Exit To Community

Grafting the lessons of old cooperatively owned companies to the online economy.

Klaus Vedfelt/Getty
Credits

Nathan Schneider is an assistant professor of media studies at the University of Colorado. He is the co-author of “Exit to Community: A Community Primer” and author of“Everything for Everyone: The Radical Tradition That Is Shaping the Next Economy.”

Perhaps you remember the time, in the fall of 2016, when rumors swirled that Twitter might be up for sale. As a reporter, I’d come to depend on Twitter. I first started using it heavily during the Arab Spring in 2011, enabling me to follow events on the ground in real-time and connect with activists directly. It felt shocking to me that this tool, which had become so important to my work and my life, was also a commodity. The news reports said maybe Disney or Google would buy it, or maybe Salesforce. This all seemed so arbitrary.

With some nudging from friends, I wrote an article in the Guardian suggesting Twitter should be owned by its users instead. What if that mere feeling that I had of ownership, as a participant in the Twitter-verse, was actually expressed in how the company was owned? I’ve written a lot of articles over the years that propose fantastical ideas; usually, they just plunk in the water. But this time, people around the world actually started organizing. I wasn’t the only person who felt this way. And we had a head start: Many of us had already been working together under the banner of “platform cooperativism,” trying to graft the lessons of old worker-owned factories and member-owned credit unions into the online economy.

First, we circulated a petition campaign, asking Twitter to consider user ownership. Then we realized we had shares of Twitter stock in our community, and we could use those to make a shareholder proposal to the company itself. We had a bit of a legal battle with the company, which tried to strike down our proposal through the American Securities and Exchange Commission. But we won, and the proposal went forward. At the annual meeting in early 2017, we won nearly 5% of the vote — which doesn’t sound like a lot, but it was more than we expected, and it meant we could resubmit later. We also got a lot of attention: articles covered us in Wired, Vanity Fair and about a hundred other places. The press thought what we were doing was crazy enough to be interesting. And then life went on.

Or did it? Something weird keeps happening. Those of us working on platform cooperativism have tended to think of ourselves as radicals working against Big Tech. So it came as a surprise to me when, in 2018 and 2019, several of the biggest, baddest tech companies began filing letters to the Securities and Exchange Commission, asking it to allow them to share ownership with their users — to distribute company stock among the very user-workers the platforms refused to classify as employees.

These platforms included the likes of Airbnb and Uber. Like Twitter, these are multi-sided marketplaces that leverage user loyalty to entice fickle investor-owners. Their demand was a modest step, to say the least, yet it looked eerily like a step in the direction of the radicalism we had been calling for — building democracy into the businesses themselves, into how they are owned and governed. The Airbnb letter even articulated a lot of the rationales that we had been harping on in our little radical world for a few years.

The companies, of course, had their own interests at heart — several seemed to think it might buttress going public on Wall Street — but the letters made me wonder if a deal could be struck. At a time when regulators are starting to clamp down on the big platform companies, what if we could meet the companies in the middle and create a new kind of paradigm, a kind of race to the top? What if companies competed with each other not to extract more data from their users or suppress working conditions, wiping away rights that people have fought for over generations, but instead to offer better deals through shared ownership?

For the same reasons that startups often share stock with early employees, the platforms would get economic alignment with their users and loyalty. Society as a whole would get more built-in accountability for our platforms, a kind of accountability that doesn’t rely on any one jurisdiction or legislature. What if, in turn, we had a new kind of story available to tech startups, a new mythology?

Twitter CEO Jack Dorsey testifies in the Senate in 2018. (Drew Angerer/Getty)
Another Narrative

If you’re forming a startup, there are generally two kinds of stories to choose from about what the startup is for. Keep in mind that startups are companies that are trying to take over some subsection of the world. It’s ambitious stuff, so they tend to take on lots of early investment. They get somebody to give them a lot of money so they can hit the market with disproportionate force. And in order to pay that investment back, they need what’s called an “exit,” which usually comes in two forms: You’re either acquired by a larger company or you go public, selling your company on a market where people can trade your shares based on their speculations about what it’s worth. In both cases, you’re passing the company that you’ve worked to build off to new owners, who in turn might be buying it just to convince future buyers to pay even more for it later. It’s a weirdly normal pyramid scheme of capitalism.

What if there were another way? What if a startup that successfully builds a community could opt for an exit to ownership by that community?

This is not so crazy. The more I look, the more I find activists, lawyers, investors and entrepreneurs around the world who are exploring options along these lines, often while seeking to address some of the deep problems of internet culture. Community ownership could be a way of rooting out the abuses that we often see with the billion-dollar “unicorn” startups that make it big and lose their souls to satisfy financial markets, building accountability into the structure of the companies themselves. Too often, tech platforms get caught in the position of trying to extract as much as they possibly can from their users on behalf of their investor-owners, getting away with as much as they think they can.

“What if a startup that successfully builds a community could opt for an exit to ownership by that community?”

There’s also a category in the Silicon Valley lexicon called “zombies” — the companies that don’t quite become unicorns but aren’t failures either. They’re a dead weight on the investors’ balance sheets because, for one reason or another, nobody wants to buy them. Even if a company like that is making a profit and pleasing users, it can be worthless in the eyes of financial markets if it isn’t poised for the kind of exit options that investors need. What if the exit to community strategy could enable these zombies to be cured of their condition?

I have been working with Morshed Mannan, a legal scholar at the Leiden Law School, to sketch out different strategies for this based on what has been done before and what seems at least approximately plausible. Allow me to introduce you to our imaginary company, CoSocial. It’s a social network that managed to sprout a gig marketplace and payment system. It’s more or less a combination of Reddit and TaskRabbit — or, if you’re familiar with the Chinese online ecosystem, just think WeChat.

CoSocial has positioned itself as an upstart alternative to the big platform giants. It has taken a small, energetic slice of the market share, especially among people who want to opt out from the giants. Fees from marketplace payments have even made it at least mildly profitable. But its growth has stalled. Every pathway toward unicorn status seems to point in a direction that would drive away its most loyal users — incentivizing clickbait, adopting dismal labor standards — which puts the whole enterprise at risk. The founders and early investors are at a loss about what to do until they started thinking about their user community itself as the answer, and they come up with three ideas for how to proceed.

Option 1: Trust Buyout

The first is for users to buy out the investors. But let’s be reasonable: Even the most loyal users probably wouldn’t have the means or inclination to pony up cash to buy the company. Instead, CoSocial could follow the pattern of the U.S. employee stock ownership plan, in which a trust operating on the employees’ behalf borrows money to buy shares in the company — except in this case, on behalf of users.

At first, commercial banks might be reluctant to participate in the deal, but if a foundation committed a significant portion of mission-related investment funds, for instance, other institutions might join in. Within a few years, CoSocial’s profits would pay off the loan and the users’ trust would become able to keep and distribute the dividends. At first, the company would appoint an initial trustee for the trust, but within five years, the position would be elected by user-members.

This kind of shared pool could take a few forms. For instance, CoSocial’s gig workers might want to supplement their incomes with a proportional cut of the profits on a regular basis. But that benefit might be less important for users primarily interested in the social media features; their contribution to the profits might be much smaller when divided among individuals, but significant when pooled.

In that case, perhaps the dividends derived from their use could become a user-controlled fund for developing features or other initiatives that users choose through a participatory budgeting process. Those funds could also be used to support an independent oversight board — a user-accountable version of the one Facebook has opted to develop and pay for, which incidentally also uses a trust to ensure its autonomy from the company. A users’ trust could look different in different contexts. In CoSocial’s case, it could enable investors to exit at least as comfortably as they otherwise might, while also attracting new waves of users by doubling down on high-road practices.

Option 2: Federation

CoSocial’s leaders might, however, decide they couldn’t win at the game of monopoly. Big tech’s market power is overwhelming. So, they realize that CoSocial could grow by distributing power rather than centralizing it. This way, they could foster deeper trust with a wider range of stakeholders.

To get started, CoSocial would begin opening up its software platform to independent companies and organizations. These “nodes” run open-source CoSocial software to create their own distinct CoSocial communities. They could do this at no cost if they wanted, but in order to interoperate with other nodes more smoothly and to use CoSocial’s robust payment network, nodes could join a new cooperative federation called CoNet, which is co-owned by the nodes. Through it, they pay dues that cover an access fee to CoSocial, and they establish standards that member nodes would have to meet. Nodes proliferate as organizations see benefits in creating their own white-labeled social networks and marketplaces. Some of these organizations might themselves be cooperatives, co-owned by their users, but they are not required to be.

As CoNet grows and contract negotiations with CoSocial become more complex, the cooperative might recognize that there is a better way. Based on robust membership growth, CoNet could obtain financing to acquire a controlling stake in CoSocial, buying out the early investors and making the original platform, in effect, a subsidiary of the cooperative. The groups that had come to rely on CoSocial’s network would then be in control of it.

“To create a powerful network, sometimes the best strategy is to spread power out across it.”

This kind of strategy does have precedents in legacy media. Notably, for instance, the Associated Press was formed as a cooperative of competing newspapers in the mid-19th century. Today, it remains a bastion of non-polarized, non-fake news, with member organizations across the U.S. political spectrum. Similar cooperative agencies exist worldwide, like Agence France-Presse or the Press Trust of India. These cooperatives enable local news providers to have access to global economies of scale — while keeping their local focus and accountability.

A similar thing happened when the visionary banker Dee Hock persuaded Bank of America to spin off its struggling Bank Americard product into a cooperative co-owned by the banks that used it, giving birth to Visa and the credit card system as we know it. (Visa has since demutualized and adopted increasingly predatory practices, an outcome that Hock mourned.) CoSocial might learn the same lesson as Bank of America did: To create a powerful network, sometimes the best strategy is to spread power out across it.

Option 3: Tokenization

CoSocial’s third option might seem even more remote than the others. Here, we turn to the blockchain technology that birthed Bitcoin and its thousands of cryptographic stepchildren. This technology creates opportunities to enable shared ownership among users of a platform through digital tokens, and — depending on how the regulatory situation develops — token-based ownership could be simpler and more flexible for trans-national platform communities than conventional stock.

CoSocial could start allowing the use of cryptocurrencies for payments in its internal marketplace, which means lower transaction costs for micropayments and users working across borders. Users would begin relying on CoSocial as a wallet for holding their tokens in multiple cryptocurrencies, giving leadership an opportunity to solve their investors’ liquidity problem. They could change the company’s articles of incorporation to represent all shares of stock as tokens on a public blockchain, CoShares, which users could then buy with their holdings in other cryptocurrencies.

The company would also issue these tokens as rewards for activity and loyalty on the platform. As the value of CoShares tokens grows on cryptocurrency markets, users might find it a worthwhile investment while they weren’t otherwise using their crypto-tokens for marketplace transactions. To encourage the system’s adoption, CoSocial could build a dashboard that enables users to easily manage, trade and spend their CoShares tokens. Before long, most early investors would sell their CoShares tokens to users.

The dashboard would enable CoShares holders to participate in governance, such as by electing directors and making shareholder proposals. But in order to prevent outside interference in this unusually transparent governance process, the company could allocate governance powers to users who hold both tokens and reputation on the platform. Effectively, those most involved as both users and investors would become those in charge.

Facebook CEO Mark Zuckerberg leaves a meeting on Capitol Hill in 2019. (Samuel Corum/Getty)
Zebras, Unite

CoSocial is an imaginary company, but these strategies are not so far-fetched. In early 2019, for example, the founder of the Shanghai electric-car company NIO transferred 50 million shares into a “user trust.” While the founder retains voting rights over the shares, the trust enables its members to help decide how to use a portion of the company’s eventual profits. This was not an act of charity; in the disclosure documents, you see a clear business rationale: an attempt to align the interests of the company with those of its customers and potential customers.

Or consider Twitter CEO Jack Dorsey’s plan, announced in late 2019, “to develop an open and decentralized standard for social media” — one in which Twitter as we know it would be a mere client. Dorsey said that he saw potential in spreading out responsibility for moderation, recommendation and other processes that were becoming unwieldy under the control of a single company.

And then there’s Facebook’s Libra proposal, which originally set out to create a massive public cryptocurrency network that would rival conventional, state-backed currencies — a terrifying idea, but one that demonstrates the appeal of crypto-networks for mainstream platforms. This proposal wouldn’t represent Facebook shares in tokens, but it would at least conjoin the company’s fate in some significant ways to a system under the control of active stakeholders. In both cases, the crypto-network could swell in importance to the point of dwarfing the conventional corporations that gave rise to them.

At a time when politicians in the U.S. and Europe are starting to get serious about antitrust enforcement, companies could use federations, user ownership or tokenization to preemptively spread out their market power while continuing to grow their networks. If they don’t, regulators could use a structure of this sort for forced restructuring.

“Enabling different kinds of exits could enable different kinds of people and places to participate in creating ambitious startups.”

Strategies like this, however, could matter most of all for the kinds of tech companies that we haven’t seen yet. Right now, we’re tapping such a tiny subsection of the creativity that could be in the startup ecosystem. The overwhelming whiteness and maleness of Silicon Valley indicate that a lot of potential energy is missing. By far, most of the population of the world doesn’t have the chance to participate in the conventional venture-capital world. Enabling different kinds of exits could enable different kinds of people and places to participate in creating ambitious startups.

I take inspiration from a group called Zebras Unite, which is an international network of startup founders who are largely women and people from underrepresented groups. (Unicorns are imaginary, zebras are real; unicorns are solitary, zebras run in packs; unicorn horns point at skyward profits, while zebras wear the black-and-white of business and purpose.) A lot of people recognize that tech startups have a diversity problem, but the Zebras community goes further — they say the reason equitable representation never seems to work is that we don’t have equitable business models.

As long as we rely on financing models that depend on maximum power and privilege, the participants will keep on looking like those with the most power and privilege. The Zebras say that they need different kinds of models because they are trying to build companies that are actually focused on their communities. They’re trying to solve real problems, not just disrupt them away with superficial apps. To do that, they’re exploring better forms of finance, including by building a cooperative that co-owns a venture fund. And they’ve been partners with me in trying to make exit to community, or E2C, into a reality.

A community-oriented startup economy is not something that can happen overnight. There are big companies out there trying to do things as I’ve described, and there are earlier-stage startups that would love to have this kind of trajectory. But exiting to community is harder than going into the capitalist pyramid scheme. Building a startup is hard enough, and rejiggering a whole system at the same time seems nearly impossible. Those of us who aren’t throwing ourselves into building a startup day after day can help make it easier.

The challenges can be divided into three parts: pioneering, policy and culture.

The pioneering part is that we need to find opportunities to experiment where we can — especially among those who hold privileged positions in the startup economy, like investors, lawyers and accelerators. Take a chance on an opportunity where an E2C looks like it might be possible. In order to change the system, we need examples that point the way and offer lessons. There’s an investment fund called Purpose Ventures that has staked its model on facilitating exits to “steward-ownership,” as the company puts it. By creating a fund, Purpose also enables other investors to try the model by investing their own money, generating returns by collecting portions of revenue, often, rather than through sell-offs.

“What a group of people creates together and relies on should be those people’s common possession.”

Policy is not something most startup people like to think about, but it structures everything. Before an intentional policy change in 1979, for instance, venture capital in the U.S. could barely get off the ground. The employee stock ownership plan wasn’t able to spread to millions of workers until, after a few successful experiments, it became an official retirement plan in 1974.

As pioneering experiments with an E2C strategy proceed, we should recognize how much the present system is slanted toward creating companies owned by outside investors and what kind of policy changes could overcome that bias. For instance, a public loan-guarantee fund could make it easier to finance community exits, since there won’t often be one rich person who can pony up collateral. There could be tax incentives too, based on the recognition that widely distributing capital ownership is a public good.

Exiting to community could even become a kind of expectation. Like Elizabeth Warren’s proposal that companies over a certain size should have employees on their boards, what if platforms over a certain size had to have users on their boards and some means of sharing their profits with the users who create them? We will also need to contend with how international the make-up of platforms’ user bases are. The challenge of E2C comes with an added and long-overdue challenge of making the governance of platform companies as transnational as the companies and the networks themselves.

We won’t get anywhere without culture. Culture is where new business models begin, and it’s what enables them to take off. I’ve been trying to spin some mythology that we can tell and retell. There need to be many cultures, as many as there are communities, roughly united in the common-sense recognition that what a group of people creates together and relies on should be those people’s common possession. Like any good mythology, we will need to share our stories and add new ones to the mix, learning from them and adjusting them — until what once was a batch of out-there ideas has become a commonplace reality, an available opportunity, a thing that reasonable people can work toward and achieve.