l6
Revieu> of Islamic Economics, Vol. 8, No. 2, 2004
The easier way to find the maximum or minimum value of
exp(-uj) is to use the maximization likelihood method (MLM), which
is automated in several computer programmes. The general problem
is often solved by the minimization of [-Iog(L)] where L is the
likelihood function.4
V. Efficiency Models and Islamic Banks
The foregoing discussion may help understand the structure,
achievements, and limitations of the efficiency studies that seem to
multiply apace in the area of Islamic banking as elsewhere. We shall
restrict the appraisal to include the works of Majid et al. (2003), Saaid
et al. (2003), and the article of Darrat et al. (2002), the only to use the
DEA approach. These are seminal works and deserve credit for
introducing Islamic economics to some new methods of analysis, that
appeared rather late even in mainstream applied economics. The
techniques used are quite sophisticated and these writings do not seem
to have received from the readers the attention they deserved.
However, being among the first of their sort in Islamic banking,
and that too in an area where more haze than light still prevails even
in mainstream economics, these efforts could not have been entirely
free of weaknesses. A few general remarks may not be out of place
though they do not necessarily apply to all these writings uniformly.
A common feature of these writings, as alluded to earlier, is the
insufficiency of background information they provide about the
nature of the economy, and the state of its monetary and fiscal
policies affecting the financial sector. Even discussion on the
structure, growth and place of Islamic banking in the overall financial
set-up of the country leaves much to be desired, especially in the first
two works. A number of more recent writings exemplify the point.’5
Another difficulty with these exercises is that they do not provide
clear explanations of the sources, nature, limitations and editing of
the data they use. The definitions with reference to the content, at
times even the number, of outputs and inputs adopted for the study
remain unclear. We shall return to this point later. Appropriate, and
unambiguous, explanations of method used are lacking; at some
points yawning gaps greet the reader. Reasons supporting the
conclusions arrived at are often missing, their policy implications
seldom adequately stated.