At the heart of Hayek’s idea of the market economy is the thinking that an economic order (“equilibrium,” to use the terminology of modern economics) is created as a result of the transmission of the benefits of pieces of tacit knowledge relating to the circumstances that exist in a particular time and place across the entire market through economic transactions. Although those pieces of tacit knowledge themselves cannot be universally disseminated, their benefits can spread through the entire market in the form of changes in price as an economic variable. In traditional economics, based on the idea that price is the only information that can be transmitted across the entire market, analysis has focused exclusively on the sort of economic order in which supply and demand volumes are equal (that is, a state of equilibrium). To be sure, it is true that price is the only information that can be quickly transmitted across the entire market.
However, on closer inspection, some pieces of tacit knowledge are transmitted from one part of the market to people in the vicinity through market transactions, leading to changes in their activity. Those changes may be ones that cannot be expressed in terms of quantifiable variables such as price. However, there is no doubt that the transmission of tacit knowledge is occurring and causing local changes in some particular parts of the market, and we may describe this as a process of the market “learning” in the same sense that AI learns.
As a result of the transmission of pieces of knowledge other than price information that are related to the circumstances that exist in a particular time and place in the vicinity through networks of people in the market and their contact with other pieces of knowledge, a reaction that is similar to a chemical reaction occurs. It is well known that networks of people and companies in the market are complex, with some particular structures of linkages (Masuda and Konno, 2010). According to Masuda and Konno, in a social relationship network of friends and acquaintances, for example, two people who are total strangers are six or fewer social connections away from each other in many cases. In other words, two randomly chosen people are likely connected with each other through a line of a few to a dozen acquaintances. This nature of a network of people is known as “small-worldness” under complex network theory.
As economic relationship networks of individuals and companies are also considered to have small-worldness, naturally, knowledge other than price information (particularly unquantifiable tacit knowledge) is assumed to spread very widely to individuals and companies through economic transactions. Local interactions caused by such contact involving scattered pieces of tacit knowledge are probably the driving force of innovation. In other words, innovation occurs through the following mechanism. Patterns that cannot be expressed in language are identified through economic transactions, and the knowledge of the identified patterns is used by individuals and companies to develop new business approaches in order to increase their own profits. The new approaches developed are innovations. If the approaches can be expressed in language or mathematical formulas, they become scientific knowledge that may be universally disseminated.
Conventional economics has developed theoretical frameworks with an exclusive focus on features such as price and quantity. Until now, the mechanism behind innovation—whereby new local features are identified through a certain learning process as a result of the transmission of tacit knowledge that cannot be expressed in language through a network of transactions with a small-world nature—has been outside the understanding of economics. Whereas economic analysis has so far focused on price information that can be shared throughout the entire market, a new breed of economics that analyzes local innovation activity that occurs in the circumstances that exist in a particular time and place may be conceptualized in the future.
The development of theories regarding the deep learning mechanism has only just begun, but if we unleash our imagination, it will be possible to imagine that a new breed of economics may be conceptualized within the paradigm of a general learning theory for machine or human learning, to which deep learning belongs. That sort of economics would probably be the new breed of economics that Hayek wanted to conceptualize.
Understanding the abovementioned nature of the innovation process makes it clear why freedom is essential to realizing innovation. Innovation is a process whereby people and companies (they may be equipped with AI-enhanced intelligence) develop a new approach to influencing the world in a more effective manner by acquiring local bits of tacit knowledge. A governmental central planning authority in principle would not have access to pieces of tacit knowledge that only exist in the circumstances that exist in a particular time and place. Only when bits of tacit knowledge are allowed to be freely utilized by the individuals who possess them can they trigger innovation by being transmitted to other people and companies in the vicinity. In a society and economy without freedom, it is extremely difficult to create innovation because tacit knowledge is not transmitted through a network of economic activity. That is one reason why individual freedom should be guaranteed if innovation is to be realized in an effective manner.
However, from a contrarian viewpoint, we could argue that if efficiently transmitting tacit knowledge is the only important point, individual freedom is not necessarily essential. In the future, as a result of the development of scientific theories regarding how to acquire tacit knowledge most efficiently, it could be discovered that a controlled economy that forcibly reshuffles transaction partners based on governmental rules that are set in advance is more likely to create innovation than a free market economy.
In other words, the possibility cannot be ruled out that a more effective creation of innovation alone does not serve as a standard for fully justifying individual freedom and that a totalitarian society without individual freedom could create innovation depending on the design of institutional frameworks.
However, there is another reason why individual freedom should be guaranteed, which is that all sorts of ideas created by humans or AI are fallible. Put simply, even if a theory is discovered which states that a totalitarian society could be the best vehicle for creating innovation, the possibility cannot be ruled out that the theory itself is wrong.
