技術擴散與外部性

Abstract

本文建構一個人策略互動的搜尋模型(search-theoretic model) 以探討技術採用。技術先驅者(pioneers) 在技術市場中販賣技術予技術執行者(workers), 技術執行者以所學得之技術在財貨市場上生產商品並彼此交換。其中,技術費用(tuition)起初為外生給定, 最後將放寬此一假設, 改以奈許協商(Nash Bargaining Problem)而得。技術市場與財貨市場均為隨機配對之非集中交易市場。本文探討在不同均衡下,何種技術以何種價格被採用。本文亦同時探討政府政策對技術採用之影響。 本文發現在某些參數設定下,本文模型只存在單一均衡;而在另一些參數設定之下,將存在複均衡。若技術費用為外生給定, 在只有一種技術被此經濟體所接受的情形下, 此費用至少必須能夠支抵技術先驅者的採用成本現值; 同時, 此費用亦不能過高, 以致技術執行者無法負擔。而技術執行者利用學得之技術在財貨市場生產,彼此交換商品 消費所得之效用必須足以支付由生產成本與技術費用現值所構成之總成本。另一方面, 若所有技術同時被接受, 不同技術之技術先驅者因傳授技術而得之淨收益不可差距過大。同樣地, 而技術執行者利用學得之技術在財貨市場生產,彼此交換商品消費所得之效用必須足以支付總成本與預期總成本差異。 若技術費用是由交易雙方以奈許協商所決定,技術費用將隨著技術執行者的總數減少或財貨市場的交易便利性增加而增加。然而,技術先驅者發現技術的機率對技術費用有著非線性之影響。本文亦發現, 若技術有成本外部性的特性, 技術費用將比沒有成本外部性為高。 最後,當政府干預市場的程度夠深,政府將可隨心所欲地挑選任一技術成為市場上唯一的技術。若財貨市場的交易便利,政府應採取需求面政策干預市場以減少干涉的程度。

We explicitly consider strategic interaction between agents to study technology adoption and its effects on the prices of technologies,welfare and government policies in a two technologies search-theoretic model. We model the technology market where pioneers can sell technologies to workers, who can implement them, and the goods market where workers can trade their consumption goods produced by the technology. The markets are decentralized with random matching and the prices of technologies are specified exogenously at first and decided by bargaining in the end. The technologies which get adopted and their prices under different equilibia are discussed in the paper. In terms of results, steady-state equilibria were characterized in several versions of the model. For some specified parameters there is a unique steady-state equilibrium; for others there are multiple equilibria. If tuition paid to a technical pioneer agent is specified exogenously,with only one technology adopted by the economy, tuition should be able to cover the present value of adoption cost and should not be so high that an unskilled worker can not afford it. The utility gained from consuming in the goods market should be able to cover the total cost, which is composed of the production cost and the present value of tuition. On the other hand, if all technologies are adopted, the difference between the net gain from imparting the two technologies can not be too large. Likewise, the utility gained from consuming should be able to cover total cost and the expected cost difference. If tuition is determined endogenously, tuition increases with an decrease in the size of workers population or with a decrease in trading friction in the goods market. The arrival rate of technology, however, has an nonlinear effect on tuition. This model also shows that tuition under externality of production cost is higher than not being under it. At last, the government can choose any technology it wants when the government size is big enough. It was determined that the government should implement the demand side policy when trading friction in the goods market is not severe.