The next obvious question - regardless of the level at which we start workweek reduction, what should determine the best level of the workweek and how fast should that level change - so we don't get any more solution than we need anywhere in the process?

But don't market forces take care of this? Can't, shouldn't, we leave this to "the dictatorship of the Market"?

No, because the market is a game and this is part of the definition of the game, part of the ground rules. The market can't redefine itself. The workweek defines the power gradient, the slope of the market's playing field, and the more level the slope, the more dynamic and fast-growing the game, because the better the feedback loop and the faster the mid-course corrections away from terminating imbalances. The workweek defines the size/intensity and number of basic game pieces (working hours) for each player. But then game theory breaks down, because most games are designed to end, while the market is an unusual kind of game that we want to continue and spread. So we have to confront the termination mechanism of most games as it rears its toxic head in the market = the tendency of mid-course winning to facilitate further winning which snowballs into total domination and finish, the tendency of the powerful to use their power to accrue more power and develop a stiflingly steep power gradient (such as between employer and employee), the tendency that "the rich get richer." How do we keep the playing field level? How do we repeatedly reset the level en route? How do we design effective power centrifuges to keep the game going, growing, and interesting? How do we replace the default concentration-termination mechanism with one or a series of specially designed sharing-spreading-continuing mechanisms? This is ideally a once&forall design challenge, not an invitation for repeated ad-hoc regulations that themselves then multiply into a stifling mass, a killing cancer.

That being said, there is a way to make the length of the workweek non-arbitrary - an almost market call. Here's how.

The whole point here is to define a dynamic range of work per person (i.e., workweek) that is automatically appropriate to our level of technology, so that as technology increases or decreases, our work level decreases or increases - an inverse variation.

But how do we gauge our level of technology in an employment-relevant fashion? The easiest answer is to broaden our goal. We want work for everyone on a level appropriate to our technology because we don't want technology to continue splitting us into overworked and unemployed - "unemployed" as in non-self-supporting. So our real goal is the ecological goal of sustainability. And our real enemy is non self-support, alias, individual non-sustainability. If we really have an adequate economic design, we will achieve societal sustainability by having virtually every individual self-supporting, that is, by zeroing non-frictional unemployment.

So non self-support is the ideal determinant of the reinvestment threshold. And non self-support is unemployment, but unemployment in a comprehensive sense that includes welfare, disability, homelessness, crime, forced part time, forced self employment.... This comprehensive unemployment is exactly the variable that we defined by referendum in Phase 1 of the Timesizing program in order to wake us up and motivate the other 4 steps in the program.

So what we're coming up with here - comprehensive unemployment - is the thing that we're going to use to determine our reinvestment threshold and through it, to size our reinvestment of overtime corporate profits and individual earnings. Thus the problem will determine the solution at each point, homeostatically, and we'll never get more solution than we need. Comprehensive unemployment will determine the level of the workweek and through it, the amount of overtime and overwork and through that, the amount of corporate and individual (or by default only, government) reinvestment in training and hiring.

This means that initially, the workweek will gradually come down to squeeze the market-demanded employment out onto more and more people until the current referendum-set target unemployment level is achieved. But from that point until the regularly scheduled referendums change the target level, the workweek will be reverse-indexed to the unemployment rate.  If unemployment goes up, the workweek will go down, and vice versa.
In the preceding private-sector program, it is corporate revenues rather than unemployment that determine the (corporate) weekweek. Instead of firing and hiring all the time and playing havoc with its skillset and morale, a company merely accordions its workweek and institutes pervasive cross-training to facilitate job reassignment. Lincoln Electric's three principles are brought into play (all sacrifice together, complete cooperation, and acceptance of job reassignment).

Whether we initially set our reinvestment threshold and start our overtime conversion at 40 or at 80, as long as we have too much comprehensive unemployment in our participating constituency, as defined by referendum, the threshold will gradually lower to create more training and hiring from reinvested overtime (and overwork). And the rate of lowering too will be determined by referendum.

Quick Reference. The 5 phases of the public-sector stage of the Timesizing program (bear in mind there's a long private-sector stage preceding that) are:
1. Referendums, to broadly define unemployment and set target rates
2. Corporate overtime tax with an exemption for OJT and hiring
3. Individual workoholic tax with an exemption for mentoring and employing
4. Making the workweek vary inversely with unemployment, newly defined to include welfare, disability, homelessness, prisons, forced part time and self employment...
5. If the workweek gets too low too fast, shifting the pressure to imports, immigrants, or births

6.=new 1. If the public doesn't want to squeeze imports immigrants or births, we move on to the next program, "Paysizing," and go through the same private and public sector stages of 5 phases apiece with "income and poverty" instead of "employment and joblessness".

For more details, see our campaign piece alias social-software manual, Timesizing, Not Downsizing, which is available from *Amazon.com online.