Re Simon Crean, Minister of Primary Industries and Energy and Australian Fisheries Management Authority v Antonio Musumeci; Tony La Macchia; Frank La Macchia; Elizabeth La Macchia; Ruth La Macchia; Andrew La Macchia; Rocco Musumeci; Vincenzo Musumeci [1993] FCA 47 (19 February 1993)

FEDERAL COURT OF AUSTRALIA

Re: SIMON CREAN, MINISTER FOR PRIMARY INDUSTRIES AND ENERGY and AUSTRALIAN
FISHERIES MANAGEMENT AUTHORITY
And: ANTONIO MUSUMECI; TONY LA MACCHIA; FRANK LA MACCHIA; ELIZABETH LA
MACCHIA; RUTH LA MACCHIA; ANDREW LA MACCHIA; ROCCO MUSUMECI; VINCENZO MUSUMECI
and LUCIA MUSUMECI
Nos. N G701 to 703 of 1992
FED No. 47
Number of pages - 11
Fisheries and Fishing Rights

COURT

IN THE FEDERAL COURT OF AUSTRALIA
NEW SOUTH WALES DISTRICT REGISTRY
GENERAL DIVISION
Lockhart(1), Beaumont(2) and Hill(2) JJ.

CATCHWORDS

Fisheries and fishing rights - implementation of a Plan of Management - statutory interpretation - use of formula for the allocation of quotas - expert evidence - formula containing a statistical fallacy that produced an irrational or absurd result - consequences flowing from such a finding.

Fisheries Act 1992, s.7B

HEARING

SYDNEY, 22,23 October 1992
19:2:1993

Counsel and Solicitors for the
First and Second Appellants: Mr R.J. Burbidge QC with

Mr P. Roberts instructed by
Australian Government Solicitor

Counsel and Solicitors for the
Respondents: Mr R.W.R. Parker QC with

Mr P.W. Bates instructed by
Thomson Rich O'Connor

ORDER

THE COURT ORDERS:
That the appeals be dismissed, with costs, other than the costs of the motion dealt with on 16 February 1993.
Note: Settlement and entry of orders is dealt with in Order 36 of the Federal Court Rules.

DECISION

LOCKHART J. I have had the benefit of reading the reasons for judgment of Beaumont and Hill JJ. I agree with them and with the orders which they propose. The question whether the South East Fishery (Individual Transferable Quota) Management Plan 1991 is delegated legislation or administrative action was not argued in these appeals. Nevertheless the question arises here. I considered the question in the Austral Fisheries Case (NG699 of 1992) and adopt what I said there about it.

INTRODUCTION
BEAUMONT AND HILL JJ. These are appeals, brought by leave, from a declaratory order made by Burchett J. that para.11 of the South East Fishery (Individual Transferable Quota) Management Plan 1991 ("the Plan") was void (see (1992) 110 ALR 201). The Minister, now the appellant, purportedly determined the Plan pursuant to the provisions of s.7B of the Fisheries Act 1992 ("the Act"). The respondents to the appeal, who were the applicants at first instance, operate in the fishery dealt with by the Plan. The respondents catch gemfish. These appeals were heard, and judgment given, at the same time as the related appeal in the Austral Fisheries matter (No. NG 699/92). Although the evidence was not the same as the material before the Court in Austral Fisheries, and although no argument was advanced in these matters on the question whether the Plan was delegated legislation or administrative action, these appeals raise for determination several of the questions which arose in Austral Fisheries.

THE CASE IN CHIEF FOR THE RESPONDENTS AT FIRST INSTANCE
2. So far as is presently relevant, at the hearing at first instance the respondents adduced expert evidence from Dr. Nicholls. His affidavit evidence was to the same effect as his evidence in Austral Fisheries.

3. In cross-examination, Dr. Nicholls gave the following evidence:

"Which is the fairer method or the method open to the
government, is a matter ultimately of course for the court
to decide, you perceive that?---It may be either of these
methods, that's right.
Or perhaps neither?---That's correct.
If I understand you, the reason that you have advanced so
far why the alternative method is to be preferred is because
in your judgment the other method leads to an unfair result,
correct?---It can lead to an irrational result or an unfair
result.
...
You see, may I understand, Doctor, what you are asserting is
that taking an extreme case, a situation may result which is
'unfair'?---Not only from the extreme case.
I see. Well, then do you say that in every case the result
obtained is unfair?---No.
...
If it was your intention and I do not mean you personally,
of course, if it was a decision maker's intention to accord
to those involved in low yield years, a weighting or
advantage in consequence of their so doing, one method of
doing so in abstract or theoretical terms, would be to
accord to them for that year the percentage of the market
which they had obtained in that year and aggregate it
together with percentages in other years, would
it not?--- Well, that's the method that they've adopted in the
plan.
And the effect of it is to accord an advantage to those who
participated in low yield years, correct?---That is correct,
whether they be early or later years in the five years.
Whenever it was?---That's correct.
So what we are looking at then is whether or not looking at
the results, the results are so outrageous or unfair as to
suggest that such a weighting ought not be accorded to them,
are we not?---There's so many boats in each species that are
affected as a result of that, that I agree with that.
HIS HONOUR: Is it just a matter of the results or is it a
matter of the statistical theory that is involved?---It is
the statistical theory ..."

THE CASE FOR THE MINISTER AT FIRST INSTANCE
4. In addition to the evidence of Mr. Scott, which was similar to his evidence in Austral Fisheries, the Minister relied on the expert evidence of Dr. David Georges Pederson, Senior Lecturer in Statistics, University of Canberra. In his affidavit sworn 1 September 1992, Dr. Pederson said:

"3. ...Dr Nicholls has presented an example involving road
deaths over Easter in order to support his contention
that the method of averaging percentages contains a
statistical fallacy and produces an irrational result.
I propose that this example has no relevance to the
case of (the Plan). Human populations have the
property, in the short term, of being relatively
constant in size. However, in the case of gemfish
(East) the annual fleet totals are currently
controlled by (the Plan) and this is one aspect which
throws doubt on the relevance of Dr Nicholls road
fatality example to the present case. Furthermore,
the road deaths example does not lend itself to a
discussion of weights in estimation. For example, it
is difficult to think of a sensible calculation in
which the road death data for the ACT. would be
given, say, five times the weight of the data for
N.S.W.
4. An example in the sphere of economics would be more
relevant to the present case. I now present such an
example:
Company A is involved in the selling of firewood for
domestic use. Total sales depend on whether the
winter is cold, warm or hot. When it is cold the
demand is high and other sellers enter the marketplace
because sales are easily made. In three successive
years the sales of company A (tonnes) and the total
sales are:
Year 1 (warm) 2 (cold) 3 (hot) Total
A: 1600 3000 800 5400
Total: 4000 15000 1000 20000
%: 40 20 80 27
The market percentage method, the method used in (the
Plan), gives an average of (40 + 20 + 80)/3 = 46.7%
for the percentage of sales by company A. The overall
averaging method, based on totals across years and
preferred by Dr Nicholls, gives an average of 27%.
Suppose that it is now decided that the supply of
firewood should be limited for environmental reasons.
The total sold in any one year shall not exceed 2000
tonnes. The question arises as to which percentage to
use to determine the future quota allocation for
company A. Since sales are to be restricted in the
future, it is rational to choose deliberately an
average which reflects past market percentage in low
sales years. This is the average of 46.7% given by
the market percentage method, since this method has
the property of giving greater weight to years with
low totals. Alternatively, suppose that the market
were not to be restricted in size but that it was
simply desired to favour company A because of its
persistent presence in the market-place in all years,
relative to the ephemeral operators. This result
would be achieved by the use of the market percentage
method rather than the overall averaging method.
5. The example given in paragraph 4 is of greater
relevance to the present case than is Dr Nicholls'
example on road fatalities since the example in
paragraph 4 introduces the idea of anticipated or
planned changes in annual totals. It also allows for
a discussion of the use of weights in estimation..."

Dr. Pederson then quoted some figures regarding the applicants' boats and said:

"7. Dr Nicholls has claimed that the market percentage
method 'contains a statistical fallacy and produces an
irrational result'... In his example, he implies that
the overall averaging method is not fallacious and
produces a rational result. From the above figures,
it is clear that, in the case of gemfish (East), the
differences between the quotas produced by the two
methods are not significant in terms of the quantities
involved. It would therefore be difficult to justify
the conclusion, on the grounds of calculated quotas
for gemfish (East), that one method produced a
rational result and the other an irrational result."

5. In cross-examination, Dr. Pederson gave the following evidence:

"Now, you say (in your affidavit of 27 August) Doctor, that
the method of averaging percentage is not inherently
fallacious, and then you tell us that there are two
situations in which the method produces the same quotas,
correct?---That is true.
The first of these is when the fleet catches is constant
over years and the second is when the proportion of the
fleet catch for each year - - -
HIS HONOUR: Each boat.
MR PARKER: I am sorry, your Honour, for each boat remains
constant over the year. Now, have you made an inquiry as to
whether either of those events have occurred?---I haven't
made an inquiry. I know that the first of the events is not
true for gemfish east. I don't know whether the second
event is true for gemfish east.
I see, but the first proposition, that the first of these,
when the fleet catches are constant over the years, that in
fact has not occurred?---That is true.
And you do not know whether the proportion of the fleet
catch for each boat remains constant over the years has in
fact occurred, if I may use that phrase?---No, no.
Those would be the only two, would they, situations which
you could think of when you could show that the method of
averaging percentages was not inherently fallacious. That
is correct?---No, if I could reply, those are the two
situations in which these two methods which we are
discussing give exactly the same result.
But they are the only two you know of, that is my point I am
seeking to put to you?---Yes.
And I take it that - well now, what do you mean when you say
in paragraph 5, that although it is not inherently
fallacious there is a potential for producing results for
which high weight is given to years with a low fleet
catch?---I am combining there my two statements; firstly, that the
method is not inherently fallacious, by which I mean that
there are situations if the test of fallibility or
fallaciousness is the comparison with Dr Nicholls' formula,
if that is the test for fallaciousness then this particular
formula - this method of averaging percentages - since it
agrees with Dr Nicholls formula in some circumstances cannot
be inherently fallacious. I was combining that comment with
the comment that it is true that the method that has been
used has the potential for producing results for which a
high weight is given to years with a low fleet catch. There
is that particular quality of the formula that has been used.
You then say:
'But this potential does not invalidate the use
to which the formula is put provided one is
aware of the effects of variation in fleet
catch.'
?---That is true.
What do you mean when you say 'provided one is aware of the
effects of variation in fleet catch'?---I mean that if a
person presented with a situation of using such data as are
available wished to estimate quotas, as is the case here,
then presumably that person would investigate various
possibilities. If they then decided upon using the method
of averaging percentages then I can see that that would be
as a statistician I would say that that was a reasonable
conclusion. I would hope that the person who chose that
method was aware of the effects of using that formula."

CASE IN REPLY FOR THE RESPONDENTS AT FIRST INSTANCE
6. In his affidavit sworn 2 September 1992, Dr. Nicholls referred to Dr. Pederson's affidavit and said:

"...the statistical fallacy demonstrated in that
'Easter road deaths' example referred to above is the
same statistical fallacy employed by the
abovementioned management plan. In both cases the
statistical fallacy arises as a result of averaging of
percentages of populations which vary from year to
year. It is irrelevant whether the populations used
are road deaths over Easter or annual fleet catches
just as it is irrelevant whether the percentages cited
refer to increases in road deaths or boat catches, the
statistical fallacy in applying the formula adopted by
the management plan is present in both cases.
2. ...Mr Pederson gives an example based on 3 years of
data to show that in the Management plan's statistical
method 46.7 per cent is an appropriate measure for
company A's sales rather than 27 per cent when one
takes into account varying annual total sales over the
three years. It is argued by Mr Pederson that if the
supply of wood in a future year is to be limited to
2000 tonnes that company A should get 934 tonnes (46.7
per cent) rather than 540 tonnes (27 per cent) on the
basis that it is rational to choose deliberately an
average which reflects past market percentages in low
sales years. ...(T)his is an irrational argument
(leading to unfair results) ... On the figures applied
hypothetically for ('Boat A') in the 'S' area, it
gained 100 percent of a catch of 1 tonne in 1985.
Since each of the 5 years is given equal weighting
this figure produces 20 per cent of the total. That
is 1 tonne out of a total of 9017 tonnes over 5 years
gets this boat 20 per cent (of its total of 43 per
cent) of the quota.
...
4. With respect to paragraph 6 of the affidavit of David
Georges Pederson referred to above I say that the
choice of three boats to compare methods of quota
allocation for gemfish is inadequate in that he has
not taken into account the impact of the quotas on
other species for which these boats also have quotas
nor has he considered the impact on other boats some
of which get higher ITQ's and some of which get lower
ITQ's when comparing the two methods of quota
allocation.
5. In reply to paragraph 7 of the affidavit of Mr
Pederson referred to above, in annexure A to his
affidavit of 27 August 1992 it is seen that the
gemfish east annual fleet catch was 2408152 kilograms
in year OY1 (presumably 1984) and 4052969 kilograms in
year OY4. Such variations have lead to his comparison
of two methods of distributing quota allocations
showing differences of up to 37.23 per cent.
6. I say further in respect of paragraph 7 of the
affidavit of Mr Pederson of 1st September 1992 that
small changes in individual quota allocations can be
most important in cases where the TAC's are increased.
For example if the TAC for a particular species is
increased by 100 tonnes a 2 per cent increase in quota
would result in an ITQ increase of 2 tonnes under the
Management Plan. I understand that the ITQ is the
owner's asset and such variations would be of extreme
importance to him."

THE REASONING AT FIRST INSTANCE
7. After referring to the evidence of Dr. Nicholls and the reasoning of O'Loughlin J. in Austral Fisheries, the learned primary Judge said (at 206):

"For the respondents, evidence was called, to answer the
evidence of Dr Nicholls, from Dr D.G. Pederson, a Senior
Lecturer in Statistics at the University of Canberra. Dr
Pederson disputed that the method in question is 'inherently
fallacious', but conceded 'there is the potential for
producing results for which a high weight is given to years
with a low fleet catch.' He added: 'In my opinion this
potential does not invalidate the use to which the formula
is put provided one is aware of the effects of variation in
fleet catch.' But awareness of a distortion is of little
use unless the formula is adjusted. Although it was argued
for the respondents that the effects of the method may have
been deliberately chosen by the Minister, and there was some
remarkably general evidence that the officer responsible for
developing the method had the benefit of computer
simulations (how complete they were was not revealed) and
'was fully aware of the effects of the formula on
individuals', I can only conclude to the contrary. If such
a method had been deliberately adopted in order to achieve
some designed weighting effect, there must have been
calculations demonstrating the validity of going about the
task in that way; none were produced, nor was their
existence suggested. Nor were the respondents able even now
to suggest, except as a generalisation, how the weighting
would work out to produce some rational or desirable result.
I note that O'Loughlin J. remarked upon the same difficulty
about the evidence before him. Furthermore, the terms of
s.7B may have required a statement of the weighting
objectives and the measures (including the suggested
deliberate bias of the formula) by which those objectives
were to be attained; if the section did not, on its true
construction, require this, it certainly made it likely
consideration would have been given to such a statement.
There is no hint of any consideration of this matter,
although measures were stated which included 'allocating ...
units to individual operators in the fishery.'
As between the experts, I should say that I prefer the
evidence of Dr Nicholls, and I accept the views expressed in
his affidavit about the fallacy involved in para.11. In
reaching this conclusion, I have not failed to take account
of what I think were, in the main, verbal infelicities
rather than substantial concessions extracted during his
cross-examination. Dr Pederson appeared to me to be
evading, rather than dealing with, the real issue."

8. His Honour went on to refer to the objectives stated in s.5B and to the provisions of s.7B of the Act, and said (at 207-8):

"Neither these provisions nor the objectives of the Act
suggest that the allocation is to be arbitrary or determined
by the capricious consequences of a kind of statistical
lottery. Where a statute provides for an allocation of a
scarce resource amongst participants in the relevant
industry, in general, and failing some clear indication to
the contrary, the statute should be understood as
authorising a method of allocation in accordance with some
intelligible principle appropriate to achieve a reasonable
division as between those participants. On the expert
evidence before me, the principle upon which the method here
in question was selected is not intelligible (and if there
was some intelligible basis the respondents were in a
position to prove it in detail, so that I am entitled to
give full weight to the inferences arising from the evidence
of Dr Nicholls: Jones v Dunkel ... and the result is a
demonstrably arbitrary and unequal sharing."

9. His Honour concluded that the Act -

"...on its true construction, did not authorise the adoption
of a management plan providing for an allocation in
accordance with para.11. That paragraph 'could not
reasonably have been adopted as a means of attaining the
ends of the power': Williams v The Mayor, Aldermen,
Councillors and Citizens of the City of Melbourne .... See
also The Mayor, Councillors and Citizens of the City of
Brunswick v Stewart ...."

CONCLUSIONS ON THE APPEAL
10. For the reasons we have given in Austral Fisheries, this conclusion was reasonably open to the primary Judge on the material before him. In particular, it was clearly open to the trial Judge to prefer the expert evidence of Dr. Nicholls. Since we are not persuaded that there was any error of principle in the approach taken by his Honour, it follows that these appeals should also be dismissed.