Although Tunisia’s retail banking sector delivers vital services to consumers and businesses, this report identifies several areas in which competition is not working as well as it could. The report finds several features of the market that weaken competition, limiting the ability of banks to compete and the ability of consumers and small businesses to exert competitive demand-side pressure on banks. The report also identifies several regulatory provisions that reduce banks’ incentives to compete.
Competition Market Study of Tunisia's Retail Banking Sector
8. Summary of findings
Abstract
This chapter brings together the key findings of the market study and competition assessment, forming the basis for a discussion of the recommendations. It summarises the market outcomes emerging from the analysis and their key drivers, starting from those more likely to negatively affect competition and create harm to consumers.
The report finds that the market delivers for many consumers. However, while there are undeniably a range of prudential and public policy objectives to consider in relation to banking services, competition is not as strong as it could be in a number of areas, creating a market that does not function as well as it should and leading to worse outcomes for consumers and small businesses than could be the case.
8.1. Market structure and regulatory framework stifle competition
Competition between banks in Tunisia is not as vigorous as it could be, and there is a risk of co‑ordinated conduct among them. Several legal provisions, market practices and ownership structures may facilitate the sharing of commercially sensitive information and the monitoring of price strategies. For example, Article 34 of Circular No. 1991‑22 states that banks need to consult with the banking association before introducing new fees. This increases opportunities for banks to share information, making it easier for them to operate in a co‑ordinated manner rather than competing. Stakeholders interviewed by the OECD said that banks indeed tend to share their plans to increase fees on financial products with competitors. Such communications are likely to undermine incentives for banks to compete vigorously. Section 3.4 also discusses elements of the codes of conduct designed by the CBF such as principles of fair competition vis-à-vis other banks and restrictions on hiring rival banks’ personnel that weaken competition. The OECD has also been made aware of at least two separate allegations that banks have engaged in illegal conduct in relation to information sharing. It is not the role of this report to assess these allegations, but the OECD recommends that the allegations should be considered by the appropriate authorities, as well as a broader strengthening of the role of the competition authority in the sector.
The existence of connections among listed banks reduces banks’ incentives to compete vigorously. Sections 3.2 and 3.3 show that large domestic industrial groups are shareholders in several listed banks and that, in some instances, listed banks have minority shareholders in common. Sections 3.2 and 3.3 also show that the boards of listed banks include individuals connected to common shareholders. The connections shown in Section 3.3 are an underestimate of the actual prevalence of such linkages, as information on non-listed banks and on informal connections between directors was not available. The OECD was also unable to assess the impact on competition of common ownership and interlocking directorates. However, the economic literature suggests that these connections may reduce banks’ incentives to compete and increase the risks of co‑ordinated conduct, leading to higher prices, lower quality and lower levels of innovation.
The large presence of the state in the banking sector does not increase incentives to compete. State‑owned banks seem to be less efficient than other listed banks and have higher proportions of non-performing loans. Furthermore, stakeholders interviewed by the OECD said state‑owned banks are still used by the government to fund state‑owned enterprises and typically follow the strategies of other large banks, for example when setting or changing fees.
The analysis found several instances in which regulatory oversight by the Central Bank of Tunisia has been inconsistent. The OECD understands that at least some banks do not comply with many legal provisions requiring disclosures of fees to consumers or limiting the total sums loaned to related borrowers. Although in 2016 legislation was introduced to slow the trend of increasing bank fees and increase transparency around tariffs, the OECD understands that these consumer protection safeguards have not yet been enforced with the required rigour. The OECD has been informed that the BCT has already prepared a draft circular to provide a framework for commercial and pricing practices. The aim of the circular is to guarantee commercial conduct in the interests of customers and to ensure that customers are provided with clear information throughout the marketing phases. The adoption of this text should enable the BCT to devise a disciplinary procedure for offenders breaching the reporting and transparency obligations on products offered and conditions applied to consumers as per Article 84(1) and Article 84(2) of Law No. 2016‑48 that gives the power to BCT to impose fines. The fact that BCT enforces this provision systematically should increase the incentives for market players to comply with regulation. Limited control and a lack of sanctions for regulatory breaches create the risk of fostering or worsening an uneven playing field, particularly with respect to incumbent banks and potential new entrants.
The evidence shows market outcomes that are consistent with weak competition. Fees on current accounts have increased over the past decade or more. The Observatoire de l’Inclusion Financière has calculated that fees on current accounts increased 67% between 2010 and 2017 in nominal terms. OECD analysis of data provided by the BCT found that between 2015 and 2021, revenues per account increased above the rate of inflation in three out of five years, or 65% in nominal terms. Over at least the past decade, banks’ overall profitability has increased steadily. In addition, consistent with the limited incentives to compete described above, listed prices advertised on banks’ websites do not vary significantly between lenders. Finally, innovation in the finance sector has been low, for example as shown by Tunisians’ very low take‑up of online banking and mobile payments.
Finally, co‑operation between the Competition Council and the BCT is very limited. This hinders the ability of the conseil to stop potentially anti-competitive practices, limits its advocacy role when new regulation is introduced, and ultimately risks reducing compliance with competition law.
8.2. Customer engagement is low
For competition to work well, customers must be sufficiently well informed to buy products or services offering the best value for money. Several aspects of consumer behaviour are important drivers of competition. Consumers need to be sufficiently aware to consider shopping around, and they need to be able and willing to access and understand information on features of products and assess their characteristics to identify the most suitable products. Consumers must also be able and willing to choose, or switch to, their preferred products. And finally, firms need to perceive consumers as likely to act on better offers to have incentives to make such offers. If consumers do not engage with products and if banks expect them not to act, incentives to compete are weakened, which may result in higher prices and poorer quality.
Evidence shows that the engagement of consumers and small business in Tunisia’s retail banking sector is low. In the market for current accounts, four in five consumers and two of three small businesses do not compare fees when opening accounts, and two out of three consumers in the OECD’s consumer survey said they did not know how much they paid. Only 3% of consumers and 4% of small businesses had switched current accounts in the previous year. When seeking finance, small businesses tend to use their current account providers.
Many factors may contribute to this. Consumers and small businesses may find it costly to gather, understand and act on information about financial products. In addition, firms have incentives to weaken competition and make it more difficult for customers to shop around. For example, evidence shows that banks in Tunisia do not make it easy for consumers to find comparable, meaningful information on fees, and that no tools are available to help consumers, such as price comparison websites for current accounts. Banks also create monetary and non-monetary barriers to closing accounts, which ultimately reduces the ability of customers to switch, undermining competitive pressure on banks to lower prices or improve service levels. For example, the process of closing a current account is slow and uncertain, as banks do not always close accounts promptly upon being instructed to do so. In some instances, banks also require the payment of fees for account closures, which may discourage customers from switching.
The report also finds that the existing mediation mechanism is largely unused and ineffective. In 2020, only 301 complaints were made and, of these, only 17% reached a resolution. The OECD’s consumer survey indicated that fewer than one in three individuals was aware of the service and, among those who were aware of it, 18% said they did not use it because they did not expect any benefit, with 15% describing the process as cumbersome. A potential cause of the ineffectiveness of the service is a lack of actual and perceived independence among mediators, who are appointed by the banks (see Section 3.5).
Regulation and market practices that make individuals and micro, small and medium-sized enterprises (MSMEs) less likely to respond to differences in prices or quality also represent a barrier to entry and expansion. Customer inactivity and the inability to choose preferred products and providers increases costs for new entrants and costs for smaller banks seeking to acquire new customers.
8.3. Limited lending to MSMEs
The proportion of lending to MSMEs is low, and surveys have consistently found that access to financing and its costs are a major obstacle to their growth. Chapter 5 provides an assessment of competition in the market for bank lending to MSMEs.
The limited availability of granular data did not allow an accurate assessment of market concentration levels. However, analysis of aggregate data suggests that the market for business loans is concentrated, with the five largest banks accounting for 70‑75% of all lending in 2021, if state‑owned banks were considered as separate entities. The aggregate share of the five largest banks was much higher when state‑owned banks were treated as a single entity. The market shares of these banks have been stable over at least the past decade.
The results of the OECD’s MSME survey show the importance of relationship lending and MSMEs’ limited propensity to shop around for financial products. Establishing banking relationships was an important reason for choosing particular business current account (BCA) providers, and for around 45% of small businesses, their BCA providers were their only finance providers. MSMEs were more likely to seek and obtain finance from their existing BCA providers for a variety of reasons. They may value the convenience of using their BCA providers as a one‑stop provider for all their financial needs. MSMEs may also face high costs to shop around and compare fees and conditions for loan products among providers. MSMEs may also place significant weight on their banking relationships with their BCA providers because they perceive they may be more likely to be able to access financing. In fact, BCA providers typically have more information about the credit and payment histories of their customers and may be able to better price risk for existing customers to finance more quickly and cheaply. The MSME survey also showed that MSMEs applying for financing from their BCA providers were less likely to compare fees and conditions associated with loan products. This reduces the competitive pressure that small businesses could exert on finance providers.
The lack of a private credit information bureau can exacerbate the information advantages enjoyed by larger banks and further increase small businesses’ costs of shopping around for credit and switching banks, especially for newer customers, which are more likely to use single financial services providers. In Tunisia, information on MSMEs’ credit histories is available from the BCT, but stakeholders said that this information is limited, for example because it does not include information on individuals or enterprises that have never used loan products. The lack of tools for sharing accurate credit information reduces the ability of lenders to make accurate lending decisions, reduces the ability of smaller banks to compete, and increases borrowers’ switching costs. The BCT has slowly introduced a framework to regulate private credit information bureaus that includes burdensome requirements, such as unnecessarily high capital requirements, that stifle entry (see Section 5.5).
The analysis also shows that Tunisian banks rely heavily on collateral when granting loans. According to the World Bank Enterprise Surveys, the average collateral requested of MSMEs in Tunisia is almost 300% of loan value, the highest among the countries in the World Bank’s enterprise surveys dataset (see Figure 5.8). A factor contributing to this may be the cap on lending interest rates, which aims to protect vulnerable customers but prevents banks from accurately pricing risk. The effect of the cap is likely to vary between banks. Smaller providers, which may have higher financing costs, seem to price on average closer to the cap than larger banks. In some periods, the average interest rates charged by some banks on some lending products is equal to the cap, implying that the cap is binding for all loans in that category that are granted during that period (see Section 5.3.1).
Several stakeholders interviewed by the OECD shared concerns about practices of banks favouring related borrowers. (OECD, 2022[1]) found that in 2019, five industrial groups controlled more than 60% of the turnover of the most important private companies in Tunisia. These five industrial groups also have direct links to banks, a situation that has the potential to reduce access to credit for firms not connected to these groups. Stakeholder evidence suggested that this may be affecting lending for many companies. Section 5.2.3 shows that state‑owned banks and banks controlled by Tunisian industrial groups account for around three‑quarters of business loans, a significant proportion of total loan volume. Although the regulatory framework in Tunisia contemplates safeguards to mitigate these concerns, according to annual reports, at least three large banks have exceeded the limit on lending to connected parties imposed by Circular No. 2018‑06. Given the lack of granular data, the OECD was not able to analyse the effects of this issue in detail but notes the significant potential for links between banks and industrial companies to affect lending in the wider economy.
The circular introduced by the BCT to rein in these practices and enforce a limit on the volume of lending to related borrowers is not as effective as it could be. Evidence suggests that some banks do not comply with these measures, either exceeding the threshold or by engaging in agreements with other banks to lend to one another’s related parties.
Finally, stakeholders described informal decision-making mechanisms, especially at banks controlled by Tunisia’s large industrial groups, which could reduce the importance of boards and increase large industrial groups’ influence over banks.
8.4. Unnecessary regulatory restrictions on payment service providers
The licensing process for payment services providers creates unnecessary barriers to entry. Some of the requirements and elements of the process, such as higher capital requirements than other countries, ad hoc provisions, and uncertainty about the length of the process, increase the costs of obtaining licences and limit market entry. For example, Law No. 2016‑48 introduces capital requirements that are between 12 and 76 times higher than the capital requirements for similar service providers in EU countries, and the length of process and ad hoc requirements increase the discretionary power of the BCT. Law No. 2016‑48 also requires that payment service providers operate governance structures like those of traditional banks, which increases costs for potential entrants.
These barriers are likely to discourage smaller firms from applying for licences, and they risk favouring affiliates of existing banks. In fact, the only providers that have successfully applied for licences are either traditional banks or institutions with ties to traditional banks. Three of the four establishments that have been licensed by the BCT so far are directly linked to banks, while the fourth is linked to a microfinance institution. This has a negative effect on competition and innovation as it reduces incentives for new entrants to compete and eliminates players that could vigorously compete with traditional banks. Section 6.2.2 highlights the fact that banning fees for mobile payments below TND 15 may result in payment services providers excluding customers that are more likely to make small payments.
Tunisia is home to 22 banks, yet a large proportion of its population is not served by them. Two-thirds of individuals do not have personal current accounts and three in four do not have savings accounts. High fees are one of the reasons why individuals remain unbanked. Mobile payment providers have boosted competition and financial inclusion in other developing countries, but regulation in Tunisia risks excluding independent providers that could challenge established banks, reducing innovation and the provision of newer and cheaper banking services.
8.5. Other findings
The analysis finds that Tunisia’s geographical markets are local, and that competition is defined by the density of banks’ branch networks, upon which consumers and small businesses rely to open and manage their accounts. Customers are very unlikely to choose banks that do not operate branches nearby, and banks have competed by extending their branch networks. However, the growth rate of branch networks has fallen significantly, which is consistent with a weaking of banks’ incentives to compete. A few banks have recently started to offer online banking services (often at an additional cost), but take‑up of online banking services remains very limited.
The cost of building branches is significant and represents a barrier to entry and expansion for smaller and newer banks. Consistent with the slow pace of growth of banks’ branch networks, shares of the current account market and of the bank loan market have been very stable since at least 2015 (see Sections 4.3 and 5.2.2).
The report also discusses the role of La Poste Tunisienne in the retail banking sector. La Poste plays an important role by offering cheaper banking services and increasing financial inclusion in Tunisia. However, its lack of a banking licence prevents it from providing a full range of financial products. National postal operators have played an important role in boosting financial inclusion in many countries, as they are more likely than traditional financial institutions to serve clients with lower incomes, lower levels of education and who are more likely to be unemployed. Given La Poste’s extensive branch network, the fact that its fees are lower than those of banks, and its high market share among young people, it has the potential to exert significant competitive pressure on traditional banks. However, this potential is hindered by its lack of a banking licence.
The report also discusses the low usage of cards to make payments. Around one in five individuals have card to make payments, but even those who have them do not use them often. High fees seem to be one of the reasons for the low take‑up. According to the OECD’s consumer survey, around 30% of individuals without cards said they would obtain one if fees were substantially lower. The small number of merchants accepting cards is another potential reason for low card usage. Around 40% of respondents to the consumer survey said they would be more likely to use payment cards if more shops accepted them. However, the high cost of card payment processing equipment may be preventing merchants from accepting cards. High fees for consumers and for merchants may be partly due to existing market practices whereby banks, via the Societé Monétique Tunisie, source cards and other services from a single provider. This increases costs, which banks then pass on to consumers.
References
[1] OECD (2022), “OECD Peer Reviews of Competition Law and Policy: Tunisia”, https://www.oecd.org/competition/oecd-peer-reviews-of-competition-law-and-policy-tunisia-2022.htm.