Ownership and the Worker

Volume V — Labour, Value, Flourishing

Nine movements, one stake.

THE PLATE

A woman in a soft knit sitting on a sofa, drinking from a cup held in both hands, pale window light. — Plate V.03The Statement on the Kitchen Table.A share in the company arrives once a year as a number on a page. What that number is worth to the person holding it is a different number, and almost nobody has been shown how to compute it.

THE LETTER

You have probably been told, at some point in the last ten years, that employee ownership is the answer — and told it by somebody sincere, with a case study and a warm anecdote about a janitor who retired with a million dollars.

The janitor is real. So is the case study. And the sincerity is not the problem.

The problem is that almost every account of employee ownership you will encounter is written by somebody who wants more of it, which means the evidence arrives pre-selected, the effect sizes arrive pre-rounded, and the one piece of arithmetic that matters most to the person actually receiving the shares — what is this worth to me, given that my job is already in this building — is almost never done at all.

This chapter does it. Not to argue against employee ownership, which on the evidence is a good thing that we should have more of, but because a case made without the hard number is a case that collapses the first time somebody in a finance function runs it. And the hard number turns out to be the most useful thing in the whole subject, because it does not say stop. It says exactly how much, and funded which way — and those two answers, taken together, rearrange the entire design.

So: the ESOP record with its real identification problems. The British employee ownership trust, now growing faster than anyone planned for. Worker co-operatives' measured survival, which is genuinely better and less cleanly better than you have been told. The productivity evidence at its honest size. And then the instrument: a ledger that prices a given stake against the concentration risk it creates, and a ceiling that falls out of it.

Chapter I.11 dealt with the seller's side of this — succession, steward ownership, what it costs an owner to hand a company on. This chapter stands on the other side of the same table and asks the employee's question instead.

— The Editors

DISCOVERY

What is already working

Start with the largest working example in the world, because it is not a pilot and it is not a foundation's demonstration project. It is a supermarket.

Publix Super Markets established its PROFIT Plan in 1974. Its Form 10-K for the year ended 27 December 2025 describes it as the largest employee-owned company in the United States, with more than 260,000 employees. The same filing says something more interesting than any ownership percentage: the common stock is not traded on an established securities market, and its price is determined by the board, informed by an independent valuation, and restated quarterly. Substantially every transaction in it is between the company, its people, their families and the retirement plans.

Half a century of that, through every recession in living memory, in an industry with famously thin margins and famously high turnover. Whatever else is true of employee ownership, it is not fragile at scale and it is not confined to professional services and craft bakeries. (You will see "about 80 percent employee-held" quoted for Publix in many places. It is not in the filing we read, so it is not in this chapter.)

In France, the sociétés coopératives et participatives — the SCOPs — have been measured against the rest of the French economy for decades, and they survive better. Virginie Pérotin's synthesis for Co-operatives UK reports 80 to 90 percent of worker co-ops still trading after three years, against 66 percent of French firms generally; and for co-ops formed out of existing businesses, five-year survival of 61 to 82 percent against roughly 50 percent. The oldest SCOP still trading was founded in 1882.

In Italy, the Marcora Law of 1985 did something unusual: it let workers capitalise their unemployment benefit as equity in a co-operative that buys the firm that was about to make them unemployed, with an institutional investor matching what they put in. Research at Euricse found buyout co-operatives formed after 2007 showing a three-year survival rate of 87.16 percent against 48.30 percent for Italian enterprises generally. The state's money went into a going concern instead of into unemployment insurance, which is the same money, routed differently — the move Chapter I.01 called the discovery move.

In Britain, something has happened fast enough that the statistics are still catching up with it. The employee ownership trust was created by the Finance Act 2014, and in that year the Employee Owned Business Register counted 135 employee-owned businesses in the whole country. As at March 2026 it counts 2,824, with about 548,000 employee owners — 20.9 times the 2014 figure, a compound 28.8 percent a year for twelve years.

Read that alongside the other number, because they point different ways in the same year: 560 transitions completed in 2024 and 500 in 2025, a 10.7 percent fall. The stock grew while the flow fell. Both are true, and a report that quotes one without the other has told you half of it.

The names are ordinary businesses and that is the point. Riverford Organic Farmers moved 74 percent into trust in June 2018 for its 650 staff. Aardman Animations — Wallace and Gromit — sold 75 percent into trust on 5 November 2018. Richer Sounds transferred 60 percent on 10 May 2019, and Julian Richer separately paid every employee below director level £1,000 for each year worked — about £4 million across some 500 people, implying an average of eight years of service, which tells you about the retention that preceded the transfer rather than followed it.

And the finding that matters most for a volume about labour: employee-owners lose their jobs less. Kurtulus and Kruse, working across more than eight thousand public companies from 1999 to 2011, put it as an elasticity rather than a rate — which is why it is the sturdiest number in this chapter. For every one percentage point rise in national unemployment, employee-ownership firms reduced employment by 1.3 percent; comparable firms reduced it by 3.0 percent. That is a ratio of 0.43, and on a workforce of a thousand it is seventeen people still in work for every point the economy falls. The same study finds employee-ownership firms over 21 percent more likely to survive, and about 23 percent more likely where worker ownership exceeded five percent of shares.

Co-operatives show the same behaviour from the other direction — Pérotin's finding is that when demand falls they adjust pay first and employment second, which is the exact inverse of the conventional firm's reflex.

That is five bodies of evidence, on three continents, pointing the same way. The structure survives, it employs, and it holds people through downturns. Everything that follows is an attempt to say how much — because a good thing whose size is unknown is a good thing nobody can budget for.

THE ARITHMETIC

What works, what does not, and where the line sits

First, the effect, at its honest size.

The most careful synthesis is O'Boyle, Patel and Gonzalez-Mulé (2016), across 102 samples and 56,984 firms: a small, positive, significant relation to firm performance, mean correlation r = 0.04 — 0.16 percent of the variance, leaving 99.84 percent to everything else. Kruse's wider count says it differently: of 129 studies, about 86 positive, 13 negative, 30 neither. Consistent in direction, small in magnitude, and no field experiment has ever been run.

Kruse's own estimate is that productivity rises 4 to 5 percent in the year an ESOP is adopted and stays there; Census data linked to ESOP records puts it at 6 to 7 percent in American manufacturing plants, strongest where profit sharing runs alongside. That is a step, not a trend. Ten years of a 5 percent step is worth 50 percent of one year's output; ten years of 5 percent growth would be 62.9 percent in the final year alone.

Second, what carries the effect. Doucouliagos's 1995 meta-analysis of 43 studies separated ownership from participation in decisions, and both columns matter because secondary citations print one:

                                  mean r   weighted r    K
  worker ownership, all firms       0.06       0.03      17
  participation in decisions        0.21       0.06      19
  codetermination only             -0.14      -0.11       3

Participation outperforms ownership 3.5× unweighted, 2× weighted, and codetermination alone is negative on both. A share of the result without a voice does little; a voice without a share does slightly less than nothing.

The field's most-quoted null agrees. The GAO's 1987 study of firms adopting ESOPs in 1976–79 found no consistent improvement — except that of every factor examined, only employee participation showed a significant relationship with productivity. And the average stake in its sample was 8.5 percent, one twelfth of the company. An 8.5 percent stake is not employee ownership; it is a benefit. A null from an underpowered test on a tiny dose is a statement about the test, and the GAO said so itself.

Third, the identification problem — arithmetically, not rhetorically.

Wiefek's 2017 study of NLSY97 respondents aged 28–34 is the most-quoted work in the field: employee-owners showed 92 percent greater median household wealth, 33 percent higher wage income, 53 percent longer tenure. The wage and tenure absolutes are published and reproduce exactly — $40,000 against $30,000, 5.2 years against 3.4.

The wealth absolutes are not, so what follows substitutes a base from a different survey and age band: an order-of-magnitude check, not a replication. The Federal Reserve's 2022 Survey of Consumer Finances puts median net worth for a household headed by someone under 35 at $39,000; 92 percent above is $74,880, an implied gap of $35,880. Build the account — eight years of an employer contributing 8 percent of a $45,000 salary at 8 percent — and it is $38,292.

The whole measured wealth advantage is 1.07 times the account balance. The study may be observing the plan rather than a behaviour, which makes the money no less real and the causal reading unsupported. The honest claim is narrower and still good: employee ownership puts an asset in the hands of people who would otherwise hold almost none.

The survival figures have the same shape. Italian buyouts are measured against all Italian enterprises, a population including firms in their first year, while a buyout is a going concern on the first morning. Correct for it crudely — assume 10, 15 or 20 percent of the comparison died in year one — and the 1.80× ratio falls to 1.62×, 1.53×, 1.44× and does not disappear. Both halves carry weight. Pérotin says the same herself: co-operatives may simply be more selectively created.

Fourth — the instrument. What a stake is worth to the person holding it.

A median participant at 55. Pay $60,000. The employer has put 8 percent of pay into the ESOP for 15 years at 8 percent; she has saved 5 percent herself into a diversified account at 7 percent.

  ESOP account                    $130,330
  diversified 401(k)               $75,387
  total financial wealth          $205,717
  stake as a share of it             63.4%

Her employer's equity runs about 40 percent volatility a year; a diversified index about 17. At a beta of 1 the market pays her for the covariance term and nothing else:

  total variance of the stock          0.1600
  covariance term                      0.0289
  idiosyncratic variance               0.1311
  idiosyncratic volatility, per year    36.2%

She carries thirty-six points of volatility a year for no expected return at all. Under lognormal returns and constant relative risk aversion the certainty-equivalent ratio has one term, exp(−γ · w² · σ_ε² · T / 2). At γ = 2 over ten years:

    w          value of $1 of employer stock
  0.10                 87.0 cents
  0.20                 74.5 cents
  0.35                 57.6 cents
  0.634                35.4 cents      <- our median participant
  1.00                 27.0 cents

Her $130,330 stake is worth about $46,160 in diversified terms. The concentration costs $84,170, and nobody ever put that on the statement.

Check it by a route sharing none of these assumptions. Lisa Meulbroek priced company stock for an undiversified holder by the capital asset pricing model and found employees give up on average 42 percent of market value over five years. This model at five years gives up 37 percent. Two methods, no shared machinery, a few points apart — and that agreement is the only reason to believe either.

Note the shape, because the shape is the design. The benefit scales with w; the cost scales with w². Double the stake and you quadruple the risk charge. So there is a hurdle — the excess return the shares must earn over an index merely to justify holding them, λ*(w) = γ · w · σ_ε² / 2: 1.31 points a year at a tenth, 2.62 at a fifth, 4.59 at a third, 8.31 at her 63 percent.

She needs her employer to beat the market by 8.31 points a year, every year, for ten years. Nothing in the productivity literature supports that. Invert it and the ceiling falls out, w* = 2λ / (γ σ_ε²): at a premium of 1.0 points a year, 7.6 percent of financial wealth; at 1.5, 11.4 percent; at 2.0, 15.3 percent. Somewhere around a tenth to a fifth is where the arithmetic stops. She is at 63.

Fifth, the correlation the variance cannot see. All of that priced the variance of her wealth, not the thing that makes this different from any other concentrated position: the stake and the wage fail in the same week. At a 2 percent annual hazard the chance of failure across her ten remaining years is 18.3 percent, and the expected loss of stake alone is $23,841 — arriving alongside a displacement that Davis and von Wachter price at 2.8 years of pre-displacement earnings when unemployment is above 8 percent, which on her salary is $168,000. The lost pay is not a cost of owning; it would befall a worker who owned nothing. What ownership adds is that the savings die in the state of the world where a dollar is worth most to her — the same risk read a second way, never a second charge to be added.

Now the honest negative, and there are two.

United Airlines. In 1994 employees acquired about 55 percent of UAL for wage and benefit concessions reported at $4.8bn to $4.9bn — roughly $89m per percentage point of equity. In December 2002 the company filed for Chapter 11 and the equity was extinguished: −100 percent on the largest purchase of shares American workers have ever made. In 2005 the Pension Benefit Guaranty Corporation took over United's plans — $9.8bn of underfunding, a $6.6bn claim on the insurer, about 122,000 participants. Per participant, $80,328 of underfunding, of which $54,098 fell to the insurer and $26,230 simply did not get paid. Pilots lost up to about half their benefit. The wages, the equity and part of the pension went in one event.

Enron. In January 2001 the 401(k) plan held $2.1bn, 62 percent of it in Enron stock — about $1.302bn. The stock fell 94 percent that year; roughly $1.22bn was destroyed inside the plan. Enron was not an ESOP. It was employees' own elective deferrals plus an employer match paid in stock. Hold that distinction, because it does all the work.

Now the cut: price the stake by how it was funded, not by how large it is.

Ask what the employee gave up.

At Enron, a diversified dollar. Every dollar in company stock could have been in an index. One-for-one, the stake was worth a third to a half of the alternative. The trade was a loss before the fraud.
At United, a wage — cash, contractually owed, already earned, with no volatility at all. They paid a dollar of certainty for a stake worth cents, and the cents went to zero.
At Publix, Riverford, Aardman, Richer Sounds and every properly built ESOP or EOT, the employee gave up nothing. The employer funded it, the trust borrowed against future profit, the founder transferred shares. The alternative was not an index fund. It was an empty account.

And thirty-five cents is not zero. A gifted stake is $46,160 of diversified equivalent our participant would not otherwise have had, with no hurdle to clear because there is nothing on the other side of the trade.

So the concentration critique — correct, omitted by advocacy, and made here as sharp as it will go — is not an argument against employee ownership. It is an argument against paying for it. Kruse says it in one line in his own review: the financial risk is substantially reduced when ownership supplements rather than substitutes for regular pay. The arithmetic turns that caveat into a design rule.

Three rules follow, and they are the chapter.

Never buy it with wages. Concessions, deferrals, matched contributions in company stock — each converts a certain or diversified dollar into a stake worth a fraction of it.
Cap the concentration. Above roughly 15 to 20 percent of a worker's financial wealth, the stake needs a premium nobody can promise.
Pair it with a voice, and fund the exit from the first day. Ownership without participation measures near zero, and a stake that cannot be converted cannot be capped.

DREAM

What becomes ordinary

In the firm that has understood this, the annual statement has two numbers on it and the second one is new.

The first is the familiar one: the value of the employee's shares, as valued by the independent appraiser, as of the plan year end. The second sits directly underneath it and reads your stake as a share of your total retirement savings, and the ceiling this plan holds itself to. If the first number has pushed the second above the line, the statement says so, and it says what happens next — because the next contribution goes to the diversified side rather than the concentrated one, automatically, without anybody having to ask.

Nobody experiences this as a restriction. It reads the way a seatbelt reads.

The 401(k) alongside the plan holds no company stock at all, and nobody proposes that it should, because the firm has understood that a match paid in its own shares is not generosity — it is asking the employee to fund the ownership out of their own diversification, which is the one structure the evidence condemns without qualification.

The diversification election at 55 is not a form buried on an intranet. It arrives in an envelope with a covering letter, a worked example using the employee's own balance, and a named person who will sit down and do it with them. Take-up is measured and published in the annual report beside the safety statistics, because the firm regards a participant who reached 65 at 90 percent concentration as an operational failure of the same kind as a lost-time injury.

And the repurchase obligation has a fund behind it. Not a projection in a consultant's model — a segregated account, contributed to annually, sized against the actuarial profile of the workforce, disclosed. When a machinist retires after thirty years the money is there on the day, and the company does not have to choose between paying her and buying the next machine.

The result is not a different company. It is the same company, with the risk placed where it can be carried. The people in it own a share of what they build. They also own a diversified portfolio, a pension that is not their employer, and a wage that was never traded for equity. Those things are not in tension. They were only ever in tension because nobody did the arithmetic.

DESIGN

The structure that gets there

Six components, in this order, and the order is load-bearing.

One: the funding source, decided first and written down.

Before any structure is chosen, answer one question in a sentence: what does the employee give up to get this? The permitted answers are nothing and a share of future profit that does not yet exist. Every other answer — a wage concession, an elective deferral, a matched contribution in stock, a purchase out of take-home pay — puts the transaction on the wrong side of the arithmetic above. Write the answer into the plan document, where a future board will find it.

Two: the vehicle, chosen by jurisdiction rather than by enthusiasm.

In the United States, an ESOP under ERISA, with the company contributing or the trust borrowing and repaying out of profit.

In the United Kingdom, an employee ownership trust under TCGA 1992 ss.236H–236U — and the terms have moved twice in eighteen months. The Autumn Budget of 30 October 2024 required trustees to be UK resident as a single body, to take all reasonable steps to pay no more than market value, and to be fewer than half former owners or connected persons, and extended the clawback of the seller's relief to the end of the fourth tax year after disposal. The Budget of 26 November 2025 then cut the relief from 100 percent of the gain to 50 percent — an effective 12 percent against a 24 percent main rate, where it had been nil.

The incentive is worth half what it was, and the flow of transitions fell in the same year. Chapter I.11's conclusion — that the deferral, not the relief, decides the transaction — is strengthened by the halving, not weakened: the smaller the tax advantage, the more the deal turns on how fast the seller is paid.

Where co-operative law is strong — France, Italy, Spain — a worker co-operative with an external financing partner is the better instrument, and Marcora is the template: the worker's capitalised unemployment benefit as equity, matched one-for-one.

Three: the concentration cap, in the plan document.

State the ceiling as a share of the participant's total retirement savings, not as a share of the company. Fifteen to twenty percent is where the arithmetic puts it; a plan may choose higher and should then name the return premium it believes justifies the choice. Enforce it by redirecting future contributions, never by forced sale — a forced sale is a tax event the participant did not choose.

Four: the diversification machinery, built as though somebody will use it.

The American statutory floor is IRC §401(a)(28)(B): a qualified participant — age 55 with ten years of participation — may diversify up to 25 percent of the post-1986 account across the first five years of a six-year election window, and up to 50 percent cumulative in the sixth.

Know what that is worth, because it is the most valuable unexercised right in the field. Our participant, holding 63.4 percent concentration for ten years, has a stake worth 35.4 cents on the dollar. Take the election in full and the same stake is worth 67.5 cents.

  the election is worth  32.1 cents on the dollar  =  $41,867

It costs nothing, and at a beta of 1 it gives up no expected return — the expected return was never the compensation for the risk being removed. Take-up among eligible participants is not published anywhere we could find, and that gap is reported as a gap rather than filled with a guess.

Five: the put, and the money behind it.

Shares in a private company are worth nothing to a retiring employee unless somebody must buy them. In an American ESOP that is the §409(h) put option; its economic content is the repurchase obligation, and that is where mature plans fail — the next movement.

Six: participation, which the evidence says does more than the ownership.

Last in sequence, first in effect. Doucouliagos measured participation outperforming ownership by two to three and a half times; the GAO, finding nothing else significant, found participation significant. A stake with no voice is a lottery ticket with a payroll number on it — and a voice with no stake is worse. So build both: a beneficiary representative on a trustee board whose majority does not report to the chief executive; a pass-through vote on sale, merger and dissolution, because those are the decisions that convert the stake into something else; standing work groups with real authority over method, scheduling and quality, which is the exact form the GAO found to matter; and an annual meeting where the valuation is explained by whoever performed it, in language a warehouse supervisor can check and argue with.

Ownership buys the right to run this. Participation is what turns it into a number.

DESTINY

How it holds when nobody is pushing

The failure mode of employee ownership is not ideological. It is a cash flow.

Every share the plan issues is a share it must one day buy back. The NCEO's 2023 survey of 248 ESOP companies found a median of 5 percent of outstanding shares repurchased in the most recent year, a mean of 6 percent, and annual contributions averaging 11.7 percent of payroll. Check that from the other direction: a company on a 10 percent EBITDA margin at a 5× enterprise multiple, with payroll at 30 percent of revenue, has equity worth half a year's revenue — so repurchasing 5 percent of shares a year costs 2.5 percent of revenue, or 8.3 percent of payroll.

Two routes, one order of magnitude. The obligation is roughly a tenth of payroll, every year, forever — and it grows with the share price. A plan that succeeds in raising its own valuation has thereby raised its own liability, which is a sentence worth reading twice.

Three things make the arrangement hold.

A reserve that is actually funded. Segregated, contributed to annually, sized against the age profile of the workforce rather than against last year's redemptions. An unfunded repurchase obligation is a defined-benefit promise wearing a defined-contribution costume.

A valuation nobody has a reason to distrust. Independent, consistent in method year to year, and explained out loud. When the price moves, the appraiser says why. The fastest way to kill employee ownership is a valuation that rises smoothly while the company struggles, because the first retiree to be paid out at that price is being paid by everyone who is staying.

A concentration cap that is enforced when it is inconvenient. It will be inconvenient exactly when the share price has been rising, which is exactly when it matters.

And here is where it fails. It fails when the stake was bought with wages, so the workforce has no diversified savings at all and the plan becomes the whole of retirement. It fails when the repurchase obligation arrives as a demographic wave — a founding cohort retiring within five years of each other — and the reserve was sized on a five-year average. It fails when the valuation becomes a morale instrument. And it fails, quietly and most often, when nobody ever tells a 55-year-old that the election exists, and she retires ten years later holding 35 cents on the dollar when she could have been holding 67.

DELIGHT

What it feels like

There is a specific, unglamorous pleasure that people in employee-owned firms describe, and it is not pride. It is the disappearance of a particular kind of suspicion.

You stop wondering who the saving was for. When a supervisor finds a better way to run a line and it takes forty minutes out of a shift, the question and who gets that has an answer that everyone in the room already knows, and so the question does not get asked, and so nobody spends the afternoon being careful. That absence is worth more than it sounds. A great deal of ordinary working life is spent quietly calculating whether a contribution will be captured, and an ownership stake, properly built, retires the calculation.

Then there is the statement itself. Once a year, a piece of paper with your name on it and a number that got larger because of something you can point to. Not a bonus, which is a verdict on you. An asset, which is a record of the place. The pleasure is closer to reading a tide table than to receiving a prize — it is the quiet satisfaction of a thing that has been accumulating whether or not you were watching it.

And on the day somebody retires and the cheque clears on time, the whole building learns something it cannot be told in a meeting: the promise was real. That is one afternoon a year, and it does more for the culture than every poster in the corridor combined.

OPERATIONALIZE THIS

At the level of finance

The instrument: an employer-funded stake with a concentration covenant and a funded repurchase reserve.

Three components, one document, and it is designed so that the treasurer, the auditor and the shop floor are each reading terms they recognise.

The structure. A leveraged ESOP in the United States or an EOT in the United Kingdom, acquiring a stated percentage of equity. The trust borrows; the company services the debt out of pre-tax profit; no employee contributes cash, defers wages, or accepts a reduction in any existing term of employment. That last clause is written into the plan document as a covenant, not stated in a presentation, because presentations do not survive a change of chief executive.

The concentration covenant. The plan commits to a ceiling on each participant's stake as a share of their total retirement assets — 15 percent unless the board minutes a specific and defended higher figure. Enforced by redirecting future contributions to the diversified plan, never by compelled sale. The companion clause is the one that costs nothing and is refused most often: the 401(k) alongside holds no employer stock, and the match is paid in cash.

The repurchase reserve. A segregated account, contributed to annually, sized by an actuarial projection of redemptions over fifteen years rather than by last year's outflow. Disclosed in the accounts. Reviewed when the valuation moves by more than ten percent in a year.

The balance-sheet treatment. The trust's borrowing is a liability of the sponsor at amortised cost and the interest runs through the income statement. The repurchase obligation is not a liability under most frameworks until the put is exercisable — which is exactly why it must be disclosed and funded voluntarily. An obligation that is certain, large, growing, and invisible to the balance sheet is the definition of the thing this edition exists to find. Talk to the auditors in the first month, not the eleventh.

The counterparties. An independent appraiser retained by the trustee and not by management. An institutional trustee, or a trustee board with a majority who do not report to the chief executive. A lender who has done this before — in Britain the EOT market is deep enough that this is now an ordinary credit; in the United States, an ESOP lender with a portfolio. In Italy, CFI. In France, the SCOP federation's own financing arm.

The first ninety days.

Day	Action	Artifact
1–15	Answer the funding question in one sentence	The funding memo — "the employee gives up nothing"
16–30	Independent valuation commissioned by the trustee, not by management	Engagement letter
31–45	Model the repurchase obligation over fifteen years on the real age profile	Redemption projection
46–60	Draft the concentration covenant and the cash-match clause	Plan document, marked
61–75	Auditor conversation on disclosure; lender term sheet	Accounting memo, term sheet
76–90	Diversification election pack drafted for every participant over 50	The envelope, with their own numbers in it

The number that decides it. One inequality, and it belongs on the front page of the board paper:

         participant's stake                     2 · lambda
       ------------------------------   <=   ------------------
       participant's total retirement           gamma · sigma_eps^2
                savings

At a plausible ownership premium of 1.5 points a year, the parameters in this chapter put the right-hand side at 11.4 percent. Choose your own premium, choose your own volatility, and defend them in the minutes. What is not defensible is running the plan without computing the number at all, which is what almost every plan in existence currently does.

If the inequality holds, the stake is a gift and it is worth having. If it does not, the difference is being paid for out of the employee's risk budget rather than out of the firm's performance — and the fix is never to cancel the ownership. It is to change how it was funded and to open the diversification door.

APPRECIATIVE QUESTIONS

Twelve, for a room

Discovery — what is already working

Think of a time when someone here made an improvement and did not have to ask who would capture it. What made that possible, and where else is it already true?
Which of our people have stayed far longer than the industry average — and when we ask them why, what do they actually say?
Where in this business does someone already behave like an owner without holding a share? What are they responding to, and could we give them more of it tomorrow?

Dream — what becomes possible

If every person here could open one envelope a year and understand exactly what they owned and what it was worth to them, what would change in how this place talks about money?
Imagine a retirement in this company ten years from now that goes perfectly. Describe the week. What had to be true, and when did it have to become true?
If our annual report carried a diversification take-up figure beside the safety statistics, what would we want it to say — and what would it take to say that honestly?

Design — what we build

What does an employee here give up to get a stake? If the answer is anything other than nothing, what would it take to make it nothing?
What is the largest share of someone's retirement savings we would be comfortable having in this company, and who in this room would have to sign that number?
Who should sit on the trustee board who does not report to anyone in this room, and what would we need to give them to make the seat genuinely theirs?

Destiny — how it holds

When the founding cohort retires within five years of each other, where does the cash come from? Who is the first person we would tell if the answer became unclear?
What would we want the valuation conversation to sound like in a year when the number falls — and who would we want explaining it?
What is the first sign we would see that this had quietly become a benefit nobody understands, and who on the floor would notice it before we did?

WORKS CITED

Blasi, J. R., Freeman, R. B. and Kruse, D. L. (2013). The Citizen's Share: Reducing Inequality in the 21st Century. Yale University Press.

Blasi, J. R. and Kruse, D. L. (2026). Employee ownership research brief, The Aspen Institute, Washington, DC, May 2026. (Analysis of 2023 Form 5500 filings; General Social Survey 2014/2018/2022; Census-linked productivity estimates. The exact published title of this brief was not confirmed at the time of writing and is given here descriptively rather than quoted.)

Board of Governors of the Federal Reserve System (2023). Survey of Consumer Finances, 2022. Washington, DC. (Median family net worth by age of head.)

Burdín, G. (2014). "Are Worker-Managed Firms More Likely to Fail Than Conventional Enterprises? Evidence from Uruguay." Industrial and Labor Relations Review, 67(1), 202–238.

Davis, S. J. and von Wachter, T. (2011). "Recessions and the Costs of Job Loss." Brookings Papers on Economic Activity, Fall 2011, 1–72.

Doucouliagos, C. (1995). "Worker Participation and Productivity in Labor-Managed and Participatory Capitalist Firms: A Meta-Analysis." Industrial and Labor Relations Review, 49(1), 58–77.

Employee Ownership Association and White Rose Employee Ownership Centre. The UK Employee Ownership Business Register and annual sector figures. Brough, UK.

HM Revenue & Customs. Taxation of Chargeable Gains Act 1992, ss. 236H–236U, as inserted by the Finance Act 2014 and amended following the Autumn Budget of 30 October 2024.

Internal Revenue Code, 26 U.S.C. §§ 401(a)(28)(B), 409(h) and 1042.

Jacobson, L. S., LaLonde, R. J. and Sullivan, D. G. (1993). "Earnings Losses of Displaced Workers." American Economic Review, 83(4), 685–709.

Kruse, D. L. (2016). "Does employee ownership improve performance?" IZA World of Labor, 311.

Kruse, D. L. (2002). Research Evidence on Prevalence and Effects of Employee Ownership. Testimony before the Subcommittee on Employer-Employee Relations, Committee on Education and the Workforce, US House of Representatives, 13 February 2002.

Kurtulus, F. A. and Kruse, D. L. (2017). How Did Employee Ownership Firms Weather the Last Two Recessions? Employee Ownership, Employment Stability, and Firm Survival: 1999–2011. W. E. Upjohn Institute for Employment Research.

Kurtulus, F. A. and Kruse, D. L. (2018). "An Empirical Analysis of the Relationship between Employee Ownership and Employment Stability in the US: 1999–2011." British Journal of Industrial Relations. (The peer-reviewed companion to the Upjohn monograph above; volume and pages not confirmed at the time of writing.)

Meulbroek, L. K. (2005). "Company Stock in Pension Plans: How Costly Is It?" Journal of Law and Economics, 48(2), 443–474.

National Center for Employee Ownership. ESOPs by the Numbers. Oakland, CA. (Counts derived from Form 5500 filings; data lag approximately two years.)

National Center for Employee Ownership (2023). ESOP Repurchase Obligation Survey. Oakland, CA. (248 valid responses.)

O'Boyle, E. H., Patel, P. C. and Gonzalez-Mulé, E. (2016). "Employee ownership and firm performance: a meta-analysis." Human Resource Management Journal, 26(4), 425–448.

Pension Benefit Guaranty Corporation (2005). Termination of the United Airlines pension plans. Washington, DC.

Pérotin, V. (2006). "Entry, exit, and the business cycle: Are cooperatives different?" Journal of Comparative Economics, 34(2), 295–316.

Pérotin, V. (2016). What do we really know about worker co-operatives? Co-operatives UK, Manchester.

Publix Super Markets, Inc. Employee Stock Ownership Plan (the PROFIT Plan) and Annual Report. Lakeland, FL.

United States Government Accountability Office (1987). Employee Stock Ownership Plans: Little Evidence of Effects on Corporate Performance. GAO/PEMD-88-1.

Vieta, M., Depedri, S. and Carrano, A. (2017). The Italian Road to Recuperating Enterprises and the Legge Marcora Framework: Italy's Worker Buyouts in Times of Crisis. Euricse Research Report, Trento.

Wiefek, N. (2017). Employee Ownership and Economic Well-Being. National Center for Employee Ownership, Oakland, CA. (NLSY97, respondents aged 28–34.)

Note on figures. Every figure above is computed in lib/verify/V_03.py and reproducible with python3 lib/verify.py V.03. Volatility, risk aversion, hazard rate and horizon are stated assumptions, printed as assumptions in that module and given with sensitivity tables so a reader can move them. Chapter I.11 holds the seller-side arithmetic of an employee ownership trust and is not duplicated here.