本文基于高管创新努力视角，检验政府补贴对高管在研发投资上的激励作用。选取2009—2018年A股制造业上市公司数据，通过固定效应面板回归进行实证检验。研究发现：政府补贴降低了薪酬业绩敏感性，提升了薪酬研发投资敏感性，考虑内生性以及改变核心变量的计量方式后，研究结论保持稳健。拓展分析表明，上述作用在政府补贴强度更高的企业表现显著，对业绩压力大的企业或者连续获得政府补贴的企业效果不显著，表明在补贴方式上，连续补贴不如加大补贴强度效果好。政府补贴降低了企业经营风险，为高管从事风险相对更高的研发投资提供了激励。本文拓展了政府补贴影响企业研发投资的作用机制，为政府补贴政策制定和薪酬契约设计提供了启示。

   R&D investment is a long-term investment with high risks and it faces strong financing constraints, so the incentive is weak for corporate executives to increase R&D investment. Government subsidy can mitigate financing constraints, but the conclusions are mixed regarding the link between government subsidy and R&D investment, because there is a gap from the alleviation of financing constraints to the realization of R&D investment, and it is not clear whether government subsidy will spur executives′ efforts in R&D investment. Thus, based on the perspective of executives′ innovation efforts, the paper investigates directly the incentive effects of government subsidy on R&D investment. Government subsidy can reduce the performance pressure induced by the expensing of R&D expenditure, but also weaken the link between performance and executives′ efforts, then may reduce the sensitivity of pay-performance. At the same time, the performance evaluation is more innovation-oriented as enterprises attach more importance to innovation activities, government subsidy helps change the attitude of the corporate executive to the risk of R&D investment, then may increase the sensitivity of pay-R&D investment. The paper takes the A-share listed manufacturing firms from 2009 to 2018 as a sample, selects the data from CSMAR and CCER databases, and employs a fixed-effect panel-based regression approach to test the predictions. The result shows that government subsidy can reduce the sensitivity of pay-performance while increasing the sensitivity of pay-R&D investment, and the conclusion is robust when considering the endogeneity issue and changing proxies of the main variables. In the section of further analysis, it shows that the above effect is significant in companies with stronger intensity of government subsidy, while the effect is not significant for companies with higher performance pressure based on industry comparison or companies that continuously receiving government subsidy. Government subsidy reduces corporate business risk and provides incentives for executives to engage in R&D investments with relatively higher risk.This paper contributes to the existing literature in the following three aspects. First, it expands the mechanism of how government subsidy affects corporate R&D investment. Prior studies mainly focus on the alleviation effect of government subsidy on financial constraints, this paper focuses on the incentive effect of government subsidy on executives′ efforts in R&D investment with the aid of the sensitivity of pay-R&D investment. Second, it provides empirical evidence on which forms of government subsidy are more effective. The intensity of government subsidy is more effective than its persistence, and the subsidy providing to the enterprises with low performance pressure is more effective. Lastly, it analyses the possible path of how government subsidy affects executives′ innovation efforts. Government subsidy changes the attitude of the corporate executive to the risk and motivates them to put more efforts into R&D investment with higher risk. The research conclusion has three implications in practice. First, the use efficiency of government subsidy should be maximized; the formulation of relevant policy should consider the psychological effect of the corporate executive who receives the subsidy; and the spillover effects should be demonstrated in project selection. Second, it is more effective to motivate executives′ innovation efforts in the form of investing before subsidizing or subsidizing based on innovation results. Lastly, it should consider adding back the expensed R&D investment or adding clauses directly to motivate R&D investment when design compensation contracts.