在一个具有研发学习效应的国际寡头框架下,本文将动态不一致、研发学习效应引入到一个两时期四阶段的离散模型中,考察政府与厂商预先承诺能力的差异对产业战略性研发政策的影响。结果表明:完全预先承诺均衡情形下,研发政策纯粹表现为租金转移的激励;仅政府预先承诺均衡情形下,本国厂商为了获得第二期研发优势,第一期将选择过度研发,由此研发政策将包含一项课税激励;序贯均衡情形下,本国厂商为了获得更多的第二期补贴,第一期研发投资更为过度,由此进一步增强了研发政策中的课税激励。 Under an international oligopoly framework with R & D learning effect, this dissertation introduces dynamic inconsistency and R & D learning effects into a four-stage discrete model of two periods and examines the impact of differences in pre-commitment capacity between government and firms on the strategic R & D policy of the industry. The results show that R & D policy is purely incentive for rent transfer under the completely pre-commitment equilibrium. Only when the government pledges equilibrium in advance, in order to obtain the second R & D advantage, domestic manufacturers will choose to over R & D in the first phase, so R & D policy Will include a tax incentive; In the case of sequential equilibrium, the domestic manufacturers in order to get more second subsidies, the first phase of R & D investment more excessive, thereby further enhancing the R & D policy of tax incentives.