Page 318 - 360.revista de Alta Velocidad - Nº 6

Kim, Junghwa. Schmöcker, Jan-Dirk. Li, Yeun-Touh. Demizu, Fumiaki.

4. Adoption Ratio and City Heterogeneity

We assumed that diffusion effects are different by year after opening as well as by
city. Based on diffusion model, the city-specific annual values of and are obtained as
following.

Equation (3) could be regarded as quadratic equation Y of thus the coefficients could be
t
estimated by a Least Square Method. This estimation was carried out by city and year with the
results shown in Appendix. All estimates by city and year are significant at 0.01% error level,
except for 2007. Table 2 shows the mean value of diffusion rates by city. Here, the diffusion
rates in 2007 were not calculated due to small sample size, therefore we assumed that Adoption
Effects in 2007 is the same as in 2008 for model estimating. Chiayi is showing the highest p value
followed by Taipei/Banciao and Taichung. Taoyuan has the lowest rate of innovative diverted
demand among the cities. For diverted and induced demand, Hsinchu has a high value followed
by Taichung and Taoyuan. Zuoying and Tainan, the cities farthest from the capital, show the
lowest rates of q.

For our focus on expected long-term demand growth, we now suggest the concept of “Adoption
Ratio” obtained as the ratio between p and q:

The “Adoption Ratio” shows how strong the latent impact for increasing a long-term demand
is compared to the initial demand growth during the first few months since HSR operation
starts. The higher the ratio the more we can expect a continuous, fairly steady growth.
Therefore a high ”Adoption Ratio” means strong Adoption Effects.

As shown in Table 2, Taoyuan and Hsinchu both show high adoption ratio. We note that
both cities can be classified as comparatively small cities and both are located close to the
capital. Also when we see Figure 5, we observe that roughly two city groups are divided
that lie above or below the curve for Taipei/Banqiao. Each group has a strong Adoption
Effects the order of city size, Small, Small and Middle i.e. Based on the line of Taipei/
Banqiao(L), Taoyan(S) and Hinchu(S) as well as Taichung(M) are belong to the group above.
And the group below include Tainan(S), Chiayi(S) and Zuoying(M). From these observations
we hypothesize that the city characteristic elements such as size and distance from Taipei,
the capital and economic centre of Taiwan, affect Adoption Effects. Therefore, in the next
section we analyse what factors, which reflect city heterogeneity, determine the level of
Adoption Effects.

316 360.revista de alta velocidad