27
focus for future research.
Conclusions
The analyses reported here provide further, systematic support for the proposition that
public management matters for performance. Further, some of the ideas at the heart of the
O’Toole-Meier model and research agenda receive support from this study – in particular, the
inertial aspect of the performance function, the importance of examining multiple performance
measures since public organizations and their managers must juggle multiple goals, and the key
notion that managers have influence through multiple functions and via multiple causal channels.
Public management is clearly not a simple, undifferentiated activity. The top managers in
our sample do report, as Moore’s analysis anticipated, interactions upward, downward, and
outward. While these distinctions are simplifications of a broader and more differentiated set of
managerial tasks, they are also distinct enough to allow analysts to begin to sort through some of
the ways that management might matter.
Management downward, at least as measured by the relatively blunt management metric
included in this analysis, is clearly no panacea. Sometimes it seems to help performance,
sometimes it is associated with negative performance impacts, fairly often it shows no significant
effect. The lack of a clear pattern here should not be particularly surprising, despite the fact that
this aspect of management has received the lion’s share of attention in field. As indicated earlier,
managers may produce mixed results across performance criteria by emphasizing or
deemphasizing particular goals; attainment of one goal might come at the expense of others.
Also, explicit interaction is only one channel top managers can use to shape what happens in and
through their organizations. Further, school-system superintendents may be managing downward
through contacts with individuals other than school principals. Their interactions with other
Convention