12
The analysis also shows that some analytical practices, which were
essential to controllers in the past, are now also performed by other
actors in the agency. Moreover, the analytical support function which
controllers had, likened by researchers to that which a business
partner supplied to local managers, has been weakened and at times
taken over by other actors at the SIA. Controllers appear thus to be
losing authoritative power as the controlling function is spread to
other groups, such as specialists, in the organization.
Lastly, opportunities and risks involved with the new decentralized
management accounting and control practices of the SIA are
discussed. On the positive side, management accounting and control
practices seem more inclusive, which also makes them everybody’s
responsibility. This may produce commitment and motivation as well
as conditions for increased creativity, adaptation to specific situations
and customization, especially at the SIA officer level where actual
meetings between the clients and the agency take place. On the
negative side, however, the ambition to democratize and include more
groups in the management accounting and control process can be
difficult to realize and take responsibility for. Moreover, management
accounting and control practices may become fragmented and thus
also unsystematically controlled. The new direction, which includes
a deviation from the use of quantitative measurements of specific
goals and the transformation of the controller role into a consultative
and coaching role, has in some cases caused uncertainty among
controllers, managers and co-workers, who sometimes feel the
need for clearer direction. In effect, in the practice of management
accounting and control that is being instituted at the SIA it is
concomitantly up to everyone to lay down outlines and levels of
what they should strive for.
There are, moreover, potential risks in terms of sub-optimization
because self-managing teams with similar problems solve them in
different ways. By extension this could also mean increased variation
and that the principle of equal treatment would not be obtained. Lack
of knowledge about methodological issues could be problematic,
given that measurements are performed more often and by more
actors, which could lead to insufficient and non-generalizable results
and poorer decision-making.