Second chamber as a site of legislative intergovernmental relations: An African federation in comparative perspective

ABSTRACT

South Africa, a country that does not recognize itself as a federation, has established a second chamber that is probably ideal for legislative intergovernmental relations. The National Council of Provinces is explicitly mandated to represent provincial interests. This is so both in terms of composition and the authority it enjoys in influencing national legislation. This article argues that the functioning of the second chamber tells a different story. The South African experience reveals that a properly designed second chamber may not deliver the desired result of facilitating legislative intergovernmental relations owing to internal operational rules that do not allow subnational governments to properly consider a bill, formulate a mandate that reflect subnational concerns and instruct their delegation to vote accordingly. Intergovernmental relations, as a result, continues to be the domain of the executive, denying the federation the benefit of an institution for intergovernmental relations that is open to public scrutiny.

KEYWORDS:

• Legislative intergovernmental relations
• second chambers
• federalism
• South Africa
• National Council of Provinces

Introduction

The interaction between different levels of governments is often portrayed as one that is limited to politicians and administrators associated with the executive branches of the respective levels of government. The situation is not any different in African federations. When countries discuss the establishment of intergovernmental forums or the adoption of laws that regulate such interactions, what they typically have in mind are interactions between members of the executive branches of the respective orders of government. The talk of intergovernmental relations [IGR] in Africa ignores other sites of power that are used to facilitate intergovernmental interactions. One such site of IGR that has not been given enough attention in Africa is the legislative arena.

There are different reasons that underscore the need for legislative IGR. In many federations, subnational governments serve as implementers of national laws and policies. It is thus logical that subnational governments participate in the making of laws that they are responsible to implement (Swenden 2010). Given the impracticality of creating neatly defined divisions of powers between national and subnational governments, it is also likely that laws enacted by the national government may encroach on powers allocated to subnational governments. Legislative intergovernmental relations that give subnational governments a say in some or all of the laws passed by national government can help avoid laws that interfere with the powers of subnational governments. The existence of an effective legislative IGR forum might also render litigation over division of powers unnecessary or keep it at a minimum. On a positive note, it also ‘has the potential to bind the nation together, minimize the dangers of fragmented decision–making and encourage common position to be found which are to the benefit of both the nation and its component territories’ (Russell 2001, 109)

In federations with a bicameral parliament, which is the case with most federations, the second chamber is usually designed to serve as the voice of different interest groups. It can, however, serve as a site of legislative IGR in those federations where it is designed as the voice of the constituent governments. On the other hand, there is a view that emphasizes the ‘unfitness of [second chambers], irrespective of their setups and powers, to serve as the voice of subnational units’ (Palermo 2018, E-51). The argument is that a second chamber is not the right place to facilitate legislative IGR. A question that has not been addressed is whether the experience of African federations tells the same story of second chambers that are an unlikely place for legislative IGR.

This article sets to examine the capacity of the second chamber to serve as a site of legislative IGR in the context of one African federation, South Africa. The discussion is not, however, limited to the institutional design of the South African second chamber. It goes beyond the institutional design and examines the functioning of the second chamber to determine whether it actually serves as a site that facilitates interactions between different levels of governments. To examine the actual role that the second chamber plays in facilitating IGR, reliance is made on minutes of the meetings of the second chamber,¹ reports published on the websites of the parliamentary monitoring group, newspaper articles and secondary literature.

The rest of the article is structured in four parts. The next section provides, in a comparative perspective, the elements that, in terms of institutional design, make for a good second chamber that serves as a site of legislative IGR. It then moves to South Africa to examine whether the second chamber serves as a forum for intergovernmental interactions. The article first discusses the constitutional design of the second chamber followed by an examination of its actual functioning. The article completes the discussion by drawing some conclusions.

As a matter of design: What makes a good legislative IGR forum

As mentioned above, a second chamber in a federal system has the capacity to serve as a site of legislative IGR. This does not mean that all second chambers are forums for intergovernmental relations. Lijphart famously categorized bicameral systems into strong and weak bicameralism (Lijphart 2012). His two points of references were the composition and powers of the two houses of a parliament. According to him, the degree of congruence between the two houses of parliament in terms of political composition and the powers they exercise determines whether the second chamber is in a position to advance subnational interests. A federation is deemed to have a strong bicameral system when the second chamber is identical with the lower house in terms of political composition and enjoys co-equal power or is significantly powerful.²

For a second chamber to serve as a site of legislative IGR, the composition of the chamber and, in particular, the method used to select members of the second chambers is important. This means, among other things, that the second chamber must be designed to be composed of representatives of the subnational units. Second chambers like that of Spain that is not composed of the relevant subnational units cannot be expected to serve as a site of IGR (Fessha 2019). Despite the fact that the Constitution regards the Senate as the ‘chamber of territorial representation’, the overwhelming majority members of the latter are elected from the provinces (an administrative unit of the central state) rather than the autonomous communities that are regarded as the custodian of Spain’s quasi federal arrangement and are only represented by one-fifth of the members of the Senate (Guibernau 1995, 239).

Yet the fact that the second chamber is composed of representatives of the subnational entities does not on its own guarantee that it becomes the site of legislative IGR. For example, even the twenty percent of the members of the Spanish Senate that are elected by the autonomous communities vote along party lines rather than the autonomous community they represent (Russell 2001). The same is true in Canada. Although the Senate is constitutionally mandated to represent provincial interests, it rarely serves as the voice of provincial governments. This basically has to do with the way in which members are appointed to the senate. Members are appointed by the Governor General upon the recommendation of the federal Cabinet and can serve until the age of 75(Hogg 1997). Since 2016, the federal Cabinet is assisted by an independent and non-partisan body, known as the Independent Advisory Board for Senate Appointment. The Advisory Board, which includes ‘two members from each of the province or territories where a vacancy is to be filled’, is mandated to provide recommendations to the Prime Minister on senate appointments. Its recommendations are not, however, binding. The ultimate power to make appointment rests with the federal government. Because of this, accountability runs not to the provincial governments but to the federal government that appointed them. As a result, the Senate has not served as a conduit for the interaction between provincial and federal governments let alone used to advance provincial interests. As noted by Meg Russell (2001, 109), ‘since members are appointed from the centre to represent the provinces … the public feel little ownership of their upper house representatives’. The role of the Senate as a house of provincial representatives is simply nominal. The same is true in Australia (Russell 2001).

The composition of the second chamber in Germany, the Bundesrat, seems to be closer to the ideal composition for the purposes of facilitating interaction between the different levels of governments. The Bundesrat is composed of representatives of the Länder, the subnational government as it is known in Germany. It, in fact, consists of members of the Land governments (i.e. Länder ministers), which are appointed by the governments of the Länder. The direct link between the Länd governments and their representatives is evident. This is partly the reason why the second chamber in Germany is referred to by some as an ‘ambassadorial second chamber’ (Palermo and Kössler 2017, 174). That is also the reason why Meg Russell (2001, 108) argues that members of the Bundesrat ‘may be considered to be “indirectly elected”’ although, in her more recent work, she suggests that the Bundesrat is neither elected nor appointed and ‘presents something of a borderline case between election and appointment’ (Russell 2012, 3)

Related to composition, an important condition for a second chamber to serve as an effective site of legislative IGR is the overrepresentation of smaller subnational units . If a second chamber is going to be appealing as a means of intergovernmental collaboration for smaller subnational units, ‘each subnational unit should also have equal or at least disproportionate representation compared to the share it might earn based solely on its population’ (Parker 2015, 28). In USA, Brazil and Australia, where the constituent units are represented equally, irrespective of their population size, the second chambers have the potential to serve as an IGR forum of equal partners. The same is also true with federations, like Germany and Switzerland, which have adopted a weighted representation that nevertheless allows for the overrepresentation of smaller subnational units. In the Bundesrat, according to Article 51(2) of the Basic Law, ‘each Länder has at least three votes, Länder with more than two million inhabitants have four, Länder with more than six million have five, and Länder with more than seven million have six’. Second chambers are less likely to be sought by smaller subnational units as an IGR forum when the constituent units are, more or less, represented based on the respective populations of the subnational units, as it is the case in India, Austria and Belgium . The same is true when smaller subnational units are not meaningfully overrepresented, as it is the case in Canada although, formally speaking, the Canadian Senate provides for equal representation of the four regions. In those federations, smaller subnational units, fearing the domination of larger units, might eschew the second chamber as a forum of intergovernmental relations and seek collaboration through executive-centered intergovernmental structures (Parker 2015).

In addition to composition, the powers that the second chamber is authorized to exercise also crucial. For a second chamber to serve as an effective locus of IGR, it must have significant influence in the federal legislative process, if not enjoy co-equal powers with the lower house. In Spain and Austria, although the second chambers participate in the national legislative process, the last word belongs to the lower houses (Swenden 2010). That makes the second chambers an ineffective forum for IGR. The Senate in USA (Dinan 2006, 316), Canada (Knopff and Sayers 2005, 103) and Australia (Stone 2006, 529) enjoy co-equal power with the lower house. A bill cannot become a law unless it secures the support of both houses of parliament. Although these second chambers have the potential to become an effective site of legislative IGR, the potential is again not realized, as indicated earlier, because of the absence of direct link between the senators and their respective subnational governments.

As noted by Stepan (1999), the second chamber of Germany is less powerful than its USA or Australian counterpart. The Bundesrat does not enjoy co-equal power with the lower house. It ‘cannot participate in the two most important legislative votes, those for government formation and government termination’. (Stepan 1999, 26) It only enjoys suspensive veto if the bill under consideration does not affect the power of the Länder. For the purposes of legislative IGR, however, it is not necessary for the second chamber to have influence on all legislation. It suffices if it has significant influence on bills that affect the powers and functions of subnational governments (Swenden 2010). And with respect to those bills that affect the powers of the Länder, the Bundesrat enjoys absolute veto power. Its objections cannot be overridden by the lower house. Furthermore, members of the Bundesrat that represent the Länder do not enjoy independent vote. Although not legally obliged, they are expected to vote as a block based on the instructions they receive from the Länder.³ This makes the Bundesrat an ideal site for legislative IGR as long as it decides to use the chamber to advance the interests of the Länder governments (Russell 2012).

In Africa, the capacity of second chambers to serve as a forum of intergovernmental relations is also affected by their composition and the powers they exercise. In Nigeria, the oldest federation in Africa, the constitution gives the senate as much legislative scope as it gives to the lower house (Osieke 2006). A bill has to secure the blessing of both houses before it becomes a law. In addition, the states, irrespective of their population size, are represented equally by three senators with the exception of Abuja, the federal capital, which is only represented by a single senator. Yet the fact that the Senate is a directly elected body means that there is no direct link between the senate and the state governments. Perhaps that is also why ‘the disposition, especially on the part of senators, has been more toward being federal statesman rather than state representatives’ (Osaghae 2015, 284). The other quasi-federal African country that has a directly elected second chamber, Kenya, explicitly links the senators with the counties by requiring the latter to ‘represent the counties’ and ‘protect the interests of their county governments’ (The Constitution of Kenya, Article 96(1)). It seeks to further entrench the link by requiring the senators from the same county to cast a single vote when it comes to bills that affect counties. This, however, does not force the senator to vote based on the wishes of the county governments as voting may still take along party lines. As a result, the Senate does not serve as a forum of legislative IGR. Perhaps, in so far as African federations are concerned, it is the second chamber of the Ethiopian federation that is the most unlikely place for legislative intergovernmental relations. Formally speaking, it has no link with the state governments as it is composed of representatives of ethnic communities (Bihonegn 2015). More importantly, it plays little or no role in the legislative process. Let alone override the lower house, it does not even have the power ‘to provide inputs or recommend the reconsideration of a bill adopted by the latter before it becomes a law’ (Fessha 2016).

From the foregoing, it is clear that the capacity of second chambers, both in Africa and beyond, to serve as a site of IGR is a function of the manner in which members are selected and the influence they enjoy over laws that affect the powers and functions of subnational governments. However, it is important to note that a good constitutional design does not necessarily guarantee the desired result. A strong link with subnational governments and strong powers over legislation that at least affect the powers of subnational governments may not translate into a second chamber that serves as effective intergovernmental forum. Other political interests may dictate the manner in which members of second chamber exercise their powers. Members may use their strong powers in the second chambers to advance the agenda and interest of their political parties, which may or may not align with the interests of their respective subnational governments. As mentioned earlier, Lijphart (2012) has, for example, indicated that the degree of congruence between the two houses of parliament in terms of political composition is an equally important factor in evaluating the effectiveness of a second chamber. In a system of bicameral parliament, where the second chamber is identical with the lower house in terms of political composition, the latter can hardly be expected to serve as a site of legislative intergovernmental relations. This is irrespective of whether the second chamber enjoys co-equal power with the lower house.

With this brief comparative observation in background, the article now moves to its main business and discuss the role of the second chamber in South Africa as a site of legislative IGR. The question is whether the second chamber is designed as a forum for legislative IGR and whether it actually serves as such.

Designed as a forum for legislative IGR?

The 1996 Constitution of the Republic of South Africa [Hereinafter South Africa Constitution] has established three levels of government. Its nine provincial governments and 278 local governments enjoy original constitutional powers. Legislative and executive powers are divided among the three levels of governments. None of the governments can unilaterally alter the division of powers entrenched in the constitution. The power to settle intergovernmental disputes between the national and provincial governments is reserved to the Constitutional Court while disputes between local and other levels of government can be adjudicated by other courts. The constitution also provides for shared rule by establishing a second chamber through which the provinces are brought to the national decision making process.

Yet, South Africa does not describe itself as a federal state. It has, however, established a second chamber that is the ideal arena for legislative IGR. The National Council of Provinces [NCOP], the second chamber as it is known in South Africa, is explicitly mandated to represent provincial interests. It is also designed as a house of provinces. This is so both in terms of the composition and the authority it enjoys in influencing national legislation.

In slight departure from the Bundesrat after which it is modelled, provinces are represented equally with each province allowing to send a delegation of ten people.⁴ That gives each province equal influence in the negotiation between the two orders of government. Each provincial delegation is composed of six permanent delegates and four special delegates, headed by the Premier of the province. The six permanent delegates are the permanent representatives of the provincial government in Cape Town, where the NCOP is located. The special delegates, which include the Premier of the province or a representative, are basically temporary delegates that will change as and when required by the province. They are appointed from time to time by the provincial legislature with the concurrence of the Premier and the leader of the parties entitled to special delegates in the province’s delegation. They are selected based on the nature of the bill before the NCOP and their expertise.

In another departure from the German model, the provincial delegation to the NCOP is drawn not only from the provincial executive but also both from the provincial legislature and executive. The four special delegates include members of the provincial legislature, headed by the Premier of the province or his or her nominee, while the six permanent delegates can be drawn from outside the provincial legislature.⁵ Unlike in the case of Germany, where the interaction is between subnational government and federal government that is represented through its legislative branch, the interaction in South Africa involves federal legislatures, on the one hand, and subnational legislatures and executives, on the other. Furthermore, the provincial delegation must reflect the diversity of political parties represented in the provincial legislature.⁶ In short, the NCOP is a second chamber that is composed of ‘multi-party delegations from each province drawn from the provincial executive and legislature’ (Powell 2015, 316).

More importantly, on matters that affect provinces, the provincial delegation, although required to reflect the diversity of political parties represented in the provincial legislature and include members of provincial legislature and provincial executive, is expected to act as a coherent unit of a provincial government. Members of the delegation do not enjoy individual autonomy and are expected to vote in bloc based on the instructions they receive from their respective provincial legislatures (South African Constitution, Section 65). In other words, there is only a single delegation vote, allowing the province to speak in unison. This helps the provincial delegation to present a unified front despite the fact that the delegates might belong to different political formations and different branches of a provincial government. This means that when the delegates are voting on matters affecting provinces, they are not casting an independent vote but single vote, acting on behalf of the provincial government, making the second chamber a forum for legislative IGR.

Like the case of Germany, the position of the NCOP as a site of legislative IGR is further concretized with the power that it enjoys over the legislative process. Unlike some second chamber in federal system, the role of the NCOP in the national legislative process is not limited only to bills that affect provinces. It also considers bills that do not affect provinces, the so-called section 75 bills. It, however, enjoys stronger powers over bills that affect provinces, commonly known as section 76 bills.⁷ A bill that comes through the National Assembly and affects the provinces needs the approval of the second chamber before it becomes a law (Steytler 2017, 360).⁸ The National Assembly can adopt a section 76 bill that is rejected by the NCOP only if a mediation committee that is composed of equal members of both houses fails to come up with a bill that is acceptable by both houses and if it is able to secure the support of its two third of its members.⁹ It is, however, ‘unlikely that a party or coalition that cannot acquire the approval of five provinces in the NCOP could assemble a two-thirds majority in the National Assembly’ (Murray 2006, 268) in which case the bill lapses. It must be added that the NCOP, although it rarely uses it, has also the power to initiate legislation on matters that are listed in Schedule 4 of the Constitution (The Constitution of South Africa, 1996, Section 68 (b)). It may not, however, initiate money bills.

The foregoing suggests that the NCOP, compared to some of the federations discussed above, has a better capacity to serve as a site of legislative IGR. This largely has to do not only with the manner in which it is composed but also the authority it enjoys with respect to national legislation. To begin with, a direct link between the second chamber and the subnational government is created as the former is composed of the representatives of the latter. It is members of provincial legislature and executive that sits in the NCOP representing their respective provincial government. This obviously creates a direct link between provincial governments and the NCOP, making the latter a forum for negotiation between the two orders of government. This link is further concretized by the fact that members of the NCOP vote based on the instruction they receive from their respective provincial governments. Added to this is the stronger position that the NCOP finds itself when it comes to bills that affect provincial governments and federalism. In short, the NCOP do not only provide the forum for legislative IGR but also provides participants of IGR real influence on matters of mutual interest. The position of the NCOP as the site of IGR seems to be also appreciated by the government. Every year, the President of the Republic ‘delivers a special annual address on IGR to the NCOP’ (Powell 2015, 318).

Functioning as a forum of legislative IGR?

Now that we have established the capacity of the NCOP to serve as a site of legislative IGR, the remaining question is whether the NCOP has, in fact, served as effective forum of legislative IGR. As mentioned earlier, an important element that makes the NCOP a potential site of legislative IGR is that those representing the provinces are required to act based on the mandate they receive from the provincial legislature. This raises the question whether the provincial legislature have managed to duly consider the proposals before the NCOP, manage to articulate a provincial position on the matter and instruct the provincial delegation accordingly. Let’s begin with the first question of whether they have managed to duly consider the proposals before the NCOP.

The time element

Normally, the NCOP works with a six-week legislative cycles.¹⁰ The way it works is that the NCOP, after being briefed by the Department that initiates the bill, refers the draft legislation automatically to the provincial legislatures. The provincial process commences with the relevant provincial portfolio committee considering the bill and conducting public hearing to obtain public input on the bill. It might also require consulting the relevant provincial minister and holding discussions with local government. The committee is then expected to take a provincial position on the bill and formulate a negotiating mandate based on which the provincial delegation in the NCOP must negotiate on the bill.¹¹ It is at this point that the four special delegates, which, as mentioned earlier, are supposed to be chosen according to their expertise and knowledge of the Bills being debated before the NCOP, join the six permanent delegates to represent the province and participate in the negotiation on the bill. The provincial delegation is then expected to brief the provincial legislature, sitting in plenary on the outcomes of the negotiation after which the latter must formulate a Final mandate. The provincial delegation then casts the provincial vote in the relevant select NCOP Committee based on the Final mandate.¹² Unless an issue that necessitates consultation with the provincial legislature arises from the deliberations in the select committee, the Final Mandate can also be considered as the Voting mandate,¹³ based on which the head of the provincial delegation, or another member of the provincial delegation appointed by the head, casts a single vote on behalf of the province in the NCOP’s plenary.¹⁴

From the foregoing, it is clear that this is a process that involves public hearing and consultation with relevant stakeholders. It demands provincial legislatures to carefully study the bill, identify the implications of the bill for their respective provinces and ensure that provincial needs are taken into account. This means that adequate information and time must be provided to provincial legislatures to deliberate on the bill and formulate a mandate. As noted by the Independent Panel,

[i]f section 76 Bills are to be properly considered before they are passed, the NCOP should have a programme which ensures that provincial decision-makers have adequate information about the Bills to make informed decisions about them. Furthermore, adequate time must be allowed for discussions involving both Ministers of Executive Councils and Members of Provincial Legislatures in each province, so that the final Bill addresses the particular needs of each province. (Parliament of the Republic of South Africa 2014)

Unfortunately, provincial legislatures are not given adequate time to formulate a voting mandate. To begin with, the NCOP’s six-week legislative cycle in which the bills have to be passed is simply not sufficient. This becomes evident when one compares the parliamentary days that Parliament takes to adopt a Section 76 Bill versus a section 75 bill. According to a report by a Parliamentary Monitoring Group, it takes Parliament 73 parliamentary days to adopt a Bill while it ‘takes only three additional parliamentary days to adopt a Section 76 Bill’ (Parliamentary Monitoring Group 2018). This is despite the fact that a section 76 Bill must be processed by both houses of parliament and the provincial legislatures that must conduct public hearing and consultation while a section 75 bill must only be processes by both houses of parliament.¹⁵ The Constitutional Court has also expressed ‘serious doubts’ on the sufficiency of the six-week legislative cycle (Land Access Movement of South Africa and Others v Chairperson of the National Council of Provinces and Others 2016).

Further, the short legislative cycle is not also always adhered to and, as a result, it is not uncommon for bills to be fast tracked. For instance, the Constitutional Court, with respect to a legislation that sought to amend the land rights act, observed the ‘truncated timeline’ that the NCOP adopted:

from start to finish, the provinces had less than one calendar month to process fully a complex piece of legislation with profound social, economic and legal consequences for the public. The timeline gave the provinces a mere three to five calendar days to notify the public of the hearings, from the date the Provincial Legislatures were briefed until the date the public hearings commenced. The provinces had only eight calendar days to conduct the hearings, consider public comments and confer appropriate negotiating mandates, from the start of the hearings until the negotiating mandate meeting. (Land Access Movement of South Africa and Others v Chairperson of the National Council of Provinces and Others 2016)

There were a number of other cases where bills were rushed through the NCOP that amendments proposed by provincial legislatures were not considered. In the most extreme cases, provinces were expected to formulate a mandate within a few hours after receiving a notice (Parliament of the Republic of South Africa 2014). The provinces complain of ‘little thought for provincial responsibilities in NCOP scheduling’, including ‘insufficient provision of details on the scheduling of meeting by the NCOP’ (Francis 2011, 269). These means that provincial legislatures are often faced with a deadline for a provincial voting mandates that they can hardly meet. Under such circumstances, a provincial legislature can hardly formulate a well-informed mandate. It has often resulted either in ‘delay or the inability [of the provincial legislature] to be properly prepared in the preparation of the voting mandates and for meetings’ (Francis 2011).

It is not surprising that the NCOP, on a number of occasions, has been chided by the Constitutional Court for ramming through legislations within the framework of ‘inherently unreasonable timeline’ without adequate public participation. In 2016, the Constitutional Court invalidated the Restitution of Land Rights Amendment Act, mentioned above, on the ground that the NCOP adopted the law without adequate public participation, which, the Court believed, was largely attributed to the ‘truncated timeframes that provincial legislatures were given’ (Land Access Movement of South Africa and Others v Chairperson of the National Council of Provinces and Others, 77; Pretorius 2016). A mere two weeks was given to provincial legislatures to advertise and hold public hearings, consider oral and written submissions from the public, and formulate negotiating and final mandates. This was not the only bill for which the NCOP is criticized for not taking the time to engage the public.

A mandate that represents the wishes of provincial legislature

Another issue that can be raised with regard to the mandate that delegates receive from their respective provincial legislature is whether the mandate represents the wishes of their respective provincial legislatures. This, of course, depends on how the mandate is given and, in particular, on the majority that is required to endorse a mandate. Unlike the uniform procedure that the South Africa parliament has provided with respect to the way provincial legislatures confer mandate on their respective delegation, there is no equivalent national law regulating the vote that is required to endorse a mandate. It varies from one province to another. The provinces are not also consistent in the procedure they use to formulate a provincial mandate.

In the small but economically strong province of Gauteng, for example, the vote that is required to formulate a mandate depends on the nature of the Bill (Maloka 2000, 107–118). Once the NCOP refers a bill to the provincial legislature, the bill is directed to the proceeding committee that classifies bills as either important, ordinary or technical. The group to which a bill is classified into determines the vote that is required to endorse a mandate. Accordingly, only bills that are classified as important will be debated at the Provincial Legislature’s Plenary and require the support of majority members of the legislature before a mandate is conferred on the provincial delegation. The procedure is not strict with bills that are classified as Ordinary. These bills do not necessarily have to be considered in the province’s plenary. If the house is not in session, the decision of the relevant committee is distributed by the Speaker to all members. In the absence of at least four written objections, the decisions of the Committee will be considered as the decision of the House based on which a mandate will be conferred on the provincial delegation. In KwaZulu Natal, the formulation of mandates is done by the National Council of Provinces Standing Committee, which ‘coordinates provincial voting mandates’ (Francis 2011). Based on the reports it receives from the relevant portfolio committees, the Committee approves the negotiating and the final mandate that is conferred on the provincial delegation. It takes decisions by a vote. The same approach seems to be, more or less, followed by some of the other provinces. A brief examination of some of the mandate conferral letters submitted to the NCOP by Limpopo, Eastern Cape, Mpumalanga, Free State and KwaZulu-Natal indicate that it is the relevant portfolio committees that often formulate and confer voting mandates on provincial delegations. The provinces where a provincial plenary is involved in conferring mandates are the Western Cape, Northern Cape and North west.

From the foregoing, it is clear that the vote and at times even the process that is required to formulate both a negotiating and voting mandate is not that complicated. There is not a single provincial legislature that requires more than a simple majority to endorse a mandate. In fact, many of the provinces, for at least with respect to certain bills, do not even bother to table the bill before a provincial plenary.¹⁶ The decision on and formulation of negotiating and voting mandate is left to the relevant committees. This type of procedure that relies on a simple majority to endorse a mandate can easily result in the alienation of small parties. The wishes and interests of small parties can be easily ignored.

The effect of the simple majority that is required to endorse a mandate on the unity of the provincial delegation wasevident, at least, once when a member of the delegation that does not belong to the ruling party of a province expressed her reservation on a mandate that is already submitted by her provincial legislature. In June 2018, when the NCOP was processing the Traditional Leadership and Governance Framework Amendment Bill, a member of the Gauteng delegation that belongs to the Democratic Alliance, the official opposition party in the province, expressed unwillingness to present the negotiating mandate from the province. The member argued that ‘it is an irregular mandate’ and is seeking legal opinion on the matter (Parliamentary Group 2018). This was rejected by the committee on the ground that it has received the negotiating mandate from the provincial legislature, which indicates that the provincial legislature has accepted the bill with few proposed amendments. The Chairperson of the Committee insisted that ‘the Committee’s approach should be based on formal submissions they get from the side of the [provincial] Legislature’. It is clear that the provincial representative was trying to win a battle that her party probably lost in the provincial legislature.

In short, there is a danger that the simple majority requirement to endorse a mandate might frustrate members of a provincial delegation that are representing smaller parties.¹⁷ Under such circumstances, it is very difficult to claim that a provincial delegation speaks for a province. Perhaps, a qualified majority would help a province to make decisions based on provincial interest and present a truly unified face in the national legislative process.

Mandates that reflect provincial concerns

A more important question is whether mandates given by provincial legislatures reflect substantive provincial concerns. This is important because a second chamber can only be deemed to serve as effective forum of legislative IGR if it is used to inject the needs of subnational units into the national legislative process. A report that documents the performance of the parliament from 2006 to 2017 reveals that, out of the 391 Bills that were adopted by parliament,¹⁸ only 23 percent of them were classified and processed as bills affecting provinces. During the same period, the Parliament processed about seven bills amending the Constitution that all, except one, were classified as amendment that affect provinces and required the blessing of the NCOP. The remaining, about 70% of them, were ordinary bills that do not affect provinces.¹⁹

Despite the fact that the NCOP has dealt with more section 75 bills, it has only proposed few amendments with respect to such bills. In 2018, for example, the section 75 bills that the Parliament adopted were approved by the NCOP without amendments.²⁰ In 2017, of the twelve section 75 bills that were approved by the NCOP, only two were accompanied by some amendments.²¹ In 2016, NCOP proposed amendment only to the Refugee Amendment Bill despite the fact that it has adopted close to 14 section 75 Bills (Parliament of the Republic of South Africa 2017). The situation was not different in 2015. More importantly, in the rare cases where the NCOP proposed amendments, the amendments were largely inconsequential and only served to fine tune the bills as they were often grammatical or technical in nature.

The cursory examination of section 75 bills by the NCOP is not, of course, new to the current parliament. Earlier reports have also confirmed that ‘the NCOP’s section 75 legislative role was mostly very superficial’.²² This is also evident from the fact that the NCOP often deliberates on and adopts a number of section 75 bill in a single meeting, sometimes up to four bills within a single session. This is often justified by the fact that the NCOP does not just have the manpower to effectively scrutinize all section 75 bills that are forwarded by the National Assembly. ‘The Permanent delegates served on up to eight committees at a time compared to [the National Assembly] where a member served on only two committees’ (The Ministry and Department of Provincial and Local Government, 1999, 95). This means that bills processed by different select committees of the National Assembly often end up in a single committee of the NCOP.

As mentioned earlier, the Parliament has also processed bills amending the Constitution. As the majority of the constitutional amendments involved the demarcation of provincial boundaries, the involvement of the provinces through the NCOP was deemed necessary. According to the Constitution, a constitutional amendment that involves changes in provincial boundaries must be approved by six out of the nine provinces in the NCOP; it must also be supported by the particular provinces that are affected by the demarcation of provincial boundaries (Section 74(3b) (ii), South Africa Constitution). The Bills were referred to the provinces for ‘further consultation’. And the provinces were actively involved in the process of adopting the amendments. Most of them, in particular those affected by the provincial demarcation, held public hearings based on which they instructed their respective delegation to support the bills. Part of the Twelfth Constitutional Amendment that transferred Matatiele from KwaZulu Natal to Eastern Cape was, however, declared constitutionally invalid on the ground that the province of KwaZulu Natal did not comply with the public participation requirements of the Constitution (Matatiele Municipality and Others v President of the Republic of South Africa and Others 2007). The Court, however, suspended its order of invalidity. That allowed the Parliamentto rectify the procedural defect, reintroduce and pass the bill in the form of the Thirteenth Constitutional Amendment. The Provinces have not, however, proposed any changes to the constitutional amendment bills. It was clear, however, that the demarcations were, at the end, effected with the active involvement of the provinces. Although they have not countered or proposed change to the constitutional amendment bills proposed by the National Assembly, they have been able to engage in a process that allowed the provincial legislatures to formulate a mandate that is preceded by a public consultation process, although that sometimes required a court order.

Yet a decision on whether the NCOP has served as effective site of legislative IGR where provincial needs are injected into the national legislative process cannot be based on the performance of the latter with respect to section 75 bills and bills that amend the Constitution. It depends on its performance with respect to the so-called section 76 legislation, legislations that affect provinces. As mentioned earlier, only 23 percent of the bills that were processed between 2006 and 2017 were classified and processed as bills affecting provinces. The NCOP has proposed amendments to some of those bills, suggesting a more serious preoccupation of the NCOP with bills that affect provinces. At first sight, this might suggest that, with respect to these bills, the NCOP has proved to be an effective legislative IGR forum. This would, however, only be the case if the amendments represent substantive provincial concerns.

In order to determine whether the amendment proposed by the NCOP represent provincial concerns, I examined all section 76 bills that were considered by the NCOP during South Africàs fifth parliament. The fifth parliament was elected in the general election of 7 May 2014. Its term was expected to end on 21 May 2019. The Parliament, however, agreed to dissolve itself on 6 May 2019 in order to allow for election to be held on 8 May 2019. The report of the Independent Panel is relied upon to provide a general overview of the performance of the NCOP with respect to section 76 Bills that were considered prior to 2014.

The examination of the section 76 bills that were considered by the NCOP reveals a similar trend. It reveals that the amendments proposed by the NCOP are not limited to grammar and technical errors. Some of the amendments affect the substance of the proposed bills. The problem is that even those amendment do not reflect provincial concerns. This is not to suggest that the amendments are unnecessary or useless. But, with few exceptions as in the case of the annual Division of Revenue Bill, they do not show a house of parliament that understand the unique role it is expected to play in the legislative process, a role that is distinct from that of the lower house. For instance, the minutes of NCOP meeting on the adoption of the National Health Laboratory Services Amendment Bill (2017), a bill that affects provinces, indicates active engagement of the provinces through their provincial delegation.²³ The provincial negotiating mandates submitted to the NCOP expressed a number of concerns with the bill and proposed some amendments. However, the concerns that the provinces expressed and the amendments they proposed were not different from the kind of amendments that members of the lower house would propose.²⁴ The mandate from Eastern Cape, for instance, suggested that ‘a board member must not serve more than two terms’ to avoid ‘having lifetime board members’. Limpopo proposed, among other things, that ‘each discipline of laboratory medicine should be represented in the board so that it is constituted by a variety of expertise in pathological services’. Mpumalanga proposed amendments to to the composition of the board. North West wanted the bill to be ‘explicit on the development of the career path of young people who want careers in medical and laboratory services’. The only exception was KwaZulu Natal that expressed provincial concerns by requesting the reconsideration of the removal of provincial representatives in the board as that might make it difficult for provincial concerns to reach the board. The other provinces supported the bill without proposing amendment.

The same is evident from the minutes of the NCOP’s deliberation on the adoption of the Traditional Leadership and Governance Framework Amendment Bills that was adopted in 2018.²⁵ Five provinces proposed amendments to the bill. Eastern Cape expressed concern about the ‘time period allowed to establish Kings and Queens councils’ and proposed for it to be reduced from two years to one year. KwaZulu Natal expressed concern with the use of terminology and proposed ‘that the term “traditional council” be deleted and replaced by the term “traditional authority”’. Limpopo expressed concern about traditional councils that operate ‘outside the law’ and legal costs incurred by traditional leaders. Mpumalanga proposed that the bill defines some terms including ‘principal traditional community’. North West rejected the bill claiming the lack of adequate consultation with Kings and queenships on the Bill. The proposed amendments obviously do not reflect provincial concerns. The same can be observed in the deliberations of the Traditional and Khoisan Leadership Bill and the Public Service Commission Amendment Bill, section 76 bills, that were adopted in the same year.²⁶

It is evident that the provincial focus that should form the basis of the concerns that members of the NCOP express and the amendment they proposes to the bills that affect provinces is visibly absent. It is also not limited to the fifth parliament. In its 2014 report, the Independent Panel that assessed parliament recommended that the mandate that the provincial legislatures formulate ‘must reflect substantive provincial concerns’ (Parliament of the Republic of South Africa 2014, 27). It is only then that the NCOP can fulfil its role as a site of IGR. A mandate from the provincial legislatures that simply provides an input that aims at improving legislation without a special focus on the need of provinces does not represent an effective use of the second chamber as a site of IGR. This failure of the NCOP to focus on its unique role is not limited to its law-making role but also extends to the nature of the debates that the house holds both in the plenary debates and the work of the committees. As noted by the Independent Panel,

the topics for debate do not reflect the NCOP’s unique mandate to serve as a forum for the discussion of issues affecting the provinces. While most subjects chosen for debate were certainly of interest from the point of view of national debate, in several cases the topics bore no clear link to provincial interests. (Parliament of the Republic of South Africa 2014, 51)²⁷

There are different explanations as to why the NCOP has not fulfilled its role as a site of legislative IGR. The fact that most bills that affect provinces are first introduced in the National Assembly than the lower house might have made it not easy for members of the NCOP to meaningfully interact with bills. The argument is that once ‘the senior house’ passes a bill and forwarded it to the NCOP, the latter is under ‘political pressure’ to pass the Bill (Powell 2015). It is also argued that provincial legislatures do not bother to provide substantive input on the ground that provincial concerns are taken care of in the executive intergovernmental structure. They assume that the executive intergovernmental structure that brings together the relevant national minister and the corresponding provincial ministers is used by the latter to inject provincial interest and the needs of the provinces into the proposed national bill. They, as a result, rarely engage effectively with national bills before the NCOP (Murray and Nijzink 2002, 49; Powell 2015, 316). This, however, points to another major problem that undermines the effectiveness of the NCOP as intergovernmental forum, namely the dominance of executive IGR. ‘Long before bills are tabled for consideration before the NCOP, provincial concerns have already been communicated by the relevant provincial officials in the different intergovernmental structures’ (Fessha 2016, 129).

Yet, the role of executive intergovernmental structures as a site of interaction among equal partners should not be exaggerated. It must be noted that even the executive intergovernmental structures may not be a forum for negotiation but rather a forum through which decisions made by the centre are communicated to the provinces. The ANC, with the exception of the Western Cape, controls all provincial governments. The dominance of ANC also means that provincial politicians are often willing to toe the lines of national departments and engage in less or no scrutiny of bills initiated by the later and is before the NCOP for further approval. This is especially the case in relevant provincial committees that are headed by a member of the provincial legislature that belongs to the ANC (Francis 2011).

Conclusion

From the foregoing, it is clear that the second chamber in South Africa is in a good position to truly represent and work for subnational units. In so far as design is concerned, it seems to be properly positioned to serve as a forum for legislative IGR. NCOP enjoys significant influence in the national legislative processes. Although the level of influence it has varies depending on the nature of the bill before it, it enjoys real influence on legislation that affect the powers and functions of provincial governments. More importantly, it has a direct link with provincial governments.

As the experience of South Africa shows, however, a good design does not guarantee that a second chamber will serve as a forum that facilitates interaction between the different levels of government. This depends on a host of factors. A properly designed second chamber may not deliver the desired result owing to internal operational rules of the second chamber that do not give enough time to subnational governments to properly consider a bill, formulate a mandate that reflect subnational concerns and instruct their delegation to vote accordingly. The nature of politics that characterizes second chambers may also go a long way in explaining why second chambers may not really serve as a forum for IGR. When the same partisan politics that dominates lower house also characterizes the debate in second chambers, members of the latter cannot but be expected to be driven by ‘the political over the territorial element’. That is why the issues they raise and the amendments they propose do not reflect subnational concerns. The domination of national partisan politics in the second chambers could also be because of party politics and tight party discipline as is the case in South Africa.

The experience of South Africa confirms the experience elsewhere that second chambers, ‘irrespective of their set ups and powers’, are not ‘serv[ing] as the voice of subnational units’ (Palermo 2018, E-51). The functioning of the NCOP reveals little or no legislative IGR. This also means that IGR continue to be the domain of the executive. This is unfortunate as it denies the South African federation the benefit of an intergovernmental relation that is transparent and open to public scrutiny.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Funding

This project has received funding from the European Union's Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement 752098.

Notes

1 In South Africa, the Parliamentary Monitoring Group keeps a comprehensive account of the meetings of the South African Parliament, both plenary and committee meetings. The minutes are made available on the website of the organization, both audio and written.

2 Swenden (2010) observes that weak bicameral legislatures are most common in federations that have adopted a parliamentary system of government while strong bicameral legislatures are largely located in presidential system of government. The only two exception, according to him, are the second chambers in Australia, Germany and Switzerland that are strong but operate within parliamentary federations.

3 According to the Federal Constitutional Court in Germany, in the event that representatives of a Land do not agree on voting ‘yes’ or ‘no’ to a bill, the votes of that Land are rendered invalid (2 BvF 1/02 (2002) (Ger)). It is also interesting to note that the seating arrangement in the Bundesrat is not based on party affiliation but the state that the members represent (Russell 2011).

4 In addition to the representation of each province by a ten-member delegation, included in the NCOP is the non-voting ten-member delegation that represents organized local government.

5 If a member of a provincial legislature is appointed as a permanent delegate, that person vacates the seat in the provincial legislature (section 62(2), South Africa Constitution).

6 Each political party that has a seat in the provincial legislature is entitled to one permanent delegate provided that the party has enough seats in the provincial legislature to warrant its inclusion in the NCOP delegation. The formula provided in schedule 3 of the Constitution and the Determination of Delegates (NCOP) Act 69 of 1998 provides the guidelines to determine the parties that are entitled to ‘delegates in the provincial delegation’ and how many each is entitled.

7 The power of the NCOP extends to constitutional amendments that affect provincial boundaries, institutions, powers and functions.

8 The Constitutional Court has again and again indicated that it is not only bills that affect the authority of provincial government that are regarded as section 76 bills but ‘any Bill whose provisions substantially affect the interests of provinces’ (Tongoane v. Minister of Agriculture and Land affairs, para 72). This gives the NCOP broader involvement in the national legislative process. As noted by Steytler (2017, 360), the Court ‘rejected the argument provincial interests are synonymous with provincial competences’.

9 In matters that do not affect provinces, the National Assembly can override an objection from the NCOP by a simple majority. In those cases, members enjoy an independent vote and often vote along political party lines.

10 NCOP’s Rule 240 provides:

1. All section 76 or 74(1), (2) and (3) Bills should be dealt with in a manner that will ensure that provinces have sufficient time to consider the Bill and confer mandates.

2. Depending on the substance of the Bill, the period may not exceed six weeks.

3. In the event that the substance of the Bill requires sufficient time beyond the six-week period, the cycle may be extended with the approval of the Chairperson of the Council.

11 The Constitution, under section 65(2), envisages the enactment of an Act of Parliament that ‘provide[s] for a uniform procedure in terms of which provincial legislatures confer authority on their delegations to cast votes on their behalf’. This was realized in 2008 when the Parliament enacted the Mandating Procedures of Provinces Act 52 of 2008 [Hereinafter Mandating Act]. According to the Act,

“negotiating mandate” means the conferral of authority by a committee designated by a provincial legislature on its provincial delegation to the NCOP of parameters for negotiation when the relevant NCOP select committee considers a Bill after tabling and before consideration of final mandates and may include proposed amendments to the Bill.

12 According to the Mandating Act, final mandates refer to the conferral of ‘authority on [a] provincial delegation to the NCOP to cast a vote when the relevant NCOP select committee considers a Bill prior to voting thereon in an NCOP plenary’.

13 The Mandating Act defines voting mandate as ‘the conferral of authority by a provincial legislature on the head of its provincial delegation to the NCOP to cast a vote on a Bill in an NCOP plenary’.

14 The provincial legislatures are, of course free to change their mind and formulate a voting mandate that departs from the Final Mandate. The Mandating Act includes a number of forms that must be used by provincial legislatures when conferring a mandate. The Act clearly states that the conferral of Final mandate must be drafted in the letterhead of the provincial legislatures, indicate whether the provincial legislature votes in favour, against or abstains from voting on the Bill, and amendments, if any, and include the signature of the Speaker of the provincial legislature. The prescribed form, which must be directed to the Chairperson of the NCOP, is included in schedule 2.

15 There have been calls to extend the six-week legislative week to eight-week legislative cycle with respect to bills that affect provinces. To date, the NCOP rules have not been amended to extend the legislation cycle (Polity 2018).

16 It is not also clear if this practice of conferring a mandate without involving the provincial plenary is consistent with the Constitution. According to the Mandating Act, a negotiating mandate is conferred by a committee designated by a provincial legislature while the ‘Final mandate’ must be conferred by a provincial legislature. Section 65 of the Constitution seems to also envisage the conferral of mandate by a provincial legislature. For more on this, see also Murray and Nijzink (2002, 49).

17 An earlier NCOP report on the NCOP Needs Assessment has already noted that permanent delegates from minority parties ‘feels isolated and say that provincial committees make not attempt to use them’ (Murray and Nijzink 2002, 49).

18 A 2018 report prepared by Parliamentary Monitoring Group indicates that, between January 2006 and until December 2017, 486 Bills were processed.

391 of the 486 Bills were passed, 13 lapsed, 5 were rejected, 32 were withdrawn and 45 Bills were still under consideration at the time of writing. For the period under review, the efficiency percentage is 80% (i.e. 80% of all Bills introduced were successfully adopted by Parliament). See Parliamentary Monitoring Group (2018)

19 It is important to note that the tagging of bills, which is the classification of bills as section 76 and section 75 bills, is crucial as it affects the opportunity of the provincial governments to properly scrutinize bills that affect provinces. The classification is done by the Joint Tagging Mechanism (JTM) that involves equal number of members from both houses. A bill that is incorrectly classified as section 75 bill denies the provinces the influence they should have on a bill that affects the interest of provinces. This has happened on a number of occasions. It has also been brought before a court, which ruled that the incorrect classification of the bill renders the adopted law invalid. See Tongoane v. Minister of Agriculture and Land affairs (2010).

20 See, for example, NCOP Finance (2018), Public Audit Amendment Bill- Briefing and Finalization. Available at https://pmg.org.za/committee-meeting/26696/ (accessed 18 February 2019); NCOP Economic and Business Development (2018), Competition Amendment Bill. Available at https://pmg.org.za/committee-meeting/27609/ (accessed 18 February 2019); NCOP Social Services (2018), Electoral Laws Amendment Bill. Available at https://pmg.org.za/committee-meeting/27756/ (accessed 18 February 2019).

21 See, for example, NCOP (2017), Consideration of the Criminal Procedure Amendment Bill (22 June 2017). Available at https://pmg.org.za/hansard/24784/ (accessed 18 February 2019); NCOP Security and Justice (2017), International Arbitration Bill & Legal Practice Amendment Bill: finalization, Deputy Minister; Suspension of Magistrate V Gqiba. Available at https://pmg.org.za/committee-meeting/25569/ (accessed 18 February 2019); NCOP Communications and Public Enterprise (2018), FPB Amendment Bill; National Research Foundation Amendment Bill & Protection, Promotion, Development and Management of Indigenous Knowledge Bill: Adoption. Available at https://pmg.org.za/committee-meeting/27611/ (accessed 18 February 2019); NCOP Economic and Business Development (2018), Labor Bills (NMW; BCEA; LRA & LLA): Finalization. Available at https://pmg.org.za/committee-meeting/26791/ (accessed 18 February 2019); Ad Hoc Committee on the Funding of Political Parties NCOP (2018), Political Party Funding Bill. Available at https://pmg.org.za/committee-meeting/26701/ (accessed 18 February 2019).

22 In the third parliament, for example, NCOP’s amendment to section 75 bills account for only 6% of the amendments made to section 75 bills. See Department and Ministry of Provincial and Local Government (1999).

23 Parliamentary Monitoring Group (2018), Public Service Commission and Traditional Leadership and Governance Framework Amendment Bills: Department Response to Negotiating Mandates. Available at https://pmg.org.za/committee-meeting/26679/ (accessed 19 February 2019). The same with Housing Development Agency Bill (2008), Responses to NCOP – Provincial Negotiating Mandates – Amazon AWS. Available at pmg-assets.s3-website-eu-west-1.amazonaws.com/docs/080611response.doc. (accessed 19 February 2019).

24 Parliamentary Monitoring Group (2018), National Health Laboratory Services Amendment Bill [B15B-2017]: Negotiating Mandates. Available at https://pmg.org.za/committee-meeting/27134/ (accessed 25 November 2018).

25 Parliamentary Monitoring Group (2018), Public Service Commission and Traditional Leadership and Governance Framework Amendment Bills: Department Response to Negotiating Mandates. Available at https://pmg.org.za/committee-meeting/26679/ (accessed 19 February 2019).

26 NCOP Cooperative Governance and Traditional Affairs (2018), Public Service Commission Amendment Bill & Traditional and Khoisan Leadership Bill: Final Mandates. Available at https://pmg.org.za/committee-meeting/27705/ (accessed 19 February 2019)

27 The Independent Panel, p.51.