Nonparametric estimation of the marginal effect in fixed-effect panel data models

Impact of digital finance on energy-based carbon intensity: Evidence from mediating effects perspective

This research measures energy-based carbon intensity based on energy consumption of 30 Chinese provinces and investigates the impact of digital finance. First, the baseline results were examined through panel data model. Second, we reveal the "black box" system of digital finance and energy-based carbon intensity through mediating effects model. Third, by employed panel quantile regression model, we examine the heterogeneity of the effects of various factors under different carbon intensity quantile. The main quantitative results are as follows: (1) Digital finance shows significant sustainable effect, and its growth of 1% will reduce the carbon intensity by about 0.092%. (2) The intermediary effects of technological innovation and energy structure are -0.0663 and 0.007, respectively, accounting for 41.88% and 19.36% of the total effects. Both are significant transmission mechanisms. (3) When the carbon intensity in >0.9 quantile, digital finance shows a positive coefficient. On the contrary, the coefficient at <0.75 quantile is negative and its absolute value increases with the decrease of quantile. It shows that with the reduction of carbon intensity, the positive impact of digital finance gradually increases. (4) In the eastern, central and western regions, a 1% increase in digital finance will reduce the carbon intensity by 0.147%, 0.096% and 0.089% respectively. This research provides a reference for regional governments to use digital finance tools to promote sustainable development.

Flypaper effect assessment methods in the expansion of regional autonomy

Flypaper Effect is a public finance term that indicates a government grant given to recipient cities increases the local community spending level more than an increase in local income of equivalent size. This paper analyzed the Flypaper Effect Assessment Method in the Expansion of Regional Autonomy. It employed 210 New Autonomous Regions (NARs) in Indonesia during 1999-2021 as a case study, where Indonesia became the country with the highest number of new autonomies in the world. Panel Data Regression was utilized to determine the Flypaper Effect. Flypaper Effect analysis was carried out using the BLUE model selection method. The selected models in this study were Pooled Least Square (PLS), Fixed Effect Model (FEM), and Random Effect Model (REM). Several tests, such as Chow Test, Lagrange Multiplier Test, and Hausman Test, were conducted. Furthermore, the procedures to get the data in BLUE were carried out, such as Heteroscedasticity and Autocorrelation Test. Koenker-Bassett test was used for ascertaining Heterocedascity.•Panel Data Regression is used as a method to determine the Flypaper Effect in the autonomous region.•Each stage in this method is discussed with a calculation/process example.•The method utilized in this paper is recommended to determine the Flypaper Effect of New Autonomous Regions (NARs) for various parties.