[larsonf]

Why not go meta?

Sure, a company like Google directly does more good than Goldman. (Although, trivially, you could argue a taco stand does something arguably more important than inverse-indexed search) Non-trivially, though, in order for a company like Google to get financed, it needs to have investors. Now, you'd be right that this doesn't necessarily require bankers, but it does present an interesting problem.

When an investor makes an investment, she wants to make more money than she put in. Else, why do it? Ok, so the simplest possible way to get more money than you put in would be to just let the invested business produce cash from profits and receive distributions over time. Ok, great, but what if that takes 20 years and you need the cash sooner?

The other option is to sell the stake to someone else. But the second you do that, in fact, the second you even offer to do that, the person on the buying end has the same worry. The only possible way for this to work is for there to be a bunch of other people who are willing to buy or sell at a certain amount at a certain price. That, my friends, is a market.

But what if there's no one out there to buy or sell your stake? Or what if there's very few? You need a market but there is none. Well, what if a person could somehow make a market for you? Someone, who, because of their deep knowledge of your company, companies like yours, the way markets work generally, and a view of the future state of markets could come in and say, "Hey, listen, I know there's not a lot of people out there right now who are willing to buy or sell, but I'm willing to buy X shares of your company at Y amount and sell X shares of your company at Y + Z amount." You, as the investor, are loving it since you don't have to tell someone upfront, "Hey, I'm saying my entire stake, what's the price?" Because that would be extreme adverse selection and would lower your price. And further, you don't have to say, "I'd like to double my share," thereby hinting you know something and the price should be higher. A market is made: there's a standing price to sell at X, and a standing price to buy at Y.

This actually happens all over the place. Facebook, amazon, google, et al. make money by being platforms for other people to create content. Facebook as a platform is obvious. For google, they are really the platform for the internet itself. No matter what firm w're talking about, the only way for content-creators to be willing to share is if there are enough people to consume/listen. And for there to be enough people to listen, there has to be someone who aggregates. And for someone to aggregate it necessitates a huge amount of upfront investment. Meaning, for example, that Facebook had to build itself before a billion people used it. Yes, this was in stages. But at each stage, it required a huge amount of risk before other people were there to use it. The same with google. Same with amazon. You see, these firms are underwriting the act of sharing content. Their knowledge of the eventual payoff allows them to buy the time, commitment, etc before there is an obvious customer. They have made a market.

Now, while Google, Amazon, and Facebook are doing this for things on the internet. Goldman is doing it for the Googles, Amazons, and Facebooks themselves. Goldman is providing a platform for the businesses in general. And, not only that, for anything that has substantial future cash flows!

It might seem that anybody could do this. Like why not Tim Ferriss on Angellist-Underwrite? That isn't so clear. The only reason Goldman can afford to post a bid and offer (i.e. make a market in cash-flows generally) is to have a better idea of the value of that something than whomever they are transacting with. Goldman has to think, "Ok, we'll buy/sell at x/y price because we know the real price that is z and so we can make money by offering the spread x - y." And guess what? In order to know that, they have to spend what seems like a life-wasting amount of time getting into the minutiae of esoteric financial knowledge. Because, remember, it's not that the market-maker is just connecting buyers and sellers. The market-maker is actually fronting the money--bearing the risk, by posting a bid and offer--in order for there to be a guarantee of the market. So the stakes are huge.

Now, in the future, it will be Tim Ferriss on AngelList-Underwrite. But he will hire 10 guys/gals to help him analyze markets. And then will he will get a lot of software to help aggregate information. And then he will want to be near the action, so he'll get an office in SF. And before you know it, he will be Goldman Sachs in jeans.

In the end, the only way for a platform to exist of any type is for someone somewhere to bear the initial risk on an unconditional (not knowing if a person is buying or selling) basis. It doesn't matter if it's a Facebook, a taco stand, the Hoover Dam or asset-backed securities. Goldman (and the other mega dealer-banks) are facilitating a platform for the entire world to unload cash-flows: the platforms' platform. And if you pick Finance as a career, it will be thankless, often boring, hard and people will likely not understand what you do and will demonize you for it. But you will know that you are doing something special: you are building the platform for everything.