The Finnish Journal of Business Economic 4/1988, 263-277.
With the admission that financial ratios are commonly used for
intra-industry comparisons, and that this practice is frequently
ill-advised because of the heterogeneity of the industry branches
with respect to financial ratios, there arises two consecutive
problems which constituted the objectives of the present study: 1)
What is the magnitude of the risk of biased information from
undertaking the said comparisons, and 2) what can be inferred about
the nature of the criteria which lie behind the present (official)
industry branch classifications. These objectives can be attained by
tackling the problem of our study: How reliably is a firm's industry
branch predicted (predefined) by financial ratios.
We used discriminant analysis as our method. The stronger the
predictive ability of a financial ratio to discriminate an industry
branch, the more homogeneous the industry branch with respect to the
financial ratio, and the less risk of making comparisons within an
imaginary coherent group. We put forward an index of the predictive
ability as a measure of the risk.
In going through the results for individual financial ratios one
finds that the variation of the predictive index is large. Indeed,
there are incidents where comparisons seem safe. It appears that,
generally, the risk is smaller in trade and service industries than
in manufacturing industries.
As for the official industry branch classification it seems that it
does not have a basis ideally suited for evaluation of firms'
performance. This is inferred from the vast variance of the index of
predictive ability both for the financial ratios and the industry
branches. The poor performance of labor intensiveness, which comes
closest to the nature of the production function, lends further
support to our assertion. Summing up, it is probably the criteria of
industry classification which ought to be reconsidered rather than
the soundness of financial ratios.
[To the fulltext paper ]