Engels remarked (citation and precise quote needed) that calculation of labor times required for each product would be straightforward and when this became so society would express these costs directly in units of time rather than by the roundabout route of prices.
At the same place he remarked that was analogous to the fact that when chemistry had advanced to the point of actually knowing the masses of atoms the normal standard measures of mass would be used instead of relative atomic weights.
In my view he was wrong on both points.
The second is quite clear. The Dalton is now known to bekg but not much work is done in terms of 1.66 yoctograms.
Even the atomic mass unit, the mole and Avogardro’s number are treated differently for different purposes.
There is even more than one system based on fundamental physical constants.
We do still have various different standard system of units measure for good reasons.
This is partly related to the transformation of quantity into quality.
Many important non-standard units are necessarily accepted for use with the SI.
For toy theoretical models here we will use the international standard currency sign for an unspecified currency, as the symbol for an unspecified unit of value (which could be thought of as somewhere between a yocto second and a yotto second of total human global value added:
For reasons discussed elsewhere I do view the relevant “labour time” concept as a fraction of total global human value added per SI second rather than individual hours or years of labour time with adjustments for skill etc. This fits in with the necessity of dual variables in constrained resource allocation over time.
Likewise intertemporal price conversion ratios and growth rates etc will be expressed as pure exponentials with respect to 1 SI second.
The metre-tonne-second system of “industrial” units will be used as more convenient SI variant than the usual metre-kilogram-second. It uses only single letter units – m, t and s.