Two points I found interesting:

The labour movement has long understood that raising output per hour of work through more investment in machinery and equipment (as opposed to work intensification) lays the basis for negotiating rising wages and living standards and investing in social programs and public services. Higher productivity can also support the reduction of working time.

In the US, governments have more often let the economy “run hot.” Low unemployment results in higher wages which pressure companies to invest in new capital equipment and skills, raising productivity.

I don't know if there's data but it could explain why the correlation between wage growth and productivity decoupled over the last 40-50 years. The standard assumed causation is productivity -> wage growth, but that's shown to be false by the decoupling. If however we assume wage growth -> productivity causation then at least that doesn't violate the decoupling data.