Interdependence is a big deal.

Some real, present revenue to big companies comes from the startup world. Maybe it's where I live and what I do, but startups seem to pay for a ton of the Twitter and Facebook ads I see. Amazon makes good money running datacenters for them. Apple and Google see a lot of the value of their mobile platforms created by startup app developers. The big companies' current revenue helps determine how much they're willing to invest, including investments in the form of acquisitions, so the dollars invested into the system can themselves contribute to exit amounts in a weirdly circular way. I'm not the first to observe this.

That in itself proves very little; both sustainable and unsustainable systems can feed on themselves. And all these large companies I'm mentioning are certainly sticking around.

But it does suggest there are paths were one thing goes bad first--new investment falters, the market starts pricing ads drastically lower, big regulatory interventions shake up some subsector or other--and the ripples are bigger and reach further than might be expected.

Yeah, this is definitely a reasonably-sized elephant in the room. It's clear that a big chunk of tech revenues come from spending by other tech companies. You can't throw a stone in SOMA without hitting the fancy offices of a startup-servicing startup.

In the late 90s, this happened because all of the companies were buying ad contracts from one another, booking the total value of the contract as revenue, and using that to pad revenue growth. Today's version is the deferred ARR, which turns a tiny cash flow from subscription software into a magically big top-line number. But what goes up quickly, can fall just as fast...it doesn't take many companies to pare back on spending before your fictional deferred revenue graph falls off a cliff.

Yeah, I thought the same thing when reading through the slides. It's pretty disingenuous of them to do that, since I think when most people say "tech bubble" they're referring to the inability of most of the VC funded companies to go public. A couple of months ago there was a blog post by Mark Cuban that made a similar argument. It wasn't well received on HN largely because most commenters just looked at the words tech bubble and instantly disagreed, rather than seeing that his argument was really about liquidity.

Anyways, I don't have a strong opinion either way as to whether we're in a tech bubble or not and it doesn't strongly affect me since I'm not heavily invested in tech companies. But I certainly am skeptical of the high valuations that currently exist.

One last thing HN readers should be aware of. VC firms are like hedge funds. The people running them make money whether the fund does well or poorly and typically the amount invested by the general partners is quite small relative to the size of the fund (often around 1%), so when you look to VC guidance for how the VC market is doing keep in mind what their incentivizes are.

Edited to add on VC firms: Typically the investors in such funds aren't rich people either. They're often (probably in most cases) institutional investors such as pension funds and university endowments.

GP commit is always at least 2%. Also, I think its pretty disingenuous to say that partners at VC / PE make money whether the fund does well or not. The management fee (2%) almost all goes into operating expenses - mostly salaries for mid-level / junior folks, professional / legal fees, research, consultants, etc. Partners make almost no money off of the management fee (LPs make sure that the management fee is just sufficient to cover the operating expenses of the fund) The way that partners make real money is through carried interest. Most funds have a hurdle rate of return (around 8%) below which, no carry gets paid. Furthermore, the vast majority of LP agreements have clawbacks associated with early carry paid in the event that later investments prove unsuccessful. So, unless the fund returns 8% per year to its investors, the partners make only their salaries.

Public tech company's price to earnings does look historically reasonable.

And private companies are slowly going public at current valuations.

Yet public tech company's earnings to revenue is unusually high.

http://www.philosophicaleconomics.com/2015/01/explosion/

Open source doesn't fully explain this, as this alone is not a barrier to entry.

As US rates slowly increase, I expect to see an increase in currency headwinds.

Mid-sized mature software businesses have figured out how to run very very profitably (25%+ EBITDA margins). Tech management teams have learned from the giants (Oracle, SAP, etc.) about how to extract maximum value from their IP and past investments. This innovation helps drive earlier stage investment, as investors know that there is a potential for a ton of cash flow available down the road.