Written evidence from Michael Gordon,
Lecturer in Law, University of Liverpool
RE: PART 1 OF THE EUROPEAN UNION BILL (RESTRICTIONS
ON TREATIES AND DECISIONS RELATING TO THE EU)
1. The call for written evidence issued by the
European Scrutiny Committee acknowledges the potentially important
impact that Part 3 of the EU Bill could have on the doctrine of
parliamentary sovereignty. It is important to note, however, that
the doctrine of parliamentary sovereignty may also be affected
by Part 1 of the EU Bill. For, in requiring referendums to be
held in a range of situations before further competence or power
can be transferred from the UK to the EU, Part 1 of the Bill can
be seen as attempt by the present Parliament to bind its successors,
something traditionally thought to be constitutionally impossible.
2. This submission will contend that this traditional
understanding of parliamentary sovereignty can no longer be supported,
and seek to explain why the steps taken in the EU Bill to place
"referendum locks" on future Parliaments are constitutionally
permissible. Nonetheless, given the undoubted significance of
the measures here proposed, the constitutional desirability of
making these fundamental changes to the legislative process must
be fully explored.
3. In what sense can the EU Bill be said to
be an attempt by Parliament to bind its successors? By Clause
2, a treaty amending or replacing TEU or TFEU can only be ratified
once approved by an Act of Parliament which must satisfy either
the referendum condition or the exemption condition. The approving
Act must therefore either (i) provide for a referendum to be held
on the proposed treaty changes; or (ii) provide that the proposed
changes do not have the kind of impact caught by Clause 4. An
Act of Parliament which purported to approve the ratification
of a treaty amending or replacing TEU or TFEU but which satisfied
neither requirement (i) nor requirement (ii) would be insufficient
to authorise the proposed alterations. Therefore, when a future
Parliament seeks to legislate to authorise the amendment or repeal
of TEU or TFEU, whether a referendum is deemed to be required
or not, an additional legislative hurdle has been put in place
by Clause 2.
4. Similarly, by Clause 3, a Minister may not
confirm an Article 48(6) decision of the European Council unless
it has been approved by an Act of Parliament which must satisfy
the referendum condition, the exemption condition or the significance
condition. In relation to Clause 3, the approving Act must therefore
(i) provide for a referendum to be held on the proposed decision;
or (ii) provide that the proposed decision does not have the kind
of impact caught by Clause 4; or (iii) provide that the decision
falls within a specified part of Clause 4 only, and will not have
a significant impact on the UK. An Act of Parliament which purported
to approve the adoption of an Article 48(6) decision but which
satisfied neither requirement (i) nor requirement (ii) nor requirement
(iii) would be insufficient to authorise the change proposed.
Again, therefore, as with Clause 2, when a future Parliament seeks
to legislate to authorise the adoption of an Article 48(6) decision,
whether a referendum is deemed to be required or not, an additional
legislative hurdle has been put in place by Clause 3.
5. The actual impact in practice of these additional
legislative hurdles will of course vary depending on whether the
Minister deems a referendum to be required or not. When a referendum
is deemed to be required, because in relation to Clause 2 the
proposed change is not exempt, or in relation to Clause 3 the
proposed change is neither exempt nor insignificant, the additional
hurdle will be substantial: a majority of those voting in the
referendum will have to be in favour of the change proposed for
it to be lawfully authorised by Act of Parliament. Where a referendum
is not deemed to be required by the Minister, the additional hurdle
will be insubstantial: all the approving Act need do here is explicitly
provide that the planned measure is either exempt or insignificant
(where the latter option is available) for the proposed change
to be lawfully authorised. While it may therefore appear that
the legislative process has only actually changed where a referendum
is deemed to be required by the Minister, due to trivial nature
of the additional action required of Parliament to satisfy either
the exemption condition or the significance condition, this is
not strictly accurate. That the additional legislative hurdles
put in place by Clauses 2 and 3 will necessarily vary in the extent
of their practical impact does not detract from the fact that
all still constitute additional legislative hurdles. Whether a
future Parliament legislates subject to the requirement that a
referendum must be held, or subject to the requirement that the
proposed measure must be declared to be exempt or insignificant,
the legislative procedure which must be followed to produce approving
Acts for the purposes of Clause 2 or 3 will be altered by the
EU Bill.
6. That the EU Bill constitutes an attempt by
the present Parliament to bind its successors is even more clearly
demonstrated by Clause 6. In the circumstances covered by Clause
6, a Minister can only support a decision when it has been approved
by an Act of Parliament which satisfies the referendum condition.
There is no opportunity for a Minister to declare the decision
exempt from the requirement that a referendum be held: a decision
which falls into Clause 6 will always require a majority of those
voting in a referendum to support it before an approving Act of
Parliament can lawfully authorise the proposed change. In relation
to decisions covered by Clause 6, it is presumed that their adoption
would have a significant impact on the UK, either through a considerable
transfer of power or competence from the UK to the EU, or through
the removal of the UK's veto in a range of policy areas. As a
result, a future Parliament will only be able to legislate to
authorise a decision covered by Clause 6 subject to the proposed
change being approved in a referendum. The additional legislative
hurdle put in place by Clause 6 of the EU Bill will always be
substantial, and is also unavoidable, due to the absence of a
mechanism enabling a Minister to judge a decision exempt.
7. The EU Bill can therefore be seen as an attempt
by the present Parliament to bind its successors. Clauses 2, 3
and 6 limit the capacity of future Parliaments to legislate to
approve the ratification of treaties which amend TEU or TFEU or
the adoption of decisions in a range of areas which transfer power
or competence from the UK to the EU without further steps also
being taken: future Parliaments will be subject to additional
legislative hurdles. In many cases, the additional legislative
hurdle will be that a referendum must be held. This would be a
substantial practical limitation on Parliament's freedom to legislate,
as a negative result in the referendum would prevent the lawful
authorisation of the relevant treaty or decision.
8. A serious question must thus be confronted:
is this constitutionally possible? The UK constitution is underpinned
by the doctrine of parliamentary sovereignty, according to which
Parliament has legally unlimited legislative authority. Does this
doctrine mean that the present Parliament can use its sovereign
power to alter the legislative process that must be adhered to
by future Parliaments? Or is a future Parliament entitled to exercise
its sovereign power without regard to the requirements of the
EU Bill, and legislate to authorise the ratification of a treaty
or the adoption of a decision covered by Clause 2, 3 or 6 without
complying with the referendum condition, or, where relevant, the
exemption or significance conditions? This is likely to be a matter
of fundamental constitutional importance, for a future government
which is seriously inhibited from taking action which would expand
the competence or power of the EU, either as a result of a defeat
in a referendum or an unwillingness even to hold a referendum
due to the prospect of defeat, could well consider attempting
to legislate to authorise the proposed change disregarding entirely
the provisions of the EU Bill.
9. So can a sovereign Parliament bind its successors?
The notion of sovereignty itself provides no inherent solution
to this constitutional conundrum. Past opinion and practice must
therefore be briefly considered. A.V. Dicey, the classic authority
on parliamentary sovereignty, believed that Parliament was not
empowered to bind its successors, and argued that "a sovereign
power cannot, while retaining its sovereign character, restrict
its own powers by any particular enactment". However, Sir
Ivor Jennings, among others, disagreed. According to Jennings,
Parliament could bind its successors in a particular way: by changing
the manner and form required for legislation to be produced. While
absolute limits could not be placed on Parliament's sovereign
power, changes made by one Parliament to the legislative procedure
would have to be adhered to by future Parliaments for statutes
to be validly enacted. For, according to Jennings, if Parliament
"has for the time being power to make laws of any kind in
the manner required by the law", any law made by the Queen-in-Parliament
"will be recognised by the courts, including a rule which
alters this law itself". For Jennings and others, that Parliament
possessed the power to alter the law-making process was a manifestation,
not a limitation, of its sovereign legislative authority. This
"manner and form" view was, however, rejected by Sir
William Wade who influentially insisted that the rule that the
courts will always accept Acts produced by the Queen-in-Parliament
was a "political fact" which could not be altered by
legislation, as it was the very source of legislative authority.
10. Wade further took the case of Ellen Street
Estates v. Minister of Health [1934] 1 K.B. 590 to preclude Parliament
from binding its successors as to the future legislative procedure.
In this case, Maugham LJ claimed that "[t]he Legislature
cannot, according to our constitution, bind itself as to the form
of subsequent legislation". However, as Jennings noted, this
observation was made obiter, and therefore was not legally binding,
for no genuine attempt had been made by Parliament in the statute
considered by the Court of Appeal in Ellen Street Estates to alter
the manner and form for legislating that future Parliaments would
have to follow. An alternative perspective to that provided obiter
in Ellen Street Estates can be seen in a trilogy of merely persuasive
authorities from Australia (Attorney-General for New South Wales
v. Trethowan [1932] AC 526), South Africa (Harris v. Minister
of the Interior 1952 (2) SA 428) and Sri Lanka (Bribery Commissioner
v. Ranasinghe [1965] AC 172). In all three cases, additional legislative
hurdles similar to those set out in the EU Bill were upheld, and
in the Ranasinghe case, decided by the Privy Council, Lord Pearce
suggested that even a sovereign legislature "has no power
to ignore the conditions of law-making that are imposed by the
instrument which itself regulates its power to make law".
11. Nonetheless, this is far from definitive.
The lack of a clear authority indicating whether the UK Parliament
has the power to alter the manner and form which must be adhered
to for legislation validly to be enacted (and in so doing binding
its successors) is primarily due to the fact that Parliament has
mostly refrained from interpreting its sovereign power in this
way. Few attempts have been made by Parliament to modify the legislative
process in an Act of Parliament. Section 1 of the Northern Ireland
Act 1998, according to which a referendum would have to be held
in Northern Ireland before the nation could cease to be part of
the UK, is a notable exception, but it does remain an exception.
12. In recent years, however, it has become
increasingly plausible to assert that Parliament has interpreted
its sovereign power as including the authority to change the manner
and form for the enactment of valid legislation, an interpretation
which it was never conceptually prohibited from adopting. For
Parliament has now legislated to change the law-making procedure
applicable in certain contexts both implicitly and explicitly,
and these changes have been upheld as lawful by the courts.
13. Parliament has done so implicitly by passing
the European Communities Act 1972 which has, in effect, changed
the manner and form for Parliament to produce successfully legislation
which violates EU law. As Clause 18 of the present EU Bill will
confirm, Parliament still retains the capacity to contradict EU
legal norms. However, following R. v. Secretary of State for Transport,
ex p. Factortame (No. 2) [1991] 1 A.C. 603, in which the House
of Lords had to afford domestic supremacy to EU law, Parliament
must now adopt a particular manner and form to enact legislation
which violates EU law: to be valid, a statute which substantively
breaches EU law must expressly state that it is to take effect
regardless of section 2(4) of the ECA 1972.
14. Parliament has further altered the law-making
procedure explicitly by passing the Parliament Acts 1911 and 1949,
replacing the absolute legislative veto of the House of Lords
with a delaying power. In the critical recent case of R. (Jackson)
v. Attorney General [2005] UKHL 56; [2006] 1 A.C. 262, the House
of Lords was called on, in its judicial capacity, to consider
the legal status of the Parliament Acts. The argument derived
from Wade that the Parliament Acts had created a subordinate legislature
of the House of Commons and Queen was decisively rejected by the
Law Lords. Instead, the court held that Parliament had, in the
words of Lord Bingham, created "a new way of enacting primary
legislation", or what Lord Nicholls called a "parallel
route by which... any public Bill introduced in the Commons could
become law as an Act of Parliament". This decision is of
profound importance for present purposes, for here the House of
Lords held that Parliament had the power to effect a statutory
alteration of the manner and form to be adhered to for the enactment
of valid primary legislation. As Baroness Hale observed obiter,
although arguably explaining the necessary implications of the
court's decision in Jackson, "[i]f the sovereign Parliament
can redefine itself downwards, to remove or modify the requirement
for the consent of the Upper House, it may very well be that it
can also redefine itself upwards, to require a particular parliamentary
majority or a popular referendum for particular types of measure".
15. The EU Bill can therefore be seen as further
confirmation of a recalibration of our understanding of the doctrine
of parliamentary sovereignty. If it was arguable prior to the
creation of this Bill, as I submit it was, that the manner and
form understanding of parliamentary sovereignty offered the best
available explanation of contemporary constitutional practice,
in particular following the decision of the House of Lords in
Jackson, then now the constitutional position seems clearer still.
Our understanding of the nature of the doctrine of parliamentary
sovereignty has shifted, and Parliament must now be understood
to possess the power to change the manner and form which must
be followed for valid legislation to be lawfully enacted. In short,
Parliament can now bind its successors, not absolutely, but as
to the manner and form of future legislation. As a result, the
changes to the legislative process envisaged by the EU Bill can
be seen as constitutionally permissible.
16. The changes to the legislative process proposed
in the EU Bill will therefore bind future Parliaments. It is of
course the case that the EU Bill makes no attempt to entrench
the requirement that referendums be held in the circumstances
provided for, by stipulating, for example, that Clause 2, 3 or
6 could only be repealed by legislation approved at a referendum.
This lack of entrenchment ensures that a future Parliament could
in principle, if it were judged to be necessary, repeal, entirely
or partially, the EU Bill by an ordinary Act of Parliament. However,
on the assumption that the EU Bill is passed in its present form,
until such a repeal was effected (if ever), any future Parliament
would be required to adhere to the terms of Clause 2, 3 or 6 to
enact valid legislation on any matter covered by these provisions
while they remain on the statute book. The changes to the manner
and form for the valid enactment of future legislation made by
the EU Bill will therefore be of full legal effect unless or until
repealed, with future Parliaments unable to legislate to authorise
the ratification of treaties or the adoption of decisions covered
by Clause 2, 3 or 6 other than in accordance with the modified
legislative procedure.
17. Nonetheless, while the enactment of these
"referendum locks" is not constitutionally prohibited,
and can be reconciled with the doctrine of parliamentary sovereignty,
the matter of the constitutional desirability of these changes
remains. The changes to the manner and form required for legislative
action in the areas covered by Clauses 2, 3 and 6 must be examined
in context. Parliament has historically been extremely reluctant
to introduce onerous additional legislative hurdles which would
bind future Parliaments, and the EU Bill deviates radically from
this past practice.
18. The EU Bill will require referendums to
be held to authorise the ratification of treaties or the adoption
of decisions in a significant range of situations. The breadth
of areas in which a referendum will be required is striking, especially
when it is considered that the UK-EU relationship is being singled
out for this exceptional treatment. The Bill will give UK citizens
far more say in the structure and development of the EU's constitutional
arrangements than those of the UK itself. It is therefore especially
difficult to justify adequately the extensive use of referendums
to authorise the further alteration of the powers and institutional
structure of the EU when the population of the UK is provided
with no similar potentially wide-ranging involvement in the development
of the domestic system of governance. While in recent years referendums
have begun to be utilised more frequently as a way of validating
domestic constitutional change, with national referendums being
held in Scotland and Wales in 1997 and in Northern Ireland in
1998 prior to the devolution settlement being established, and
a UK-wide referendum now planned on the voting system of the House
of Commons, there is no guarantee that even the most critical
areas of the UK constitution will in future be reformed in accordance
with the wishes of the population expressed in a specific plebiscite.
The EU Bill can therefore be seen to treat the development of
the EU as if a matter of unique interest to UK citizens, a judgment
which seems unsustainable.
19. Further, the extensive range of situations
in which a referendum would be required under the EU Bill could
exacerbate voter apathy. Can it be said that a referendum to protect
the UK's veto in any one of a number of contexts, such as those
delineated, for example, in Schedule 1, would be likely to animate
citizens in such a way as to ensure something approaching a sufficient
turnout could be achieved? While there may be a strong democratic
case for guaranteeing the use of referendums in the circumstances
of a palpable, substantive shift of power from the UK to the EU,
the EU Bill seems to go well beyond this in requiring referendums
to be held on what could be viewed as relatively esoteric issues.
Consequently, the question of how adequate voter turnouts could
be obtained, to ensure that the result of any referendum held
is seen as legitimate, requires full consideration. Yet there
are no easy solutions, for legislating to impose a minimum turnout
threshold which would have to be met for the result of a referendum
to have legal effect would be likely to cause further problems,
with the prospect of repeat referendums inducing even greater
voter fatigue.
20. The EU Bill therefore takes the UK into
uncharted constitutional territory. While the present Parliament
binding its successors as to the manner and form of future legislation,
in the way outlined above, should not be understood to be constitutionally
prohibited, this dramatic divergence from past practice must be
comprehensively evaluated. Given the UK's traditions of constitutional
flexibility and the primacy of political as opposed to legal limits
on legislative authority, Parliament must be wary of placing itself
in a straightjacket. While it may have the constitutional power
to do so, Parliament should not too readily provide that its legislation
shall take effect subject to the approval of the general population
in a referendum. The fact that the EU Bill does not purport to
entrench itself, noted above, does serve formally to preserve
Parliament's ultimate constitutional authority, for any future
Parliament could outright repeal the EU Bill by an ordinary Act
of Parliament. Nevertheless, it may be difficult in practice for
a government to seek to justify an attempt to remove legal rights
to popular participation in the legislative process from citizens.
As such, Parliament should proceed cautiously when making such
profound alterations to the future legislative process. While
the ordinary law-making procedure which requires only the assent
of the House of Commons, House of Lords, and Queen for a statute
to be enacted should not be viewed as sacrosanct, extensive changes
to the legislative process must be fully justified. If this consideration
is taken into account, it would seem to indicate that the kind
of "referendum locks" proposed by the EU Bill should
be reserved for truly exceptional cases of the utmost constitutional
or political significance only, when an informed and engaged population
can quite properly resolve questions of critical national importance,
rather than being employed excessively in relation to a single
area of political activity.
7 December 2010
|