3 ms·
What I don't understand. How do these figures justify a $70 share price ? Companies tend to be valued at 15 * revenue, on large aggregate. With these earnings
by waps 12y ago
What I don't understand. How do these figures justify a $70 share price ? Companies tend to be valued at 15 * revenue, on large aggregate.
With these earnings their PE ratio will go from 72 to 57, assuming the stock price doesn't increase. If this stock deleverages at the same speed as google stock did, over 2 years, the share price will be $35-$30 by the end of the year ...
If the pe ratio gets in line with S&P 500 "normal" pe ratio of 15 (actually more like 10-15, but recently it's been nearer 15 than 10), facebook is only worth $17 per share.
Now the default argument is "but growth", so I fit a third degree curve to their earnings. So I fit a second-degree curve to their eps (this assumes it's on an exponential growth trajectory, quite generous I would say), and at what point would their valuation become justified at standard S&P 500 ratios ? Q1 2017. This is assuming FB's exponential growth holds up. That's not as bad as I feared it would be, but still it's pretty bad.
EPS curve for facebook:
0.005 x^2 - 0.0015 x + 0.12 (x is measured in number of quarters since Q2 2012. Data from streetinsider.com)
- stormbrew 12y agoIf share prices were determined by a simple mathematical formula based on earnings reports, there'd be very little profit in buying and selling stocks. People (and organizations) who buy fb believe it is undervalued in the long run. They may be wrong or they may be right, but you can't math that perception out of existence.
- lutusp 12y agoThe tl;dr: anyone who thinks equity valuations are based on rational decision-making are going to be taken to the cleaners.
- stormbrew 12y agoWell, for a certain definition of rational. I think the operative definition in this subject amounts to roughly "not random," though. People (or nowadays trading bots) have reasons, they just aren't necessarily good reasons.
- lutusp 12y ago> Well, for a certain definition of rational. I wasn't going too far afield -- by "rational" I meant on the basis of P/E ratios and other conventional sources of information, rather than mass psychology or hunches. As to trading bots, depending on how much capital they move, they can twist a small market until it cries uncle, and in a matter of minutes in the worst cases. That's rational by some definitions. :)