2023-08-012024-05-18https://scholars.lib.ntu.edu.tw/handle/123456789/701334此二年期研究計畫，衡量將租車服務在時間與空間上化整為零的價值。隨著通訊科技發達，與智慧型手機普及，共享租車在全世界許多大城市持續大幅擴張。共享租車以短時、多點為特色。使用者以手機應用程式，在鄰近的停車場或路邊車位解鎖共享車，便可以小時（甚至15分鐘）計費租用該車。第一年計畫中，我將估計引進共享租車，對於租車需求的影響。第二年計畫中，我將估計結構需求模型，並在反事實分析中，量化共享車帶來的消費者福利變化。第一年的計畫，將使用一般化合成控制模型，估計新增共享車，對於整體租車需求之影響。我將利用2021至2022年，雙北地區漸次引進共享車的自然實驗，研究這種新型態、零散式的租車模式如何影響租車需求，並討論共享車的福利意涵。相較於差異中之差異法，一般化合成控制模型擁有不需平行趨勢假設與兼容異質處理效果兩項優勢，適合估計引進共享車對於消費需求與廠商收益之影響。第二年的計畫，將估計結構租車需求模型。我將建構一個隨機係數離散選擇模型，此模型考量消費者異質性，可用來估計價格與地點之需求彈性。以個人層級訂單、停車位置、車輛特徵、消費者特徵等數據，估計得到之模型係數，將用來進行反事實分析。我將假設消費者只能以天為計價單位租車，或是只能在租車門市租車。這些假設下的模擬結果，將可量化將租車服務化整為零，所帶來的消費者福利變化。 The proposed two-year research project estimates the value of disaggregating rental time and space for car rental services. With the advances in communication technology and widespread popularity of smartphones, car-sharing rental service has been expanding into major cities around the world. Car sharing is characterized by short-term and decentralized rentals. Users can unlock a shared car at a nearby parking spot on the street or in a parking lot using a mobile app and rent the car by the hour or the quarter hour. I intend to estimate the effect of adding a shared car on car rental demand in the first year and estimate a structural demand model to quantify changes in consumer welfare in a counterfactual analysis in the second year. The first-year study estimates the effect of adding a shared car on car rental demand by estimating a generalized synthetic control model. I will exploit a natural experiment resulting from the staggered introduction of car-sharing services in Taipei City and New Taipei City from 2021 through 2022 to study how the new, disaggregated car rental service affects demand for car rental and discuss its welfare implications. Compared to the difference-in-differences approach, the generalized synthetic control method has the advantage of not requiring the parallel-trend assumption and allowing for heterogeneous treatment effects, which will adequately quantify the effects of car-sharing introduction on consumer demand for car rentals and firm revenue from car rentals. The second-year study focuses on the structural estimation of demand for car sharing. I will develop a random coefficient discrete choice model that allows for consumer heterogeneity and estimate the demand elasticities with respect to rate (price) and location. The model parameters estimated using micro-level data on orders, locations, vehicle characteristics, and consumer characteristics will be used for counterfactuals in which consumers are allowed to rent a car by the day only or from car-rental outlets only. These exercises can show changes in consumer welfare attributable to the flexible pricing or convenient locations of car sharing.汽車租賃;共享租車;消費者福利;化整為零;Car rental; car sharing; consumer welfare; disaggregation