Conclusions

In evaluating the performance of the fire safety policy, one can easily notice that performance of the actions is a mosaic of several actors and several streams of action.

The performance of the fire safety policy is a sum of several factors. As in the garbage can

model by March and Olsen (1976), where the decision is an outcome or an interpretation of several relatively independent streams of problems, solutions, participants, and choice opportunities, performance of the fire safety policy is an outcome of an abundant group of actors from fire brigades, rescue services, municipalities, social services, schools in welfare for intoxicants, and numerous crews of other, different safety authorities, and other actors outside the public sector.

The performance of the fire safety policy was ambiguous in terms of evaluation, since the data covered only documentary data from regional state administrative agencies, the ERC and the Emergency Service College. Other data and data covering other fire safety organizations are outside the analysis. The studied data also covered only governmental authorities and, thus, fire departments are outside the analysis. This makes performance evaluation ambiguous. Whose performance do the actual results relate to if the performance of the actions is analyzed in only a few organizations?

What is also ambiguous is the level of targets. What can be said about the actual performance of the fire safety policy and the rescue services when the overall objective level to reduce fire deaths to by 2015 is 50? The optimal level is that nobody would die in fires and that level would be zero. An objective level of 50 reduces the number of fire deaths, but it still tells us that 50 people will die in fires and it is not a good result. The objective level would be zero to be the best possible objective level.

Then what can be done to prevent accidents and operate efficiently in the case of an emergency being at hand for the rescue services; however, in cooperation with different authorities. The question of whose performance should be evaluated is an essential subject of examination. Financial statements and annual reports of the ERC or the Emergency Service College tell a detailed story of targets reached. When considering the achieved targets, both agencies are performing well. However, people are still dying in fires. This is the result of a series of events that are linked to each other and can neither be predicted nor excluded. However, examining what the consequences were of intended actions (positive and negative) and if there were any non-intended side effects (positive and negative) that were caused as a result of the actions, is an essential aspect of performance evaluation.

The logic of performance is familiar, for example, from school exams. Students may be successful in tests but they may fail in actual learning. Tests do not tell us much about the actual learning. Targets are of limited use and people will try to manipulate them to gain good results. In management, this means “gaming” in that managers may try to make results look good by any means by “hitting the target but missing the point.” Their work is evaluated in the performance management cycle by reaching the targets and reaching the targets should look good. Actors will change their conduct when they know that the data they produce will be used to control them (Bevan & Hood, 2006, 521).

Outcome measures help policy makers control public sector resources. Smith (1996) defines internal, managerial control, political control, contractual control and the science of control. In managerial control, the central management wishes to control the periphery,

and outcome measures can be used as an instrument to control. In political control, the outcome measures are used to inform external persons about the performance of the organization. In contractual control, the principal is a central purchasing organization, and the agent is a separate provider organization, and under these arrangements, control is by means of a formal contract such as a performance contract. The science of control derives from the Greek word for a steersman, and it is illustrated in a cybernetic model of measurement, analysis, action, and system (Smith, 1996, 7–8).

The rescue services, the Emergency Service College, and ERC, for example, have set targets in a professional and skillful manner. However, there are many ways to reach these targets and there are many ways to improve performance outside these targets.

What can be done to intensify cooperation and who are the most important actors in terms of preventing accidents? In alcohol–related fires caused by a cigarette, an important feature is the alcohol consumption and behavior; the victim’s ability to react is one of the most important features in fires. Alcohol education could play an important role in individual behavior. However, it is beyond the rescue services’ sphere of influence. It is not impossible to affect people’s awareness of the accidents and the risky situations leading to casualties. However, this requires much cooperation between several authorities and other parties.

8 Conclusions

This dissertation explored ambiguity in the performance management cycle and in the fire safety policy. This chapter presents the major findings of the study and discusses them in the light of theoretical findings. Ambiguity was analyzed in a performance management cycle of three phases: ambiguity in identification of fire deaths as a fire safety policy, ambiguity in target setting in fire safety policy and ambiguity in evaluating performance toward the targets of fire safety policy. The fourth research question was the coherence of the performance management. Is the performance management cycle a coherent management cycle, and are mutually reinforcing policy actions supporting each other?

The main aim is to understand ambiguities in the model of performance management and what causes ambiguity in the context of political-administrative decision making environments. In addition, the question of coherence in the performance management cycle and how different intercrossing policies go together when considering fire safety policy were questions that needed to be clarified. Fire safety policy and fire deaths offered an empirical case through which to analyze these questions, but the same questions and phenomenon could be studied in some other policy field as well. This is a possibility for further studies.

Thus, we start in section 8.1 by examining at what ambiguities there are in the performance management cycle. The specific research questions are what are the ambiguities in identifying fire safety policy, in target setting and in evaluating performance of the actions? The specific research question considering coherence studied the coherence of the performance management cycle and mutually supporting policy actions as policy coherence. The theoretical implications are used to elaborate the empirical results.

Possibilities for further research are presented in the end in section 8.5.