Green and Low-Carbon Efficiency Assessment of Urban Agglomeration Logistics Industry: Evidence from China’s Beijing-Tianjin-Hebei Metropolitan Area (2008–2020)

Abstract

With the advent of the post-industrial era, the rapid development of e-commerce has propelled the logistics industry to become the lifeline of the national economy, supporting the orderly flow of resource elements between cities. However, the concerning issues of excessive energy consumption and low logistics efficiency in the transportation process have come to the forefront. The introduction of China’s dual-carbon policy goals indicates that enhancing regional logistics’ green and low-carbon efficiency is key to solving the global logistic sustainability problem. Nowadays, the logistics sector’s efficiency in producing green and low-carbon emissions has been quantified using an input-output measurement index. Using data from 2008 to 2020 from the dynamic panel of the logistics sector in the urban agglomerations of Beijing, Tianjin, and Hebei, the three-stage SBM-DEA and Malmquist index quantitative evaluation models are selected to estimate the logistic green and low-carbon development efficiency comprehensively. The analysis discovered that green and low-carbon logistics in the Beijing-Tianjin-Hebei metropolitan agglomeration are relatively efficient overall, and the urban siphon effect of Beijing and Tianjin is noticeable. Once the impact of environmental variables and random errors is eliminated, it becomes evident that these factors tend to inflate the overall technical efficiency. Technical efficiency levels are the primary factor leading to regional logistics inefficiencies. Additionally, it is essential to note that scale efficiency positively affects urban development, leading to a rebound effect, summarizing the existing problems combined with the visualization map, and putting forward corresponding policy suggestions, which is of great practical significance.

1. Introduction

With the advent of the post-industrial era, the rapid development of e-commerce has propelled the logistics industry to become the lifeline of the national economy, supporting the orderly flow of resource elements between cities. However, the concerning issues of excessive energy consumption and low logistics efficiency in the transportation process have come to the forefront. The introduction of China’s dual-carbon policy goals indicates that enhancing regional logistics’ green and low-carbon efficiency is key to solving the global logistic sustainability problem. The Chinese government has formulated a carbon peak in 2030 and a carbon-neutral strategy in 2060, attracting widespread academic attention [1,2]. Maps and coordinates of the Beijing-Tianjin-Hebei urban agglomeration clarify the geographical location of the region, which is located in northern China and consists of Beijing, Tianjin, and parts of Hebei Province; Beijing: 39.9042° N, 116.4074° E Tianjin: 39.0842° N, 117.2009° E; Hebei: 38.0428° N, 114.5149° E; its location is shown below Figure 1. It is one of the regions with the most active economic development in China and one of the most important political, economic, and cultural centers in China. The unique geographical location provides important infrastructure and convenience for logistics development [3]. In 2020, the GDP of the Beijing-Tianjin-Hebei region will be 8639.3 billion yuan, an increase of 2.4% over the previous year, accounting for 8.5% of the national total; the added value of the logistics industry will account for 7.8% of GDP, an increase of 0.2 percentage points over the previous year; the express business volume and express business revenue will increase by 60.7% and 38.2%, respectively. However, the entire life cycle of logistics and transport often produces significant greenhouse gas emissions, especially carbon dioxide [4]. According to data surveys, the total carbon emissions of express packaging in China in 2020 will be 23.9584 million tons of carbon dioxide, which requires planting trees equivalent to Beijing’s land area to offset these carbon emissions. In 2020, China’s per capita express plastic packaging consumption will be 1 kg, equivalent to 100 medium-sized supermarket plastic bags per person on express packaging. This shows that the logistics and green efficiency of urban agglomerations are still relatively low. Therefore, it is of great significance to analyze the green and low-carbon efficiency status of the logistics industry in the Beijing-Tianjin-Hebei urban agglomeration and to put forward relevant suggestions to promote the development of the logistics industry. The environmental protection awareness and technological innovation capabilities of regional residents and enterprises provide a basis for developing favorable conditions for green and low-carbon logistics. The significance of the research is that by measuring and evaluating the green and low-carbon efficiency of logistics, the quantification of the green and low-carbon efficiency of logistics can also be used to monitor and assess the effects of government policies and measures, to promote the development of the logistics industry in a green and low-carbon direction, implement the “double Carbon” policy and the goal of sustainable development, and timely put forward corresponding improvement measures to promote the formation of a dynamic and long-term feedback mechanism between industry, academia, and research.

For the Beijing-Tianjin-Hebei urban agglomeration, the efficient operation of green and low-carbon logistics will not only help protect the environment and improve air quality but also help promote economic development and improve urban competitiveness [5,6]. By using SBM-DEA (Slack-Based Measure Data Envelopment Analysis) [7] and the Malmquist product index [8] to analyze the construction of input-output measurement indicators comprehensively, the purpose is to quantify the green and low-carbon efficiency of logistics in the Beijing-Tianjin-Hebei metropolitan area, evaluate its sustainable development level, and monitor the double carbon target green and low-carbonization efficiency level of the logistics industry under the policy. In general, SBM-DEA and Malmquist’s methods are two commonly used econometric models to evaluate and measure efficiency and productivity. It is especially suitable for green low-carbon logistics because it provides an objective, data-driven way to evaluate and measure efficiency and productivity. Unlike the traditional DEA method, the SBM-DEA method considers the problem of resource slack between regions, removes environmental variables and random errors, and can help us discover resource utilization problems and provide improvement directions. Whereas the Malmquist method is a productivity index based on measuring changes in unit productivity. It assesses regional productivity changes by comparing productivity indices from different years. It can help us compare the improvements in reducing environmental impact in different regions and provide a reference for decision-makers and policymakers. An in-depth analysis of the impact of technical efficiency and scale efficiency on the green and low-carbon efficiency of logistics can further study the specific contributions of different factors to the green and high-carbon efficiency of logistics, including the impact of changes in technical efficiency and scale efficiency on overall efficiency [9].

Based on the above analysis, this paper comprehensively evaluates the green and low-carbon efficiency of logistics and puts forward appropriate policy recommendations and implementation measures to promote the high-quality, coordinated development of the logistics industry in this urban agglomeration. Emphasized the importance of sustainable development and green low-carbon technologies and highlighted the value and practical significance of the research. Since there is not much research on the green and low-carbon development of logistics in developing countries and the research understanding is not thorough, this research aims to provide the necessary scientific basis and policy support for developing countries to achieve sustainable development and a green and low-carbon transformation of the logistics industry. The main contributions of this study are as follows: (I) panel data of 13 prefecture-level cities in the Beijing-Tianjin-Hebei region from 2008 to 2020 are selected to assess the green low-carbon efficiency of the logistics industry in Beijing-Tianjin-Hebei by combining the three-stage SBM-DEA model and Malmquist index model with more accurate measurements, and then GIS technology is used to visualize and deeply analyze the efficiency value of the green low-carbon efficiency of the logistics industry. Furthermore, GIS technology is used to visualize and deeply analyze the efficiency value of the logistics industry, which extends the research method of logistics industry efficiency. (II) Based on the need for international green and low-carbon city construction and coordinated development of logistics in city clusters, we establish a city cluster logistics input-output evaluation index system based on the perspective of dual carbon objectives. This system expands the perspective of logistics evaluation in urban clusters and is conducive to promoting sustainable development. (III) Under the strategic requirement of high-quality, coordinated development of the Beijing-Tianjin-Hebei logistics industry, corresponding policy suggestions and implementation measures are proposed, which are conducive to accurately grasping the laws and trends revealing the development of green logistics in city clusters, enhancing the green logistics construction capacity of city clusters, promoting the transformation of the logistics industry, and optimizing industrial structure.

The rest of the components are organized as follows: Section 2 reviews the relevant literature. In the third part, we expound on this study’s assessment index and data construction and analysis of the logistics industry carbon emissions concept and the three-stage SBM-DEA and Malmaquist index model construction processes. The fourth part analyzes the research results through tables and visual maps and discusses the limitations and possible research directions. The fifth part summarizes the main conclusions of the full-text study and puts forward policy recommendations to promote the green development of the logistics industry.

2. Literature Review

In existing research, literature related to green and low-carbon logistics has conducted a more in-depth exploration of the impact of logistics activities on the environment. The focus is usually on carbon emissions, energy consumption, and waste generation. Various methods and models are used in the study, such as Carbon footprint analysis, Life-cycle assessment, and input-output analysis. However, there are still some gaps and limitations in current research. For example, there is a lack of research on specific industries or regions, limitations in methodology, and the effectiveness of implementing particular strategies and policies. These gaps provide opportunities for future research to fill knowledge gaps and promote the practice of green and low-carbon logistics. Mainly focused on the following aspects: (1) research on different logistics nodes, (2) build a low-carbon logistics index system, and (3) evaluation method Model selection.

First, research on Different Logistics Nodes: Research nodes on logistics efficiency in recent years have focused chiefly on logistics parks [10], provinces [11], railways [12], aviation [13], and sea transportation [14]. With the arrival of the post-industrial era, the logistics industry has gradually become the artery of the national economy, supporting the orderly flow of resource elements between cities, and resulting energy consumption and environmental pollution issues have become prominent; from a vertical perspective, academics began to focus on the issue of carbon emissions in logistics after 1995, [15] believing that the definition of carbon-neutral logistics is the low-carbon management of forward and reverse logistics processes. According to the carbon footprint, to cut down on carbon emissions throughout the supply chain, the study [16] cut down on carbon emissions across the board in the operations of the supply chain.

Second, construction of a green and low-carbon logistics index system: From a horizontal perspective, the green and low-carbon logistics industry’s evaluation indicators are produced primarily from three directions: input, expected output, and non-expected output indicators. Some typical articles choose total transportation mileage, logistics capital investment, and employee count as input indicators, freight volume, cargo turnover, logistics industry GDP, and CO₂ emissions as output indicators; logistics industry density, urbanization level, and logistics specialization level as environmental indicators variables analyzed the static differences and dynamic changes in low-carbon logistics efficiency [5].

Third, the evaluation model of carbon efficiency. Regarding efficiency evaluation methods, parametric and non-parametric methods are the mainstream research methods for measuring efficiency. For example, Data Envelopment Analysis [17,18], Stochastic Frontier analysis [19], Tobit model [20], Analytic Hierarchy Process (AHP) [18], Kernel Density estimation [21], Hot Spot analysis [22], and Regression analysis ([11,19,23] are commonly used to measure logistics efficiency. Regarding parameter research: [24] used the SBM-DEA model to accurately measure the efficiency of green logistics. Using the flat-panel data of 285 cities in China from 2005–2019 as a natural experimental time-varying double-difference simulation to identify the effect that it has on the effectiveness of green logistics, we found that innovative city pilot policies can drive green logistics efficiency by accounting for regional, city-scale, and distribution network differences. Ref. [10] proposed the ecological efficiency of logistics parks; it was quantitatively evaluated for environmental efficiency and sustainable development, and the link between many factors and ecological efficiency was investigated. In terms of non-parametric research: [25] used the fuzzy analytic hierarchy process (FAHP) to determine the criteria used in the MCDM model, picked the weight of each standard, and used geographic information system (GIS) technology to analyze the MCDM model, making spatial data visualization possible. Ref. [23] employed projection pursuit regression and neural networks to study China and its provinces’ logistics industry CO₂ emissions from 2010 to 2019. Carbon emissions, input surplus, output deficit, economic development, and environmental quality in Chinese provinces were detected [17]. The hyper-relaxation data envelopment analysis (DEA) model was coupled with the global Malmquist index, and the spatial autocorrelation of international and local Moran indices was used to analyze the eco-logistics efficiency of 11 provinces and cities in the Yangtze River Delta from 2004 to 2016. [26] combined cross-efficiency evaluation (CEE) with decision-maker psychological aspects to estimate urban logistics efficiency in my country’s central region of the Yangtze River Delta urban agglomeration in 2019 and presented a combination of prospect theory and ordered weighted average (OWA). Operators’ mixed CEE technique broadens urban logistics performance evaluation.

To sum up, the current research on the efficiency evaluation of the logistics industry has achieved fruitful results, showing a variety of methods and technical applications for the efficiency evaluation of the logistics industry and, at the same time, providing a research approach for the development of evaluation indicators for the logistics industry. Among them, some pieces of literature pay special attention to the research issues of eco-efficiency and sustainable development and the effectiveness of logistics industry policies. However, there are few studies on the logistics of high (county) level cities and underdeveloped areas, especially urban agglomerations. In terms of research methodology, most research methods are based on a static investigation of the current situation of logistics development in the Beijing-Tianjin-Hebei metropolitan area. Static analysis cannot scientifically examine urban agglomeration logistics development laws and trends. In addition, questions that consider comparative assessment results or carbon emissions inputs are missing. Therefore, against the background of China’s implementation of the dual-carbon policy, this paper combines cutting-edge quantitative models and takes the Beijing-Tianjin-Hebei urban agglomeration as a research case to evaluate the status quo of the green and low-carbon efficiency of the logistics industry, which has theoretical value and practical significance.

3. Evaluation Index and Methodologies

Based on the above analysis, input and output indicators are constructed to evaluate the logistics industry’s green and low-carbon efficiency and define the logistics industry’s carbon emission calculation and research data processing. After that, we introduce a three-stage SBM-DEA and Malmquist production index as logistic efficiency evaluation methods and the principles and formulas of the evaluation model in detail, highlighting the value of the choice of research method—there is no need to quantify the data before building the model using the DEA method. There is no need to assign weights manually.

3.1. Logistics Industry’s Green and Low-Carbon Evaluation Index System

3.1.1. Variable Selection

Due to the lack of a distinct segment within the logistics business, this paper refers to most literature that currently uses the transportation, warehousing, and postal industries, which account for more than 85% of the logistics industry, as representatives. Based on the principles of representativeness, objectivity, and availability in variable selection, literature research and data query are carried out in combination with the green and low-carbon development direction of the logistics industry to select the following relevant variables:

Input factors usually include capital and labor. Through the investigation of the factors that affect the development of the logistics industry, the investment in fixed assets and the total wages of employees in the logistics industry are selected to measure the investment in logistics capital and labor force. In 2019, the logistics sector in China accounted for 11.8% of China’s total energy consumption, according to figures, and the logistics industry’s carbon dioxide emissions accounted for 12.2% of the carbon dioxide emissions of the entire community. To meet the requirements of high-quality economic development, logistics carbon emissions should be used as an input indicator for green and low-carbon logistics.

Table 1. Green and low-carbon evaluation index system of the logistics industry.

Index Type	First-Order Index	Secondary Index	Unit
Input index	Labor force	Total wages of logistics employees [27]	Thousand yuan
	capital	Fixed assets of the logistics industry [28]	Million yuan
	Energy consumption	Carbon emission	megaton
Output index	Tangible output	Freight volume [5]	megaton
		Freight turnover [29]	Hundreds of millions of tons of kilometers
	Intangible output	The gross product of the logistics industry [30]	Ten thousand yuan
Environment variable	Market share	Total retail sales of consumer goods [31]	Hundred million yuan
	Opening to the outside world	Actual utilization of foreign investment [32]	Ten thousand dollars
	Scientific and technological innovation level	Number of patents granted per 10,000 people [33]	pieces

provides an overview of the indicators of inputs and outputs in the logistics industry, as well as the selected environmental variables. The tangible inputs in the logistics industry include freight volume and freight turnover, which represent the physical aspects of the industry’s operations. These indicators quantify the amount of goods being transported and the efficiency of the logistics processes. On the other hand, the intangible output in the logistics industry is represented by the total output value. This variable measures the overall development and performance of the logistics industry within a specific region. It considers various factors such as revenue, profitability, and economic contribution. Additionally, the table includes three selected ecological variables that are considered to impact the logistics industry’s management efficiency. These variables are (I) Actual Use of Foreign Investment: This variable measures the degree of foreign investment in a city or region. A higher use of foreign investment indicates greater openness to the outside world. The presence of foreign businesses brings technology, management expertise, and potential spillover effects that can promote the development of the local logistics industry. (II) Total Retail Sales of Social Consumer Goods: This variable reflects the total size and geographical distribution of a region’s consumer market and indicates economic prosperity. When people’s purchasing power is high, the scale of the logistics market for commodity circulation tends to increase correspondingly. (III) Number of Approved Patent Applications: This variable indicates a city’s capacity for technical and scientific advancement. More approved patent applications suggest a higher level of scientific and technological innovation. This innovation can lead to optimizing and upgrading the industrial structure in the logistics industry, promoting the development and application of new logistics equipment. By considering these indicators of inputs, outputs, and environmental variables, stakeholders in the logistics industry can assess and monitor the development and efficiency of logistics operations within a specific region. Indicators of inputs and outputs are displayed in Table 1.

3.1.2. Data Description

By manually collecting and sorting out the panel data of the Beijing-Tianjin-Hebei urban agglomeration from 2008 to 2020, measuring the development level of green logistics in the Beijing-Tianjin-Hebei urban agglomeration, and obtaining the efficiency value of green and low-carbon logistics, the China Statistical Yearbook, the Hebei Economic Yearbook, Statistical Bulletins from various cities, and the ESP Global Database all contribute to the body of knowledge used in this investigation. For some missing data, refer to the practice of some scholars for interpolation.

3.1.3. Concept Definition

This paper refers to the previous research. [34,35] to convert the energy consumption coefficient, carbon emission coefficient, and freight turnover data of various transportation modes into a carbon emission accounting method. The carbon emission coefficients of diesel, gasoline, and aviation kerosene, respectively, come in at 3.096, 2.925, and 3.033 when compared to one another. The carbon emission coefficient of conventional coal is equivalent to 2.4, while the carbon emission coefficient of electrical energy is equivalent to 8.726. The precise formula is as follows:

$$E_{mode}={\sum }_j{T}_{ij}{F}_{ij}{C}_j$$

(1)

3.2. Model Building

3.2.1. Three-Stage SBM-DEA Model

The three-stage SBM-DEA model is an advanced method used to evaluate the efficiency of decision-making units, particularly in the context of the logistics industry in the Beijing-Tianjin-Hebei region. This model is an extension of the traditional DEA (Data Envelopment Analysis) approach and incorporates the concepts of Super-efficiency and SBM (Slack-based Measure) models. The traditional radial DEA model can decompose technological efficiency into purely technical efficiency as well as scale efficiency. However, it does not eliminate the effects of environmental effects, random interference items, and management inefficiency items. Therefore, ref. [36] proposed a three-stage DEA model that can eliminate the effects of environmental factors, management inefficiency, and random interference on productivity, and more accurately reflect the productivity level of decision-making units. At the same time, in order to eliminate the influence of input-output variable relaxation and selection errors in the three-stage DEA, the impact of different input-output scales is excluded, making the evaluation results comparable. Ref. [37] put forward a non-radial SUPER-SBM model when calculating the efficiency value.

The first stage evaluates the initial efficiency using an input-oriented, scale-variable SBM model, which more accurately reflects the efficiency of the decision-making units and random interference on efficiency. This stage focuses on capturing the pure technical efficiency and scale efficiency of the units while considering multiple inputs and outputs, but due to the simple application of the model, it is not described in detail.

In the second stage, the input relaxation variable obtained in the first stage is taken as the explained variable, and the external environment is taken as the explanatory variable to construct the stochastic frontier SFA regression equation. In the meantime, each decision-making unit’s input relaxation variable is adjusted to exclude the influence of environmental effects, random interference terms, and management inefficiency terms on efficiency. This stage helps eliminate the influence of these factors on efficiency, allowing for a more accurate evaluation. The formula is as follows:

$${\Phi }_{ik}=F^i\left({\alpha }_m;{\beta }_n\right)+{\mu }_{ik}+{\nu }_{ik}$$

(2)

where: ${\Phi }_{ik}$ is the slack variable input by the $k$ decision-making unit of the $i$ item; ${\alpha }_m$ is the environmental variable; ${\beta }_n$ is the environmental variable coefficient; ${\mu }_{ik}+{\nu }_{ik}$ is the mixed error item; ${\nu }_{ik}$ is the random error item.

Adjusted formula:

$${\xi }_{ik}={\mathsf{{\rm Z}}}_{ik}+[Max{F}^i\left({\alpha }_m;{\beta }_n\right)-F^i\left({\alpha }_m;{\beta }_n\right)]+[Max\left({\mu }_{ik}\right)-{\nu }_{ik}]$$

(3)

In the formula: ${\xi }_{ik}$ is the input value excluding environmental and random factors; ${\mathsf{{\rm Z}}}_{ik}$ is the input amount before the adjustment; $\left[Max{F}^i\left({\alpha }_m;{\beta }_n\right)-F^i\left({\alpha }_m;{\beta }_n\right)\right]$ are the adjusted external environmental factors; $[Max\left({\mu }_{ik}\right)-{\nu }_{ik}]$ is used to optimize random variables.

In the third stage, the new input data obtained after the regression adjustment of the SFA model in the second stage replaced the initial input data, and the SUPER-SBM-DEA model was constructed again to calculate the real efficiency value. It identifies the most efficient decision-making units and uses their performance as a benchmark to assess the relative efficiency of other units. This stage helps in comparing and ranking the efficiency levels of different units in the logistics industry.

3.2.2. Malmquist Index Model

Compared with the three-stage DEA model, the Malmquist index is inconvenient for comparing the characteristics of DMU production efficiency at different time points. It can dynamically analyze the efficiency change of decision-making units at other times. At the same time, the production efficiency index can be further decomposed into the technical efficiency change index and technological change index. The model can be applied to various sectors, including the logistics industry, to assess changes in productivity and identify factors that contribute to productivity improvement or decline. By using data on input and output variables, the Malmquist index model provides a quantitative measure of productivity change and helps identify areas for improvement and benchmarking against best practices.

Its calculation formula is as follows:

$${ML}_k^{k+1}=\sqrt{\left[\frac{1+s_0^k\left(x^{k+1},y^{k+1},c^{k+1},g^{k+1}\right)}{1+s_0^k\left(x^k,y^k,c^k{,g}^k\right)}\times \frac{1+s_0^k\left(x^{k+1},y^{k+1},c^{k+1}{,g}^{k+1}\right)}{1+s_0^{k+1}\left(x^k,y^k,c^k{,g}^k\right)}\right]}$$

(4)

$${EC}_k^{k+1}=\frac{1+s_0^k\left(x^{k+1},y^{k+1},c^{k+1}{,g}^{k+1}\right)}{1+s_0^k\left(x^k,y^k,c^k{,g}^k\right)}$$

(5)

$${TC}_k^{k+1}=\sqrt{\left[\frac{1+s_0^k\left(x^{k+1},y^{k+1},c^{k+1}{,g}^{k+1}\right)}{1+s_0^{k+1}\left(x^k,y^k,c^k{,g}^k\right)}\times \frac{1+s_0^k\left(x^{k+1},y^{k+1},c^{k+1}{,g}^{k+1}\right)}{1+s_0^{k+1}\left(x^{k+1},y^{k+1},c^{k+1}{,g}^{k+1}\right)}\right]}$$

(6)

Among them: $ML$ is the change index of total factor productivity, which can observe the law of dynamic efficiency change; $EC$ reflects the evolution of technical efficiency, which indicates technical efficiency change from t period to t + 1 period; $TC$ is the change of technological progress, which mainly reflects the degree of influence of production technology progress on DMU.

4. Empirical Analysis

Based on the above index system establishment, concept definition, evaluation model setting, etc., this section uses the three-stage SBM-DEA and Malmquist index method to evaluate the calculation results related to the development level of green and low-carbon logistics in the Beijing-Tianjin-Hebei urban agglomeration in 2008–2020 and to analyze the green logistics of the Beijing-Tianjin-Hebei urban agglomeration more scientifically and comprehensively.

4.1. Static Efficiency Analysis

4.1.1. Phase 1: Initial SBM Model Evaluation

Input-Output Classification: classify the collected data into input and output variables based on their respective characteristics. Inputs are the resources or factors utilized in the logistics industry, while outputs are the results or outcomes generated by these inputs. Making use of MAXDEA to determine the value of efficiency of the SBM-DEA model with input orientation and VRS variable returns to scale in each region of the Beijing-Tianjin-Hebei urban agglomeration from 2008 to 2020, run the SBM model on the input data to calculate the efficiency scores for each area. The efficiency scores represent the relative efficiency of each region in utilizing inputs to generate outputs; analyze the efficiency scores obtained from the SBM model to identify the regions with the highest and lowest efficiency levels; see Figure 2 for the results. From the results, there are significant differences in the green logistics efficiency values among regions, and the fluctuations are large, which shows that the logistics development level of each city is not balanced. The logistics efficiency may increase or decrease over time.

To observe the differences in the distribution of efficiency values more intuitively, using GIS tools, the regional efficiency value classification in Figure 3 can be obtained. The two municipalities immediately subordinate to the central government and the southern part of Hebei enjoys the most benefit, with the average efficiency value reaching more than 0.9, because the municipalities directly under the central government have developed economies, superior infrastructure, and strong public service capacity. During this time, the southern region of Hebei places a significant emphasis on the growth of the logistics business. Logistics enterprises and transportation teams promote the development of local industrialization and become demonstration bases, continuously enhancing logistics economy carrying capacity, while the rest of Hebei is in the second echelon with average efficiency values between 0.7 and 0.8.

Further analysis of pure technical efficiency and scale efficiency in terms of time is shown in

Table 2. Green and Low-Carbon Efficiency Value of Beijing-Tianjin-Hebei Regional Logistics Industry.

Area	Index	2010	2011	2012	2013	2014	2015	2016	2017	2018	2019	2020	Average Value
Municipality	TE	0.840	0.960	1.019	1.020	0.987	0.974	0.921	0.916	0.920	0.793	0.799	0.917
	PE	0.847	0.974	1.035	1.034	1.022	0.979	0.948	0.943	1.020	0.819	0.807	0.940
	SE	0.992	0.986	0.985	0.986	0.965	0.995	0.973	0.972	0.910	0.959	0.982	0.976
Eastern Hebei Province	TE	0.697	0.789	0.801	0.801	0.713	0.671	0.647	0.707	0.730	0.789	0.777	0.768
	PE	0.758	0.833	0.851	0.998	0.770	0.736	0.713	0.792	0.775	0.826	0.811	0.838
	SE	0.908	0.936	0.931	0.819	0.916	0.904	0.906	0.893	0.939	0.947	0.950	0.912
Northern Hebei Province	TE	0.749	0.828	0.813	0.820	0.738	0.679	0.691	0.746	0.782	0.711	0.730	0.716
	PE	0.885	0.854	0.831	0.830	0.812	0.787	0.827	1.010	1.002	0.961	1.015	0.880
	SE	0.862	0.967	0.975	0.985	0.913	0.866	0.836	0.738	0.778	0.738	0.720	0.824
Central Hebei Province	TE	0.753	0.829	0.868	0.885	0.757	0.663	0.675	0.692	0.692	0.696	0.685	0.765
	PE	0.801	0.846	0.895	0.942	0.777	0.680	0.694	0.705	0.709	0.709	0.696	0.798
	SE	0.942	0.981	0.970	0.943	0.974	0.974	0.975	0.982	0.981	0.985	0.986	0.965
Southern Hebei Province	TE	0.830	0.980	1.060	0.990	0.928	1.009	0.916	0.997	1.012	0.987	0.940	0.958
	PE	0.843	0.997	1.083	0.998	0.964	1.078	0.952	1.017	1.056	1.147	0.976	0.996
	SE	0.984	0.983	0.979	0.993	0.963	0.938	0.961	0.980	0.960	0.881	0.962	0.965

, where TE and PE in 2012 and 2013 in the municipality directly under the Central Government reached the efficiency frontier surface, TE in 2017 and 2018 in Northern Hebei Province reached the efficiency frontier surface, scale efficiency in Eastern Hebei Province came the efficiency frontier surface in 2013, 2015, and 2018, and TE and PE in 2012, 2015, and 2018 in Southern Hebei Province reached the efficiency frontier surface, in addition, only pure technical efficiency reached the efficiency frontier surface in 2017 and 2019, indicating that the technology and management level of the logistics industry in the Beijing-Tianjin-Hebei city cluster is well developed overall. However, each region had a distinct degree of efficiency reduction in 2015, as China’s economic development model shifted from high-speed to high-quality development after 2015 and the logistics industry regularly modified its industrial structure. It prioritized environmental protection to usher in the future of the Internet of Things, technology, and data-driven logistics. The fact of logistics is that large-scale development of the status quo is required to generate conflict, resulting in decreased logistics efficiency.

4.1.2. Second Stage: Use SFA to Analyze the First Stage Relaxation Variables

Using carbon emissions, total wages of logistics employees, and fixed assets of the logistics industry as the explained variables and foreign investment, retail sales of all consumer products, and patent authorization as the explanatory variables, this paper selects the relaxation variables of panel data input in the first stage of the Beijing-Tianjin-Hebei region from 2008 to 2020 and builds the SFA regression model. The regression results are shown in

Table 3. SFA regression results.

Variable	Coefficient	Actual Use of Foreign Investment	The Total Retail Sales of Social Consumer Goods	Number of Patents Granted	${\mathit{\sigma }}^2$	$\mathbf{Log}\mathit{V}\mathit{a}\mathit{l}\mathit{u}\mathit{e}$	LR
carbon emission	1.452 ***	−1.91 × 10−6	−0.0048 ***	0.00036 ***	1.59 × 102 *	−547.25	50.90
	1.25	−0.938	−4.269	5.029	1.703
Total logistics salary	253,154.39 ***	−0.447 *	−373.4 ***	52.477 ***	9.86999 × 1011 ***	−2542.9	23.99
	2.51 × 105	−1.905	−2.841	3.880	9.86999 × 1011
Logistics fixed asset investment	23.9960 ***	−0.003 **	0.939 *	−0.08	1.73 × 108 ***	−1749.3	139.7
	1.78 × 103	−2.066	1.898	−1.40	1.73 × 108

Note: The lower row is the t value; *, **, and *** represent the significance levels of 10%, 5%, and 1% respectively.

. It can be seen that: (I) The return of different input variables is not consistent; the carbon emissions, the total salary of logistics employees, and the LR value of logistics fixed asset investment are all more remarkable than the inspection threshold, indicating that the model setting is practical; (II) The amount of foreign investment actually utilized is negatively correlated with the three input relaxation variables, and it is negatively correlated with the significant tests of 10% and 5% of the total salary of logistics employees and the fixed asset investment in logistics, respectively, which indicates that the increase in foreign investment is conducive to reducing the input relaxation variables, and the absorption of foreign investment not only saves the resource consumption of the invested areas but also is conducive to the development of low-carbon logistics efficiency and promotes the goal of high-quality economic development; (III) the total retail sales of consumer goods and the three input variables all pass the significance test of different degrees, and the regression coefficient of carbon emission and total logistics wages is negative, which indicates that the increase in residents’ consumption will result in a growth in both the workforce and the amount of capital invested in the logistics industry consume the logistics salary quota, and is not conducive to the improvement of low-carbon logistics efficiency; the increase in social consumer goods accompanied by the increase in logistics fixed asset investment will promote the development of logistics infrastructure and increase freight volume and freight turnover. (IV) Patent authorization, carbon emission, and total wages of logistics have all passed the 10% significance test, showing a positive correlation between carbon emission and total wages of logistics, indicating that the improvement of science and the technology will obviously improve the quality of the labor force and then improve the salary level of employees, which will have a positive impact on the efficiency of low-carbon logistics; however, the impact on fixed asset investment in logistics is not apparent.

4.1.3. The Third Stage: Adjusted SBM

After analyzing the findings of the second regression stage and reintroducing the modified input variables and the actual output into the SBM-DEA model, we can calculate the green and low-carbon efficiency values of the Beijing-Tianjin-Hebei city logistics industry. This is accomplished by removing the impact of environmental variables and random errors from the equation. Compared with the efficiency values obtained in the first stage, the change trend results are shown in Table 3.

After a comparative analysis of efficiency trends and comprehensive technical efficiency rankings, it can be seen that:

The levels of productivity in thirteen prefecture-level cities in the Beijing-Tianjin-Hebei region have undergone significant shifts in recent years. The adjusted efficiency values have increased, indicating that environmental factors positively impact efficiency; conversely, they have a negative effect.

Table 4. Green and low-carbon logistics efficiency of the Beijing-Tianjin-Hebei logistics industry in the first and third stages (2008–2020).

	Comprehensive Technical Efficiency			Pure Technical Efficiency			Scale Efficiency
Area	A	B	Trend	A	B	Trend	A	B	Trend
Beijing	0.9435	0.9432	↓	0.951	0.992	↑	0.992	0.949	↓
Tianjin	0.891	0.924	↑	0.929	0.965	↑	0.959	0.958	↓
Qinhuangdao	0.614	0.341	↓	0.713	0.992	↑	0.860	0.344	↓
Tangshan	0.923	0.917	↑	0.963	0.999	↑	0.964	0.920	↓
Zhangjiakou	0.657	0.411	↓	0.766	0.991	↑	0.868	0.414	↓
Chengde	0.775	0.297	↓	0.994	1.002	↑	0.781	0.297	↓
Shijiazhuang City	0.892	0.889	↓	0.930	0.989	↑	0.963	0.899	↓
Baoding	0.737	0.641	↓	0.745	0.960	↑	0.991	0.664	↓
Hengshui	0.721	0.343	↓	0.824	0.994	↑	0.889	0.345	↓
Langfang	0.744	0.428	↓	0.747	0.964	↑	0.996	0.445	↓
Cangzhou	0.732	0.679	↓	0.743	0.949	↑	0.986	0.715	↓
Handan	0.970	0.897	↓	1.025	1.019	↓	0.953	0.878	↓
Xingtai	0.945	0.605	↓	0.967	0.999	↑	0.977	0.604	↓
average value	0.811	0.640	↓	0.869	0.986	↑	0.937	0.649	↓

Note: Efficiency values are on average for 2008–2020. A is the first-stage efficiency value, and B is the third-stage efficiency value. ↓ represents a decrease in efficiency values, ↑ opposite.

makes it clear, as can be seen here. The average comprehensive technical efficiency is 0.811, the average pure technical efficiency is 0.869, and the average scale efficiency is 0.937. When the influence of environmental variables is not considered, the overall technical efficiency value of the broad green logistics integrated technology of the Beijing-Tianjin-Hebei urban agglomeration is between 0.614 and 0.97. Pure technical efficiency and average scale efficiency precede the average comprehensive technical efficiency, indicating that interregional capacity for resource allocation and efficiency of resource usage are comparatively insufficient. The overall average technical efficiency and scale efficiency decreased by 0.17 and 0.288, respectively, after the influence of external environmental variables and random error items were removed.

On the other hand, the average pure technical efficiency increased by 0.117 while the scale efficiency dropped significantly, indicating that the level of technology management still had an advantage after the environmental variables were removed. Under the strategic layout of China’s Beijing-Tianjin-Hebei collaborative development, the siphoning effect of high-level cities such as Beijing and Tianjin was pronounced. The level of regional scientific and technological innovation and government support was strong, while the scale of the Beijing-Tianjin-Hebei logistics industry limited the efficiency level.

From Figure 4 of the adjusted space efficiency classification, the efficiency results of the third stage were more objective. Among them, the comprehensive technical efficiency of Qinhuangdao, Zhangjiakou, Chengde, Hengshui, and Langfang all fell below 0.5, indicating that the three environmental variables greatly influenced the investment relaxation variables of their complete technological efficiency. Further analysis of purely technical and scale efficiency changes is needed to explain the differences in these cities. That is, optimizing resource allocation and improving the development of scientific and technological innovation and logistics management capabilities are necessary. Furthermore, the green logistics efficiency values in Beijing, Tangshan, Shijiazhuang, Baoding, Cangzhou, Handan, and Xingtai have declined slightly. After taking out the impact of external environmental variables and random error items, the overall technical efficiency and scale efficiency both declined by 0.17 and 0.288 percentage points, respectively, and the average pure technical efficiency went up by 0.117. In contrast, the scale efficiency went down significantly, suggesting that the level of technology management still had an advantage after the environmental variables were taken out of the equation.

4.2. Dynamic Efficiency Analysis

After removing environmental variables and random mistakes using the MAXDEA program and the Malmquist index model, it is possible to deduce the dynamic variations of green logistics efficiency in Beijing, Tianjin, and Hebei. The complete results are provided in

Table 5. Dynamic change index of green and low-carbon efficiency in the Beijing-Tianjin-Hebei logistics industry from 2009 to 2020.

Area	ML	PEC	SEC	TC	EC
Beijing	1.031	1.043	1.04	0.955	1.085
Tianjin	0.993	0.979	1.024	0.998	1.002
Qinhuangdao	1.037	1.001	0.992	1.047	0.993
Tangshan	1.033	1.003	1.005	1.031	1.008
Zhangjiakou	1.088	1.005	1.055	1.03	1.06
Chengde	1.023	1.002	0.995	1.027	0.997
Shijiazhuang	1.037	1.013	0.993	1.039	1.005
Baoding	1.054	1.014	1.029	1.018	1.043
Hengshui	1.069	0.999	1.033	1.037	1.032
Langfang	1.045	1.001	1.015	1.032	1.016
Cangzhou	1.032	1.006	1.066	0.976	1.072
Handan	1.024	1	1.038	0.997	1.038
Xingtai	1.085	1.012	1.052	1.026	1.065
average value	1.042	1.006	1.026	1.016	1.032

Note: Each efficiency value is the average value from 2009 to 2020. EC is the technical efficiency change index; TC is the index of technological progress change; PEC is the pure technical efficiency change index; SEC is the scale efficiency change index; ML is the change index of total factor productivity, and the relationship of the above change indexes is as follows: ML = TC × EC = TC × PEC × SEC.

.

When the change index value of total factor productivity is greater than 1, it indicates that production efficiency is at the top level; otherwise, it falls behind. As demonstrated by Table 5: (I) in general, the average value of the total factor productivity change index of the green logistics industry in the Beijing-Tianjin-Hebei region is 1.04, the average value of the pure technical efficiency change index is 1.006, the average value of scale efficiency is 1.026, the average value of the technological progress change index is 1.016. The average value of the technical efficiency change index is 1.032, all of which is greater than 1. It shows that, with the support of the coordinated development strategy of Beijing, Tianjin, and Hebei, the regional logistics production efficiency is gradually reaching the perfect stage. (II) In particular, the city of Tianjin has a total factor productivity that is lower than 1. The growth rate is the slowest, indicating that the development efficiency of green logistics in Tianjin is declining. The resource utilization efficiency of labor and capital needs to be improved, which is related to the decline of logistics competitiveness caused by the sluggish rate of economic development that Tianjin has been experiencing in recent years. (III) The mean value of the pure technical efficiency change index is less than 1, while the mean value of other efficiencies is greater than 1, indicating that Tianjin and Heng Shui should increase their investment in technological innovation; The mean value of the scale efficiency change index is less than 1 in Qinhuangdao City, Chengde City, and Shijiazhuang City, so the current level of logistics allocation should be improved. (IV) The amount of technological advancement is a significant contributor to inefficient logistics. This is evidenced by the fact that in Beijing, Tianjin, Cang zhou, and Handan, the mean value of the technological progress change index is less than one, and in Qinhuangdao City and Chengde City, the mean value of the technical efficiency change index is less than 1.

As can be seen from Figure 5: (I) the change index of total factor productivity of the logistics industry in 2014, 2015, and 2020 is less than the efficiency frontier, indicating that the green and low-carbon efficiency of the logistics industry in the Beijing-Tianjin-Hebei urban agglomeration is in a declining state in recent years, which reflects that the long-term rapid development has lost the carrying capacity of essential logistics services. The change in the public environment will affect the low-carbon efficiency of the logistics industry. (II) The fluctuation range of the technological progress change index was prominent in 2014 and 2019, which was highly consistent with the changing trend of the total factor productivity change index, indicating that technological progress change mainly affected the change of total factor productivity when the change of technical efficiency was in a relatively stable state; that is, technological progress could reduce the carbon consumption of logistics and improve the efficiency of logistics development. (III) From 2008 to 2020, the green and low-carbon total factor productivity of the Beijing-Tianjin-Hebei logistics industry has generally declined. This is due to the heavy industry-based industrial structure in the region for many years, and the level of pollution emissions is higher than the government’s environmental regulations.

4.3. Discussion

Carbon neutrality and carbon peak construction play an essential role in the sustainable development of the logistics industry in China. Studying the green efficiency of the logistics industry helps grasp the reasons for the high cost and low efficiency of the current logistics industry. The shortcomings of this paper are as follows: First, due to the difficulty in obtaining data, the selection of indicators combined with previous research results, and only collecting 12 years of relevant data in the Beijing-Tianjin-Hebei region. Secondly, in constructing the index system of green and low-carbon influencing factors in the logistics industry, only qualitative descriptions have been carried out, and no quantitative research has been carried out. Nevertheless, it is undeniable that the dynamic and static evaluation of the green and low-carbon efficiency of the logistics industry in the Beijing-Tianjin-Hebei urban agglomeration can provide a specific scientific basis and empirical reference for global green and low-carbon development businesses to a certain extent and promote regional logistics development. The study has important practical implications and application value for the green and low-carbon efficiency of logistics in the urban agglomerations of Beijing, Tianjin, and Hebei. Specifically, the results of the study can provide important references for government departments, enterprises, and relevant stakeholders to formulate and improve policies and strategies for green and low-carbon development in logistics. By assessing the green and low-carbon efficiency of logistics, potential room for improvement and problems can be revealed, and guidance can be provided to promote sustainable development and high-quality synergistic development. In addition, the study provides useful lessons for other city clusters or regions to promote the advancement of green and low-carbon development in logistics nationwide.

Based on the current study, several future research directions deserve further exploration. The following are a few possible directions:

Spatial analysis based on the GIS system: Further utilize the functions of the GIS system to carry out a spatial analysis of logistics green and low-carbon efficiency and explore the spatial differences and potential optimization paths within city clusters and between different cities.

Comparative analysis: It would be valuable to compare the green and low-carbon efficiency of the logistics industry in the Beijing-Tianjin-Hebei region with other urban agglomerations or regions in China. This can help identify regional differences and best practices for promoting green and low-carbon development in logistics.

Cross-disciplinary cooperation and innovation: further strengthen cooperation among academia, government, enterprises, and social organizations to promote exchange and innovation, and jointly promote research and practice on green and low-carbon development in logistics.

By addressing these future research directions, a more comprehensive understanding of the green and low-carbon efficiency of the logistics industry in China can be achieved. This will contribute to the sustainable development of the logistics industry and support the country’s efforts toward carbon neutrality and carbon peak construction.

5. Conclusions

As my country puts forward the “double carbon” goal, the logistics industry faces the urgent task and historic opportunity to accelerate the green and low-carbon transformation. Based on China’s green and low-carbon city construction needs and the integrated development of urban agglomeration logistics. It constructs the input indicators of total wages of logistics employees, fixed assets of the logistics industry, and carbon emissions, output indicators based on freight volume and freight turnover, and environmental variables based on total retail sales of social consumer goods, actual foreign investment, and patent authorization per 10,000 people, using the three-stage SBM-Malmqusit model to analyze the Beijing-Tianjin-Hebei cities. Overall, the three-stage SBM-DEA model provides a comprehensive evaluation of efficiency by considering various factors and eliminating their effects. It offers insights into the development status of the logistics industry in the Beijing-Tianjin-Hebei region and helps in identifying areas for improvement and optimization. The measurement and in-depth analysis of the green and low-carbon efficiency of the logistics industry found that:

(I) The growth of the logistics business as well as its industrial structure will have an impact on the development of low-carbon and environmentally friendly communities. The logistics industry’s energy usage and carbon emissions will impede the growth of the urban green economy. At the same time, technological development’s backward level is the main reason for the inefficiency of logistics in some cities and is also essential for establishing quality urban development.

(II) Due to the impact of random errors and environmental variables, the first-stage measurement will exaggerate the comprehensive technical efficiency. After adjustment, the extensive technical efficiency value will decrease, indicating that optimizing resource allocation between regions can reduce economic development differences caused by different environments. At the same time, the level of total factor productivity is relatively high, which produces positive external effects.

(III) Under the strategic background of the coordinated development of Beijing, Tianjin, and Hebei, the overall green and low-carbon efficiency value of the logistics industry is relatively high, indicating that the green and low-carbon development of the logistics industry is gradually making progress at this stage. To a certain extent, high-level cities such as Beijing and Tianjin have apparent siphon effects, leading regional technological innovation levels, and strong government support. In contrast, the scale of the logistics industry will have a rebound effect, which will restrict efficiency improvement to a certain extent.

(IV) Absorbing foreign investment can save resource consumption in the input area, which is conducive to developing low-carbon logistics efficiency and promoting high-quality economic development. The increase in resident consumption will increase the labor force and capital investment in the logistics industry and consume the logistics wage quota, which is not conducive to improving low-carbon logistics efficiency. The increase in social consumer goods, accompanied by the increase in logistics fixed asset investment, will promote the development of logistics infrastructure and increase freight volume and turnover. Improving the level of science and technology will significantly improve the quality of the labor force, thereby increasing the wage level of employees and positively impacting the efficiency of low-carbon logistics. However, the effect on the investment in logistics fixed assets is not apparent.

Based on these findings, to accelerate the green and low-carbon transformation of the development mode of the logistics industry and form a sustainable development model, the green, low-carbon, and high-quality development of the logistics industry in the Beijing-Tianjin-Hebei urban agglomeration can refer to the following policy recommendations:

(I) Taking into account the concept of industrial synergy, logistics companies must make strategic choices in terms of upstream and downstream industries that align with their operational processes. Prioritizing establishing economic development goals, eliminating outdated capacity, and leveraging industrial agglomeration will enhance the capacity of green and low-carbon logistics. Policy guidance should be emphasized to foster the growth of leading enterprises, enhance their resilience, strengthen alliances to invigorate the market, and promote strategic cooperation within the Beijing-Tianjin-Hebei region. These efforts will drive the industry’s adaptation to green and sustainable development.

(II) From the perspective of government macroeconomic regulation, it is essential to refine the carbon emissions trading system and intensify environmental supervision to incentivize market participants to reduce emissions. Clear guidelines should be established to outline the developmental trajectory of the logistics industry, gradually extending its reach to encompass a wider range of industries and sectors. Special attention should be given to fostering a green environment by planning and constructing logistics parks, improving the logistics infrastructure network, and promoting the development of multimodal transportation. Establishing effective rules and service standards for multimodal transportation will facilitate its market-oriented growth and contribute to structural reforms in freight supply. Additionally, expediting the establishment of a unified national carbon market is of utmost importance.

(III) Considering the aspect of green and low-carbon technological innovation, it is imperative to drive the development of the logistics industry in an environmentally friendly and sustainable manner. This involves the development of resource-efficient, energy-saving, environmentally friendly, and low-carbon materials. Significant emphasis should be placed on implementing green and intelligent solutions in logistics equipment and facilities. Adopting electrification technologies for logistics engineering equipment and exploring carbon capture and storage techniques can contribute to the industry’s low-carbon objectives. Implementing cloud-based management and development initiatives will provide comprehensive support and protection for regional logistics operations.