Performance Expectation Gap and Risk-Taking of Agricultural Enterprises: The Moderating Effect of Institutional Environment

Abstract

In recent years, the operational performance of agricultural enterprises has been influenced by both natural conditions and market environments, resulting in high uncertainty and volatility. When performance falls below expectations, agricultural enterprises consciously engage in strategic change and proactive risk-taking to alleviate performance pressures. Based on Firm Behavioral Theory, Performance Feedback Theory, and Prospect Theory, we examine how performance expectation gap affects risk-taking of agricultural enterprises by using panel data of Chinese A-share listed agricultural firms from 2007 to 2023. The results show that performance expectation gap has a positive effect on risk-taking, which means the greater the gap, the higher the level of risk-taking. And the better developed the institutional environment, the greater the tendency for risk-taking. Further analysis shows that performance expectation gap promotes risk-taking by driving strategic change within agricultural enterprises. This research enriches the study on the influencing factors of risk-taking in agricultural enterprises, offering decision-making insights for them to prudently assess and manage risks.

1. Introduction

Driven by global population growth and the inelastic demand for food security, agriculture has long been regarded as a foundational industry with stable growth potential. However, as the key actors in agricultural production, agricultural enterprises have long been confronted with dual risks arising from both natural conditions and market dynamics. This “dual pressure” is not unique to Chinese agricultural enterprises but represents a widespread challenge facing the global agricultural sector in the course of modernization. In the context of global climate change, the heavy reliance of agricultural production on natural conditions and its vulnerable responses have become a worldwide concern [1]. However, dual risks in China demonstrate more complex institutional and structural characteristics. In terms of natural risks, compared with the large-scale, standardized farming models prevalent in Western developed countries, Chinese agricultural enterprises remain in a unique system based on small-scale family operations, achieving large-scale connectivity through socialized services. The fragmented production process and weak disaster-resilience infrastructure magnify the spillover effects of natural hazards. In terms of market risks, Chinese agricultural enterprises are caught in a structural squeeze between rising costs and shifting demand dynamics. On one hand, the sticky rises in land-transfer costs, labor expenses, and input prices continually raise the operational baseline for firms. On the other hand, the rapid shift from quantity-driven to quality-driven demand in the domestic consumer market, coupled with cyclical price fluctuations of agricultural products, has significantly increased the decision-making costs for enterprises [2].

Consequently, under the pressures from both natural conditions and market dynamics, it is difficult for agricultural enterprises to achieve expected performance targets. China’s agricultural sector serves as a pertinent example. Data released by the Ministry of Agriculture and Rural Affairs at the end of 2023 shows that the median net profit margin for the A-share agricultural enterprises was only 3.8%—a drop of 0.6 percent from 2022, which was significantly lower than the sector’s five-year average. Within this sector, livestock breeding companies were impacted by the production cycle, with over 60% of them reporting losses. Leading companies, such as Wens Foodstuff Group Co., Ltd., saw their net profit plummet by 152% year-on-year in 2023, with performance falling more than 30% below market expectations. The U.S. agricultural giant Corteva also faced considerable operational pressure due to lower sales volumes and higher input costs. According to its 2023 financial report, its full-year net sales declined by 9%, while segment operating EBITDA fell by 18% year-on-year. Similarly, ADM, one of the world’s four major grain traders, experienced profit shrink due to increased global supply and softened demand. Its full-year net profit for 2023 dropped sharply by 49% compared to the previous year, significantly under expectations. These cases demonstrate that agricultural enterprises worldwide are commonly facing performance difficulties. When a firm’s actual performance falls below its expected level, a performance expectation gap emerges, indicating that the current performance has failed to meet the minimum level expected by managers [3].

The existing literature has primarily explored the impacts of the performance expectation gap on corporate innovation behavior [4,5]: digital transformation [6]; digital strategy choice [7,8]; corporate risk preference; and investment behavior [9]. Most of them focus on the behavior mechanisms of manufacturing and high-tech enterprises, while paying little attention to agricultural enterprises. The inherent natural dependence and market volatility of agricultural enterprises may lead to distinctive behavioral mechanisms in response to performance expectation gaps. Their performance fluctuations often stem not only from managerial decisions but are also constrained by the growth cycles of biological assets and uncontrolled natural disruptions. According to Performance Feedback Theory, when a firm’s actual performance falls below its target level, decision-makers will receive negative feedback, which increases their risk tolerance and makes them more inclined to adopt high-risk strategies in order to improve their performance [10]. For agricultural enterprises, strategic change can help them adapt to market changes, enhance innovation capabilities, identify and manage potential risks, thereby raising their risk-taking capacity to achieve performance leaps. Based on the risk aversion logic of Threat Rigidity Theory, when external environmental pressures become overwhelming, such as severe natural disasters affecting agricultural production or persistent market downturns, which will lead to a sharp decline in performance, the enterprises may tend to reduce high-risk investments to ensure survival. Particularly when agricultural enterprises face performance shortfalls, they often cut risk investments and tighten controls [11]. From the micro perspective, risk-taking is a vital component of strategic decision-making for agricultural enterprises. It is not only crucial for building competitive advantages and expanding market shares but also serves as a key driver of long-term development [12,13]. From the macro perspective, risk-taking behavior facilitates capital accumulation and contributes to economic growth [14]. Thus, in the complex economic environment, how to guide agricultural enterprises experiencing a performance expectation gap to reasonably undertake risks has become a pressing practical issue.

Therefore, this research analyzes the influences of the performance expectation gap on risk-taking of Chinese agricultural enterprises and examines the moderating role of the institutional environment. Additionally, a heterogeneity analysis compares the effects between state-owned and non-state-owned enterprises, as well as between enterprises with high and low risk-taking levels. Furthermore, this research also examines the mechanism through which the performance expectation gap affects the risk-taking of agricultural enterprises, specifically investigating whether such gaps can enhance risk-taking capacity by promoting strategic change. The potential contributions are as follows: First, this research analyzes the impacts of performance expectation gap on risk-taking in agricultural enterprises based on Firm Behavioral Theory and Performance Feedback Theory, fully considering the characteristics of agricultural enterprises and the current institutional environment. Due to dual pressures from rising costs and weak market demand, many agricultural firms are experiencing performance challenges. And a sound institutional environment can help them to undertake risks more reasonably. So the institutional environment is introduced as a moderating variable [15,16] in order to investigate the moderating role in the relationship between performance expectation gaps and risk-taking. Thereby a new framework is offered for studying how agricultural enterprises can proactively manage risks. Second, the existing studies have examined the direct impacts of performance expectation gaps on corporate risk-taking [17,18]. But the research on the mechanisms remains limited, and has mainly focused on corporate governance factors, such as managerial behavior [19,20,21] and governance mechanisms [22] as well as innovation dimensions like R&D investment [23] and innovation strategies [24]. To address this gap, this research introduces strategic change as a mediating variable to further investigate the mechanism through which performance expectation gaps affect risk-taking in agricultural enterprises. This provides a new perspective for understanding how agricultural enterprises can enhance their risk-taking capacity through internal adjustments.

2. Theoretical Analysis and Hypothesis Development

2.1. Performance Expectation Gap and Risk-Taking of Agricultural Enterprises

Based on Firm Behavioral Theory, performance feedback influences managerial decisions in agricultural enterprises. Managers evaluate operational conditions by comparing actual performance with expected levels, subsequently adjusting strategic decisions [25]. When actual performance deviates from the expected level, the resulting performance expectation gap significantly influences a firm’s strategic choices and risk preferences.

The existing research on performance expectation gap and risk-taking of enterprise shows inconsistent conclusions. Some scholars argue that larger performance expectation gaps encourage greater risk-taking. This can be explained through Firm Behavioral Theory and Prospect Theory. According to Firm Behavioral Theory, decision-makers’ cognitive limitations make it difficult for them to conduct entirely objective assessments based solely on absolute measures of firm performance [26]. When performance falls below reference points, it signals organizational dissatisfaction, prompting problem analysis and innovative strategies. Prospect Theory further indicates that losses loom larger than gains, leading to risk-seeking behavior in loss domains [27].

Growing research in this area provides support for this perspective. Studies indicate that when firms experience performance expectation gaps, managers actively pursue high-risk, high-return activities [28], identify organizational problems [29], and search for “satisfactory” alternatives [30] to improve performance. These actions consequently promote greater risk-taking. Moreover, although innovation involves risks, managers are more likely to engage in innovative activities when facing performance expectation gaps to improve performance. The larger the shortfall, the greater the innovation efforts, thereby encouraging the firm to undertake higher levels of risk. Specifically, in agricultural enterprises, performance expectation gaps make managers more inclined to select high-return, high-risk projects, thereby promoting risk-taking.

So this research proposes the following hypotheses:

H1a. 

Performance expectation gap positively influences risk-taking in agricultural enterprises. A larger performance expectation gap promotes higher risk-taking.

Other scholars contend that larger performance expectation gaps lead to risk aversion [31]. Threat Rigidity Theory suggests that managers view high-risk activities as additional threats to survival during performance expectation gaps, reducing willingness for risky investments. Performance expectation gaps also increase managerial anxiety due to information asymmetry, leading to less rational decisions under external threats [19]. Empirical studies show that persistent shortfalls prioritize organizational survival, prompting conservative behavior and discouraging risk-taking [32].

Thus, this research proposes a research hypothesis:

H1b. 

Performance expectation gap negatively influences risk-taking in agricultural enterprises. A larger performance expectation gap encourages risk avoidance.

2.2. The Moderating Effect of Institutional Environment

The institutional environment comprises the formal rules and informal norms governing economic activities and organizational behavior, including macroeconomic policies, laws and regulations, industry standards, and social norms and culture [33]. Agriculture is inherently vulnerable and highly dependent on natural conditions, national policies, and market dynamics so that a well-developed institutional environment is crucial. It can bolster agricultural managers’ confidence in risk-taking, thereby promoting such behavior and fostering firm development. Specifically, rational industrial policies can guide firms to accurately identify opportunities and risks following performance expectation gaps. This enables a more scientific response through strategic ventures and risk investments, ultimately facilitating sustainable development [34]. When agricultural enterprises face performance expectation gaps, sound macro-policies can directly increase managerial risk tolerance and encourage greater risk-taking. Similarly, well-established legal systems reduce the costs and bolster the confidence associated with risk-taking. Complete legal frameworks lower transaction costs, making managers more willing to take risks by restructuring resources to improve performance. Moreover, mechanisms that mitigate losses from failure can promote R&D investment and further encourage risk-taking [35].

Clear industry regulations can guide agricultural firms toward compliant and sustainable risk strategies, such as quality safety standards and market access rules. This increases their risk tolerance and promotes higher risk-taking. Developed social norms and culture shape managers’ fundamental perceptions of risk [36]. In the context of performance expectation gaps, these influences can shift agricultural enterprises from “passive risk-taking” to “proactive risk resistance,” thereby enhancing risk-taking and supporting sustainable agricultural development. Thus, a sound institutional environment provides a favorable external context for risk-taking when agricultural enterprises experience performance expectation gaps. It reduces the cost of risk-taking, strengthens managerial confidence, and ultimately promotes corporate risk-taking.

Thus, this research proposes a research hypothesis:

H2. 

The institutional environment positively moderates the relationship between the performance expectation gap and risk-taking in agricultural enterprises.

3. Research Design

3.1. Data Sources

The sample consists of agricultural companies listed on China’s Shanghai and Shenzhen A-share markets. Data were primarily sourced from the CSMAR and WIND databases. The observation period spans from 2007 to 2023. This start date was chosen because China’s enterprise accounting standards underwent a major reform on 1 January 2007, ensuring the comparability of financial data across the sample period. The sample was processed as follows: (1) Exclude companies labeled ST or *ST (indicating financial distress). (2) Remove companies with serious missing values in key financial variables. (3) Omit companies that conducted their initial public offering (IPO) after 2007. (4) Winsorize all continuous variables at the 1% and 99% to mitigate the influence of extreme values. The final sample comprised 65 listed agricultural firms, resulting in 952 firm-year observations. Furthermore, to account for the fact that corporate decisions are often based on prior performance feedback, we lag all independent, moderating, and control variables by one period relative to the dependent variable. This process yielded 870 observations for empirical testing.

3.2. Measures

(1) Dependent Variable

Risk-taking (Risk). Referring to the research of Boubakri et al. (2012), this research uses the volatility of corporate earnings as a proxy [37,38,39]. Since high-risk projects or decisions often lead to significant profit fluctuations, greater volatility typically indicates a stronger risk-taking propensity. In the measurement process, the return on assets (ROA) is adopted as the core financial metric and constructs the observation interval using a rolling time window approach. This metric reflects a company’s ability to generate profits using all available assets, and its ratio-based nature facilitates cross-scale comparisons while directly capturing performance changes driven by operational decisions. Given that the average tenure of executives in Chinese listed companies is approximately three years, a three-year rolling window is adopted for calculations. To account for the susceptibility of agricultural enterprises to external shocks such as industry cycles and natural disasters, the following procedure is implemented: First, the firm-level ROA is adjusted by subtracting the annual industry-average ROA, yielding an industry-adjusted ROA. Then, within each three-year observation window, the standard deviation of the firm’s adjusted ROA is calculated. Finally, this standard deviation is used as the quantitative measure of firm-level risk-taking. This methodology effectively filters out systemic risks at the industry level (e.g., changes in agricultural policies, cyclical fluctuations in agricultural product prices), thereby more accurately capturing the differences in managerial decisions driven by performance aspiration pressures. The specific calculation formula is as follows:

$$Riski,t=\sqrt{{\displaystyle \frac{1}{N-1}}{\sum }_{t=1}^N({adj\_ROA}_{i,t}-{\displaystyle \frac{1}{N}}{\sum }_{t=1}^N{adj\_ROA}_{i,t}{)}^2}\hspace{1em}N=3$$

(1)

$$adj\_{ROA}_{i,t}={\displaystyle \frac{{EBITDA}_{i,t}}{{ASSETS}_{i,t}}}-{\displaystyle \frac{1}{n}}{\sum }_{i=1}^n{\displaystyle \frac{{EBITDA}_{i,t}}{{ASSETS}_{i,t}}}$$

(2)

Here, $Risk$_i,t denotes the risk-taking for firm i in year t and n stands for the total number of firms within a given industry. ${\mathrm{a}\mathrm{d}\mathrm{j}\_\mathrm{R}\mathrm{O}\mathrm{A}}_{\mathrm{i},\mathrm{t}}$ refers to the industry-adjusted return on assets for each firm after removing the industry average effect. Return on assets is defined as the ratio of earnings before interest, taxes, depreciation, and amortization (${\mathrm{E}\mathrm{B}\mathrm{I}\mathrm{T}\mathrm{D}\mathrm{A}}_{\mathrm{i},\mathrm{t}}$) to total assets at the end of the year (${\mathrm{A}\mathrm{S}\mathrm{S}\mathrm{E}\mathrm{T}\mathrm{S}}_{\mathrm{i},\mathrm{t}}$) for firm i in year t.

(2) Independent variables

Performance expectation gaps are categorized into historical performance expectation gaps and industry performance expectation gaps [40,41].

Historical Performance Expectation Gap (Hps). It is defined as the absolute value of the negative deviation between actual firm performance and historical expected performance. Return on assets is used as the key indicator of actual performance. The historical expected performance level is calculated as a weighted average of the firm’s prior actual performance and its prior expected performance [42]. The formula is as follows:

$$A_{i,t-1}={\alpha }_1{P}_{i,t-2}+(1-{\alpha }_1)A_{i,t-2}$$

(3)

$${Hps}_{i,t-1}=I_1 \ast (P_{i,t-1}-A_{i,t-1})$$

(4)

Here, A_i,t−1 represents the historical performance expectation level of firm i in t − 1 year, and P_i,t−2 denotes the actual performance of firm i in t − 2 year. α₁ reflects the relative weight between the prior period’s actual performance and its expected level, with a value range of [0, 1]. Following Chen, this research sets α₁ = 0.6 to measure the historical performance expectation level [42]. It should be noted that for the first period, the historical performance expectation gap is replaced by the actual performance of that period. Additionally, when actual performance is lower than the historical expectation, the indicator I₁ is set to 1; otherwise, it is set to 0. To facilitate further analysis, the absolute value of the historical performance expectation gap is used.

Industry performance expectation gap (Ips). Similar to the calculation of historical performance expectation gap, the industry performance expectation gap is measured as the absolute value of the negative deviation between a firm’s actual performance and the industry expected performance level. The industry expected performance level is computed using a weighting factor of 0.6, consistent with the approach applied in historical shortfall measurement. Drawing on the method of Chen, the median actual performance of all firms within the industry is used as the basis for determining the industry expected performance level [42].

(3) Moderating Variable

Institutional environment (Market). This research adopts the measurement method of Shi to assess the institutional environment [43], using the comprehensive marketization index from the Marketization Index Report compiled by Fan [44]. This index evaluates regional market development levels through five dimensions: Government-market relationship index, development index of the non-state economy, product market development index, factor market development index, market intermediary development and legal environment index.

(4) Control Variables

This study selects firm size (Size), Tobin’s Q (TobinQ), board size (Board), leverage ratio (Lev), firm age (ListAge), fixed asset ratio (FIXED), and revenue growth rate (Growth) as control variables [45].

3.3. Baseline Model Specification

This research employs the following model to analyze the impact of performance expectation gap on agricultural firms’ risk-taking. The historical performance expectation gap (${\mathrm{H}\mathrm{p}\mathrm{s}}_{\mathrm{i},\mathrm{t}-1}$) and industry performance expectation gap (${\mathrm{I}\mathrm{p}\mathrm{s}}_{\mathrm{i},\mathrm{t}-1}$) are included as independent variables, while risk-taking (${\mathrm{R}\mathrm{i}\mathrm{s}\mathrm{k}}_{\mathrm{i},\mathrm{t}}$) serves as dependent variable. Control variables (${\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}_{\mathrm{i},\mathrm{j},\mathrm{t}-1}$), year fixed effects (Year), industry fixed effects (Ind), and the error term (${\mathsf{ϵ}}_{\mathrm{i},\mathrm{t}-1}$) are also incorporated. The model is specified as follows:

$${\mathrm{R}\mathrm{i}\mathrm{s}\mathrm{k}}_{\mathrm{i},\mathrm{t}}={\mathsf{\alpha }}_0+{\mathsf{\alpha }}_1{\mathrm{H}\mathrm{p}\mathrm{s}}_{\mathrm{i},\mathrm{t}-1}+\sum {\mathsf{\alpha }}_{\mathrm{j}}{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}_{\mathrm{i},\mathrm{j},\mathrm{t}-1}+\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}+\mathrm{I}\mathrm{n}\mathrm{d}+{\mathsf{ϵ}}_{\mathrm{i},\mathrm{t}-1}$$

(5)

$${\mathrm{R}\mathrm{i}\mathrm{s}\mathrm{k}}_{\mathrm{i},\mathrm{t}}={\mathsf{\alpha }}_0+{\mathsf{\alpha }}_1{\mathrm{I}\mathrm{p}\mathrm{s}}_{\mathrm{i},\mathrm{t}-1}+\sum {\mathsf{\alpha }}_{\mathrm{j}}{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}_{\mathrm{i},\mathrm{j},\mathrm{t}-1}+\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}+\mathrm{I}\mathrm{n}\mathrm{d}+{\mathsf{ϵ}}_{\mathrm{i},\mathrm{t}-1}$$

(6)

To test the moderating effect, the interaction term between the institutional environment and historical performance expectation gap is incorporated into model (1), and the interaction term between the institutional environment and industry performance expectation gap is added to model (2). The models are specified as follows:

$${\mathrm{R}\mathrm{i}\mathrm{s}\mathrm{k}}_{\mathrm{i},\mathrm{t}}={\mathsf{\alpha }}_0+{\mathsf{\alpha }}_1{\mathrm{H}\mathrm{p}\mathrm{s}}_{\mathrm{i},\mathrm{t}-1}+{\mathsf{\alpha }}_2{\mathrm{M}\mathrm{a}\mathrm{r}\mathrm{k}\mathrm{e}\mathrm{t}}_{\mathrm{i},\mathrm{t}-1}+{\mathsf{\alpha }}_3{\mathrm{M}\mathrm{a}\mathrm{r}\mathrm{k}\mathrm{e}\mathrm{t}}_{\mathrm{i},\mathrm{t}-1} \ast {\mathrm{H}\mathrm{p}\mathrm{s}}_{\mathrm{i},\mathrm{t}-1}+\sum {\mathsf{\alpha }}_{\mathrm{j}}{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}_{\mathrm{i},\mathrm{j},\mathrm{t}-1}+\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}+\mathrm{I}\mathrm{n}\mathrm{d}+{\mathsf{ϵ}}_{\mathrm{i},\mathrm{t}-1}$$

(7)

$${\mathrm{R}\mathrm{i}\mathrm{s}\mathrm{k}}_{\mathrm{i},\mathrm{t}}={\mathsf{\alpha }}_0+{\mathsf{\alpha }}_1{\mathrm{I}\mathrm{p}\mathrm{s}}_{\mathrm{i},\mathrm{t}-1}+{\mathsf{\alpha }}_2{\mathrm{M}\mathrm{a}\mathrm{r}\mathrm{k}\mathrm{e}\mathrm{t}}_{\mathrm{i},\mathrm{t}-1}+{\mathsf{\alpha }}_3{\mathrm{M}\mathrm{a}\mathrm{r}\mathrm{k}\mathrm{e}\mathrm{t}}_{\mathrm{i},\mathrm{t}-1} \ast {\mathrm{I}\mathrm{p}\mathrm{s}}_{\mathrm{i},\mathrm{t}-1}+\sum {\mathsf{\alpha }}_{\mathrm{j}}{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}_{\mathrm{i},\mathrm{j},\mathrm{t}-1}+\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}+\mathrm{I}\mathrm{n}\mathrm{d}+{\mathsf{ϵ}}_{\mathrm{i},\mathrm{t}-1}$$

(8)

4. Empirical Results

4.1. Descriptive Statistics and Correlation Analysis

Table 1. Descriptive Statistics.

Variable	N	Mean	SD	Min	Median	Max
Risk	870	0.0415	0.0533	0.0003	0.0263	0.7150
Hps	870	0.0221	0.0533	0.0000	0.0000	0.3670
Ips	870	0.0238	0.0521	0.0000	0.0002	0.3180
Market	870	8.0274	2.1501	3.2590	8.2660	12.8640
DS	870	0.6550	0.3440	0.0949	0.5820	2.7760
FIXED	870	0.2670	0.1370	0.0170	0.2600	0.6620
Growth	870	0.1730	0.5120	−0.6770	0.0869	4.5930
Board	870	2.1140	0.2190	1.6090	2.1970	2.7080
TobinQ	870	2.3030	1.3580	0.8270	1.9400	12.6300
Lev	870	0.4790	0.1940	0.0496	0.4860	1.0300
Size	870	22.1300	1.2100	19.0500	22.0600	26.0100
ListAge	870	2.7510	0.4230	0.0000	2.8330	3.4010

presents the descriptive statistics for the 870 observations. The results show that the mean value of historical performance expectation gap (Hps) is 0.0221, with a maximum of 0.3670 and a minimum of 0. The mean value of industry performance expectation gap (Ips) is 0.0238, with a maximum of 0.3180 and a minimum of 0, indicating that most agricultural enterprises experienced performance expectation gaps, and historical performance expectation gaps were greater than industry performance expectation gaps. The mean value of risk-taking (Risk) is 0.0415, with a standard deviation of 0.0533, a maximum of 0.7150, and a minimum of 0.0003, suggesting that agricultural firms generally exhibit low levels of risk-taking, with substantial variation among them. The mean value of the institutional environment (Market) is 8.0247, with a maximum of 12.8640 and a minimum of 3.2590, reflecting a relatively favorable institutional environment in the agricultural sector, albeit with considerable fluctuation.

Table 2. Correlation Analysis.

	Risk	Ips	Hps	DS	FIXED	Growth	TobinQ	Board	Lev	ListAge	Size
Risk	1
Ips	0.337 ***	1
Hps	0.468 ***	0.792 ***	1
DS	0.179 ***	0.213 ***	0.101 ***	1
FIXED	0.069 **	0.087 ***	0.055 *	−0.088 ***	1
Growth	0.015	−0.095 ***	−0.085 ***	−0.026	−0.090 ***	1
TobinQ	0.068 **	−0.019	−0.017	0.128 ***	0.020	0.005	1
Board	−0.106 ***	−0.092 ***	−0.114 ***	0.028	−0.024	0.011	−0.007	1
Lev	0.151 ***	0.334 ***	0.177 ***	0.086 ***	0.048	0.018	−0.206 ***	0.131 ***	1
ListAge	0.043	0.051	0.011	0.068 **	−0.175 ***	−0.006	−0.056 *	−0.093 ***	−0.023	1
Size	−0.171 ***	−0.126 ***	−0.064 **	−0.196 ***	−0.121 ***	0.003	−0.381 ***	0.150 ***	0.206 ***	0.188 ***	1

Standard errors in parentheses. * p < 0.1, ** p < 0.05, *** p < 0.01.

presents the correlations among the variables. The results show that both historical performance expectation gap and industry performance expectation gap are positively correlated with agricultural firms’ risk-taking at the 1% significance level (p < 0.01), suggesting a statistically significant positive relationship between performance expectation gaps and risk-taking. Additionally, the variance inflation factor (VIF) for each variable was calculated, and all values were found to be below 5, indicating no severe multicollinearity issues among the variables.

4.2. Regression Results and Analysis

The results are provided in

Table 3. Baseline regression test.

	(1)	(2)	(3)	(4)
	Risk	Risk	Risk	Risk
Hps	0.538 ***		0.498 ***
	(0.094)		(0.090)
Ips		0.562 ***		0.525 ***
		(0.082)		(0.083)
FIXED			0.030 **	0.025 **
			(0.013)	(0.012)
Growth			0.007 *	0.007 *
			(0.004)	(0.004)
Board			−0.007	−0.011
			(0.009)	(0.008)
TobinQ			0.002 **	0.002 **
			(0.001)	(0.001)
Lev			0.047 ***	0.020 **
			(0.010)	(0.009)
Size			−0.006 ***	−0.002
			(0.002)	(0.002)
ListAge			0.002	0.004
			(0.004)	(0.004)
_cons	0.032 ***	0.036 ***	0.122 ***	0.067 **
	(0.006)	(0.006)	(0.029)	(0.032)
N	870	870	870	870
R2	0.31	0.32	0.36	0.34
ind	YES	YES	YES	YES
year	YES	YES	YES	YES

Standard errors in parentheses. * p < 0.1, ** p < 0.05, *** p < 0.01.

. Column (1) estimates the impact of historical performance expectation gap on agricultural firms’ risk-taking using a fixed-effects linear model, while column (2) examines the effect of industry performance expectation gap. In column (3), the regression coefficient for historical performance expectation gap is 0.498 and significant at the 1% level, indicating a significant positive effect on agricultural firms’ risk-taking. A larger historical performance expectation gap encourages higher risk-taking. Unlike the findings of Jiang and Verhees, who observed that different types of performance expectation gaps had varying impacts on corporate innovation investment, our results suggest that when agricultural firms face historical performance expectation gaps, managers perceive developmental pressure and recognize deficiencies in existing strategies [45,46]. They prioritize internal improvements such as optimizing resource allocation and enhancing operational efficiency, while also actively pursuing change. Historical performance expectation gaps can drive exploratory innovation in agricultural enterprises. Managers tend to make risk-oriented decisions and select high-risk, high-return projects to rapidly improve performance. This process ultimately promotes risk-taking in agricultural firms.

In column (4), the regression coefficient for industry performance expectation gap is 0.525 and significant at the 1% level, indicating that a larger industry performance expectation gap also promotes higher risk-taking in agricultural firms. This result aligns with Chen and Miller [47]. The finding is consistent with Firm Behavioral Theory, Prospect Theory, and Performance Feedback Theory, which suggest that when an agricultural firm’s performance falls below the industry average, it perceives a threat to its competitive position and experiences “catching-up pressure.” Managers are more inclined to adopt high-risk strategies to quickly close the gap and avoid being eliminated from the industry, thereby promoting risk-taking. The above proves support for hypothesis H1a.

To further examine the role of the institutional environment in the relationship between performance expectation gap and risk-taking in agricultural enterprises, this research introduces interaction terms into the regression analysis. As shown in

Table 4. The moderation effect of the institutional environment.

	(1)	(2)	(3)	(4)
	Risk	Risk	Risk	Risk
Hps	0.498 ***	0.535 ***
	(0.090)	(0.095)
Ips			0.525 ***	0.603 ***
			(0.083)	(0.104)
Market		0.001		0.002 **
		(0.001)		(0.001)
Hps*Market		0.078 **
		(0.038)
Ips*Market				0.076 **
				(0.035)
FIXED	0.030 **	0.028 **	0.025 **	0.025 **
	(0.013)	(0.012)	(0.012)	(0.011)
Growth	0.007 *	0.006	0.007 *	0.007 *
	(0.004)	(0.004)	(0.004)	(0.004)
Board	−0.007	−0.006	−0.011	−0.009
	(0.009)	(0.008)	(0.008)	(0.008)
TobinQ	0.002 **	0.003 **	0.002 **	0.003 ***
	(0.001)	(0.001)	(0.001)	(0.001)
Lev	0.047 ***	0.051 ***	0.020 **	0.025 **
	(0.010)	(0.010)	(0.009)	(0.010)
Size	−0.006 ***	−0.006 ***	−0.002	−0.003
	(0.002)	(0.002)	(0.002)	(0.002)
ListAge	0.002	0.003	0.004	0.005
	(0.004)	(0.004)	(0.004)	(0.004)
_cons	0.122 ***	0.117 ***	0.067 **	0.057 *
	(0.029)	(0.029)	(0.032)	(0.032)
N	870	870	870	870
R2	0.361	0.388	0.339	0.375
industry	YES	YES	YES	YES
year	YES	YES	YES	YES

Standard errors in parentheses. * p < 0.1, ** p < 0.05, *** p < 0.01.

, the coefficient of the interaction term between historical performance expectation gap and the institutional environment is 0.078, and that between industry performance expectation gap and the institutional environment is 0.076. Both are positive and significant at the 5% level, indicating that the institutional environment positively moderates the relationship between performance expectation gap and risk-taking.

Specifically, in regions with more developed institutional environments, agricultural enterprises are more inclined to pursue higher-risk investments or strategic actions when facing historical or industry performance expectation gaps. This is primarily reflected in two aspects: First, a robust institutional environment ensures stable and transparent agricultural support policies, such as direct subsidies, target price insurance, and disaster compensation. These policies can effectively mitigate potential losses from strategic initiatives like technological innovation and market expansion, thereby encouraging proactive responses to performance gaps. Second, such environments feature more efficient policy implementation and resource allocation. For example, where property rights are well-protected and government services are standardized, firms facing performance pressure gain better access to policy-backed loans, dedicated development funds, and technology extension services. These resources significantly enhance the feasibility of catch-up through strategic change.

Thus, a stronger institutional environment enhances agricultural firms’ responsiveness to performance expectation gaps, encouraging riskier initiatives to improve performance. This reinforces the positive relationship between performance expectation gap and risk-taking. These findings support research hypothesis H2.

4.3. Endogeneity Test

A stronger risk-taking capacity enables firms to improve operational performance and narrow the performance expectation gap by mitigating financing constraints, enhancing equity incentives, optimizing resource allocation, and increasing decision-making efficiency [48,49]. Therefore, there may be endogeneity problems caused by reverse causality. Existing research on farmer performance–risk behavior suggests that unaddressed endogeneity may lead to biased estimates and misleading policy implications [50]. Therefore, this research employs two-stage least squares (2SLS) and limited information maximum likelihood (LIML) estimation to address potential endogeneity concerns. Two-way clustered standard errors (at both the firm and time dimensions) are applied in the model to control for serial correlation. The performance aspiration gap lagged by two periods is selected as the instrumental variable. The 2SLS regression results are presented in

Table 5. Results of endogeneity test.

Variable	2SLS
	(1) First-Stage Hps	(2) Second-Stage Risk	(3) First-Stage Ips	(4) Second-Stage Risk
HpsIV	0.146 *
	(0.084)
Hps		2.445 *
		(1.446)
IpsIV			0.160 **
			(0.063)
Ips				2.993 **
				(1.002)
FIXED	0.018	0.008	0.018 *	−0.027
	(0.011)	(0.035)	(0.011)	(0.032)
Growth	−0.010 **	0.023	−0.010 ***	0.029 **
	(0.003)	(0.014)	(0.003)	(0.012)
TobinQ	0.0002	0.002	0.0005	0.001
	(0.001)	(0.003)	(0.0016)	(0.004)
ListAge	0.002	−0.005	0.00004	0.002
	(0.005)	(0.009)	(0.0038)	(0.010)
Board	−0.024 **	0.048	−0.017 **	0.034
	(0.011)	(0.033)	(0.009)	(0.028)
Lev	0.049 ***	−0.061	0.091 ***	−0.240 **
	(0.013)	(0.084)	(0.013)	(0.108)
Size	−0.004 *	0.003	−0.009 ***	0.026 **
	(0.002)	(0.007)	(0.002)	(0.012)
_cons	0.117 **	−0.125	0.223 ***	−0.578 **
	(0.036)	(0.201)	(0.037)	(0.269)
Cluster level	firm and year		firm and year
N	870	870	870	870
R-sq	0.1061		0.2401
Prob > F	0.0844		0.0118

Standard errors in parentheses. * p < 0.1, ** p < 0.05, *** p < 0.01.

. After accounting for endogeneity, the performance aspiration gap continues to exhibit a statistically significant positive effect on agricultural firms’ risk-taking. As a robustness check, the model is also estimated using the LIML method. The resulting coefficients and significance levels show no substantial differences from the 2SLS results reported in Table 5. Given that the model is exactly identified, the 2SLS and LIML estimates are equivalent, confirming the robustness of the findings and thereby validating the reliability of the baseline regression results. The 2SLS results are reported here.

4.4. Robustness Tests

This research employs multiple robustness checks to verify the reliability of the results, as presented in

Table 6. Regression results of robustness tests.

	(1)	(2)	(3)	(4)	(5)	(6)
	Risk	Risk	Risk1	Risk1	Risk	Risk
Hps1	0.513 ***
	(0.092)
Ips1		0.535 ***
		(0.085)
Hps			0.942 ***		0.561 ***
			(0.172)		(0.091)
Ips				0.998 ***		0.529 ***
				(0.161)		(0.086)
FIXED	0.034 **	0.026 **	0.059 **	0.049 **	0.026 *	0.024 *
	(0.014)	(0.013)	(0.025)	(0.024)	(0.014)	(0.014)
Growth	0.006 *	0.007 *	0.013 *	0.014 *	0.007	0.005
	(0.004)	(0.004)	(0.007)	(0.008)	(0.005)	(0.005)
Board	−0.009	−0.012	−0.011	−0.019	−0.013	−0.014
	(0.009)	(0.009)	(0.016)	(0.016)	(0.009)	(0.010)
TobinQ	0.003 **	0.002 **	0.005 **	0.004 **	0.001	0.002 *
	(0.001)	(0.001)	(0.002)	(0.002)	(0.001)	(0.001)
Lev	0.052 ***	0.021 **	0.090 ***	0.037 **	0.043 ***	0.018 *
	(0.010)	(0.009)	(0.018)	(0.018)	(0.009)	(0.009)
Size	−0.006 ***	−0.002	−0.011 ***	−0.004	−0.005 ***	−0.001
	(0.002)	(0.002)	(0.003)	(0.003)	(0.002)	(0.002)
ListAge	0.002	0.004	0.004	0.006	0.002	0.001
	(0.004)	(0.004)	(0.008)	(0.007)	(0.005)	(0.005)
_cons	0.129***	0.066 **	0.230 ***	0.125 **	0.133 ***	0.067 *
	(0.029)	(0.033)	(0.054)	(0.060)	(0.031)	(0.035)
N	870	870	870	870	746	746
R2	0.327	0.322	0.364	0.345	0.403	0.353
ind	YES	YES	YES	YES	YES	YES
year	YES	YES	YES	YES	YES	YES

Standard errors in parentheses. * p < 0.1, ** p < 0.05, *** p < 0.01.

. First, the measurement weights of the independent variables were adjusted (Hps1 and Ips1). The parameter ${\mathsf{\alpha }}_1$, which captures the relative weight between prior actual performance and historical performance expectation gap, was varied to examine its differential impact on risk-taking. Similarly, the parameter ${\mathsf{\alpha }}_2$, reflecting the weight allocation between prior actual performance and industry performance expectation gap, was adjusted to assess the heterogeneous effects of industry performance expectation gap on risk-taking. The values of ${\mathsf{\alpha }}_1$ and ${\mathsf{\alpha }}_2$ were changed from 0.6 to 0.5 [51]. The regression results, shown in columns (1) and (2), confirm the robustness of the main findings. Second, an alternative measure for the dependent variable was adopted. The proxy for risk-taking was replaced from the standard deviation of industry-adjusted ROA ($\mathrm{A}\mathrm{d}\mathrm{j}\_{\mathrm{R}\mathrm{O}\mathrm{A}}_{\mathrm{i},\mathrm{t}}$) to the range of adjusted ROA (Risk1). The regression outcomes, reported in columns (3) and (4), remain consistent with the baseline results. Third, the sample period was shortened to 2011–2023. As shown in columns (5) and (6), both historical and industry performance expectation gaps remain statistically significant at the 1% level, further supporting the robustness of the initial conclusions.

4.5. Heterogeneity Analysis

Based on the ownership type, the sample is divided into state-owned and non-state-owned agricultural enterprises. The regression results, presented in

Table 7. Performance expectation gap and property right heterogeneity test of risk-taking.

Variable	State-Owned Agricultural Enterprise		Non-State-Owned Agricultural Enterprise
	(1)	(2)	(3)	(4)
	Risk	Risk	Risk	Risk
Hps	0.387 ***		0.570 ***
	(0.044)		(0.135)
Ips		0.393 ***		0.666 ***
		(0.050)		(0.137)
FIXED	0.008	0.006	0.056 ***	0.051 **
	(0.014)	(0.013)	(0.022)	(0.021)
Growth	0.005	0.006	0.009	0.008
	(0.005)	(0.005)	(0.006)	(0.007)
Board	−0.017 *	−0.013	0.002	−0.009
	(0.010)	(0.010)	(0.013)	(0.013)
TobinQ	0.002	0.001	0.003	0.003
	(0.001)	(0.001)	(0.002)	(0.002)
Lev	0.027 **	−0.000	0.071 ***	0.045 ***
	(0.012)	(0.012)	(0.015)	(0.014)
Size	−0.005 ***	−0.003	−0.008 ***	−0.003
	(0.002)	(0.002)	(0.003)	(0.003)
ListAge	0.009 **	0.011 **	−0.005	−0.002
	(0.004)	(0.004)	(0.009)	(0.008)
_cons	0.152 ***	0.103 **	0.125 ***	0.063
	(0.043)	(0.046)	(0.042)	(0.045)
N	496	496	374	374
R2	0.291	0.285	0.472	0.460
Chow Test	2.600	3.220	2.600	3.220
p-Value	0.0001	0.0000	0.0001	0.0000
ind	YES	YES	YES	YES
year	YES	YES	YES	YES

Standard errors in parentheses. * p < 0.1, ** p < 0.05, *** p < 0.01.

, show that both historical and industry performance expectation gaps have a significantly positive effect on risk-taking at the 1% level for both types of enterprises, and the between-group coefficient difference tests are statistically significant.

Specifically, the regression coefficients for historical performance expectation gap and industry performance expectation gap are 0.570 and 0.666, respectively, for non-state-owned enterprises, compared to 0.387 and 0.393 for state-owned enterprises. This indicates that the positive effect of performance expectation gap on risk-taking is stronger in non-state-owned agricultural enterprises than in state-owned ones. This finding is consistent with prior research [52], suggesting that non-state-owned firms are more inclined to take risks than state-owned enterprises.

This phenomenon can be explained by the fact that profitability is fundamental to the survival and development of non-state-owned enterprises. When facing performance expectation gaps, executives in non-state-owned firms exhibit stronger opportunistic motivations and are more likely to engage in risky investments to improve profitability. Their higher risk tolerance facilitates greater risk-taking [53]. In contrast, state-owned enterprises must balance the demands of multiple stakeholders, leading to greater risk sensitivity and a tendency to adopt conservative strategies to avoid risks [54].

To investigate the different impacts of performance expectation gaps on risk-taking, this research divides agricultural enterprises into high risk-taking and low risk-taking groups based on ROA volatility, with regression results presented in

Table 8. Performance expectation gap and heterogeneity test of risk-taking.

Variable	High Risk-Taking Enterprise		Low Risk-Taking Enterprise
	(1)	(2)	(3)	(4)
	Risk	Risk	Risk	Risk
Hps	0.461 ***		0.025 ***
	(4.890)		(3.550)
Ips		0.400 ***		−0.006
		(4.510)		(−0.290)
FIXED	0.062 *	0.061 *	0.002	0.002
	(2.510)	(2.340)	(1.230)	(1.010)
Growth	0.009	0.008	0.002 ***	0.002 ***
	(1.700)	(1.480)	(4.690)	(4.790)
Board	−0.030	−0.029	−0.000	−0.001
	(−1.950)	(−1.780)	(−0.240)	(−0.670)
TobinQ	0.001	0.002	0.000	0.000
	(1.290)	(1.720)	(0.620)	(0.440)
Lev	0.053 ***	0.037 **	−0.000	−0.000
	(4.420)	(2.860)	(−0.030)	(−0.150)
Size	−0.003	0.001	−0.000	−0.000
	(−1.030)	(0.350)	(−0.460)	(−0.440)
ListAge	0.003	0.003	0.001	0.001
	(0.260)	(0.230)	(1.500)	(1.720)
_cons	0.119 *	0.050	0.014	0.016 *
	(2.500)	(0.930)	(1.890)	(2.090)
N	422	422	448	448
R2	0.370	0.300	0.120	0.100
adj. R2	0.330	0.260	0.070	0.060

Statistics in parentheses. * p < 0.05, ** p < 0.01, *** p < 0.001.

. For high risk-taking enterprises, both historical and industry performance expectation gaps significantly positively influence risk-taking at 1% significance level, with regression coefficients of 0.461 and 0.400, respectively. These findings are aligned with Ling (2008) and Wu (2005), demonstrating that high risk-taking enterprises are more likely to consider potential returns from risky decisions and adopt more aggressive risk-taking behaviors [55,56]. In contrast, for low risk-taking enterprises, the impacts of industry performance expectation gaps on risk-taking are not statistically significant, while the influence of historical performance expectation gaps is significantly weakened. This observation supports the Threat Rigidity Theory, which posits that when facing systemic risks at the industry level, managers tend to engage in cognitive contraction by actively ignoring external industry performance information to avoid decision-making burdens caused by information overloading. Consequently, low risk-taking enterprises exhibit institutional neglect toward industry performance expectation gaps, leading to a significant reduction in the impact of historical performance expectation gaps on their risk-taking behavior [57].

4.6. Further Analysis

According to the Firm Behavioral Theory, when agricultural enterprises experience a performance expectation gap, managers become more aware of the need for strategic change and believe that risk-taking decisions can improve performance. Agricultural production is inherently risky due to its susceptibility to natural conditions and market fluctuations. If an agricultural enterprise is performing well, it generally avoids unnecessary strategic changes to prevent additional risk. However, when actual performance falls significantly below expectations, creating a performance expectation gap, the enterprise faces survival pressure. Managers are then more likely to initiate strategic change to improve performance [58,59]. Specifically, performance expectation gaps alter existing strategic decisions and increase the degree of strategic change. Enterprises also adjust the direction of strategic change based on the extent of the performance expectation gap. Strategic change positively influences risk-taking in agricultural enterprises by helping them adapt to market changes, promoting technological innovation, and optimizing resource allocation, thereby enhancing sustainable development capabilities. Research based on the Firm Behavioral Theory and Optimal Distinctiveness Theory shows that the more a firm’s strategic positioning deviates from industry norms, the greater its innovation investment and risk-taking [60,61]. Thus, when performance expectation gaps occur, managers tend to pursue strategic change, which in turn promotes risk-taking [62]. To verify the above hypothesis, we further investigated the role of strategic change.

Following the approach of related research, we measure strategic change (DS) as the degree of fluctuation in an enterprise’s strategic resource allocation across consecutive years. The measurement process involves three steps: (1) Select six key resource allocation variables: advertising investment, R&D investment, the indirect cost ratio, inventory level, fixed asset renewal rate, and the financial leverage ratio. (2) Calculate annual changes and standardize: For each variable, compute its year-to-year change (from year t−1 to year t). Then, standardize these changes by their annual industry median. (3) Compute the index: Take the absolute value of each industry-adjusted change. The degree of strategic change is the average of these six absolute values.

The results for the mediating effect of strategic change are presented in

Table 9. Test results of mediating effect.

	(1)	(2)	(3)	(4)	(5)	(6)
	Risk	DS	Risk	Risk	DS	Risk
Hps	0.498 ***	0.591 **	0.489 ***
	(0.090)	(0.260)	(0.088)
Ips				0.525 ***	1.375 ***	0.514 ***
				(0.083)	(0.350)	(0.082)
DS			0.015 ***			0.008 *
			(0.004)			(0.004)
FIXED	0.030 **	−0.317 ***	0.035 ***	0.025 **	−0.339 ***	0.028 **
	(0.013)	(0.085)	(0.013)	(0.012)	(0.084)	(0.013)
Growth	0.007 *	−0.003	0.007 *	0.007 *	0.005	0.007 *
	(0.004)	(0.031)	(0.004)	(0.004)	(0.030)	(0.004)
Board	−0.007	0.072	−0.008	−0.011	0.080 *	−0.012
	(0.009)	(0.047)	(0.008)	(0.008)	(0.046)	(0.008)
TobinQ	0.002 **	0.012	0.002 **	0.002 **	0.012	0.002 *
	(0.001)	(0.014)	(0.001)	(0.001)	(0.013)	(0.001)
Lev	0.047 ***	0.165	0.045 ***	0.020 **	0.052	0.019 **
	(0.010)	(0.104)	(0.010)	(0.009)	(0.111)	(0.009)
Size	−0.006 ***	−0.058 ***	−0.005 ***	−0.002	−0.045 ***	−0.002
	(0.002)	(0.013)	(0.002)	(0.002)	(0.013)	(0.002)
ListAge	0.002	0.095 ***	0.001	0.004	0.096 ***	0.003
	(0.004)	(0.030)	(0.004)	(0.004)	(0.029)	(0.004)
_cons	0.122 ***	1.522 ***	0.099 ***	0.067 **	1.280 ***	0.057 *
	(0.029)	(0.271)	(0.031)	(0.032)	(0.267)	(0.034)
N	870	870	870	870	870	870
R2	0.361	0.083	0.369	0.339	0.109	0.342
ind	YES	YES	YES	YES	YES	YES
year	YES	YES	YES	YES	YES	YES

Standard errors in parentheses. * p < 0.1, ** p < 0.05, *** p < 0.01.

. According to the regression results of Model (2), the coefficient of historical performance expectation gap is positive and significant at the 5% level, indicating that agricultural firms implement strategic changes when facing historical performance expectation gaps. Model (3) results preliminarily suggest a positive partial mediating effect of strategic change in the relationship between historical performance expectation gaps and risk-taking. This implies that historical performance expectation gaps promote risk-taking by driving strategic change. When such gaps occur, managers tend to initiate strategic changes to modify existing business models and improve performance. Model (5) shows that the coefficient of industry performance expectation gap is positive and significant at the 1% level, indicating that industry performance expectation gaps also stimulate strategic change. Based on Model (6), strategic change exhibits a positive partial mediating effect between industry performance expectation gaps and risk-taking. That is, industry performance expectation gaps promote risk-taking through strategic change. When faced with industry performance expectation gaps, managers sense competitive pressure and adjust strategies promptly to avoid elimination [63]. Therefore, the performance expectation gap promotes the risk-taking of agricultural enterprises by facilitating their strategic change.

Given the potential endogeneity issues associated with the three-step method, this study employs the Bootstrap approach to further examine the mediating effect of strategic change. The results indicate that strategic change plays a significant partial mediating role between performance expectation gaps and risk-taking in agricultural enterprises. Specifically, when agricultural firms face performance expectation gaps, managers exhibit higher risk tolerance and are more inclined to pursue strategic changes to improve performance, thereby enhancing risk-taking, as shown in

Table 10. Bootstrap test of mediating effect of strategic change.

Effect	The mediating effect of strategic change between historical performance expectation gap and risk-taking of agricultural enterprises			The mediating effect of strategic change between industry performance expectation gap and risk-taking of agricultural enterprises
	Estimated value	Boot SE	Bias-corrected 95%	Estimated value	Boot SE	Bias-corrected 95%
Indirect effect	0.008952	0.004808	[0.000987, 0.018216]	0.011467	0.006626	[0.001117, 0.036346]
Direct effect	0.489352	0.094839	[0.350489, 0.762811]	0.513881	0.082232	[0.356801, 0.679577]
Overall effect	0.498304	0.096226	[0.355373, 0.772245]	0.525348	0.082579	[0.365674, 0.667803]

.

Thus, performance expectation gaps can promote risk-taking in agricultural enterprises by driving strategic change.

5. Conclusions

Grounded in the Firm Behavioral Theory, Performance Feedback Theory, and Prospect Theory, this research constructs a theoretical framework to examine the relationship between the performance expectation gap and the risk-taking in agricultural enterprises. The main research conclusions are as follows:

(1) Performance expectation gap has a positive impact on the risk-taking of agricultural enterprises. A larger performance expectation gap motivates agricultural enterprises to undertake higher risks. Firstly, agricultural enterprises face high natural risks. When they experience a performance expectation gap, managers develop a sense of crisis to counteract the performance decline caused by natural risks. Driven by this sense of urgency, managers invest in high-risk, high-profit projects to overcome performance difficulties and restore corporate performance, thereby promoting risk-taking [64]. Secondly, the emergence of a performance expectation gap indicates potential issues in current operational strategies and methods. If performance pressure continues to increase, agricultural enterprises are inevitably compelled to alter their original operational, investment, and financing approaches, engaging in strategic change to address current performance challenges. Finally, these negative signals subject agricultural enterprises to skepticism from various sectors of society, damaging their reputation. To protect corporate reputation and enhance enterprise value, managers of agricultural enterprises adopt high-risk decisions and actively innovate [65,66].

(2) The institutional environment positively moderates the relationship between performance expectation gap and risk-taking in agricultural enterprises. That is, the more developed the institutional environment, the greater the promoting effect of performance expectation gap on risk-taking. When agricultural enterprises face a performance expectation gap, a well-developed institutional environment enables them to access low-interest loans, enjoy subsidies, and obtain special funds, thereby reducing risk costs, improving decision-making capabilities, and enhancing their confidence and capacity to take risks [67]. The combination of the high-risk nature of agricultural enterprises and performance expectation gaps creates survival pressure, prompting them to seek investment opportunities and take on more risks. A well-developed institutional environment can shift risk-taking from “blind adventure” to “rational innovation,” thereby reducing risks for agricultural enterprises [68].

(3) From a heterogeneity perspective, the positive impact of performance expectation gap on risk-taking is greater in non-state-owned agricultural enterprises compared to state-owned ones. When facing a performance expectation gap, non-state-owned agricultural enterprises, which primarily aim for profit, have fewer stakeholders and bear less social responsibility. This makes them more motivated by opportunism and inclined to take risks to improve performance. Therefore, non-state-owned agricultural enterprises are more willing to engage in risk-taking when experiencing a performance expectation gap.

(4) Performance expectation gap promotes risk-taking in agricultural enterprises by driving strategic change, which plays a partial mediating role between performance expectation gap and risk-taking. When agricultural enterprises encounter a performance expectation gap in production and operations, managers reflect on whether there have been issues such as unreasonable allocation of resources and strategic missteps. As the performance expectation gap widens and the enterprise faces survival and development crises, managers often proactively choose strategic change to improve performance. Thus, managers are more inclined to pursue strategic change [69].

Furthermore, our research provides practical management insights. From an internal perspective, agricultural firms should strengthen organizational risk management capabilities. This can be achieved through regular seminars to raise risk awareness among managers, engaging risk consultants for strategic evaluation and advice, and building a knowledge base from past experiences to support future decision-making. These steps can help mitigate performance expectation gaps and enhance adaptability. When implementing strategic changes, firms should thoroughly assess internal issues and align changes with the intensity and persistence of performance expectation gaps to ensure consistency with long-term goals and avoid uncontrolled risk exposure. From a managerial perspective, it is essential to set scientifically grounded target performance expectations. By fully integrating internal and external information resources, managers should establish performance goals that align with the enterprise’s development stage, thereby reducing strategic misjudgments caused by expectation biases. Additionally, managers need to enhance their ability to dynamically interpret performance information and make rational decisions based on continuous assessment to improve long-term performance. From an external standpoint, agricultural enterprises should actively monitor and engage with the institutional environment, including fiscal subsidies, tax incentives, industrial guidance policies, and financial support mechanisms. Leveraging such policy support can help reduce risk costs, steer firms away from reckless adventurism and toward reasoned innovation and unlock their potential to respond to performance pressures and undertake risks within a supportive institutional framework.

This research contributes to the existing literature but still has several limitations. First, the research sample requires further supplementation. The data in this study primarily relies on listed agricultural enterprises, excluding unlisted ones. It means that the sample does not cover all agricultural enterprises in China and may not accurately represent the industry. Second, the measurement for indicators needs further refinement. While the study measures performance expectation gap based on historical performance expectation gap and industry performance expectation gap, other measurement approaches should also be directly applied. This discrepancy warrants improvement in future research. Finally, although institutional environment is introduced as a moderating variable, agricultural enterprises face distinct asymmetric policy interventions such as targeted subsidies, etc. Future studies may consider other non-market impacts of these specific policies.