Global Macro Trading Journal
- Thread starter Daal
- Start date
This the "David Rosenberg" symdrome. The idea that short-term economic news should matter anything at all to the stock market. Here is a fun way to disprove this with numbers
Damodaran has an spreadsheet that values the stock based on analysts estimates for the next 5 years and then calculates an terminal value for the market. The whole method is based on the Dividend Discount Model, the idea that a stream of dividends is worth the value of those dividends (say $10 per share) divided by a discount rate minus the growth rate of those dividends
Example:
$10/6%-4%
10/0.06-0.04 = 500
The Dividend Discount Model is essentially that terminal value. Damodaran added figures for the next few years to make the calculation more accurate (instead of having a fixed growth rate for the whole calculation) and then he adds that terminal value.
But whats interesting is that the forecast for the next 5 years matter little to the final intrinsic value of the market. Usually its only around 20% of the figure. You change the forecast and the intrinsic value for the market doesn't change all that much. There is SOME effect but its unlikely to change someone's decision from buy to sell, etc
What this means in english is that the vast majority of the value stock market is derived, not from little changes in the forecast for the next 5 years but from the actual valuation level of the market (the terminal value of the market), what Damodaran calls the Equity Risk Premium and what I would call the Expected Value of the market.
This bears repeating, the vast majority of the value of a stock market comes from that terminal value, not little changes in forecasted earnings for next year or even 5 years.
What's interesting is that Rosenberg is pretty terrible at valuation as well, he is a huge fan of the Shiller PE, which has serious flaws, especially in times like right now (during earnings booms at 'high valuations', I wrote about that in the past).
Damodaran ERP right now is around 7%, my expected value (which I right now I would favor deriving from the forward PE, again, due to the earnings boom) would be around 6% or so. So this market is on the richer side but I believe its far from this bubble that the Shiller cultists seem to think
In the future, I might favor a different indicator than the forward PE to try to derive an Expected Value for the market but right now, I believe that analysts working all day to try to figure out the earnings for the next year or 2 are far more likely to get it right than these macro observers will. And once they tell me what will be the overall value of earnings, I can have a PE and with that PE, using some historical data, I can build some kind of estimate for the next 10 years.
I will stop using forward PEs when I believe that analysts will get it wrong, say, if if I think there is a recession coming. But right now, I trust stock analyst a lot more than Rosenberg looking at the rear view mirror with that Shiller nonsense
I will stop using forward PEs when I believe that analysts will get it wrong, say, if if I think there is a recession coming. But right now, I trust stock analyst a lot more than Rosenberg looking at the rear view mirror with that Shiller nonsense
Maybe I simplify things too much but it seems to me that there are 2 types of stock observers these days in the US.
1) The Shiller believers, they look at PAST conditions and try to figure out what is the expected value of the stock market (which is another say of saying the expected return). Frequently they believe there will be little to no return in the next 10 years. They not only look at the Shiller PE but also to things like Price to Sales (which ignores high profit margins), price to book (which ignores high ROEs), Tobin Q and other things
2)The Damodaran/David Tepper type folks that look at CURRENT and FORWARD conditions. They don't care as much about where earnings were in the past. They are more flexible, they are not following some silly Shiller system (that is based on tiny historical samples in very unstable 'extremistan' datasets) and closing their eyes to what's right in front of them. They care more about current and forward PEs, cashflows
Ironically, I believe that the stock market itself is a lot more of the second type than of the first. That's why it keeps marching despite the fact that the Shiller PE is at 1999 levels or whatever. It does that because smart investors simply can't ignore the facts that are right in front of them (namely that based on current and expected earnings, the market is not very overvalued)
1) The Shiller believers, they look at PAST conditions and try to figure out what is the expected value of the stock market (which is another say of saying the expected return). Frequently they believe there will be little to no return in the next 10 years. They not only look at the Shiller PE but also to things like Price to Sales (which ignores high profit margins), price to book (which ignores high ROEs), Tobin Q and other things
2)The Damodaran/David Tepper type folks that look at CURRENT and FORWARD conditions. They don't care as much about where earnings were in the past. They are more flexible, they are not following some silly Shiller system (that is based on tiny historical samples in very unstable 'extremistan' datasets) and closing their eyes to what's right in front of them. They care more about current and forward PEs, cashflows
Ironically, I believe that the stock market itself is a lot more of the second type than of the first. That's why it keeps marching despite the fact that the Shiller PE is at 1999 levels or whatever. It does that because smart investors simply can't ignore the facts that are right in front of them (namely that based on current and expected earnings, the market is not very overvalued)
I have been playing around with the monetarist formula and its applications to valuying cryptotokens and assets. Its tough because most articles are very confusing, so I'm trying to clarify my own thoughts.
The classic formula is
Money * Velocity = Nominal GDP (which is much simpler than the commonly used formula of M * V = P * Q)
That is, all of the units of money (usually measured in dollars) multipled by how often they change hands is equal to the nominal value of an economy
If there is an economy inside a single room with $100 in the total of money (say with ten $10 bills), if the turn over of those bills is 5x per year (they change hands 5 times per year), the total $ value in that economy would be $500. Every purchase has a seller and if there is turnouver (economic activity), the formula will indicate that
For tokens it seems that a crude application it is to first to try to estimate how much a token will take over in the real world, in terms of economic activity. As an example, I will use Filecoin (a blockchain storage coin). Lets say that one estimates that the cloud storage industry will be worth $500B in 5 years, and Filecoin will be 5% of that (so $25B). Given that we already have the dollar value of how much economy activity Filecoin will be supporting ($25B), it becomes easy to estimate how much is the coin worth IF you can resonably estimate its future velocity. Lets assume its 5. The simplified formula that I came up with is:
TokenMarketCapInUSD * Velocity = Nominal economic activity of the Token
X * 5 = $25B
X = $5B
If Filecoin market cap is $2.5B, one would make 100% holding for 5 years. A 14.8% compounded return, which is not all that high considering that the volatility of these assets is gigantic (50-70% drops seem routine every year or so). In other words, if one were to discount back that $5B in expected value at an appropriate discount rate, one would probably pass on the investment.
A typical discount rate in the stock market is around 5-8%. BTC is 5x more volatile and alt coins are even more volatile. So an appropriate rate is probably around 30-40% a year.
The tricky part is figuring out the likely velocity of the token (and things like hodling, store of value properties affect that velocity) and how much of real world economic activity will the token be involved in.
But whats interesting is that with this formula, one can understand better why would these tokens and cryptoassets have value. They have to have value because they are a gateway in which one can access an application/software/system that supports real world economic activity. The same way that they find it worth it to pay for Dropbox or GoogleDrive premium services and collectively will generate $500B in revenues to that industry
Lower velocity (due things like hodling, hoarding) implies a higher TokenMarketCapInUSD. If Filecoin will take $25B over the cloud storage market but its velocity is 1 because a lot of people are hodling/hording like crazy, the total market cap must rise to balance that equation. Presumably, the people that are buying the token to access that software/application/system (and collectively support its market value at $25B) find that its worth doing that because of the advantage of the system/software over conventional cloud software
Playing around with this figures and formula helps one understand the variables that give theses assets value and why would they a lot of value despite the fact that there is not a lot of real world use just yet. Its all in the optionality that it will happen and they will take a certain % market share from existing industries or create entire new industries/economic activity.
Since the tokens/coins will be the gateway to access those applications/software/systems, their value will rise as a result.
Any feedback or criticisms are welcome
The classic formula is
Money * Velocity = Nominal GDP (which is much simpler than the commonly used formula of M * V = P * Q)
That is, all of the units of money (usually measured in dollars) multipled by how often they change hands is equal to the nominal value of an economy
If there is an economy inside a single room with $100 in the total of money (say with ten $10 bills), if the turn over of those bills is 5x per year (they change hands 5 times per year), the total $ value in that economy would be $500. Every purchase has a seller and if there is turnouver (economic activity), the formula will indicate that
For tokens it seems that a crude application it is to first to try to estimate how much a token will take over in the real world, in terms of economic activity. As an example, I will use Filecoin (a blockchain storage coin). Lets say that one estimates that the cloud storage industry will be worth $500B in 5 years, and Filecoin will be 5% of that (so $25B). Given that we already have the dollar value of how much economy activity Filecoin will be supporting ($25B), it becomes easy to estimate how much is the coin worth IF you can resonably estimate its future velocity. Lets assume its 5. The simplified formula that I came up with is:
TokenMarketCapInUSD * Velocity = Nominal economic activity of the Token
X * 5 = $25B
X = $5B
If Filecoin market cap is $2.5B, one would make 100% holding for 5 years. A 14.8% compounded return, which is not all that high considering that the volatility of these assets is gigantic (50-70% drops seem routine every year or so). In other words, if one were to discount back that $5B in expected value at an appropriate discount rate, one would probably pass on the investment.
A typical discount rate in the stock market is around 5-8%. BTC is 5x more volatile and alt coins are even more volatile. So an appropriate rate is probably around 30-40% a year.
The tricky part is figuring out the likely velocity of the token (and things like hodling, store of value properties affect that velocity) and how much of real world economic activity will the token be involved in.
But whats interesting is that with this formula, one can understand better why would these tokens and cryptoassets have value. They have to have value because they are a gateway in which one can access an application/software/system that supports real world economic activity. The same way that they find it worth it to pay for Dropbox or GoogleDrive premium services and collectively will generate $500B in revenues to that industry
Lower velocity (due things like hodling, hoarding) implies a higher TokenMarketCapInUSD. If Filecoin will take $25B over the cloud storage market but its velocity is 1 because a lot of people are hodling/hording like crazy, the total market cap must rise to balance that equation. Presumably, the people that are buying the token to access that software/application/system (and collectively support its market value at $25B) find that its worth doing that because of the advantage of the system/software over conventional cloud software
Playing around with this figures and formula helps one understand the variables that give theses assets value and why would they a lot of value despite the fact that there is not a lot of real world use just yet. Its all in the optionality that it will happen and they will take a certain % market share from existing industries or create entire new industries/economic activity.
Since the tokens/coins will be the gateway to access those applications/software/systems, their value will rise as a result.
Any feedback or criticisms are welcome
So the key to invest in Tokens is to find ones that will try to disrupt big industries and be involved in a lot of economic activity. To the extend that they do that, they will be worth a lot of money. That huge pontential in the token SHOULD lead to things like hodling and lower velocity. The net result of that is a big premium on the token market cap. As time passes and the token realizes its full economic potential, hodling should decrease and the token will be priced more on reality rather than a big premium.
In a lot of ways this seems to be very similar to how tech startups get priced. In the beginning, there is no revenue, there are no tangible assets. All you have is an idea and a team. People think 5-6 times price to book is a lot but in startups, the book value is close to $0 and they usually raise $2-3M with no issue, assuming they have a strong team. Its a huge premium based on potential/optionality. As the company matures and its full potential becomes more clear, those gigantic premiums dimish
A key difference between tech startups and token/coin startups is that the latter cannot change its plans very much. Once they went down a certain route, its very hard to "pivot" into something else. A lot of the capital and effort is made to create a certain type of code, its hard to switch to something else if they realize that it wont work. Teams are important but bad token ideas are to recover from
In a lot of ways this seems to be very similar to how tech startups get priced. In the beginning, there is no revenue, there are no tangible assets. All you have is an idea and a team. People think 5-6 times price to book is a lot but in startups, the book value is close to $0 and they usually raise $2-3M with no issue, assuming they have a strong team. Its a huge premium based on potential/optionality. As the company matures and its full potential becomes more clear, those gigantic premiums dimish
A key difference between tech startups and token/coin startups is that the latter cannot change its plans very much. Once they went down a certain route, its very hard to "pivot" into something else. A lot of the capital and effort is made to create a certain type of code, its hard to switch to something else if they realize that it wont work. Teams are important but bad token ideas are to recover from
That tends to be true in Angel investing as well. At least according to some Angels. One that I saw an interview from talks about how VCs (which invest at later financing rounds) have the luxury of backing really strong teams even if they dont love the idea but angels dont have that option. Usually the startup will run out of money and struggle raising more. So unless the pivot comes early on, its unlikely to suceed simply due lack of funding