How does the risk of gridlock affect the type of legislative output? Do bureaucratic agents expand their activities when they can expect that the principals are unable to overrule them? This article introduces a novel approach for calculating the risk of gridlock in bicameral legislatures in order to estimate its impact on bureaucratic activities, combining data on all secondary and tertiary acts of the European Union (EU) from 1983 to 2009. The findings reveal that bureaucratic activities expand when the risk of gridlock increases and an overruling of tertiary acts becomes less likely. This may sustain the EU's overall decision-making productivity, but its bureaucratic nature may raise further questions about democratic legitimacy and principal-agent problems in the representation of interests.Legislative scholars have intensely studied the power distribution within legislatures and the factors that promote gridlock. 1 Some scholars have argued that gridlock can improve the quality of legislative output, 2 but a common view in legislative research is that gridlock reduces the likelihood of policy change. 3 Drawing on theories of veto players, 4 party cartels, 5 presidential approval, 6 coalition governments, 7 unified or divided government, 8 and ideological conflict in parliaments, 9 the empirical research mainly supports the effect of this risk on the scale and scope of adopted legislative acts. 10 But how does the risk of gridlock affect the power of the bureaucracy and the type of legislative output? We argue that the risk of gridlock may also change the distribution of