Supreme Court declares Part III of Industrial Relations Act 1946 invalid for unconstitutional delegation of legislative power.
Sintesi Legislator
Generato dall'IAThe Supreme Court allowed the appeal, declaring Part III of the Industrial Relations Act 1946 invalid. Electrical contractors challenged a registered employment agreement, arguing the Labour Court's power to register such agreements constituted an unconstitutional delegation of legislative authority. The court held that because the registered agreement binds non-parties and carries criminal enforcement, it functions as legislation, which the Constitution reserves exclusively to the Oireachtas.
Regesto Legislator
Generato dall'IAs. 27 of the Industrial Relations Act 1946; Article 15.2.1 of the Constitution of Ireland: The power to enact legislation which binds non-parties and carries criminal sanctions cannot be delegated to a court or body, as such delegation violates the exclusive legislative competence of the Oireachtas.
Parole chiave
Generato dall'IADisposizioni normative citate
Generato dall'IAGiurisprudenza citata
Generato dall'IATesto completo
THE SUPREME COURT
310/10
Denham C.J.
Murray J.
Fennelly J.
O’Donnell J.
Clarke J.
Between:
Benedict McGowan, Kieran Marshall, Padraig Brady, Brian Kelly, Donal Phelan, David McCarthy, Jimmy Lee, Camlin Electric
Limited, and
Cavanty Electrical Limited
Applicants
And
The Labour Court, Ireland and
The Attorney General
Respondents
And
The Technical Engineering and Electrical Union and
The Association of Electrical Contractors (Ireland), the
Electrical Contractors Association and
The National Electrical Contractors of Ireland
Notice Parties
Judgment of the Court delivered on the 9th of May, 2013 by O’Donnell J.
1 It is possible, although not without some difficulty, to detect in these proceedings an important point of constitutional law
concerning the Industrial Relations Act 1946 (hereinafter the “Industrial Relations Act”, the “Act” or the “1946 Act”), albeit almost
entirely obscured in a thicket of procedural complication and confusion.
2 The appellants are, or at least are alleged to be, electrical contractors and as such, affected by an employment agreement
registered by the Labour Court on the 24th of September 1990 pursuant to the provisions of Part III of the Industrial Relations Act
1946 setting out certain terms and conditions of employment of electricians within the construction sector. The operation,
effectiveness and validity of that Registered Employment Agreement (hereafter “REA”) has been sought to be challenged in multiple
proceedings which have led, by a circuitous route, to this appeal.
3 It is necessary to explain in some detail the legal operation of Registered Employment Agreements in general, and indeed the wider
operation of the Industrial Relations Act 1946, before it is possible to commence the analysis of the constitutional issue raised in
these proceedings.
4 The 1946 Act contained two mechanisms under which a general sectoral agreement made in respect of terms and conditions of
employment in a specified industry or sector of an industry may become legally enforceable both in civil and criminal law. Although this
case concerns the provisions of Part III, it is also necessary to have regard to the provisions of Part IV which have been the subject
of the case law relied on in this appeal.
5 Part IV of the Industrial Relations Act (ss. 34-58) permitted the Labour Court to establish Joint Labour Committees (“JLCs”) either
where there was substantial agreement among groups representing employers and employees, or where it was considered that the
existing mechanism for the regulation of remuneration and other conditions made it expedient to establish such a body. A JLC could
then make a submission to the Labour Court which, if accepted by the Labour Court, would result in the making of an Employment
Regulation Order (“ERO”) giving effect to the proposals of the JLC. The effect of such an ERO was to make its provisions concerning
remuneration and conditions of employment part of the contract of employment between an employer and an employee within the
sector (whether represented in the JLC or not) and failure to comply with such terms was not only enforceable in civil law, but also
gave rise to a criminal offence punishable by a fine. Provision was also made for an inspectorate to assist in the enforcement of the
provision.
6 The provisions of Part IV of the Act can be traced back to the Board of Trades Act 1909 which was part of the reforming labour
legislation introduced at that time in response to the growing power of the trades unions. JLCs tended to be utilised in industries with
a transient work force and often lower paid employees, and where traditional collective bargaining could not take hold. The perceived
benefit of a JLC was that it provided a structure for a form of collective bargaining in such industries which might not otherwise arise
naturally because of the structure of the industry and the nature of the workforce.
Part III of the Industrial Relations Act 1946
7 The provisions of Part III of the Act of 1946 have similarities of structure to those of Part IV. However it appears that Part III is
unique to the Irish code of industrial relations and cannot be traced back to any pre-existing body of legislation. Under Part III an
employment agreement, defined as an agreement regulating remuneration and conditions of employment of work and made between
trade unions and an employer or a group of employers or at a meeting of the registered Joint Industrial Council, may on the application
of the parties thereto, be registered by the Labour Court. On any such application the Labour Court is obliged to register the
agreement if it is satisfied that the conditions of six subparagraphs of s.27 of the Act have been complied with. It will be necessary to
return to this section in some detail later in this judgment. Once registered, an REA, like an ERO under Part IV, becomes incorporated
in the contract between the employer and employee and is enforceable by criminal prosecution. The agreement may be varied by
application brought by the parties to the original agreement, and may be cancelled by the court either on a joint application of the
parties, or if the Labour Court is satisfied that there has been substantial change in the trade or business. But most significantly for
present purposes, an REA like an ERO applies not just to the parties thereto and those they represent, but to every worker and
employer in the sector, whether or not they were a party at the original agreement, or represented in the conclusion of the
agreement, or even in existence at the time it was made. Thus, s.30(1) of the Act provides:
“A registered employment agreement shall, so long as it continues to be registered, apply, for the purposes of this
section, to every worker of the class, type or group to which it is expressed to apply, and his employer, notwithstanding
that such worker or employer is not a party to the agreement or would not, apart from this subsection, be bound
thereby.”
8 The provisions of Part III appear somewhat anomalous today. In general collective agreements are considered not to be legally
enforceable either by the parties thereto, or the persons represented by the parties to any such agreement. Indeed in general the
fact that collective agreements exist outside a precise legal framework and are not enforceable by either the civil or criminal law, is
normally seen as a desirable feature in industrial relations. Furthermore, since 1946 there has been considerable further development
of statutory regulation of employment but it has been away from private sectoral agreements and towards legislation having general
effect imposing obligations such as setting minimum terms and conditions of employment in employment. But the most striking feature
of Parts III and IV of the 1946 Act to modern eyes is the fact that both EROs and REAs are made part of the criminal law and bind
everyone who participates in the relevant sector. Furthermore, the relevant provision of the criminal law is made not by the
Oireachtas, but rather by private individuals, themselves participants in the industry to be regulated. Not only therefore does the
scheme of the 1946 Act confer a high degree of autonomy on participants to an ERO or an REA, in the sense that they are
empowered to make law for themselves, they are also empowered to make law for others giving rise to the prospect of burdensome
restraints on competition for prospective employers and intrusive paternalism for prospective employees. Given these unusual features
it was perhaps inevitable that this scheme would come under increasing scrutiny.
9 The first significant challenge arose in the context of Part IV of the Act in Burke v. The Labour Court [1979] I.R. 354 (“Burke”).
That case arose in a narrower context than that which has given rise to the present litigation. In Burke an ERO had been made in
relation to the hotel industry without regard to submissions made on behalf of the employers. The ERO was challenged on non-
constitutional grounds by representatives of the employers and the Supreme Court held that the relevant JLC had failed to comply
with fair procedures. In the course of the judgment however, the court took the opportunity to make some pointed observations
about the legislation and in particular the extent of the law-making power which appeared to have been conferred on the parties to
the JLC, or on those parties and the Labour Court conjointly. Thus, Henchy J. said of Part IV:
“It will be seen, therefore, that the power to make a minimum-remuneration order is a delegated power of a most
fundamental, permissive and far-reaching kind. By the above provisions of the Act of 1946 Parliament, without reserving
to itself a power of supervision or a power of revocation or cancellation (which would apply if the order had to be laid on
the table of either House before it could have statutory effect) has vested in a joint labour committee and the Labour
Court the conjoint power to fix minimum rates of remuneration so that non-payment thereof will render employers liable to
conviction and fine and (in the case of conviction) to being made compellable by court order to pay the amount fixed by
the order of the Labour Court. Not alone is this power given irrevocably and without parliamentary, or even ministerial,
control, but once such an order is made (no matter how erroneous, ill judged or unfair it may be) a joint labour committee
is debarred from submitting proposals for revoking or amending it until it has been in force for at least six months. While
the parent statute may be amended or repealed at any time, the order, whose authors are not even the direct delegates
of Parliament, must stand irrevocably in force for well over six months.” (pp. 358 and 359)
Later in the judgment he returned to this issue:
“As I have earlier observed, the delegated power that was vested in the Committee was of the most extensive nature. It
enabled the Committee to formulate the proposals for an order fixing minimum rates of remuneration. All the Labour Court
was to do was to refer the proposals back to the Committee with observations. The Labour Court is given no power of
initiation or amendment. It could but make or refuse to make the order. Essentially, therefore, the order making body was
the Committee. Apart from the skeletal provisions in the second schedule of the Act of 1946 as to its constitution,
officers and proceedings, the Act of 1946 is silent as to how a committee are to carry out their functions in making
orders.” (p. 361)
In the particular case, the court held that given what was described as the extensive nature of the delegation, it was necessary to
conclude that the Oireachtas had intended that the power would be exercised within the terms of the relevant Act and based on
fairness and reasonableness and good faith. In the circumstances, the particular order was quashed.
10 It is clear that the court was considerably exercised by the apparent scope of the law-making power conferred by Part IV. It is
perhaps noteworthy that, although reported in different volumes of the Irish Reports, the Burke case was argued almost
contemporaneously with the landmark case on the validity of subordinate regulation (Cityview Press v. An Chomhairle Oiliúna [1980]
I.R. 381 (“Cityview Press”)) and the judgments delivered within a day of each other. It seems clear therefore that the observations
made in Burke were influenced by, and reflective of, the consideration the court was giving at the time to the proper limits of the
delegation of law-making authority.
11 Thereafter, a number of challenges were brought to the provisions of Part IV of the Act but most of them were compromised or
resolved without reaching the issue latent in the observations made in Burke. Ultimately, in John Grace Fried Chicken v. The Labour
Court [2011] 3 I.R. 211 (“John Grace”), the matter reached the High Court. Feeney J. held that the Act lacked any principles or
policies for the exercise of the power of law-making conferred on JLCs and that accordingly, the provisions of Part IV were repugnant
to the Constitution. That decision was not appealed to this court. Instead the Industrial Relations (Amendment) Act 2012 was
introduced. The background to the that decision appears to be that the provisions of the 1946 Act and in particular those establishing
the EROs had been under review not least as part of the Memorandum of Understanding under which structural reforms are to be
made in return for funding to this country from the EU, ECB and IMF. The 2012 Act sets out much more elaborate principles and
policies and indeed provisions for review and reconsideration by the Minister and the Oireachtas, and thus addresses (and appears to
accept as correct) the decision in John Grace. Significantly for present purposes, the opportunity was also taken to made wide-
ranging changes to Part III to similar effect to the changes introduced in respect of Part IV. This case concerns however the
provisions of Part III in their unamended form.
These Proceedings
12 The proceedings involved in this appeal are an amalgam of a series of cases all of which relate to the provisions of the REA of the
24th of September 1990 made in respect of the electrical trade in the construction sector. That agreement has been the subject of
repeated disputes in particular involving complaints by a group of employers organised in the National Electrical Contractors
Association, the fourth named notice parties hereto. They have persistently complained that they have been bound by an agreement
to which they were not a party, and in which the employers’ interests were represented by parties which they do not consider
represent their interests.
13 The impetus for this latest round of litigation was a proposal made to the Labour Court by the employee representatives to vary
the REA and to increase the minimum pay of electricians in the construction sector. At the same time, a District Court prosecution
had been commenced against Camlin Limited for breach of the existing REA. On the 27th of May 2008 those proceedings were
adjourned and a consultative case stated prepared for the High Court. The District Court proceedings and case stated did not raise,
as they could not, any issue of the constitutional validity of Part III. Meanwhile, an application had been made to the Labour Court on
the 22nd of May 2008 by, it was said, 500 contractors seeking a cancellation of the REA. Some procedural skirmishing took place and
the Labour Court refused the applicant contractors’ request for an adjournment to await the outcome of the case stated proceedings.
A large number of applicants then sought judicial review of the Labour Court decision and an injunction restraining further hearing.
These proceedings (which were then known as the “Sullivan Proceedings” after the then first named plaintiff) were commenced, and
leave to seek judicial review was granted on the 13th of June 2008 together with an interim injunction restraining a further hearing.
However, on the 20th of October 2008 the injunction was lifted by O’Keeffe J. because, we are informed, of the unwillingness of the
applicants to offer an undertaking as to damages, and also because of concerns about the constitution of the applicants. The
proceedings nonetheless remained in being. Since there was now no injunction restraining the proceedings, the Labour Court
proceeded with an eleven day hearing and on the 26th of February 2009 issued a determination which refused the initial application to
vary the agreement by increasing the remuneration, but also refused the application made on behalf of the discontented contractors
for a cancellation of the existing REA. Yet again the representatives of the contractors sought judicial review to challenge the
decision refusing cancellation and a further set of proceedings (“the Bunclody Proceedings”) were commenced which ultimately were
heard with these proceedings in the High Court. The respondents to these proceedings sought to clarify the identity of the multitude
of applicants named, and that they were indeed electrical contractors and were not members of any body which was a party to the
1990 REA. Accordingly they raised this issue by way of particulars. This proved to be anything but a straight forward task. In the end
the applicants’ response was not to provide particulars but to seek to reduce dramatically the number of named applicants. Even
then, of the seven applicants remaining one was not an employer at the time of the institution of the proceedings and three had
previously been members of organisations that were party to the REA. One further party, Camlin (which as already observed was the
party to the case stated), had been struck off. In the High Court, Hedigan J. observed that:
“…it must be noted that throughout the evidence before this Court there has been some degree of confusion as to the
identity of the applicants. Many of them were alleged not to be employers, thus calling into question their locus standi
and their representativity. The drastic reduction in the number of plaintiffs still leaves open this question and it is a very
undesirable aspect of this case to date.” (para. 10.2)
Undesirable is, if anything, a mild adjective in the circumstances. It is of the essence of any litigation, but particularly constitutional
litigation, that it be firmly based on the ascertainable facts relating to real persons who can claim to have suffered a measurable and
identifiable injury which requires a remedy which may even extend to the striking down of legislation enacted by the Oireachtas. It is
the factual matrix of an individual’s situation which gives real focus and reality to a claim of unjust infringement of a constitutional
provision and which justifies the court in addressing the validity of legislation of general application which in other circumstances, and
for many other citizens, may conceivably be beneficial. A claim is not a classroom hypothetical and the identification of real claimants
who have the identified complaint is not an optional extra. It is truly disturbing that a claim initiated in the names of a legion of
parties could shrink so dramatically on the simplest inquiry and request for verification. In addition, passage of time and the economic
downturn have had their own effect on the proceedings. The party to the case stated, Camlin Ltd., was struck off the register of
companies for failing to make returns thus ending those proceedings. Similarly the Bunclody proceedings which were heard with these
proceedings in the High Court were struck out in July 2012 in the Supreme Court following the appointment of a liquidator to the last
remaining appellant in that case. The result is that what remains in this appeal is a far reduced number of applicants to the original
Sullivan proceedings (now the McGowan proceedings) and which retain their original structure as a challenge to the Labour Court
commencing a hearing even though in the event the hearing did proceed, and which became the subject of a separate challenge,
which, itself although heard in conjunction with the Sullivan/McGowan proceedings, is not now before this court.
14 The foregoing is a necessarily truncated account of only some of the unsatisfactory aspects of these proceedings. Of even more
significance is the fact that the issue which might have been thought to be central to this case, namely the question of alleged
unauthorised delegation of legislative power contrary to Article 15 of the Constitution, was by no means central to the argument in
the High Court either the Sullivan/McGowan proceedings, or indeed the Bunclody proceedings. In the light of the John Grace decision
and the apparent acceptance of that decision by the State authorities that argument has certainly moved centre stage but it must
be said that there was little trace of it in the voluminous pleadings in either the Sullivan or the Bunclody case. What did happen
however was that the point was canvassed at some length in very comprehensive submissions exchanged between the parties both
at the opening of the High Court case and at its closing. As a result, the position is that the question was undoubtedly argued in the
High Court, and in some detail, and was addressed in the High Court judgment.
The Judgment of the High Court
15 The learned High Court judge dismissed the applicants’ claim on all grounds. He held that in so much as the claim was a challenge
to the REA on non-constitutional grounds, it was out of time in that the proceedings had not been commenced within three months of
the decision of which complaint was made. That decision is not under appeal and accordingly it is not necessary to express any view
upon its correctness. The judge also rejected a series of other challenges to the Labour Court decision and which are now not before
this court. In respect of the constitutional challenge he said the following:
“This aspect of the case has been brought in judicial review proceedings and I can see no reason why it should not be
subject to the same requirement as to time limits as any other application. That being so, it is brought well outside the
period of three months from the date when grounds for the application first arose and for that reason alone ought not to
be allowed to proceed. It would also, I think be undesirable that the constitutional challenge should proceed alone in
judicial review proceedings. There are many issues concerning the balancing of interests and, as noted above, the
identification of the plaintiffs that need much greater elaboration than they have had in these proceedings. I will not
therefore proceed to consider the constitutional issues raised herein. They should be brought if the applicants wish to do
so by way of plenary proceedings.” (para. 11.10)
This Appeal
16 It may be appreciated that the appeal which arrived in the Supreme Court was one with myriad complications. As a result however
of active case management by a member of this court, the parties ultimately agreed to narrow the issues that would be addressed on
this appeal to two:
(a) Are the appellants entitled to raise the issue at (b) and; is the Supreme Court entitled to determine this issue in
circumstances where although the issue was raised in the proceedings from which the appeal is taken, and argued before
the High Court, the learned High Court judge expressly did not determine the issue?
(b) Does Part III of the Industrial Relations Act of 1946 or any section thereof contravene Article 15.2.1 of the
Constitution by delegating the making, variation and cancellation of registered employment agreements to the Labour
Court and the parties to such agreements?
The appellants agreed to abandon all other grounds of appeal and it is apparent from the formulation of issue (a) that the
respondents also agreed not to rely on any of the other grounds mentioned in the High Court judgment for declining to hear the
constitutional claim. The first issue raised here therefore, is the question of the entitlement of the appellants to argue, and this court
to decide, the core constitutional issue as set out at (b) namely, the question of the alleged excessive delegation of legislative
powers. Furthermore the issue to be considered is simply whether this court can, and should, hear an issue which was argued but not
decided in the High Court.
17 It is indeed easy to sympathise with the High Court’s frustration with the manner in which the case has advanced and its belief
that it would be better advanced in plenary proceedings with a full elaboration of facts and argument focussed on this specific issue.
However, there is no doubt that the Supreme Court can hear and determine an appeal on an issue which for whatever reason, the
High Court has heard but not determined. The respondents accepted that the views expressed by Keane C.J. in A.A. v. Medical
Council [2003] 4 I.R. 302 at p. 308, represent a reasonable statement of principle:
“…the court is not automatically precluded in every case from considering such an issue simply because it has not been
subject of a determination by the High Court Judge. Whether a party is to be precluded from advancing again arguments
which were relevant to an issue in the case and on which he relied on in the High Court must, in the interests of justice
be determined according to the circumstances of the particular appeal before this court.”
Keane CJ also said:
“This court is a court of appeal only and cannot exercise any jurisdiction other than an appellate jurisdiction, save under
Articles 26 and 12.3 of the Constitution and then it is exercising the case stated jurisdiction vested in it pursuant to s. 38
of the Courts of Justice Act 1936 and s. 16 of the Courts of Justice Act 1947. For that reason, this court has consistently
declined to consider an issue of constitutional law which, though arising in a case not yet determined by it, has not been
fully argued and decided in the High Court, save in the most exceptional circumstances.” (p. 307)
This case, it should be said, is not one where the point was not argued or fully argued. It is one where the point was fully argued, but
not decided. Chief Justice Keane also referred to the observations of Murray J. (with whom Denham and Murphy JJ. concurred ) in
Dunnes Stores v. Ryan [2002] 2 I.R. 60 where he suggested that rather than remitting it again to the High Court where the common
experiences of such issues are almost invariably further appealed to this court for final determination, the Supreme Court ought to
have jurisdiction to determine a point which arose for resolution in proceedings but on which no decision on any disputed question of
fact is required.
18 This is a sensible and pragmatic approach. In a perfect world it would undoubtedly be preferable to have meticulous and detailed
argument in the High Court followed by a comprehensive judgment and a speedy appeal to this court, further comprehensive argument
on the issue as addressed in the judgment of the High Court and, a final, as it is to be hoped, conclusive determination of all issues in
this court. But the court cannot ignore the reality that the course of litigation is often very far from the ideal, and indeed this case is
one example. Here there are a number of factors which suggest that the point should be considered and determined by this court.
The issue is one which has been mooted for a considerable time, since at least the judgment in Burke. The relevant REA is still in full
force and effect. Indeed, the third named appellant has been the subject of a District Court prosecution which was commenced in
2008 and which is awaiting the outcome of this decision. The REA will continue to have effect therefore and the uncertainty over its
validity and indeed the validity of the underlying statutory scheme is undesirable. There have been three separate pieces of litigation
in relation to this REA alone and a lengthy hearing both in the Labour Court and in the High Court. Considerable costs have been
incurred on all sides. The point was fully argued and it was not adjudicated on not because, as sometimes occurs, the trial court had
decided the case in the plaintiffs’ favour on non-constitutional grounds but rather because the court considered that it was preferable
that the case be brought by plenary procedure. It is not at all clear that this is a valid ground for declining to address a point
otherwise properly before the court. To decline to hear and determine this issue would mean requiring the parties to incur substantial
costs without the issues between the parties being resolved, and exposing the plaintiffs to the possibility of ongoing criminal
prosecution and a choice between having to recommence proceedings or submitting themselves to a regime which they consider
unconstitutional. Such an outcome would not be consistent with the administration of justice. Accordingly, albeit reluctantly, the
court considers it necessary to address the central issue raised in this appeal.
Article 15.2.1 of the Constitution
19 Article 15.2.1 of the Constitution is in very clear terms:
“The sole and exclusive power of making laws for the State is hereby vested in the Oireachtas: no other legislative
authority has power to make laws for the State.”
This Article while striking and emphatic, might appear almost tautologous. If the “sole and exclusive power” for making laws is vested
in the Oireachtas, then it follows that no other body has power to make laws for the State. The interesting historical background to
this Article, and its predecessor in Article 12 of the Free State Constitution, is set out both in Morgan, The Separation of Powers in
the Irish Constitution, (Dublin, 1997) p. 261 and, Hogan, The Origins of the Irish Constitution: 1928-1941, (Dublin, 2012) pp. 335-339
(hereinafter “The Origins”). As Keane C.J. observed in Laurentiu v. The Minister for Justice [1999] 4 I.R. 26:
“Historically, this Article can be seen as an uncompromising reassertion of the freedom from legislative control by the
Imperial Parliament at Westminster of the new State.” (p.83)
It is also (and perhaps in part for the same historical reason) an assertion of a core democratic principle. Since all power comes from
the People, the only body with power to make legislation binding the People, is the Oireachtas containing as it does the chosen
representatives of the people. As Keane C.J. continued;
“But it [Article 15] is also an essential component in the tripartite separation of powers which is the most important
feature of our constitutional architecture and which is enshrined in general terms in Article 6.” (p.83)
As recorded in Hogan, The Origins, a specific question was raised by Gavan Duffy J. at the time of the drafting of the Constitution as
to the necessity to include in Article 15.2.1 some saver in respect of statutory instruments and orders. The view was taken in a
memo produced by Philip O Donoghue legal assistant to the Attorney General one the principal drafters of the new Constitution, and
which received the express agreement of the Attorney General Patrick Lynch however, that such rules and orders were not “laws”
within the meaning of the Article. Such rules and orders were;
“intimately related with legislative enactments. They are considered part of the law and have the force of the law but
alone do not constitute legislation. They must always be referred back to the enabling statute under which they are
made. Very little consideration will indicate the abuses which will grow up if the legislature contented itself with an
enacting loose and indefinite principles adding that the Minister could give effect to such principles by rules and
regulations.
Statutory Rules and Orders as the title suggests, are intimately related with legislative enactments. They are considered
part of the law and have the force of law but alone do not constitute legislation. They must always be referred back to
the enabling Statute under which they are made. Very little consideration will indicate the abuses which would grow up if
the legislature contented itself with enacting loose and indefinite principles adding that the Minister could give effect to
such principles by rules and regulations.”
Indeed the position in Ireland was contrasted favourably with the position in the United Kingdom where attention was drawn to the
views of the then Lord Chief Justice Sir Gordon Hewart who in his book The New Despotism, (London, 1929) had recently expressed
the view, trenchantly and possibly intemperately, that the proliferation of delegated legislation was a threat to democracy.
20 The understanding of Article 15.2.1 and the place of subordinate regulation within the constitutional scheme was illustrated by the
thoughtful judgment of Hanna J. in an early case which arose after the coming into force of the Constitution, Pigs Marketing Board v.
Donnelly (Dublin) Ltd. [1939] I.R.413 (“Pigs Marketing Board”). There a challenge was made to the price fixing function which had
been delegated to the board by the Pigs and Bacon Acts 1935 and 1937. In rejecting that claim Hanna J. said:
“It is axiomatic that powers conferred upon the Legislature to make laws cannot be delegated to any other body or
authority. The Oireachtas is the only constitutional agency by which laws can be made. But the Legislature may, it has
always been conceded, delegate to subordinate bodies or departments not only the making of administrative rules and
regulations, but the power to exorcise, within the principles laid down by the Legislature, the powers so delegated and
the manner in which the statutory provisions shall be carried out. The functions of every Government are now so
numerous and complex that of necessity a wider sphere has been recognised for subordinate agencies such as boards and
commissions. This has been specially so in this State in matters of industry and commerce. Such bodies are not law
makers; they put into execution the laws as made by the governing authority and strictly in pursuance therewith, so as
to bring about, not their own views, but the result directed by the Government.
Now, what is the complaint here? What is the legislative power which it is suggested the Pigs Marketing Board exercises?
It is the fixing of the hypothetical price. It has been submitted that, when the Pigs Marketing Board has constitutional
power to fix the appointed price because they are directed to consider certain matters in determining it, as there is no
schedule of topics to be considered by the Pigs Marketing Board in fixing the hypothetical price, they are in a position of
legislators in that respect. But I cannot accept this view of the duties of the Pigs Marketing Board in reference to the
hypothetical price, where the Legislature has directed them to fix, not any price, but the price which, in their opinion,
would be the proper price under normal conditions. That is a statutory direction. It is a matter of such detail and upon
which such expert knowledge is necessarily required, that the Legislature, being unable to fix such a price itself, is
entitled to say: “We shall leave this to a body of experts in the trade who shall in the first place determine what the
normal conditions in the trade would be apart from the abnormal conditions prescribed by the statute, and then form an
opinion as to what the proper price in pounds, shillings and pence would be under such normal conditions.” The Pigs
Marketing Board, in doing so, is not making a new law; it is giving effect to the statutory provisions as to how they should
determine that price.” (pp.421 and 422)
21 The judgment in Pigs Marketing Board is an early and sophisticated analysis of the issue. The leading modern authority is Cityview
Press. There the plaintiff challenged s.21 of the Industrial Training Act 1967 which empowered the defendant, An Chomhairle Oiliúna,
to make a levy order fixing the amount of the levy to be collected from each enterprise in a specified industry and used for training
recruits to that industry. The Levy Order for the printing industry which the plaintiff was refusing to pay imposed a levy of one per
cent of total emoluments on all employees less £20,000. The argument was that the Act did not provide the defendant body with any
precise guidelines as to the basis on which the levy should be made, i.e. whether by reference to turnover, total salaries and wages
profits, or some other basis. This argument was rejected both in the High Court and Supreme Court. The test to be applied was
identified in the judgment of O’Higgins C.J.:
“In the view of this Court, the test is whether that which is challenged as an unauthorised delegation of parliamentary
power is more than a mere giving effect to principles and policies which are contained in the statute itself. If it be, then it
is not authorised; for such would constitute a purported exercise of legislative power by an authority which is not
permitted to do so under the Constitution. On the other hand, if it be within the permitted limits - if the law is laid down in
the statute and details only are filled in or completed by the designated Minister or subordinate body - there is no
unauthorised delegation of legislative power.” (p.399)
The principle has been stated by Casey, Constitutional Law in Ireland, 3rd edition, (London, 2000) p. 255:
“The Oireachtas, it seems, may delegate a power to put flesh on the bones of an Act; but anything going beyond this will
be constitutionally suspect.”
22 It is said that the differences between those cases in which there has been held to be a breach of Article 15.2.1 and cases in
which it is accepted that there has been no such breach is largely one of degree and that it is difficult to detect a precise dividing
line between what is permissible and impermissible. That may be so on a case by case basis although there is a clear principle as set
out above. However, it seems that the difference in this case is not just one of degree, rather it is structural in nature. In that
respect it is instructive to compare the extensive area and nature of authorisation involved here with that in the Cityview Press case,
itself considered by some commentators to be close to the limits of permissible rule making.
23 In the Cityview Press case, the delegation or authorisation under s. 21 of the Industrial Training Act 1967 may be said by some to
be vague, but a number of important features were identified, particularly in contrast to the position which applies under the
Industrial Relations Act 1964. The area of authorisation was narrow. It was the power to fix the amount of the levy, the Oireachtas
having already made the decision that An Chomhairle Oiliúna was to be funded by a levy on the relevant designated industrial activity.
The body which was authorised to fix the levy was itself public law body exercising powers constrained by statute. Accordingly any
order made would be subject to consultation with the relevant industrial training committee (s.21(3)), review and approval by the
Minister (s.21(4)) and laid before each house of the Oireachtas, either of which was entitled to annul it within 21 days (s.21(6)).
Furthermore, as McMahon J. in the High Court observed:
“ There can be no doubt that s21 is so expressed as to as to confine the use of any money raised by a levy ordered to
meeting any expense of AnCo in relation to the performance of its functions under the Act in respect of the designated
industrial activity in respect of which the levy order was made”.
For that reason, and indeed for more general reasons of public law, there could be no question of the money raised being used for any
other activity or for example, as a form of taxation or covert revenue raising. The area of decision making accorded therefore to An
Chomhairle Oiliúna under s.21 was limited in a number of respects. Its power to fix the quantum of the levy was restricted by the
object for which the levy was to be fixed. It retained a discretion as to the precise manner in which the levy should be raised as
indeed was argued, whether by reference to turnover, profits, or otherwise. But given the broader constraints just identified, that is a
very limited power and furthermore raises no obvious issues of policy. It was argued that this was still an excessive delegation but it
was accepted by the court that this was akin to the price fixing function upheld in Pigs Marketing Board. For present purposes
however, it is only necessary to identify the significantly limited scope of authorisation that was in issue in that case.
24 The contrast with the scope of power afforded under the 1946 Act is instructive. If the 1946 Act conformed to the same pattern
as that established in the 1967 Industrial Training Act, then the relevant terms would be set by the Labour Court perhaps after
consultation with other public bodies and subject to ministerial approval and Oireachtas review. Even if such a structure were in place
the breadth of the power afforded would still be telling. An REA can make provision not merely for remuneration, as was the case in
Burke, but can make provision for any matter which may be regulated by a contract of employment. Thus, it can determine wages,
pensions, pension contributions, hours of work, health insurance, grievance procedures, discipline procedures, staffing levels,
production procedures, approved machinery or equipment, and anything else in the employment relationship. It is in the words of
Henchy J. in Burke, a delegation of a “most fundamental and far-reaching kind”. It involves a fundamental part of the person’s life (if
an employee), and their business (if an employer).
25 The extent of the delegation is also of significance. What is unusual and possibly unique is that the law making power granted
under the 1946 Act is granted over a broad area of human activity to private persons, themselves unidentified and unidentifiable at
the time of the passage of the legislation. When an employer such as the third named appellant is the subject of prosecution for
breach of a registered employment agreement, that amounts to a clear allegation that a part of the law of the State has been
breached. In such a case the particular provision which it is alleged has been breached has been made by the private parties to the
employment agreement which has been registered by the Labour Court. The Labour Court itself has no power of consultation or even
(as is the case of an ERO made under Part IV of the 1946 Act) a power to comment and return the proposed order to the joint
industrial council. Therefore, it is clear that this specific provision is being made, not by a subordinate public body governed by public
law, but by participants in the industry who were empowered to make regulations for themselves and for all others within that
industry who may be competitors and whose interests may not be aligned with the makers of the REA. This is not a grant of a power
to make regulations over a limited area subject to explicit or implicit guidance and review. It is an unlimited grant of power in relation
to employment terms, made to bodies unidentifiable at the time of the passage of the legislation and without intermediate review. On
its surface therefore, this appears to be a facial breach of Article 15.2.1. “Law” is undoubtedly being made for the State, and by
persons other than the Oireachtas. No direct statutory guidance is given for the exercise of the power. On its face, the Act does not
define who might be parties to the agreement, or impose any limitation on the content of such agreement other than that it should
relate to the conditions of employment. Such a far-reaching conferral of law making authority, can only be valid if it can be brought
within the test outlined in Cityview Press. In the context of this case that can only be achieved if the process of registration by the
Labour Court (which is essential to give statutory effect to an employment agreement) introduces sufficient limitation on the
regulation making power granted by the statute to render that regulation no more than the filling in of gaps in a scheme established
by the parent statute.
26 From a structural analysis of the Act , it is firstly significant that the power of approval and registration is itself delegated to an
intermediate body, in this case the Labour Court. The degree of autonomy and discretion afforded to that body and the lack of a
mechanism for appeal or review on the merits, means that any control of the exercise of the regulation making power is necessarily
attenuated. A further noteworthy feature of s. 27 is that the only limitation imposed upon the regulating power is limited, indirect and
negative. The structure of the section is such that registration is mandatory subject only to compliance with the subheadings of s.27.
Section 27(3) provides that on an application being made to the Labour Court, that body “shall, subject to the provisions of this
section, register the agreement”. There is therefore almost a double delegation: first of the power to set the terms; and second of
the power to control those terms by refusing registration. Given the scope of any possible regulation, it is particularly significant that
the power of the Labour Court in this regard is essentially limited and negative.
27 Thus the restrictions imposed by s.27 are critical if the scope of the power conferred by s.27 is to be brought within constitutional
boundaries. However no express guidance is given to the Labour Court as to how it should exercise its powers, and any implicit
guidance to be deduced from the provisions of s.27 is necessarily limited. Section 27(3) provides as follows:
“Where an application is duly made to the Court to register in the register an employment agreement, the Court shall,
subject to the provisions of this section, register the agreement in the register if it is satisfied -
(a) that, in the case of an agreement to which there are two parties only, both parties consent to its registration
and, in the case of an agreement to which there are more than two parties, there is substantial agreement amongst
the parties representing the interests of workers and employers, respectively, that it should be registered.
(b) that the agreement is expressed to apply to all workers of a particular class, type or group and their employers
where the Court is satisfied that it is a normal and desirable practice or that it is expedient to have a separate
agreement for that class, type or group,
(c) that the parties to the agreement are substantially representative of such workers and employers,
(d) that the agreement is not intended to restrict unduly employment generally or the employment of workers of a
particular class, type or group or to ensure or protect the retention in use of inefficient or unduly costly machinery
or methods of working,
(e) that the agreement provides that if a trade dispute occurs between workers to whom the agreement relates
and their employers a strike or lockout shall not take place until the dispute has been submitted for settlement by
negotiation in the manner specified in the agreement, and
(f) that the agreement is in a form suitable for registration.”
Subparagraphs (a), (b), and (f) are essentially procedural and formal in nature and clearly therefore do not provide any principles or
policies to guide the exercise of the power of registration by the Labour Court, still less the power to make the relevant agreement by
the parties in question. While paragraph (e) is a matter going to the substance of the agreement, it provides no limitation on, or
guidance for, the exercise of the power by the regulation making parties. Instead, it imposes a statutory requirement as a condition of
registration. Subparagraph (c) does contain a condition, but it does not go to the substance of the agreement and is not itself
guidance and does not set out any principle or policy for the content of that agreement. While it is directed towards the Labour
Court, it provides no guidance on something which might be considered fundamental, being the representative nature of an agreement
which will have the effect of binding persons who are not parties to it. Accordingly, most attention on this appeal was directed
towards the requirements imposed by s.27(3)(d). There is no doubt that it imposes certain limitations of substance upon the Labour
Court’s power to register an agreement, and therefore can indirectly be said to give guidance to the parties to the agreement as to
the content thereof. The question is however, whether in the light of the extensive scope of the power conferred, such guidance is
sufficient.
28 It is plain however that subs.27(3)(d) is not adequate to provide sufficient limitation on the regulation making power of the parties
to an agreement the registration of which is sought pursuant to s.27, to render that exercise of power compatible with Article 15.2.1.
It was sought to be argued on behalf of the respondents that the word “intended” should not be given its natural and ordinary
meaning and that the Labour Court was empowered to refuse to register an agreement which it considered would have any of the
effects set out in s.27(3)(d). There is no reason however to give a broad and artificial meaning to the phrase for the purposes of
giving greater restrictive capacity to a section which is in its content extremely permissive. While the promotion of employment, and
the avoidance of inefficiency and costly machinery, are laudable and desirable objectives, they do not constitute a sufficient
restriction on an otherwise unlimited power of regulation to bring the power conferred by s.27 within the constitutional limits. In
particular, there is no guidance given in relation to the concept of representativity. There is no obligation on the Labour Court or the
parties to the agreement to consider the interests of those who will be bound by it and who are not parties to it. Furthermore, while
the agreement once registered is binding on everyone in the sector, it may only be varied on the application of the original parties.
29 The provisions of s.28 relating to variation of an REA are a further illustration of what is absent from the scheme. Because the
regulation made by the parties to the agreement is, and retains its character as, an agreement between private parties, s.28 is at
least consistent in limiting the power of variation of the agreement to the original parties thereto. The logic of this position however
ignores the fact that by registration, the agreement becomes part of the law binding on all present and future employers and
employees within the sector. Parties who were not represented at the time of the making of the agreement are nevertheless at risk of
enforcement by prosecution or civil claim, but cannot seek a variation of the agreement. Nor can the Labour Court itself initiate any
process of variation even if that court considered that the agreement had become unduly restrictive of employment or now involves
an inefficient and unduly costly machinery or mechanism of work, or was otherwise generally undesirable. The Oireachtas cannot vary
the provision or direct that variation be considered. It was argued that the Labour Court had power to cancel the agreement and that
non-parties to the agreement could invite the Labour Court to invoke the power of cancellation under s.29(2). It appears the Labour
Court now takes the view that cancellation is not limited to the parties. Even if that is correct, that power is limited , and only
exercisable if the Labour Court considers that there has been “such substantial change” in the circumstances of the trade or business
that it is now undesirable to maintain registration. Even this power of cancellation highlights therefore rather than cures the absence
of a similar power of variation.
30 There can be little doubt therefore that Part III of the 1946 Act raises serious issues of compatibility with Article 15.2.1. What
appears to be law is being made by persons other than the Oireachtas. But this case does not really raise the troublesome questions
of detail and degree that can sometimes arise in this area. There is not here a grant of a limited power to a subordinate body subject
to review as there was for example in the Cityview Press case. Instead there is a wholesale grant, indeed abdication, of lawmaking
power to private persons unidentified and unidentifiable at the time of grant to make law in respect of a broad and important area of
human activity and subject only to a limited power of veto by a subordinate body. In effect, Part III allows the parties to an
employment agreement to make any law they wish in relation to employment so long as the Labour Court considers them to be
substantially representative of workers and employees in the sector, and does not consider the agreement itself to be unduly
restrictive of employment or make provision for unduly costly or inefficient methods of work or machinery, and otherwise complies with
the formal requirements of s.27. No guidance or instruction is given to the Labour Court as to how the matters of representativity or
restriction on employment or inefficiency or costly methods of work, are to be gauged. The process permitted by Part III cannot be
said to be merely the filling in of gaps in a scheme already established by the Oireachtas: in truth the Oireachtas which enacted the
1946 Act could have no idea of even those areas which may be subject to regulation in an employment agreement sought to be
registered under the Act, and no conception still less control of the possible range of regulation that might be made in respect of
each such matter. Nor did the Oireachtas retain any capacity for review either by the Oireachtas or by a member of the Executive
responsible to it, of the agreements actually made. Whatever may be thought of a scheme which permits parties to an agreement to
clothe that agreement with certain legal consequences including the possibility of enforcement by criminal proceedings, once such an
agreement purports to become binding on non-parties pursuant to s.30 of the Act, it passes unmistakably into the field of legislation
which by Article 15 is the sole and exclusive preserve of the Oireachtas. The limited and essentially negative limitations imposed by
s.27(3)(d) are plainly inadequate to bring the exercise of such power within constitutional limits.
31 Finally, it may be worth considering the use of terminology in this area. The term “delegated legislation” is, as a description,
perhaps unexceptional. It has entered our law from the law of the neighbouring jurisdiction. In a constitutional regime where a
parliament is supreme, any provision may be made including presumably, the delegation to others of part of its law making function.
But it is worth recalling however, that in the constitutional dispensation created in 1922 and extended in 1937, the position is
somewhat different. As Hanna J. observed in the Pigs Marketing Board case and as the Gavan Duffy, O’Donoghue, and Lynch
exchange in 1937 illustrates, if in truth any piece of regulation amounted to truly delegated legislation, it would offend Article 15,
since it is plain from the very language thereof, and indeed the constitutional structure, that the function of legislation is one that
cannot be delegated by the Oireachtas to any other body. Indeed the case law since that time can be understood as an attempt to
seek to delineate the boundary between permissible subordinate regulation, and the abdication, whether by delegation or otherwise,
of the lawmaking authority conferred on the Oireachtas by the People, through the Constitution.
32 Accordingly, this appeal will be allowed, and the court will make a declaration that the provisions of Part III of the Industrial
Relations Act 1946 are invalid having regard to the provisions of Article 15.2.1 of the Constitution of Ireland.