Filtered for the future of the firm

17.57, Tuesday 23 Dec 2014


There are these cars around London that seem a bit like Zipcar - the car sharing service - but you can leave them anywhere. Or at least, in most parking bays. So they kind of float around. The app tells you where the nearest car is.

Which is like Uber with no drivers?

How far could this go?

No parking bays. I know Zipcar give you credit for getting the car cleaned. Or at least they used to. Could that be included… and gradually raise the credit you get until somebody is motivated to do the cleaning?

And to get them repaired? What if you get credit for car maintenance?

This is why I’m into the idea that companies are resource allocation markets.

What if the company doesn’t even supply any cars, just the marketplace. You get credit if you supply cars. Or rather, you get car-marketplace-currency, which is itself a fraction of the ownership of the marketplace, and the value of your share goes up as the entire system is utilised.

What else could this be applied to?

I’m playing with a startup idea at the moment that involves inventory and many concession stands, each of which need to be staffed. So I treated this as a toy… what if Bitcoin was used for resource allocation, what then? If someone supplied a concession stand, they would get paid in credit. If some staffed a stand, they would receive a wage in credit. If someone repairs a stand: credit.

Credit is backed by a certain number of shares in the company, so they’re worth something. When somebody purchases some of the inventory, that’s profit for the company; when that dividend is paid out, it is paid in proportion to the shares, and so the credit can be exchanged for dollars.

This would be simple, using Bitcoin.

Then I asked myself: What would be the benefits of running the business like this?


Currently, a business like this would track its assets, liabilities, etc, using double-entry book-keeping. The prepared accounts are used to manage the business and allow it to invest in more assets and achieve more income. But the accounts are a model. They’re a map of the territory, not the territory itself. They’re hard to maintain, and they don’t integrate well with the business.

But use Bitcoin? There is no book-keeping because your activities in running the business are the same as recording it.

More importantly for a startup business: The job of finding the right prices can be thought of like game balancing… like finding the right cost of wheat in the economy of Farmville. We know how to do that.

The accounts - and the business model itself - become something that can be iterated in code to achieve financial growth, just like A/B testing the code of a website to get user growth.

Okay, this is a long way off. The costs of setting up this system would be prohibitive. But when someone makes the tools…


So here’s a classic long read for the holidays: A Brief History of the Corporation: 1600 to 2100.

(…which was last in my head when I used to write weeknotes at Berg, back in week 315.)

The piece ends by speculating what happens after the traditional corporation is a spent force…

And when that shift happens, the Schumpeterian corporation, the oil rig of human attention, will start to decline at an accelerating rate. Lifestyle businesses and other oddball contraptions — the solar panels and wind farms of attention economics — will start to take over.


Without realizing it, the hundreds of entrepreneurs, startup-studios and incubators, 4-hour-work-weekers and lifestyle designers around the world, experimenting with novel business structures and the attention mining technologies of social media, are collectively triggering the age of Coasean growth.

Coasean growth? Coasean growth is fundamentally not measured in aggregate terms at all. It is measured in individual terms. An individual’s income and productivity may both actually decline, with net growth in a Coasean sense.

I don’t know what this means. But it makes me wonder.

The author names “Coasean growth” after the Nobel prize-winning economist Ronald Coase: He is best known for his work on transaction costs, social costs and the nature of the firm.


Why do companies exist? In The Nature of the Firm (1937), Ronald Coase put it down to transaction costs. In short, companies exist because it’s cheaper to have an organisation that does the necessary activity internally than to use the free market outside it. Why? Because using the market - the price mechanism - itself has costs.

Here’s a summary of Coase’s paper:

There are costs to using the price mechanism for coordinating economic activity. ‘transaction costs’ or ‘marketing costs’

For example, if you want someone to carry your goods from the warehouse to your shop, first you have to find someone. That’s tough. It’s easier when lots of people who need carriage and lots of people who provide carriage come together – on a website or in the Yellow Pages. That’s a marketplace. But at a certain threshold, it’s easier still to just employ those people.

Firms exist to economize on the cost of coordinating economic activity.

Firms are characterized by the absence of the price mechanism.

This insight is old, but it makes my head spin.

Because it has two implications.

First is that the internet has made much easier both forming marketplaces and negotiating prices. Amazon is a marketplace where buyers and sellers are brought together, and prices change fluidly. Uber has a bottled marketplace: It doesn’t employ its drivers, but they transact via Uber. And pricing is dealt with half by algorithm (increase the price till there are enough drivers) and branding (passengers never negotiate).

So as markets and pricing get easier still, firms can get much smaller – ad hoc value chains assembled out of code and culture, barely anyone working at the company at all.

(Though we have to ask: If online marketplaces are so efficient, how can Amazon afford to buy the third party book sellers on their platform? There must be some efficiencies to being inside the firewall of the firm.)

The second implication is that the firm, no matter how small, can have a kind of hinterland of value providers - a community of users who post pictures, drivers who transport passengers - who, although technically not inside the firm, are as part of it as a spider’s web is part of the spider, or a beaver’s artificial lake is part of the colony.

If the firm is thought of as contouring transaction costs, and these costs are radically lowered…

And if…

Not only are we realising that functions that used to be part of a firm are now outside it,

but also that functions that have always been outside of the firm should actually be thought of as being part of it - for instance, realising that the community of a website should be rewarded like workers or owners, such as when reddit is giving partial ownership of its company to its users - then we need a new understanding of what a firm is.


There’s some kind of fight online about some kind of technology something something. Dunno.

But in a smart piece asking people maybe to just chillax, Quinn said something that caught my eye:

This age has put a group of maladjusted geeks, of which I would happily count myself one, into an historical role of giving input into what human agency will mean long after we’re all dead.


We should take that seriously.

So, y’know, the idea that a corporation - an organisation for orchestrating human endeavour to deliberate ends - an entity invented in the 1600s, and entity which BY ITS INESCAPABLE LOGIC forces us into mass production, mass consumption, mass media, alienation and the loss of individuality, and all kinds of ugly inhuman shit… the idea that we can re-invent the corporation, and create new forms of it: That’s interesting.

The idea that we might create a type of organisation which is empowering, has local value which doesn’t mean everything gets coerced into the value of giant companies, is smaller, can be interrogated and critiqued because it’s just code, that avoids the priesthoods of capital and law.

That a company might be a fuzzy-edged thing, where consumers are owners too…

Back in the 1960s, the US Department of Defense funded the development of ARPANET which became the internet, a made-out-of-whole-cloth dropped-into-history collection of protocols, practices, computers and networks, which I reckon probably had to be created all at once, because it couldn’t evolve, incubated just like that.

And if DARPA came along now and said, Hey Matt, What Next?

I’d say: Make a little bottle-city company that embodies all of this. Consumer-owners, internal currencies for resource allocation, corporate governance as executable code, doing an actual interesting tractable not-too-ambitious thing. Half co-op, half lifestyle business, half startup. Show what happens when we use capital, instead of capital using us. Do it simply and elegantly. Make a little nest of these companies.

Then sit back and see what happens.

More posts tagged:
Follow-up posts: