Suppose you’d written the same post about whether buying popcorn creates popcorn:
When I buy popcorn, the trade doesn’t actually create any popcorn, and the conclusion seems pretty weird if you think about it. Namely, you are saying that just by making some trades that at the time don’t create any popcorn, you end up creating some popcorn.
The popcorn argument is mistaken for exactly the same reason. In fact people buying popcorn causes people to make popcorn, by driving up demand. The mechanism is not speculative or mysterious. And people buying investments causes people to make investments by the same mechanism.
Another way to think about it is to think about loans rather than equity, you might find these less confusing. If I loan someone $10, and get paid back $11 tomorrow, it’s not crazy to assume that they now have more stuff than they had before, otherwise why did they pay me the $1? (I might have instead duped someone into taking out a loan that they’ll regret, in which case I’ve transferred money from them to me, but if you are saying “most of the time when people buy something, it makes them worst off” then I think you are the one with the weird claim.)
Now what if you loan someone money, and then I merely buy the loan from you, so that I give you money today and then I’ll get paid back instead of you? Now am I doing anything?
Yes, because you were going to sell the loan to someone else. I’ve increased the number of people “buying loans” by 1, which means that I also need to increase the number of people “selling loans” by 1, i.e. I need to increase the number of loans by 1. (This isn’t quite right—I can also raise the price of loans and thereby discourage another buyer, but the basic analysis ends up nearly the same.) It’s just like buying popcorn from someone who doesn’t make it themself, in a world where popcorn is non-perishable. It’s all the same.
Loans and equity are very close substitutes, the main difference is the risk profile of repayment (loans pay a fixed amount, equity pays an amount that depends on how well the company does).
Let me see if I understand what you are saying. People have set up supply chains that can produce popcorn from nothing, so when you buy popcorn then it sends a message to this supply chain that it should create one more, and then it does. So it sounds like you are saying there is a similar type of supply chain for the abstract good “use some resources now to create more resources in the future”. When you buy a loan or equity it sends the signal “we need one more of those” and so one more ends up being produced.
Of course I was not saying “most of the time when people buy something, it makes them worse off”, but I was concerned that this might happen often enough that if the way investment works is by initiating a long chain of transactions, then maybe there is a high likelihood that at least one of those transactions is a bad one.
What I understood from your comment is that we can probably ignore this concern at least until the point where people are no longer just trading loans or equities. Because there may be a bunch of people that you have changed who they are buying/selling loans from, but not how many they bought/sold because they were pretty sure about how many they wanted to buy/sell and won’t change that in response to the infinitesimal change in market price. The one person who does change what they do is not a typical buyer/seller, but someone who was just on the edge of deciding what to do, which means that the money now or the more money later are of roughly equal value to them. This presumably means that they have a way of getting the more money later from the money now that doesn’t involve just trading loans, so then they will do that.
So, this does make me update in the direction of that the value created is mostly real, but I still don’t feel completely comfortable with it. “Use some resources now to create more resources in the future” is still pretty abstract, it would be nice to have a better idea of what is a typical example of it, and how there could be a supply chain for it similar to the supply chain for popcorn. Although maybe it is wrong to try to focus on examples.
Also if I’ve understood you right, then cata’s comment is almost completely a red herring? It seems like your argument would apply even if you were just trading abstract financial instruments with a defined interest rate, without any direct connection to companies.
I do think that you get the full effect by trading abstract financial instruments, because investors are comparing those opportunities to other opportunities in individual companies and other places, ready to take the best return slash do arbitrage of various types, so you absolutely move investment into everywhere that investment is being demanded by increasing the overall investment supply. You also push up the value of the thing you purchase, and the things it’s tied to, relative to other things.
I don’t think that makes cata’s comment a red herring, though, because it’s another way to think about it slash another method of action. And it points out that exactly how you invest does, to some extent, matter—if you want the maximum effect you’ll want something as tied to the things you want as you can, which you trade off against returns.
But it is sounding like the main effect is just “create more demand for investment” and that infinitesimally changing the price of the particular instrument you buy is just a side effect. Right now my impression is that the side effect is much smaller than the main effect—and also more sensitive to questions like does the company own equity in itself or plan on issuing more equity, which don’t matter for the main effect.
To put it another way, in the case where the company itself is the marginal seller/issuer of equity, then it matters which company you invest in. And yes, this is more likely if you buy the company’s stock rather than an abstract financial instrument. But it’s still not very likely, if the company has issued a lot of stock.
From your response it’s clear that I didn’t do a good job explaining.
Another short-form attempt: The details matter, but the magnitudes aren’t that different. If you invest in a financial instrument, the people who sell it will hedge or redirect their funds into other instruments until it’s all diffused into various places, driving down remaining supply of available investment by the amount you invested. Now supply and demand must match, so price goes up to do that. If there’s lots of similar available investment opportunities this will mostly be more supply and the price won’t move much. If there’s no remaining places to invest, it will mostly raise prices and reduce demand, instead. If the additional value in the companies slash making the funds available to those companies doesn’t itself increase supply, a fair prior would be 50⁄50 here, so it’s somewhat more than that, plus in addition there’s the anticipation effect—you’ve made past returns look better and future returns look better as well, so investment is now more attractive. (Plus any acausal effects, if you think they count).
The price change being a ‘side effect’ isn’t quite right, it’s more like it’s an intermediate effect.
I hope that helped but I worry the best way to understand this is actually to just understand microeconomics, and that would require a full blogpost. Although I’m not convinced it requires two!
It sounds like you are responding to my first paragraph, in which case I agree with your criticism. I think my second paragraph is closer to the way I am thinking about things. For me the main question here is whether it matters what company you invest in, not whether the price change should be counted as a “side effect”.
We can even say something more. When they manufacture popcorn, they do so because they expect to sell popcorn. If you choose to be the type of person who would buy popcorn, this is plausibly correlated to others’ similar decisions to buy popcorn and to the perception of demand for popcorn by those making popcorn. One way they observe popcorn demand is through popcorn sales, and there are many others. So if you buy a unit of popcorn, you are not only causing them to create the popcorn to replace it (since they now own less popcorn to sell to others or future you) but you also raise anticipated future demand for popcorn (and you’d also do that in any number of other ways). And you can make a case that you actually acausally caused the manufacturing of more popcorn in the past, as well. Markets can be weird, man.
Suppose you’d written the same post about whether buying popcorn creates popcorn:
The popcorn argument is mistaken for exactly the same reason. In fact people buying popcorn causes people to make popcorn, by driving up demand. The mechanism is not speculative or mysterious. And people buying investments causes people to make investments by the same mechanism.
Another way to think about it is to think about loans rather than equity, you might find these less confusing. If I loan someone $10, and get paid back $11 tomorrow, it’s not crazy to assume that they now have more stuff than they had before, otherwise why did they pay me the $1? (I might have instead duped someone into taking out a loan that they’ll regret, in which case I’ve transferred money from them to me, but if you are saying “most of the time when people buy something, it makes them worst off” then I think you are the one with the weird claim.)
Now what if you loan someone money, and then I merely buy the loan from you, so that I give you money today and then I’ll get paid back instead of you? Now am I doing anything?
Yes, because you were going to sell the loan to someone else. I’ve increased the number of people “buying loans” by 1, which means that I also need to increase the number of people “selling loans” by 1, i.e. I need to increase the number of loans by 1. (This isn’t quite right—I can also raise the price of loans and thereby discourage another buyer, but the basic analysis ends up nearly the same.) It’s just like buying popcorn from someone who doesn’t make it themself, in a world where popcorn is non-perishable. It’s all the same.
Loans and equity are very close substitutes, the main difference is the risk profile of repayment (loans pay a fixed amount, equity pays an amount that depends on how well the company does).
Let me see if I understand what you are saying. People have set up supply chains that can produce popcorn from nothing, so when you buy popcorn then it sends a message to this supply chain that it should create one more, and then it does. So it sounds like you are saying there is a similar type of supply chain for the abstract good “use some resources now to create more resources in the future”. When you buy a loan or equity it sends the signal “we need one more of those” and so one more ends up being produced.
Of course I was not saying “most of the time when people buy something, it makes them worse off”, but I was concerned that this might happen often enough that if the way investment works is by initiating a long chain of transactions, then maybe there is a high likelihood that at least one of those transactions is a bad one.
What I understood from your comment is that we can probably ignore this concern at least until the point where people are no longer just trading loans or equities. Because there may be a bunch of people that you have changed who they are buying/selling loans from, but not how many they bought/sold because they were pretty sure about how many they wanted to buy/sell and won’t change that in response to the infinitesimal change in market price. The one person who does change what they do is not a typical buyer/seller, but someone who was just on the edge of deciding what to do, which means that the money now or the more money later are of roughly equal value to them. This presumably means that they have a way of getting the more money later from the money now that doesn’t involve just trading loans, so then they will do that.
So, this does make me update in the direction of that the value created is mostly real, but I still don’t feel completely comfortable with it. “Use some resources now to create more resources in the future” is still pretty abstract, it would be nice to have a better idea of what is a typical example of it, and how there could be a supply chain for it similar to the supply chain for popcorn. Although maybe it is wrong to try to focus on examples.
Also if I’ve understood you right, then cata’s comment is almost completely a red herring? It seems like your argument would apply even if you were just trading abstract financial instruments with a defined interest rate, without any direct connection to companies.
I do think that you get the full effect by trading abstract financial instruments, because investors are comparing those opportunities to other opportunities in individual companies and other places, ready to take the best return slash do arbitrage of various types, so you absolutely move investment into everywhere that investment is being demanded by increasing the overall investment supply. You also push up the value of the thing you purchase, and the things it’s tied to, relative to other things.
I don’t think that makes cata’s comment a red herring, though, because it’s another way to think about it slash another method of action. And it points out that exactly how you invest does, to some extent, matter—if you want the maximum effect you’ll want something as tied to the things you want as you can, which you trade off against returns.
But it is sounding like the main effect is just “create more demand for investment” and that infinitesimally changing the price of the particular instrument you buy is just a side effect. Right now my impression is that the side effect is much smaller than the main effect—and also more sensitive to questions like does the company own equity in itself or plan on issuing more equity, which don’t matter for the main effect.
To put it another way, in the case where the company itself is the marginal seller/issuer of equity, then it matters which company you invest in. And yes, this is more likely if you buy the company’s stock rather than an abstract financial instrument. But it’s still not very likely, if the company has issued a lot of stock.
From your response it’s clear that I didn’t do a good job explaining.
Another short-form attempt: The details matter, but the magnitudes aren’t that different. If you invest in a financial instrument, the people who sell it will hedge or redirect their funds into other instruments until it’s all diffused into various places, driving down remaining supply of available investment by the amount you invested. Now supply and demand must match, so price goes up to do that. If there’s lots of similar available investment opportunities this will mostly be more supply and the price won’t move much. If there’s no remaining places to invest, it will mostly raise prices and reduce demand, instead. If the additional value in the companies slash making the funds available to those companies doesn’t itself increase supply, a fair prior would be 50⁄50 here, so it’s somewhat more than that, plus in addition there’s the anticipation effect—you’ve made past returns look better and future returns look better as well, so investment is now more attractive. (Plus any acausal effects, if you think they count).
The price change being a ‘side effect’ isn’t quite right, it’s more like it’s an intermediate effect.
I hope that helped but I worry the best way to understand this is actually to just understand microeconomics, and that would require a full blogpost. Although I’m not convinced it requires two!
It sounds like you are responding to my first paragraph, in which case I agree with your criticism. I think my second paragraph is closer to the way I am thinking about things. For me the main question here is whether it matters what company you invest in, not whether the price change should be counted as a “side effect”.
We can even say something more. When they manufacture popcorn, they do so because they expect to sell popcorn. If you choose to be the type of person who would buy popcorn, this is plausibly correlated to others’ similar decisions to buy popcorn and to the perception of demand for popcorn by those making popcorn. One way they observe popcorn demand is through popcorn sales, and there are many others. So if you buy a unit of popcorn, you are not only causing them to create the popcorn to replace it (since they now own less popcorn to sell to others or future you) but you also raise anticipated future demand for popcorn (and you’d also do that in any number of other ways). And you can make a case that you actually acausally caused the manufacturing of more popcorn in the past, as well. Markets can be weird, man.