positive effect of state legislatures on OSHA enforcement: but where is the variable that
negatively affects enforcement?
Interestingly, all the other political-institution coefficients in Table 1 are negative.
However, they cannot be clearly said to negatively affect OSHA enforcement, because none of
these terms are statistically significant. The negative coefficients for the political-institution
variables, then, are suggestive, but they cannot definitely be said to suppress the effect of state
legislatures on enforcement.
Conclusions
This paper confirms that the effects of political institutions on regulatory enforcement can
be concealed unless one controls for the effects of other political institutions. However, further
research is needed as to the nature of these relationships. Union percentage of workforce,
ideology of Congressional contingent, and party of governor cannot clearly be described as
suppressor variables: although the effect of state legislatures did not appear until they were added
to the model, none of them had a significant negative impact on OSHA enforcement. The
findings intriguingly suggest that different political institutions can have opposing effects on
enforcement, even when they are dominated by politicians of the same party or ideology.
However, more research is needed to fully investigate the nature of the relationships among
OSHA enforcement and the various political institutions.
12
Convention