Bands and Futures settle how people are paid. This chapter covers who is on the payroll in the first place. Two tests decide it: one for whether a piece of work belongs inside the company, and one for whether a new role is worth adding.
The in-house test
Chapter 5 asked what it costs when a job does not get done, and found that the roles paid least are often the ones a company can least do without. FairShare turns that into a rule. If the company needs a piece of work done regularly, the person doing it is on the payroll, on the ladder, earning Futures. No category of staff is “support” and paid by a different logic.
The question to ask of any work is the one from Chapter 5: what happens if this is not done? If the answer is that other people stop being able to work, the role is part of the company.
Cover for the unexpected
Some needs cannot be scheduled. A pipe bursts, a server fails, someone is hurt. Having someone on hand who can respond is worth a great deal, as the sick day in Chapter 5 showed, and on most days it asks nothing of that person except to be there. FairShare treats being available as work and pays for it. What it does not do is make availability a whole role. Every role on the ladder has a full schedule of its own work, and nobody is paid a band and a Future only to wait.
Keeping watch is different from waiting. A guard on a gate, a lifeguard or an operator monitoring a control room is working for the whole shift, because the attention is the work and cannot be set down to do something else. That is a role. It passes the in-house test like any other regular work and is placed in a band by the three questions in Chapter 8.
The test is whether the person can do other work while they cover. If they can, the cover is a duty. If the cover takes their whole attention, it is a role.
Cover comes from three places.
- Preventive work inside a real role. Most of what looks like cover is inspection, maintenance and monitoring. That is regular work, and it passes the in-house test on its own.
- A duty, held by people who already have full roles.
- A supplier on retainer, with a guaranteed response time.
A duty is a published addition to a role, such as building responder, first aider or out-of-hours cover for the company’s systems. Each duty has a name, a training requirement and a credit, and they are the same for anyone who holds it. Most of the time a duty consists of being reachable and ready and nothing more. The credit pays for exactly that.
- Training is paid, scheduled time, open to anyone whose role allows them to hold the duty.
- Time on duty earns a credit. Each hour on duty counts as a fraction of an extra hour worked, which earns salary and Futures in proportion. The example company counts a tenth of an hour during the working day and a quarter outside it. A callout outside working hours counts in full.
- The duty rotates among everyone qualified, so the credit is shared and more than one person knows what to do.
- Band is unaffected. A band measures mastery of a function, and a duty sits beside it.
Someone in the example company who holds a daytime duty one week in four is on duty for ten of the year’s forty working weeks. At a tenth of an hour for each of 400 hours, the credit is 40 hours, or five standard days: five extra days of salary and five extra Futures.
A duty makes a first responder, and a first responder is not a specialist. The person on duty can shut off the water, reset the breaker, make the area safe and call the right supplier. The repair may need a licensed trade, which is why a duty is paired with a supplier on retainer. The company keeps what has to be inside it, which is knowledge of the place and a fast first response. It buys the depth it rarely needs.
A duty does not work everywhere.
- Roles that cannot be interrupted. Someone on a live call with a customer, running a machine, teaching or driving cannot leave, and neither can anyone who is the only cover for their own function at that moment.
- Work the law reserves for licensed or certified people.
- Very small companies, where there are too few people for a rota.
- Duties that are called on too often. Every callout is logged. When a duty regularly takes real time from people’s main work, it has become regular work, and the in-house test makes it a role.
The hiring test
Under FairShare a new role costs everyone something they can see. It adds a salary, which reduces net income, and it adds Futures, which divide the pool among more people.
Take one extra Standard hire in the example company’s 2027, working the full year.
| Effect of the hire, if it adds no income | Figure |
|---|---|
| Salary | $100,000 |
| Reduction in the pool (70 percent of it) | $70,000 |
| Extra Futures | 200 |
| Value of one Future | $397.15, down from $400 |
| Cost to each full-year colleague | About $570 |
For everyone else to come out even, the hire has to add about $214,000 to the company’s income before their own salary: $100,000 to cover the salary and about $114,000 to cover the dilution. The sum ignores tax and benefits, which push the figure higher.
This gives every hiring decision a number that the whole company has a reason to care about. Three practices follow from it.
- Proposals are published. Whoever wants a new role states the band, the cost and what the role is expected to add.
- A committee decides. It is drawn from several functions and bands, so that no single manager is growing a team at everyone else’s expense.
- A departure is a question. When someone leaves, the role is tested again before it is refilled. The work may have changed, or it may now belong to a different function.
The test cuts both ways. It discourages hiring for status or comfort. It also shows when a team is short: if a missing Standard role is costing more than $214,000 a year in lost or delayed work, everyone is better off filling it.
The test has a cost, and economists described it long before FairShare. In 1958 Benjamin Ward showed that a firm whose members share its income has a reason to turn away a hire who would add to profit, if that hire would add less than the average member already does.1 The table above is that calculation. A conventional employer gains from the extra Standard role at anything over $100,000. The people in a FairShare company gain only above $214,000. Between those two figures lies work that would be worth doing and that nobody inside has a reason to vote for. A company run on FairShare should expect to grow more slowly than one run for its shareholders.
How much more slowly is not known. Ward’s model went further and predicted that such a firm would shrink when the price of its product rose, and studies of real firms have not found that. The plywood cooperatives of the Pacific Northwest, set beside conventional mills over nearly twenty years, adjusted what they paid where the conventional mills adjusted how many they employed.2 A study of every cooperative in Uruguay’s social security records from 1996 to 2005 found the same pattern and added to it. When prices rose, conventional firms took on more people and paid a little more. The cooperatives paid more and took on no one: their employment did not respond to price in either direction. In the crisis of 2002, employment fell by less in the cooperatives than in the firms beside them.3
So the firms that have been studied did not shrink in good times, as Ward feared. They did not grow in them either. If FairShare companies behave the same way, they would hold on to their people through a bad year better than their competitors, and add to them more slowly in a good one. That is the cost to expect, and the hiring test is where it comes from.
The committee and the published proposal are FairShare’s answer, and they are a partial one. They make sure that a role worth more than it costs everyone is argued for in the open, by people from outside the team that wants it.
When a role ends
The hiring test runs in both directions, and the other direction is harder to look at. If adding a role costs every colleague something, removing one pays them. Take away a Standard role that brings in nothing and each remaining person is about $570 a year better off, the same sum the hire would have cost. In a conventional company the gain from a layoff goes to the shareholders. Here it would go to the people who vote.
FairShare does not pretend that incentive away. It sets four rules against it.
- While the ladder is being paid in full, a role ends for one reason: the work has gone. The case is made the way a hire is made, in a published proposal, and the same committee decides it. Where several people hold the role, the order in which they leave follows a rule published before it was needed, with volunteers first.
- A shortage of money is met in a fixed order, and roles come last. The pool takes it first, since Futures pay nothing in a year without profit. The reserve takes it next. Then the council declares a lean period, which defers pay from the top band down (Chapter 12), and a function with too little work can shorten everyone’s schedule so that the shortage is shared. Roles end for want of money only after all of that.
- Whoever leaves keeps earning Futures for a published period, as though their schedule had been worked. The example company uses six months. For that time the people who stayed gain nothing from the pool by the departure. Everything already earned is paid on the usual dates, logged pay stays with the person who is owed it, and severance follows one published scale, counted in weeks of the person’s own band figure.
- A sale cannot reward it. The proceeds of a sale are divided by the Futures earned over five years (Chapter 14), so nobody’s share of a sale can be taken from them by ending their role first.
These rules slow the incentive and put it in plain sight. They do not remove it. After the published period, the people who remain are better off for every role that has gone, and a majority that wanted to shrink the company for its own gain could do it within the rules, one published proposal at a time.
What can be said is that the firms nearest to FairShare have behaved the other way. In the plywood cooperatives and in Uruguay, people who shared the income kept one another in work through bad years and took the loss in pay. Those firms were able to cut pay. The lean period is how a FairShare company would do the same without touching the floor. Whether it would is not known, and Chapter 16 names the result that would show it had not.
Contractors
The hiring test creates a temptation, and it would be dishonest to hide it. Work done by a contractor adds no Futures, so the workforce does better, on paper, by keeping people off the payroll. This is the route by which companies usually come to underpay the people who clean, guard and maintain their buildings. Studies of outsourcing find that cleaners and security guards earn less when their jobs are contracted out: 4 to 7 percent less for cleaners and 8 to 24 percent less for guards in the United States, and 10 to 15 percent less across outsourced service jobs in Germany.4 5
FairShare closes the route in two steps.
- Regular work comes in-house. A company should set its own threshold. FairShare suggests treating anything that amounts to more than a month of one person’s work in a year as regular.
- Contracted work has a pay floor. Where regular work does stay with a supplier, the contract requires that the people doing it are paid at least the band figure the same work would earn inside the company, in proportion to their hours.
The floor has to be the band of the work. If it were the lowest band, a company could still save money by sending out any work that ranks above Junior. Set at the band of the work, it removes the saving on wages. It does not remove the saving on Futures, because a contractor’s staff earn none. The floor is therefore a backstop. The in-house test is the rule, and it holds only if the council and the hiring committee apply it honestly.
Occasional and specialist work is a different matter. A lawyer consulted twice a year or a firm brought in to replace a roof is a supplier, paid whatever the two sides agree.
Notes
- Ward, B. (1958). The Firm in Illyria: Market Syndicalism. The American Economic Review, 48(4), 566–589.
- Craig, B., & Pencavel, J. (1992). The Behavior of Worker Cooperatives: The Plywood Companies of the Pacific Northwest. The American Economic Review, 82(5), 1083–1105.
- Burdín, G., & Dean, A. (2009). New evidence on wages and employment in worker cooperatives compared with capitalist firms. Journal of Comparative Economics, 37(4), 517–533. The findings described here were read in the working-paper version: Instituto de Economía, Universidad de la República, DT 02/09.
- Dube, A., & Kaplan, E. (2010). Does Outsourcing Reduce Wages in the Low-Wage Service Occupations? Evidence from Janitors and Guards. ILR Review, 63(2), 287–306.
- Goldschmidt, D., & Schmieder, J. F. (2017). The Rise of Domestic Outsourcing and the Evolution of the German Wage Structure. The Quarterly Journal of Economics, 132(3).