In recent years, the rapid advancement of smart manufacturing, particularly through the adoption of industrial robots, has transformed production processes globally. As a core component of the Fourth Industrial Revolution, robot technology enhances efficiency, precision, and flexibility in manufacturing, potentially mitigating various enterprise risks. This study investigates whether the application of industrial robots suppresses enterprise bankruptcy risk, drawing on empirical evidence from manufacturing firms. By analyzing data from 2011 to 2019, we explore the mechanisms through which robot technology influences financial stability, including supply chain financing effects and dual innovation effects. The findings indicate that robot adoption significantly reduces bankruptcy risk, with variations across firm types and market conditions. This research contributes to understanding the economic implications of smart manufacturing and offers insights for policymakers and managers aiming to leverage robot technology for sustainable growth.
The proliferation of robot technology in manufacturing aligns with global trends toward automation and digitalization. Industrial robots, characterized by their ability to perform repetitive and complex tasks with high accuracy, have become integral to modern production systems. Their adoption is driven by the need to improve productivity, reduce costs, and enhance competitiveness. However, the impact of robot technology on enterprise risk, particularly bankruptcy risk, remains underexplored. Bankruptcy risk, often a culmination of financial distress, operational inefficiencies, and market volatility, poses significant threats to firm survival. This study addresses this gap by examining how robot technology application affects bankruptcy risk through empirical models and robustness checks.
Prior research on robot technology has primarily focused on labor market outcomes, such as employment effects and wage dynamics. For instance, studies have shown that robot adoption can lead to job displacement in some sectors while creating opportunities in others. Additionally, robot technology has been linked to increased export competitiveness and innovation capabilities. However, limited attention has been paid to its role in managing financial risks. Concurrently, literature on enterprise risk highlights factors like corporate governance, digital transformation, and environmental, social, and governance (ESG) performance as determinants of bankruptcy risk. This study bridges these domains by analyzing robot technology as a key factor in risk mitigation.
Theoretical frameworks suggest that robot technology influences bankruptcy risk through multiple channels. Firstly, by enhancing production efficiency and reducing error rates, robots improve operational stability and profitability. Secondly, robot technology facilitates better information sharing and coordination in supply chains, enabling firms to secure trade credit and alleviate liquidity constraints. Thirdly, robots foster innovation by enabling rapid prototyping and process optimization, strengthening competitive advantages. These mechanisms are formalized in our hypotheses, which posit that robot technology directly and indirectly suppresses bankruptcy risk.
To test these hypotheses, we employ a dataset combining industrial robot penetration data from the International Federation of Robotics (IFR) with financial data from Chinese manufacturing firms listed on the A-share market. The baseline regression model is specified as follows:
$$ OR_{i,t} = \alpha_0 + \alpha_1 EIRP_{i,t} + \alpha_2 X_{i,t} + \delta_i + \varphi_t + \mu_{i,t} $$
where \( OR_{i,t} \) represents the bankruptcy risk of firm \( i \) in year \( t \), measured by the Altman Z-score. \( EIRP_{i,t} \) denotes the industrial robot penetration level, and \( X_{i,t} \) includes control variables such as firm size, leverage, and profitability. Firm and year fixed effects are captured by \( \delta_i \) and \( \varphi_t \), respectively. The Z-score is calculated as:
$$ Zscore = 1.2 \times \frac{Working Capital}{Total Assets} + 1.4 \times \frac{Retained Earnings}{Total Assets} + 3.3 \times \frac{EBIT}{Total Assets} + 0.6 \times \frac{Market Value of Equity}{Total Liabilities} + 0.999 \times \frac{Sales}{Total Assets} $$
Industrial robot penetration is derived from industry-level data and firm-specific employment structures:
$$ PR_{k,t}^{CH} = \frac{MR_{k,t}^{CH}}{L_{k,t=2010}^{CH}} $$
$$ EIRP_{i,k,t} = \frac{PMP_{i,k,t=2011}}{ManuPMP_{t=2011}} \times PR_{k,t}^{CH} $$
where \( PR_{k,t}^{CH} \) is the robot penetration in industry \( k \), \( MR_{k,t}^{CH} \) is the stock of robots, and \( L_{k,t=2010}^{CH} \) is the base-year employment. \( EIRP_{i,k,t} \) is then scaled for regression analysis.
| Variable Category | Variable Name | Symbol | Definition |
|---|---|---|---|
| Dependent Variable | Bankruptcy Risk | OR | Altman Z-score, inversely related to risk |
| Independent Variable | Robot Technology Application | EIRP | Industrial robot penetration level |
| Control Variables | Firm Size | Size | Natural logarithm of total employees |
| Leverage | Lev | Total liabilities divided by total assets | |
| Return on Assets | Roa | Net income divided by total assets | |
| Operating Expense Ratio | Ofee | Operating expenses divided by revenue | |
| Board Size | Board | Natural logarithm of board members plus one | |
| Independent Director Ratio | Indep | Percentage of independent directors on the board |
Descriptive statistics for the key variables are presented in Table 2. The sample consists of 10,743 firm-year observations from 2011 to 2019. The mean bankruptcy risk (OR) is 5.021, with a standard deviation of 5.588, indicating significant variation across firms. The average robot technology application (EIRP) is 0.331, suggesting room for increased adoption. Control variables exhibit distributions consistent with prior studies, supporting the robustness of our data.
| Variable | N | Mean | Median | Std. Dev. | Min | Max |
|---|---|---|---|---|---|---|
| OR | 10743 | 5.021 | 3.239 | 5.588 | 0.103 | 35.07 |
| EIRP | 10743 | 0.331 | 0.122 | 0.597 | 0 | 3.739 |
| Size | 10743 | 22.10 | 21.96 | 1.143 | 19.93 | 25.43 |
| Roa | 10743 | 0.038 | 0.035 | 0.060 | -0.199 | 0.211 |
| Lev | 10743 | 0.411 | 0.404 | 0.197 | 0.051 | 0.886 |
| Board | 10743 | 2.131 | 2.197 | 0.189 | 1.609 | 2.639 |
| Indep | 10743 | 37.34 | 33.33 | 5.324 | 33.33 | 57.14 |
| Ofee | 10743 | 0.173 | 0.139 | 0.122 | 0.021 | 0.628 |
The baseline regression results, shown in Table 3, demonstrate a significant positive coefficient for EIRP, indicating that robot technology application reduces bankruptcy risk. Column (1) reports results without fixed effects, while column (2) includes firm and year fixed effects. The coefficient of EIRP increases and remains statistically significant at the 1% level, confirming the robustness of the relationship. Control variables such as firm size and leverage exhibit expected signs, aligning with economic theory.
| Variable | (1) OR | (2) OR |
|---|---|---|
| EIRP | 0.248** (2.442) | 0.496*** (4.420) |
| Size | -0.627*** (-7.025) | -1.668*** (-10.326) |
| Roa | 11.436*** (7.358) | 8.862*** (7.519) |
| Lev | -13.296*** (-19.158) | -12.135*** (-18.011) |
| Board | -0.433 (-1.086) | -0.628 (-1.362) |
| Indep | 0.003 (0.240) | -0.009 (-0.614) |
| Ofee | 4.640*** (5.218) | 1.063 (0.770) |
| Constant | 23.827*** (11.919) | 47.867*** (12.961) |
| Firm FE | No | Yes |
| Year FE | No | Yes |
| N | 10743 | 10743 |
| Adjusted R² | 0.394 | 0.698 |
To address endogeneity concerns, we employ an instrumental variable approach using U.S. industry-level robot penetration as an instrument. The two-stage least squares (2SLS) results, presented in Table 4, confirm the causal relationship. The first stage shows a strong correlation between the instrument and EIRP, while the second stage indicates a significant effect on bankruptcy risk. Additionally, propensity score matching and higher-order fixed effects models yield consistent results, reinforcing the main findings.
| Variable | (1) 2SLS First Stage EIRP | (2) 2SLS Second Stage OR | (3) PSM OR | (4) High FE OR | (5) High FE OR |
|---|---|---|---|---|---|
| EIRP | 0.207*** (1.677) | 0.368*** (3.191) | 1.090* (1.845) | 0.485*** (2.819) | |
| EIRP_IV | 0.005*** (96.570) | ||||
| Controls | Yes | Yes | Yes | Yes | Yes |
| Firm FE | Yes | Yes | Yes | Yes | Yes |
| Year FE | Yes | Yes | Yes | Yes | Yes |
| N | 6905 | 6905 | 7859 | 10733 | 9842 |
| Adjusted R² | 0.258 | 0.703 | 0.700 | 0.688 |
Robustness checks include replacing core variables, adjusting clustering levels, and altering sample composition. For instance, using digital application scores as an alternative measure of robot technology yields similar results. Lagging the explanatory variable accounts for dynamic effects, showing that the impact of robot technology on bankruptcy risk strengthens over time.
Mechanism tests reveal that robot technology reduces bankruptcy risk through supply chain financing and dual innovation effects. The supply chain financing effect is measured by trade credit (CCF), calculated as the sum of accounts payable, notes payable, and advance receipts divided by total assets. Regression results show a positive relationship between EIRP and CCF, supporting Hypothesis 2. Dual innovation is assessed using patent data: breakthrough innovation (BI) is the natural logarithm of invention patents plus one, and incremental innovation (PI) is the natural logarithm of utility and design patents plus one. Both BI and PI are positively associated with EIRP, confirming Hypothesis 3. The mechanism models are specified as:
$$ MV_{i,t} = \beta_0 + \beta_1 EIRP_{i,t} + \beta_2 X_{i,t} + \delta_i + \varphi_t + \mu_{i,t} $$
where \( MV_{i,t} \) represents the mechanism variables (CCF, BI, PI).
| Variable | (1) CCF | (2) BI | (3) PI |
|---|---|---|---|
| EIRP | 0.007*** (3.005) | 0.167*** (5.895) | 0.094*** (2.597) |
| Constant | 0.257*** (3.456) | -6.122*** (-7.071) | -7.253*** (-6.806) |
| Controls | Yes | Yes | Yes |
| Firm FE | Yes | Yes | Yes |
| Year FE | Yes | Yes | Yes |
| N | 10743 | 10500 | 10500 |
| Adjusted R² | 0.836 | 0.711 | 0.730 |
Heterogeneity analysis examines how the effect of robot technology varies by ownership, market pressure, and ecological support. Non-state-owned enterprises (non-SOEs) exhibit a stronger risk-reduction effect due to greater market sensitivity and flexibility. Market pressure, measured by the Herfindahl-Hirschman Index (HHI), shows that higher competition amplifies the benefits of robot technology. Ecological support, assessed through environmental support scores, enhances the risk-mitigating impact of robot adoption. Interaction terms in extended models confirm these findings, as shown in Table 6.
| Variable | (1) OR | (2) OR | (3) OR |
|---|---|---|---|
| EIRP | 0.789*** (5.195) | 0.806*** (5.086) | 0.187* (1.779) |
| EIRP × SOE | -0.720*** (-4.391) | ||
| SOE | 0.425 (1.189) | ||
| EIRP × HHI | -6.121*** (-3.875) | ||
| HHI | -2.271 (-1.259) | ||
| EIRP × EES | 0.016*** (3.061) | ||
| EES | 0.022*** (3.191) | ||
| Constant | 43.261*** (11.960) | 43.735*** (11.719) | 43.520*** (11.707) |
| Controls | Yes | Yes | Yes |
| Firm FE | Yes | Yes | Yes |
| Year FE | Yes | Yes | Yes |
| N | 10543 | 10727 | 10743 |
| Adjusted R² | 0.702 | 0.699 | 0.699 |
Further analysis explores economic consequences, specifically the impact on firm value. Using Tobin’s Q as a proxy, we find that robot technology application increases firm value, and this effect is mediated by reduced bankruptcy risk. The interaction term between EIRP and OR in a moderated regression model is positive and significant, indicating that risk suppression enhances value creation. This aligns with resource-based theory, where efficient risk management frees up resources for growth investments.
| Variable | (1) TobinQ | (2) TobinQ |
|---|---|---|
| EIRP | 0.123*** (3.608) | 0.112*** (2.670) |
| EIRP × OR | 0.046*** (3.161) | |
| OR | 0.206*** (8.540) | |
| Constant | 16.716*** (14.252) | 7.953*** (4.823) |
| Controls | Yes | Yes |
| Firm FE | Yes | Yes |
| Year FE | Yes | Yes |
| N | 10727 | 10581 |
| Adjusted R² | 0.675 | 0.753 |
In conclusion, the adoption of robot technology in smart manufacturing significantly suppresses enterprise bankruptcy risk through mechanisms like supply chain financing and dual innovation. The effects are more pronounced in non-state-owned firms, competitive markets, and supportive ecological environments. Moreover, risk reduction translates into higher firm value, underscoring the strategic importance of robot technology. Policymakers should promote robot adoption through incentives, innovation ecosystems, and infrastructure development. Firms should invest in robot technology and complementary skills to harness its full potential. Future research could explore longitudinal effects and cross-country comparisons to deepen understanding of robot technology’s role in economic resilience.