With that context in mind, I'll try to answer your questions:

> what or how would such a company form in the first place?

With private equity firms buying up land and businesses for rent-seeking, the main leverage for workers is their participation. So rather than creating new companies in already saturated markets, workers could organize and unionize to create uncertainty and manipulate a company's valuation before its sale. Basically bring its projected value low enough that it's no longer appealing to the buyer. Or even bring it into a range where workers can afford to buy it and make it a cooperative.

Of course that can backfire if the lower valuation makes it more attractive to vulture capitalists. In that case, it might make more sense for workers to leave and use their expertise to form a new cooperative that the old company can't compete with. This might work best in sectors like tech where experience and connections can matter more than equipment.

I've faced this lack of capital for entrepreneurs my whole life. And feel that there's perhaps half the capital available for Millenials and Gen Z as there was for my generation Gen X, who had perhaps half as much available as the Baby Boomers, due to ever-widening wealth inequality. So there's a huge opportunity here to create a means for workers to use their potential to borrow capital.

Since banks, venture capital firms and even HN have largely removed themselves from small business funding, if someone created something like the Building and Loan from the movie It's a Wonderful Life on a broad scale, they would find a huge untapped market.

> The initial capital would essentially be from the workers themselves. Then, what about the newly hired workers - do they have to pay to get in? or do they get a share of the equity once they're hired?

If we go by the Overthrowing Our Tech Overlords article in my parent comment, then the ownership equation might consider tenure (like for bonuses), not just quantity of funds invested. With relative value decided by democratic vote with one vote (voting share) per worker. Basically time becomes currency instead of just productivity, credentials, job title, etc, to honor the dignity of each individual.

With no tenure, new hires would have what amounts to a loan against their eventual share of the cooperative. But keep in mind that since the cooperative is the means of production, then once it's paid off, it produces income rather than requiring funds from investors. Most likely, employees would have to work for a time before participating in profit sharing, rather than having to come up with funds for their buyin. There are existing models for this:

https://www.wincofoods.com/employee-owned

https://www.publix.org/retirement-and-stock/profit-plan

https://www.nceo.org/research/employee-ownership-100

Unions do sometimes create barriers to entry by controlling how job openings are filled, which can affect unemployment, and cooperatives could be viewed as fully-unionized businesses. They can be gamed just like any other zero-sum game. So cooperatives would need to be mindful of this and be proactive in creating policies that aren't exclusionary. But since private companies aren't any better at this, it can be regarded as an open problem, and best solved through a democratic process.

> What about the initial workers who have paid capital - do they get it back if they choose to leave? or do they now own equity, regardless of being an employee there?

Judging by the WinCo and Publix models above, it sounds like employees are compensated in company stock instead of just money. So they can sell their stock when they leave the company and retire:

https://strixus.com/entry/how-winco-foods-turns-its-employee...

Twenty percent of what each WinCo employee makes goes into an ownership account, so they own part of the company. In six years, it is theirs to keep whenever they leave or retire. One employee said "knowing it's there and it's continuously growing is a huge contributing factor to me staying here to further my career."

At that rate of 18% compounded, by the rule of 72, it only takes about 72/18 = 4 years for the stock's value to double.

Workers in a cooperative might be compensated perhaps $20/hr, with $16/hr as pay and $4/hr as stock. I just ran the math for each year's compound gains:

  year    wage     stock   gain18% total
  1       16       4       0.72    4.72
  2       16       4       1.57    10.29
  3       16       4       2.57    16.86
  4       16       4       3.76    24.62
  5       16       4       5.15    33.77
  6       16       4       6.80    44.57
  7       16       4       8.74    57.31
  8       16       4       11.04   72.34
  9       16       4       13.74   90.09
  10      16       4       16.94   111.02

Employees earn more in compound gains on stock by about year 10 than in hourly compensation!

Looks like WinCo made about $7.2 billion for 20,000 employees in 2024, or $360,000 per employee:

https://www.zippia.com/winco-careers-1115631/revenue/

So it appears that more of their proceeds go to paying into stock than wages.

Note that this is comparable to how company pensions worked in years past, before financialization used bankruptcy to break up companies and avoid paying those "entitlements".

Contrast this with a typical private company job, where workers produce about 3-4 times their hourly compensation. So a private company that charges $50/hr might pay workers $20/hr. That other $30/hr goes to overhead, then owners and investors who enjoy perhaps a 10% return on stock markets.

Could workers save some of their money for the stock market, and/or set some aside in a 401(k)? Sure. Do they do that in practice, especially for middle-class, blue collar jobs lately? Perhaps not.

I think this fits with our notions of how much partners and the board are paid in private companies vs workers who never receive compound returns. It's why CEOs get millions of dollars in bonuses while workers rarely see raises that beat inflation. It brings up questions around why we structure companies the way that we do, and whether executives are really worth what they are compensated.

Keep in mind that none of this even considers the compound returns available through automation. We could ask questions like: is manual labor even necessary anymore, when we could all make more from compound gains, as capitalists suggest?

The key is that the owner class has capital while the worker class doesn't, so workers are currently unable to effectively capitalize on the returns of economic growth.

-

My personal opinion is that I don't think it's safe for retirements to be tied to employment at one company over a long period, because the company can fail. I also don't think that it's safe for workers to have to use their wages to invest for retirement individually, because without diligent regulation on capitalists, wages eventually stagnate and we see the late-stage capitalism of today with small or missing retirements.

So I'd vote for employee ownership so that workers can participate in economic gains that quickly exceed hourly wages, as well as give them a vote in company policies, while keeping the option to cash out at any time.

As usual, this got way too long, sorry. I couldn't stop until I went down the rabbit hole.