Digital transformation and corporate financialization in emerging markets: Evidence from China

Abstract

Corporate financialization poses serious challenges to the development of the real economy. In the context of promoting the deep integration of the digital economy and the real economy, it is crucial to explore whether digital transformation can inhibit corporate financialization. Using data from Chinese listed companies from 2009 to 2021, we construct a fixed effects model and find that digital transformation significantly reduces the level of corporate financialization, a conclusion that still holds after a series of robustness tests such as propensity score matching and adding control variables. Channel analysis shows that that digital transformation inhibits corporate financialization by enhancing the information mobility and operational capability of corporations. In addition, this effect is more pronounced at higher levels of industry competition as well as marketization. Finally, we also find structural differences in the impact of digital transformation on corporate financialization. Our study explores the determinants of corporate financialization in terms of a firm's mode of operation and type of strategy, and the findings provide a theoretical basis for the active development of digital technologies in emerging markets that are undergoing economic transitions, as well as for guarding against the shift of the economy from the real to the virtual.

1. Introduction

Over the past few decades, the financial sector has consistently ranked among the top industries in terms of high return on investment in developed countries. However, there is a growing consensus among academics that the financial sector's dominance is encroaching on the real economy, a phenomenon commonly referred to as the "financialization of the economy" [1]. This encroachment has raised concerns about the prioritization of short-term financial investments over long-term projects by entity corporations, leading to a phenomenon known as corporate financialization. Scholars have increasingly focused on studying corporate financialization at the micro level, as it has been observed that this preference for short-term financial investments hinders real and innovative investments in long-term projects [2,3].

In many emerging markets, with the disappearance of the demographic dividend and the transformation and upgrading of manufacturing industries, how firms can gain a competitive advantage in order to dominate the market position is the key issue, so the research related to the corporate financialization is particularly important. China, as the world's second-largest economy, is actively seeking to accelerate its economic growth mode transformation. However, the prevalence of cross-sector arbitrage behavior among entity corporations has led to a structural imbalance between the real economy and the virtual economy. This imbalance poses a significant obstacle to the stable and healthy development of the economy. Consequently, inhibiting corporate financialization has become a crucial issue across various sectors. The existing literature has mainly focused on exploring the macro-level determinants of corporate financialization, such as macro-economic uncertainty and regulatory policies [4,5]. However, there is a relative scarcity of studies that investigate the determinants of corporate financialization in terms of corporate strategies and operational modes. In particular, against the backdrop of the reality of a booming digital economy on a global scale, can the digital transformation of companies become a powerful tool to steer the economy from virtual to physical? This is a meaningful theoretical and practical topic.

In recent years, changes in digital technology have permeated all walks of life. The new trend of informationization and digitization has triggered large-scale enterprise and even industry changes. Digital transformation refers to the adoption of digital technology to improve business processes, business models and customer experience for intelligent development, which empowers corporations to collect and process information more efficiently and helps them to gain a competitive advantage [6]. The Global Digital Economy White Paper (2022) highlights that the economic volume generated through digital transformation accounted for 45 % of the GDP of major economies in 2021, underscoring its significant contribution to the global economy. Indeed, for companies, digital transformation has become a daily agenda item for boards [7]. A study by Harvard Business Review Analytic Services (2014) showed that 50 % of business conductors said that their organizations have missed out on business opportunities presented by new technologies. Moreover, a study by MIT Sloan Management Review (2019) showed that boards of directors with digital executive members significantly outperformed their companies in all aspects of financial metrics, such as profit growth, return on assets, etc., than companies without digital executive members. In the era of digital economy, the integration of digital technology and traditional elements breaks down organizational boundaries, reshapes the business forms and value creation models of organizations, and greatly promotes internal management changes [8]. At the operational level, it has been found that digital transformation promotes strategic investment [9] and enhances innovation performance [10]. At the governance level, evidence suggests that digital transformation enhances the incentives for corporate disclosure and helps to reduce the level of information asymmetry, thereby weakening the agency problem [11]. From the above perspectives, digital transformation largely affects the investment decisions and business conditions of corporations, and also has an important impact on corporate governance, so it is necessary to explore the changes in corporate financialization from the perspective of corporate digital transformation.

Our research hypothesis is mainly based on the following inferences. First, digital transformation is a long-term process, and firms will inevitably "push information" to the market in the process of digital transformation, which will help external investors to pay attention to the real business situation of corporations, especially corporate investment in a timely manner. That is to say, digital technology can realize the perception, traceability and real-time sharing of business processes [12]. In the face of unpredictable market risks, investors will demand rigorous operations [13]. As a result, management's short-sightedness through financial investments is limited in the face of stronger external scrutiny. Second, digital transformation improves the efficiency of organizational operations, opens up the company's investment channels, and increases competitive advantage. With the positive feedback of digital transformation, the firm's willingness to invest in finance decreases.

We choose China as the focus of our study for several compelling reasons. Firstly, China serves as a representative country that has implemented corporate digital transformation on a large scale. The Digital China Development Report (2022) reveals that China's digital economy has reached an impressive scale of $7 trillion in 2022, ranking second globally in terms of total volume. Furthermore, its share of GDP is projected to increase to 42 %. Therefore, studying the outcomes of China's digital transformation can provide valuable insights for other countries, particularly emerging economies, that are undertaking similar digital transformation initiatives. Secondly, over the past decade, China's market demand space is shrinking, the growth rate of the real economy has slowed down significantly, on the contrary, the financial sector by virtue of the interest rate control policy has a very high rate of return on investment, which contributes to the motivation of cross-sector arbitrage of real enterprises, driving the economy from the real to virtual. In response, the Chinese government has repeatedly emphasized the importance of preventing the economy from becoming excessively virtualized and has advocated for a focus on the development of the real economy within the financial sector. In the context of the country's promotion of industrial restructuring and upgrading, research on the real economic effects of digital transformation can help deepen the perception of the actual effects of digital transformation in all circles.

Specifically, based on the annual report data of Chinese listed companies, we find that there is a negative correlation between digital transformation and corporate financialization. In order to enhance the reliability of our results, we used methods such as difference-in-difference model, replacing variables and adding control variables. Second, we examine the influence channels and find that digital transformation inhibits corporate financialization by enhancing the information mobility and operational capability of corporations. Considering the possible environmental factors, we conduct a cross-sectional analysis based on the level of industry competition as well as marketization, and we find that the effect of digital transformation in inhibiting corporate financialization is more pronounced when the level of industry competition is higher and the level of regional marketization is higher. Finally, we also find structural differences in the effect of digital transformation on corporate financialization.

This study makes several significant contributions to the existing literature. Firstly, it expands the research on the factors influencing corporate financialization by taking digital transformation as a focal point. While previous studies have mainly focused on macro factors such as economic policy uncertainty and regulatory policies, there is limited research on the impact of corporate strategy types on financialization. By examining the content of annual reports of listed companies, this paper enriches the understanding of the factors influencing corporate financialization from the perspective of digital transformation. Secondly, this study extends the research on the microeconomic effects of digital transformation by investigating its impact on corporate financialization. It explores the micro mechanisms through which digital transformation affects financialization, specifically by analyzing the role of information flow and operational capacity. This deepens our understanding of how digital transformation can contribute to the growth of the real economy and provides a theoretical foundation for further promoting digital transformation in the current stage. Compared to the study of Zhang et al. [14], we validate both external and internal mechanisms of digital transformation acting on corporate financialization. Thirdly, this paper explores the heterogeneous consequences of digital transformation. While previous literature has primarily focused on individual firm characteristics such as property rights and firm size [15,16], this study goes beyond that by examining the environmental dependence and structural differences in the consequences of digital transformation. It investigates the varying impact of digitalization on corporate financialization across different levels of industry competition, regional marketization, and segmentation. This adds a new dimension to the existing literature and provides valuable insights into the heterogeneity effects of digital transformation on corporate financialization.

2. Theoretical background and hypothesis development

2.1. Motivations for corporate financialization

Corporate financialization first emerged in developed countries, led by the U.S. Since the 1970s, the neoliberal slowdown in global demand and destructive product competition have led to a decline in the profitability of large U.S. non-financial corporations. Chronic overcapacity forced NFCs to turn to "impatient" financial markets, which in turn shortened their planning horizons [17]. Several literature has documented the increasing share of financial assets in the portfolios of US non-financial firms and the increasing share of financial profits in firms' profits [18]. On the other hand, the rise of the concept of "maximizing shareholder value" is an intrinsic driver of management's preference for corporate financialization. Agency theory suggests that moral hazard between management and shareholders can lead to poor corporate performance. In order to reconcile the interests of shareholders and management, firms need to take measures to improve agency efficiency, and one way to do this is through stock option-based compensation [19]. That is, management's compensation is directly linked to the company's stock market performance. However, this approach has driven large-scale financial payments by management, especially share buybacks, to maximize the firm's short-term share price.The hostile takeover movement that emerged in the 1980s and the shareholder revolution that followed invariably reflected managers' short-termist tendencies. Strong evidence suggests that shareholder value orientation crowds out fixed investment [20].

In most emerging markets, financial liberalization is often accompanied by high macroeconomic volatility and increased uncertainty, leading to greater performance pressure on managers [21,22]. As a result, firms seem to have incentives to hold financial assets for capital reserves and risk aversion. Compared to fixed assets, financial assets are more liquid with lower adjustment costs, and firms can invest as well as sell financial assets to cope with future business difficulties [23]. Tornell [24] argues that when faced with uncertainty, real firms in developing countries may be more willing to invest in financial assets that are more reversible and that can bring higher returns on financial assets. However, the literature provides evidence to the contrary. Many studies show the predominance of the "crowding out" effect of corporate financialization, which discourages real investment and is a drag on innovation investment and future business performance [25,26].

On the other hand, in the context of poor industry outlook, managers are likely to engage in risky financial investment activities as their expected revenue from product operations declines [4]. It has been found that in the Chinese context, firm managers allocate financial assets to hide negative information such as declining performance in the main industry, and this information "muffling" behavior increases the future stock price crash risk [27]. The above results imply that under the current poor investment environment, non-financial corporations are likely to invest in financial products for profit-seeking motives.

2.2. Economic consequences of digital transformation

The world today is experiencing an unprecedented wave of digital transformation. The development of emerging technologies represented by artificial intelligence, big data, and cloud computing has led companies to explore transformation routes to gain new competitive advantages. These changes in digital technologies have a disruptive impact on companies' organizational forms, operational models, and customer behaviors [28]. Evidence suggests that companies have been paying more attention to digital transformation in recent years, as evidenced by a significant increase in the frequency of keywords related to digital transformation in company annual reports [15]. Prior studies have documented the impact of digital transformation on firms in terms of productivity and performance. For example, Wu et al. [29] found that data analytics techniques can integrate knowledge from different domains leading to greater productivity benefits. Brynjolfsson et al. [30] argued that the development of artificial intelligence has spawned complementary innovations that have contributed to an increase in total factor productivity. Mikalef et al. [31] found that digital technology enhances firms' dynamic capabilities thereby positively affecting competitive performance by using data from Norwegian firms. Li et al. [32] obtained similar findings using data from China. Some studies have also examined the internal governance effects of digital transformation, and evidence suggests that digital transformation can alleviate the financing constraints of small and medium-sized enterprises (SMEs) [33] and reduce firm's stock price crash risk [11].

Based on existing research, it is easy to see that digital transformation has important implications for both the operational and governance levels of firms. In this paper, we focus on whether the digital transformation of firms can reduce the ability and willingness of firms to invest financially thereby inhibiting corporate financialization, which is a useful addition to the literature on the economic consequences of digital transformation.

2.3. Hypothesis development

The previous analysis shows that management myopic behavior triggered by agency problems is an important determinant of corporate financialization, and this performance is more pronounced when firms face poor investment opportunities and industry sector outlooks. We argue that digital transformation can in turn inhibit corporate financialization in terms of both the ability and willingness of firms to invest financially.

Digital transformation refers to the integration of digital technology into various aspects of a corporation's production, operations, and activities. By optimizing resource allocation and transforming existing processes, digital transformation aims to enhance organizational efficiency. According to Wu et al. [34], corporations that undergo digital transformation are more likely to attract market attention and effectively disseminate information to market participants.Based on this premise, it can be inferred that digital transformation has several positive effects. Firstly, by increasing the liquidity of information, digital transformation enhances the visibility of corporations in the market, thereby strengthening external monitoring. Through digital transformation, corporations promote internal and external information sharing, leading to improved information transparency. This reduces the ability of management to conceal internal company information through financial asset allocation, thereby alleviating principal-agent problems and enhancing management efficiency. Secondly, digital transformation empowers organizations with new business models and operational modes, enhancing their technological innovation capabilities. This, in turn, helps corporations optimize production factor inputs and reduce production costs. Consequently, digital transformation enhances organizational operations, improves operational efficiency, and increases profitability at the firm level. Taken together, these two aspects create favorable conditions for reducing management's short-sighted motives and mitigating their tendency to engage in financial sector arbitrage. Therefore, we propose the following hypothesis.

H1

Digital transformation can inhibit corporate financialization.

The crucial advantage of digital transformation is its ability to handle vast amounts of non-standardized and unstructured data, both internally and externally, within corporations. This ensures a smooth flow of information within and outside the organization. When internal information remains unexplored, external investors lack insight into the corporation's internal affairs, providing an opportunity for management to pursue personal gain. However, through digital transformation, corporations can actively share business information with the outside world, accelerating the dissemination of information among investors [35]. This reduces the information asymmetry between internal and external stakeholders, enabling them to better understand the true state of affairs within the corporation and curbing management's opportunistic behavior. Research by Jiang et al. [11] demonstrates that digital operations can inhibit management's manipulation of information, thereby reducing the risk of stock price crashes. Financialized investment behavior, on the other hand, is often used by management to conceal negative news or information [27]. As the digital transformation process continues to advance, the market gains more comprehensive information about corporations, empowering investors to play a stronger role in external supervision. This mitigates the principal-agent problem and curtails management's short-term financialized investment behavior. Moreover, in the context of China's high-quality development of the digital economy, corporations aligned with digital transformation initiatives receive increased attention from the market, particularly from analysts and research institutes. This heightened exposure creates a stronger monitoring effect, making it challenging for management to manipulate news through financialized investments. As a result, short-term profit-seeking behavior is suppressed. In light of these arguments, we propose the following hypothesis.

H2

Digital transformation can inhibit corporate financialization by enhancing the information mobility of corporations.

Digital transformation plays a crucial role in improving the operational efficiency and profitability of corporations through organizational structure changes and business process transformations. By adopting digital technologies, corporations can create a networked and flattened organizational structure, facilitating cross-border entry and vertical integration among functional departments [36]. This enables corporations to effectively integrate and leverage their resources, capture market demand in a timely manner, and enhance the overall efficiency of the supply chain. In terms of production, corporations can optimize production plans by leveraging market information, enabling flexible capacity utilization and faster inventory turnover. This leads to improved capital efficiency. Similarly, in transportation, corporations can leverage digital platforms to obtain optimal logistics solutions, significantly enhancing operational efficiency. For instance, Bright Dairy has implemented environmentally conscious management and breeding platforms, precision feeding systems, and milk source traceability systems across their farms and cows nationwide. Additionally, they have leveraged big data and cloud computing to establish "smart manufacturing" factories and "just-in-time delivery" shopping platforms. These initiatives enable data interoperability and intelligent synergy across the entire industry chain, ensuring traceability from field to table. In addition, corporations acquire rich user information through digital operation, which helps further improve product quality. Digital corporations can make use of big data, cloud computing and other digital technologies to fully explore the potential needs of users and personalized needs, so as to provide users with more accurate and comprehensive services, and further enhance the competitiveness of company products. The above initiatives have undoubtedly created corporate profit growth space, on the one hand, good profitability reduces the management's incentive to pursue excessive returns on financial assets; on the other hand, the new profit growth space implies a higher quality of investment channels, which promotes the rational allocation of resources and reduces the demand for the behavior of moving "from real to virtual". Based on the above arguments, we propose the following hypothesis.

H3

Digital transformation can inhibit corporate financialization by enhancinge the operational capacity of corporations.

3. Sample, variables and methodology

3.1. Sample selection and data sources

We select the data of Chinese listed companies from 2009 to 2021 as the original research sample. There are two main reasons for starting our sample period in 2009. First, China implemented a new corporate accounting standard in 2007, and in order to avoid the impact of measurement bias caused by the alternation of the old and new accounting standards on the study's conclusions, we chose to delay the year by two years to start the observation. Second, after the financial crisis in 2008, countries began to realize the necessity of digital transformation, and thus digital transformation policies began to receive more attention. This can be seen in the amount of literature on digital transformation over the years [37]. Then we refer to existing studies to screen the samples as follows: (1) eliminate samples of financial and real estate companies; (2) eliminate samples handled by ST(Samples with special treatment due to two consecutive years of losses), *ST (Sample with special treatment due to three consecutive years of losses) and PT (Sample of stocks suspended due to three consecutive years of losses) during the observation period; (3)eliminate samples of insolvent companies; (4) eliminate samples of companies listed in the year of observation; (5) eliminate samples with missing main variable observations, and finally we obtain 28,253 annual observations. The data used in the paper are mainly from the CSMAR (China Stock Market & Accounting Research) database and the CNRDS (Chinese Research Data Service) database. Finally, all continuous variables at the firm level are deflated at the upper and lower 1 % quartiles.

3.2. Variable definitions

3.2.1. Corporate financialization

Existing studies mainly measure corporate financialization from the financial asset allocation perspective and the financial channel profitability perspective. In this paper, we focus on the choice of corporate investment behavior in the context of digital transformation, so we adopt the financial asset allocation perspective, i.e., the proportion of financial assets to total assets, to measure it. Consistent with previous studies [38,39], financial assets encompass various categories, including money funds, trading financial assets, derivative financial assets, loans and advances, available-for-sale financial assets, held-to-maturity investments, and investment real estate. It is worth noting that China implemented a new accounting standard in 2018, which necessitates slight adjustments in measuring corporate financialization. Therefore, considering the updated accounting standards, we modify the measurement for samples after 2018. In addition to the aforementioned categories, we include other investments in equity instruments, investments in debt, other investments in debt, and other non-current financial assets in the calculation of financial assets. Conversely, we exclude available-for-sale financial assets and held-to-maturity investments from the measurement. We utilize the proportion of financial assets to total assets as the measure of corporate financialization. Through the above methods, we can capture the extent to which corporations allocate their assets to financial instruments.

3.2.2. Digital transformation

Previous research has investigated various indicators to measure firms' digital transformation. However, it is widely acknowledged that a firm's strategic characteristics and operating conditions are better reflected in its annual reports. Therefore, we adopt a proxy variable for digital transformation based on the frequency of keywords related to digital transformation in firms' annual reports, following the methods used by Ren et al. [16].To ensure the appropriateness of the data distribution, we address the right-skewed nature of the data. Firstly, we add one to the frequency count, then, we take the natural logarithm of the adjusted frequency count to obtain an overall indicator of digital transformation. By using this approach, we can capture the level of digital transformation within firms and incorporate it into our analysis. This indicator allows us to assess the extent to which firms have embraced digital technologies and transformed their operations.

3.2.3. Control variables

Referring to existing studies, we control for some firm-level variables and the details are as follows: First, we control for management shareholding (manage), which reflects compensation incentives and management's rights. Second, we control for the nature of ownership (soe), which is unique to China's institutional context. Existing evidence suggests that SOEs and non-SOEs do not exhibit the same financialization performance [40]. We also control for corporate governance variables such as board size (board) and the proportion of independent directors (indep), who reflect the governance structure. In addition, controlling for financial indicators is necessary because they have an impact on the firm's resource allocation, including indebtedness (lev), corporate growth (growth), and cash flow position (cash). Finally, we control for company size (size) because size affects corporate financialization [41]. The variables are described in Table 1.

Table 1.

Variable definition.

Variable	Definition
fin	Corporate financial assets/total assets
digital	The number of occurrences of keywords about digital transformation in the firm's annual report is added to one, and the natural logarithm is then taken
manage	Number of shares held by management/company's total number of shares in
soe	State-owned enterprises take the value of 1, otherwise 0
board	The number of members of the Board of Trustees is added and repeated in natural logarithms
indep	Number of independent directors/number of company's board of directors
lev	Total liabilities/total assets
growth	Operating income for the year divided by operating income for the previous year, less one
cfo	Net operating cash flow/total assets
size	Natural logarithm of total company assets

3.3. Baseline regression model

In order to test the hypotheses presented in the previous section, we construct the following model：

$$fi{n}_{i,t}={\alpha }_0+{\alpha }_{1}digita{l}_{i,t}+Contro{l}_{i,t}+Firm+Year+{\epsilon }_{i,t}$$ (1)

where fin represents the level of corporate financialization, ditgital represents the level of digital transformation, and Control represents a set of control variables that are random error terms. In order to mitigate the effects of omitted variables as much as possible, we also control for firm effects (Firm effects) and year effects (Year effects).

4. Analysis of empirical results

4.1. Descriptive statistics

Table 2 reports the descriptive statistics of the variables. fin has a mean of 0.22 and a median of 0.18, indicating that overall China's entity corporations have a high level of financialization and it is right-skewed; the maximum value is 0.71, indicating that some of the firms have a high level of financialization. digital has a mean of 1.29, and about 56.97 % of the firms have their digitization level below the sample mean; the standard deviation is 1.38, indicating that there is a large difference in the level of digitization between the samples. The values of the other variables are all within a reasonable range.

Table 2.

Descriptive statistics of variables.

Variable	Obs.	Mean	Sd	Min	P50	Max
fin	28,253	0.2227	0.1478	0.0241	0.1844	0.7112
digital	28,253	1.2902	1.3824	0	1.0986	5.0039
managehold	28,253	0.1308	0.1929	0	0.0047	0.6750
soe	28,253	0.3671	0.4820	0	0	1
board	28,253	2.2454	0.1762	1.7918	2.3026	2.7726
indep	28,253	0.3744	0.0529	0.3333	0.3333	0.5714
lev	28,253	0.4227	0.1994	0.0558	0.4170	0.8729
growth	28,253	0.2986	0.6948	−0.6542	0.1300	4.5598
cfo	28,253	0.0489	0.0673	−0.1454	0.0470	0.2417
size	28,253	22.1430	1.2320	19.9142	21.9746	25.9911

4.2. Basic regression result

In order to examine the impact of digital transformation on corporate financialization, we take a step-by-step approach of adding control variables to estimate model (1), as shown in Table 3. Column (1) shows the results without adding any control variables, and it can be found that the coefficient of fin is significantly negative, indicating that digital transformation can inhibit corporate financialization when other control variables are not considered. Columns (2)-Column (4) show the regression results with the gradual addition of control variables, and it can be found that the sign and significance level of fin remain unchanged, indicating that the conclusion that digital transformation reduces the level of corporate financialization is still robust. In terms of economic significance, for each 1-unit standard deviation increase in the level of digital transformation, the reduction in the level of corporate financialization is equivalent to 2 % of the standard deviation of the sample.

Table 3.

The result of hypothesis H1.

	(1)	(2)	(3)	(4)
	fin	fin	fin	fin
digital	−0.0052***	−0.0050***	−0.0034***	−0.0024***
	(-5.8002)	(-5.5899)	(-3.9523)	(-2.8383)
managehold		0.1137***	0.0682***	0.0681***
		(13.6697)	(8.5121)	(8.5495)
soe		−0.0121***	−0.0037	−0.0039
		(-2.7173)	(-0.8708)	(-0.9206)
board		0.0038	0.0039	0.0078
		(0.5703)	(0.5257)	(1.0551)
indep			−0.0280	−0.0233
			(-1.3796)	(-1.1577)
lev			−0.2737***	−0.2622***
			(-47.6871)	(-44.0928)
growth				0.0003
				(0.2812)
cfo				0.1930***
				(18.4345)
size				−0.0015
				(-1.0157)
Firm effects	Yes	Yes	Yes	Yes
Year effects	Yes	Yes	Yes	Yes
Observations	28,253	28,253	28,253	28,253
R2	0.5859	0.5893	0.6238	0.6274

Note: ***, ** and * represent t-values significance at the 1 %, 5 % and 10 % statistical levels, respectively. The following tables are the same.

4.3. Influence channel testing

According to the findings above, digital transformation can inhibit corporate financialization, but the underlying mechanism of this effect has not been verified. We argue that the impact of digital transformation on corporate financialization mainly relies on two channels: the information flow channel and the operational capability channel, i.e., digital transformation inhibits corporate financialization by facilitating the information liquidity and improving operational capability.

4.3.1. Information mobility channel

To measure the liquidity of corporate information, we employ three indicators: investor attention (investor), analyst attention (analyst), and report attention (resport). Firstly, as corporations engage in digital transformation and disseminate information to the market, they attract increased attention from external investors. This heightened investor attention serves as an effective form of supervision, curbing opportunistic behavior by management. Therefore, investor attention is an ideal indicator for measuring information mobility. We use investor attention (investor), analyst attention (analyst), and report attention (resport) to measure the mobility of corporate information. With the widespread adoption of the Internet and the use of search engines in China, investors increasingly rely on online news sources to access corporate information. To capture this trend, we utilize the corporate internet search index as a measure of investor attention. This index reflects the search volume of keywords such as stock codes and company abbreviations associated with listed companies, providing a more accurate reflection of investor attention. To obtain the annual index of investor attention, we aggregate the daily index values, add one to the sum, and finally take the natural logarithm. Furthermore, digital transformation has the potential to reduce information asymmetry, leading to increased attention from analysts and research reports. A higher number of analysts tracking a company and a greater number of research reports indicate a higher degree of external supervision and information flow. Therefore, we also incorporate analyst attention and research report attention into our analysis. These indicators are calculated by summing the number of analysts and annual reports tracking a company over a year, respectively, and then taking the natural logarithm of the sums.

The test results of information flow channels are shown in Table 4, columns (1)-column (3), which shows that the regression coefficients of digital transformation (digital) and investor attention (investor), analyst attention (analyst), and research report attention (report) are positive and highly significant, indicating that the digital transformation helps to improve the smoothness of internal and external information flow, which brings stronger external monitoring effect and then inhibits the financialized investment behavior of corporations.

Table 4.

The result of hypothesis H2.

	(1)	(2)	(3)
	investor	analyst	report
digital	0.0133***	0.0290***	0.0415***
	(2.8780)	(4.2632)	(4.8156)
Control	Yes	Yes	Yes
Firm effects	Yes	Yes	Yes
Year effects	Yes	Yes	Yes
Observations	25,464	19,993	20,225
R2	0.9548	0.5858	0.5741

4.3.2. Operational capability channel

We choose the inventory turnover ratio (inventory) and the main business profit margin (mbc) as two indicators to measure the operational capability of corporations. Firstly, inventory turnover reflects the operational efficiency of the firm in all aspects of production and operation, i.e., whether the use of funds in each aspect is reasonable or not, so the level of inventory turnover is an important indicator of whether the digital transformation can improve the operational efficiency of the organization. Secondly, whether digital transformation can bring stronger competitiveness and better product market performance will ultimately be reflected in the firm's profitability, so we use the profitability of main business margin to portray the profitability of the corporation, which is measured as follows: (main business revenue - main business cost - main business taxes and surcharges)/main business revenue.

The test result of the operational capability channel are shown in columns (1)-columns (2) of Table 5, the regression coefficients of digital transformation (digital) and inventory turnover (inventory) as well as main business profitability (mbc) are all significantly positive, which indicates that corporations can effectively improve their operational capability through digital transformation, which helps to open up the firm's good investment channels and reduces the need for short-term excess returns on corporate financial investments.

Table 5.

The result of hypothesis H3.

	(1)	(2)
	inventory	mbc
digital	1.4973***	0.0019*
	(4.9758)	(1.6790)
Control	Yes	Yes
Firm effects	Yes	Yes
Year effects	Yes	Yes
Observations	27,962	28,248
R2	0.6735	0.6218

4.4. Robustness tests

4.4.1. PSM-DID

Considering that the firm's transformation from non-digital to digital is a quasi-natural experiment, we use a multi-period double-difference model to mitigate the possible sample selection bias problem, as follows：

$$fi{n}_{i,t}={\beta }_0+{\beta }_{1}Trea{t}_{i,t}\times Afte{r}_{i,t}+Contro{l}_{i,t}+Firm+Year+{\epsilon }_{i,t}$$ (2)

In this case, Treat indicates that the firms that went from non-digital transformation to digital transformation in the observation period are set as the treatment group and take the value of 1; at the same time, the samples that did not undergo digital transformation at all times in the observation period are set as the control group and take the value of 0. After indicates the dummy variable for digital transformation change, and the year in which the firms underwent digital transformation and the year after that takes the value of 1, while the year before digital transformation is set as 0. In order to improve the validity of the model estimation, we conduct one-to-one nearest-neighbor matching through propensity score matching to select appropriate samples before regressing the model (2), and the results of the test Table 6 column (1) show that the coefficient of the interaction term Treat × After is significantly negative, indicating that the level of corporate financialization is significantly reduced after the firm undergoes digital transformation.

Table 6.

The result of robustness tests.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)
	fin	fin	fin	fin	fin2	fin	fin
Treat_After	−0.0146***
	(-4.6991)
L.digital		−0.0022**
		(-2.4327)
digital			−0.0019**		−0.0042***	−0.0021**	−0.0023***
			(-2.2665)		(-5.1264)	(-2.4254)	(-2.6842)
invest			0.0743***
			(11.8221)
finback			0.0016
			(0.4407)
digital_max				−0.0051***
				(-4.8947)
Control	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Firm effects	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year effects	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Industry ×  Year effects	No	No	No	No	No	Yes	No
Region× Year effects	No	No	No	No	No	No	Yes
Observations	10,897	23,628	28,217	28,253	28,253	28,253	28,253
R2	0.5854	0.6465	0.6297	0.6276	0.6048	0.6339	0.6328

4.4.2. Selection of lagged explanatory variables

Considering the possible reverse causality problem between digital transformation and corporate financialization, we lag the explanatory variables by one period, and the regression results are shown in column (2) of Table 6, the regression coefficients are still significantly negative, which indicates that the digital transformation can reduce the level of corporate financialization, and the conclusion remains unchanged.

4.4.3. Adding control variables

Alexiou et al. [42] found that institutional investor shareholding has a positive impact on corporate financialization; Du et al. [43] found that CEO financial background also has a significant impact on corporate financialization. In this regard, we add two variables, institutional investor shareholding (invest) and CEO financial background (finback), to the existing model. The test results, as shown in column (3) of Table 6, show that the coefficient of invest is significantly positive, while the coefficient of finback is also positive but not significant. More importantly, the coefficient of digital is still significantly positive, which indicates that the conclusion that digital transformation inhibits corporate financialization is unchanged with the addition of control variables.

4.4.4. Replacing explanatory and interpreted variables

Considering that when corporations disclose information related to digital transformation, they disclose more in the year of digital transformation and less in the year after the completion of digital transformation, in this regard, we firstly determine the maximum value of the digital transformation level of each corporation, and then set all the sample values of the year after this maximum value to this maximum value, which is used to replace the original digital transformation level, named digital_max. The regression results are shown in column (4) of Table 6, and the coefficient of digital_max is significantly negative. We replace the original level of corporate financialization with the proportion of money funds and trading financial assets summed up to total assets, named fin2, and the regression results are shown in column (5) of Table 6 and the results are not significantly changed. The above results show that the conclusions of this paper are robust after considering the measurement bias of variables.

4.4.5. Considering annual trends by industry as well as by region

The development of the industry during the sample observation period may be characterized by cyclical changes due to monetary policy, industrial policy and other factors, while regional economic growth may also be characterized by cyclical changes, which in turn have an impact on the investment decisions of corporations. In order to control the impact of these macro-level factors, this paper further controls for the "industry × year" effect and the "region × year" effect on the basic regression. The regression results are shown in column (6) and column (7) of Table 6, and the regression coefficients of digital are still significantly negative, indicating that the effect of digital transformation in inhibitig corporate financialization still exists after considering the influence of macro-level factors.

4.5. Further discussion

The relationship between digital transformation and corporate financialization has been discussed in the previous section. However, it is important to acknowledge that the implementation of digital transformation in real-world scenarios can be influenced by external factors. Additionally, the digital transformation index used in this study is a comprehensive measure, and the impact of digital transformation on corporate financialization may vary across different internal levels of the organization. Therefore, we aim to explore the heterogeneity of the effects of digital transformation on corporate financialization from three perspectives: industry competition, regional marketization, and structural differences in the corporation's digital transformation efforts.

4.5.1. Considering the level of competition in the industry

The level of competition in an industry plays a crucial role in shaping firms' investment decisions. In highly competitive markets, firms often experience a decline in profitability and face increased pressure to perform. As a result, they may be more inclined to seek opportunities for arbitrage in the financial and real estate sectors, where they can potentially earn excess returns. Guo et al. [44] have highlighted that in more competitive industries, management is more willing to prioritize short-term benefits over long-term profitability, which can hinder firms' innovative activities. However, digital transformation offers a potential solution to this challenge. By integrating digital technology into their production and operational processes, corporations can enhance their efficiency and improve the quality of their products. This, in turn, can boost the profitability of their core business and reduce their reliance on short-term arbitrage opportunities. Therefore, digital transformation has the potential to mitigate the incentives for corporate financialization. Based on this logical framework, it can be expected that the impact of digital transformation on inhibiting corporate financialization would be more pronounced in industries facing intense market competition. In such competitive environments, digital transformation can play a pivotal role in driving the development of corporations by enhancing their competitive advantage and reducing the need for short-term arbitrage.

In this regard, we follow the median of the HHI (Herfindahl-Hirschman Index) in the same year to classify the level of competition in the industry in which the corporation is located, and samples below the median take the value of 1, which indicates that the market competition is more intense; otherwise, it is 0, which indicates that the competition is less intense in the market. The group regression results are shown in Table 7, columns (1)-column (2), in the group with more intense market competition, the coefficient of digital is −0.0047 and significant at the 1 % level; while in the group with less intense market competition, the coefficient of digital is −0.0021 and significant at the 10 % level. This suggests that digital transformation has a greater marginal effect when corporations face more intense competition.

Table 7.

The result of heterogeneity test for industry competition.

	(1)	(2)
	High level of industry competition	Low level of industry competition
	fin	fin
digital	−0.0047***	−0.0021*
	(-3.7029)	(-1.7289)
Control	Yes	Yes
Firm effects	Yes	Yes
Year effects	Yes	Yes
Observations	13,319	14,524
R2	0.6319	0.6515

4.5.2. Considering the level of regional marketization

China is a vast country with regional variations in the level of marketization, which is influenced by differences in economic development and policy orientation. The level of marketization has a significant impact on resource allocation by corporations and its effects. Firstly, a high level of marketization indicates more mature factor and product markets [43], leading to firms relying more on market price signals to make investment decisions. This enhances the sensitivity of firms' investment levels to investment opportunities, making them more inclined to undertake digital transformation aligned with economic practices.Secondly, increased regional marketization expands corporations' financing channels. Digital transformation often requires substantial capital investment, and regions with a high level of marketization can effectively alleviate financing challenges associated with digital transformation. This reduces the need for corporations to enhance their financing capacity through financial investment, thereby leveraging the role of digital transformation in boosting the real economy. Furthermore, a high level of marketization implies a better legal environment, which has a fundamental governance effect at the firm level [45]. This promotes information flow and reduces agency costs, providing institutional support for the implementation of digital transformation. Corporations can actively promote digital transformation to obtain effective feedback in regions with a high level of marketization. Based on the above analysis, it can be inferred that the role of digital transformation in inhibiting corporate financialization is more pronounced in regions with a high level of marketization.

We use China Sub-Provincial Marketization Index Report (2021) compiled by Wang et al. [46] for measuring regional marketization levels. Since the index is currently updated only until 2019, in this regard, we refer to the current practice of most of the literature, using the average growth rate of the marketization index in previous years to extend the marketization index to 2021, and taking the value of samples higher than the median of the marketization index in the same year to be 1, which indicates a high level of marketization, and 0 otherwise, which indicates a low level of marketization. The results of the subgroup test are shown in columns (3)-columns (4) of Table 8, where the coefficient of digital is −0.0027 in the group with a high marketization level and is significant at the 5 % level, while the coefficient of digital is −0.0016 in the group with a low marketization level, but is not significant. This result supports the above conjecture.

Table 8.

The result of heterogeneity test for regional marketization.

	(1)	(2)
	High level of marketization	Low level of marketization
	fin	fin
digital	−0.0027**	−0.0016
	(-2.1465)	(-1.3504)
Control	Yes	Yes
Firm effects	Yes	Yes
Year effects	Yes	Yes
Observations	13,279	14,507
R2	0.6305	0.6325

4.5.3. Considering the structural differences in digital transformation

The baseline regression used holistic indicators of digital transformation to test the effect of digital transformation on corporate financialization, however, different levels of digital transformation may have different effects. In this paper, digital transformation is divided into two levels: the underlying technology application (digital_underlying) and the practical technology application (digital_practice). The regression results, as shown in Table 9, indicate that the regression coefficient of digital_underlying is significantly negative, while the regression coefficient of digital_underlying is negative but not significant, which indicates that the digital transformation at the level of underlying technology application has a more significant inhibitory effect on corporate financialization than that at the level of practical technology application. This may be due to the fact that the underlying technology, such as big data and cloud computing, focuses on the embedding of digital technology and the integration of digital technology with the original organizational structure, production process, operation and management mode. The practical application level focuses on the application of digital technology in complex scenarios. Therefore, digital transformation at the level of underlying technology has a more direct and effective impact on corporate governance and production and business models, and helps to inhibit corporate financialization.

Table 9.

The result of structural differences in digital transformation.

	(1)	(2)
	fin	fin
digital_underlying	−0.0036***
	(-6.9235)
digital_practice		−0.0002
		(-0.2609)
Control	Yes	Yes
Firm effects	Yes	Yes
Year effects	Yes	Yes
Observations	28,253	28,253
R2	0.6280	0.6273

5. Conclusion

The trend towards financialization under the liberalization of the global economy poses a threat to the economic performance of many countries. Corporate financialization not only crowds out productive expenditures and R&D investments, but also poses a hazard to the stability of the financial system. In our study, we examine how one such model of corporate operations, digital transformation, affects corporate financialization.

Our results support the class of arguments that digital transformation benefits the real economy. We find that digital transformation significantly inhibits corporate financialization because it reduces the ability and willingness of firms to engage in financial speculation. Although this study only uses data from China, the conclusion has implications for other countries and regions with rapidly developing financial sectors. Especially for emerging market economies, the uncertainty of the international environment brings more potential problems of national governance innovation, they need to be more alert to the risks associated with the shift of the economy from the real to the virtual in the process of optimizing and upgrading their economic structure. In this regard, relevant policymakers may consider introducing policies to stimulate the digital transformation of corporations and help traditional industries gradually transform into high value-added and intelligent ones. This result is also relevant for corporations, and we recommend that they focus more on investing in digital technology, as it helps reduce information asymmetry as well as improve operational capabilities.

Our results also emphasize the importance of environmental conditions for pursuing digital transformation. Cross-sectional tests show that the dampening effect of digital transformation on corporate financialization is more pronounced at higher levels of industry competition as well as marketization. This also holds intuitively. In highly competitive markets, firms help gain a competitive advantage through better digital performance, reducing the need for financial speculation. Higher levels of marketization imply better factor markets as well as property rights protection systems, which contribute to the effects of digital transformation. At the same time, given the structural differences in digital transformation, the synergies of digital technologies at all levels should be given special attention.

Finally, while this study deepens our understanding of how digital technologies affect the real economy, it still has several limitations. For example, we only tested the impact of firms' digital transformation on themselves, however, digital transformation may have a peer effect, which requires further testing. In addition, the role played by the characteristics of the management as the most important human resource among the company's operations in the digital operation process also deserves further research in the future.

Data availability statement

The data associated with our study been deposited into a publicly available repository. Please see: Fang, Xusheng; Ju, Chunhua (2023), “Data for Digital Transformation and Corporate Financialization”, Mendeley Data, V1, https://doi.org/10.17632/ts4wgg3zbb.1.

Funding

This study was supported by National Social Science Foundation Major Project (Grant No. 21&ZD119)

CRediT authorship contribution statement

Xusheng Fang: Writing – review & editing, Writing – original draft, Software, Data curation. Chunhua Ju: Writing – review & editing, Writing – original draft, Data curation.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.