According to the historical record, the increased use of energy in gen
eral and electric power in particular stands as one of the key factors
in productivity growth. Growth accountants, however, find little supp
ort for this view, rejecting claims that the productivity slowdown can
be attributed to the lower rate of growth of the energy input in US m
anufacturing. In this study, an attempt at reconciling the historical
record with the empirical facts is made. Specifically, the commonly us
ed factor income shares in the relevant Divisia input-growth index, ar
e replaced by the estimated output-input elasticities. The latter attr
ibute a considerably larger role to electric power than previous studi
es indicated. These are then used to reevaluate the sources of growth
in US manufacturing.