Hacker News Reader

Comment by pdq | Hacker News Reader

It's quite simple, actually:

  - The capital investment required for Instagram is well under a million dollars
  - The capital investment required for Hyperloop is well over a billion dollars (R&D, Materials/Prototypes, Insurance, Land, Construction)

The best-case returns are:

  - 1 billion for Instagram
  - 20 billion for Hyperloop

Thus, the best-case ROI for both are:

  - 1000x for Instagram
  - 20x for Hyperloop

The timelines to get to production are:

  - less than 1 year for Instagram
  - minimum 5 years for Hyperloop

The only way Hyperloop will happen is if there is a way to incrementally get there. In other words, if you can make a 1/1000-size scaled-down prototype, iterate and improve upon the design, and then scale up to a full-sized model.

View on HN

This is wrong on many levels. Firstly funding over a billion dollars is not remotely intimidating to the large VC funds who understand about capital intensive businesses and managing the investment lifecycle.

When you put together very large sums of investment you don't have a 100x expectation for return. The institutionals which will take on later stages of this financing just want to outperform the other investment options they have for that amount of casH (not that many).

Your 20Bn upside for Hyperloop is very low. If done to scale and over the long haul it's a very big opportunity. Even then I know of no investors - ZERO - who wouldn't take a 20x return on a billion dollar placement.

Finally. Instagram. Really? An outlier like that is what you're going to base your idea of investment on. Look at VC returns. They take a portfolio strategy for a reason.

It's a common misunderstanding that VCs want to put small amounts of money to work for huge returns. They'd much rather put much bigger amounts of money to work. It's how they're structured. Do they want the odd moonshot Instagram in their? Sure. They'll take it. But they'd take 1billion -> 20billion over 1million -> 1billion every day.

"Even then I know of no investors - ZERO - who wouldn't take a 20x return on a billion dollar placement."

Sure, if there was any sort of guarantee it would work. When there's a serious chance of throwing your money into a black hole where cash goes to die, then I'm not sure the upside is worth it. And even then, if you can make 1,000 investments into Instagrams for the cost of one investment into Hyperloop, I'd do that every time. "But they'd take 1billion -> 20billion over 1million -> 1billion every day." sounds wrong to me. If you have the $1B to invest, wouldn't you want to spread that out, make 1,000 $1M investments? I'd imagine your return would be better with that than an outside shot you could hit 20x your money on one investment.

Right. But no investment has a guarantee of working. There aren't 1,000 Instagrams in tech. There are roughly 10-15 moonshot investments made a year out of thousands. So you can't find 1,000 instagrams.

Investment doesn't scale like that. You can "imagine" return would be better but it isn't is it? Look at VC returns. If you make 1,000 investments of $1million how many do you think make it to being a billion dollars vs just losing your money? How many of those investments need ZERO more capital along the way - if they need more capital you either get diluted out or you're putting a ton more cash in.

Instagram is a ridiculous outlier to build an investment strategy on. How many other Instagrams have you heard of this year? Yeah... not 1,000.

People investing in Hyperloop type investments don't put all their money on one investment either. If you're an investment manager at a pension fund having to find places to park $500BILLION you cannot be managing million dollar investments, they're meaningless. Even an Instagram doesn't move the needle. You have to place lots of billion dollar bets.

funding over a billion dollars is not remotely intimidating to the large VC funds

I think the idea is that some of the companies in the $1 billion fund will succeed and others will fail. But, funding a single company/public works project to the tune of $1 billion would give some pause.

Not really. Funding comes in stages. No VC with a billion dollars in their pocket is making one bet. But for folks with a lot of money to put in this is not unusual for lifecycle.

Look at the actual amounts raised through multiple rounds by - not just the initial seed investment the time they IPOd by any big tech name of the last decade or two: Google, Cisco, Facebook... hell look how much Groupon burned through.

The idea that you cannot raise a billion dollars if you can deliver a return on it greater than capital markets would otherwise provide is nonsense.

Exactly. I'd even call 20x for Hyperloop in 5 years excessively optimistic. you'd need 15-20 years.

Then you run into the whole playing of the odds. Invest in 20 companies, hope 1.5ish make great returns. Even the largest VC firms can't afford that on mulitbillion dollar industrial ventures.

There is investment for new and groudbreaking technology, but it is much MUCH slower moving. We're talking 5 years of simply drumming up investors and getting everybody onboard before any ground breaking even happens.

Also, you have to be much MUCH more connected to your industry than say, an SV programmer. There is A LOT of checkboxes one has to hit before one takes on large scale industrial ventures. This shit doesn't get a proof of concept built up in a dorm room over the weekend, and it's user base isn't going viral to signal interest in the product if there is no product to use until you're already $4,000,000,000 in the hole....

"This shit doesn't get a proof of concept built up in a dorm room over the weekend, and it's user base isn't going viral to signal interest in the product if there is no product to use until you're already $4,000,000,000 in the hole...."

Exactly. And on top of that, there is no making serious pivots. If you're an Instagram and you realize that people don't just want to share photos, they want filters, you can make that change in minimal time. In a Hyperloop situation, you're designing something that you're hoping to convince people that they want, and if you build it and no one comes, you're massively screwed.

Not to mention that Hyperloop would have a massive cost over-run--I don't care what his estimates are. The cost of land-use alone is going to be just outright insane. If you run a pilot in CA, you're going to blow through several billion just securing basic land rights, running years-long ecological and wildlife surveys, determining impacts to groundwater/rivers, etc.

Then the real work begins. This would make the Big Dig look well planned & executed.

You just described in a nutshell why it's usually governments that make big infrastructure-scale and basic research investments.

Governments either don't care about ROI at all and are motivated by other factors (war, national security, national development, political points, etc.) or can afford to look at ROI over exceedingly long time scales.

However, governments are piss-poor investors. So what we get is private capital -- which invests in a lot of tiny incremental things -- and a lumbering beast called "public capital" that invests in big things in a clumsy, ineffective way.

Your argument is spot on, however, Hyperloop is just a different game altogether. It is an infrastructure project and those projects are usually bankrolled by governments (local + federal funds). Even SapceX seems to be largely supported by NASA contracts.

The only large SV project I've heard about in recent time is Planetary Resources, the company that wants to mine asteroids. But, the returns for that expensive project would approach the 1000x that SV firms wish for.

Admittedly both of these projects are quite ambitious, but I have to think that delivering equipment to asteroids, maintaining that equipment while it works, and then returning the output of that work to Earth, seems much more involved than building and operating a 300-mile evacuated pipe on the Earth's surface.

The investment capital for Instagram was about $57.5M, not $1M: http://www.crunchbase.com/company/instagram

Otherwise, this is the Occam's Razor answer to me, and I agree.

Private industry is just now beginning to fund things like Hyperloops and space ships - this was the role of nation-states in the past, because making trains and going to space was more costly and un-risk-manageable than any real business. The government doesn't have to build the trains, but they incentive the building.

Direct government construction of infrastructure is evolving, and in the coming decades, private industry will be more involved in the investment end of it too, because the price of production and the amount of public knowledge is increasing. I don't think our modern day robber barons have that much more real capital than their forefathers (comparing like people like Rockefeller to Gates), but it's cheaper to build a railroad, and there's the whole open internet and body of academic knowledge to consult.

Good analysis but I think its missing a few key factors:

1) The risk profile of each investment. It may be the case that Instagram is a riskier proposition. Specifically, the likelihood of success of Instagram once it is funded may be smaller than the likelihood of success of a funded Hyperloop project

2) There aren't enough instagrams to invest in or they are harder to identify than the hyperloop type projects

3) Even though the percentage return may be lower than the hyperloop, in absolute dollars, the hyperloop has a higher return. For investors looking to put more of their capital to work, this is attractive.

But an Instagram has a high risk of failure whereas a transportation infrastructure is a very low-risk investment. The Hyperloop would be better compared with a huge Internet infrastructure like Google Search than with Instagram.

> transportation infrastructure is a very low-risk investment.

That's a good one. Building out a technology that has never been proven in a lab, let alone at production scale, in an industry where 2x cost overruns are inevitable and 10x not uncommon, that is dominated by government spending and not private capital for infrastructure, you consider the hyperloop a low-risk investment.

It might work, despite the skepticism, but there is no sense in which it can be considered low-risk.

Yes, I was talking about expensive infrastructures in general. It would be crazy to start building the Hyperloop before some working prototypes have proven that it works. But the high-speed train has low risks of loosing money if highways and airlines are saturated (i.e. if the demand is there, which can be predicted by studies).

Another way is to get the governement to subsidize those kind of work.

Reply on news.ycombinator.com