5 ms·

P/E is a very meaningful metric. It's not that Amazon has infrastructure that matters, it's that they are reinvesting their earnings at a higher rate than other

P/E is a very meaningful metric. It's not that Amazon has infrastructure that matters, it's that they are reinvesting their earnings at a higher rate than other companies. If they returned capital to investors because they didn't have enough potential projects, then the investors would need to decide how to reinvest it. Your comment about reflecting the infrastructure in valuation would lean toward looking at Price/Book Value instead. Amazon would look quite expensive on this measure as well.

garmaine 9y ago
My “not meaningful” point should perhaps been “can’t conclusively draw data from it alone” or “don’t make investment decisions based on this metric only” and I’d stand by that.
- prklmn 9y ago
  P/E is without a doubt meaningful, and looking at it in the context of growth i.e. PEG, makes it even more meaningful. Amazon is selling at an objectively absurd PEG ratio of 5. Anything over 2 is fundamentally considered overvalued by people who use this metric. However, You could argue that none of the fundamentals are meaningful because we are in a bubble, with Amazon on the forefront.
__blockcipher__ 9y ago
I generally prefer free cash flow yield (fcf / dollar invested), although both PE and FCF can be fucked with depending on how certain expenses are classified. That being said, if I had to choose one of the two, I would go with FCF yield every time.
- prklmn 9y ago
  It's also very common now for companies to buy back a ton of stock thus lowering the share denominator in the earnings portion of the ratio, while not actually growing revenue much if at all.
  - hidenotslide 9y ago
    You are referring to Earnings/Share, not Price/Earnings. It doesn't matter much anyway, if they can make the same amount of money with less invested capital that should increase their metrics. And there are good tax reasons to prefer share buybacks over dividends as a way to return cash to shareholders.
    - prklmn 9y ago
      I was referring to both because EPS is wrapped up into P/E... P/E = stock price / (earnings / # of shares oustanding) And it actually does matter because a company can steadily grow EPS while not growing revenue meaning that something might be amiss, but it can easily be overlooked because EPS has been growing. Also, there are plenty of reasons not to want buybacks. Many times companies will buy their stock when it's at highs, thus destroying value for shareholders.
- hidenotslide 9y ago
  I agree, I should have mentioned free cash flow. Of course the real answer is to look at many metrics and get a broad picture. But I doubt you are going to find a "value investing" reason to buy AMZN at these levels, it's clearly a growth stock.
matt4077 9y ago
Increases in "book value" are reflected in earnings. Example: you make a profit of $100. You use those $100 to buy the Brooklyn Bridge. Your profit is still $100. It's only your cash flow that's now at $0.
- hidenotslide 9y ago
  The difference is stocks vs. flows. $100 capex only shows up in year one, but the balance sheet goes up permanently (depreciating of course). When the parent mentions "level of infrastructure" it refers to the balance sheet account, which is why I mentioned P/B.

garmaine 9y ago

My “not meaningful” point should perhaps been “can’t conclusively draw data from it alone” or “don’t make investment decisions based on this metric only” and I’d stand by that.

prklmn 9y ago
P/E is without a doubt meaningful, and looking at it in the context of growth i.e. PEG, makes it even more meaningful. Amazon is selling at an objectively absurd PEG ratio of 5. Anything over 2 is fundamentally considered overvalued by people who use this metric. However, You could argue that none of the fundamentals are meaningful because we are in a bubble, with Amazon on the forefront.

__blockcipher__ 9y ago

I generally prefer free cash flow yield (fcf / dollar invested), although both PE and FCF can be fucked with depending on how certain expenses are classified. That being said, if I had to choose one of the two, I would go with FCF yield every time.

prklmn 9y ago
It's also very common now for companies to buy back a ton of stock thus lowering the share denominator in the earnings portion of the ratio, while not actually growing revenue much if at all.
- hidenotslide 9y ago
  You are referring to Earnings/Share, not Price/Earnings. It doesn't matter much anyway, if they can make the same amount of money with less invested capital that should increase their metrics. And there are good tax reasons to prefer share buybacks over dividends as a way to return cash to shareholders.
  - prklmn 9y ago
    I was referring to both because EPS is wrapped up into P/E... P/E = stock price / (earnings / # of shares oustanding) And it actually does matter because a company can steadily grow EPS while not growing revenue meaning that something might be amiss, but it can easily be overlooked because EPS has been growing. Also, there are plenty of reasons not to want buybacks. Many times companies will buy their stock when it's at highs, thus destroying value for shareholders.
hidenotslide 9y ago
I agree, I should have mentioned free cash flow. Of course the real answer is to look at many metrics and get a broad picture. But I doubt you are going to find a "value investing" reason to buy AMZN at these levels, it's clearly a growth stock.

matt4077 9y ago

Increases in "book value" are reflected in earnings. Example: you make a profit of $100. You use those $100 to buy the Brooklyn Bridge. Your profit is still $100. It's only your cash flow that's now at $0.

hidenotslide 9y ago
The difference is stocks vs. flows. $100 capex only shows up in year one, but the balance sheet goes up permanently (depreciating of course). When the parent mentions "level of infrastructure" it refers to the balance sheet account, which is why I mentioned P/B.