國立臺灣大學政治科學論叢

Abstract / Chinese PDF Download

The   present   study   explores   whether   the   pattern   of   geographic distribution  of  policy  benefit  among  cities  and  counties  in  Taiwan  is minimum  winning  coalition  or  universalism.    By  looking  at  the  grant allocation of    “Creating New Features of City and Rural Areas” offered by Department  of  Interior  in  the  Taiwan  central  government  from  fiscal  year 1999  to  2000,  we  find  that  the  pattern  of  the  distribution  of  grant  is  either minimum winning coalition or universalism. We say it is minimum winning coalition because the minority exploits the majority of constituencies either we define the policy benefit as the total amount  of  money  or  define  the  policy  benefit  as  the  number  of  projects respectively.    We find that, on the one hand, the disproportionate amounts of money flow to the cities or counties with higher ratio of KMT legislators or with junior legislators. On the other hand, we find that most of projects go to  the  cities  or  counties  with  higher  ratio  of  non-KMT  legislators  or  with senior legislators. When  we  put  these  two  things  together,  we  say  it  is  universalism because  the  members  of  winning  coalition  who  receive  disproportionate amounts  of  money  and  members  who  receive  most  of  the  projects  are exclusive   to   each   other.   In   other   words,   cities   or   counties   receive disproportionate  amounts  of  money  tend  not  to  receive  disproportionate number  of  projects,  and  vice  versa,  in  the  manner  that  each  coalition  gets whatever it needs and would not be deprived of the other. However, we still allege that cities or counties with higher ratio of KMT legislators or with junior legislators are the real winners.    Because we find that  the  coefficient  of  correlation  between  the  amount  of  money  and  the number  of  projects,  which  cities  and  counties  receive,  is  significantly negative  in  the  sense  that  cities  or  counties  receive  more  money  tend  to receive less projects, and vice versa.    It implies that cities or counties with higher  ratio  of  KMT  legislators  or  with  junior  legislators  tend  to  receive more  money  despite  they  got  few  projects.    Based  on  these  results,  we conclude that minimum winning coalition is still at work in the substantive sense (monetary value).