Sammendrag:

The Norwegian bureaucracy is subjected to structural changes on a regular basis. The frequency at which this occurs begs the question: Why does it happen? In the literature, termination (organizational death) has been devoted more attention than other organizational changes. However, such studies have thus far only been carried out in majoritarian systems, and are not immediately transferable to consensus democracies like Norway. This thesis represents an exploratory attempt at adapting theories of termination to a Norwegian context. By running a series of discrete survival regressions on 142 Norwegian agencies between 1980 and 2014 (based on a unique and constructed dataset), I find that agencies are at much higher risk of being terminated after a change in government. While this finding suggests that political executives have considerable influence over government structure, another finding indicates that there are limits to their power. For example, large, geographically dispersed and young agencies seem to at lower risk of being terminated. I find no evidence that fiscal pressure or the attributes of the government matters. Since termination is a disputed concept, I also empirically examine whether terminations are conceptually different from reorganizations. I develop an alternative hypothesis, which argues that they can be merged and referred to collectively as ``structural reforms.’’ By running the same regressions with reorganizations as a dependent variable, I find that there should be a clear distinction between terminations and reorganizations. First, very few effects are found. Second, large and geographically dispersed agencies are at higher risk of being reorganized, an empirical pattern that is contrary to what termination theory expects. Thus, the viability of structural reform as a fruitful theoretical concept is debunked.