COST‐BENEFIT ANALYSIS AND BRITISH DEFENCE EXPENDITURE
| Date | 01 February 1963 |
| DOI | http://doi.org/10.1111/j.1467-9485.1963.tb00912.x |
| Published date | 01 February 1963 |
| Author | Roland N. McKean |
COST-BENEFIT ANALYSIS AND BRITISH DEFENCE
EXPENDITURE
ROLAND
N.
McKEA"
I
As
indicated by Professor Peacock, we must seek substitutes
for
the
price mechanism where it does not or cannot carry out its function.
One partial substitute that can be helpful in the government sector
of the economy is the use of cost-benefit analysis to compare courses
of action. Such analyses are by no means
a
perfect substitute for the
price system, for
at
best they permit the occasional comparison of
selected policies, not the frequent examination of numerous alterna-
tives at all levels that
a
properly functioning price mechanism encour-
ages. Moreover, quantitative analyses of possible policies, like market
prices of alternative inputs and outputs, can help identify better courses
of action, but unlike the price mechanism they do not at the same
time provide strong incentives towards the adoption of the improve-
ments. Still, it is extremely important at least to be able to identify
preferred policies and practices. Unless wecan point to the better actions,
we cannot possibly expect officials to choose better rather than worse
ones, nor can we devise institutional modifications that would provide
incentives leading to the better policies. Indeed unless we can identify
preferred actions we cannot even tell where political
or
administrative
processes are distorting incentives
or
leading to poor decisions. There
is reason thus to count cost-benefit analysis as
a
tool worth trying
to
develop.
In this paper
I
shall try
to
outline the nature of this type of
analysis2 and to survey the possibilities and difficulties
of
applying
such quantitative analysis to the appraisal of alternative British defence
expenditures. For some of these evaluations, analyses of this general
type are no doubt already used, but
I
shall look at the possibilities in
terms
of
several broad categories regardless of existing applications.
1
In connection with the preparation
of
this paper
I
am
indebted
to
the
Fulbright programme and the University
of
Glasgow, which has been my host
during the Fulbright grant.
2
Several names-such
as
cost-benefit analysis, operational research, and
systems analysis-have been applied
to
particular applications
of
this type
of
analytical activity, and later in the
paper
I
shall
look
briefly at the rel:tion-
ships among these analyses. It begins
to
appear
to
me,
however, that cost-
benefit analysis
'
is
the title that is most widely understood and that most nearly
suggests the nature
of
this work.
2
17
18
ROLAND
N.
MCKEAN
In conclusion
I
shall comment briefly on the case for this
‘
substitute
for the price mechanism
’.
11.
THE
NATURE
OF
COST-BENEFIT ANALYSIS
Cost-benefit analyses are essentially exhibits of the costs,
or
dis-
advantages, and the benefits,
or
advantages, entailed by each of several
alternative courses of action. The courses of action might be alterna-
tive government investments-for example, different flood-control
projects on which a resource-development budget could be spent.
They might be alternative designs
for
a
particular project (a few big
dams in
a
river basin versus more numerous small dams and levees)
or ways of operating
a
network of flood-control dams.
Or,
as
we
shall
see, the competing courses of action could be alternative investments
or
modes of operation in national defence.
The exhibits present in as systematic and useful a format as possible
the costs,
or
what would have to be sacrificed
to
have each
of
the
alternatives, and the benefits, or what would be gained by having each
one.
If
these costs and benefits were exhaustive, certain, and commen-
surable, they would point clearly to the preferred policy. It would be
the one that yielded the greatest net benefit. Ideally, one would come
up with such clear cut results, e.g., exhibits showing that Policy
A
would cost
f50
million and yield
f60
million, while Policy B would
cost
&50
million and yield only
f45
million.
Or,
if gains could not be
measured in monetary units, one might fix the budget and see which
way
of
spending it yielded the larger benefit. Again one would like
the results to point unambiguously to one policy-e.g., to find that
Policy
A,
costing
f50
million, would produce
100
units of gain (say
a certain kind of defence capability) while Policy B, costing
f50
million, would give
50
units
of
gain. It might be more convenient,
or
more pertinent, to
fix
the gain to be achieved and find which policy
would accomplish this at lower cost. In that case, one would hope
for
a
result such as: Policy
A
would produce the desired capability,
100
units, at a cost
of
f50
million, whiIe Policy
B
could produce the
100
units
of
gain only at
a
cost of
El00
million.
Since many effects of the rival policies are uncertain and incommen-
surable, the exhibits do not always point conclusively to the best
choice. Nonetheless they often reveal
a
very convincing case
for
the
‘right’ policy, and on a sounder basis than intuition. Moreover, by
weeding out irrelevant effects and putting
some
of them in terms
of
a
common unit, the exhibits can reduce the number of considerations
that decision-makers must keep in mind and can make possible better
decisions.
To continue reading
Request your trial