The policy efficiency evaluation of the Beijing–Tianjin–Hebei regional government guidance fund based on the entropy method

In this paper, we mainly searched the keyword of ‘government guidance funds’ on the China Laws & Regulations Database and pkulaw.cn and collected policies related to government guidance funds issued by various government departments in Beijing, Tianjin and Hebei Province from 2005 to 2018. After excluding irrelevant policies, 518 policies of government guidance funds were finally included in the sample, including 137 policies in Beijing, 56 in Tianjin and 325 in Hebei Province.

This paper uses ‘Government Guidance Funds’ as the search term for the full text of laws and regulations, and collects and sorts out the policies of government guidance funds issued by multiple government agencies in the Beijing–Tianjin–Hebei region from 2005 to 2018 in the Chinese Law Search System (pkulaw.cn). After necessary screening and elimination, 518 policies of government guidance funds were finally included in the research sample, including 137 in Beijing, 56 in Tianjin and 325 in Hebei Province.

The quantitative results of the policies of government guidance funds in the three regions of Beijing, Tianjin and Hebei from 2005 to 2018 are shown in Table 1. From its data, it can be seen that the quantitative values of the policies of government guidance funds in the three regions of Beijing, Tianjin and Hebei are showing an overall upward trend, indicating that the effectiveness of policies is gradually strengthened over time. According to the results of policy quantification among the three regions, the policy level of Hebei Province is higher, followed by Beijing City and Tianjin City.

Fig. 1

As the main manifestation of government guidance funds, policies are generally displayed in verbal form. However, since words cannot be directly substituted into formulas and models for calculation, we need to process the policies as data. The specific methods for the quantification of policies of government guidance funds mainly refer to a set of policy quantification standards formulated by Peng Jisheng et al. (2008) in the process of quantifying and quantitatively analysing the national technology innovation policies in the past 30 years [13]. This paper draws on the manual to numerically process the policies of government guidance funds in Beijing, Tianjin and Hebei. Besides, to make the policy content can be fully displayed by the data, this article scores each policy in terms of policy intensity, policy objectives and measures. Among them, according to the characteristics of the policy content of the government guidance funds, this paper has formulated 12 secondary indicators in three aspects: intensity, objectives and measures to score the policy in an all-round way. The detailed policy scoring standards are shown in Table 2.

The scoring standards established in Table 2 mainly quantify policies in a comprehensive manner from the three aspects of policy intensity, objectives and measures. The intensity is mainly based on the effectiveness of the level of documentation; the scoring standard for objectives is mainly based on the level of detail of the goal description in the policy and the degree of the tone; the measures are mainly scored from 0 to 5 based on the detailed level of the measures mentioned in the policy so as to determine the final score of a single policy. Finally, the final score of a single-year policy is added to the final quantitative value of all policies in the corresponding year, and the specific calculation is as follows: EAMpt=∑i=1n(mi+ai)ei {EAM}_{pt} = \sum\limits_{i = 1}^n ({m_i} + {a_i}){e_i}

In the formula, p is the province where the scoring policy was published; t refers to the published time of the scoring policy, i refers to the i-th policy promulgated in the p-th province in year t; stands for the total score of the i-th policy measure (measure) and the policy objective (aim); refers to the score of the policy intensity of the i-th policy; and refers to the overall status of the policy contents, objectives and measures of the government Guidance funds for the year t in the P province.

This paper establishes an output indicator system in terms of innovation performance and economic performance. First, the first-level indicators are weighted by the AHP method, and then the second-level indicators are weighted by the entropy method. Finally, the policy output of government guidance funds in the Beijing–Tianjin–Hebei region was evaluated comprehensively.

This paper adopts the GRA to analyse the relationship between the input and output of the policies of government guidance funds, chooses the final quantitative value of the policies in Beijing, Tianjin and Hebei Province as the reference sequence and the secondary indicator data of the policy output as the comparative sequence. Suppose the reference sequence is X₀, X₀ = {x₀(1), x₀(2),. . . ,x₀(n)} and the comparative sequence is X_i, X_i = {X_i (1), X_i (2),. . . , X_i (n)}, i = 1, 2,..., m.

To avoid the inability to make comparisons due to the inconsistent units and quantity levels of the indicators in the system, it is necessary to preprocess the data of the secondary indicators so that the indicators can be directly compared. Therefore, data standardisation is processed for all indicators in this paper: γ(k)=Yi(k)Y¯i \gamma (k) = {{{Y_i}(k)} \over {{{\bar Y}_i}}} where k=1,2,3,. . . ,n; i = 1,2,. . . ,m

This paper uses the standardised data of the reference sequence and the comparative sequence to solve the correlation coefficient. The formula is: ξx0(k),xi(k)=minimink|x0(k)−xi(k)|+ρ×maximaxk|x0(k)−xi(k)||x0(k)−xi(k)|+ρ×maximaxk|x0(k)−xi(k)| \xi {x_0}(k),{x_i}(k) = {{\mathop {\min}\limits_i \mathop {\min}\limits_k |{x_0}(k) - {x_i}(k)| + \rho \times \mathop {\max}\limits_i \mathop {\max}\limits_k |{x_0}(k) - {x_i}(k)|} \over {|{x_0}(k) - {x_i}(k)| + \rho \times \mathop {\max}\limits_i \mathop {\max}\limits_k |{x_0}(k) - {x_i}(k)|}}

Here, Δi(k)=|x0′(k)−xi′(k)| {\Delta _i}(k) = |x_0^{'}(k) - x_i^{'}(k)| ; the following formula can be obtained: ξx0(k),xi(k)=miniminkΔi(k)+ρ×maximaxkΔi(k)Δi(k)+ρ×maximaxkΔi(k) \xi {x_0}(k),{x_i}(k) = {{\mathop {\min}\limits_i \mathop {\min}\limits_k {\Delta _i}(k) + \rho \times \mathop {\max}\limits_i \mathop {\max}\limits_k {\Delta _i}(k)} \over {{\Delta _i}(k) + \rho \times \mathop {\max}\limits_i \mathop {\max}\limits_k {\Delta _i}(k)}} wherein miniminkΔi(k) \mathop {\min}\limits_i \mathop {\min}\limits_k {\Delta _i}(k) is the two-level maximum difference of the data, maximaxkΔi(k) \mathop {\max}\limits_i \mathop {\max}\limits_k {\Delta _i}(k) is the two-level maximum difference of the data, ξ x₀(k);x_i(k) is the correlation coefficient between the k-th indicator of the comparative sequence x_i and the k-th indicator of the reference sequence X₀ and ρ is the resolution ratio. The smaller the value of ρ, the larger the resolution. Generally, the value range of ρ is in the range of 0 to 1 and ρ is usually taken as ρ = 0.5.

Finally, the GRA is solved according to the previous data, and the formula is: ∂(x0,xi)=∑k=1n(x0(k),xi(k))n. \partial ({x_0},{x_i}) = {{\sum\limits_{k = 1}^n ({x_0}(k),{x_i}(k))} \over n}.

According to previous studies, when ρ = 0.5, ∂ (x₀, x_i) > 0.7, the effect is ideal, indicating that the correlation between the two factors is relatively strong and the greater the GRA, the stronger the correlation.

The units and quantity levels of the four secondary output indicators of the policies of government guidance funds are not uniform, which will affect the results of data analysis, so they cannot be directly compared by weighted sum. First, this paper standardises the output indicators of government guidance funds. Since the four output indicators in this paper are all positive indicators, we standardise indicator data so that the result value is between 0 and 1. The formula is as follows: Tab=tab−tmintmax−tmin {T_{ab}} = {{{t_{ab}} - {t_{\min}}} \over {{t_{\max}} - {t_{\min}}}}

T_ab in the formula is the result of normalisation of the data, and T_ab refers to the original value of each indicator; t_min and t_max are the minimum and maximum values of various indicators, respectively.

The results of the normalisation of the policy output indicator data of the Beijing-Tianjin–Hebei region are shown in Table 5. From Table 5, it can be seen that the technology market has the largest transaction volume in Beijing (2018) and the smallest transaction volume in Hebei Province (2005); Beijing City (2018) has the largest number of domestic invention patent applications, and Hebei Province (2005) has the smallest number; Hebei Province has the largest regional GDP (2017), and Tianjin has the smallest GDP (2005); Beijing has the largest regional GDP per capita (2018), and Hebei Province has the smallest GDP per capita (2005). From the above, we can find that the maximum values of innovation performance output indicators all appear in Beijing, and all the minimum values appear in Hebei Province and all are distributed in 2005; while the maximum values of economic output indicators are in Beijing and Hebei, the minimum values are in Tianjin and Hebei.

According to the above-standardised data and the combination weights obtained by AHP and entropy method, the weighted sum is conducted to obtain the standardised policy output values. The results are shown in Table 6. It can be seen from Table 6 that the policy output of Beijing, Tianjin and Hebei is showing an overall upward trend; among them, Beijing has the highest policy output book reviews. The policy output level of Beijing in 2018 was >8 times that of 2005, reflecting the remarkable achievements of Beijing's policy output. The policy output level of Tianjin was slightly higher than that of Hebei Province, ranking in the middle of the three cities, showing a steady upward trend on the whole. Hebei's policy output level lags far behind Beijing with the growth rate lower than that of Beijing and Tianjin.

This paper uses the GRA method to measure the policies of government guide funds and the second-level indicators of policy output. The results are shown in Table 7. From Table 7, we can see that the GRA range of government guide fund policies and policy output in Beijing, Tianjin and Hebei is 0.614, 0.814, indicating that there was a strong correlation between policy and policy output indicators during the period from 2005 to 2018, which showed that the government guide fund policies of the three areas were effective on the whole.

Taking the GRA value of 0.7 as the criterion, there are seven combinations with significant influence and five combinations with insignificant influence, and all insignificant combinations appear in Beijing. This shows that the policies of the Beijing municipal government guidance funds tend to be less effective than the other two places. The policies of government guidance funds have the strongest impact on the turnover of the technology market in Hebei Province, while the impact on the per capita GDP of Beijing is the weakest.

According to the GRA ranking of the policies of government guidance funds of Beijing, Tianjin, Hebei and the various policy output indicators, the first two value indicators of each region ranked from largest to smallest are taken, respectively: Beijing (domestic invention patent application > turnover of technology market), Tianjin (turnover of technology market > domestic invention patent applications and regional GDP) and Hebei Province (domestic invention patent applications > domestic invention patent applications > turnover of technology market). It can be seen from the above that the policy output effects of Beijing and Hebei are relatively similar, and the government guidance funds in the Beijing–Tianjin–Hebei region are more effective in promoting innovation performance output. It also proves that government guidance funds have a positive impact on corporate innovation and entrepreneurship.

After analysing the policy efficiency of the government guidance funds on the four secondary indicators of policy output, the result is not intuitive enough. Therefore, in this paper, the four indicators were weighted and averaged by the overall weight to synthesise a comprehensive output index, which was analysed by GRA with the quantitative value of the policy, and the result was taken as the policy efficiency of government guide fund. The results of the study are shown in Table 8.

From the perspective of the policy efficiency of government guidance funds in Table 8, with the grey correlation value >0.7 as the criterion, the government guidance funds of the Beijing–Tianjin–Hebei region and the policy output show a strong correlation, while the policy efficiency of government guidance funds in the Tianjin and Hebei is similar, reaching a relatively ideal level. Although the policy efficiency of the government guidance funds in Beijing has exceeded the judgement standard, there is still a big gap compared with the other two places. Therefore, although Beijing has the lowest policy efficiency, its innovation output is the highest among the three places. Therefore, it can be seen that Beijing's policies of government guidance funds are not the main factor for its innovation output. For Tianjin and Hebei, although the policy efficiency is relatively large, there is still a big gap from fully exerting the policy effect, and there is still more room for improvement in the policy efficiency of government guidance funds.

The GRA analysis of government guidance fund policy and policy output is carried out in three aspects: intensity, objectives and measures. The results are shown in Table 9.

Taking the GRA value >0.7 as the criterion for judging the ideal effect of policies, it can be seen that Tianjin, as the city with the most ideal output effect among the three places, is in a leading state in terms of policy intensity, objectives and measures. Although Hebei Province has achieved a relatively ideal state in terms of policy intensity and measures, there is still a lot of room for improvement in terms of policy objectives. However, Beijing has failed to achieve the desired results in terms of policy intensity, objectives and measures. Combined with its innovation indicator data, the main reason for the low-efficiency results may be the result of Beijing's relatively large number of resources in innovation and the influence of other policies.