y, those which have to
do with the choice of the actual policy of adjustment during times of
rising and falling prices. The same division and order is maintained in
the following attempt to sketch out a good plan of adjustment of living
wage rates.

First, then, these wage rates should be varied in accordance with the
movement of a price index number. This index number should represent the
prices of all the important commodities produced within the country, but
so weighted as to give a defined importance (50 per cent. was suggested)
to the prices of those classes of foodstuffs, clothing, housing
accommodations, and other commodities upon which the wage earners tend
to spend the bulk of their income. It was sufficiently emphasized in the
earlier discussion of this subject that this basis of calculation was in
the nature of a compromise, and was not beyond criticism. Adjustments
should not be undertaken unless the index number of prices has moved at
least 5 per cent. (the figure is meant to be merely a suggestion) and
adjustment should not be more frequent than twice a year (again a
suggestion, only).

Secondly, as to the policy of adjustment to be pursued in times of
rising and falling price levels, respectively. The policy for a period
of rising prices can be very briefly stated. All wage rates prescribed
under the living wage policy should be increased by the same percentage
as the index number of prices moves upward. There is one case in which
this policy cannot be justified theoretically. That is when the increase
of prices can be wholly or mainly accounted for by a falling off in the
general level of industrial productivity. However, in my opinion, it
will be hardly practicable to attempt to distinguish this case from
other cases of price increase,--save in an entirely exceptional
circumstance, such as a period of war invasion.

The policy to be pursued during a period of falling prices cannot be
stated so briefly. The difficulties involved have already been discussed
at length.[133] The following policy based upon that analysis is
tentatively suggested. The complexities of the subject are too great to
permit of dogmatism. Firstly, the occasion for the price decline may be
such as was termed "natural," as for example when it is brought about
by a general advance in the arts of production, or by the development of
the means of transport. In this case, it will be satisfactory to keep
wage rates unchanged, though pr