amazon's valuation fluctuation in response to?

Again, we went over this - being a "moat" company doesn't help when your valuation is sky high relative to interest rates.

When people are talking about interest rates in response to stock valuation, they are talking about long-term rates (10-year and out), not short-term rate. This is quite embarrassing even for you. Here's an actual comparison expressed as risk premium:

https://www.yardeni.com/pub/stockmktequityrisk.pdf

It's probably come down a bunch since this was published so it may be a little frothy, but nowhere near as bad as 2000.

Retract what? You're completely wrong and trillion/week in lost wealth/economic output during lockdowns is a pretty solid ballpark estimate.
My claim was that every week of lockdown was reducing global wealth/economic output by a trillion dollars. A little more than 1% of yearly GDP (which is $87 billion) per week was being lost for every week of global lockdown. It's an eminently reasonable estimate. That the effects happen across time isn't surprising. For example, the ~10% of yearly global car output that didn't happen in the lockdown weeks when the factories were shut won't all show up in Q2 GDP, but it will show up throughout the year as inventory gets sold down and supplier chains report the lower orders/lower future investments that flows through onto reported GDP eventually.

So where will it go? Back through lows or not?
I'll try to find the graph of projected 2 year earnings vs price I've posted before. It's a pretty strong relationship, strong enough that I think your point can almost conclusively be called wrong.
Perhaps I'm a little primed toward price (I am a trader looking to make bets), but I just don't buy this. I don't buy that a 30 P/E Microsoft should go to 50 P/E because the the interest rate drops from 5% to 2% or from 2% to 1%. DCF is only part of the equation, and a small part near the lows imo. Given me a P/E of 5 and rates of 10% and halve rates? Great, it will shoot up, I'm right with you, that's in the sweet spot of the theory. Given me a P/E of 30 and rates of 2% and halve rates and I don't think the same applies. You can see this in global valuations, for example Europe (which has plenty of moat companies), hasn't had this DCF model be validated as rates plummeted in years past. It's a model and it breaks down at the tails.

Will the rate valuation equation make people bid up a $1.5 trillion company like Microsoft (that makes $40 billion/year in profit pre-corona, far less after), to much higher because the rates have gone from 2% to zero? I don't buy it. The US stock market isn't just competing with rates, it's competing with business investment and lending (safe corporate bonds for example which pay 2-4%), global stocks, global currencies that offer much higher than 0%, etc. At some point the relationship between rates and moaty sensitivity to rates breaks down.

I will have to find that graph matching 2 year projected earnings vs stock price. It might change your mind.

Yeah of course I know how DCF works, but it's not the whole picture. Short term rates are perfectly relevant for valuation decisions unless you're working off an unrealistic model. Not least of which is the capital flow these short term rate changes cause.

That seems revisionist.

If there was an inventory build-up, this would reduce future GDP. If the inventory was depleted, it would increase future GDP. You seem to be hinting at both these things - I don't know what it is that you're saying. Either way, this doesn't have a meaningful impact - this will differ significantly from one sector to another and I don't think most economists are actually factoring this into Q2 vs Q3 projections because you'd have to know what the inventory situation looks like at Q2 end.

Either way, if you think this was from the damage that was already done as of Q2, say the virus disappears completely at the beginning of July and everyone knew this back in May. Do you think Q3/Q4 GDP projections would be where they are?

And my earlier point was that if your loss estimates are at this level (trillions in the US, 5 trillions globally) as of May 6, which is when we had this conversation, you're most certainly talking about the future, not the loss that has already happened. The claim that I was addressing was that this kind of damage already happened / is-happening, and look at all these additional things that are going to happen. If you're now saying, all these things in aggregate is going to result in a 5T GDP drop for the whole year, it seems like you're agreeing with me now. Also, if this was your point, why do you mention this as the best case, as though the worst case scenario from the second wave has any bearing on the damage caused by the initial lockdown?

grunching here...

the economy is going to be much lower the next few years than it would have been.

i would say that a reasonable portions of our economy will never get back to full capacity.

therefore, i think we have lost some of potential gdp...

you can through all kinds of activities. the haircuts was a good example.

and there is alot of discretionary spending that can be delayed.. a cruise, a trip to the casino, a new car, going to a restaurant........... consumer staples generally can't be delayed much.

So we talked about the retail theory and if this is true, for the retail bull market to peak, we'll probably need some corrections in the meantime. Maybe that's what we're going to get now but depends on lots of things. In a generally bullish population, the correction brings in additional people that thought they missed the boat earlier. Since corrections don't usually churn people - a prolonged crash does - each correction makes the bull market stronger. At least that's my theory - we don't really have a lot of data points on this so I'm guessing a lot here.

I think at this point, going through the lows likely requires a negative shock that's not baked in - for example: 1) a severe second wave, 2) a credit crisis. A complete lack of federal coordination and guidance, not to mention international coordination, increases the risk of a severe outcome. The further out the credit crisis is, the higher the risk of a severe mraket outcome, since the political will to do what's necessary to soothe the market will be harder to muster.

The problem with this theory is that typically recessions are caused by the same things that cause the earnings to drop in recessions. Businesses and investors typically overextend at the same time at the peak of the market, so valuation gets stretched precisely as businesses are making speculative malinvestments that lead to an oversupply and will have to be written off. You talked about how businesses are sensitive to small drops in GDP. But if that's the case, why weren't they equally sensitive to small increases in GDP? This is the main reason, GDP drops usually happen when businesses are maximally sensitive to GDP drops. Meanwhile, this also usually happens while the monetary/fiscal policy is working against them - so leverage in the private sector increases to compensate for this.

There are really two parts to my theory here - 1) earnings won't drop by way more than they did in 01/02 or 08/09, which is what you'd naively predict if you're trying to go from drop in GDP to drop in earnings. 2) a short-term drop in earnings isn't what moves the market anyway and this is especially true if the earning drops aren't existential (as was the case for many financials) and are explained away as a short-term thing (this was much harder in 2001-2002 because one of the bull theories about how the internet is going to change everything also became a bear theory - maybe the internet is going to compress corporate margin forever through increased efficiency).

I think your general intuition is correct but the real reasons IMO are: 1) Not all profits are fixed monopoly profits, so obviously not all of it should be considered interest-rate sensitive. 2) Equity isn't risk free, so, say, given a fixed 3% equity risk premiun, a long-term rate drop of 2% to 1% changes the effective discount rate from 5% to 4%, so it won't have the same impact as going from 10% to 5%, which would change the discount rate from 13% to 8%.

I think you're finally starting to think about the things that matter. The better way to think about it is that equity exists on a continuum of similar risk assets that compete for the same money. Obviously if the risk premium for all risk assets go up, the risk premium for equity needs to go up. And that's exactly what happened in 2008-09. Btw, I've already talked about the risk premium and how the real discount rate is risk free rate + risk premium way back. The correct framing here isn't that risk-free rates don't matter, but whether the implied equity risk premium makes sense historically or in comparison to other risk assets. As I mentioned way back, the reason why a credit crisis matters is that it dramatically increases the risk premium for all credit assets. So the equity risk premium must also go up - if there are a bunch of attractive credit assets that are yielding 10% above risk free rate, why shouldn't I demand something similar or more for equity, which may be even riskier? This lowers equity valuation.

Looks like continuing claims were much worse than expected. This looks like a good investigative opportunity - I really suspect some type of fraud. I don't think I've seen a non-fraud explanation for the discrepancy between the two series that makes any sense. Tooth, I think this is your specialty right?

fundamentally EV/Noplat = (1-g/roic) / (wacc - g)

plot it to wacc to get a quick view of the sensitivity, roic should be 50%+
you will see that it's very steep (almost impossible to price in) when you go from wacc of 8% to wacc of 6%, especially when growth is 4 or 5%

market is not pricing this 1-to-1 but only partially

good growth companies in europe with a moat trade on >25x pe as well
with much slower growth than msft

naive probably rhetorical question,

trump and some hypocritical nut-job he nominated for fed reserve board both love the "gold standard" (judy shelton is her name)

is there any way the USA could have had its currency backed by gold with QE, QE2, Covid stimulus?.......... it just seems like the USA has been printing money for a long time now........... don't get me wrong, other countries have printed money too.

as per judy, she thought zero interest rates were horrible when obama was president, but she was pushing for negative interest rates with trump and a strong economy........... you can't make this stuff up. even the republican senators interviewing her were disdainful.

We can't be more sure about any company actually.

The awesomeness of the gold standard is a myth to start with.

Problems in the financial sector unresolved since the last crisis are coming home to roost



https://www.ft.com/content/284fb1ad-ddc0-45df-a075-0709b36868db

It's not even the same bullshit all over again. In never stopped.

As for the economy was amazing before Covid meme

I would say companies are continually degrading service and increasing price. Anyone who lives in 1990s until now can see difference in quality of both retail items and food. Go to dunkin donuts for example look how small their donuts are.

What data shows food inflation is out of control?