Compensation Incentive of Executives under the Situation of Synergy or Mutual Exclusion of Corporate Profit and Innovation Tasks: Based on Incentive Game Model between Principal and Agent

. Based on the synergistic or mutually exclusive relationship between corporate profts and innovation tasks, such that shareholders can predict these two scenarios, this study constructed a multitask principal-agent mode considering “the degree of synergy or mutual exclusion between proft and innovation,” “salary reduction penalty,” and other factors. Based on mathematical logic, we determined the correlation and infuence of synergistic or mutually exclusive degrees on executive proftability compensation and salary reduction penalties. Based on the 2016–2022 annual reports of A-share listed companies registered in Shanghai and Shenzhen, China, this research empirically verifes the efects of the aforementioned degree of synergy and mutual exclusion on managers’ compensation incentives. Te research conclusions are stated as follows: (1) If shareholders can predict the synergy between proftability and innovation activities, then the salary reduction penalty (representing executive compensation stickiness) exhibits a signifcant negative correlation with the degree of synergy μ , whereas if the coefcient of executive proftability compensation β i (compensation sensitivity) bears a moderately positive relationship with μ , the shareholders will, to a certain extent, improve the performance-sharing coefcient of senior executives and reduce the penalty coefcient of salary reduction to increase the overall income level of senior executives via the incentive efect. (2) If shareholders can predict the mutually exclusive efect of proftability and innovation activities, the salary reduction penalty (representing executive compensation stickiness) is

Based on the synergistic or mutually exclusive relationship between corporate profts and innovation tasks, such that shareholders can predict these two scenarios, this study constructed a multitask principal-agent mode considering "the degree of synergy or mutual exclusion between proft and innovation," "salary reduction penalty," and other factors.Based on mathematical logic, we determined the correlation and infuence of synergistic or mutually exclusive degrees on executive proftability compensation and salary reduction penalties.Based on the 2016-2022 annual reports of A-share listed companies registered in Shanghai and Shenzhen, China, this research empirically verifes the efects of the aforementioned degree of synergy and mutual exclusion on managers' compensation incentives.Te research conclusions are stated as follows: (1) If shareholders can predict the synergy between proftability and innovation activities, then the salary reduction penalty (representing executive compensation stickiness) exhibits a signifcant negative correlation with the degree of synergy μ, whereas if the coefcient of executive proftability compensation β i (compensation sensitivity) bears a moderately positive relationship with μ, the shareholders will, to a certain extent, improve the performance-sharing coefcient of senior executives and reduce the penalty coefcient of salary reduction to increase the overall income level of senior executives via the incentive efect.(2) If shareholders can predict the mutually exclusive efect of proftability and innovation activities, the salary reduction penalty (representing executive compensation stickiness) is signifcantly positively correlated with the degree of mutual exclusion μ, whereas the coefcient of executive proftability compensation β i (compensation sensitivity) negatively afects the degree of synergy μ to a certain extent; however, this negative relationship is not prominent.Tis indicates that to avert the short-sighted behaviour of executives, shareholders will drive them toward investing in innovation by reducing the proftability compensation coefcient as well as increasing the innovative compensation coefcient.Simultaneously, the shareholders will increase the penalty coefcient of salary reduction to urge the executives to strengthen the operation and management of other proftable activities of the enterprise to maintain short-term performance at a reasonable level.(3) If shareholders cannot predict the synergy between proftability and innovation activities, they will attribute the source of excess earnings to the eforts of the senior executives.In particular, a greater degree of synergy is more likely to reduce the penalty for the senior executives.If the shareholders cannot predict the mutually exclusive efect of proftability and innovation tasks, the correlation between the penalty coefcient of salary reduction and the degree of mutual exclusion is related to the decline in performance.In such a scenario, the shareholders attribute the short-term deterioration in corporate performance to the negligence of the senior executives in corporate operations.Consequently, they urge senior executives to seriously operate the enterprise by increasing the penalty for senior executives.If the decline in performance exceeds the critical value, the shareholders will efectively motivate the senior executives by reducing the penalty coefcient of their salary reduction and moderately increasing their proft-making compensation sharing coefcient to avoid negative investment or their resignation.

Introduction
Innovation is a principal driving force of development, and a nation must rigorously promote its innovation level to improve its competitiveness.In this regard, the sole reliance on scientifc research institutions is insufcient, as enterprises are key drivers of innovation.Moreover, continuous innovation is indispensable for the long-term development of an enterprise.Regardless of the enterprise's proft-making endeavours, it must innovate to achieve progress.Successful innovation by the enterprise will considerably increase its market share and leverage its competitive advantage in the same industry.Specifcally, for enterprises with gloomy development prospects, innovation activities are highly preferred to improve the performance of the enterprise.
Owing to the separation of management rights and proprietary rights and the information asymmetry caused by the enterprise management system, diferences arise in the interests pursued by shareholders and the salary expectations of managers.Terefore, certain diferences will exist between shareholders and executives in terms of risk perception and long-term and short-term interest demands.To this end, Jensen and Meckling [1] proposed the concept of principal-agent, wherein managers and owners form a contractual relationship by signing an agreement, and agents follow the contents of the contract for performing activities such as operation and management.A comprehensive compensation contract should attempt to converge the interests of both the executives and shareholders, such that executives are encouraged to implement the management activities from the perspective of maximizing the value of the company.Under this agency-by-agreement relationship, the avenue of innovation enables shareholders and managers to employ diferent countermeasures.Shareholders preside with an attitude of guiding and supporting the innovation activities, whereas the high risk and high uncertainty involved in executing the innovation projects can degrade the short-term performance of the enterprise, which consequently poses a negative impact on the personal remuneration of senior executives.Terefore, managers tend to be cautious regarding the innovation requirements of shareholders.As observed in several instances, the key to resolving any problem is the convergence of the interests and goals of senior management and shareholders.
Te incentive mechanism for managers is vital to alleviate the conficts of interests between senior executives and shareholders, and its design and implementation have widely concerned scholars and practitioners.In case the salary incentive is not adequately efective, further incentives will motivate opportunistic behaviour among the senior management.To coordinate the confict of interests between the operator and owner, the optimal compensation contract theory forms the basis for providing an optimal compensation incentive contract and minimising the agency cost.Compensation incentives can efectively alleviate the longstanding principal-agent confict of the listed companies, reduce the risk aversion of managers, and efectively restrain the generation of opportunistic behaviour of managers.Terefore, to implement the enterprise innovation strategy considering the current economic background, how to improve and develop the innovation incentive mechanism for managers, adjust the incentive methods in a targeted manner, and guide senior executives to account for both personal interests and enterprise innovation goals constitute the core issues for leveraging innovation advantages at the enterprise level, which demands further research.Tus, this study aims to construct a multitask principal-agent mode that contains both proft and innovation tasks to analyse the most optimal management incentives under the mutually exclusive and cooperative state of enterprise innovation as well as proftable tasks.Furthermore, we propose a targeted incentive adjustment scheme to develop a theoretical basis as well as an empirical reference for improving the senior management incentive mechanism to act as a key driver of enterprise innovation.
Te contributions of this research are stated as follows: (1) overcome the limitations of previous empirical research and construct a multitask principal-agent model considering both proft-making and innovation tasks; (2) deduce the optimal solution of managers' incentive under the guidance of innovation; and (3) mathematically express the logic between enterprise innovation goals and managers' compensation incentives.Tis study aims to provide theoretical support for subsequent related research and promote relevant theoretical research at a deeper level.(2) In the basic principal-agent model, factors such as "mutual exclusion or degree of synergy between innovation and proft" and "salary reduction penalty" are included.Tus, the innovation incentive mechanism of managers should be analysed from the perspective of the performance sharing coefcient and salary reduction penalty, which act as theoretical support for corporate managers' incentive mechanism to develop an adjustment plan and provide practical value along with theoretical signifcance.

Relationship between Innovation and Corporate
Performance.Currently, the relationship between innovation and corporate performance has not achieved consensus among scholars.Certain scholars believe that the relationship between innovation input and frm performance is complementary, i.e., high performance is conducive towards promoting innovation input decision-making, and the performance of a frm will improve with a greater innovation input [2][3][4].Camisón and Villar-López's [5] research results confrmed that organisational innovation is crucial for the development of technological innovation capability, and the adoption of organisational innovation improves the technological capability of enterprises to develop new products and processes, thereby enabling them to achieve higher performance.In addition, Lee et al. [6] inferred a similar research conclusion, reporting that for 2 Complexity high-tech enterprises, the relationship between new products and corporate performance increases with market innovation.In low-tech enterprises, process innovation and organisational innovation pose a direct and positive impact on corporate performance.Boubakri et al. [7] investigated that risk-taking is a crucial driving force for promoting corporate performance.Interestingly, certain scholars have drawn contrasting research conclusions.Tey believe that enterprise innovation input requires a large amount of capital, is difcult to predict, and is often in confict with enterprise performance.For instance, Tong and Zhang [8] demonstrated that high research and development expenditures will create an impact on the short-term fnancial performance of enterprises but will result in higher expected future cash fows.Markarian et al. [9] studied the relationship between innovation input and incremental return on assets, highlighting that enterprise management would consider changes in the current return on assets before implementing R&D investment decisions.If the return on assets diminishes, enterprise innovation input would be reduced as well.Other scholars state that the relationship between innovation input and corporate performance is not a simple linear relationship.For instance, Laursen and Salter [10] considered large industrial enterprises in their sample and identifed that the relationship between the innovation input and corporate performance follows an inverted Ushaped trend.Upon examining the relationship between R&D input, innovation cost, and productivity of manufacturing enterprises, Trachuk and Linder [11] identifed that innovation investment and enterprise performance growth are related nonlinearly, and these two factors develop a strong positive correlation only when innovation investment attains a critical quality.Tus, the relationship between innovation investment and enterprise performance cannot be easily summarised.Relevant innovation activities aid in improving the long-term performance of enterprises.Nonetheless, the high demand for funds for innovation investment will inevitably diminish the short-term performance of the enterprise and might promote myopic behaviour among managers.Tus, managers would continue to strive toward improving the performance of the enterprise, maximise current profts, and disregard innovation decisions that are discussed later in this paper.However, continuous innovation is conducive to the improvement of the core competitiveness of enterprises and will certainly yield positive economic benefts to enterprises in the long term.

Relationship between Management Incentives and Corporate Innovation.
Te essence of motivation is to enable managers to produce positive behavioural responses and achieve the goal through external stimulation.Te research and development investment in enterprise innovation involves high risk and high uncertainty.Te efective incentive for managers aids in improving their sense of responsibility, mobilising the enthusiasm of enterprise afairs participation, and focusing on the proftability of the enterprise.It can restrain the short-sighted behaviour infuenced by the low rate of return on innovation in the short term.Prugsamatz [12] observed that with the correct incentive measures such as those rewarding long-term performance, dominant CEOs can be incentivized to pursue ventures such as innovative projects that are critical towards promoting the long-term growth of their company.External incentives include various management methods and infuencing factors.Currently, scholars categorise management methods into explicit incentives (salary incentives and equity incentives) and implicit incentives (job promotion incentives, etc.).As the fundamental measure of alleviating the principal-agent problem, the compensation contract is dominated by explicit incentives and supplemented by implicit incentives [13].Terefore, this literature review focuses on summarising the impact of salary incentives and equity incentives on innovation.

Impact of Salary Incentives on Corporate Innovation.
Stakeholder management theory holds that the development of a company is closely related to the input or participation of various stakeholders, such that the enterprise pursues the interests of the stakeholders as a whole and not only the interests of certain subjects.Compared with corporate shareholders, senior executives are more risk-averse to innovation.Compensation incentives in listed companies can efectively restrain the long-term confict between principal agents and risk-aversive managers, which can avoid unfavourable situations arising from risky, uncertain projects.Terefore, certain scholars have demonstrated that providing compensation incentive to managers can enhance the enthusiasm of innovation activities and positively afect the innovation input for innovation performance of enterprises [14][15][16][17].Coles et al. [18] believed that increasing executive compensation can indemnify the losses caused when executives abandon short-term high-yielding projects to focus on innovative activities.Terefore, monetary compensation incentives are essential for promoting the innovation activities of enterprises.Cheng [19] argued that a reasonable compensation incentive mechanism can mitigate the risk-aversive attitude of senior executives and encourage them to engage in innovative investment activities.Xu et al. [20] concluded that compensation incentives pose a signifcantly positive impact on the relationship between innovation investment and fnancial sustainability, especially in technology-intensive enterprises.Haider et al. [21] demonstrated that management compensation is positively correlated with frm innovation (represented by R&D investment and intangible assets), and the total compensation provided to executives and specialists motivates them to invest more in R&D activities and innovative products.Similarly, Phung et al. [22] reported that the compensation level of top managers is directly related to the score of ecoinnovation participation in the company.Furthermore, certain scholars believe that managers' compensation is directly linked to enterprise performance, and the confict between innovation input and short-term performance can easily induce short-sighted behaviour among managers, which is not conducive to enterprises' innovation decisionmaking.For example, Tosi et al. [17] stipulated that corporate innovation is a risky long-term investment, whereas compensation incentive is a short-term incentive that urges corporate managers to attend more to projects with short duration of innovation and high returns and abandon innovative investments with longer duration and higher risks, i.e., compensation incentive will inhibit corporate innovation.Te analysis conducted by Zeng et al. [23] reported that executives with a strong salary comparison mentality exhibit less risk-taking and more short-sighted behaviours, thereby reducing their motivation to innovate.Liang et al. [24] reported that salary incentives can motivate managers to strive for improving the short-term performance of enterprises.However, it negatively impacts enterprise innovation, and the salary incentive of managers bears a signifcant negative relationship with the efciency of enterprise innovation as well as research and development.

Impact of Equity Incentives on Corporate Innovation.
An equity incentive is a medium-and long-term incentive tool that combines incentives with constraints.Primarily, it aims to build a sense of ownership among the employees by providing them with certain shareholders' rights and interests, which forms a community of interests within the enterprise, helps the enterprise to achieve the long-term goal of stable development, and motivates investment in research and development.Stulz [25] used this model for the frst time to prove that when the manager's shareholding ratio is excessively high, it will result in a lack of acquisition premium and reduce the value of the company.If managers' shareholding ratio is extremely low, investors are not willing to pay a higher acquisition premium to gain control of the company.Instead, the company should have a unique value of managerial ownership to maximise its value.Currently, several scholars have indicated that equity incentive is positively related to enterprise innovation and research, i.e., executive ownership and innovation benefts display a convergence efect.For instance, Smith and Stulz [26] considered that the incentive method involving the addition of stock options to the executive compensation portfolio can increase the convexity of executive compensation, thereby alleviating the agency problem caused by the inconsistent interests of shareholders and executives.Currim et al. [27] analysed the data of 842 listed companies from 1993 to 2005 and reported that an increase in managers' equity incentive is conducive to the improvement of R&D expenditure by enterprises, thereby promoting the enterprise innovation level.Liu et al. [28] empirical results demonstrated that correctly designed equity incentive plans can promote enterprise innovation activities; however, short-sighted managers will damage enterprise innovation.Only a few scholars such as Que and Ren [29] verifed that the impact of equity incentives on corporate performance exhibits an inverted U-shaped relationship based on the sample data of A-share-listed companies from 2013 to 2016.Although compensation incentives positively impact the relationship between restricted stock incentives and corporate performance, they also have a negative impact on the relationship between stock-option incentives and corporate performance.

Literature Evaluation.
Tere are abundant researches on enterprise innovation, enterprise performance, and manager incentives in the existing literature, and their theoretical achievements provide theoretical support for the follow-up research.However, it can be found that there are still some limitations in the existing theoretical achievements through reviewing the previous literature: (1) Few literature explore manager incentive from the dual perspectives of proft and innovation tasks.Most of the existing literature focuses on the impact of compensation incentives and equity incentives on enterprise innovation, but the literature research on manager incentives from the perspective of the synergistic or mutually exclusive relationship between enterprise proft and innovation is still shallow.( 2) Te synergistic or mutually exclusive relationship between proft and innovation tasks will have an important impact on incentive contracts, for example, by afecting the salary reduction penalty (executive compensation stickiness) and coefcient of executive proftability compensation (compensation sensitivity) in incentive contracts, resulting in diferences in the incentive efect of managers.Te research on this part is very weak, and relevant research literature has not been consulted so far, which needs further exploration.

Research Ideas and Model Building
Te German physicist Hermann Haken frst proposed the "synergy efect," which refers to the enterprises in a cluster because of mutual cooperation, sharing business and specifc resources, and thus achieving higher proftability; in simple terms, the "1 + 1 > 2" efect.Considering this as a reference, the synergy or mutual exclusion between proft and innovation tasks proposed herein refers to whether the innovation activities can yield additional gain for proftmaking activities in the short term or the same period, such that the enterprise's performance produces a synergistic efect of "1 + 1 > 2." On the contrary, if the augmentation of innovation activities requires a disposition of a large corpus of working capital, resulting in a signifcantly lower proft level than the expected target, then the proft and innovation tasks considered to be mutually exclusive.In event that an enterprise opts to perform innovation activities, the executive revenue function depends on whether the mutual exclusion or synergy between the innovation and proft tasks can be predicted by the shareholders.If the shareholders can predict the degree of synergy or mutual exclusion between these two types of activities, shareholders can identify that the signifcant increase or decrease in the corporate performance is related to innovation activities, but not to the executives' eforts.Terefore, the surplus benefts resulting from the innovation activities will not be shared among the senior executives, and the decline in performance caused by innovation will not be attributed to the senior executives.Tus, in both cases, the actual income of the senior executives is only the sum of the fxed remuneration, the remuneration for proft-making activities, and the normal remuneration for innovation activities, excluding the shares of excess or reduced benefts resulting from the synergy or mutual exclusion of profts and innovation tasks.4 Complexity However, if the shareholders cannot predict the synergistic or mutually exclusive relationship between the activities of proftability and innovation, they cannot clearly distinguish whether the increase or decline in performance is caused by the innovation activities or is related to the eforts of the executives.Shareholders can distribute the overall earnings including the excess earnings or declining earnings owing to the synergy or mutual exclusion efect of innovation and proftability to senior executives based on the performance sharing coefcient.In addition to the aforementioned three rewards, the actual revenue of the senior executives shall be additionally provided with a reward based on the performance variations caused by the interaction between innovation and proft.
According to the aforementioned analysis and based on the conventional principal-agent mode, this study considers the synergy or mutual exclusion of corporate proftability tasks and innovative tasks and further subdivides into whether shareholders can predict the synergy and mutual exclusion of two kinds of tasks.Te multitask principalagent model is reconstructed by introducing factors such as "the degree of synergy or mutual exclusion between proft and innovation" and "salary reduction penalty."Assuming that the level of senior executives' proftability efort is a i , the proftability output coefcient is k i , the company's proftability revenue π i can be derived as a linear function of senior executives' proftability efort, expressed as π i � k i a i + ε i , where ε i denotes a random factor afecting the company's output, Similarly, assuming that the level of the top management's innovative efort is a j and the innovative output coefcient is k j , the company's innovative revenue π j can be derived as a linear function of the top management's innovative efort, the linear function expression of the innovative eforts of senior executives is π j � k j a j + ε j , ε j ∼ N(0, σ 2 j ).Assuming that the fxed salary of the company's senior executives is C, the decline in performance is n, n ∈ [0, 1], β i and β j denote the proftability sharing coefcient and innovation sharing coefcient of the executives respectively, β i and β j ∈ [0, 1], λ denotes the salary reduction penalty coefcient, λ ∈ [0, 1], and μ represents the degree of synergy or mutual exclusion between proft and innovation, μ ∈ [0, 1].

Principal-Agent Mode in Which the Degree of Synergy or Mutual Exclusion between Proft and Innovation Can Be
Predicted by Shareholders 3.1.1.Synergies Can Be Predicted by Shareholders.If the proft-making tasks are coordinated with innovation tasks, then the innovation activities will produce additional gains to the proft-making tasks.In case the shareholders can predict the synergy, they can identify the origin of such additional gains as the synergistic efect of the mutual promotion of the innovation and the proft-making tasks, which is unrelated to the individual eforts of the senior executives.Terefore, this portion of the additional gains will not be shared with the senior executives, and the degree of synergy will not impact executives' actual income.As such, the executives' actual income includes fxed remuneration, remuneration for proftable tasks, and remuneration for innovative tasks, expressed as follows: Assuming that the executive efort cost is , (where b denotes the executive efort cost factor), the actual executive beneft can be derived as follows: Terefore, the executive expectation of earnings E(C) can be expressed as follows: Assuming that the executive is an absolute risk avoidant and ρ denotes the absolute risk aversion, the expected utility of the executive can be expressed as follows: According to the relations We can obtain Equation ( 6) expresses the executive incentive compatibility constraint (IC).
Te actual income of the shareholders is the sum of the company's proft income, innovation income, and the Complexity additional income of these two tasks subtracting the executive compensation expenses.As it is in a synergistic relationship, the sum of the additions is μ(π i + π j ), i.e., the actual income of shareholders is expressed as follows: Expected benefts of shareholders is expressed as follows: Assuming that the executive opportunity income is W, the optimal incentive model can be expressed as follows: By solving the above equation, we can obtain According to equation (10), the salary reduction penalty coefcient λ is negatively correlated with the degree of synergy between proft and innovation μ, the proftability sharing coefcient β i is positively correlated with μ, and the innovation sharing coefcient β j is positively correlated with μ, i.e., a smaller salary reduction penalty coefcient increases both the proft and innovation sharing coefcients.
Te results revealed that if the shareholders can predict the synergy between the proft and innovation tasks, a higher degree of synergy that can be predicted by shareholders promotes the tendency of shareholders to enhance incentives by reducing the salary reduction penalty coefcient and improving the proftability sharing coefcient and innovation sharing coefcient of executives.Moreover, the shareholders can predict the economic benefts to the enterprise resulting from the investment in innovation.To motivate senior executives toward attending to innovation and actively enacting decisions on innovation projects, their proftability sharing coefcient will be increased to a certain extent, the salary reduction penalty coefcient will be decreased, and their overall revenue level will be improved.In principle, efective salary incentives will motivate executives to implement business decisions that converge with the shareholders' interests.Consequently, executives will adopt a more positive attitude towards the shareholders' innovative goals, which is benefcial for increasing the shareholders' earnings.
Based on the above model derivation results and analysis, hypotheses H1 and H2 are proposed.H1: When the proft and innovation tasks are synergistic and can be predicted by shareholders, the salary reduction penalty λ is signifcantly negatively correlated with the degree of synergy μ.H2: When the proft and innovation tasks are synergistic and can be predicted by shareholders, the coefcient of executive proftability compensation β i is signifcantly positively correlated with the degree of synergy μ.

Mutually Exclusive Efects Can Be Predicted by
Shareholders.Te executive's actual income is the sum of the fxed salary, proft-making compensation, and innovation compensation.If the proft-making and innovation tasks are mutually exclusive and the shareholders can predict this exclusivity, they can identify that the decline in short-term proft-making income is caused by the innovation activities and is unrelated to the executive's behaviour.Terefore, the degree of mutual exclusion will not infuence the composition of the executive's actual income, i.e., the executive's actual income can be expressed as follows: Terefore, the executive's expected earnings E(C) can be derived as follows: 6 Complexity Assuming that the executive is absolutely risked aversive and ρ denotes the absolute risk aversion coefcient, the expected utility of the executive can be derived as follows: According to the relationship, We can obtain Equation ( 15) formulates the executive incentive compatibility constraint (IC).
If the shareholders can predict the degree of mutual exclusion, the actual return of shareholders is the sum of the company's proft income, innovation income, and the additional income of proft and innovation task subtracting the executive compensation expenses.Moreover, if the proft and innovation tasks are mutually exclusive, the resultant of this sum is μ(− π i + π j ), i.e., the actual income of the shareholder can be expressed as follows: Te shareholders' expected earnings E(V) is Assuming executive opportunity returns as W, the optimal incentive model can be expressed as follows: By solving the above equation ( 18), we can obtain According to equation (19), the degree of mutual exclusion of the task μ is positively correlated with the salary reduction penalty coefcient λ as well as the innovation sharing coefcient β j , whereas it is negatively correlated with the proftability sharing coefcient β i .
Te results demonstrated that if the proft and innovation tasks are mutually exclusive and can be predicted by shareholders, a higher degree of mutual exclusion that can be predicted by shareholders increases the salary reduction penalty coefcient set for senior executives and decreases the proft sharing coefcient.Nonetheless, the innovative sharing coefcient will increase.In the shortterm period, innovation activities may decrease proft if the innovation investments consume an extensive amount of funds that could otherwise be used for proftable activities.To maintain both performance and income levels, senior Complexity executives may be inclined to abandon projects with a more gradual return on innovation investments.Terefore, to efectively avoid myopic behaviour of senior executives, if the shareholders can accurately predict the mutually exclusive relationship between proft and innovation, they will motivate senior executives to invest in innovation by reducing the proftability sharing coefcient and promoting the innovation sharing coefcient.Simultaneously, they will urge senior executives to strengthen the operation and management of other proftable tasks of the enterprise by increasing the salary reduction penalty coefcient, which aims to maintain short-term performance at a reasonable level.
Based on the above model derivation results and analysis, hypotheses H3 and H4 are proposed: H3: When the proftable tasks and the innovative tasks are mutually exclusive and can be predicted by shareholders, salary reduction penalty λ is signifcantly positively correlated with the degree of mutual exclusion μ.H4: When the proftable tasks and the innovative tasks are mutually exclusive and can be predicted by shareholders, the coefcient of executive proftability compensation β i is signifcantly negatively correlated with the degree of mutual exclusion μ.

Principal-Agent Mode in Which the Degree of Synergy or Mutual Exclusion between Proft and Innovation Cannot
Be Predicted by Shareholders 3.2.1.Synergies Cannot Be Predicted by Shareholders.If the shareholders cannot predict the synergy and fail to identify the additional revenue generated by the synergistic efect of the innovation and proftability tasks on the company's overall revenue, they will share the overall income with senior executives according to a certain proportion and rules without distinction.In addition to the sum of fxed salary, proft-making compensation, and innovation compensation, the actual income of the senior executives should include the additional proftability and innovation compensation that is infuenced by the degree of synergy.Terefore, the actual income of senior executive can be expressed as follows: Te executive's expected earnings E(C) can be expressed as follows: Moreover, the executives expect utility as follows: Upon partial derivation for a i and a j , we obtain Te actual income of shareholders is the sum of the company's proft income, innovation income, and the additional income from proft and innovation task subtracting the executive compensation expenses.Te sum of the bonuses in the synergistic relationship is μ(π i + π j ).Tus, the expected income of the shareholders can be stated as follows: Assume that the executive opportunity return is W.
8 Complexity Terefore, β * j bears no relationship with μ and λ.Analysis of derivation results (26): ; thus, if the proftmaking and innovation tasks are coordinated and unpredictable, the salary reduction penalty coefcient λ is negatively correlated with the degree of synergism μ.
, the relationship between β i and μ at this instant cannot be derived owing to the loosely defned positive and negative conditions in the numerator k 2 i − λk 2 i − λρb i σ 2 i .If the shareholders cannot predict the synergy between the proft-making and innovation tasks, the relationship between the degree of synergy μ and the penalty factor λ is negatively correlated, i.e., the greater the degree of synergy, the less the shareholders will penalise the executives on salary.Shareholders do not predict that the reason for the signifcant improvement of the company's short-term performance corresponds to the excess returns generated by the synergy of innovation and proft tasks.Terefore, they may attribute this additional revenue to the eforts of the senior management.To maintain enthusiasm and reward the efforts of the senior executives in the operation process, the shareholders will reduce the salary reduction penalty of the senior executives.In addition, as shareholders are unaware of the additional gains resulting from the synergistic efect of innovation and proft tasks, they cannot assess the necessity of further motivating the senior executives to enact innovation decisions.Terefore, they may not continue to motivate the senior executives by adjusting the innovative sharing coefcient.

Mutually Exclusive Efects Cannot Be Predicted by the
Shareholders.If the mutual exclusion efect cannot be predicted by the shareholders, in the total calculation of fxed salary, proft-making compensation, and innovative compensation, the loss of proft-making compensation caused by the mutual exclusion shall be deducted, and the innovative compensation afected by the degree of mutual exclusion shall be included.Terefore, the actual income of senior executives can be expressed as follows: Executive expected revenue E(C) can be expressed as follows: Te expected utility of senior management is stated as follows: Upon the partial derivation of a i and a j , we obtain Complexity a i � Te expected earnings of the shareholders can be calculated as follows: Assuming that the executive opportunity income is W, the optimal incentive model can be expressed as follows: Terefore, β * j bears no relationship with μ and λ.Analysis of derivative results (33): If the proft-making and innovation tasks are mutually exclusive and cannot be predicted by shareholders, the derivative of λ * with respect to μ can be obtained as , which indicates that the derivative yields a critical value.For (zλ * /zμ) ≥ 0, we obtain 0 ), the salary reduction penalty coefcient λ is positively correlated with the degree of mutual exclusion μ, and for 1 − (β i /A) < n ≤ 1, they are negatively correlated.Moreover, the proft sharing coefcient β i of executives is positively correlated with the degree of mutual exclusion μ.
If the proft-making and innovation tasks are mutually exclusive and not predicted by the shareholders, the correlation between the salary reduction penalty coefcient and the degree of mutual exclusion depends on the critical value of the performance decline range.For a small range of performance decline such as 0 ≤ n ≤ 1 − (β i /A), the salary reduction penalty coefcient is positively correlated with the degree of mutual exclusion.In contrast, for 1 − (β i /A) < n ≤ 1, the salary reduction penalty coefcient exhibits a negative correlation with the degree of mutual exclusion, i.e., a greater degree of mutual exclusion reduces the penalty coefcient imposed by the shareholders on the remuneration of senior executives.Based on this analysis, if the shareholders cannot estimate the mutually exclusive efect and if the decline in corporate performance is greater than or equal to 0 but does not reach 1 − (β i /A), they attribute the marginal decline in short-term corporate performance to the negligence of the senior executives in corporate operations.Te salary reduction penalty coefcient will be increased to urge senior executives to seriously operate the enterprise reduction.In case the performance decline exceeds 1 − (β i /A) and is less than or equal to 1, the shareholders may realize that the corporate performance declined primarily because of the extensive amount of capital investment in innovation activities.Concurrently, shareholders should reduce the penalty coefcient to avoid negative actions among the executives, e.g., avoiding investment in innovation tasks or resigning.Moreover, shareholders can moderately strengthen the proft-sharing coefcient of the executives to strengthen the incentive efect.
Te aforementioned research results are summarised in Table 1.
Based on the results of model derivation and critical values, uncertain and irrelevant situations may arise in case the proftable and innovative tasks are synergistic or mutually exclusive, which cannot be predicted by the shareholders.Tus, this research did not propose research hypotheses for these two cases.In the following section, we empirically tested only two cases in which the proftable and innovative tasks are synergistic or mutually exclusive and can be predicted by shareholders.Accordingly, the empirical results are further analysed.Te data were primarily acquired from the Guotai'an database.In certain cases, the wind database was used to supplement missing data.Initially, Microsoft Excel was used for simple data processing, and later, Stata17.0 was used for empirical analysis.We winsorised up or down 1% of each continuous variable to eliminate the infuence of extreme values.

Explained Variables and Explanatory Variables.
Te economic connotation and function of the salary reduction penalty λ, incorporated in the stated model, are extremely similar to the role of executive compensation stickiness, both of which refer to the incentive systems formulated for tolerating the short-term failure behaviors of senior executives.Terefore, in the empirical process, the executive compensation stickiness (ECS) was used as a proxy index for the salary reduction penalty λ.As such, the coefcient of executive proftability compensation is related to enterprise performance, and an increase in enterprise performance directly results in higher remuneration for senior executives and encourages them to focus on enterprise development, which is consistent with the defnition and connotation of the sensitivity of executive compensation.Terefore, in the following empirical analysis, we employed compensation sensitivity as an alternative index of executive proftability compensation, and the calculation method is based on the practice of Li and Guo [30].Furthermore, a regression model between executive compensation and the enterprise's performance is established, considering compensation sensitivity as the regression coefcient of the corporate performance index.Te corporate performance is measured using return on equity (ROE).
(1) Explained variables in the empirical model (a) Executive compensation stickiness (ECS).Executive compensation stickiness provides rewards to executives when performance increases, but it does not penalise the executives when performance declines [31].Executive compensation stickiness refers to the phenomenon that executive compensation is not easily afected by external factors and variations in a short period of time.Numerous studies have explored executive compensation stickiness worldwide.Based on the practices of Lei and Guo [32] and Xu et al. [33], this study measures the executive compensation of the top-three executives and adopts the following steps to calculate executive compensation stickiness: frst, we evaluated the annual growth rate of executive compensation and the growth rate of corporate net proft (2012-2022) relative to previous year; second, the annual growth rate of executive compensation was divided by the growth rate of net proft to obtain the sensitivity of the annual variation in executive compensation relative to that in net corporate proft.Tird, the mean sensitivity of variations in executive compensation relative to those in corporate net proft was evaluated when the net proft increases and decreases across each year of the sample period (2016-2022) and the rolling fve-year period of the frst four years; fourth, the rolling fve-year executive compensation stickiness (ECS) of executives of each company is obtained by subtracting the increase in rolling average of fve-year sensitivity of net proft from the decrease in rolling average of fveyear sensitivity of net proft.Stata 17.0 is used to perform the aforementioned calculation steps.(b) Executive compensation (Lnpay).In the calculation method, the natural logarithm of the average compensation of the top-three executives of the listed company was obtained, and the topthree executives' compensation disclosed in the annual report of the company was used as the data source.In the process of examining the infuence of the degree of synergy or mutual exclusion between proftable and innovative tasks on the compensation sensitivity, executive compensation was selected as the explained variable and measured by the natural logarithm of the average compensation of the top-three senior managers.
(2) Explanatory variables in the empirical model (a) Te degree of synergy between proft and innovation and the degree of mutual exclusion between proft and innovation (μ) were Complexity considered the explanatory variables in the empirical method.Te synergy between the proftability and innovative tasks proposed herein indicates that conducting more innovation activities in the short term or the same period can yield additional gain from proftable activities.In contrast, if the augmentation in innovation activities produces a signifcantly lower proft of the enterprise than the expected target, the proft and innovation tasks are deemed as mutually exclusive.Te performance growth rate is selected as the index to measure the variations in the enterprise's proftable tasks results, and the innovation input growth rate is selected as the primary index to refect the changes in the enterprise's innovation task results.Prior to testing, the size of μ should be quantifed.In this research, the relative value of the performance growth rate/innovation input growth rate was selected to measure the degree of synergy and mutual exclusion of enterprises.(b) Return on equity (ROE).Te return on equity refects the level of return on the shareholders' equity and measures the efciency of a company's utilization of its own capital.A higher index value of ROE corresponds to a higher return on investment.Herein, the ROE is used to measure the corporate performance, and the regression coefcient between ROE and Lnpay can refect compensation sensitivity.

Control Variable.
To improve the explanatory power of the empirical research results, the following control variables are selected in this paper: the impact of assetliability ratio (LEV), corporate growth (GROWTH), shareholding ratio of the largest shareholder (GQP), proportion of independent directors (RATIO), board size (LNBOARD), and double duty (DUAL) on corporate performance.Concurrently, three dummy variables are set: province (PROVINCE), industry (IND), and year (YEAR).
Te specifc defnitions and measurement methods of each variable in this paper are listed in Table 2.

Construction of a Model of the Infuence of the Degree of Synergy between Proft and Innovation Tasks on Managers'
Compensation Incentive.To verify the above-mentioned hypothesis H1, when proft and innovation tasks are synergistic and can be predicted by shareholders, the salary reduction penalty λ (executive compensation stickiness, ECS) is signifcantly negatively correlated with the degree of synergy μ.Te regression model is established as depicted in In model (34), we emphasise on coefcient α 1 , which should be signifcantly negative according to the proposed hypothesis.In particular, ε denotes a random item, whereas the remainder are control variables, as listed in Table 1.
To verify hypothesis H2, when proft and innovation tasks are synergistic and can be predicted by shareholders, the salary reduction penalty β i (compensation sensitivity) is positively correlated with the degree of synergy μ.A regression model is established, as depicted in In model (35), μ × ROE, the cross-multiplication coefcient μ × ROE of μ and ROE, is used to refect the relationship between the degree of synergy and compensation sensitivity.Terefore, we focus on the cross-multiplication coefcient β 3 .If it is positive, the degree of synergy will promote the improvement of the compensation sensitivity, which is in line with hypothesis H2.

Construction of a Model of the Infuence of the Degree of Mutual Exclusion between Proft and Innovation Tasks on
Managers' Compensation Incentive.To verify the proposed hypothesis H3, when proft and innovation tasks are mutually exclusive and can be predicted by shareholders, the salary reduction penalty λ (ECS) is signifcantly positively correlated with the degree of mutual exclusion μ.Te regression model is established as expressed in (36) In model (36), we focus on coefcient c 1 , which should be signifcantly negative according to the above hypothesis.ε denotes a random item.
To verify hypothesis H4, when proft and innovation tasks are mutually exclusive and can be predicted by shareholders, the salary reduction penalty λ (ECS) is signifcantly negatively correlated with the degree of mutual exclusion μ.Te regression model is established as expressed in (37) In model (37), μ × ROE, the cross-multiplication coefcient μ × ROE of μ and ROE, refects the relationship between the degree of mutual exclusion and compensation sensitivity.Terefore, focusing on the cross-multiplication coefcient θ 3 , if it is negative, the model conforms to hypothesis H4.

Descriptive Statistical Analysis.
In this study, the ratio of performance growth rate to innovation input growth rate (i.e., μ) was used to segment the synergistic and mutually exclusive research samples of enterprises.Samples with μ > 0 were categorized into the synergistic group of proft and innovation tasks, and those with μ < 0 were classifed into the mutually exclusive group of proft and innovation tasks.Among them, the synergistic group contained 5238 observations, whereas the mutually exclusive group comprised 1968 observations.Te descriptive statistical analysis results of the two groups of sample data are listed in Tables 3 and 4.
Te descriptive statistical results of the fundamental variables when the enterprise proft and innovation tasks are in synergy are listed in Table 3.As observed, the average value of executive compensation (Lnpay) among the listed companies in China was 13.660 from 2016 to 2022, the minimum value was 10.310, and the maximum value was 17.100, indicating a large diference in executive compensation among the listed companies in China.Te average value of ECS is 0.000, indicating almost no diference between the increase in executive compensation when the performance increases and the decrease in executive compensation when performance declines to the same degree.Te minimum and maximum values were 6.904 and 5.583, respectively.Note that a larger value in this regard indicates a weaker tendency to "penalise bad," or even "rewarding bad."Te mean value of ROE is − 0.032, indicating that the performance of the selected sample companies is relatively normal, whereas the minimum and maximum values are − 186.600 and 1.598, respectively, indicating a large gap in proftability among listed companies in China.
Te descriptive statistical results of the fundamental variables when proft and innovation tasks are mutually exclusive are listed in Table 4.As observed, certain diferences existed in executive compensation (Lnpay) among the listed companies in China from 2016 to 2022.Te average value of executive compensation stickiness (ECS) is − 0.029, indicating that the increase in the rate of executive pay in cases of improved performance is 0.028% lower on average than the decrease in the rate when performance declines to an equivalent degree.Te minimum and maximum values are − 7.222 and 3.874, respectively, and the overall value of the ECS is lower than that when the enterprise proft and innovation tasks are in synergy.Terefore, under the condition of mutual exclusion of proft and innovation tasks, the shareholders will constrain the executive's compensation, and the higher stickiness of compensation will not prevail under the synergistic condition.Te average return on equity (ROE) is − 0.078, implying that the performance of the selected sample companies is relatively normal, whereas the minimum and maximum values are − 176.400 and 64.060, respectively, suggesting a large gap between the proftability of listed companies in the state of mutual exclusion.
As evident from Tables 3 and 4, whether the proftable tasks and the innovative tasks are in the synergistic state or the mutually exclusive state, the absolute value of μ (after processing) is ∼0.2, indicating a marginal diference between the level of synergy and mutual exclusion.Overall, the sample data selected by the two are reasonable.

Correlation Analysis.
Correlation analysis was conducted on the sample data of the synergy group and the mutual exclusion group, and the results are summarised in Tables 5 and 6.
As observed from the correlation analysis results in Table 5, when the proft and innovation tasks are in cooperation, μ is signifcantly negatively correlated with ECS, with a correlation coefcient of − 0.076, which passes the Pearson test with a signifcance level of 1%.In addition, a signifcant positive correlation was detected between ROE and Lnpay, with a correlation coefcient of 0.232, which passed the Spearman test with a signifcance level of 1%.Moreover, in the Spearman test, μ is significantly negatively correlated with ECS, with a 1% signifcance level.
As observed from the correlation analysis results stated in Table 6, when proft and innovation tasks are mutually exclusive, μ is signifcantly positively correlated with ECS, with a correlation coefcient of 0.039, which passes the Pearson test with a signifcance level of 10%.In particular, a signifcant positive correlation exists between ROE and Lnpay, and the correlation coefcient is 0.190, which passes the Pearson test with a signifcance level of 1%.Furthermore, 14 Complexity  Upon analyzing the overall control variables, the following conclusions can be inferred: when the proftable tasks and the innovative tasks are synergistic, μ is negatively correlated with LEV, GQP, LNBOARD, GROWTH, and RATIO, and positively correlated with DUAL.If the profitable tasks and the innovative tasks are mutually exclusive, μ is positively correlated with LEV, GQP, LNBOARD, and GROWTH.In particular, it is signifcant between LNBOARD, DUAL, and GROWTH, attaining the signifcance level of 1% and 10%.Tere is a negative correlation between DUAL and RATIO.As the coefcients among the explanatory variables, explained variables, and control variables are relatively small, no multicollinearity problem between the variables is selected in this paper.

Regression Analysis
(1) Efect of the Degree of Synergy on the Incentive of Managers' Compensation.Based on the previous mathematical logic reasoning, regression analysis is conducted on hypotheses H1 and H2, and the year, industry, and province are respectively controlled.Te regression results are listed in Table 7.
Model (34) derives the regression between the salary reduction penalty λ (ECS) and the degree of synergy μ when the proft and innovation tasks are synergistic and can be predicted by shareholders.As observed from Table 5, μ is signifcantly negatively correlated with ECS, the correlation coefcient is − 0.041, and the signifcance level attained 1%.Terefore, a higher degree of synergy predicted by shareholders indicates a stronger inclination among shareholders to motivate executives by reducing the salary reduction penalty, which is consistent with hypothesis H1.Model (35) tested the regression of the relationship between compensation sensitivity (coefcient of executive proftability compensation β i ) and degree of synergy μ.According to the empirical test results, the coefcient of cross multiplication term μ × ROE was evaluated to be 0.004, but failed the signifcance test.In case of regression considering multiple factors, although μ was positively correlated with the compensation sensitivity, it was not signifcant.Tus, Hypothesis H2 is only partially verifed.However, the cross-multiplication coefcient is positive, which explains the positive relationship between β i and μ to a certain extent.Section 4.4.4 will conduct a supplementary test of the correlation between the two without considering other factors.
(2) Efect of the Degree of Mutual Exclusion on the Incentive of Managers' Compensation.Tereafter, regression analysis was performed on hypotheses H3 and H4, and the regression results are listed in Table 8.
Model (36) expresses the regression between the salary reduction penalty λ (ECS) and the degree of mutual exclusion μ when the proft and innovation tasks are mutually exclusive and can be predicted by shareholders.Table 8 reports that μ is signifcantly positively correlated with ECS, the correlation coefcient is 0.037, and the signifcance level attains 10%.Terefore, a higher degree of mutual exclusion that can be predicted by shareholders corresponds to a stronger inclination among shareholders to encourage executives to invest by increasing the penalty coefcient of salary reduction, which is consistent with hypothesis H3 above.
Model (37) is a regression to test the relationship between compensation sensitivity (coefcient of executive proftability compensation β i ) and degree of mutual exclusion μ.According to the results shown in the table, the coefcient of the cross multiplication term μ × ROE is

Research Conclusions
Based on the development goals of the shareholders' innovation activities and the current interests of shareholders and executives, this research proposes the coefcient of executive proftability compensation, considering the degree of synergy or mutual exclusion between proft and innovation tasks.Based on the classical principal-agent model and the results from the derived mathematical model, we discussed the incentive mechanism of managers under the guidance of innovation.Based on the 2016-2022 annual reports of A-share listed companies registered in Shanghai and Shenzhen, China, this research explores the impact of the degree of synergy and mutual exclusion on managers' incentives.Te conclusions are stated as follows: (1) When the proftable and innovative tasks are synergistic and can be predicted by shareholders, salary reduction penalty λ (executive compensation stickiness ECS) is signifcantly negatively correlated with the degree of synergy μ; (2) When the proft and innovation tasks are synergistic and can be predicted by shareholders, the coefcient of executive proftability compensation β i (compensation sensitivity) exhibits a positive relationship with the degree of synergy μ to a certain extent; (3) When the proftable and innovative tasks are mutually exclusive and can be predicted by shareholders, the salary reduction penalty λ (salary reduction penalty ECS) is positively correlated with a mutual exclusion degree μ; (4) If the proft and innovation tasks are mutually exclusive and can be predicted by shareholders, the coefcient of executive proftability compensation β i (compensation sensitivity) and the mutual exclusion degree   Complexity μ are engaged in an inverse relationship to a certain extent; however, this negative relationship is not prominent.Te research results will provide a reference for shareholders to improve the innovation management system, strengthen the manager's innovation incentive scheme, and provide a theoretical basis for the design, optimization, and adjustment of the compensation incentive contracts of enterprises' managers.

Management Implications
As enterprise owners, shareholders consider the long-term benefts of innovation activities to the enterprise.Simultaneously, considering that innovation decision-making is conducive to promoting the enterprise's strategic position, shareholders are more expected to participate in the decision-making of innovation tasks compared to senior executives.In particular, executive compensation is linked to the short-term performance of the enterprise, which will prioritise self-interest in handling innovation tasks with high risk and possible long payback periods, and the executives may act cautiously toward innovation decision-making, which conficts with the long-term proft target pursued by shareholders.To converge the interests of shareholders and senior executives, shareholders should adjust the remuneration contracts of senior executives according to various situations.On the premise of predicting the relationship between proft and innovation tasks, the selection of reasonable salary incentives such as adjusting the profitability and innovation sharing coefcients for senior executives can motivate senior executives towards the innovation decisions of the enterprise or urge senior executives to strengthen operation management by varying the salary reduction penalty.Te aforementioned incentives can efectively improve the consistency of interests and goals between the shareholders and senior executives, which is benefcial for the long-term development of the enterprise.
In addition, the shareholders should completely grasp the relevant information about the innovation activities, strengthen the comprehensive assessment of the innovation projects, accurately assess whether the innovation activities can yield additional proftability gains or diminish the proftability gains in the short term, adjust the synergy or mutual exclusivity between proftability and innovation tasks from unpredictable to predictable, and timely select the appropriate means to adjust the senior executives' incentive model in a targeted manner.Tus, enhancing the enthusiasm of senior executives toward enacting innovative decisions can achieve win-win results between shareholders and senior executives.
Based on the development goal of shareholders' innovative activities and the present situation of shareholders and senior executives' interests, this study contributes certain key factors such as the performance sharing coefcient of innovative tasks, the degree of mutual exclusion between proftability and innovative tasks, and the salary reduction penalty coefcient based on the classical principal-agent model to discuss the incentive mechanism of managers under the guidance of innovation according to the results derived from the mathematical model.Te research results will be insightful for shareholders to improve the innovation management system, enhance the innovative incentive scheme of managers, and provide a theoretical basis for designing, optimizing, and adjusting the compensation incentive contract of enterprise managers.

Table 2 :
Names, symbols, and measurement methods of variables.Shareholding ratio of the largest shareholder GQP Number of shares held by the largest shareholder/total number of shares in the company Proportion of independent directors RATIO Number of independent directors/Number of board members Board size LNBOARD Logarithmic number of board members Double duty DUAL When DUAL � 1, the chairman and the general manager are the same individual; when DUAL � 0, they are not the same person Province PROVINCE Virtual variable.If it belongs to the province of the sample, PROVINCE � 1; otherwise, it is 0 Industry IND Virtual variable, which belongs to the industry of the sample, IND � variable, which belongs to the year in which the sample resides, YEAR �

Table 1 :
Derivation results of principal-agent model.Synergy between proft and innovation tasks is predictable Negative correlation Positive correlation Positive correlation Mutual exclusion between proft and innovation tasks is predictablePositive correlation Negative correlation Positive correlation Synergy between proft and innovation tasks is unpredictable

Table 3 :
Descriptive statistical results of major variables (synergistic).

Table 4 :
Descriptive statistical results of major variables (exclusive).

Table 5 :
Correlation analysis between proft and innovation tasks (synergistic).Note.* * * signifcant at 1% level, * * signifcant at 5% level, * signifcant at 10% level.Te lower-left corner presents the result from Pearson correlation analysis, and the upper-right corner represents the results from Spearman correlation analysis.

Table 6 :
Correlation analysis between proft and innovation tasks (exclusive).
Note. * * * : signifcant at 1% level; * * : signifcant at 5% level; * : means signifcant at 10% level.Te lower left corner is the Pearson correlation analysis result, and the upper right corner is the Spearman correlation analysis result.Complexity  15in the Spearman test, μ is signifcantly positively correlated with ECS, with signifcance at the 10% level.
Note. * assets is selected as the index to measure the compensation sensitivity, expressed by ΔLSalary/ΔTotalAsset. Without considering other factors, when the proftable and innovative tasks are mutually exclusive, μ is signifcantly negatively correlated with ΔLSalary/ΔTotalAsset at the 10% signifcance level, and the correlation coefcient is − 0.08.Hypothesis H4 is further verifed.When the proftable tasks are synergistic with the innovative tasks, μ is positively correlated with ΔLSalary/ΔTotalAsset, but not to a signifcant level.Terefore, H2 is only partially verifed.