Exploring the Role of Digital Transformation and Breakthrough Innovation in Enhanced Performance of Energy Enterprises: Fresh Evidence for Achieving Sustainable Development Goals

Abstract

The energy sector’s digital evolution is a critical micro-reflection of the digital economy’s architecture and an essential tactical pathway for achieving sustainable development goals. However, the value of digital change in regard to how effectively energy firms’ core business functions is not yet apparent. This research utilizes textual analysis to quantify the textual intensity of energy enterprises’ digitization. Applying data gathered from A-share listed firms in Shanghai and Shenzhen from 2010 to 2021 and based upon the fixed-effects panel model and mediated effects model, we assess the impact of digitization on critical business performance and evaluate the research themes’ variability from business and geographical viewpoints using a full-sample strategy. We derive three notable findings. First, the digital development of energy enterprises significantly improves the performance of their core businesses and exhibits some fluctuating characteristics. Second, the digitization of energy enterprises greatly increases the input and output of enterprise innovation, thereby improving the efficiency of their core business in the two main ways of breakthrough innovation. Third, there is a nonlinear relationship between the performance of energy enterprises’ core businesses and their digital transformation, meaning there is just one threshold consequence that diminishes after crossing the threshold. Digital transformation enables energy companies to carry out cross-border cooperation and integrate innovative resources, thereby improving corporate performance and promoting sustainable development. This paper offers relevant insights for more effective fostering of digital transformation and can help spur enterprises to seek out distinctive and ground-breaking innovation models.

1. Introduction

Global energy faces multiple challenges from the economic recession, accelerating climate change, and increasing structural social injustice. If new methods can be found to reduce, recycle, and reuse carbon emissions, such as increasing access to clean, reliable, and affordable energy, then the world will have more choices and opportunities for sustainable development. According to data from the World Energy Council, China is both the world’s biggest energy user and the largest renewable energy power-generating country. The nation’s recovery from the crisis is crucial for optimizing the global energy structure due to the particularly important performance of its energy sector. At the same time, China’s energy sector is making serious efforts to reduce greenhouse gas emissions, especially carbon emissions, as a component of the global economy’s transition to green progress and to ultimately achieve climate neutrality. As the main body of its domestic energy industry, energy enterprises bear the heavy responsibility of achieving the sustainable development of China’s energy structure through transformation.

Digital transformation has been carried out in various fields around the world [1]. Against the backdrop of the digital tsunami, China will inevitably play an increasingly important role in this process. Digital transformation promotes enterprise development and is an important component of enterprise transformation and upgrading [2]. It is indeed crucial to utilize all the potential provided by modern technology to establish a transformation strategy that adapts to its endowment characteristics through digital transformation [3,4]. Among the methods to accomplish breakthrough innovation is digital transformation. Breakthrough innovations that are tailored to the characteristics of energy companies may assist in bringing about new changes, broaden their experience, motivate high-level social development, aid in the formation of organizational and individual values, and improve the economic competitiveness of energy companies [5]. Based on the survey, businesses that implement revolutionary inventions are better equipped to leverage innovation resources, which opens up more financial avenues for the enterprise to thrive [6].

The concept of low-carbon green development has come to the fore with green innovation emphasizing the production of renewable energy and green management of the industrial chain [7]. The relationship between energy innovation and environmental performance has been extensively studied by academics both domestically and internationally in recent years. The majority of these studies have viewed technological innovation as a valuable resource, particularly one that can assist with lower carbon emissions and enhance the environmental performance of businesses [8]. Putting transformation into practice to increase value is a key issue for effective growth in the digital economy, but most enterprises’ digitization at the micro-level only develops to the surface level. According to statistics, only 16% of Chinese enterprises have made significant progress in digital transformation. In fact, the new concept of digital transformation lacks penetration at all levels, angles, and characteristics of enterprise change. Technological innovation has the potential to greatly improve the environmental performance of energy enterprises and present appealing opportunities for the low-carbon growth of traditional energy enterprises by speeding up the transformation of energy production methods and increasing the efficiency with which resources are utilized [9]. Therefore, using technology as a carrier to promote the structural reform of energy enterprises and to encourage low-carbon and green development of energy enterprises is currently the development path for many enterprises and even the entire energy industry.

In summary, the arrival of the Internet provides enormous opportunities for the energy industry’s digital transformation, which may become the next frontier of the energy economy [10]. Digital transformation has become a powerful measure to effectively drive the accelerated development of energy companies. So, what specific impact will digital transformation have on the performance of energy companies? What is the impact mechanism and action path of digital transformation on energy enterprise performance? At present, there is little research on this issue at home and abroad. The purpose of this study is to examine how the digital revolution affects the performance of energy companies’ core businesses. Based on the above research background and research questions, this paper takes A-share listed energy companies from 2010 to 2021 as the research object to investigate the relationship between digital transformation, breakthrough innovation, and enterprise performance, the specific impact mechanism, and the action context.

There are three contributions to the literature herein. First, this paper strengthens and expands the discussion around this issue by utilizing big data and annual reports of listed companies. Through data text recognition, the total number of keyword frequencies for digital transformation is summarized and used as an important benchmark for evaluating the digitalization and financial strength of energy companies. The term digital transformation of energy companies is used to describe the survey. Compared to other business performance studies, our research on the main corporate outcomes of energy companies affected by key processes of digital transformation is greatly relevant and feasible. Second, this study establishes a research framework of benchmark heterogeneity testing mechanism analysis to illustrate the economic effects of energy enterprises’ digital transformation. Third, this study examines the nonlinear threshold relationship between the digital revitalization of energy companies and their main business performance. The research findings offer guidance for the government and the entire industry to promote a sustainable development path for the digital development of energy companies.

2. Literature Review

The relevant research in the academic community mainly focuses on the meaning of digitalization, the impact of digital change on organizational performance, and the path of digital transformation to improve enterprise performance. Below, we present its various forms.

(1)

Connotation of digital transformation. There is no consensus in the academic community on the meaning of digital transformation. At the digital level, Fitzgerald et al. [11] explained it as the use of digital technology (such as social media, embedded devices, etc.) for significant business transformation, which needs to be carried out as a means to maintain a company’s business model, generally referred to as critical business functions, such as improving user experience, simplifying operational models, and creating new business models. Reis [12] noted that digital transformation is the use of new digital technologies to achieve significant business transformation, affecting all aspects of user life. Mergel et al. [13] defined digital transformation as the need to utilize new technologies to provide products and services online and offline in the Internet age so as to maintain competitiveness.

The value that a company creates for itself and its stakeholders via the application of strategic management is known as the full sample strategic value. At the transformation level, many scholars have stated that this is a dynamic transformation process, attempting to integrate digital technology into the company to promote operational and business strategy adjustments while improving management and output. Chew et al. [14] argued that the term digital transformation or digitization refers to social changes brought about by technological advancements. Citrix [15] stated that the ability to implement digital transformation denotes the strategic adoption of new digital tools and technologies to improve business productivity, customer satisfaction, and employee satisfaction. Schallmo et al. [16] and Verhoef et al. [17] concluded that digital transformation is the use of digital technology to analyze and organize collected data, transform it into actionable information, and use it for evaluation, decision making, and the development of new digital business models, helping enterprises create value and improve performance and influence. Digital transformation signifies a strategic shift that builds upon digital transformation and upgrading. It involves the flexible use of knowledge and information as building blocks for production, which subsequently impacts the company’s fundamental operations and ultimately leads to the creation of a new business model [18]. In short, digital transformation can be understood as Enterprise + Technology + Data, which has the characteristics of model innovation, value creation, and new economic forms. In order to spur growth in performance, energy businesses are embracing digital change at an exponential rate and using new digital technologies and business models.

(2)

How digital transformation affects the performance of enterprises. There is scant research on whether digital transformation can improve corporate performance, and the conclusions are also inconsistent. Studies have concluded that the application of traditional digital technologies has no significant impact on business performance [19]. Vial [20] suggested that companies use digital technology to realize their value and to change and advance their operations, exuding both positive and negative impacts on company change. By reducing costs and improving operational efficiency, digital transformation can improve business performance and enhance innovation [21,22,23]. Some scholars have also questioned the idea that digital transformation can directly help improve business performance [24]. Buttice et al. [25] found that if there is a forgery issue, then digital technology can reduce the economic efficiency of enterprises. Awan et al. [26] argued that enterprises’ digital transformation is gradually evolving into a strategic decision that affects their environmental awareness and sustainable growth.

Some scholars held positive views on this issue through qualitative and quantitative analyses [27,28,29]. A few have pointed out that the higher the quality of digital transformation is, the greater the productivity of an organization [30,31]. The procedure of digital transformation encounters the potential to lower inter-firm communication costs, strengthen innovation network connectivity, quicken the pace of digital convergence, widen the scope of convergence, raise the need for heterogeneity and aggregation of knowledge, and shift how knowledge is created and shared within innovation networks [32]. Digital technology improves the quality of enterprises’ products and services, utilizes big data technology to analyze personalized user needs, reshapes the value creation mechanism of stakeholders in traditional business models, and expands the breadth and depth of enterprise users through new participants [33]. Therefore, the sole means by which digital transformation could enhance company performance is through technology advancements. In summary, although the direction of the impact of digital transformation on corporate performance is not clear enough, most studies have pointed out the promoting effect of digital transformation on corporate performance. Simultaneously, the significance of digital technology in advancing the growth of green innovation sectors and low-carbon operation and management practices in businesses has been demonstrated.

(3)

The path of digital transformation to improve enterprise performance. The literature has shown that the digital transformation of physical enterprises can improve enterprise performance by reducing costs, improving efficiency, and encouraging innovation. First, digital transformation helps reduce operational costs for enterprises. The characteristics of digital technology such as connectivity, sharing, and openness determine that enterprises can effectively reduce the adverse effects of information asymmetry among trading parties [34] and reduce the costs of information search, negotiation and signing, transaction supervision, and conversion [35]. At the same time, by integrating digital technology into business processes, enterprises can cut resource matching and channel operation costs in procurement, marketing, logistics, and other fields and even meet customer-personalized needs at extremely low costs, significantly improving the double-high problem of past costs and energy consumption [36].

Second, digital transformation helps improve the efficiency of enterprise operations. The structured and unstructured information contained in emerging digital technologies has broadened the data mining space [37], accelerated the response speed to the long-tail needs of enterprise customers, and promoted industrial specialization and collaborative operation. Numerous researchers have tested various scenarios in this regard, and their outcomes reveal that through digital transformation, businesses can fortify their capacity for innovation, boost R&D spending, optimize their supply chain structure, and improve factor allocation efficiency—all of which contribute to an advancement in energy firms’ carbon performance [38]. Such information is also conducive to improving the overall operational efficiency of enterprises. To put it briefly, when corporate innovation technology advances, energy companies’ resource allocation capacity and overall input–output factor structure will be tuned, which will boost their overall performance [39].

Third and finally, digital transformation is beneficial for promoting enterprises’ innovation and upgrading. A three-dimensional, networked, and informationized value chain is constructed through the use of digital technology in digital transformation, which stimulates the establishment of a new, dynamic network relationship between businesses and their clients, partners, suppliers, and other subjects [40]. New and old enterprises can promote the advancement of the multiplier creation effect in resources, technology, products, experiences, and customers, thus providing incremental contributions for value discovery and value creation [41]. Moretti et al. [42] pointed out that enterprises can carry out digital or intelligent upgrading and transform existing products through digital transformation, thereby promoting innovation output. Digital transformation also makes the R&D activities of enterprises repeatable and flexible. The effects of digital technology [43], digital investment [44], knowledge heterogeneity in digital enterprises [45], and the rise of digital platforms on innovation performance have all been studied by certain academics. With the help of digital technology, enterprises can add new functions in the product life cycle, promote innovation iteration based on digital technology, and greatly improve their innovation output [46,47]. Many scholars have stated that enterprises generally improve their innovation capabilities through digital transformation that enhances their performance. Consequently, the digital shift has the potential to reorganize creative production modes within businesses, enhance input–output efficiency, and subsequently propel the expansion of company performance. Furthermore, most of the research that has already been carried out on the effects of digitalization on businesses is still qualitative.

As far as energy companies are concerned, Zhao et al. [48] found that digital transformation provides large-scale information evaluation for public policies and diagnostic measures and greatly improves the efficiency of communication with people, which has a certain impact on enterprise performance. Midtun et al. [49] noted that energy companies with digital transformation capabilities to integrate operations can utilize digital technology to transform their operations, enabling better customer interaction and collaboration and thereby improving business performance. However, Peng et al. [50] pointed out that digital transformation often involves building big data analysis capabilities, which pose significant challenges for many energy companies. To summarize, a considerable body of literature has been composed on the effects of digital transformation on enterprises and the selection of innovation models. Yet, most of this literature has focused on the macro or enterprise level of digital transformation. Relatively little research has been conducted on the specific ways that digital transformation affects the performance innovation of energy enterprises, and there are also few mechanistic analyses of this interaction. Thus, even after refining to energy companies alone, existing research still cannot determine the direction and path of the impact of digital transformation on corporate performance, and further research is needed to determine it. In this research, we investigate the processes of heterogeneity by firm type and geographic location, as well as the effects of firm innovation and digitalization on the important business performance of energy enterprises.

3. Theoretical Model and Research Hypotheses

Under the digital transformation model, energy enterprises carry out breakthrough innovations to empower performance improvement from the two paths of input–output efficiency and market performance. The two paths can be subdivided into three-dimensional mechanisms of information efficiency improvement, resource structure optimization, and technology digital upgrade. In the digital upgrade of technology, there are subtle differences in the focus of traditional energy companies and clean energy companies. Therefore, this paper proposes three hypotheses.

Hypothesis 1:

The digital transformation of energy enterprises significantly promotes the improvement of enterprise performance.

Energy businesses experience a boost in information efficiency after going digital, which helps the primary business function better. Under the breakthrough innovation path of input–output efficiency, the digital transformation of energy companies is helpful for the development and improvement of the ecological system. A breakthrough cooperation model for innovation that encompasses a multilevel and multidimensional cooperation–collaboration platform can carry out collaborative technology research and development, collaborative resource complementarity, and collaborative industry linkage. The market performance path may be more significantly impacted by the disclosure of the digital evolution of knowledge platforms and the betterment of enterprise resource usage [51]. These changes are important signals to attract external investors, as attractiveness increases chances for cooperation and to realize the performance of the primary business. Additionally, when external investors to an enterprise have sufficient information, the exposure effect will attract their attention, and the probability for stock trading will also be improved, thereby promoting investment and financing for energy companies to go public. This is crucial for both economic growth and company-scale expansion. The breakthrough innovation of digital transformation promotes the enhancement of the effectiveness of information disclosure across the board, upgrades the cooperation model, broadens the channels of capital injection, and injects innovative vitality into the listing and development of energy companies.

Hypothesis 2:

Breakthrough innovation is an important way for energy enterprises to improve their performance through digital transformation.

Energy organizations are receptive to optimizing resource structure and enhancing input–output efficiency to better the effectiveness of their core business following digital transformation. Under the limitation of scarce financial resources, the optimal boundary of capital utilization efficiency can be reached, so as to re-promote the output and technological innovation of enterprises. Mergers and acquisitions, technology upgrades, and other actions can significantly reduce costs and improve performance to achieve breakthrough innovation in products and markets. The specific mechanism by which these elements enhance the performance of energy enterprises is that through digital transformation, energy enterprises can effectively reduce communication costs between departments and improve the efficiency of business information circulation. The improvement of information efficiency can enable decision-makers to understand the real situation of enterprise operation in a timely manner, quickly adjust the direction and intensity of enterprise resource investment, and thus complete the optimization of enterprise resource allocation. The efficient internal processes formed from this can enable enterprises to complete the iteration of their own production technology level and product service level more accurately. From the perspective of other production factors, Acemoglu [24] believed that the traditional production model and digital technology are integrated and combined [24]. Moreover, creative technology is used to estimate, evaluate, and reorganize the constrained resources used for production, breaking through the boundary constraints of traditional factors and maximizing under limited conditions of value function output. Therefore, breakthrough innovation affects the performance of energy enterprises mainly by improving the input–output efficiency.

Hypothesis 3:

The effect of digital transformation on the performance improvement of energy enterprises exhibits heterogeneity of enterprise type and location.

Traditional energy companies are mainly labor-intensive, and compared to clean energy companies, they have a weaker digital foundation. The intelligent upgrading of production equipment helps to improve mining efficiency and control labor costs. Clean energy companies are mainly knowledge-intensive and technology-intensive. The digital update of renewable energy technology presents the biggest opportunity for their primary competitors and their company’s performance. The efficiency advantage brought by the digital upgrade brings forth more resource investment. The digital technology of clean energy enterprises is superior to traditional energy enterprises in many aspects such as efficiency, electrification, low carbonization, and interconnection, and so the impact on enterprise performance also significantly varies.

The nature and location of an energy enterprise are two potential influencing elements from the standpoint of its performance following digital alteration. When considering the nature of the businesses, we can further analyze the differences between the two types of enterprises under the superimposed influence of micro-elements and correlation effects. State-owned enterprises (SOEs) are relatively stable for downstream demand-side firms, have low market competitiveness, and pay attention to digital transformation. The opposite is true for non-state-owned energy companies, which are under great pressure from free market competition.

In terms of corporate location, the eastern developed region of China has obvious natural endowment advantages in terms of geographical location, market share, and resource acquisition that offer a good foundation. The input/output ratio of digital transformation is high, and the performance improvement rate is also relatively high. If one considers the actual situation in China, policy support gives more driving force to digital transformation. For example, the new energy demonstration city policy (NEDC), an essential economic system aiming to implement innovative development strategies and energy structure transformations in China, is crucial for solving the current plight of resource-based cities [52]. While the market for the conservation of energy services is expanding, it is clear that there is a very serious issue with uneven regional development. Due to the differences in technology, capital, and talent between the east and the west of China, the layout of enterprises in the industry is unreasonable, and it is apparent that the east as well as the west have different levels of prosperity.

In Figure 1, input–output efficiency and market performance are dual approaches for digital resurgence that energy organizations use to produce innovation, and the two are mutually restrained and interact. Assuming that the input–output efficiency is low, the growth rate of the enterprise will be small, and it will enter a plateau period, making it difficult to attract rent injection. Conversely, when the corporate value and financial situation are not good, sufficient resources cannot be given to energy mining, excavation, and production chains, and even production safety technology and facility guarantees will be threatened. Therefore, the two paths can effectively avoid development hazards such as insufficient production and R&D capabilities or disconnected capital chains.

4. Methods and Data

4.1. Data Origin

This study selects A-share listed energy organizations from the Shanghai and Shenzhen stock exchanges as the initial research sample to assess the effects of digital evolution on corporate performance. This is because the A-share markets of the Shanghai and Shenzhen stock exchanges are the most common stock markets in China, while other exchanges have not yet matured due to their short establishment time. This research selects the initial sample time as 2010–2021 to assure consistency of the financial data caliber in light of the 2007 implementation of new accounting standards in China. On this basis, this paper deals with the data as follows: (1) remove companies with special financial data of ST and * ST during the sample period; (2) avoid the IPO effect, and remove the sample of public offerings; and (3) keep those samples that for at least five years in a row do not have any missing data. The reason for deleting ST/ST * company and IPO is that both of these situations can be considered as non-recurring phenomena in the operation of the enterprise, and the data during this period cannot reflect the normal operating status of the energy enterprise. The deletion of incomplete data entries is to ensure horizontal comparability among enterprises. Original information comes from the China Stock Market & Accounting Research (CSMAR) Database. Pertinent corporate annual report information is from the websites of the Shanghai and Shenzhen stock exchanges.

4.2. Variable Setting

4.2.1. Interpreted Variable

Energy enterprise performance (EEP) is the interpreted variable. EEP is measured by return on assets after subtracting financial income—that is, EEP = (operating profit − investment income − income from changes in fair value + investment income from associates and joint ventures)/total assets.

4.2.2. Explanatory Variables

Energy companies’ digital transformation (EDT) is measured by the total number of keywords for the digital transformation of energy enterprises. The construction steps of the digital transformation lexicon in this article are as follows. First, identify a series of keywords based on the digital transformation of energy enterprises (Appendix A). These sources include the Big Data Industry Development Standard (2016–2020), China Fintech Operation Report (2019), and Fintech Development Strategy (2019–2021). Second, with the help of the Python crawler function, sort all annual reports of A-share companies traded on the Shanghai and Shenzhen marketplaces based on the acquisition of specific keywords for the digital shift. Third, use the Java PDF box library, extract all text content, and carry out the same operations as the keywords in Appendix A. Search and count the number of words to come up with the final aggregated word frequency, build an index system for digital change in energy firms, and then classify and gather the word frequency of major technical directions. The total word frequency obtained by the sum is used as a variable indicator for the digital transformation of energy enterprises.

4.2.3. Control Variables

In

Table 1. Descriptive statistics.

Variable	Obs.	Mean	Std. Dev.	Min	Max
EEP	10,956	0.3164	3.0911	−24.6816	86.8322
EDT	10,956	44.3978	79.7664	0.0000	1166.0000
SCALE	10,956	22.1921	1.4960	14.3226	28.6365
OI	10,956	21.4693	1.6152	−22.6034	28.7183
EGE	10,956	2.6964	0.5004	0.0000	4.1431
NPR	10,956	0.0446	0.2410	−4.7821	19.6743
ROE	10,956	0.0749	1.2514	−26.1794	64.0564
TDR	10,956	0.4639	0.2790	−3.2277	11.9950
TAT	10,956	0.6636	0.4966	−0.5304	9.3799

, multiple parameters for control are introduced in this paper, such as enterprise size (SCALE), enterprise operating income (OI), enterprise age (EGE), net profit ratio of total assets (NRP), return on equity (ROE), asset/liability ratio (TDR), and total asset turnover (TAT), to increase research accuracy.

Table 2. Benchmark regression of EEP and EDT.

	M (1) EEP	M (2) EEP
L.EDT	0.0103 *** (3.63)	0.0108 *** (3.75)
CVs	NO	YES
year	YES	YES
Ind	YES	YES
N	10,043	10,043
adj. R2	0.0786	0.0958

Notes: *** indicates statistical significance at the 1% levels. The corresponding t statistics are in brackets.

displays the precise variable data structure.

4.3. Model Setting

To be able to demonstrate the implications of the digital revolution on the efficiency of energy enterprises’ core business, the following theoretical model is developed herein:

$$EE{P}_{it+1}={\alpha }_1+{\beta }_{1}ED{T}_{it-1}+\sum {\beta }_{i}CVs+\sum Year+\sum Ind+\epsilon$$

(1)

Here, EEP and digital shift are the explained variables. The main explanatory factor is EDT. CVs is the control variable group. Covering the aforementioned control variables, there is a random error item $\epsilon$. This study delays the processing of EDT by one period since the effect of the digitization of energy companies on the efficiency of the primary business has a certain time delay. This not only takes into account the time-consuming transmission of variables in practice but also effectively avoids reverse causality, creating an endogenous interference problem.

5. Empirical Findings and Economic Analysis

5.1. Benchmark Regression of EEP and EDT

The major test results of the correlation between the digital revolution and EEP are presented in Table 2. The regression coefficient of the energy enterprise digital transformation index (L.EDT), which passes the test at the 1% significance level, is 0.0103 in model M (1), which only controls time and industry fixed effects. In model M (2), which introduces the control variable set on the original basis, the regression coefficient of the energy company digital change index (L.EDT) is still highly positive. This suggests that the effectiveness of energy businesses’ main operations will be dramatically enhanced with the acceleration of the digitization process, and there is a large connection between both of them. Based on the agreement with other research conclusions, this paper refines the positive association of the digital evolution of energy firms, and EEP is increasing. The relationship also effectively verifies the previous mechanism speculation and forms the foundation for the correct direction of further mechanism deconstruction and splitting.

5.2. Assessing the Impact of Digital Evolution on EEP

5.2.1. Deletion of Some Samples

Globally significant financial shocks could affect how energy companies are transforming digitally. Digital evolution encourages enterprises to actively participate in economic globalization competition. Accordingly, the interaction between the digital transformation of energy enterprises and global economic fluctuations forms a multiplier effect. For example, after being impacted by a major adverse financial event, the digital transformation process of energy companies themselves will also stagnate. The 2008 global financial crisis and the 2015 meltdown of the China stock market were two very large monetary shocks that have occurred in recent years and are related to the time series of the sample data used in this study. However, as far as the existing research situation in academia is concerned, it is not feasible to eliminate the influence of such factors using variable construction, and it is extremely difficult to achieve the expected effect. Thus, it has become a risk factor for model construction.

This study tests the reliability of the conclusions by excluding the consequences of the global financial crisis. First, the sample of businesses from 2010 is excluded, and the samples from 2011 to 2021 are retained for additional examination in light of post-crisis characteristics. Second, we keep samples from 2011 to 2014 and 2016 to 2021 for separate testing. The statistical regression results in

Table 3. Test of robustness: deletion of some samples.

	M (1) EP	M (2) EP
L.EDT	0.0061 *** (3.64)	0.0073 *** (3.67)
	Exclude international finance crisis impact	Exclude the impact of the international financial crisis + China’s stock market crash
CVs	YES	YES
year	YES	YES
Ind	YES	YES
N	9130	7304
adj. R2	0.0770	0.0934

Notes: *** indicates statistical significance at the 1% levels. The corresponding t statistics are in brackets.

reveal that, after the two methods of treatment, the digital alteration of energy firms considerably enhances EEP, validating the main finding of the research outlined herein.

5.2.2. Extension of Observation Window

This study extends the time window for the effect of energy companies’ digital evolution on the success of their core businesses in

Table 4. Robustness test: extension of observation window.

	M (1) EEP	M (2) EEP	M (3) F 2. EEP	M (4) F 3. EEP
EDT			0.0069 *** (3.74)	0.0030 *** (3.75)
L2.EDT	0.0064 *** (3.68)
L3.EDT		0.0028 *** (3.66)
L4.EDT
CVs	YES	YES	YES	YES
Year	YES	YES	YES	YES
Ind	YES	YES	YES	YES
N	9130	8217	9130	8217
adj. R2	0.0816	0.0674	0.0869	0.0689

Notes: *** indicates statistical significance at the 1% levels. The corresponding t statistics are in brackets.

. The core explanatory variable (EDT) in models M (1) and M (2) is delayed by two to three periods, while in models M (3) and M (4), the explained variable (EEP) undergoes two to three phases of pretreatment. No matter whether EDT is lagged or EEP is preprocessed, the results demonstrate that the regression coefficients of the digital evolution of energy firms are all favorable and pass the threshold test at the 1% significance level. This demonstrates that EEP is impacted by the digitization of energy businesses. There is a very strong beneficial promotion effect, and it does not considerably decline with the lengthening of the time window. The initial finding of this work is further reinforced by the observation that EEP can be positively impacted by the digital transformation of energy firms over a lengthy time horizon with overlaid characteristics, thereby stimulating the performance growth of energy companies to some extent. It is incredibly stable, and an energy firm’s digital transformations can help them grow and perform better in their core businesses. This result is also a breakthrough and an addition to the ongoing research and investigation in the time dimension.

5.2.3. Quantile Test

As the process of energy businesses’ digital transformation advances, there may be obvious gaps in the behavior of energy companies under different performances. For this reason, this research further disassembles and analyzes the performance of energy companies at the quantile level and brings them into the model, respectively. The outcomes are displayed in Figure 2. The test results suggest that depending on where in the distribution of circumstances an energy company is located, its digital change has a variable impact on the efficiency of its key business. However, the fitting lines of the effect intensity coefficients under most digital transformation levels are significantly higher than the horizontal lines, showing that although the performance intensity of energy companies has changed, an energy company’s transformation to digital has consistently had a favorable impact on it. The basic conclusion that energy companies’ digital shift can help improve their performance remains unchanged. In this regard, under the differentiated intensity achieved under different operating conditions, it has been proven once more that energy companies’ digital transformations have very high robustness in terms of enhancing the performance of their core businesses. From this, it can be extended in the theoretical analysis part that the efficiency of innovation output can increase to some extent by innovation investment in energy firms’ digital transformation. The factors affecting the conversion rate of input and output and the intensity of transformation will be discussed later.

5.2.4. Winsorize

The figures on the digital shift of energy companies that are reported in this part are shrink-tailed to 1% and 5%, respectively—that is, the first and last 1% and 5% of outliers are removed, respectively, and the values beyond the specified range are replaced with the percentiles, respectively. This is implemented to ensure that the data are more stable. Regression tests are carried out using the data after the winsorize treatment.

Table 5. Robustness check: shrinking treatment.

	M (1) Indent 1%	M (2) Indent 5%
L.EDT	0.0051 *** (5.58)	0.0001 * (1.77)
CVs	YES	YES
year	YES	YES
Ind	YES	YES
N	10,043	10,043
adj. R2	0.1318	0.6432

Notes: *** and * indicate statistical significance at the 1% and 10% levels, respectively. The corresponding t statistics are in brackets.

presents the effect of the digital evolution of energy corporations on the success of their core businesses. Models M (1) and M (2), respectively, deal with 1% and 5% reductions in the tail of L. EDT. The performance of energy firms is unaffected by the influence of their digital change, but the absolute value and significance of each regression coefficient after the winsorize treatment decrease with an increase in the winsorize treatment percentage when compared to the results of the benchmark regression. This demonstrates the reliability of the benchmark regression results from this study and offers fresh proof in support of the main finding.

5.3. Role of Digital Shifts on EEP: Heterogeneity Analysis

Based on the viewpoints of the full sample, we investigate how the digital shift of energy companies affects EEP, and the impact effect between the two is confirmed by numerous robustness tests. However, it is worth noting that under different corporate attributes and regional differences, the transfer of power from businesses’ digital evolution to the efficacy of their core business may have asymmetric consequences. Analysis of this problem will help form differentiated policy orientations to refine the factors affecting different corporate endowments and to improve the accuracy of the mechanism path analysis and the implementation effectiveness of the solution suggestions for the change in energy operations. As a result, this study runs a subsample test on the entire sample based on geographical and property rights traits. The results are presented in

Table 6. EEP and digital transformation: heterogeneity test of enterprise types.

	M (1) EP	M (2) EP
L.EDT	0.0022 * (1.78)	0.0130 *** (3.49)
Division basis	State-owned energy companies	Non-state-owned energy companies
CVs	YES	YES
year	YES	YES
Ind	YES	YES
N	3267	6776
adj. R2	0.0935	0.1273

Notes: *** and * indicate statistical significance at the 1% and 10% levels, respectively. The corresponding t statistics are in brackets.

and

Table 7. EEP and digital transformation: regional heterogeneity test.

	M (1) EP	M (2) EP	M (3) EP	M (4) EP
L.EDT	0.0135 *** (3.47)	−0.0001 (−0.13)	0.0063 (1.56)	0.0022 (1.63)
Division basis	east	northeast	middle part	west
CVs	YES	YES	YES	YES
year	YES	YES	YES	YES
Ind	YES	YES	YES	YES
N	6776	484	1485	1298
adj. R2	0.1226	0.8574	0.0599	0.1136

Notes: *** indicates statistical significance at the 1% levels. The corresponding t statistics are in brackets.

.

The influence of the major business performance fluctuation and change effect brought on by the digital empowerment of state-controlled energy companies and privately owned energy providers is further broken down for analysis. The empirical results display that, in the group of government-owned energy corporations, the regression coefficient of the digital alteration of energy firms is 0.0022, which passes the statistical significance test at the 1% level. In contrast, the regression coefficient of digital change in energy enterprises in the group of non-state-owned energy firms is 0.0130 and passes the 1% significance level. This leads to the observation that government-run and non-state-owned energy companies’ digitization could drastically enhance the efficiency of their core business. This finding is in line with earlier theoretical analyses’ observations. The development of both types of organizations has been positively impacted by digital transformation throughout the study period.

Based on the differences in the primary data of the two types of Chinese enterprises, it can be concluded that the digital resurgence of non-state-owned energy operations has a noticeably higher promotion influence over efficiency than that of the state energy industries. The performance of the primary business is impacted by the contract energy arrangement between SOEs and non-SOEs at the macro-level corporate structure. The underlying reason is mainly the platform support guarantee given by the type of enterprise, which leads to a lack of endogenous motivation, resulting in weak subjective awareness of market competition. There is even a more common situation. Upstream and downstream companies all account for a relatively high proportion in the same system, and at the same time, the internal information resource interaction channels are smoother than those of external non-SOEs, resulting in information gaps in non-SOEs. Their subjective willingness to compete in the market is low, and their market competitiveness is also poor. This is reflected in another notable characteristic of SOEs—their production decision-making power is not in their hands, and they cannot make independent decisions.

The incentives for innovation and R&D are insufficient, resulting in very slow technological innovation and breakthroughs in product quality and variety, which lead to insufficient motivation for inherent production innovation. In terms of the anticipated improvement in performance under the breakthrough innovation following digital transformation, there is a difference between SOEs and privately held companies. Private energy companies have a stronger subjective willingness to engage in innovation and transformation activities and expect to optimize resource allocation structure, improve information efficiency, and achieve better profits through digital transformation. State-owned energy firms, however, frequently experience less market competition and have fewer subjective motivations to support digital transformation.

With the regional variation of the enterprise as one of the variables, the regression coefficient for the digital alteration of the east region’s energy companies is 0.0135, and it passes the threshold for the significance test at the 1% level. This points out how the digital resurgence of these firms has greatly aided the enhancement of their primary business performance. The outcomes of the analysis also validate the results of the preceding theoretical inquiry. Although the significance test fails, the digitization of firms in northeast China has a regression coefficient of −0.0001, proving that it has no appreciable inhibitory effect on performance. The central and western regions’ corporate energy transition to digital operations has little impact on performance, according to the regression coefficients, despite having shown that they are both positive and fail the test of relevance. This may be because energy providers in the eastern region already have a competitive advantage in terms of setting, market share, and obtaining resources. They have much more investment in innovation and transformation than firms in other regions, and they have taken on the leadership role in digital transformation, which accelerates notable improvement in the performance of their core business.

5.4. How Digital Transformation Affects EEP

The aforementioned study empirically tests the fundamental link between the success of the primary business and the digitization of energy companies, but it does not examine the mechanism’s “black box”. This section identifies and examines the channel mechanism that affects the problem between the two. For this reason, this paper selects two channels of “innovation input and innovation output” and “enterprise value and financial stability” for inspection. The mediation effect model is used for identification and trying to describe the mechanical path of how the performance of the main company is influenced by the digital alteration of energy organizations. Its mathematical expression is:

$$EE{P}_{it+1}={\alpha }_1+{\beta }_{1}ED{T}_{it-1}+\sum {\beta }_{i}CVs+\sum Year+\sum Ind+\epsilon$$

(2)

$$M\mathrm{e}diato{r}_{it}={\alpha }_2+{\theta }_{1}ED{T}_{it-1}+{\displaystyle \sum {\theta }_i}CVs+{\displaystyle \sum Year}+{\displaystyle \sum Ind}+\tau$$

(3)

$$EE{P}_{it+1}=\alpha +{\phi }_{1}Mediato{r}_{it}+{\phi }_{2}ED{T}_{it-1}+\sum {\phi }_{i}CVs+\sum Year+\sum Ind+\xi$$

(4)

Here, Mediator is an intermediary variable. Considering that the variable conduction of the mediation effect model has a certain time lag and to avoid potential reverse causal interference among the variables, this study preprocesses the interpreted variable EP in the first period, and the mediator variable maintains the data structure of the current period. The core explanation is that the variable EDT is treated with a lag of one period.

To characterize the input–output effect of energy firms’ digital shift on R&D creativity, the intermediary variables in this paper are innovation input (II, the ratio of enterprise R&D investment to operating income) and innovation output (IO, the number of enterprise patent applications), which can be disassembled and analyzed into three potential intermediate media. First, the digital transformation of energy companies can carry out technological digital upgrades, in particular, to how digital tools and technology are used. For example, the use of cutting-edge digital technology or intelligent industrial machinery can enable energy companies to accurately determine the location of minerals, conduct a comprehensive evaluation of geological conditions, and achieve accurate and safe unmanned mining of minerals. Second, the digitization of operations facilitates the visualization and transparency of internal information flows. Digital training of employees is conducive to improving the efficiency of personnel training and skill output of employees. Third, the digital transformation of energy companies effectively integrates their resource allocation, thereby strengthening the emphasis on innovation and forming a good innovation ecological scene. Increasing the level of innovation essentially is more conducive to promoting EEP.

In

Table 8. Identification of pivotal business factors of digital transition in the energy sector: innovation input and innovation output.

	M (1) F. EP	M (2) II	M (3) F. EP	M (4) IO	M (4) F. EP
L.EDT	0.0067 *** (3.71)	0.0016 (1.58)	0.0066 *** (3.71)	−0.0001 (−0.49)	0.0067 *** (3.71)
II			0.0295 *** (3.74)
IO					−0.0161 * (−1.70)
Sobel test	Intermediary variable: innovation input 1.5725 ** Mechanism is effective—forward conduction			Mediator variable: innovation output 0.4743 Ineffective mechanism
CVs	YES	YES	YES	YES	YES
year	YES	YES	YES	YES	YES
Ind	YES	YES	YES	YES	YES
N	9130	10,043	9130	10,043	9130
adj. R2	0.0853	0.0778	0.0879	0.0123	0.0853

Notes: ***, ** and * indicate statistical significance at the 1%, 5% and 10% levels, respectively. The corresponding t statistics are in brackets.

of this study, the path of “digital change in energy firms—breakthrough innovation—EEP” is recognized and verified. The discoveries disclose that while the performance of energy firms significantly improves by both the digital resurgence of those firms and their innovation input, the regression coefficient of those firms’ digital revolution on innovation input is not statistically significant. Additional Sobel tests are necessary in this regard to confirm if innovation input acts as a mediating factor linking the digital shift of energy enterprises and EEP. The results of this study illustrate that there is a strong connection between EEP and the digital evolution of energy organizations, and this association is greatly facilitated by creative investment.

The Sobel test also reveals no substantial association between the success of the core business and the intermediary effect of innovation output in the digital shift of energy companies. One possible reason is that the digitalization of energy enterprises is specialized system engineering. Energy companies are currently undergoing a delayed digital transformation process with little support from data collection, interpretation, or output. Assisting energy companies to improve technology output is challenging, and EEP has not significantly improved.

5.5. Test the Threshold Effect of Digital Transformation on EEP

The last discussion of the repercussions of energy industries’ digital shift covers the efficiency of the main company. What variations exist between the outcomes of the digitization of energy firms’ core businesses concerning varying levels of innovation output and input? Is there a typical threshold effect? In this section, Hansen’s (1999) panel threshold model theory will be expanded upon, a fixed-effect panel model will be set up, and the consequences of heterogeneity in various situations will be examined [53]. The following is the mathematical formula:

$$EE{P}_{it}={\beta }_0+{\beta }_{1}ED{T}_{it}•I(q_{it}\le \gamma )+{\beta }_{2}ED{T}_{it}•I(q_{it}>\gamma )+{\beta }_c{X}_{it}+{\lambda }_i+{\epsilon }_{it}$$

(5)

Here, $q_{it}$ represents the threshold variable, which includes innovation input (II) and output (IO). $\gamma$ represents the threshold value. $I(q_{it}\le \gamma )$ represents the indicator function, which takes a value of 1 when the conditions in the brackets are satisfied and otherwise 0.

To figure out the output and input thresholds for innovation, this study sets up single, double, and triple thresholds and uses bootstrap sampling to calculate the number of thresholds.

Table 9. Threshold estimation results.

Threshold Variable	Threshold Quantity	Threshold	F Value	p Value	Critical Value
					10%	5%	1%
II	single threshold	3.0100	305.86 ***	0.0000	18.4161	25.0733	45.2827
	double threshold	5.2700	89.89 ***	0.0000	17.1170	20.6356	38.5970
	triple threshold	6.4216	13.66	0.6750	36.2598	46.7717	65.5426
IO	single threshold	1.7918	42.42 ***	0.0075	6.1657	10.3116	32.9191
	double threshold	2.1529	−1.92	1.0000	8.0076	52.5663	301.5807

Notes: *** indicates statistical significance at the 1% levels. The corresponding t statistics are in brackets.

displays the outcomes of the threshold estimation. Conclusions imply that when utilizing innovation input as the threshold variable, the single-barrier and double-boundary tests both pass at the 1% significance level, while the triple-threshold test fails the significance test. This fully demonstrates that the influence of energy company digitization on performance has a double-limit effect based on creativity input. Making use of innovation output as the criterion variable, the single threshold passes the significance test at the 1% level, but the double threshold does not. The estimated values of the innovation output threshold 1 are 1.7918, and the innovation input threshold 1 and threshold 2 are 3.0100 and 5.2700, respectively. The likelihood ratio function diagrams for the threshold variables in Figure 3 and Figure 4.

First, according to

Table 10. Threshold effect regression results.

	II Threshold		IO Threshold
L.EDT ($q_1\ge \gamma$)	0.0050 *** (8.46)	L.EDT ($q\ge \gamma$)	0.0115 *** (24.56)
L.EDT ($q_1<\gamma \le q_2$)	0.0112 *** (18.71)	L.EDT ($q<\gamma$)	0.0053 *** (5.16)
L.EDT ($q_2<\gamma$)	0.0187 *** (29.40)
CVs	YES	CVs	YES
year	YES	year	YES
Ind	YES	Ind	YES
N	10,043	N	10,043
R2	0.1270	R2	0.0964
f	132.71	f	108.14

Notes: *** indicates statistical significance at the 1% levels. The corresponding t statistics are in brackets.

, when the innovation input is lower than the first threshold value of 3.0100, the impact coefficient of digital transformation of energy enterprises is 0.0050 and passes the significance test at 1% level. When the innovation input is more than the first criterion (3.0100) but less than the second threshold (5.2700), the impact coefficient of digital transformation of energy enterprises is 0.0112 and passes the significance test at 1% level. When the innovation input is higher than the second threshold value of 5.2700, the impact of digital transformation of energy enterprises is significant at 1% level and the coefficient is 0.0187.

This demonstrates that the positive effect of the digital evolution of energy companies on the performance of their core businesses is not static, but rather, as innovation investment gradually rises, those energy companies with higher levels of investment in innovation are encouraged by the digital transformation, and the performance of their core businesses is enhanced. The reduction impact that energy firms may gain directly or indirectly grows more and more visible with the continuous expansion of innovation investment, and so the encouraging impact of digitization on the company’s efficiency has steadily increased.

Second, when innovation output is greater than the 0.7918 initial threshold value, the regression coefficient of the digital transformation of energy companies drops to 0.0053, or a decrease of 53.9% from the first threshold. When innovation output is less than the 0.7918 original threshold value, the regression coefficient of the digital transformation of energy enterprises at the 1% significance level is 0.0115. One possible reason is that with the continuous increase in innovative production, the risk of sinking money into research and development for energy companies shows a rapid rise. The subjective willingness of energy companies with high innovation output to digital transformation weakens, and the process gradually slows down, thus affecting their performance. The positive promotion effect changes from strong to weak. At this time, energy companies need to reduce R&D output moderately.

6. Discussion and Limitations

Most studies in the literature have reached a consensus that digital shifts may boost businesses’ creativity and foster the advancement of fresh digital technologies. Compared with the results in other papers, this research finds some similarities and differences.

First, the influence of the digital shift on the success of a company’s primary business, particularly in the energy industry, is expanded upon in this study. Other studies generally noted that the success of energy companies’ core activities and the influence and mode of digital change do not directly correlate; rather, they can only be deduced from appropriate studies. This paper demonstrates that digital growth develops into a game-changing innovation for businesses that may boost the performance of their core industries. And this innovation that changes the rules of the game eliminates the information asymmetry between various business departments of the enterprise and the external market environment. The internal communication channels of the enterprise and the communication channels between the enterprise and the external market are coordinated and shared, which better improves the efficiency and direction of product and service innovation for the enterprise and ultimately improves the business performance of the enterprise. This finding provides important insight into earlier research on enterprise digital transformation.

Second, the digital evolution of non-government energy enterprises far exceeds that of state-owned energy firms in terms of boosting main profitability; it is also more probable that this will take place in the east region’s energy firms. This paper shows that the impact of digital transformation on EEP varies in different types of enterprises, providing a deeper discussion than other research conclusions.

Third, digital transformation has an important impact on information exchange between energy enterprise systems. The allocation of responsibilities for activities executed during the integration process can be more clearly expressed [54], and the effectiveness of communication can be improved. This is one of the important channel mechanisms for digital transformation to improve the efficiency of energy companies’ core business.

The investigations herein set the groundwork for additional study. Although this study focuses on energy companies that are more representative and useful, it may have certain limitations owing to restrictions on the research topic and available space. First, the study excludes domestic unlisted corporations and companies that are not listed on A-share exchanges, taking only energy enterprises listed in Shanghai and Shenzhen. More energy-related companies should be included in the study sample in order to further test the conclusions discussed above. A future study could perform in-depth analyses of representative businesses. Second, it may be informative to analyze how the digital revolution affects the performance of energy companies in their core businesses in other countries, again offering an opportunity for additional research.

7. Conclusions and Policy Recommendations

This paper outlines the level of digitization of energy enterprises and empirically presents the consequences and mechanisms of digitization as well as its influence on the threshold of performance via the data of listed energy enterprises in Shanghai and Shenzhen from 2012 to 2020. Employing crawler data, this study also collects and assesses terms associated with digital shifts found in the annual reports of all publicly traded corporations. The results appear as follows.

First, the efficiency of energy companies’ core businesses is considerably enhanced through digitization. The digital evolution of non-government energy enterprises far exceeds that of state-owned energy firms in terms of boosting main profitability. In addition, this effect is more likely to take place in the east region’s energy firms. This may be because non-governmental energy enterprises do not have the advantages of government relations and financial support of state-owned energy enterprises and are more directly facing market competition pressure. At the same time, market competition pressure has also created more efficient resource adjustment mechanisms for non-state-owned energy enterprises. The eastern region of China is in a coastal economically developed region, with a deeper perception of the market compared to other regions and closer communication with the outside world. At the same time, the market competition pressure brought by economic development will also improve the digital transformation efficiency of local energy enterprises.

Second, input–output efficiency and market performance are two crucial paths underpinning breakthrough innovation in terms of channel mechanism pathways. Through the industry’s digital shift, EEP can be substantially increased. Digital transformation has an important impact on information exchange between energy enterprise systems. This indicates that digital transformation mainly improves the efficiency of information exchange between various departments and markets within the enterprise, thereby enhancing the innovation ability and accuracy of energy enterprises and ultimately improving enterprise performance.

Third, EEP and digital change have a relationship that is not linear. This is mostly caused by the illogical correlation of the input–output benefits approach, which is related to the willingness of enterprises to digitally transform. As innovation expenditure rises, so does the capacity of digital advancement to lift the fundamental operational performance of energy companies. However, after reaching a critical value for innovation expenditure gains, this efficiency may decline. This may be because as a heavy asset enterprise, energy enterprises need to pay a great deal of cost to carry out their own innovation process. If the relationship between innovation and cost control cannot be balanced, excessive pursuit of innovation will affect the actual operational performance of the enterprise.

The research findings in this report propose three implications for energy enterprise management procedures and policy support.

First, China should properly understand the development potential presented by the energy industry’s transition to digital technology. At the technical level, digital tools should be fully applied to improve productivity and enhance performance. Surplus funds and other factors after optimization at the organizational level should be invested in technological research and development and upgrading. Under this digital transformation-enabled breakthrough innovation mechanism, a positive feedback loop will be created to facilitate the deeper integration of digital innovations across all enterprise dimensions. As such, the enterprise-benefiting policy will be strengthened in favor of energy enterprises.

Second, the digital evolution of energy firms should follow a distinguishing principle. Based on the differentiated characteristics of the enterprises [55], the right remedy should be supported to motivate the state-owned energy enterprises to widely use digital transformation-related technologies. Unique digitalization paths with different objectives and standards for technological upgrading should be developed according to the differences in labor-intensive, technology-intensive, and knowledge-intensive characteristics and attributes of traditional energy enterprises and clean energy enterprises, respectively. Cooperation should be made in the areas of similarities in terms of resource structure optimization and information efficiency enhancement to create a digital ecosystem for energy companies.

Third, to avoid becoming caught in a digital contradiction, energy companies should make balanced investments in R&D. Energy companies should investigate novel breakthrough innovation models appropriate for their technology levels, increase the effectiveness of the use and allocation of innovation resources, and refrain from heedlessly increasing their R&D investments to improve EEP. The digital transformation of the energy sector has as its primary goals the development of a value system that examines the data components of this sector as well as the discovery and release of the value of big energy data.