This research is to do a study on the technology sourcing strategy for the information technology or the technology of information products while the high-tech domestic corporations are choosing to develop the information system technology. These three strategies are: the way of technology sourcing, the characteristics of technology, and the factors that are beyond the products. To discuss the strategic affected factors while choosing the technology sourcing strategy is the main thesis.The information electronic industry is the main object in this research. Each corporation discusses for the mode of information system technology sourcing. The result shows that ¡§the in-housing R&D ¡¨ is the priority to be considered while high-tech domestic corporations are thinking about information system technology sourcing. But the result also shows that professional knowledge in cost and in the information system technology personnel is insufficient. The studied corporations used the way of ¡§in-housing R&D¡¨ at the beginning, and then they get the technology via outsourcing. The main reason for that is to shorten the period of research time and make the cost lower. In the affected factors among the decided ways of information system technology, the most conspicuous way is the ¡§availability of licensing-in¡¨. The author through deeply interviewing with studied corporations found out that the technology sourcing strategies of information system are multi-channel. The ¡§technology outsourcing¡¨ and ¡§in-house R&D¡¨ are not substitute, however, are complementary. Among the technology sourcing strategies, ¡§the availability and urgency of licensing-in¡¨ plays an important role, that is, the main part of transaction cost items of each cost. As a whole, transaction cost theory could explain the evaluation while the corporations are making the information system strategy.

本研究的樣本資料來自2006年台灣地區第二次產業創新調查，以及後續追蹤調查所建立的資料庫。主要目的為探討政府資助廠商進行創新活動的政策，對於廠商的附加性程度。 在變數的選擇上，本研究以「在民國97-98年獲得政府資助」作為應變數，而以廠商創新能力、廠商技術來源以及前期（民國93-95年）是否曾獲得政府資助，作為本研究的三構面。本研究的分析分為兩階段，第一階段利用Probit迴歸模型，分析三個構面與政府資助之間的相關性，第二階段則是使用Multi-variate Probit模型，分析廠商在無政府資助情況下，對於其創新活動的因應措施。結合兩階段分析結果，定義出四種典型的情境。 實證結果發現： （1）廠商規模愈大、創新能力愈強、技術來源為自行研發或取得外部知識，具以上特質的廠商，與獲得政府資助具正向關係，且廠商規模對於獲得資助具有顯著正向影響。在無政府資助的情況下，具有以上四種特性的廠商較不會改變其創新活動的行為，因此資助政策之附加性低，屬於情境3：所補非人的情況。 （2）技術來源為購買機器設備則與獲得政府資助具有正向關係，當無政府資助時，會「選擇風險較低的創新活動」。代表政府資助對於此類型廠商具有附加性，屬於情境1：雪中送炭的情況。 （3）委外研發對於獲得政府資助具有顯著負向影響，且當無政府資助時，廠商不會「照原訂計畫進行創新活動」，因此資助政策能為此類廠商帶來較高的附加性，屬於情境4：嗷嗷待「補」的情形。 （4）前期獲得政府資助對於後期獲得政府資助並無顯著正向影響。而廠商在無政府資助時，較可能會改變其創新活動的行為，尤其對「縮小創新活動的資金規模」的選項為顯著正相關，因此政府資助對於此類型廠商存在附加性。 本研究著重於廠商成本面的考量，來詮釋補助政策之附加性，然而若要依此制定政策方向，僅以附加性為依據將顯得不足，為此，本研究最後依據建立的架構與實證模型結果，提出若干後續研究方向的建議。 / The study provides insights into the additionality of government subsidy on firms’ innovation activities. Previous research has focused on the relationship between government funding and private R&D input and output, but no consensus were made. Using a sample of 581 observations in the database of Taiwan Technology Innovation SurveyⅡ, the study contains a two-stage analysis. First, the Probit regression model is used, and the dependent variable is "whether a firm gaining government subsidy in 2008-2009 or not". The explanatory variables include firm size, innovation capacity, technology sources and subsidy history. In the second stage, we used the counterfactual scenario about what could happen if there existed no government-sponsored programs intervention. Firms could choose from four alternatives:1. Abandoning innovative activities. 2. Choosing projects which contain lower risk. 3. Downsizing the scale of capital and 4.Keeping the same project. If firms choose the first one, then the additionality of government subsidy is the highest. If firms choose the second or the third one, then the additionality is in the middle. If firms choose the fourth one, then the additionality is the lowest. In this stage, the Multi-variate Probit model is used. The empirical results show that there is a significant positive correlation between firm size and gaining government subsidy, and a significant negative correlation between outsourcing R&D and gaining government subsidy. In the second stage, we find that subsidy toward the firms which "outsourcing R&D and "purchasing machines" have the highest additionality. Besides, subsidy toward firms which received subsidy in 2004-2006 has median additionality.