Sure, policy matters. It's important to have people in public leadership who understand it and can work on tuning it to avoid or eliminate unintended consequences.
But complaining about either regulation or bureaucracy at a general level is the policy equivalent of approaching software by making general complaints about how terrible existing code is: sure, some of it is probably sub-optimal and should be rewritten, but anyone who's worked in the field for a while knows that the sweep-it-away/ground-up-rewrite impulse is more often wrong than applying Chesterton's fence (make sure you understand why it's there), profiling, and examine potential alternatives against studied outcomes.
> Markets would love to build more housing.
Sure. And would love to chase the wealthiest end of demand and externalize costs of community displacement or ecological impact.
> Markets would love to build nuclear power plants
And would love to externalize the costs of risk and waste management.
Note that these are not arguments against building more housing or nuclear power. There are many metro areas that do need to build housing. Nuclear power has its merits as well as hazards.
They're arguments about specific incentives absent intentional oversight, supporting the general case that it is often irresponsible to believe or advocate that incentives tend towards optimal on their own, and that the primary problem is oversight itself.