Therefore, the model appears reasonably accurate for up to two statements prior to
failure but not accurate for earlier periods. These findings are quite similar to those
of Altman’s (1968) model and we can suggest that the similarities in accuracies are
partially related to the similarities of the data quality and the somewhat diverse in-
dustries represented in the sample.
A&L also simulated their results for various assumptions of prior probabilities of
group membership and costs of error. Their findings were that Type I errors could be
reduced, even eliminated, but that the resulting Type II error was unacceptably high
and vice versa for eliminating the Type II error. The Z model’s results were also com-
pared to a naive classification strategy of assigning all observations to the nonbank-
rupt category or assuming that the resulting errors would be realized in proportion to
the actual experience of bankrupts and nonbankrupts (proportional chance model.
They concluded that, in every case, the Canadian Z model was more efficient; that is,
it had a lower expected cost than a naive model.
Finally, A&L observe that the industry affiliations of the misclassified firms were
predominantly retailers amongst the failed group and manufacturers among the non-
failed. It appeared that one of the variables, sales/assets (X 1 ), was particularly sensi-
tive to industry effects, with the misclassified failed retailers all having high asset
turnovers and the misclassified manufacturers all with low turnovers.
(d) Implications. A&L attempted to reestimate the model without the sales/assets
variable, but the results actually were worse. One can conclude that the Canadian in-
vestigations are at an early stage and follow-up work is needed in subdividing a
larger sample into manufacturers and retailers-wholesalers and/or improving the in-
formation on critical industry differences, such as lease usage and capitalization.
Only additional time will permit analysts to construct models with sufficiently large
samples or to witness an improvement in the quality of reported data. We are aware
of a move with the Canadian government to set up an early warning system to iden-
tify potential large publicly traded firm crisis situations, for instance, Massey-Fergu-
son. Authorities are currently considering available models such as Altman (1968)
and A&L (1980) as alternatives to building their own model.
10.7 THE NETHERLANDS
(a) Bilderbeek (1977). Bilderbeek (1977) analyzed a sample of 38 firms which went
bankrupt from 1950 through 1974 and 59 ongoing companies. They found that 85
firms had sufficient data for analysis. Bilderbeek analyzed 20 ratios within a stepwise
discriminant framework and arrived at a five-variable model of the form:
where
X 5 net profit>equityX 4 sales>total assetsX 3 accounts payable>salesX 2 added value>total assetsX 1 retained earnings>total assetsZ Z-score 1 Netherlands, Bilderbeek 2Z0.455.03X 1 l.57X 2 4.55X 3 0.17X 4 0.l5X 510.7 THE NETHERLANDS 10 • 17