Estimation Results of the Cross-National Determinants of Corruption
economics
Description
Notes: Model 1–5 are based on the full sample. Model 6 drops mixed-member systems,
including MMP and Taiwan. Robust standard errors in brackets. *Significant at 10 per cent;
**Significant at 5 per cent; ***Significant at 1 per cent.
whereas under closed-list PR systems the relationship between corruption and district
magnitude is described by:
TI CLOSED
i 5.16 0.0199DM CLOSED
In other words, as district magnitude increases under open-list PR, corruption rises. Under
closed-list PR arrangements, political corruption becomes less prevalent as district
magnitude increases.
To ascertain whether the effects of district magnitude on corruption under alternative
electoral regimes are indeed significant, we use CLARIFY to enrich our understanding
graphically of how the effect of district magnitude is conditioned by the difference between
a systematic role in inducing corruption? These questions merit theoretical investigation
informed by the empirical regularities we have observed thus far.
Empirically, our study suggests the need for the collection of data on and the
development of measures of judicial independence, campaign financing laws and
government regulation of the economy, none of which we succeeded in incorporating into
our analysis. Finally, an obvious empirical extension of our concern with the personal vote
is to incorporate single-member electoral districts into the analysis. Our expectation is that
single-member districts with open primaries should resemble large districts with open lists
in the extent of corruption generated, whereas closed primaries and single-member districts
should resemble large districts under closed lists. But this awaits future analysis.
