*3.4. Empirical Model*

An effective method to explore the net effect of a policy using the DID model is by comparing treatment and control groups before and after implementing the policy. In this study, the following DID model is constructed to analyze the heterogeneous effect of the market value of listed companies in pilot regions (i.e., treatment group) and non-pilot regions (i.e., control group) before and after the implementation of the CET policy:

$$Ln(MV)\_{it} = \beta\_0 + \beta\_1 treated\_i \times time\_l + \beta\_2 \times X\_{it} + \mu\_i + \gamma\_t + \varepsilon\_{it} \tag{1}$$

where *i*, *r*, and *t* denote the listed companies, city, and time, respectively. *Ln*(*MV*)*it* is the natural logarithm of market value of the company *i* at period *t*. *treati* is equal to 1 if a company is located in one of the seven pilot provinces and cities; otherwise, it is 0. *timet* equals one for every year after 2013; otherwise, it equals 0. *Xit* represents all the control variables, including SIZE, BM, ROE, DAR, fix, ROA, MSR, lnage, cash, and subsidy. *β*<sup>0</sup> is the constant term and *β*<sup>1</sup> is the core explanatory variable that indicates the net causal impact of the CET policy on companies' market value. *β*<sup>2</sup> represents the coefficients of all control variables. *μ<sup>i</sup>* denotes the fixed effects of listed companies, *γ<sup>t</sup>* is the time fixed effect, and *εit* is the standard error term.
