The long Finnish tradition in cash-flow analysis formed the basis for this
study. In addition, the international research has began to give growing
interest to cash-flow reporting. The special interest in international
research has been the importance of cash-based information for the investors.
Among a countless number of recent studies dealing with distributional
properties of financial ratios this study is the first one where the
calculation of the ratios is entirely based on cash-flow numbers.
This study is mainly empirical by nature. When judging the properties of the
selected cash-based ratios theoretically, the basis has been that the main
task of financial statements is to give information to investors, managers and
other interested parties. Another central assumption is that the objective of
the management is to maximize the market value of the firm. This means, for
example, when dealing with profitability ratios that these ratios should
reflect the true profitability of the firm, i.e. such that the effect of all
expences, e.g. depreciation has been taken into account in defining the ratios
(the Finnish and international tradition in cash-flow research differ a bit
from each other).
When examining the cash-based ratios in Finnish literature, we found several
ratio classifications differing slightly from each other. We selected for our
study the classification presented by Aho. This is because the presented
ratios are entirely cash-based, they are numerous enough and they have been
used in many Finnish studies.
The final number of selected variables was eleven cash-based ratios. We
grouped the ratios slightly differing from Aho's textbook. They were
classified into three groups. The first group constisted of ratios measuring
profitability or adequacy of income financing (cash-flow from operations) in
some studies. The ratios were: cash margin Ia (cash operating income a) to
cash from sales, cash margin Ib (cash operating income b) to cash from sales,
cash margin II (cash net income) to cash from sales, and cash-based ROI (cash
margin Ib to liabilities).
From the theoretical point of view, the main problem in profitability ratios
is the lack of the effects of investment expenditures in these measures. This
deficiency should be removed in the future development of the ratios. In
addition, in the third ratio (cash net income to cash from sales) the
numerator measures in the first place the net cash-flow for shareholder's
equity (the effect of depreciations is also ignored) whereas the denominator
belongs both to shareholder's equity and to debt. So the interest payments
should be added back to the numerator to control better for capital structure
differences across firms or over time. On the other hand, both the
reliability and calculation properties of the ratios are good. In Finland,
the ratios do not include so many items subject to consideration as do the
accrual income numbers, and the denominator of the ratio is always positive
and large as compared with the numerator. A stable behaviour of the
profitability ratios was thus to be expected.
The empirical behaviour of the four cash-based profitability ratios was good
(if the criterion of goodness is normality as usual). The distributions are
normal already in raw data or at least affer removal of one or two outliers.
Different business cycles also reflected in the distributions. Outliers
appeared when the competitiveness of Finnish firms was at the lowest and
disappeared during "normal years" (the period studied was 1974-84). The
results were very similar to those of profitability ratios we obtained
earlier when we analyzed the distributions of accrual-based financial ratios
of Finnish firms. It must be noted here, however, that despite of their
similar distributional behaviour, the accrual-based profitability ratios and
the cash-based profitability ratios has been shown to measure different
properties in the firm's behaviour (Gombola and Ketz 1983, and Yli-Olli 1983).
The second ratio class in this study was already theoretically more
heterogeneous than the first one. The ratios in this class describe in the
first place how much different cash out-flow can be covered by the cash from
sales or by parts of it. The selected ratios were: direct labor and materials
to cash from sales, cash-based income to total cash out-flow, cash-based
payback period (total debt to cash net income), and cash net income to
cash-based investments.
The first variable, the direct labor and materials-ratio, was systematically
skewed to the left. The square-root transformation of the complementary
variable worked, therefore, well: the transformed variable became normal
with no outliers.
The second variable in this class was the income financing-ratio. Also this
variable was systematically skewed to the left. The square-root transformation
of the complement (the complementary variable can be called the equity
financing-ratio) became normal, with a couple of outliers in the years of low
competitiveness, however.
The third ratio in the second class was the cash-based payback period. This
ratio was already in advance very problematic, both theoretically and
technically. The denominator includes the problems mentioned earlier.
Technical problems are also many: the denominator has an exceptionally large
variance, it is small compared the numerator, and it changes its sign. With
negative values of the denominator the ratio has no economic meaning and
interpretation. Due to these facts, instead of the original payback
period-ratio its inverse, the cash-based payback rate, was analyzed. This
ratio was positively skewed. After square-root transformation and removal of
some outliers the distribution became approximately normal. As a summary,
however, the ratio must be regarded both theoretically and technically quite
poor.
The last ratio in the second class was the ratio cash net income to cash-based
investments. The ratio is theoretically sensible (reservations concerning the
item cash net income must be beared in mind, however). Problems was to be
expected in behaviour, however, because the variance of both the numerator and
the denominator is extremely large. This is true both across firms and
firm-specifically across years. Unstable distributions were thus to be
expected. The distributions were, without and with transformation, strongly
leptokurtic and thus non-normal.
The third ratio class consisted of different items of profit distribution
(interest payments, taxes, dividends) related to cash net income. The ratios
are theoretically sensible but technically very problematic. The problem is in
small and very unstable denominator. The general empiral conclusion about
these ratios was, that normality was not achieved. The distributions were
randomly skewed to the left or to the right and strongly leptokurtic. Instead
of the unstable cash net income a technically better denominator shoud be
find for these ratios.
As a conclusion we can state the following rough rule: the better the ratios
were theoretically motivated and the better their calculation technical
properties were, the better were also their distributional properties.
Theoretical development should be made especially in including the effects of
investment expenditures in the ratios. From the technical point of view,
attention should be paid for avoiding in the denominator items which are
extremely unstable, change their sign or are too small as compared with the
numerator.
(The Finnish Journal of Business Economics 41989, 374404)