Abstract: A major current focus in e-commerce study is to explore why there are high inputs but low profits in e-commerce. Focusing on the excessive effect of inputs on outputs, market competition rate and mobile e-commerce development, input congestion amount is firstly calculated using data envelopment analysis models, and then e-commerce firm profitability determinants are explored by constructing the panel data regression model. Thirty five pure e-commerce websites from the stock markets of Shanghai, Shenzhen, Hong Kong and NASDAQ are considered, all of which utilize at least one of the two e-business models, namely business-to-business (B2B) and business-to-customer (B2C). The results of DEA models identify that e-commerce suffers from congestion, and the regression shows that asset input congestion has a negative effect on e-commerce firm profitability while market concentration rate and the development of mobile e-commerce can strengthen e-commerce firm profitability. These findings indicate that it is input congestion that leads to low profits. It also enlightens decision makers to strengthen their profitability by eliminating congestion resources and adding new marketing channels such as mobile e-commerce.

Key words: e-commerce, DEA model, congestion, profitability, market concentration