As stated above, the problem has two aspects. First, there is the problem of
measurability. Some of the items constituting costs and benefits cannot be
measured at all (nonmetricity) or have
a unit of measurement which does not
satisfy the requirements for a valid and reliable measure.
377
Even where
appropriate measures can be devised, they may be costly or difficult to apply.
Second, there is
the problem of comparability. Because items cannot be
properly measured they cannot be added or subtracted. Consequently, it is not
possible to construct a valid and reliable measure of net benefit. This means
that the net benefit of one option cannot be compared with the net benefit of
other options (the problem of incomparability). Therefore, it is not possible to
apply the net benefit rule for making the decision with any sort of precision.
Attempts to Alleviate Problems
Introduction
There are at least two means of alleviating the problem. One is to construct
measuring devices, while the other is to get by with measuring the main item.
Constructing Measurement Devices
Attempts have been made to construct methods of measurement for items that
have so
far been unmeasurable. Economists, for example, who are inherently
interested in differences and comparisons, have devised a number of derivative
means of measuring various aspects of welfare. Social scientists have devised
measures of human and sociable attributes, covering such things as
extraversion and introversion, personality and intelligence. However, most of
these devices go only part of the distance because they provide only an ordinal
and not a cardinal measure.
Average Gross Domestic Product
A simple example is average gross domestic product (GDP). This is derived
from total GDP divided by the population of the country. Average gross
domestic product can be compared from one county to another.
Distribution of Income
If income was evenly distributed throughout all countries, average gross
domestic product would be a useful measure of welfare, both absolutely and
comparatively between countries. However, income is not evenly distributed
so it is necessary to devise a measure of its distribution. One device is the
Lorenz Curve which correlates percentage of income to percentage of
population. For example it might show in a particular country that the bottom
50% of the population has 20% of income, while the top 20% of the
population accrues 50% of the total income.
___________________
377
In the context of evaluating these items they are sometimes referred to as
“soft values” – see Arup (1982).
 |
 |
 |