The paper begins by establishing the systematic nature of legal systems and the potential for conflict between laws of different levels (e.g., national laws versus local regulations). It highlights the common occurrence of such conflicts and contradictions in China's legal system, despite mechanisms like the Legislation Law aiming to resolve inconsistencies. The focus then shifts to the impact of legal revisions on the broader societal context, particularly when higher-level laws are revised without corresponding changes at lower levels. This creates a potential gap between legal intent and practical implementation due to the public's difficulty in navigating conflicting regulations, leading to a lack of compliance or inconsistent practices. The paper notes the absence of empirical research on this mechanism before focusing on the reform of the patent transaction system in Chinese colleges and universities as a quasi-natural experiment. This reform, initiated by the Law on Promoting the Transformation of Scientific and Technological Achievements in 2015 and further modified by the revised Interim Measures for the Management of State-owned Assets of Public Institutions in 2019, presents an ideal opportunity to analyze the blocking and transmission effects of legal inconsistencies. The paper posits two hypotheses: H1: Post-2015, university patent transaction efficiency decreased due to conflict. H2: Post-2019, efficiency was restored due to revised lower-level systems.

The literature review acknowledges existing research on legal conflicts, primarily focused on private international law (conflicts between different countries' laws). However, it points out the limitations of this research for understanding internal legal conflicts within a single legal system, especially in a rapidly developing legal environment like China's. The review highlights the lack of macro-level, empirical studies on domestic legal conflicts and their effects on policy implementation. While some studies analyze individual conflicts and offer legal interpretations, the paper notes the scarcity of research using empirical methods to assess the societal impact of such conflicts. This is particularly important in science and technology-related fields, where evaluation methods from management economics and statistics are more mature. Finally, the review highlights existing research on university patent transactions, which provides a foundation for the empirical study's model and variables. The review highlights the 2015 law reforming patent transactions and the 2019 revision of the asset management measures as a case study of upper and lower legal system conflict.

The study employs a difference-in-differences (DID) estimation model, a quasi-experimental approach suitable for analyzing the effects of policy changes. It uses mixed cross-sectional data from three periods: before the 2015 law, between the 2015 and 2019 reforms, and after the 2019 revision. The data covers university and enterprise patents, with enterprises serving as the control group and universities as the treatment group. The DID model allows for the estimation of the treatment effect by comparing the change in patent transaction efficiency in the treatment group (universities) with the change in the control group (enterprises). The dependent variable is patent transaction efficiency, measured as the time between patent publication and transaction filing. A shorter time indicates higher efficiency. Explanatory variables include dummy variables for time (post-2015 and post-2019) and the treatment group (universities). Control variables encompass patent transaction type (exclusive, sole, general license), and patent characteristics (number of claims, number of pages, number of applicants, IPC classifications, number of patents in the same family). Data is sourced from the Incopat database. Two DID estimations are performed, one for each reform period. Regression equations are presented to model the changes in patent transaction efficiency for both universities and enterprises before and after each policy change. The paper then explains how the second difference is calculated to isolate the effect of the policy reform. The paper then elaborates on the various variables used in the study, including the dependent variable (patent transaction efficiency), explanatory variables (treatment and time dummies), and control variables that account for patent transaction modes and individual patent characteristics. The data description details the source of the data (Incopat), the sample selection process, and the handling of missing or extreme values. Finally, the paper outlines the process of statistical analysis which includes a basic regression analysis to determine the initial impact of the reforms, and robustness tests including parallel trend tests and counterfactual tests to strengthen the validity and reliability of the findings.

The basic regression analysis (Table 3) shows that after the 2015 reform (Stage 1), university patent transaction efficiency significantly decreased, indicating a blocking effect caused by the inconsistency between the higher-level law and the lower-level regulations. Conversely, after the 2019 reform (Stage 2), which harmonized the higher and lower-level systems, university patent transaction efficiency improved significantly. The results in Table 3, both before and after the inclusion of control variables, confirm that the interaction terms are significant, supporting the hypothesis that the legal reforms significantly impacted patent transaction efficiency. The parallel trend test (Table 4) supports the assumption that the treatment and control groups followed parallel trends before the policy interventions, validating the use of the DID model. The counterfactual test (Table 5), involving a one-year shift in the policy intervention time, further supports the robustness of the findings by showing that the significant changes in patent transaction efficiency can only be attributed to the policy interventions. The difference between the magnitudes of the DID coefficients in the two phases suggests that the efficiency improvement in Stage 2 exceeded the efficiency loss in Stage 1.

The findings demonstrate the empirical existence of institutional conflict in the legal system and clarify the underlying mechanism. The inconsistency between higher and lower laws creates a blocking effect, hindering the implementation of the higher-level law and its intended benefits. Harmonizing the legal framework through revisions at lower levels produces a transmission effect, improving implementation and achieving the intended goals. The study's results underscore the importance of a coherent and well-coordinated legal system for effective policy implementation, especially in the context of dynamic reforms and evolving regulations. The study adds to the limited empirical evidence in the field of domestic legal conflict and its impact on socio-economic outcomes. The significant results of the DID analysis, supported by robustness checks, strengthen the argument that the observed changes in patent transaction efficiency are indeed a result of the legal reforms and not due to other confounding factors.

This study makes several significant contributions. It moves the study of legal conflicts from theoretical discussions to empirical analysis, providing concrete evidence of the blocking and transmission effects of inconsistencies within the legal system. It highlights the importance of aligning higher and lower-level laws to ensure policy effectiveness. Future research should extend this analysis to other areas of law and policy and explore various aspects of legal conflicts such as conflicts between laws of similar rank, or conflicts within a single law.

The primary limitation is the focus on invention patents, and the measurement of transaction efficiency based only on the time between publication and filing of the transaction. This does not consider patents that never undergo transactions, potentially underestimating the overall impact of the legal reforms. Also, the study is limited to the Chinese context, and its generalizability to other legal systems might be limited.