Abstract: In recent years, the standardization of the research method and the improvement of publishing requirements have made it more difficult for researchers to publish independently. More and more researchers choose to engage in cooperative research. Taking the three important periodicals of economics and management in China: Social Sciences in China, Economic Research Journal and Management World as examples, the proportion of cooperative papers was only 32% in 2000 and had risen to 90% by 2014, which shows that cooperative research has become the mainstream trend. However, there are costs in cooperative research, especially due to the long-standing first author evaluation rule in China, which makes the coordination cost of cooperative research increase sharply. The first author rule is mainly derived from the contribution-based signature mechanism. Compared with the first-name alphabetical order prevalent in international journals, signature by contribution highlights the contribution of the first author, which correspondingly weakens the cooperative benefits of non-first authors. In high-quality research, every collaborator needs to make a lot of efforts, and the importance of contribution is often difficult to distinguish. Signature by contribution is obviously not an optimal result. In the case that only the first author is recognized in scientific research evaluation rule in universities, the benefit from cooperation (reflected in title promotion and scientific research awards) of non-first authors is basically zero. The existence of this benefit distribution mechanism will greatly reduce the probability of cooperation among researchers. Some existing studies have qualitatively analyzed the impact of the first author rule on cooperative incentives, but so far, the quantitative research in this area is still scant.This paper collects the scientific research evaluation rule of "985" universities and universities of finance and economics by the questionnaire survey and matches it with the published data of three major economics and management journals. Based on the quasi-natural experiment of adjusting the first author rule in some universities from 2000 to 2014, this paper uses the difference-in-differences method (DID) to investigate the effect of scientific research evaluation rule on co-authored papers. The results show thatthe non-first author rule in the professional qualifications institution can significantly stimulate teachers to cooperate, increasing about 33.5%, while awarding institution has no effect. Considered the heterogeneity of cooperation, the non-first author rule mainly promotes inter-university cooperation and has no obvious impact on inner cooperation. However, the abolishing of the first author rule may also lead to the emergence of the academic "free rider" behavior like "fake cooperation". After solving this problem, we find that the non-first author rule still promotes cooperation. Based on these conclusions, we analyze the impact mechanism from the perspective of promotion pressure. The results show that the non-first author rule in the professional qualifications institution has cooperative incentives only for teachers who have not yet been promoted to professors, but has no significant impact on teachers who have become professors.The contribution of this paper is mainly embodied in three aspects. Firstly, this paper quantitatively evaluates the impact of the first author rule on the co-authored paper incentive for the first time. Previous literature mainly explores the reasons for cooperative research from the perspectives of author's personal characteristics and thesis's characteristics. A few literature on the impact of scientific research evaluation rule on the co-authored paper only analyzes qualitatively. This paper identifies the impact of the first author rule on cooperative incentives with the DID method by the quasi-natural experiments of adjusting the first author rule in some universities. Secondly, by the form of the questionnaire survey, we make statistics and descriptions on the scientific research evaluation rule of domestic universities in economics and management disciplines, which is of great significance to grasp the current situation of the construction of scientific research rule in universities. The results of the questionnaire survey show that, as time goes on, more and more universities begin to recognize the non-first-author paper, which is an important opportunity for researchers to strengthen academic cooperation and engage in important research with an international perspective. Finally, in the definition of paper cooperation, the previous literature only takes the number of paper’s authors into consideration, and we not only consider the number of paper’s authors but also the affiliation of each author. Based on it, cooperation can be divided into inner-school cooperation, inter-school cooperation, and inter-university cooperation.Finally, we put forward three policy suggestions from the perspective of bettering scientific research environment. As follows: (1) Establish a good academic reputation mechanism. The establishment of an academic reputation mechanism is the foundation. Without academic reputation mechanism, the abolishing of the first author rule may condone academic misconduct such as "free rider". On September 1 in 2016, the Ministry of Education has promulgated “the Measures for Preventing and Handling Academic Misconduct in the University” in the form of departmental regulations, but it only serves as a warning and precaution in advance. It is also necessary to establish an independent academic integrity investigation institution that publishes the list of irregular scholars and improves the self-discipline of scientific researchers. In addition, academic journals should give a warning or reject their submissions to researchers who violate academic ethics. (2) Reform the scientific research evaluation rule with papers as the core. In order to eliminate the scientific research evaluation rule with the number of papers and highly cited papers as the core, the peer review mechanism mainly carried out by the third-party evaluation institution should be gradually established. Besides that, on the establishment of an evaluation system, comprehensive and classification evaluations are equally important. Good classification evaluation mechanism means that teachers working in teaching or scientific research can be evaluated separately. At the same time, researchers working in basic or applied research can also be evaluated separately. (3) Reform the signature rule. On the basis of (1) and (2) points, the signature mechanism should be gradually changed to be sorted by initials of surnames. From the subject evaluation to academic ability evaluation, the first author rule should be broken. Only in the way, more and more researchers can be encouraged to working in the high-level of co-authored researches.

Key words: distorted institution, first author rule, co-authered paper