This paper evaluates the treatment effect of high-speed rail (HSR) operation on city consumption using the dataset of Chinese cities from 2003 to 2019. Firstly, the applicability of observations is discussed; secondly, observations with no appropriate contrast samples are dropped for more precise empirical results. Then, propensity score matching (PSM) is implied to have the database much more balanced and suitable for the difference-in-difference (DID) framework, that is, the PSM-DID approach. The main results find a novel phenomenon of Simpson's paradox regarding the HSR-consumption nexus, which indicates that even though we can observe the positive effect on the whole, the results argue negative relationships between HSR and consumption within subclass cities. In addition, a dose-response assessment (DR) and some other checks have been proposed to demonstrate robust estimation results.

In order to avoid the real economy development lags behind and the deterioration of ecological problems in the process of traditional urbanization, China has been promoting city-industry deep integration, but the connection between city-industry integration and green economic growth, especially the spatial effect, has not been systematically explained. Based on the panel data from 2007 to 2018, this paper constructs an evaluation index system and uses the SEEA method to measure city-industry integration (CII) level and green economy growth (GEG) level. Then, by employing spatial Durbin model and intermediary effect model, it further systematically investigates the spatial impact of CII on GEG and the potential mechanism. The study found that: (1) On the whole, CII shows “slow-steady integration” trend, but regional heterogeneity was obvious and accompanied by “slow gap expansion”. GEG experienced “sharp increase” with “polarization” characteristic. (2) CII can directly promote regional GEG (with a marginal effect of 0.689), more effectively than traditional urbanization, and CII has obvious spatial spillover effects on GEG in both “local effect” and “neighboring effect”. (3) Interestingly, technological innovation and consumption structure upgrading are significant mediating mechanisms. (4) The direct positive effect of CII shows the regional imbalance characteristic. Finally, the corresponding policy implications are put forward.

Based on the panel data of western central-cities from 2002 to 2021, this paper uses the intermediary effect model to analyze the relationship between urban internationalization, innovation ability and urban competitiveness. It is found that: (1) The internationalization of western central-cities has a positive and direct impact on the promotion of urban competitiveness, which has a significant impact on economic strength and infrastructure, and a weak impact on government efficiency and market efficiency The marginal growth effect of internationalization on urban scale will gradually decrease with the improvement of urban development level. (2) The continuous promotion of city internationalization can promote the innovation ability to enhance the competitiveness of the city. Compared with the internationalization of cities, the influence of innovation ability on the competitiveness of cities is small, and the innovation ability of western central-cities and each city is weak. The scale effect has become the main motivation to promote the competitiveness of cities. (3) In the process of internationalization of western central-cities, innovation ability has a part of intermediary effect on the improvement of urban competitiveness, and the effect on urban “hard environment” is significantly greater than that on urban “Soft-environment”. The results of direct and indirect impact tests show that the impact of the process of internationalization of western central-cities on urban competitiveness is exogenous.

This paper explores how the medical expenditure risk affects the households’ portfolio choice across health status theoretically in a life cycle model and empirically using machine learning methods. Medical expenditure risk, as a background risk, has the potential to influence households’ financial decisions. A higher medical expenditure risk leads to a larger fluctuation and more uncertainty in households’ consumption and therefore utility. As a result, risk-free assets become more attractive. Our machine learning analysis provides evidence that aligns with the predictions of the theoretical life cycle model. Specifically, households with better health hold a larger proportion of stocks in their portfolios. Furthermore, when facing increased medical expenditure risk, households in good health demonstrate a greater willingness to invest in safe assets.