ZO
Review of Islamic Economics, Vol. 8, No. z, z004
there? The input-output numbers, let alone their composition details,
are unclear. The section on data and variable specifications, for the
most part talks of what the mainstream writings on the subject
contain; what the study is based on is scantily mentioned. The section
is overloaded with methodological explanations, though these too are
not devoid of gaps.
Take, for example, the estimation of E[exp (- ui∣εi)∣. The
authors set up where -εj = ɪ-ʌeʃ “is the residual obtained from
equation r” (p. 130). Their equation 1 is:
/« TCi = f (yi, wi) + εi
But the sum of residuals εi from this equation must always equal
zero! That makes the formulation intriguing. Again, the authors
“define ζ,i=tnax^-εj where the maximum is introduced in order to
provide values of ς (p. ɪ 30). The statement needs elaboration: why
is max used in the equation not min, as the foregoing discussion
suggests? Maximization is required when efficiency measurement is
output oriented. A comparison of Figure 3 with Figure ɪ would make
the difference of the two approaches clear.
Shortfall of the estimated efficiency scores from one - the frontier
- is not exceptional but mostly expected. The central elements of
frontier analysis, to reiterate, consist of (i) the ranking of PUs on the
efficiency scale to compare their relative performance, and (ii) to test
the hypotheses that claim a causal relationship between efficiency and
its perceived determinants. The work of Saaid et al. is distinct from
others under review in that it falls in neither of the categories.
Therefore, it is uncertain what significance one can attach to their
conclusion: “The study as a whole shows 78 percent overall efficiency
(OE), meaning that 2.8 percentio of the Sudanese Islamic banks’ total
cost was inefficiently used compared to (if) the banks were on the
frontier (p. τ37)”. But could not the Sudanese Islamic banks still be
found to be more efficient in comparison with those in other
countries? Again, the claim that the technical component is the main
source of overall inefficiency seems to conflict with the suggestion that
public policy forced the banks to divert more of finance to the less
productive agricultural sector of the economy (p. 137)/1 If that were
true, the allocative component, not the technical, should have been
the main culprit in lowering the overall efficiency scores, which the
results show is not the case. Thus, question marks could be put on the