This paper makes an attempt to estimate the effect of different levels of urban compactness on travelling distance and time for various means of transportation, such as on foot, public transportation, or by car, in the compact cities where population is concentrated in the center of the cities. The travelling distance is shorter in the cities with high compactness where the residents are concentrated in the center of cites, because the scale of the urban area of the cities is smaller, and, as a result, the travelling time is longer where the means of transportation is public transportation or on foot, while the travelling time by car is smaller, and it is expected that urban compactness eventually contributes to maintaining the health of residents, and reducing both energy consumption and the environmental burden.
In this case, the level of urban compactness is an endogenous variable, which is influenced by the public transportation networks such as rail or bus, and the estimation by Ordinary Least Square (OLS) analysis might be biased by endogeneity. In addition, the effect of the level of urban compactness on the travelling time for various means of transportation may be different depending on age, sex, occupation and form of employment.
Therefore, we estimate the effect of "Normalized Standard Distance" (NSD), the indicator of the effect of urban compactness on the travelling distance and travelling time for various means of transportation, including on foot, public transportation (railroad, bus) or car by using instrumental variables for extracting the endogeneity bias. We also estimate effects of NSD on the travelling distance and time, focusing on age, sex, occupation and form of employment.