Crony interlockers and the centrality of banks: the network of Moroccan listed companies

ABSTRACT

In this paper, we study crony capitalism through the lens of Moroccan listed companies and the role of the financial sector. We first highlight the role of banks in the history of Moroccan capitalism from the setting up of the French protectorate over Morocco to the dominance of some major family groups. Using a network analysis of board members of the Moroccan listed companies we confirm the centrality of finance compared to other sectors. We also measure the relational proximity to demonstrate that each cluster of the four main holding families has at least one financial company. Finally, we argue that crony interlockers, who are members of royal foundations and represent an institutional investor or a holding family that owns a bank and/or an insurance company, are the most central actors in the network of listed companies. Our contribution to crony capitalism in MENA analyses the politics of finance in a neoliberal and peripheral country such as Morocco. Boone and Henry [(2004). Neoliberalism in the Middle East and Africa: Divergent banking reform trajectories, 1980s to 2000. Commonwealth & Comparative Politics, 42(3), 356–392] had described the banking sector in Morocco as a private oligopoly. Here, we provide evidence that the monarchy and its entourage are capturing corporate governance instruments such as board membership to control economic activities.

KEYWORDS:

• Interlocks
• crony capitalism
• Morocco
• holding families
• finance
• banks

JEL:

• Z13
• G21
• G38

Disclosure statement

No potential conflict of interest was reported by the author.

ORCID

Oubenal Mohamed http://orcid.org/0000-0001-5625-3662

Notes

1 The Pujo Committee considered that interlocks can engender collusion. The political implication of this was the establishment of The Clayton Act which prohibited interlocks between competing companies.

2 The first column of the table is dedicated to board members while the second comprises company names.

3 It can be 0 if two companies do not share common board members.

4 It is the sum of all common board members between each two companies from the same sector.

5 If we consider interlocks between sector A and sector B we sum up the common board members between all the companies from the sector A and all the companies from the sector B.

6 This is computed using the hierarchical clustering and the Ward method (See the section ‘The clusters of four main holding families’ for more details) which enables us to consider that two companies have high proximity if they share the same board members or they share the characteristic of having no interlocks with others.

7 In this case, degree centrality corresponds to the sum of different board members that each actor meets in the boards of listed companies.

8 The proxy used to define crony interlockers is the fact of sitting in the board of royal foundations.

9 Known in Morocco as Banque Nationale de Developpement Economique (BNDE). Because of mismanagement and misuse of public money, the government dismantled the bank in 2005.

10 Darouich (1989, p. 229) evidences that the bank SMDC who was controlled by the French group Paribas (BPPB) gives important loans to ONA due to interlocks that was inherited from the period where ONA was subsidiary of BPPB.

11 We didn’t find sufficient data for other periods.

12 Each group contains between 6 companies for Credit sector and 12 organizations for Construction sector. The Industry includes 11 companies while finance sector comprises 10 banks and insurances. Both the energy and agribusiness sectors contain 9 companies while the IT and transportation gather 8 entities.

13 This means that we don’t find a lot of board members of companies from each of this three sectors sitting in the boards of other sectors. For instance, there are very few board members from IT companies sitting in finance, transportation or any other sector.

14 Actors in a network can have similar or dissimilar relational profile. Some companies may have a high relational proximity if they have similar relational profile (the same links or absence of links). For instance, this is the case for companies sharing many board members. Companies can also be considered with high relational proximity if they share the characteristic of interlockers’ absence.

15 A company under the sphere of influence of a family group means that it is controlled by the shareholder family.

16 The clustering of data is structured (not arbitrary) using a tree or a dendrogram. Here we apply a bottom-up Hierarchical clustering. This means that we group small clusters into larger ones in an agglomerative procedure.

17 Pajek software is a free program created by Vladimir Batagelj and Andrej Mrvar. This software enables users to visualize networks, compute centrality measures and group vertices in clusters. Pajek implements some important algorithms for network analysis such as ward method.

18 In Ward’s method the distance between two clusters is measured using the sum of squares. Then the algorithm finds the closest pair of clusters and merges them.

19 The sister of the CEO who is another prominent member of the Bensaleh holding family is president of the Moroccan Business Association.

20 Surrounded by a square in the network.

21 Surrounded by a circle in the network.

22 It is the sum of different board members that each actor meets in the boards of listed companies

23 The three that do not appear as mean interlockers are: OCP which is the biggest state owned company in Morocco, Maroc Telecom owned by the Emirati group Etisalat and AKWA, a Moroccan Holding family.

Additional information

Funding

This work was supported by Economic Research Forum (ERF) [grant number #2015-030].