My favorite argument in favor a mostly-free market, and one that I think gets short shrift in popular discussion, is that the market is a phenomenal aggregator of information. It behaves like a sort of distribute processor, drawing in information possessed by each market participant from a series of one-off transactions. The effect is obvious in the case of prediction markets and other explicit wisdom-of-crowds situations, but its role in the real world is an important part of why market economies have been much more successful than centrally planned economies. The amount of information being collected, processed, and distributed is too much for a central controller to parse. Instead, the market distributes it, and information propagates across the economy to change prices, redirect the flow of goods, drive innovation, etc.
There is much to criticize about market economies in the real world. And I use the term “mostly-free market” to imply that certain aspects of the market should be centrally controlled, that certain central controls can improve the information processing function of the market. But these controls are best when they take a certain form. And it seems that focusing on the information-related consequences of any central intervention in the economy is a good way to evaluate such interventions. I have the intuition, which I’d like to flesh out in this thread, that maximizing the information processing power of the market will maximize its beneficial externalities for society. And if that’s the case, it’s important because it seems easier (though that is not to say always easy) to evaluate whether a policy will restrict or distort information flows, that to evaluate whether it will ‘improve the economy’.
In short, my claim here is that 1) a market economy is descriptively a good way of processing certain types of information about society, and 2) normatively, we should shape policy to foster that characteristic of the market.