Investment Planning

I have been rereading Democratic Economic Planning and the chapters on investment planning are throwing me off a bit. I would appreciate any clarifications to my questions.

What does the NFCC providing estimates of consumer preferences look like? And what does the NFWW providing estimates of production functions look like? Is it just one number or is this for every capital good? It seems like it would have to be the latter so that you could sum them all together to get the aggregate investment plan: the fraction of production to devote to consumption in the coming year vs investment for each year. But then, if the NFCC and NFWW are getting these estimates for each capital good, can’t the investment planning board get the marginal social product (MSP) for each capital good? In the text it seems to say that you just get one aggregate MSP and then apply it to every capital good:

“We should keep producing any capital good, in any year, up to the level at which its marginal social product is equal to the marginal social product of investment for the “best possible” level of aggregate investment for that year.”

But why would you care about the aggregate MSP if you already have the more specific MSP for each capital good?

And then it sounds like this aggregate MSP is used to calculate (1 + d), which is used to increase the indicative price for capital goods that are demanded for future years. The text says:

“And under our assumptions, in order for worker councils to make the socially efficient choice of how many drill presses produced this year to demand, they must be induced to discount their best estimate of the marginal social product of a drill press to them next year when equating it to what they will be charged for it this year. And this must somehow be accomplished during the annual planning process that takes place this year.”

Is (1 + d) the same for all capital goods? How is the aggregate MSP used to calculate (1+d)? The text says the equation for MSP is δF(2)[s&i(1)′,l(2)]/δs&i(1)′ for example for year 2, and then (1+d) to be {dU(t)[c(t)′]/dc(t)′}/{dU(t+1)[c(t+1)′]/dc(t+1)′} (which doesn’t seem to have much in common with the definition of MSP I just gave above).

And if the aggregate MSP number is not used to calculate (1+d), what is it used for? Who uses it and for what?

Finally, is there a process where producers and consumers bid on producing and consuming capital goods that takes place before the annual participatory planning procedure? Or does the former just occur during the latter? It seems like there is a separate procedure that happens first, but then it is weird why workplaces would bid on demanding capital goods during annual participatory planning, where the amount of each capital good to be produced for the upcoming year are just sort givens, like natural resources. Didn’t workplaces just bid on user rights for capital goods to use during the upcoming year? So why would we make them do it again in annual participatory planning?

Any clarifications would be greatly appreciated.