On a related note:

I really think analysts aren't looking at Google's income statement properly. I don't think Google is doing some kind of financial engineering, but their numbers doesn't look that great if you look closely:

1st Q 2011 GAAP net income $2.3B (adjusted to $1.8B after DOJ investigation, we'll use $2.3B anyway).

1st Q 2012 GAAP net income $2.89B

~$600 Million increase in GAAP net income.

BUT WAIT

1st Q 2012 Traffic acquisition cost totaled $2.51 billion

1st Q 2011 Traffic acquisition cost totaled $2.04 billion

~$500 Million increase in TAC

Google had to spend an additional $500 Million to earn an additional $600 Million net income. And it's the same thing if you compare other quarters YOY. The increase in TAC is almost equal to the amount of increase in net income. It's easier for Google to just withdraw money from it's own bank account and deposit it to another one and call it "net income". ;) jk

Net income is after subtracting TAC. Revenues increased by $2.1b, so Google paid $500m to acquire $2.1b in revenue. Of the $1.6b in "real" revenue growth, they made around $600m in profits. In other words their profit margin is roughly 40%.

Also if you're worried about such shenanigans, you can look at free-cash-flow, which was $3.1B for the quarter (astonishing for any non-oil/AAPL company).

>Net income is after subtracting TAC.

I'm aware Net income is subtracted after TAC.

>Revenues increased by $2.1b, so Google paid $500m to acquire $2.1b in revenue. Of the $1.6b in "real" revenue growth, they made around $600m in profits. In other words their profit margin is roughly 40%.

You didn't address my original post. Google's net income grew almost AS MUCH as the increase in TAC. and it's the same pattern for years. and Google's TAC has been steadily increasing over the years. I'd be more impressed if net income increased but TAC decreased.

Also I wouldn't focus on Revenue since Google's revenue INCLUDES their TAC.

> Also if you're worried about such shenanigans, you can look at free-cash-flow, which was $3.1B for the quarter (astonishing for any non-oil/AAPL company).

A limitation of using free cash flow versus the GAAP measure of net cash provided by operating activities as a means for evaluating Google is that free cash flow does not represent the total increase or decrease in the cash balance from operations for the period because it excludes cash used for capital expenditures during the period.

And if you're going to talk about free cash flow, look at Google's return on invested capital.

Google's 5-year average ROIC is only 17.1% and currently it's only at 15.9%.

TAC is largely AdSense partner payouts - and likely things like Chrome distribution deals and Android partner agreements now. Saying "net income grew almost as much as TAC" is basically saying "Google's AdSense, Chrome, and Android businesses are a healthy and growing segment of its total revenues."

Note that you can spin this to look bad for Google either way. If TAC was low, it'd be "Google gyps their AdSense partners, and soon there'll be a publisher revolt", something I've occasionally seen complained about on HN. If TAC is high, it's "Google's numbers don't look good; clearly their business is in trouble." The numbers are what they are; if you want a useful picture of whether to invest, you need to understand the business better than that.

I'm not trying to spin this to make Google look bad.

It just looks like that their spending an awful lot of money to arrive at a net income which is almost equal to the amount of money they spent.

I'm not trying to spin this to make GOOG look good, but I would love to be able to spend $1 and get $1 back as profit in addition to my $1.

In this case, Google spent $1 to get a return of $1.1

No; $1 returned $1.2, which is better than any bank at the moment; and these are marginal numbers which understate the underlying profitability (i.e. they are like the derivative of the costs and profits).

First you said - "It's easier for Google to just withdraw money from it's own bank account and deposit it to another one and call it "net income"." This is absolute nonsense, and what I was responding to.

It may well be that increasing TAC % implies a worse business for Google, but thats certainly not what your original point was.

that was a joke ;)

How is a 120% ROI in 1 quarter a problem? You just let me know where I can spend $500M and get $1.1B back in the same quarter.

I have no idea what point you are making.

It doesn't matter what the dividends are today. A rational stock valuation is the NPV of all future dividends. There must be a guarantee that dividends will be paid equally, otherwise there would be no rational way to price the class C stock.

In short, the class C stock must have identical dividends to the other classes.

I have been looking for this answer since I first heard about the idea of stock as a kid. No one else could ever provide it. Thank you.

"A rational stock valuation is the NPV of all future dividends"

Don't get too wrapped up in that answer. It assumes that investors are rational and able to see the future, when they are neither. Stock value is based on the same thing as everything else: what someone else will pay. Stock prices are no more rational than home prices.

False. There is plenty of dispute about the NPV of all future dividends and the discount rate, but the notion is a fundamental truth. It is follows from basic math and is provable.

You would have done well at Lehman Brothers, really really well.

indeed. pretty much all the points arguing with you are actually just discussing ways to value npv.

Okay, but that's not what he's claiming. He's claiming that NPV is derived exclusively (and rationally, and perhaps clairvoyantly) from the future sum of dividends.

The value of the stock is the current estimation of that npv. How it is estimated is irrelevant. It is the group estimation across all buyers and sellers.

I feel like you're taking part in some unrelated discussion. This entire discussion started because jcampbell claimed that NPV was derived from the sum of future dividends. (Actually, he claimed that stock price was NPV of those future dividends.) No one here seems to be disputing that stock has a NPV. They're disputing jcampbell's claim that NPV is derived exclusively from future dividends.

You're arguing against a claim that no one is making.

i think we have a disconnect.

what do you think an npv of a stock is?

The sum of all the cash flows related to the stock (discounted for time). Dividends are not the only factor there. For most investors and most stock, the future sale price (i.e. when I sell my share) is the major factor, and any dividends are a minor factor.

When you sell your share, someone else buys it. How do THEY value it?

someday in the distant future, the stock is going to pay a dividend. cash flows are what enable them to do it. eventually the company has nothing to do with the cash but to pay it out.

They might value it by expecting dividends, or they might value it by expecting the share price to grow, or they might value it by expecting a profitable acquisition. Hell, they might buy it hoping the company will liquidate and disburse its holdings. Dividends are only one possible payout that can come from buying stock.

You and jcampbell are making the same mistake. You're looking at dividends, seeing a reasonable explanation for stock pricing, and then simply asserting that what's reasonable must be true. But it's not true. A ton of things factor into stock pricing and value. Some are rational, like attempting to estimate the future value of dividends or an acquisition. Some are irrational, like assuming that upward trajectory will continue "just because". And others are in the middle, like believing that timing the market is possible.

Unfortunately, I am perhaps summarizing too aggressively. And this format does not really lend itself to a discussion.

Companies make money. They either pay out the revenue as dividends or invest in the company. Eventually, they are unable to invest in the company, at which point the revenue leaves the company in the form of dividends.

I'm not saying that people don't have a variety of reasons for thinking the stock will move or be worth more or less.

My point is that the UNDERLYING VALUE of the stock is that it represents a future revenue stream. The reason the price moves up or down is due to supply and demand actually changing. And the reason that happens is that some subset of people think that the stock is under or overpriced.

Thinking that the stock price represents anything other than the aggregate opinion of the value of the stock is going to be difficult unless you want to deny the efficient markets hypothesis.

(I actually spent years as an algorithmic trader. But what do I know?)

I generally agree with what you've written there. I think we're mostly arguing shades of gray.

On the topic of the efficient market hypothesis, there was a paper published a while back that purportedly disproved the hypothesis in the general case (basically, the potential number of data points can grow larger than the available computational/cognitive power of the market). I'll see if I can dig that up later. I think specific cases could potentially still hold, though, at least theoretically.

A long time ago someone came to present a paper to us that showed that people trading without information (eg momentum traders you mentioned above) impact the stock price much less than those trading with information. So I suspect that the more people know the more efficiently they themselves move the market. This might solve the "too much information" paradox you mention...

Here's the same discussion: http://ask.metafilter.com/71639/What-are-stocks-actually-wor...

It's provable? Please enlighten me.

You are welcome. When you buy a stock, you are buying two things: rights to the profits (via dividends), and rights to manage the company (stock voting). Most people you encounter don't undersand stocks at all, and I'll likely be downvoted to hell, but those fools are wrong. Don't listen.

Valuation is a super complicated subject, but what stock represents is a simple fact.

You're wrong. You're oversimplifying stock valuation to a hazard. You cannot use the dividend discount model to value stocks that pay no dividends. This is a fact. Furthermore, a dividend is not the only source of return from a stock, there's also capital appreciation. A stock appreciates for various reasons including investor sentiment and share repurchases (another way for a company to distribute profits). Also, you don't necessarily receive a voting right for a common share.

Like you said, valuation is a super complicated subject so don't mislead others with an incorrect simplification of it.

Of course you can, you just may not enjoy the answer. Without dividends, there's no actual mechanism for corporate success to benefit shareholders, there's only how "investors" assume each other will probably react to the news. Even share buybacks are priced through this hall of mirrors. A rational valuation is not an attempt to predict where the speculators' random walk will be at some point (there's every sign that can't be done) but the amount you can pay in confidence of ending up with a profit regardless of all that, based solely on how the company plans to reward buy-and-hold investors. These days the answer is often "only speculators think these shares should have any value", and Mark Cuban astutely likened those stocks to baseball cards.

Market expectations do price shares. These expectations can be assumptions about growth, dividend payouts, and such. You're claiming that there is no random walk component, which there clearly is. There is no such thing as paying an amount "in confidence of ending up with a profit" in the world of equities investment — that's an absurd statement and not a means for rational valuation. You can mitigate your portfolio risk but you will never eliminate systematic risk.

But why would capital appreciation matter if the stock holder never gets to take that capitol? Aside from voting rights, if you assume a stock will never pay dividends, I don't understand how it would have any value at all.

That's a really good question. First, to clarify, capital appreciation refers to the increase of the price of a stock. Investors buy stocks for a different reasons. The primary reason is to receive a return on that investment. Return is an all-encompassing word. Investors earn a return if the stock price goes up or if the company pays dividends.

There are many companies that don't distribute dividends, either because (1) they may be in poor financial health or because (2) they prefer to reinvest that money in assets that will help generate more return. Even without the dividends, people do buy these companies. In the second case, a company's earnings are expected to rapidly increase, thereby generating more value and lifting the price of the stock. (These are called growth stocks.)

Analysts have developed a gazillion ways to determine a stock's value and potential return. Like many other things, the value of stock is in the eye of the beholder. You could be willing to purchase a share of Google for $2, while I may only be willing to spend $1 for a share. A market for Google develops when millions of people apply their own valuations of Google into buying and selling that stock. It's the market mechanism gives the a stock value (price).

It's likely that this answer won't satisfy you, because I'm essentially telling you that a market capitalization is in the eye of the beholder. Questioning a stock's value is an existential question that applies to everything that we buy, hold, or sell.

It's likely that this answer won't satisfy you, but it shouldn't. Nobody should ever feel that they "know" the value of a stock, that's preposterous.

I don't think this is right. I don't think it's just in the eye of the beholder, and it is not turtles all the way down. It may be 5,000 layers deep, with people believing what people will believe what people will believe what people will believe. But at it's base, at the bottom, is at least the idea of people who want stocks because you get payed dividends for having them, or might hope to get payed in the future.

Imagine I was a business owner and I made up a new thing called "bleg". The way this works is that I need a new roof for my factory, but I'd rather not borrow the money for it from a bank. Instead I tell all my friends and family, "hey! Give me some money for my business, and in return I will issue you shares of bleg!" And my family says, "oh cool! great! what does bleg get me?" And I respond, "well it get's you a percentage of the total outstanding bleg, of course!" And my friends ask if owning bleg will get them the chance to vote on how my store is run. And I say "No!" And they say, "If you do really well one year, can we have some of your profits?" And I say "No! You don't get that either!"

No on would want bleg. And no one would think that others might eventually want it, so they should get in early before bleg blows up. No, bleg would flop.

> I don't think this is right. I don't think it's just in the eye of the beholder, and it is not turtles all the way down. It may be 5,000 layers deep, with people believing what people will believe what people will believe what people will believe. But at it's base, at the bottom, is at least the idea of people who want stocks because you get payed dividends for having them, or might hope to get payed in the future.

Some people might do that. Others might hope that the company they're buying shares in will be purchased and they'll make money on the deal. Others might hope that they can simply sell the shares in the future for a higher price.

> Imagine I was a business owner and I made up a new thing called "bleg". The way this works is that I need a new roof for my factory, but I'd rather not borrow the money for it from a bank. Instead I tell all my friends and family, "hey! Give me some money for my business, and in return I will issue you shares of bleg!" And my family says, "oh cool! great! what does bleg get me?" And I respond, "well it get's you a percentage of the total outstanding bleg, of course!" And my friends ask if owning bleg will get them the chance to vote on how my store is run. And I say "No!" And they say, "If you do really well one year, can we have some of your profits?" And I say "No! You don't get that either!" > No on would want bleg. And no one would think that others might eventually want it, so they should get in early before bleg blows up. No, bleg would flop.

Of course bleg would flop. You defined it to have no value. If instead bleg were a percentage ownership in the company, people might indeed buy bleg (and they do, but they call it shares of stock).

> A rational stock valuation is the NPV of all future dividends.

Investors are not rational. Nor are they able to see the future. Future dividends have next to nothing to do with the present price of GOOG.

> There must be a guarantee that dividends will be paid equally, otherwise there would be no rational way to price the class C stock

What kind of guarantee would you imagine is meaningful in a world where 3 people control all decisions? If they promise today that dividends will be equally split, what stops them from tomorrow saying that class C gets no dividends?

> Investors are not rational. Nor are they able to see the future. Future dividends have next to nothing to do with the present price of GOOG.

Blah, blah, blah. Let's assume google issued a share that had 1/10 dividend rights and 1/10 voting rights. What would the value of the share be?

A share of stock confers two things. Voting rights, and dividend rights. That is what people are buying. Voting, and dividends.

> If they promise today that dividends will be equally split, what stops them from tomorrow saying that class C gets no dividends?

bylaws

> Let's assume google issued a share that had 1/10 dividend rights and 1/10 voting rights. What would the value of the share be?

A lot more than 1/10 of a standard share, because dividends and voting rights are not the only things of value with respect to a company. In particular, there's also the potential for acquisitions and mergers.

> A share of stock confers two things. Voting rights, and dividend rights. That is what people are buying. Voting, and dividends.

A share of stock confers partial ownership of the company. That's what it confers. It might come with voting rights (or not), and it might come with dividend rights (or not).

Are you actually operating under the belief that the new class of Google will have zero value unless Google eventually issues dividends?

> bylaws

You mean the bylaws that Larry, Eric, and Sergey can change due to their supermajority voting control?

You should check how much dividend the former shareholders of Instagram or Plumtree received.

I have longstanding confusion on this subject. What if dividends are never paid? To my layman's intuition it seems like a more sensible basis for stock valuation is as a fraction of the value of the entire company. This is only equivalent to your definition if profit is equivalent to dividends.. right?

If they were never paid one day either the company finally dies or it gets wound up. On wind up day, assuming the company was profitable there'd be one huge pay out (ie dividend)

More realistically if you look at Microsoft and Apple, they were trying to never pay dividends, but they finally just had so much cash it was the only sensible thing to do

I don't agree. A company can never pay dividends then be bought out by a company that also never pays dividends.

I'm thinking now that there are two components to a stock's value: a concrete component, since a share represents a part of the company, and a volatile abstract component depending on the amount of market demand for it at the moment (eg. in case of a hostile takeover).