Premium
DIMENSIONS, TRANSACTIONS COSTS AND COALITIONS IN LEGISLATIVE VOTING *
Author(s) -
Koford Kenneth
Publication year - 1990
Publication title -
economics and politics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.822
H-Index - 45
eISSN - 1468-0343
pISSN - 0954-1985
DOI - 10.1111/j.1468-0343.1990.tb00023.x
Subject(s) - voting , legislature , ideology , dimension (graph theory) , cohesion (chemistry) , space (punctuation) , curse of dimensionality , econometrics , economics , positive economics , political science , public economics , computer science , politics , law , mathematics , artificial intelligence , chemistry , organic chemistry , pure mathematics , operating system
Dimensional studies of Congressional voting have found a single dominant “ideological” dimension, while regression estimates find that constituency variables and party are dominant. Koford (1989b) recalibrated the dimensional studies, and found that several dimensions are important. This study reviews those findings, and then considers additional reasons why dimensional studies might understate the number of dimensions. It then examines the regression estimates for biases that overstate the number of dimensions. Overall, fewer dimensions are found than seem consistent with the wide variety of constituents' preferences on issues. A model is developed in which the transactions costs of building coalitions reduces the number of dimensions visible on roll‐call votes below the underlying dimensionality of preferences in the issue space. When legislative parties build internal coalitions to pass and defeat bills, voting on randomly drawn bills has a single party‐oriented dimension. Natural ideological dimensions are reinforced when parties write bills and logroll along natural lines of cohesion. Numerical examples suggest that these effects could be important, and suggest lines for empirical investigation of the underlying issue space.