The limited lifetime of a government
Governments have limited time in power, depending on the election cycle. The political
process means that the government, and therefore the policies may change on a regular
basis so that an individual policy may not be in place for long enough for it to begin to
have any real impact.
Practical implications
The differences in the nature of the public services create the following practical diffi-
culties:
- It is much harder to establish accurate measures of efficiency.
- The results need to be more widely circulated and explained to different stake-
holders with diverse interests than is the case in the private sector. For example, the
results on agricultural policy touch all aspects of the economy and affects different
stakeholders differently. - There may be disputes about the results as a product of the political process or other
environmental factors like an improvement in the weather conditions.
Characteristics of a good measure of performance
A good performance measurement will satisfy the following criteria:
It is appropriate for the objective it is meant to assess
It should be designed to produce information that indicates progress made towards the
objective in a meaningful way. It has to measure what it is supposed to be measured.
The chain of cause and effect should be clearly established to eliminate erroneous
results
For example, a decline in the number of hospital visits may be caused by improvements
in the public health infrastructure or a change in climate, rather than by increased effi-
ciency in medical treatment and diagnosis.
The effort and expense involved in data collection should be justified
The costs of data collection and analysis should be far outweighed by the benefits of
measurement. This is about economic efficiency.
It should be focused on an activity which can be controlled
There is no point in setting a standard that cannot be controlled. For example, the supply
of electricity to a country may depend on the rains available, which a country cannot
control.
It should be clearly defined and specified
There should be no scope for error or ambiguity in what it means by, let’s say, ‘im-
provement’ or ‘better’. There must be a benchmark.
It should relate to a specific time period
The time period chosen should be appropriate to the activity being carried out.
For example, one cannot effectively measure the success of a new policy within a
month because the indicators might be just temporary.