1) No non-compete, long hours, and low salary
2) Non-compete agreement, short and flexible hours, and high salary
If this is a possible scenario, would the ban on non-competes be anti-worker?
My first piece of evidence for the proposition that employers pay for non-competes is that some of them prefer to hire employees with non-competes. Companies pay for everything they require of employees, as all unnecessary conditions reduce their options (for employees), and the only reason they ever offer money is to increase their options (from 0-1 or 100-200). In addition, it is very common for employers to value non-competes as assets (for accounting purposes). There are many other points in favor of the idea that companies do pay for non-competes, but we can start here.
I attribute the downvotes to the unpopularity of non-complete clauses on HN. In addition to that, most of the libertarian and right-leaning users seem to be in North America, so I can count on less sympathy from ~4-12 GMT. It is also possible that my writing has gotten substantially worse in the last few hours, but it seems unlikely that I am being more rude or unfair than I normally am.
The non compete is nearly entirely valuable in that it it depresses wages. In reality its entirely likely that between firm A and firm B neither has a whole lot of special sauce non competes mostly serve to keep each other from competing for employees thus driving up wages.
Why would you suggest that the employee pay a cost for accepting a fair workers agreement (no non-compete clause)? At least suggest that salaries be raised for people on non-competes.
The value the employee is generating through this IP is always going to be considerably (orders of magnitude) larger than whatever their salary is - the company can afford to pay them more.
In fact its trivial to show that non competes result in depressed wages. It in fact shows you the exact opposite of what you intend to show. Non competes are valued as assets as it enables employers to pay employees LESS than in a free market.
I agree that I am ignoring many things, as this is a limited discussion, and I am only addressing the impact of a law on a single employee, and its ethical ramifications.
Secondly: my second point answers my first.
Markets don't manifest in individual transactions, they manifest as the emergent behaviour of multiple transactions. As others have noted, setting bounds to what can be traded away has proven necessary, empirically, to avoid winding up in a highly non-optimal equilibrium point.
If you're interested in ethical ramifications, I recommend two excellent works on the topic: A Theory of Moral Sentiments and An Inquiry into the Nature and Causes of the Wealth of Nations.
Even that treatment is unsatisfactory as it neglects considerations of short-term vs. long-term trade-offs.
Your example is contrived in that it essentially poses a Sophie's Choice: give up condition of long-term viability A or because of my superior position in establishing and enforcing terms, give up condition of long-term viability B.
The optimum condition would be for no artificial constraint on A or B imposed by the employer.
Your earlier scenario made no sense as their would be no reason to suspect that the 2 employers respective pay and conditions were related you have just presented it as such. Imagine the law were changed tomorrow? Would the high paying non compete using employer suddenly start paying smaller wages? Would they have any employees tomorrow if they did? A freer labor market would be MORE competitive not less there is no reason to suspect that they wouldn't have to pay more not less to retain the same talent.
On the last point, see generally J. Doyne Farmer and John C. Holland who've done significant work on innovation and the processes by which it occurs. You'll find presentations from both on YouTube, as well as publications. Both are/were associated with the Santa Fe Institute (Holland died in 2015).