[edmundsauto]

Isn't the price set by everyone, not just the active traders? It's set each time a sale happens, but the price itself is also related to how many people are holding the stock long term.

I don't really consider it possible for pricing to be "accurate". It is what it is, but accurate implies there's a correct valuation, which I don't think there is.

[kd5bjo]

For financial products, there is a correct price but it cannot be known for certain until far in the future. Stocks in particular represent a claim on the future dividends of the company, whether issued during operation or at the dissolution of the company. If both those future dividends and future inflation were known, you could accurately calculate the present value of that cash flow and get the correct price.

[edmundsauto]

I'm over my toes here, being a programmer and not a finance person, but isn't this viewpoint controversial? IE, does everyone (relatively well informed) agree there are correct prices?

[gizmo686]

Somewhat.

There is arguably an objectivly true answer for exactly what payments a given stock will make, and so there is an objectivly true answer for what the present value of the stock is (also dependent on other aspects of the future market). This is somewhat of a philosiphical question, largly boiling down to determinism; and is largly moot because no one claims to be able to predict the future well enough for this to work.

Instead, the working position that most take (at least implicitly) is that there is an objectivly correct probability curve of what the future payouts will be, and therefore an objectivly correct probability curve of present values. How to determine what this curve is is a matter of great debate. Further, there is a sufficient lack of objective methodology, that many of the factors that people use in this calculation would be refered to as "opinion", but there is still an objective reality out there.

However, this only gets us an (unknowable) objective probability curve of present values. In general, there is no objective way to turn this into a price. That is to say it is a matter of opinion how much a 50% chance of making $100 is [0].

In an ideal market, you would be able to sell a stock for its objective value at any time. However, "the market can stay irrational longer then you can stay solvent", so you may pay a premium for stocks that you expect to not be undervalued when want to sell them.

Conversly, you may by a stock not because you think it is worth what you are paying, but instead because you expect to find a greater fool to pay you even more then you paid.

There are also cases where people value stocks not just because of their future payments, but because they actually care about the company (or, in the case of divestment, people dont buy them becausr of personal preferences).

In short, there is some matters of opinion in determining a correct price; but most of the disagreement comes from a factual disagreement about what the future looks like.

[0] this calculus changes when you have many such gambles with varying degrees of corralation (and many a financial problem have stemmed from underestimating this corralation)

[edmundsauto]

Do you consider a probability distribution function as objectively true? If it's not obviously true beforehand, I'm not sure if it's only clear in hindsight, which has all kinds of psychological issues in interpretation.

IOW, is there any practical difference between "there is no objectively correct price" and "we'll never know what it is"? The price at any given time reflects the current consensus of the objective price, distorted through the current average psychological lens of the market?

If something is unknowable before it occurs, we will never know what the objectively true measure is until it's occurred. At which point it changes, since the market is dynamic. How could we ever know which point is the correct price?

[gizmo686]

There is a difference between "truth does not exist" and "we cannot know the truth".

In theory, we can take a now worthless stock and look back in time to determine the actual present value at a given point in the past.

In theory, we can imagine an outside observer running an arbitrarily large copies of our universe from a given point in time to determine the probability curve at said point in time (under whatever model of randomness you want to use). More plausibly, we can take a set of predicted probability curves and look back to see how accurate they were (did events predicted with uncorralated 50% probability happen half the time?).

Economics is hard, because it is very difficult to determine these facts, even in retrospect, but they still exist.

[edmundsauto]

> There is a difference between "truth does not exist" and "we cannot know the truth".

If we can't know the truth, how do we know it exists (in this situation)? I'm not convinced there's an objective true value of a stock -- it seems like stock prices are the general consensus of a huge number of subjective inputs. And that will always be the case.

(Definitely appreciate the time and thought you've put into your responses, btw! Thank you.)