https://en.wikipedia.org/wiki/Exchange_value mentions 4 notions of value: economic value, exchange value, use value, price. But this article also mentions that these are Marxian concepts and the contemporary economy do not make distinction among those values and the notion of the market price is sufficient to explain the value of the good of service.
Actually I would like to follow the path of contemporary orthodox economic theory, not heterodox theory. But I feel that there can be cases, where goods can have value that is different from the market price and it would be worth to capture and determine those other values:
- value of energy unit produced by different technologies (oil or green). One can argue that the same unit of energy that is sold by the same market price have different societal value because of some environment harm
- culture goods: they are not usually appropriately priced by the society, e.g. many countries have subsidies for the concerts of contemporary classical music or for the culture periodicals
- scientific discovery service: scientific discovery usually is not priced accordingly - wages for the scientists are small but the impact of their discoveries can be enormous.
So - some valuation always happen in government subsidies etc. Are there valuation theories (orthodox, conventional, non-Marxian) that try to capture this valuation process (e.g. for green economy, for cultural goods, for science investments) in some common terms? Are there game theories that try to determine the multiple types of value for some item?
My question is very important - computer science and artificial intelligence consider reinforcement learning as the most promising approach for the emergence of intelligence. Such learning process requires reward function. One can argue that the simple market price is the best reward function. But - if governments have some subsidy process, some support process for some parts of economy then the value theory and valuation games are very important for the determination of the reward function that is used for the training artificial general intelligence?