Success and viability within a system that is mutable
> There's a large economics literature on this: worker-owned cooperatives have not taken over the market, although they are an available institutional form, because (a) they find it hard to raise capital
There are any number of ways of making it easier for co-ops to raise captital.
> (b) they tend to make decisions that maximize worker welfare rather than profit, e.g. they won't sack underperforming divisions or expand in ways that dilute existing workers' stake.
You could make a similar case that making decisions solely around stock price leads to underperformance as well, if we quantify "performance" in terms of value provided to the broader economy. Mass layoffs are great for stock price, but they deal serious damage to your institutional knowledge base. Rather than simply shuttering an underperforming division, if you buy out the ownership stakes of the people working there, they'll have cash to re-train with. If we find that co-ops have a tendency to avoid doing that when necessary, we can tweak the incentives and subsidize a portion of the buyout, under certain circumstaces. Subsidies to tweak incentives are nothing new.
> This isn't something that anyone can decide upon
Yes, it is. Policy can change which strategies predominate in the market, by shifting incentives, by controlling streams of investment, and simply by regulating undesirable strategies out of existence. Mixed economies can be very effective. The notion that the market is a force which out to go untampered with is ideological, and it doesn't serve us well.