Facebook's mega-billion-dollar bubble ... will it float?

Only time will tell if stalker book valuation is valid

So just how much is Facebook going to be worth by the time it eventually IPOs? Given that prediction, especially about the future, is very difficult, who knows?

If we take recent movements we can still come up with any number we like:

"As for the valuation spurts, Facebook was said to be worth $23bn in June 2010, nearly as much as UK retail monster Tesco at $50bn in January this year, and was then plumped up to a $60bn golden goose in February," noted The Register's Kelly Fiveash earlier this week.

Add $10bn a month for a year and we're racing through $170bn in January 2012. Add 20 per cent a month and Facebook becomes one of the most valuable companies in the world at $450bn and rising. Clearly ridiculous, as Felix Salmon has been pointing out for some time now.

Think back to the dot com boom and yes, there was indeed a Google that came out of it. As well as an Amazon, eBay, Pets.com, Webvan and numerous other now happily forgotten names. For this is what bubbles do: produce a plethora of wannabes before the rather brutal culling down to that femtopercentage that weren't complete lunacy in the first place. Salmon makes the point that yes, we do want to value all those wanna- and would-bes as if one or two of them was going to become a Google or an Amazon – but we really don't want to be valuing all of them as if all them them are going to be.

Which is why there are more than a few raised eyebrows out there about Pandora's recent IPO. A couple of $ billion for an internet radio station? OK, sure, people will like to listen to music, sure, advertising spots and rates could rise but, umm, where's the lock-in? There's not even an obvious set of economies of scale: Pandora's major cost is royalties and those get charged not per-play but per-person-per-play.

Is LinkedIn really worth whatever vast number of spondoolies the market says it is today? Well, yes, of course, because things in markets are worth what markets say they are worth, but beyond that? There are mutterings that they might be able to use that network (for they do at least have a network effect working for them) to get into the recruitment business but it's a lot to be valuing a possibility at.

Groupon*, the "fastest growing company in the history of capitalism", is internally valued at $30bn last I saw. More by the time it IPOs, obviously: but is flogging coupons really, even at 50 per cent margins, going to remain a high growth high margin business? It's not even that this is a new business: margins didn't stay high in newspaper coupons even back when they really were delivered to every home in the country.

It's a fair old bet that at least some of this is going to end in tears: unfortunately, knowing which bit is going to be a little more difficult.

But then that is the way with booms and bubbles in investment markets and one of the reasons given at times as to why markets are so bad at allocating investment. We've centuries of evidence to look at: not just the dot bomb that most (Many? Some? Just how old are you guys anyway?) here will remember from personal experience but secondary banking before that, car companies before that, railways, canals and all the way back to Thales cornering the olive presses in Miletus. Huge over-investment, precious capital sprayed all around at anyone with a half-plausible idea and most of it wasted. Surely it would be better for such things to be planned, for the wise bureaucrat to organise this for us?

That has also been tried and strangely the results seem not to be as good as those from these splurges. Yes, fortunes were lost building the canals, most of the railway companies went bust at one time or another. Dotcommery brought us such delights as Global Crossing (just, finally, RIP'd) which was going to build a whole new fibre-optic backbone. But here's the thing. The investors in all of those projects lost most/all their money. Capital was undoubtedly wasted. If it had all been spent sensibly we probably would now be better off.

Sponsored: Network DDoS protection