But for many policy purposes—including the MDGs—path controllability is the measure of
success: policy makers want not just the instantaneous achievement of targets but that
these are sustained over time and arrived at by a path that strays as little as possible from
politically and economically acceptable values of instruments and objectives. Path
controllability, a stricter requirement than point controllability, in many cases does in fact
require that the number of instruments equals or exceeds the number of targets (Petit
(1990)). For a set of instruments and objectives this is in part conditional on a given structure
of the economy, known outcome targets, and well-defined instruments. In practice, most
policy problems do not meet the conditions for straightforward solution for the following
reasons:
Available policy levers and feasible values—
1. The number of objectives often exceeds the number of available instruments. This
means that a “solution” does not exist in a static system or for path controllability in a
dynamic system.
2. There may be more than one solution when the number of instruments exceeds the
number of targets.
3. Even if a solution does exist, the values of the instruments needed to achieve it may be
politically or technically infeasible.
Interdependencies—
4. Some objectives may be related. They may be complementary, or they may be
contradictory and raise policy trade-offs.
5. Some instruments may be related. Again, they may be complementary or
contradictory.
6. Different instruments may be controlled by different policy institutions leading to
problems of coordination and conflict.
Causality, information and uncertainty—
7. There may be uncertainty about the relationships between instruments and objectives,
or even about what the instruments are.
8. Some variables may be either instruments or objectives depending on the context or
level of analysis (and whether analysis takes a static or dynamic perspective).
9. Information on the actual values of instruments and targets may be imperfect.
10. The structure of the economy may change over time. It may be affected by
movements in instruments and target variables. In turn, as it evolves the extent to
which (1)-(9) hold may change.
All of these problems are likely to bedevil progress on the MDG project. 4 In particular, they
complicate attempts to measure progress towards and devise strategies for achievement of
the targets; they also bring to the fore problems of valuing outcomes in multi-dimensional
space (section three). They raise issues about the sustainability of outcomes (section four).
4 These problems are also likely to be significant for domestically-driven planning initiatives.