In economics, industry is usually defined as a collection of enterprises that produce the same commodity (strictly speaking, it has strong substitutability). At this time, industry and "market" are synonymous, because only enterprises that produce similar products for the same market can constitute the same industry, and all enterprises in the same industry compete with each other in the same market to pursue the goal of maximizing their profits (Ma Guangqi, 2000). Of course, the market here refers to a narrow local market, not a broad general market. Therefore, at first, industrial economics only paid attention to specific industries, industrial organization, economies of scale and competitive monopoly among enterprises in industries. Later, it was gradually extended by economic theorists to industrial structure, industrial correlation, industrial policy and so on, and formed a corresponding theoretical framework. Industrial structure theory studies the relationship between industries and its evolution law; (1) industrial correlation theory describes the correlation between industries from the perspective of technology and economy; Industrial policy theory studies industrial structure policy, organization policy, technology policy, layout policy and so on. From the perspective of research scope, in Europe and America, industrial economics is essentially industrial economics in a narrow sense, that is, the research on industrial organizations is based on a mature and perfect market economy, and little attention is paid to the influence of institutional factors on industrial organizations. Its research content mainly includes three interrelated aspects: SCP paradigm under imperfect competition, the goal, scale and organization of enterprises, and public policies related to industrial organizations such as regulation and antitrust. In Japan, China and other Asian countries, industrial structure, industrial correlation and industrial policy have also been brought into industrial economics, which has expanded the research scope of industrial economics and is customarily called generalized industrial economics.

From the development track, the construction and development of industrial economics have generally gone through three stages, namely, the germination stage, the formation stage and the deepening stage.

(1) Embryonic stage (before 1930)

As far as the relationship between theory and practice is concerned, the research of industrial economic theory is rooted in the development of social division of labor and the continuous emergence of new industries. With the development of social division of labor and socialized commodity production, the relationship between industry and enterprises in industry is becoming increasingly complex, which objectively requires theoretical circles to study its operation law. In fact, the necessity of this kind of research has been strengthened with the in-depth evolution of social division of labor, and the theoretical circle has already begun the research process to meet this objective need. In this way, the germination of industrial economics has a long history, which can be traced back to the classical economist Adam Smith (A., 1776). In "The Wealth of Nations", he put forward two principles that are respected by later generations: one is the theory of market competition mechanism, and the other is the theory of division of labor. These two principles deal with the market behavior of manufacturers in a free competitive market system and have a direct impact on the study of industrial organization theory in industrial economics. Later, British neoclassical economist Marshall (1890) absorbed his theoretical nutrition, put forward the concept of "industrial organization" and introduced it into economics. When he discusses the factors of production in the book Principles of Economics, he enriches Say's theory of three factors of labor, capital and land production, puts forward the fourth factor, namely organization, and points out that the core issues of resource allocation efficiency are the internal economies of scale and external competitiveness of enterprises. He believes that the pursuit of economies of scale will inevitably lead to the development of monopoly, and monopoly will hinder and destroy the role of the price mechanism, so that prices are controlled by human factors, thus stifling the power of economic development-free competition, making economic operation lose vitality and destroying the rational allocation of social resources. Therefore, the internal contradiction between economies of scale and competitive vitality is called "Marshall conflict" by later generations.

In the Marshall era, the market economy was still dominated by free competition, and monopoly was not common. But at the beginning of the 20th century, monopolistic economic organizations gradually formed. In 1930s, Chamberlain and Robinson put forward the theory of imperfect competition, analyzed monopolistic competition and oligopoly market, and played an important role in promoting the development of industrial organization theory. Nevertheless, industrial economics at this time is still in the primary stage, and the real industrial economics has not yet formed.

At the same time, the theory of industrial structure began to sprout. As early as the17th century, william petty, a British classical political economist, discussed the relative income gap among the three industries in his Political Arithmetic (that is, agriculture is smaller than manufacturing, and manufacturing is smaller than commerce), and pointed out that this gap is an important reason for the flow of labor between industries. This laid a solid foundation for the later study of inter-industry relations and intra-industry structure.

(2) Formation stage (1930- 1970)

In 1930s, the United States was in the stage of industrialized mass production, and the economy of mass production and the substitution of enterprise organizations had a great influence on market competition. An industrial organization research group represented by Mei Sen and Bain was established in Harvard University. They inherited the research results of predecessors, summarized the practice of industrial organization policy in the United States for decades, and used empirical analysis, case analysis and econometrics methods to analyze the price and cost under the conditions of mass production, the influence of large enterprises on market prices and the conditions in contestable market, thus forming a complete theoretical system of industrial organization. Its landmark achievement is the publication of Bain's book Industrial Organization (1959), in which Bain defined the meaning of industrial organization and summarized it into various types of market relations among different enterprises in the industry, namely, the trading relationship, resource possession relationship, interest relationship and behavior relationship between them; This paper puts forward three basic categories: market structure, market behavior and market performance, analyzes their relations, and focuses on the market structure, forming a "SCP" analysis paradigm. This is what later generations call the mainstream industrial organization theory. Since then, the publication of Scheler's book Industrial Market Structure and Market Performance (1970) has put forward a more complete SCP paradigm and made the theory of industrial organization more mature.

Harvard School believes that market structure mainly includes concentration, product difference, economies of scale, entry barriers and government regulation, while enterprise behavior mainly involves collusion and strategic behavior, advertising and research and development, and market performance includes resource allocation efficiency, profit rate and productivity. They believe that there is a causal relationship between the three, market structure determines the market behavior of enterprises, and market behavior determines the performance of resource allocation. Therefore, in order to obtain good market performance, we must adopt active anti-monopoly policies and government regulation to improve the market structure and standardize the market behavior of enterprises. Because Harvard School attaches great importance to the decisive role of market structure in market behavior and market performance, it is also called Structuralism School.

While the theory of industrial organization was formed, the theory of industrial structure also developed rapidly. In 1930s, Australian economist A.G.B. Fisher established three industrial classifications for the first time, and named the industry that directly obtained material information from nature as the primary industry, that is, agriculture in a broad sense. The industry that processes and reprocesses products taken from nature is called the secondary industry, that is, manufacturing in a broad sense (including mining); All other economic activities except primary industry and secondary industry are collectively referred to as tertiary industry. Since then, the British economist C.G. Clark, on the basis of Petty and Fisher's research, has further revealed the regularity of labor distribution and transfer among three industries by using the economic statistical data of many countries, which is later called "Petty First Clark Theorem". At the same time, German economist W.G.Hoffmann also studied the relationship among industries within the industry, further subdivided the industries into consumer information industry, capital information industry and other industries, and demonstrated that the ratio of the net output value of consumer information industry to the net output value of capital information industry decreased with the process of industrialization. Through econometric analysis, it is concluded that the proportion of consumer information industry is gradually decreasing, while the proportion of capital information industry is gradually increasing, which is the "Hoffman Theorem". By the middle of the 20th century, American economist S.S. Kuznets used data from more than 20 countries to deeply investigate the evolution trend of national income distribution structure among three industries. Leontief also founded the input-output method, analyzed the relationship between industries, and made important contributions to the development of industrial correlation theory.

(3) Deepening stage (1970 till now)

Since 1970, industrial economics has entered a new stage of deepening development. On the one hand, the mainstream industrial organization theory founded by Harvard School has been continuously developed and improved, on the other hand, it has been constantly criticized and challenged. Therefore, the development of industrial organization theory is divided into two main lines:

One is the "new industrial organization", which represents the mainstream school and continues to advance along the SCP paradigm. Its representative figures are Colin, Watson, baumol and others. When they study industrial organization, they show the following characteristics: in the research direction, they no longer emphasize market structure, but highlight market behavior. In terms of research methods, a series of theoretical models are mainly established by using mathematical methods and game theory to study enterprise behavior; The analysis of economic welfare issues is more in-depth and has many unique features. Compared with Scheler's SCP paradigm, the new industrial organization theory pays special attention to enterprise behavior, regarding the initial conditions of market and enterprise behavior as exogenous variables, while regarding the market structure as endogenous variables, and there is no feedback line between them. All these can be regarded as amendments and supplements to the traditional SCP paradigm.

The other route is the rise of other non-mainstream industrial organization theories headed by Chicago School. Representative figures are stigler, Demsetz, broz and others. Among them, stigler contributed the most. In his book "Industrial Organization (1968)", he put forward the following main points: (1) discussed the motives and roots of enterprise merger, and thought that the motives of enterprise merger were not to achieve economies of scale, as long as there was no significant scale diseconomy, mergers would still occur. (2) Put forward the theory of industrial life cycle, discuss the relationship between internal division of labor and social division of labor, and explain the phenomenon of professional division of labor and vertical integration within the industry. (3) The optimal scale of manufacturers is discussed. It is believed that the optimal scale depends on many factors that are difficult to observe and measure, and the idea of determining the optimal manufacturer scale by the principle of "survivor technology" is creatively put forward: any scale that survives long-term competition is the optimal scale. (4) The problem of "entry barriers" is deeply studied. In his view, traditional barriers to entry such as economies of scale, capital requirements and product differences no longer constitute barriers, and government supervision is one of the main barriers. (5) The theory of information economics was founded and used to explain the market organization. It is believed that information can also produce benefits, and it costs money to obtain information. The information of manufacturers is incomplete, and the decision-making environment of people in the market is characterized by limited information. Therefore, complete information is not necessary to achieve competition, as long as there are many manufacturers in the market, the best degree of competition can be achieved. (6) Take the lead in systematically studying government regulation and create a precedent of "economies of scale". After that, Demsetz (1973) empirically studied the relationship between profitability and market economy in his book "Research on Market Concentration". To sum up, the theoretical characteristics of Chicago School are: in terms of research methods, it pays more attention to the analysis with strict economic theories and emphasizes the empirical proof of various theories (Taylor,1988); Oppose to emphasize the one-way causality of "structure-behavior-performance", but emphasize the two-way interaction between them; He holds a laissez-faire view of market behavior and opposes government intervention in industrial organizations.

When the research on industrial organization theory is in full swing in Europe and America, on the other side of the Pacific Ocean, Japan attaches great importance to the role of government in industrial development, and the resulting industrial policy is the concentrated expression of government intervention in industrial organization and industrial structure. 1945- 1955 is the period of Japan's post-war economic recovery. Through the implementation of industrial revival policy and industrial rationalization policy, the industrial reconstruction and economic recovery of many industrial sectors such as steel, coal, shipping, electricity and synthetic fiber have been successfully realized. After that, through the role of industrial policies, Japan achieved rapid economic growth in the 1960s and 1970s and became the second largest economic power in the world. With the emergence of Japan's economic miracle, industrial policy has increasingly attracted extensive attention from theoretical circles in various countries. 1970- 1972, the United Nations Organization for Economic Cooperation (OECD) published 14 member countries' industrial policy research reports, which promoted the global popularization of "industrial policy". Japanese scholars Ryutaro Komomiya and Kazuhiro Miyazawa of Shinohara systematically summarized the theory of industrial policy and industrial structure, and included it in the first book, Industrial Economics, which marked the formal birth of industrial economics.