Thank you for the thorough explanation that does make much more sense now!

However, trading stock doesn't increase value, so every time you buy stock and then make a profit, you prevent someone else from making profit, and if you buy stock and then make a loss, you help someone else not make a loss.

It's worth noting that creating value takes a long time. Which means that if you make money in the short term (e.g. by buying and selling stocks all day) then, in the short term, somebody else (or, rather, several somebodies) lost that money. High-frequency-trading, for instance, is absolutely a zero-sum game.

I would add that the added value is relative. If everybody who owns stocks would decide to sell them right now they would find , with the exception of the few first persons who sell, that their stocks don't have their value anymore. So, in a sense, the value of the market is potential, not real. It is real for as long as the buying and selling is not massive.

@BenMiller: re your example with the tree. It looks like you created value by planting the tree etc. But suppose I have a tree farm. When your tree is ready, all my trees lost a tiny little bit of value because now there is one more tree, yours. It looks like you created value, and you did in a way, but this was only possible by devaluing s/th else.

@gojira Supply and demand play into value. But it's also possible that I find a new buyer for my tree that you don't have. In that case, I'm not lowering the value of your trees, because demand increased. And as the economy grows and more houses are built, more wood will be needed.

I'd disagree with the assertion that a zero-sum game is "unsustainable". The derivatives market is zero-sum, and it dwarfs the size of the physical market.

@SkinnyJ The way I understand this answer, it's not that a zero-sum game is unsustainable. What is unsustainable is the idea that the stock market is a zero-sum game. That's because "The stock market as a whole, on average, increases in value over time."

Great answer. "Not Normally". If you could add a paragraph that there could be times when someone looses money. It depends, someone may loose money and someone may not profit. There is no direct co-relation between someone making a profit and someone making a loss.

The stock market does not create value, the companies do. The stock market increases in value because some of the value created by the companies is invested in stocks.

When I say "the stock market increases in value," I mean that the total value of the companies in the market increase in value over time. This is not a guaranteed future outcome, but it is historical fact. The market itself is just a mechanism for buying and selling pieces of companies.

This answer would be improved by adding a note towards the end on how Person A buying stock in Acme Corp from Person B helps Acme Corp produce Acmes. As it is, the answer does indeed answer the question, but at the same time kind of leaves the reader hanging. Adding a short explanation of the last piece to the puzzle would greatly help remedy that.

@Dheer Thanks. I've added a footnote * that addresses your point.

@MichaelKjörling I've added a footnote ** that explains further how an investor that buys a share from another investor helps provide capital for the company.

@Kaz: I am not so sure about high-frequency trading being zero-sum. Consider all the individuals engaging in HFT. Most likely, they are (in the aggregate) making money off those buyers which buy stocks but do not engage in HFT. In return, they provide asset liquidity which may enter into the utility function of the buyers.

@HRSE When I say HFT is a zero-sum game, I don't mean that for one HFT to win, another must lose. Rather, that for a HFT to make money, somebody, somewhere (or several million somebodies) who are also trading in the market at the same time, must have lost (or not gained) said money. Sometimes that is other HFT firms. More often, it is everybody else.

Also, a company might make less money by selling produce then what costs it has to pay (wages, rent, power), so its value might shrink.

@Kaz OTOH HFT can be said to provide arbitrage and liquidity. So they can be think as a trader who transport the wood from sawmill to carpenter - it doesn't produce value directly but it helps minimizing losses. (Whether HFT do that is longer discussion but it is less clear cut then you make it).

@MaciejPiechotka I appreciate the nuance. In the context of "paying for liquidity, order and price discovery", HFT firms might, potentially, be providing a real service commensurate with their gains. But in the context of the question and the OP's likely level of knowledge, "zero-sum game" seemed close enough to get the point across.

@BenMiller: You are just reiterating and restating your same argument.

@Kaz "zero-sum game" was the perfect way to get the point across! Also as other people suggested I was thinking that even though no one is losing money in your example, the money must come from somewhere and thus someone else is either losing or losing out on money. Now I am wondering where money comes from in the first place, but that is a question for another day!

@Asleepace If you think about "the stock market" on a very simple "Money in = Money out" basis. Money in = "people spending money on stocks". Money out = "people selling stocks. Plus payouts (dividends, takeovers) from companies". It then becomes very obvious that if you made money "trading", and none of the companies paid anything out over the timescale you were trading, then that money has come directly from other people.

@Kaz "Money in = Money out" this is very inaccurate assumption. Stock is priced based on perceived value. Perceptions change very quickly. The equality almost never holds.

@xiaomy I'm not talking about "value". When I have "£100,000 of stock" in my account. I don't have £100,000. I have stuff that I might be able to sell for £100,000 but maybe not.

If, however, I start with £100,000 of cash, buy lots of stocks, sell lots of stocks, and at the end of the day I have £105,000 in cash in my account. Then that money has come from somebody's account.

And since money, in the stock market, only comes from people putting it in (to buy stocks), people taking it out (by selling stocks), and companies paying out money (in the form of dividends, buy-backs, takeovers etc.)

Then if that hypothetical trading happened over a short time period, where no company transactions were involved, by definition, that cash must have come from other peoples' accounts who were also trading stocks at the time.

In reality, that tends to be tiny bits of money from hundreds of institutions and thousands of individual investors, but the zero-sum nature remains the same.

Put another way: Every time you buy a stock, somebody else is selling that stock. Therefore, absent money from the companies whose stock you are buying, if you made £10,000 trading those stocks, then everyone on the other sides of your trades, in aggregate, lost £10,000 trading those stocks.

@Kaz Yes that money has to come from somebody's account. But is that necessarily a loss? You hold 100K worth of stock A. I then take a look at company A, and think it's worth 120K. So I bought it from you at 105K. Yes you made 5K profit. But does that mean I lost 5K?

@xiaomy It's much easier to see if you imagine that the market is just 2 people. A and B. Nobody else ever buys stocks, just these 2. They buy and sell stocks to each other based on what they think they're worth at the time.

At 9am, A has £100, and B has £100. They trade stocks back and forth several thousand times over the next 5 minutes.

And at the end of the 5 minutes, A has £105 and B has £95.

That is what "trading stocks" is.

Just that rather than 2 people, it's 20 Million, and rather than 200 trades/minute, some of them only trade once/year, and some of them trade every nanosecond.

And then companies muddy the waters by injecting money into the system (dividends, buybacks, takeovers) which means that people, on average, tend to make money, but not from trading.

And therefore, if you make money by trading stocks. Some people, somewhere, have lost money buying and selling stocks.

And not even, necessarily, right that second.

If I buy stock from A at £100 today, and sell it to B next week for £105, and then buy some stocks from B at £110 a year later, and sell them to A a week after that for £120. A has lost £20, and B has lost £5 trading with me, but those losses were distributed over the course of a year, and in neither of those trades did A or B "lose money".

And yet, they are £25 poorer and I am £25 richer.

It is not always clear, or obvious, or even visible as to who is losing money, where, and how.

And the losses are so small (relative to the value of all stocks in existence) that they get drowned out by companies putting money into the system.

@Kaz I see where you are coming from. You are assuming that the sum of all market participants' balance stays the same throughout time, i.e. the size of the pie is the same. And "trading" is just re-slicing the pie.

But the losses are real. And if you make money by trading stocks, some people, some where, some when, in aggregate, are losing that money by trading with you.

@xiaomy If you take out the effects of companies putting money into the system (dividends, buy backs, takeovers) which are not trading. Then yes, cash in = cash out because there is no other source for it to come from.

And when your trades are executed over the course of nanoseconds, as opposed to years, then there are no company effects that grow the pie over that timescale.

Which is why I say trading is a zero-sum game.

What about new players? For example, the Saudis sold some of their oil and want to put their dollar in stock. So they came in and bid up the price.

All of a sudden my holdings are worth more.

@xiaomy Except that when they bought £1,000,000 of stock, somebody sold £1,000,000 of stock to them"

And your stocks might appear more valuable in the meantime, but if you actually tried to sell them, the price would start moving.

In terms of actual cash moving through the system. There are only 2 sources. People buying and selling stocks with each other. And people buying and selling stock directly with companies.

And for you to start with £100, move some pieces of paper around ("trading") and come out of the other side with £105, everybody else, in aggregate, has moved pieces of paper around and come out with £5 less. Or, rather, would come out with £5 less, if they all tried to sell at once.

Like I said, the actual mechanisms are subtle, noisy, unevenly distributed and not instantaneous. They only necessarily come out in the aggregate, but that money in your account didn't appear out of nowhere.

Here is an example where no one loses money.

We all start with S&P500 at 2009

And hold it till today

We all have pretty gains. Everybody makes money.

Dividends and buybacks cannot explain all the gains

@xiaomy Actually, they can, and they do. Historically, the long-run average of the S&P is that it pays out about 6% of total value in Dividends every single year.

Which means companies pay out the total value of all the stock in the S&P 500 in dividends every 17 years. On average.

And my point is not that stocks can't increase in apparent value.

The "pie" of stock valuations can, indeed, grow over time.

But if and when people actually try to sell that stock, for money.

They'll discover that money is finite. And if you try to actually sell all the stocks in existence, that apparent "value" will disappear as fast as you can say "stock market selloff".

To try a completely different line of explanation, consider this:

At the start of today, I have £1,000 in a trading account. In cash.

Over the course of the day, I buy and sell lots of stocks, and at the end of the day, I have £1,100 in my trading account. In cash.

Where has that £100 come from?

The actual cash.

Where was it located before I started trading this morning, and what route did it take to end up in my account?

When a HFT firm buys and sells millions of stock, and over the course of a single second, ends up with £10 than it had a second ago, where did that £10 come from?

Where was it a second ago?

The answer, in aggregate, is "somebody else's trading account".

Or, rather, lots of other peoples' trading accounts.

And, in fact, less "peoples'" trading accounts, and more "other companies'" trading accounts.

But the relationship is the same. If, by buying and selling lots of stocks over a very short time frame, you end up with more money in your account than you started with, some company, somewhere, no longer has that money in their account anymore.

And so somebody loses £1 in trading, but then their stocks pay them £50 in dividends and so they never notice the "loss", because their "portfolio" is up £49.

But the fact remains that if you strip out the dividends and the buy backs, Money In = Money out and if your account is the "money out", then somebody else's is the "money in".

And it literally is that small. The S&P 500, collectively, gives back $550 Billion a year in dividends and buybacks.

The total trading profits of all the HFT firms and prop-traders and punters and every else "trading" in the market put together is much closer to £5 Billion.

@Kaz It could be from 1. an existing market participant's account (hence zero sum) 2. new money from outside the market (not zero sum). And the second point is what I'm trying to argue here.

@xiaomy And what I'm trying to explain is that everything that comes under 2) is not trading.

Nothing under 2) works on the scale of nanoseconds.

Or even normal seconds.

Or even days.

Dividends are once every 3 months.

Everything on a shorter timescale than that is, in aggregate, zero-sum.

Not true though.

@xiaomy Care to cite an example?

A very typical example of 2) is asset allocation between stocks and bonds

lots of pension plans do that

@xiaomy How is that adding money into the system though?

The market is not just equities. It is every trade-able asset in existence.

it moves money from bond market to stock money

@xiaomy It does, but it doesn't change the total amount of money available in the entire market.

It just shifts it around.

New Money is only ever created in 3 places: 1) The central bank. 2) Bank loans. 3) Forgery (whether physical or digital).

Well if you are expanding the scope from stock to money

@xiaomy Money is the whole point.

When I say "I made £10,000 trading stocks".

I mean that all the stuff in my account is now worth £10,000 more than when I started.

And, really, you haven't made that profit until you actually sell the stocks and have £10,000 of cash in your account.

You can't spend stocks.

My point, is that there is no "new" money. It has to come from somewhere. In the long run, 99% of it comes from companies paying dividends and performing buy-backs.

But in the short-term, there is no new money.

Which means that if you acquired money, in the short term, some number of other people have collectively lost that money, in the short term.

And I'm going to go back to my nano-second example.

1 second ago, I had £1,000,000. In the past second, I bought and sold several million stocks and now, 1 second later, I have £1,000,001.

Where did the £1 come from? Because it came from somewhere.

And wherever it came from, that place now has £1 less than they did a second ago.

Yes. But now that you are talking about money, it's very different.

@xiaomy What is "profit" if not money?

I use $1 to buy stock, I am spending it. Not "losing" it.

So it's $1 less, but that doesn't mean it's a loss.

which is what OP was asking about

@xiaomy Well, no, you did lose it. You lost £1 and gained 1 stock. Somebody, somewhere, lost 1 stock and gained £1.

Going back to my original point, which is what we're arguing about here, it was this:

And then the one you originally replied to:

@Kaz So my original comments are based on the assumption that we are talking about stock market as a segment of the overall financial market. So there is this dynamic between stock and other asset class, and also geographically, e.g. US stock vs. European stock, where money moving around also lifts and drags markets up and down as a whole.

But if we are talking about money/cash instead

while there is a finite amount of real cash

money can still be created (i.e. M3 vs M2 vs M1)

in other words, it's still not a pie of fixed size

Of course if you caveat it using a very small timeframe, then yes perhaps the total amount is fixed, and zero sum can be achieved.

@xiaomy With respect, that was exactly my point. Always has been all along. When you started responding, I assumed you were arguing with the points I was making.