Abstract
While innovation-driven development has become essential for economic transformation, the specific mechanisms through which government procurement enhances enterprises’ new quality productivity remain underexplored, particularly regarding how procurement policies facilitate technological innovation, human capital development, and digital transformation. Using comprehensive data from Chinese A-share listed companies from 2015 to 2022, this study employs fixed effects models and instrumental variable approaches within a “capability-willingness-action” theoretical framework to examine government procurement’s causal impact on new quality productivity. Results demonstrate that government procurement significantly increases firms’ new quality productivity, operating through three distinct transmission mechanisms: first, by alleviating financing constraints and reducing debt costs to enhance innovation capability; second, by reducing uncertainty perception and incentivizing digital transformation to strengthen development willingness; and third, by establishing technical standards and quality requirements to promote concrete actions in R&D investment, talent cultivation, and innovation output. Heterogeneity analysis reveals particularly strong effects among state-owned enterprises, growth-stage firms, and companies with superior internal control quality. These findings provide systematic evidence of government procurement’s multidimensional impact on new quality productivity, offering crucial insights for optimizing procurement policy design and advancing innovation-driven economic development.
Plain Language Summary
To achieve high-quality development during China’s 14th Five-Year Plan period, it is essential for governments to reform the procurement system, optimizing policy frameworks to facilitate the efficient aggregation of advanced productive forces, in alignment with evolving growth paradigms. This study explores the concept of “new quality productivity” and its relationship with government procurement, analyzing its impact on enterprise productivity. Using a sample of listed companies receiving government procurement orders from 2015 to 2022, the findings suggest that government procurement enhances new quality productivity by alleviating financing constraints, boosting development confidence, and increasing technological investments and outputs through specific mechanisms. Heterogeneity analysis indicates the most significant effects in state-owned enterprises, and those with strong risk management, in growth stages, and located in municipalities, provincial capitals, or central regions. The study integrates theoretical and empirical insights, offering strategic recommendations for policymakers and firms, highlighting government procurement’s important role in improving productivity.
Keywords
Introduction
In the critical period of global economic transformation and upgrading, traditional production models relying on capital and labor inputs are facing increasingly severe developmental bottlenecks (J. Chen et al., 2025; Jin et al., 2023), urgently necessitating the search for new sources of growth momentum. China’s proposed concept of “New Quality Productivity” (NQP) emphasizes achieving qualitative leaps in productivity through technological innovation, human capital enhancement, and digital transformation, providing crucial theoretical guidance for addressing this challenge (Feng et al., 2024; Lei, 2024b; Yang et al., 2025). As a concept distinct from traditional Total Factor Productivity (TFP), NQP not only focuses on improving production efficiency but emphasizes the inherent requirements of innovation-driven development, green development, and sustainable growth, embodying the comprehensive pursuit of high quality, high efficiency, and high technological content in modern economic development. The introduction of this concept not only enriches the theoretical connotations of productivity theory but also provides a new analytical framework for enterprise transformation and policy formulation.
In research on pathways for enhancing new quality productivity, existing literature primarily focuses on supply-side factors such as technological innovation and human capital investment, with relatively insufficient attention to the mechanisms of demand-side policy instruments. Government procurement, as an important demand-side policy tool, plays an increasingly vital role in promoting enterprise innovation and productivity enhancement. International research demonstrates that government procurement can effectively incentivize enterprise innovation investment and technological progress through multiple mechanisms including creating stable market demand, providing technological standard guidance, and alleviating enterprise financing constraints (Georghiou et al., 2014; Ghisetti, 2017). Particularly in emerging technology fields and green development, the leading role of government procurement is even more significant, as it reduces the uncertainty of enterprise innovation and improves expected returns on innovation investment through early demand creation and risk sharing (Stojčić et al., 2020).
However, existing research on government procurement’s impact on enterprise productivity exhibits obvious theoretical and empirical gaps. From a theoretical perspective, most studies remain within the analytical framework of traditional productivity indicators, lacking in-depth theoretical construction and empirical analysis of NQP as a productivity form embodying contemporary characteristics. From a mechanism analysis perspective, while existing research has identified important transmission pathways such as demand-pull effects, certification effects, and financing constraint alleviation, understanding of how government procurement enhances NQP through influencing enterprise digital transformation willingness, risk-taking spirit, and talent structure optimization remains insufficient (Dai et al., 2021). From a heterogeneity analysis perspective, although existing research considers differences across dimensions such as firm size, ownership nature, and industry characteristics, studies examining how factors such as enterprise lifecycle stages, internal control governance quality, and regional development levels moderate government procurement effects remain limited, constraining theoretical guidance for precise government procurement policy design and differentiated implementation.
Based on the aforementioned research status, this paper proposes the following core research questions: How does government procurement affect enterprises’ enhancement of NQP through multi-dimensional transmission mechanisms? What are the key mediating pathways and moderating factors in this influence process? Do systematic differences exist in different types of enterprises’ responses to government procurement policies? To address these questions, this paper constructs a comprehensive NQP measurement system encompassing multiple dimensions including human capital, physical capital, technological capabilities, and operational efficiency, utilizing large-sample data from Chinese A-share listed companies from 2015 to 2022, and systematically examines the impact of government procurement on enterprise NQP and its internal mechanisms through a progressive logical framework of “capability-willingness-action.”
This paper’s marginal contributions are reflected across four dimensions—theoretical innovation, research perspective, empirical findings, and policy implications—forming clear differentiated positioning from existing literature. First, at the conceptual innovation level, this paper constructs a multi-dimensional measurement system for new quality productivity, representing an important extension of existing productivity research paradigms. Existing literature primarily employs traditional indicators such as total factor productivity (TFP) or labor productivity to measure enterprise production efficiency. While these indicators can reflect input-output relationships, they struggle to capture the qualitative characteristics of contemporary economic development. Building upon dual-factor productivity theory, this paper defines new quality productivity as a comprehensive productivity form integrating technological innovation, human capital, digital transformation, and green development, constructing a measurement system across four dimensions: human capital quality (proportion of high-skilled employees, employees with master’s degrees or above), physical capital investment (R&D intensity, equipment upgrades), technological innovation capability (patent quantity, innovation efficiency), and operational efficiency (total factor productivity, ESG performance). This measurement system not only clearly distinguishes the connotative differences between new quality productivity and traditional TFP—the former emphasizes innovation-driven, green, and sustainable development while the latter focuses on quantitative assessment of production efficiency—but also provides more comprehensive analytical tools in extension, establishing a solid conceptual foundation for related research. Unlike Feng et al. (2024) and Yang et al. (2025), who focus on theoretical elaboration of new quality productivity, this paper advances one step further by proposing an operational measurement scheme and conducting large-sample empirical testing. Second, at the theoretical contribution level, this paper proposes a three-dimensional transmission mechanism framework of “capability-willingness-action,” systematically elucidating the internal logic of how government procurement influences enterprise new quality productivity—a first in existing literature. Current research on government procurement effects primarily identifies single or partial transmission pathways such as demand-pull effects (Georghiou et al., 2014), certification effects (Dai et al., 2021), and financing constraint alleviation effects (Cappelletti et al., 2024), lacking systematic theoretical construction of how multiple mechanisms work synergistically. This paper innovatively integrates these scattered transmission pathways into a unified analytical framework: the capability dimension focuses on how government procurement enhances firms’ resource acquisition capability by alleviating financing constraints and reducing debt costs; the willingness dimension analyzes how government procurement strengthens firms’ strategic willingness by reducing uncertainty and incentivizing digital transformation; the action dimension explores how government procurement promotes concrete enterprise practices in R&D investment, talent cultivation, and innovation output by setting technical standards. This framework not only reveals the independent effects of each mechanism but more importantly clarifies their internal connections: capability is the foundation, creating possibility for willingness formation; willingness is the motivation, providing intrinsic incentives for action implementation; action is the manifestation, transforming capability and willingness into observable productivity improvements. This progressive theoretical logic offers greater explanatory power than existing literature’s parallel mechanism analyses, providing new theoretical perspectives for understanding the micro-mechanisms of demand-side policies. Third, at the empirical findings level, this paper not only validates the significant promoting effect of government procurement on enterprise new quality productivity but more importantly reveals the heterogeneous characteristics of this effect across different enterprise groups and development stages, providing crucial evidence for precise policy design. Unlike Wu and Liu (2020), who mainly focus on innovation output), and Stojčić et al. (2020), who concentrate on the synergistic effects of fiscal support and procurement, this paper systematically examines the moderating effects across three dimensions—ownership nature, internal control quality, and lifecycle stage—finding: (1) SOEs’ procurement responsiveness (0.077) significantly exceeds non-SOEs (0.035), reflecting not only institutional background differences but also revealing the comprehensive effects of information advantages, strategic alignment, and resource allocation capability; (2) internal control quality plays a critical moderating role, with firms without control deficiencies showing significant positive effects (0.072) while those with deficiencies exhibit negative effects (−0.020), emphasizing the foundational role of governance capability in policy effect realization; (3) the effect for growth-stage firms (0.074) far exceeds that of mature-stage (0.027) and decline-stage (0.025) firms, revealing an optimal action window for government procurement policy. These heterogeneity findings enrich Hoekman and Sanfilippo’s (2020) understanding of firm size and industry differences, providing more nuanced empirical evidence for understanding the boundary conditions of government procurement effects. Fourth, at the policy implications level, this paper’s research results provide crucial theoretical support and practical guidance for optimizing government procurement policy design and leveraging the role of demand-side policies in innovation-driven development, with implications transcending the general recommendations of existing literature. Specifically, based on the “capability-willingness-action” framework and heterogeneity analysis results, this paper proposes differentiated policy recommendations: (1) in policy design, a multi-tiered procurement support system should be constructed, adopting differentiated strategies for enterprises at different development stages with emphasis on growth-stage firms to maximize policy effectiveness; (2) in implementation mechanisms, fair and transparent competitive environments should be established, reducing entry barriers for non-SOEs while strengthening examination of internal control quality in supplier selection to ensure resource allocation to enterprises capable of effectively transforming resources into innovation capabilities; (3) in supporting policies, government procurement should be organically integrated with financing support, digital transformation incentives, and talent cultivation programs to form policy synergy. These targeted and operational policy recommendations provide important references for promoting high-quality development and accelerating new quality productivity formation.
Literature Review and Research Hypotheses
Literature Review
Government procurement as a demand-side policy tool has emerged as a significant focus in international research examining its role in promoting enterprise innovation and productivity enhancement. Existing studies have validated the positive impacts of government procurement on firm performance from multiple dimensions while exploring underlying mechanisms. From the innovation effects perspective, existing research has established a relatively systematic theoretical framework. Government procurement promotes enterprise innovation through two core mechanisms: demand-pull and signaling effects. The demand-pull mechanism provides stable market expectations and revenue guarantees for enterprise technological innovation, while the signaling effect improves firms’ financing environment by alleviating information asymmetry (Dai et al., 2021; Ghisetti, 2017). Further research reveals that the innovation-promoting effects of government procurement exhibit significant heterogeneity across different enterprise groups. Particularly for small and young firms facing financing constraints, the certification effect plays a more prominent role (Cappelletti et al., 2024). These findings reveal the unique advantage of government procurement as a demand-side policy tool: it can simultaneously address both the funding bottleneck and market uncertainty of enterprise innovation through market-based mechanisms.
From the efficiency and competitive effects perspective, the impact of government procurement on firm performance presents more complex patterns. Research challenges the traditional assumption that procurement discretion necessarily leads to corruption and efficiency losses, finding that moderate flexibility may actually improve procurement performance (Coviello et al., 2018). Cross-country comparative studies demonstrate that innovation-oriented public procurement generates significant additionality effects on firm innovation output, particularly when firms simultaneously receive both financial support and innovation procurement opportunities (Stojčić et al., 2020). In the globalization context, domestic firms, especially small manufacturing enterprises, experience more significant performance improvements from procurement participation (Hoekman & Sanfilippo, 2020), providing important insights for developing countries to cultivate indigenous innovation capabilities through government procurement.
Despite the accumulation of relatively rich theoretical and empirical evidence on government procurement effects, three research gaps remain that warrant deeper exploration. First, from the research subject perspective, existing studies primarily focus on traditional enterprise innovation indicators and financial performance, lacking systematic theoretical construction and empirical analysis of new quality productivity as a productivity form embodying contemporary development characteristics. New quality productivity differs from traditional total factor productivity (TFP) in that it emphasizes the inherent requirements of innovation-driven, green, and sustainable development, necessitating a comprehensive measurement system encompassing multiple dimensions including human capital quality, technological innovation capability, digital transformation degree, and ESG performance. Second, from the mechanism analysis perspective, although transmission pathways such as demand-pull effects, certification effects, and financing constraint alleviation have been identified, the internal mechanisms through which government procurement enhances new quality productivity by influencing enterprise digital transformation willingness, risk-taking spirit, and talent structure optimization remain unclear. These mechanisms involve deep-level issues such as enterprise strategic decision-making, organizational change, and capability building, requiring systematic examination within the “capability-willingness-action” analytical framework. Third, from the heterogeneity analysis perspective, although existing research considers differences across dimensions such as firm size, ownership nature, and industry characteristics, studies examining how factors such as enterprise lifecycle stages, internal control governance quality, and regional development levels moderate government procurement effects remain limited, constraining theoretical guidance for precise government procurement policy design and differentiated implementation. Building upon these research gaps, this study constructs a comprehensive measurement system encompassing multiple dimensions including human capital, physical capital, technological capabilities, and operational efficiency based on new quality productivity theoretical connotations. Through a progressive logical framework of “capability-willingness-action,” we systematically examine government procurement’s impact on enterprise new quality productivity and its underlying mechanisms, while conducting heterogeneity analysis across multiple levels including firm characteristics and regional environments, aiming to provide richer theoretical foundations and empirical support for improving government procurement policy systems and promoting high-quality development.
Research Hypotheses
In the contemporary context where new quality productivity has emerged as the core driving force for high-quality development (Y. Liu & He, 2024; Z. Liu et al., 2025), traditional productivity growth models relying on capital and labor inputs are facing increasingly severe developmental bottlenecks. This predicament has prompted governments to position innovation-driven development strategies more prominently, particularly intensifying policy support in strategic emerging industries, digital industries, and future industries that constitute the core components of new quality productivity (M. Zheng et al., 2025). Technological progress, serving as a crucial benchmark for measuring productivity development levels (Griffith et al., 2004), has become the core driving factor of advanced productivity through its innovative applications in emerging sectors such as the digital economy and green economy, playing a decisive role in shaping the development pattern of new quality productivity. These technological innovations not only generate intelligent labor to meet diverse societal needs but also redefine and expand the functional boundaries of labor tools, thereby compelling enterprises to proactively enhance employee skills and overall workforce quality. Within this developmental logic framework, government procurement, as an important state-led economic activity (Križić, 2021), plays a pivotal catalytic role in supporting enterprises’ achievement of digitalization, informatization, intelligentization, and green transformation by influencing domestic market supply-demand dynamics, thereby promoting the overall leap of enterprises’ new quality productivity and enabling them to better adapt to continuously evolving technological landscapes and contribute more effectively to broader economic and social development objectives. Based on the logic of supply-demand theory and innovation diffusion theory (Huang et al., 2023; Wonglimpiyarat & Yuberk, 2005), government procurement creates stable market demand, providing clear directional guidance and revenue guarantees for enterprise technological innovation, thereby stimulating sustained investment and innovation activity in new technologies, products, and services.
Internal innovation and development activities within enterprises require substantial financial investment as a foundational guarantee. However, expected returns from investments in human and material innovation factors often exhibit high uncertainty and are typically accompanied by extended payback periods, making financial resources more critical than other production factors. According to signaling theory and information asymmetry theory (Connelly et al., 2011; Mavlanova et al., 2012), enterprises that secure government contracts or are listed in government procurement catalogs are generally perceived by financial institutions as more trustworthy partners due to their verified technological capabilities, strong corporate reputation, and promising market prospects. Government procurement orders function as credible quality signals that effectively alleviate information asymmetry problems between banks and other financial institutions and enterprises (Lei & He, 2025), significantly improving the likelihood of enterprises being classified as reliable lending candidates, thereby effectively addressing key challenges such as financing difficulties and high costs. Furthermore, certain government procurement programs provide direct credit endorsements for enterprises, reduce information asymmetry among various parties through digital platforms, and introduce innovative financial products along with risk compensation mechanisms. This credit enhancement effect not only reduces enterprises’ financing costs but, more importantly, increases the scale of funds enterprises can access, providing adequate financial guarantees for their investments in core elements of new quality productivity such as R&D investment, talent recruitment, and equipment upgrades.
Based on prospect theory in behavioral economics and organizational learning theory (Barberis, 2013; Bilan et al., 2020), enterprises’ investment decisions and innovation behaviors depend not only on objective resource constraints but are also profoundly influenced by management’s subjective expectations of future market environments and risk perceptions. The sustained growth in government procurement’s proportion of GDP, through employing diversified policy tool combinations and coordinating with other supporting policies, plays a crucial role in maintaining product market stability, reducing supply chain risks (Hamilton, 2022), and establishing favorable and predictable business expectations. This stable market environment effectively reduces uncertainty faced by enterprises (Cheng et al., 2025; Uzkurt et al., 2012). Particularly during the commercialization stage of technological innovation, market guarantees provided by government procurement reduce enterprises’ concerns about innovation failure risks, enhance enterprises’ willingness and capacity to undertake risks, and cultivate positive entrepreneurial spirit. Simultaneously, as digital technologies deeply penetrate various economic sectors, government procurement policies increasingly emphasize digitalization and intelligentization requirements. This policy orientation provides clear incentive signals for enterprise digital transformation, stimulating enterprises’ investment willingness and transformation momentum in emerging technology fields such as data analytics, cloud computing, and artificial intelligence.
According to resource-based theory and dynamic capability theory (Alvarez & Busenitz, 2001; Chowdhury & Quaddus, 2017), the formation of enterprises’ new quality productivity requires realization through specific resource allocation behaviors and capability-building activities, including increased R&D investment, recruitment and cultivation of high-skilled talent, output of technological innovation achievements, and enhancement of green sustainable development capabilities. Government procurement guides enterprises to direct resources toward areas aligned with new quality productivity development by setting specific technical standards, quality requirements, and green standards. At the human capital level, government procurement projects often require suppliers to possess specific technological capabilities and professional personnel configurations, prompting enterprises to increase investment in recruiting and training high-skilled digital workers, particularly knowledge-based employees and R&D personnel proficient in digital technologies (Li et al., 2024). At the technological innovation level, government procurement prioritizes products with high technological content and innovative features, directly incentivizing enterprises to increase R&D investment and improve patent applications, especially invention patents, thereby enhancing innovation efficiency in converting R&D investments into tangible technological outcomes (Hirshleifer et al., 2013; C.-H. Wang et al., 2013). At the sustainable development level, procurement policies emphasizing green development encourage enterprises to invest in green technologies and sustainable production practices by favoring products with high environmental standards, enhance ESG performance, and promote comprehensive enterprise development in environmental protection, social responsibility, and corporate governance.
Figure 1 presents the complete theoretical framework of how government procurement influences enterprises’ new quality productivity. This framework is constructed based on the progressive logic of “capability-willingness-action,” clearly depicting three interconnected yet distinctive transmission pathways. Specifically, the capability dimension focuses on how government procurement enhances firms’ resource acquisition capability by alleviating financing constraints and reducing debt financing costs, establishing a solid financial foundation for new quality productivity development. The willingness dimension emphasizes how government procurement strengthens firms’ strategic willingness by boosting development confidence, reducing uncertainty perception, and incentivizing digital transformation, making enterprises more willing to commit to long-term investments in innovation and transformation. The action dimension examines how government procurement promotes concrete enterprise practices, including increased R&D investment, high-skilled talent cultivation, and innovation output improvement, by setting technical standards and quality requirements. These three dimensions form a complete chain from foundational capability building to strategic willingness cultivation, and finally to concrete action implementation. As illustrated in Figure 1, these three transmission channels converge to facilitate the enhancement of enterprises’ new quality productivity, ultimately contributing to high-tech development, high-efficiency development, and high-quality development outcomes. It is particularly important to note that these three dimensions are not mutually independent but exhibit inherent logical connections: capability enhancement creates the possibility for willingness formation, willingness strengthening provides motivation for action implementation, and action effectiveness further consolidates capability and reinforces willingness, thereby forming a virtuous cycle. This theoretical architecture provides not only a systematic analytical framework for understanding the multidimensional impacts of government procurement policy but also clear pathways for subsequent empirical testing.
Theoretical framework.
Research Design
Model Design and Variable Description
Baseline Regression Models
To empirically assess the impact of government procurement on the development of new quality productivity within enterprises, this study constructs the following baseline regression model that addresses potential unobserved firm heterogeneity:
The inclusion of firm fixed effects (
The dependent variable, NPRO, represents the new quality productivity of enterprises. Building upon established productivity measurement frameworks while incorporating contemporary perspectives on innovation-driven growth (Lei, 2024b), this study adopts a comprehensive indicator system to capture the multidimensional nature of new quality productivity (Song et al., 2024). Following the approach based on dual-factor productivity theory, the measurement system captures the synergistic effects of activated human labor, physical capital, and technological tools in the production process.
The indicator system employs a hierarchical structure with entropy-weighted aggregation to ensure balanced representation across dimensions. The first tier encompasses human capital factors, including R&D personnel compensation ratios, R&D personnel proportions, and high-education employee ratios, reflecting the knowledge intensity and skill premium that characterize modern productive activities. The second tier addresses physical capital infrastructure, incorporating fixed asset ratios and manufacturing cost proportions that capture the tangible foundation of productive capacity. The third tier focuses on technological capabilities, encompassing R&D direct investment ratios, depreciation and amortization ratios, rental expense ratios, and intangible asset proportions, while the fourth tier includes operational efficiency metrics such as total asset turnover rates and equity multiplier reciprocals as financial risk management indicators.
To address concerns regarding the extreme values observed in the NQP distribution, the study implements several validation procedures. Cross-correlation analysis with established total factor productivity measures confirms the construct validity of the NQP index. Additionally, robustness testing with logarithmic and rank-transformed versions of the dependent variable ensures that findings are not driven by distributional extremes. The scaling factor of 1,000 applied to the final index enhances coefficient interpretability while maintaining the underlying relationships within the data structure.
The explanatory variable, PROC, represents the firm’s acquisition of government procurement orders, measured by taking the natural logarithm of total procurement amounts with one added to ensure non-negativity. This transformation addresses the natural skewness in procurement data while maintaining the economic interpretation of percentage changes in response to policy interventions (Lei & He, 2025; Wu & Liu, 2020).
The control variables encompass fundamental firm characteristics that theory suggests may influence both procurement outcomes and productivity development. These include firm age and its quadratic term to capture lifecycle effects, firm size reflecting operational scale and market presence, financial performance indicators including return on equity and revenue growth, cash holdings representing financial flexibility, governance structure variables including board independence and ownership concentration, leadership structure captured through CEO-chairman duality, audit quality proxied by Big Four engagement, and ownership type distinguishing state-owned from private enterprises. This comprehensive control structure helps isolate the specific impact of government procurement from other organizational and strategic factors.
The error term
Dynamic Specification and Timing Analysis
To address concerns about the timing of procurement effects and potential anticipation effects, this study implements an event-study specification around the first receipt of government procurement orders:
where
Mediation Mechanism Analysis Framework
To explore the underlying mechanisms through which government procurement influences enterprises’ new quality productivity, this study constructs a mediation effect analysis model that systematically examines the transmission pathways from three dimensions:
where
The first pathway is the capability enhancement mechanism, which examines how government procurement strengthens enterprises’ ability to improve new quality productivity by alleviating financing constraints and reducing financing costs. This study employs the SA index (SA_index) to measure the degree of financing constraints faced by enterprises, which is constructed based on exogenous firm variables such as size and age and effectively reflects financing difficulties (Bates et al., 2018; Zaiane et al., 2025). Additionally, a debt financing cost indicator (Cost) is constructed to measure changes in enterprise financing costs (J. Zheng et al., 2013).
The second pathway is the willingness strengthening mechanism, which analyzes how government procurement influences enterprises’ subjective willingness and attitudes toward improving new quality productivity. This includes: risk-taking spirit (Adj_Roa), measured by the difference between enterprise-adjusted ROA and industry average levels (He et al., 2019; Yu et al., 2013); uncertainty perception (N_keys), extracted through text mining techniques to capture the frequency of uncertainty-related terms in corporate annual reports (Y. Chen et al., 2025; Zhang et al., 2025); and digital transformation willingness (Digitization), measured by the frequency of digitization-related terms in annual reports (Dong & Jiang, 2024; S. Wang et al., 2024).
The third pathway is the action incentivization mechanism, which examines how government procurement motivates enterprises to take concrete actions to improve new quality productivity. Key indicators include: R&D investment intensity (RD), measured as the ratio of annual R&D expenditure to total assets; talent development, reflected through the proportion of employees with master’s degrees or above (MasterD) and the proportion of technical staff (TechnicianN); innovation output effectiveness, assessed using the natural logarithm of total patent applications plus one (Patent), the natural logarithm of invention patent applications plus one (Patent_Inv), and innovation efficiency (InnoEff; Lei & Xu, 2025a, 2025b); and ESG performance (ESG_Score), measured using comprehensive ESG rating data to evaluate enterprise performance in green development and sustainable operations (Duan et al., 2025; Jin & Lei, 2023; Lei, 2024a).
This analytical framework embodies a progressive logic from “capability-willingness-action,” comprehensively revealing the internal mechanisms through which government procurement promotes enterprise new quality productivity improvement. Through systematic examination of mediation mechanisms at these three levels, this study provides deep insights into the transmission pathways and operational boundaries of government procurement policy effects.
Sample Sources and Data Construction
This study examines A-share listed companies receiving government procurement orders from 2015 to 2022, utilizing data from multiple authoritative sources to construct a comprehensive analytical dataset. Government procurement data originate from the China Government Procurement Network, which provides systematic coverage of procurement activities following the 2015 implementation of standardized disclosure requirements. Firm-level financial, governance, and operational data are extracted from the CSMAR database, ensuring consistency with established research protocols in Chinese capital markets.
The sample construction process addresses several methodological concerns raised regarding data quality and representativeness. Regional coverage restrictions are addressed by focusing on areas with complete procurement data availability, while financial sector firms are excluded due to their distinct regulatory environment and operational characteristics that may confound productivity measurements. Firms with special treatment status, delisting events, or substantial missing data are removed to ensure analytical consistency and reliability.
Data matching between procurement records and firm identities employs a systematic multi-stage verification process that addresses potential ambiguities in entity names and organizational structures. The initial algorithmic matching stage utilizes both exact string matching and fuzzy matching techniques to accommodate variations in entity name formatting and potential typographical differences, achieving approximately 89% of matches with confidence scores above 0.85. The secondary verification stage involves manual review of uncertain matches where similarity scores fall between 0.65 and 0.85, accounting for approximately 8% of the total sample. The final validation stage cross-references matched entities using unified social credit codes, business registration addresses, and key personnel information to confirm match accuracy.
Quality assurance procedures include systematic checks for data completeness, temporal consistency, and cross-validation against external sources. Missing data patterns are analyzed to ensure that sample attrition does not introduce systematic bias, while outlier detection procedures identify and flag extreme values for additional scrutiny. The matching and validation process is summarized in Table 1, which demonstrates the progressive refinement of the sample through each verification stage.
Data Construction and Validation Summary.
The final dataset comprises 4,445 unique firms contributing 23,988 firm-year observations, providing sufficient variation across firms, time periods, and procurement experiences to support robust statistical inference. Table 2 presents the definitions and calculation methods for all variables employed in the empirical analysis.
Definition and Calculation Method of the Main Variables.
Results of Descriptive Statistical Analysis
Table 3 presents the results of the descriptive statistical analysis for the key variables. The minimum, maximum, and mean values for NPRO are 0.0464402, 804.4927, and 5.53901, respectively. These values indicate substantial variation in new quality productivity across firms, reflecting differences in their adoption of advanced technologies and improvements in production processes. Similarly, the minimum, maximum, and mean values for PROC are 0, 18.50892, and 0.5423625, respectively, suggesting considerable variation in the level of government procurement support received by firms. The differences in the values for SA_index and Cost further highlight the differing levels of financing difficulties faced by enterprises. The significant variation between the maximum and minimum values for Adj_Roa, N_keys, and Digitization points to diverse firm strategies and attitudes toward their future development, especially in areas such as digital transformation and upgrading. The proportions of RD, TechnicianN, and MasterD also reveal substantial disparities in firms’ investments in research and development and their ability to leverage highly skilled talent. The variations in Patent, Patent_Inv, and InnoEff reflect significant differences in firms’ innovation outputs and the extent to which they focus on innovation-driven growth. For ESG_Score, the minimum, maximum, and mean values are 1, 8, and 4, respectively. This indicates a wide range of scores across firms, suggesting that while some firms excel in areas like environmental sustainability, corporate governance, and social responsibility, others still have substantial room for improvement. Additionally, other control variables show considerable variation across firms and over time, confirming that the sample is well-distributed and representative of different firm characteristics and market conditions.
Descriptive Statistical Analysis of the Main Variables.
Empirical Results and Analysis
Baseline Regression Analysis Results
Table 4 presents the progression of model specifications that collectively address concerns about omitted variable bias and identification strategy. The baseline specification without fixed effects (Column 1) shows a strong positive association between government procurement and new quality productivity. However, the inclusion of firm fixed effects in the preferred specification (Column 4) provides more credible causal inference by controlling for time-invariant firm characteristics that may influence both procurement success and productivity outcomes. The coefficient on government procurement in the firm fixed effects specification (0.045) is economically meaningful and statistically reliable at the 5% significance level. These findings provide initial support for Hypothesis 1, suggesting that government procurement contributes positively to enterprise productivity development through mechanisms that extend beyond simple demand effects. The firm fixed effects specification ensures that this relationship reflects within-firm variation over time rather than cross-sectional differences in firm characteristics.
Enhanced Empirical Analysis Results.
Note. Standard errors clustered at firm level in parentheses.
p < .05. ***p < .01.
The baseline regression results demonstrate a significant positive impact of government procurement on enterprises’ new quality productivity, with deep economic logic underlying this finding. Essentially, government procurement as a demand-side policy tool operates through multiple interconnected micro-mechanisms. First, government procurement orders provide stable and predictable market demand, reducing the risk of enterprise innovation investment and making firms more willing to undertake long-term, strategic technology R&D and human capital investments. Second, obtaining government procurement contracts itself serves as a quality signal, conveying positive information about enterprise technological capabilities and reputation to the market. This signaling effect not only improves firms’ negotiation positions in capital and labor markets but also enhances their bargaining power in industrial chains. Third, government procurement is often accompanied by technical standards and quality requirements, which objectively guide enterprise technology upgrading and drive firms to continuously enhance R&D intensity, optimize talent structures, and accelerate digital transformation.
The coefficient patterns of control variables also provide valuable insights. For instance, the significantly positive coefficient of cash flow (Cash) indicates that abundant internal funds remain crucial for enterprises to enhance new quality productivity, consistent with financing constraint theory predictions. The significantly positive coefficient of state-owned enterprises (SOE) may reflect the institutional advantages of SOEs in accessing innovation resources and undertaking technological risks, as well as the strong alignment between government procurement policies and SOE strategic objectives. The non-linear effect of firm age (Age) suggests that experience and resources accumulated during firm growth contribute to new quality productivity enhancement, but this effect may marginally decrease after firms enter maturity. These control variable patterns further confirm that new quality productivity enhancement is a complex process involving multiple factors, with government procurement playing a crucial catalytic role.
Identification Strategy and Robustness Analysis
Dynamic Timing Analysis
To further validate the causal interpretation of procurement effects and address concerns about potential anticipation effects or confounding trends, this study conducts an event-study analysis examining productivity changes around firms’ first receipt of government procurement contracts. This temporal analysis provides crucial evidence for establishing causality by testing whether firms exhibit differential productivity trends prior to procurement receipt and by tracing the evolution of treatment effects over multiple post-treatment periods.
The event study specification examines productivity changes from 2 years before to 3 years after the initial procurement award, focusing on firms that receive their first government contract during the sample period. The results, presented in Table 5, provide strong support for the parallel trends assumption essential for causal inference. The pre-treatment coefficients for periods t−2 and t−1 show statistically insignificant values of −0.012 and 0.008 respectively, indicating that firms eventually receiving procurement contracts do not exhibit differential productivity trends prior to treatment. This pattern suggests that procurement awards are not merely identifying firms already on improving productivity trajectories, but rather represent genuine policy interventions that create productivity improvements.
Event Study Analysis of First Procurement Receipt.
Note. Sample includes firms receiving their first procurement contract during 2016 to 2020 to allow for complete pre- and post-treatment observation windows. Standard errors clustered at firm level in parentheses.
p < .1. **p < .05. ***p < .01.
The treatment period coefficient (t = 0) of 0.041 demonstrates immediate productivity improvements following procurement receipt, while the post-treatment analysis reveals sustained and even strengthening effects over time. The coefficients increase to 0.056 at t + 1 before gradually declining to 0.048 at t + 2 and 0.038 at t + 3, maintaining statistical significance throughout the observation period. This temporal pattern indicates that procurement effects represent sustained productivity improvements rather than temporary demand shocks, with benefits persisting well beyond individual contract durations.
Enhanced Robustness Testing Framework
The robustness testing framework addresses several critical concerns regarding the stability and generalizability of the baseline findings. Temporal robustness is examined through subsample analysis focusing on the pre-pandemic period (2015–2019), which isolates procurement effects from extraordinary economic disruptions that began in 2020. This approach ensures that documented relationships reflect normal economic conditions rather than crisis-period dynamics that might confound the causal interpretation of procurement policies.
Measurement robustness is addressed through alternative specifications of both dependent and independent variables. For the dependent variable, established total factor productivity measures (TFP_LP and TFP_OLS) provide benchmarks against which the new quality productivity construct can be validated. For the independent variable, both intensive margin effects (procurement amounts) and extensive margin effects (procurement receipt indicators) are examined to ensure robustness across different conceptualizations of government procurement exposure.
Sample selection concerns are addressed through Heckman two-stage estimation that explicitly models both the procurement award process and subsequent productivity outcomes. This approach acknowledges that procurement receipt may depend on firm characteristics that also influence productivity development, requiring careful attention to selection mechanisms to obtain unbiased treatment effect estimates. Table 6 presents the comprehensive robustness analysis results across these alternative specifications.
Comprehensive Robustness Analysis.
Note. Standard errors clustered at firm level in parentheses.
p < .1. **p < .05. ***p < .01.
The robustness results demonstrate consistent support for the main findings across alternative specifications and samples. Column (1) examines the pre-pandemic subsample, showing a procurement coefficient of 0.038 that maintains statistical significance, confirming that results are not driven by pandemic-period anomalies. Column (2) uses the count of procurement contracts as an alternative measure, yielding a positive coefficient of 0.089, while Column (3) employs a binary procurement indicator with a coefficient of 0.167. Both alternative measures maintain statistical significance, demonstrating robustness to different conceptualizations of procurement exposure.
Columns (4) and (5) validate the NQP construct using established TFP measures as dependent variables. The TFP_LP and TFP_OLS specifications show positive coefficients of 0.006 and 0.005 respectively, confirming that procurement effects are consistent with established productivity measures despite different scaling. Columns (6) and (7) present the Heckman two-stage estimation results. The first stage (Column 6) shows a strong relationship between the county-level procurement average and individual firm procurement probability, with an F-statistic well above conventional thresholds. The second stage (Column 7) maintains the positive procurement effect while controlling for selection bias through the inverse Mills ratio.
Building upon these core robustness tests, additional validation procedures address concerns about spurious correlations, distributional sensitivity, and alternative error correlation structures. These extended tests provide further assurance that the documented relationships represent genuine causal effects rather than statistical artifacts or methodological limitations.
Placebo testing examines whether randomly assigned procurement timing produces similar results, thereby testing for spurious correlations in the data. The placebo procedure randomizes procurement receipt timing within industry-year cells while maintaining the overall distribution of procurement exposure across the sample. As shown in Column 1 of Table 7, this randomization produces a statistically insignificant coefficient of 0.003, supporting the interpretation that observed effects reflect genuine treatment impacts rather than coincidental timing patterns.
Extended Robustness Analysis.
Note. Placebo test randomizes procurement timing within industry-year cells. Winsorized estimates trim extreme values at 1% level. Q25 and Q75 represent quantile regression estimates. Two-way clustering allows correlation within firms over time and within years across firms.
p < .1. **p < .05.
Outlier sensitivity analysis addresses concerns about whether results are driven by extreme observations in the productivity distribution. Winsorization at the 1% level produces a coefficient of 0.043 (Column 2), maintaining both statistical significance and economic magnitude similar to the baseline estimate. Quantile regression analysis reveals consistent positive effects across different points in the productivity distribution, with coefficients of 0.034 at the 25th percentile and 0.052 at the 75th percentile (Columns 3 and 4), indicating that procurement effects are broadly distributed rather than concentrated among extreme performers.
Alternative clustering approaches examine sensitivity to different assumptions about error correlation structures. Two-way clustering by firm and year, shown in Column 5, produces slightly larger standard errors but maintains statistical significance at conventional levels. This approach allows for more flexible correlation patterns both within firms over time and within years across firms, providing conservative inference that accounts for potential common shocks affecting multiple firms simultaneously.
The comprehensive robustness testing framework, incorporating both the core tests presented in Table 6 and the extended validation procedures shown in Table 7, demonstrates that the baseline findings are not sensitive to alternative specifications, measurement approaches, distributional assumptions, or error correlation structures. The consistent patterns across these diverse tests strengthen confidence in the causal interpretation of government procurement effects on new quality productivity development.
Instrumental Variables and Causal Identification
The instrumental variables approach addresses endogeneity concerns through carefully constructed instruments that satisfy both relevance and exclusion restrictions. These instruments exploit regional and industry-level variation in procurement activity that plausibly affects individual firm procurement probability without directly influencing firm-specific productivity outcomes.
The leave-one-out construction of instruments ensures that mechanical correlation between firm outcomes and the instruments is eliminated. Regional average procurement intensity (LogAveragePur_-i) excludes the focal firm’s procurement from the regional average, while county-level procurement probability (AveragePurR_-i) similarly excludes the focal firm’s experience. These instruments capture policy-driven variation in procurement availability while maintaining exogeneity with respect to individual firm productivity shocks.
Enhanced instrument validation includes comprehensive diagnostic testing. First-stage F-statistics confirm instrument relevance, with values well above conventional thresholds for weak instrument concerns. The Kleibergen-Paap rank statistic provides additional evidence of instrument identification, while Hansen’s J-test examines over-identification restrictions when multiple instruments are employed. Table 8 reports the instrumental variables analysis with enhanced diagnostics across different specifications.
Instrumental Variables Analysis With Enhanced Diagnostics.
Note. Standard errors clustered at firm level in parentheses.
p < .05. ***p < .01.
The instrumental variables results provide strong evidence for causal interpretation of the procurement-productivity relationship. Column (1) presents the lagged procurement specification, showing a coefficient of 0.048 that remains statistically significant, addressing reverse causality concerns through temporal separation. Columns (2) and (3) present the first and second stages of the IV estimation using regional procurement intensity as the instrument. The first stage shows a strong coefficient of 0.447 with an F-statistic of 376.2, well above the threshold of 10 for strong instruments. The second stage maintains a positive procurement effect of 0.092.
Columns (4) and (5) employ county-level procurement probability as an alternative instrument. The first stage coefficient of 4.523 with an F-statistic of 248.7 confirms instrument strength, while the second stage yields a coefficient of 0.108. The Hansen J-test p-value of .128 fails to reject the null hypothesis of instrument exogeneity, supporting the validity of the over-identification restrictions. These results collectively demonstrate that the procurement-productivity relationship withstands rigorous causal identification procedures.
Mechanism Analysis
Financial Constraint Alleviation Mechanisms
The analysis of financial constraint mechanisms reveals how government procurement enhances firm capacity for productivity improvements through multiple financial channels. Government procurement awards function as credible signals to external capital markets, providing tangible evidence of firm capabilities and future cash flow prospects that reduce information asymmetries with potential lenders. This signaling effect operates through both direct contract values and the implicit government endorsement associated with successful procurement participation.
The SA index results demonstrate that procurement awards significantly reduce financing constraints faced by enterprises. Table 9 presents the financial constraint mechanism analysis results. The negative coefficient of −0.003 indicates that firms receiving government procurement contracts experience measurable improvements in access to external capital markets. This effect reflects enhanced credibility signaling that reduces perceived default risk among potential lenders. Similarly, the debt financing cost analysis provides complementary evidence through the negative coefficient of −0.001, suggesting that government contracts serve as implicit guarantees that lead to more favorable borrowing terms.
Financial Constraint Mechanism Analysis.
Note. Standard errors clustered at firm level in parentheses.
p < .01.
The empirical results of the financing constraint alleviation mechanism reveal an important financial channel through which government procurement influences enterprise new quality productivity, though the realization process is more complex than surface data suggest. The significant decline in the SA index (coefficient: −0.003) indicates that government procurement contracts indeed improve firms’ external financing environment, but this improvement does not stem from direct government funding support; rather, it is achieved by changing financial institutions’ risk assessments of enterprises. Specifically, government procurement contracts serve as reliable guarantees of future cash flows, reducing banks’ and other financial institutions’ concerns about enterprise default risk, transforming firms from “high-risk borrowers” to “quality clients.” The reduction in debt financing costs (coefficient: −0.001) further confirms this logic—lower interest rates and more lenient credit conditions save enterprises substantial financial expenses, which can be reallocated to new quality productivity enhancement activities such as R&D, talent cultivation, and equipment upgrades. Notably, while the absolute values of coefficients appear small, two factors must be considered in interpretation: first, financial market adjustments exhibit gradual characteristics—the establishment of enterprise credibility and improvement of financing conditions require time accumulation, making dramatic short-term changes difficult to observe; second, new quality productivity enhancement itself is a long-term process, and the marginal alleviation of financing constraints exerts sustained effects over multiple years through compounding, with cumulative impacts far exceeding single-period effects. Moreover, different enterprise types exhibit varying sensitivity to financing constraint alleviation. For small and private enterprises already facing severe financing difficulties, the financing improvement brought by government procurement may become a critical breakthrough for their innovation transformation, while for large enterprises and SOEs with established financing channels, this effect is relatively modest.
Willingness and Behavioral Adaptation Mechanisms
The willingness mechanism analysis examines how government procurement influences firm strategic orientation and risk tolerance. Table 10 reports the willingness mechanism analysis across different behavioral dimensions.
Willingness Mechanism Analysis.
Note. Standard errors clustered at firm level in parentheses.
p < .05. ***p < .01.
Column (1) shows the entrepreneurial spirit indicator with a coefficient of 0.002, suggesting that procurement exposure modestly enhances firm willingness to undertake innovative but uncertain projects. This behavioral change likely reflects reduced downside risk perception when firms have guaranteed government demand, enabling more ambitious productivity improvement strategies. Column (2) presents uncertainty perception results with a coefficient of −0.045, indicating that procurement awards help firms develop more accurate expectations about future market conditions by providing information about policy priorities and demand patterns.
Column (3) demonstrates digital transformation willingness through a coefficient of 0.038, representing a particularly important mechanism in the contemporary economic context. Government procurement increasingly emphasizes digital capabilities, creating incentives for firms to invest in digital infrastructure and capabilities. This positive effect suggests that procurement exposure accelerates firm adoption of digital technologies that are fundamental to new quality productivity development.
Innovation and Implementation Mechanisms
The action mechanism analysis demonstrates how government procurement translates enhanced capabilities and willingness into concrete productivity improvements through multiple implementation channels. Table 11 presents the innovation implementation mechanism analysis across seven key dimensions.
Innovation Implementation Mechanism Analysis.
Note. Standard errors clustered at firm level in parentheses.
p < 0.05. ***p < .01.
Columns (1) through (3) examine input-side mechanisms. Research and development investment shows a coefficient of 0.002, indicating sustained increases in innovation spending following procurement awards. Human capital development operates through both technical staff proportion (coefficient: 0.008) and graduate degree holders (coefficient: 0.002), demonstrating that procurement success facilitates access to high-skilled labor markets essential for implementing advanced technologies.
Columns (4) through (6) analyze innovation output mechanisms. Patent applications show coefficients of 0.028 for total patents and 0.031 for invention patents, while innovation efficiency demonstrates a coefficient of 0.002. These results indicate that procurement-induced investments generate tangible technological advances with improved efficiency in converting research inputs into innovation outputs. Column (7) presents ESG performance improvements with a coefficient of 0.033, reflecting the broader scope of new quality productivity that encompasses environmental sustainability and governance quality alongside technological innovation.
The comprehensive mechanism analysis reveals a coherent sequence through which government procurement enhances new quality productivity. Financial constraint alleviation provides necessary resources, willingness enhancement creates strategic motivation, and implementation mechanisms translate these conditions into measurable productivity improvements. This sequential process supports Hypotheses 2, 3, and 4, demonstrating the multifaceted nature of procurement impacts on firm productivity development.
Heterogeneity Analysis
The heterogeneity analysis examines how procurement effects vary across different firm characteristics and contexts, providing insights into boundary conditions and optimal targeting of procurement policies. Table 12 reports the firm characteristic heterogeneity analysis across ownership, governance, and life cycle dimensions.
Firm Characteristic Heterogeneity Analysis.
Note. Standard errors clustered at firm level in parentheses.
p < .05. ***p < .01.
The ownership heterogeneity analysis results (SOE coefficient: 0.077; non-SOE coefficient: 0.035) reveal significant differences in government procurement effects across different ownership types, reflecting the complexity of resource allocation mechanisms under China’s specific institutional context. SOEs demonstrate stronger procurement responsiveness, which does not simply reflect “policy favoritism” but results from multiple interacting factors. First, SOEs often possess information advantages and relationship network advantages when participating in government procurement, enabling earlier access to procurement demand and technical standards for advance R&D planning and capability preparation. Second, SOEs’ strategic objectives naturally align with government procurement policy orientations, particularly in strategic emerging industries and key technology fields where SOEs bear missions of technological breakthroughs and industrial leadership, with government procurement providing a market-based platform for achieving these objectives. Third, after undertaking government procurement projects, SOEs are more inclined to treat them as long-term strategic investments rather than short-term profit projects, willing to invest more resources in technological innovation and talent cultivation.
In contrast, while non-SOEs possess greater flexibility and efficiency advantages in market competition, they face more prominent entry barriers and information asymmetry problems in government procurement markets, resulting in relatively lower procurement responsiveness. However, this does not mean non-SOEs benefit less from government procurement. In fact, non-SOEs that successfully overcome entry barriers often obtain significant certification effects and financing improvements from government procurement contracts, with potentially higher marginal benefits per procurement contract. Therefore, policymakers should focus on reducing entry barriers for non-SOEs and establishing fairer and more transparent competition mechanisms to fully leverage the comparative advantages of enterprises with different ownership types.
The moderating effect of internal control quality (coefficient for firms without control deficiencies: 0.072; with deficiencies: −0.020) further emphasizes the critical role of governance structure in government procurement effects. Internal control deficiencies not only reflect enterprise management levels but more fundamentally reveal the organizational foundation for whether enterprises can effectively convert external resources into innovation capabilities. Firms with sound internal control mechanisms can better identify technological opportunities, allocate innovation resources, and prevent R&D risks, thereby transforming government procurement contracts into tangible new quality productivity improvements. Conversely, firms with internal control deficiencies may divert resources from procurement contracts to non-productive expenditures or even experience managerial opportunism, leading to negative effects. This finding provides empirical support for governments to emphasize governance quality in supplier selection while reminding enterprises that they must prioritize internal governance capability building while pursuing government procurement opportunities.
The lifecycle heterogeneity findings (growth-stage coefficient: 0.074; mature-stage: 0.027; decline-stage: 0.025) reveal the optimal action window for government procurement policy. Growth-stage firms exhibit the strongest procurement responsiveness because these firms possess both certain technical foundations and market experience while facing dual pressures of resource constraints and market uncertainty—government procurement provides precisely the critical opportunity to break through bottlenecks. Growth-stage firms typically experience crucial periods of technological maturity improvement and market share expansion, with government procurement bringing not only revenue growth but more importantly, providing opportunities for technology validation and brand endorsement, helping firms cross the “valley of death.” Mature-stage firms, despite abundant resources and complete capabilities, may have formed path dependencies and organizational inertia, with the marginal stimulation effect of government procurement being relatively limited. Decline-stage firms face deeper strategic and organizational problems, with simple demand-pull effects unable to fundamentally reverse their downward trends. Therefore, government procurement policy should focus primarily on growth-stage firms, maximizing policy effectiveness through precise identification and targeted support.
Regional heterogeneity analysis provides insights into geographic distribution of procurement effectiveness. Table 13 presents the geographic heterogeneity analysis across different regional categories. Columns (1) and (2) compare firms in direct-administered municipalities and provincial capitals with other regions, showing stronger effects in major administrative centers (coefficient: 0.059) compared to other regions (coefficient: −0.004). Columns (3) through (5) examine regional variation, with central region firms showing particularly strong procurement responsiveness (coefficient: 0.132), likely reflecting regional development policies that emphasize government procurement as an economic development tool.
Geographic Heterogeneity Analysis.
Note. Standard errors clustered at firm level in parentheses.
p < .05.
Digital Procurement and Advanced Mechanisms
The digital procurement analysis extends the main findings by examining how contemporary procurement practices that emphasize digital transformation create additional pathways for productivity enhancement. Table 14 reports the digital procurement mechanism analysis results.
Digital Procurement Mechanism Analysis.
Note. Standard errors clustered at firm level in parentheses.
p < .05. ***p < .01.
Column (1) demonstrates that digital procurement generates substantial productivity improvements with a coefficient of 0.037, operating through mechanisms that extend beyond traditional procurement effects. Column (2) shows the digital transformation mechanism with a coefficient of 1.687, indicating that digital procurement directly incentivizes supplier firms to accelerate their digital transformation processes through demand-side pressure created by government emphasis on digital capabilities. Column (3) presents the human capital mechanism with a coefficient of 0.002, suggesting that firms respond to digital procurement opportunities by enhancing their internal digital leadership capabilities through recruitment and promotion of executives with digital expertise.
These findings demonstrate that procurement policy design can incorporate technology policy objectives, creating synergies between government purchasing and broader economic development goals. The dual mechanism structure suggests that digital procurement policies may generate sustained productivity improvements that extend beyond individual contract periods through both technological adoption and organizational capability development channels.
The empirical evidence from both traditional and digital procurement analyses provides comprehensive support for the theoretical framework developed in this study. Government procurement influences new quality productivity through multiple interconnected mechanisms that operate across different time horizons and organizational levels, with heterogeneity analysis demonstrating systematic variation across firm characteristics and contexts that provides guidance for optimal policy design and implementation.
Research Conclusion and Policy Implications
This study provides comprehensive empirical evidence that government procurement significantly enhances enterprises’ new quality productivity through a multifaceted transmission mechanism that operates across capability, willingness, and action dimensions. The findings demonstrate that procurement policies function as more than simple demand-side interventions, instead serving as catalytic instruments that transform firms’ fundamental operational capabilities and strategic orientations toward innovation-driven development. The robustness of these effects across alternative specifications, instrumental variable approaches, and various temporal frameworks establishes a credible causal relationship between government procurement participation and productivity enhancement. Through systematic analysis of 23,988 firm-year observations spanning 2015 to 2022, the study reveals that the average procurement effect corresponds to a 4.5% improvement in new quality productivity, with heterogeneous impacts that vary systematically across ownership structures, governance quality, lifecycle stages, and regional contexts. The mechanism analysis illuminates three interconnected pathways through which procurement influences productivity: first, by alleviating financing constraints and reducing debt costs, thereby expanding firms’ resource capacity for innovation investments; second, by reducing uncertainty perceptions and enhancing risk-taking willingness, particularly in digital transformation initiatives that are central to contemporary productivity advancement; and third, by incentivizing concrete actions including increased R&D investment, human capital development, and innovation output generation. These mechanisms operate synergistically rather than independently, creating reinforcing effects that amplify the overall productivity impact beyond what would be achieved through any single channel alone.
The policy implications of these findings extend far beyond traditional procurement reform considerations, offering strategic guidance for leveraging government purchasing power as a comprehensive tool for economic transformation and industrial upgrading. Policymakers should recognize that procurement policies can simultaneously address multiple market failures, including information asymmetries in capital markets, coordination problems in technology adoption, and insufficient incentives for long-term innovation investment. The heterogeneity analysis provides crucial insights for targeted policy design, indicating that procurement effectiveness is maximized when directed toward growth-stage firms with strong governance capabilities, particularly in strategic regions where industrial policy objectives align with local development priorities. The particularly strong effects observed among state-owned enterprises suggest that procurement policies can serve as effective instruments for SOE reform, encouraging these firms to enhance productivity through market-oriented mechanisms rather than relying solely on policy protection. However, the differential effects across ownership types also indicate that procurement design should account for varying institutional contexts and capability structures, with policies potentially requiring customization to achieve optimal outcomes across different firm categories. The regional heterogeneity findings underscore the importance of considering geographic disparities in institutional quality and market development when implementing procurement policies, with evidence suggesting that central regions may offer particularly fertile ground for procurement-driven productivity improvements due to their intermediate development status and policy experimentation capacity.
Looking toward future research directions and policy evolution, several critical areas warrant continued investigation and policy attention. The study’s limitations, including its focus on listed companies and the 8-year observation window, suggest that future research should examine procurement effects across broader firm populations and longer time horizons to better understand persistence and lifecycle dynamics of productivity improvements. Additionally, while this study provides evidence for the “capability-willingness-action” transmission framework, future research should investigate additional mechanisms such as supply chain effects, industry spillovers, and competitive dynamics that may amplify or constrain procurement impacts across different market structures. From a policy perspective, the digital procurement analysis reveals promising directions for integrating traditional procurement objectives with emerging technology policy goals, suggesting that government purchasing can serve as a platform for promoting digital transformation, green development, and other strategic priorities simultaneously. However, successful implementation of these integrated approaches requires careful attention to potential conflicts between multiple policy objectives and the development of sophisticated evaluation frameworks that can assess performance across various dimensions. The study’s findings also highlight the importance of complementary policies that address the broader institutional environment within which procurement operates, including financial market development, intellectual property protection, and regulatory quality, all of which influence the effectiveness of procurement as a productivity-enhancing tool. Ultimately, this research demonstrates that government procurement, when thoughtfully designed and strategically implemented, can serve as a powerful catalyst for new quality productivity development, but realizing this potential requires sophisticated understanding of transmission mechanisms, careful attention to heterogeneous effects, and coordination with broader industrial and innovation policy frameworks.
Footnotes
Ethical Considerations
This article does not contain any studies with human or animal participants.
Consent to Participate
Not applicable.
Funding
The author received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
All data will be made available on request.