Note: The authors of this article are Jia Ruixue from the University of California, San Diego, Nie Huihua from the School of Economics of the Chinese People's University, and Xiao Wei from Southwestern University of Finance and Economics. These additional papers are the result of collaborations with others at the University. In other words, there are more colleagues who want to work with the dean, which leads to an increase in the number of papers published by the dean.
The study also found that most of these new dissertation topics were different from the research areas that preceded the dean's tenure. After becoming the dean, the dean has collaborated not in the field that he is good at or has been engaged in for a long time, but more in the research direction led by the collaborators. Because of the "interlacing of mountains", it is often difficult to change research fields; In addition, the dean has a lot of administrative affairs, which means that the dean is often "named" when co-publishing papers.
Notably, the study found that for top universities and top journals, the impact of power on paper publication was minimal. For scholars who have published a large number of papers before taking office, the power effect is also relatively small. Therefore, this means that the increase in the number of papers published by the dean is mainly related to the power he has over the allocation of resources, and it is difficult to say that the dean's ability or prestige is due to it. Therefore, this shows that there is a phenomenon of power distortion that deserves attention in the process of knowledge innovation. In other words, university competition and academic reputation mechanisms can inhibit the willfulness of power.
INTRODUCTION
The allocation of academic resources is an important issue.
Generally speaking, we believe that academic resources should be allocated more to academics with high academic potential and ability, but this is not the case in a society where power is paramount. Academics who have power in their hands but lack academic standards tend to have more resources. To date, however, there is still a lack of relevant research. In this article, the author examines the connection between power and the publication of scholarly outputs in Chinese academia, and attempts to discuss how administrative power affects the volume and subject of scholarly outputs. This study provides empirical evidence for rent-seeking in knowledge production.
Specifically, we can study the dean of a university, whose authority derives from administrative hierarchy rather than academic ability.
In order to provide systematic empirical evidence, this paper focuses on the field of economics and constructs a dataset of the personal information and academic publications of the deans of the School of Economics of each university. In terms of academic publications, this article only focuses on Chinese journals (because of the short review time). By studying the academic publications of these deans before and after their appointment, it is possible to quantitatively assess the influence of administrative power on their academic publications.
One of the main empirical issues in the analysis is that there is a serious endogeneity in the appointment of a dean by a higher authority, who may have considered the appointment based on the person's previous academic publications. To solve this problem, this paper uses an event research method that eliminates the effect of features that do not change over time. Specifically, the analysis includes both fixed annual effects and factors such as individual abilities that do not change over time or change slowly.
The main results of this paper have been described in the abstract, and three possible explanations for this result are presented in this paper. The first is the "competency effect", which means that academics who are on the academic upswing are more likely to become deans. Since they are on the rise, there will be more scholarly publications later. The second is the "reputation effect", in which the reputation of the deans may increase the likelihood that their articles will be published. Other scholars would argue that including deans in the authors increases the chances of publication. The last explanation is the "resource effect", i.e., power affects the allocation of resources. Fellow scholars will feel pressured and actively seek the Dean as a co-author, as this is closely related to resource allocation.
If the increase in scholarly publication is largely determined by the ability and reputation of the dean, then the findings in this paper do not necessarily imply a distorted resource allocation mechanism, as the resources and efforts of scholars at the university are not necessarily affected. However, various empirical evidence suggests that the first two channels cannot explain the results of this paper. On the contrary, the third explanation is most consistent with our results. This means that the resources available to scholars depend on their relationship with the dean. Moreover, since scholars are faced with two choices for allocating their time, namely, to conduct research or to establish a good relationship with the dean, the findings in this paper also hint at the distortions of this academic system.
Broadly speaking, our research contributes to research on the mismatch of human and physical capital resources. Some previous studies have shown that there is an improper allocation of resources between different industrial sectors. In efficient capital markets, capital should be allocated to more productive producers. Similarly, in efficiency-oriented academia, more resources should be allocated to academically more capable and efficient academics. However, in a power-oriented society, it is not the highly productive researchers who have access to resources, but those who have close ties to the higher-ups. We also found that for deans with less academic ability, access to power can increase their academic publications, suggesting that the resource effect is more important for them.
The administrative system of Chinese universities is part of the state bureaucracy, so the administrative power of the dean may have a special influence. However, the findings of this paper can also be generalized to less extreme cases, i.e., where universities have a low degree of autonomy and are heavily influenced by administrative power. For example, in surveys of universities in Japan and Hong Kong, it was found that hierarchical power relations often overrode intellectual contributions in the attribution of results. Even in Europe and the United States, the influence of executive power on research results objectively exists. Thus, although this research focuses on a specific area in China, its findings reflect a common problem in academia.
Background and data
(1) Background:
The sample of 95 universities in China that have been selected for the "211 Project" and have schools of economics in this paper is included. Of these 95 universities, 29 also belong to the "985 Project" universities. This article focuses on the heads of the economics schools of these 95 schools, i.e., the deans/executive deans, etc., collectively referred to as the deans.
In Chinese universities, the dean serves as the administrative leader of the university, responsible for teaching, academic research, communication with other institutions, and many other administrative functions. They influence the resources of other scholars through at least three channels: (1) selecting and deciding on the promotion of scholars; (2) whether they are eligible for government subsidies; and (3) eligibility for awards and funds provided by the government.
Specifically, within the college, the deans have the final say, and they decide on the salaries and benefits of other scholars, among other things. In addition to formal wages and benefits, subsidies from central and local governments have become increasingly important in recent decades through various schemes. Since these programs require a recommendation from the college, the Dean's support is critical to obtaining support for these programs.
For resources outside of the university, such as national science and technology awards and funding projects, the dean's recommendation is also important and often a prerequisite. For example, major research funding institutions such as the National Natural Science Foundation of China (NSFC) and the National Social Science Foundation of China (NSSFC) do not accept individual applications. The college collects and evaluates all applications and then decides which ones are suitable for submission. As a result, the influence of the dean is critical to most of the resources available to researchers at the school.
In addition, in public universities, the administrative system is also part of the state bureaucracy, so when their term of office expires, like government officials, they are usually promoted or transferred to other positions with more or less equal powers. As a result, their power influence often persists when they leave office. Our dataset shows that only a small number of deans do not hold executive positions after leaving office, but this also provides a useful subsample for this paper to assess the impact of losing executive power.
(2) Data:
This paper first collects the names of the incumbent deans of the faculties and departments in 2009, and then collects their academic publication records between 1990 and 2009. The year 1990 was chosen as the starting year for this paper because since 1990, the publication of journal papers has become an important criterion for evaluating researchers. The dataset constructed in this paper includes 99 deans from 95 universities (four universities have two different schools of economics).
Our data sources include: (1) resumes from the official websites of each college; (2) Baidu Encyclopedia; (3) China National Knowledge Infrastructure (CNKI); (4) The websites of the National Natural Science Foundation of China (NSFC), the National Social Science Foundation of China (NSSFC), and the Humanities and Social Sciences Foundation of the Ministry of Education of China (HSSFC). The journal we are concerned about is an economic/financial journal (it should be a C journal) that is widely followed by Chinese universities. Among these journals, Economic Research Journal, Chinese Social Sciences, World Economy, and Economics (Quarterly) are the four top journals (Management World Dissatisfaction). In addition, we use the author's affiliation to define local collaboration (collaboration with scholars from the same university) and non-local collaboration (collaboration with scholars from other universities). Collaboration here is defined in terms of author attribution. The NSFC, NSSFC and HSSFC provided information on the allocation of funds. In addition, we collected the characteristics of each university/college, including the number of faculty members and research funding. Since it is difficult to obtain data for different years, we focused on their cross-sectional changes in 2000 and allowed their effects to vary over time.
The main dependent variable of this paper is the number of academic papers published, and since very few of the sample of this paper have published articles in English (naked mockery), this paper focuses on Chinese articles. In addition, unlike English-language journals, Chinese journals typically have a publication cycle of less than one year, allowing this article to associate the year of publication with the time of becoming a dean. According to the calculations of the existing literature, the cycle of publishing a paper in China Industrial Economics is about 64 days.
In order to understand the channels by which administrative power influences scholarly publishing, this paper calculates the number of papers published by independent authors and co-authors, and distinguishes between local collaborations (collaboration with colleagues or students) and non-local collaborations (collaborations with scholars from other universities). In addition, this paper further examines the research topics of academic papers to see if the research topics of academic papers published by scholars have changed after becoming deans. Specifically, we define "topic similarity" in two steps. First, identify the 5 phrases that appear most frequently based on the title of an article published five years ago when he became the dean, and use them to define a scholar's field of study. Next, the frequency of occurrence of these 5 phrases in the title of the paper was calculated year by year by the phrase in the title.
Finally, this paper collects data on government-backed research funding, including NSFC, NSSFC, etc., which provides evidence for the impact of power on resource allocation.
The definitions of the key variables are summarized as follows:
Dean is a dummy variable assigned a value of 1 after being appointed dean, which is the primary explanatory variable;
Age is calculated based on the year of birth in the individual's resume;
The number of publications是各人每年在学术期刊上发表的文章数量；
The number of publications of the school是各院长单位所有研究人员每年的论文发表数量；
The number of single-authored publications是指各人每年在学术期刊上发表的独立作者文章数量；
The number of co-authored publications是每年与其他研究人员合作发表论文的数量；
Events of local collaboration is the sum of the collaborative activities carried out by each person with the researchers of the unit every year, and the number of collaborative events is slightly higher than the number of collaborative papers;
Cognitive similarity衡量担任院长前后所发表学术论文主题之间的相似性；
Receiving a major grant is a dummy variable that is assigned a value of 1 if the fund is received in a given year.
The descriptive statistics of the variables are shown in the following table. In all samples, the mean age of becoming a dean was 44.2 years. Of the 99 deans, only 6 are women, and 11 have received their PhDs from both the United Kingdom and the United States. On average, each person publishes 1.9 papers per year, of which 1.3 are collaborative papers. Ninety-four per cent of these publications are published in non-top-tier journals, and local collaborations are about four times more frequent than non-local collaborations. The average number of projects supported by research grants was 7.6%.
(3) Identification strategy:
The first identification strategy used in this article is shown in the following formula, which is used to identify year-over-year impact:
The second identification strategy is shown in the following equation, which is used to identify the average impact:
Empirical results
(1) The influence of power on academic publishing:
The results of the first identification strategy are shown in the following table:
Column (1) of the table above shows that the number of papers published by a scholar increases significantly after he or she becomes dean, and this effect increases over time, ranging from 0.5 to 2.9 papers per year. The results in column (2) were similar to those in column (1) after controlling for gender, overseas experience, age, and age squares, suggesting that this result could not be explained by the life cycle theory (the relationship between publication and age).
The results were similarly similar when the fixed effect at the individual level was added to column (3). Column (4) further controls the number of faculty members and research funding, and the results remain similar to before.
Visualizing the estimated coefficients for column (4) in the table above, as shown in the figure below, shows two distinct patterns. First, after becoming dean, the number of scholarly publications increased significantly. Second, contrary to the number of scholar's publications before becoming dean, the number of scholarly publications did not show an increasing trend before they became dean. That is, the first explanation, the "ability effect", is not supported by evidence.
However, one possible problem is that missing variables at the college level can lead to problems with this result. For example, an increase in human or physical capital can affect the publication of papers by all researchers over time. Therefore, in column (5) of the table above, this paper uses the faculty-wide publication as the explanatory variable for regression. The results show that there is no significant increase in the overall academic publication of these scholars after they are promoted to deans, suggesting that the benchmark results in this paper are unlikely to be due to college-level shocks.
In the last row of the table above, the average effect is given in this article. On average, serving as dean increases the number of papers published by scholars per year by 0.7-0.8, or about 37-42% of the average. In addition, nonlinear models are also used in this paper, and the table below shows the results using Poisson and negative binomial regression, and when these nonlinear models are used, the estimated result is about 0.5-0.6 articles per year.
Another way to study the role of administrative power in academic publishing is to look at the impact of the disappearance of power on academic publishing (i.e., to discuss whether there is a "human journey"). This is relatively rare, as deans tend to be promoted when they leave office. In the 1990-2009 dataset constructed in this paper, only two deans ceased to hold administrative positions after leaving office. Therefore, this paper expanded the sample from 2009 to 2015 to obtain a subsample that included 13 deans who no longer held any managerial positions after leaving office. As shown in the table below, the loss of administrative power leads to a significant reduction in the number of academic papers published. This result provides further support for the influence of executive power on academic publishing.
(2) How does executive power affect the publication of academic papers?
(1) Local cooperation VS non-local cooperation
In order to explain the reason for the increase in the number of academic papers, the papers of individual authors and co-authors were first examined separately, and the results are shown in the table below, showing that the increase in the number of papers is due to the latter. Columns (1) – (2) of Table 3 show that the number of co-author papers has increased since becoming president. Analysis of the average effect showed that serving as dean resulted in an increase of 0.71 co-author papers per year, more than 50% of the average (1.303), but had little effect on the number of single-author papers.
For co-author papers, this paper further uses the author's affiliation to define local collaboration (with academics from our university) and non-local collaboration events (with scholars from other universities), and the results are shown in columns (3)-(4) of the table above. The results showed that local cooperation did not show an upward trend before the appointment of the dean, but after the promotion to the dean, local cooperation increased significantly, and there was no significant change in field cooperation. Estimates of the average effect suggest that the increase in the number of scholarly publications is driven by local collaboration: after becoming dean, local collaboration will increase by 0.655 per year, which is about 49% of the average (1.343). Conversely, the impact on non-local cooperation is negligible.
Another way to discuss these results is to study the number of collaborators. In this dataset, the average person had 8.1 collaborators before becoming dean, and 13.7 collaborators after serving as dean, indicating that more people wanted to work with dean. Taken together, these results demonstrate how executive power influences academic outcomes: academics from the University who prefer to collaborate with or with the Dean are more likely to be influenced by the Dean's administrative power, thereby increasing the Dean's own publications.
and (2) changes in the subject of the study
This paper further examines the research fields and research topics of academic papers before and after the dean's appointment. As mentioned earlier, using articles published more than five years before taking office as a benchmark to define a scholar's field of study, and examining the topic similarity of his scholarly publications year by year, the results are shown in the following table: When a scholar becomes dean, the research topic of his academic paper deviates significantly from his earlier research. This result suggests that serving as dean affects not only the number of academic papers published by a scholar, but also the subject matter of the academic papers.
and (3) the power effect of different universities
In order to examine the heterogeneity of power on academic publishing between different universities, we categorized universities into 985 universities and non-985 universities (i.e., top universities and non-top universities), as shown in the table below. Columns (1)-(2) of table 5 provide estimates of year-on-year impacts. As with previous results, there was no significant increase in academic publications prior to the dean's tenure. However, after being promoted to dean, the academic papers published by these scholars from non-985 universities have increased significantly. The last row of the table below reports the average impact. Among the non-985 universities, the number of academic papers published increased by an average of 0.854 per year after becoming deans, which was about 54.7% of the average (1.561). However, among the 985 universities, the impact was small (only 21.4% of the average) and not significant. These results show that the influence of administrative power is weaker in China's top universities than in non-top universities. There are several possible explanations for this result: for example, there is more competition among peers at top universities that can limit abuses of power. Another possible explanation is that faculty members tend to have fewer external options at non-top universities, and therefore have a greater incentive to "bribe" those in power.
and (4) the power effect of different journals
This article is also discussed according to the quality of the journal. By performing sample regression in top journals and non-top journals, it is proved that serving as dean can only increase academic publications in non-top journals. Columns (3)-(4) of the table above show that serving as dean has almost no effect on the number of papers published in top journals, but significantly increases academic publication in non-top journals. This result shows that an increase in the number of academic papers published does not mean an increase in quality.
(3) Why does executive power affect the publication of academic papers?
(1) Ability effect
The "competency effect" refers to the fact that scholars with higher academic standards and are in the academic upward stage are more likely to be promoted to dean. But this hypothesis makes it difficult to explain the results of this paper for at least three reasons. First, the identification strategy in this paper includes individual fixed effects and individual publication trends, and the results have shown that there is no significant change in the number of academic papers published by scholars before becoming deans.
Second, after considering the scholar's publication productivity before becoming dean, the results show that competence is unlikely to be a confounding factor in the benchmark regression of this paper. Here, a scholar's productivity is measured by the average number of papers published each year during the first five years of his tenure as dean. Specifically, scholars are divided into two categories: those whose productivity was above the median and those who were below the median before becoming deans. The results of sub-sample regression are shown in the table below. On average, the high-yielding group had more than twice as many publications as the low-producing group. However, for the low-yielding group, the dean was important, while for the high-yielding group, the effect was more than 60% higher, while the latter was smaller (about 30%) and not significant. This result further suggests that the competency effect is unlikely to be a driver of the efficiency of executive power.
Thirdly, in the previous results, this paper also found that there was a situation of "people walking tea and cooling". If executive power is not related to academic publications, it is not important to expect a loss of power after controlling for life-cycle changes in productivity by adding age and age squared terms to the regression.
(2) Reputation effect
The reputation effect is that scholars who serve as deans are more well-known and therefore more likely to be co-authors to increase the probability of publication of academic papers. However, the results of this paper do not support this interpretation. For example, if the reputation effect is the main mechanism by which executive power influences academic publishing, then we can expect this effect to be even greater for deans with higher productivity, but this is the exact opposite of what our data reflects. In addition, if the reputation effect is the main driving mechanism, then the influence of executive power should extend beyond the administrative jurisdiction of the president. In reality, however, the increase in scholarly publishing brought about by the influence of administrative power is mainly due to the increase in local collaborations (fellow scholars). In addition, the previous results also showed that the increase in the number of academic papers published actually deviated from the original research field of the dean, which also reflected the unreliability of the reputation effect.
(3) Resource effect
The resource effect is an explanation that power affects resource allocation, which is also the most common consideration among Chinese scholars, and many have criticized the power-oriented approach of the Chinese university research funding system. In the School of Economics, the dean, as the most powerful person, dominates the allocation of resources in the school. Outside of the college, the dean serves as the representative of the college and is sometimes responsible for grants and funding from the university, government, or business. Therefore, our findings are most consistent with this interpretation. In particular, since the power of the deans is often confined to their own university, the increase in their collaborative publications is largely driven by the increase in local collaboration. Other hypotheses cannot explain this.
To provide more direct evidence, we looked at whether executive power would help the dean get more funding projects. The results are reported in the table below. The results show that, after including various control variables, the effect of executive power is still statistically significant at the 5% significance level, which is still considerable relative to the mean: serving as president increases the probability of receiving a grant by about 6.3 percentage points, or about 86% of the mean. It is important to note that the fund items in the table only include projects from the central government (i.e. what is often referred to as vertical projects). Due to a lack of data, there is no funding information from local governments or industry (horizontal projects). As a result, the table below underestimates the impact of the President's executive power on the allocation of resources.
It should be noted that the above evidence relates to only one of the many resource allocations, and that the President's executive authority may also bring about other types of resources. Based on case studies, some scholars argue that even if students are not adequate, some deans can use their influence to help their students earn national awards, which is also a complement to the resource effect.
Thus, even if we are unable to quantify every aspect of resource allocation, the evidence is sufficient to show that the influence of power on academic publication can be explained by resource effects. This explanation is also consistent with our previous heterogeneity analyses.
conclusion
This paper examines the link between administrative power and academic publication in China. By constructing a unique dataset, this paper explores the enormous impact of administrative power on the publication of academic papers. The evidence presented in this paper suggests that the mechanism by which the executive power of the deans influences their academic publication is that academics at the University who work with or for these deans are more likely to be influenced by their power and contribute to their academic publication through collaboration.
The results of this paper provide a new perspective on this issue and have important policy implications. Many studies have shown that the generation of new ideas is critical to a country's long-term economic growth, that the management of universities affects the output of research outputs, and that poor resource allocation can have a negative impact on economic development. Given that academia is an important contributor to knowledge production, we should try to ensure that its resources are allocated appropriately and efficiently. Unfortunately, the findings of this paper suggest that the current allocation of resources in Chinese academia is overly influenced by executive power. However, based on our analysis of different journals and different sizes, the influence of executive power can be limited through appropriate systems and methods.