Competition in Tunisia’s current account segment is not working as well as it could. Given the limited use of online banking services, banks rely on their branches to serve and attract customers, which limits consumers’ choice of suppliers, especially in rural areas. Market shares have been stable since at least 2015, and very few consumers and small businesses shop around for current accounts, compare services across providers, or know how much they pay. Very few switch providers. This is partly due to bank practices that prevent consumers from easily accessing information and making informed decisions about the products they buy. Market outcomes are also affected by the factors described in Chapter 3, which weaken banks’ incentives to compete. As a result, revenues from current accounts – a proxy for fees – and the overall profitability of banks have increased significantly in the past decade, and two‑thirds of Tunisians do not have personal current accounts; one in five say that high fees are one of the reasons for this.
Competition Market Study of Tunisia's Retail Banking Sector
4. Current accounts
Abstract
The chapter provides an assessment of competition in Tunisia’s current account segment. For competition to work well, customers must be sufficiently well informed to buy products offering the best value for money, and suppliers need to have incentives to compete vigorously. After a brief description of the main characteristics of current accounts, the chapter discusses the key role of bank branches in serving and attracting customers (Section 4.2), and market structure and concentration (Section 4.3). Section 4.4 turns to market outcomes, providing an assessment of recent price trends, quality measures, and an estimate of the proportion of individuals and micro, small and medium-sized enterprises (MSMEs) with current accounts. The chapter then presents an assessment of customer behaviour (Section 4.5) and a discussion of bank practices that exacerbate customer inaction (Section 4.5.4). It ends with a discussion of payment cards and potential reasons for their low take‑up.
4.1. Parameters of competition
Current accounts are typically used to store money, access it quicky, and make and receive payments. Additional services are linked to current accounts that consumers and businesses may or may not use, including additional payment methods such as debit or credit cards, cheques, cards to make cross-border payments, and borrowing facilities such as overdrafts. Current accounts may vary according to the characteristics of the product or the type of customer.
Current accounts offer functionalities that overlap with other products. For example, savings accounts are typically used to store money and obtain interest, but typically offer limited or no payment functionality. Some banks redirect consumers without regular incomes towards savings accounts. Other products that partly overlap with current accounts are digital wallets offering mobile payment facilities. These services have the potential to compete at least partly with current accounts (for more information on digital wallets and related payment services, see Chapter 6). However, the take‑up of mobile payments in Tunisia is currently very low.
MSMEs often use current accounts to create banking relationships and improve their access to finance to expand or obtain financial assistance. MSMEs’ needs for banking services differ from consumers’ needs and vary widely depending on their characteristics. Micro businesses tend to behave like consumers; they may use a personal current account (PCA) for business purposes, they are more likely to use cash, they are less likely to use overdrafts, business loans and credit cards, and are less likely to engage in cross-border transactions. Larger businesses are more likely to use a wide range of banking services and financial products.
Strong linkages exist between business current accounts (BCAs) and business loans. MSMEs are more likely to obtain business loans from their BCA providers than from other banks. This is more common among smaller or newer enterprises, as banks may rely more on existing banking relationships where credit information is limited, especially if a credit information bureau does not hold comprehensive information (see Section 5.5).
Different providers focus on different market segments. For example, consumers with accounts at La Poste Tunisienne are more likely to have lower incomes, to live in rural areas, and to be unemployed. Also, branch networks appear to be a very important means by which banks can attract new customers and for consumers and businesses using current accounts, due to personal preferences and the limited availability or high cost of online services. Proximity to a branch is a key reason for choosing account providers, so geographical markets are likely to be local, and competitive dynamics vary across regions, based, for example, on the density of banks’ branch networks. Given the role of branches and the different customer segments, where appropriate this chapter presents the results of analysis at different geographical levels and broken down by the main consumer demographic factors.
4.2. How banks compete
This section looks at some key characteristics of the competitive landscape, such as the importance of branch networks for both customers and banks, and the use of current accounts as a gateway for cross-selling other products (especially finance for MSMEs).
Geographical markets are local and defined by the density of banks’ branch networks. Customers are very unlikely to choose providers that do not operate a branch nearby. The analysis showed that:
Banks rely heavily on their branch networks to attract new customers and serve existing ones. According to the OECD’s consumer and MSME surveys, 95% of consumers1 and 99% of MSMEs2 opened their current accounts at bank branches. Many consumers also use cash to make payments and use branches to withdraw money, which is the most common reason for having a PCA (53% of consumers).3
Proximity to branches is the most common criterion for choosing a current account provider, with 36% of consumers and 34% of MSMEs valuing proximity to a branch when choosing their current account providers. Other reasons for choosing current account providers are fees and other terms (30% of consumers and 25% of MSMEs), and whether friends or family members have accounts at the same bank (22% of consumers), alongside whether they know someone working at the bank (22% of MSMEs).4,5
Only 7% of consumers said they typically managed their current accounts online.6
Banks interviewed by the OECD consider the expansion of branch networks a key element of their strategy for attracting new customers.7
Current account providers compete by expanding branch networks. The number of bank branches in Tunisia increased from 1 774 in 2016 to 1974 in 2020. The branch network of La Poste Tunisienne is comparable to those of banks. In 2020, La Poste had 1 046 branches and 427 ATMs (La Poste Tunisienne, 2021[1]) compared to a total of 1974 branches and 3 029 ATMs among all banks combined (Banque Centrale de Tunisie, 2022, p. 59[2]). Figure 4.1 shows that banks tend to concentrate their branches in Tunis and in the east of the country, regions characterised by higher per capita incomes. Areas with lower per capita incomes have fewer bank branches and relatively more La Poste’s branches. For example, central western Tunisia has only 80 bank branches serving around 1.5 million people, which means on average one bank branch for every 18 671 people. See Section 7.3 for a discussion of branches and barriers to entry and expansion.
Figure 4.1. Prevalence of banks’ and La Poste’s branches by region
Note: Data on La Poste branches is from July 2022. Data on bank branches is based on data provided by the BCT.
Source: La Poste; BCT; OECD calculations.
Some MSMEs use their BCAs as gateways to establish relationships with banks and obtain additional services such as finance (around 16% of the respondents to the OECD’s MSME survey said they chose BCAs for that reason).8 Banks typically rely on their customers’ credit histories to assess creditworthiness and banks may therefore be better placed to assess risks associated with existing MSME customers than those associated with other MSMEs. This is reflected in their MSME credit product holdings. MSMEs typically hold several credit products with their BCA providers. For example, 41% of MSMEs hold credit cards with their BCA provider and around 13% hold them with another provider (see Figure 4.2 for a breakdown of credit products).9
More generally, 45% of MSMEs with current accounts said they held credit products only with their BCA providers10 and 31% said they had credit products with both their BCA providers and other banks.11 Among respondents, 21% did not hold credit products12 and only 3% said they held no credit products with their BCA providers but held credit products with other banks.13
Figure 4.2. Credit products held with BCA providers vs. with other banks
Note: MSMEs could choose more than one answer.
Source: OECD MSME survey (Q27, N=804 and Q28, N=1 005).
4.3. Market structure and shares of supply
The structure of the banking sector may affect the incentives of firms to compete within it. For example, larger firms may have fewer incentives to compete with their rivals. Also, market shares that are stable over time may be consistent with weak competition.
The Central Bank of Tunisia provided annual data on the number of current accounts by bank for a six‑year period between 2015 and 2020. The data had several limitations as they did not include current accounts at La Poste, they could not be broken down by PCAs and BCAs, they could not be broken down by demographic characteristics, and they included dormant accounts, which the OECD understands represented a significant proportion of the total. However, despite these limitations, the data were useful for analysing shares of supply over time. Figure 4.3 shows that the six largest providers (excluding La Poste) accounted for between 73% and 75% of all current accounts (PCAs and BCAs) in each year between 2015 and 2020. Moreover, the share of supply of each bank remained stable over the period.
Figure 4.3. The six largest banks’ shares of Tunisia’s current account sector
Note: Each colour corresponds to a bank. Data is anonymised.
Source: BCT; OECD calculations.
The next section presents the findings on concentration at the national and regional levels, and for certain demand segments based on the OECD’s consumer and MSME surveys. Although geographical markets are likely to be local and to follow the density of branch networks, given the usage of branches and the very low take‑up of online banking, the shares of supply at the national and regional levels show differences in market participants’ strategies and overall size.
4.3.1. Shares of supply of PCAs and geographical markets
Shares of supply in the PCA market were calculated using information on current accounts that respondents to the OECD’s consumer survey identified as their main accounts.14 In 2022, La Poste was the largest provider of PCAs in Tunisia, with a share of around 22%. The largest bank and the second-largest provider of PCAs was BIAT, with around 14% of PCAs. The top ten providers accounted for around 91% of PCAs and state‑owned entities combined accounted for 44% of PCAs.
Table 4.1 shows the most granular geographical breakdown available. Shares of supply varied significantly across regions, indicating that competitive dynamics appeared to vary by locality. BIAT was the largest PCA provider in Greater Tunis, followed closely by La Poste. La Poste was the largest provider in each region outside Greater Tunis. The share of La Poste was 34% in the southwest of the country, which is more rural and characterised by lower income per capita. La Poste’s customers were more likely to have lower incomes, to live in rural areas, and to be unemployed than banks’ customers.
Some of the largest providers accounted for a very small proportion of accounts in certain regions. This could reflect the differing competitive strengths of banks in different regions, as well as business strategies, as a bank may target higher-income individuals and thus be less present in lower-income areas. State‑owned entities had a larger share of accounts in the west of the country, which is characterised by lower income per capita. Given the cost of growing branch networks, as discussed in Section 7.3, the shares could also represent historical branch networks and reflect the difficulty of expanding into new areas.
Table 4.1. Shares of supply in the PCA market by geographical region
Note: State‑owned companies are shaded.
Source: OECD consumer survey (Q3 and D2, N=1 086).
Consistently, shares of supply varied significantly also depending on the job status of consumers. Table 4.2 shows that self-employed and unemployed individuals were more likely to hold current accounts at La Poste, while BIAT was the largest provider for employed consumers.
Table 4.2. Shares of supply in the PCA market by job status (selected categories)
Note: State‑owned companies are shaded.
Source: OECD consumer survey (Q3 and D5, N=830).
4.3.2. Shares of supply of BCAs and geographical markets
Shares of the BCA market were calculated using information on the current accounts that MSMEs responding to the OECD survey identified as their main accounts. In 2022, BIAT was the largest provider of BCAs, with a share of around 16%. The second-largest provider was Attijari, with 15%. The top seven providers accounted for around 79% of current accounts.15 Table 4.3 shows that shares of supply varied significantly depending on the size of enterprises. The largest provider of BCAs for micro‑enterprises was Attijari, with 18% of accounts. The most common BCA provider for businesses with 50‑199 employees was BIAT, with 27% of accounts.
Table 4.3. Shares of supply by enterprise size
Note: State‑owned companies are shaded.
Source: OECD MSME survey (Q3 and S1, N=804).
Table 4.4 shows that Attijari was the largest BCA provider in Greater Tunis, followed by BIAT. BIAT was the largest provider in Tunisia’s northeast, central east and southeast.
Table 4.4. Shares of supply by geographical region
Source: OECD MSME survey (Q3 and S3, N=804).
4.3.3. Conclusions on market structure
Given the importance of branches, geographical markets are local. Shares of supply at an aggregate level are useful for identifying market participants’ strategies and overall size. At a national level, in 2022, the ten largest providers accounted for around 91% of PCAs and the largest seven providers accounted for around 71% of BCAs.16 Given that the state‑owned banks are controlled by one single entity (i.e. the state), it is possible to consider state‑owned banks as a single entity, and, in that case, concentration levels would be higher. Figure 4.3 shows that the shares of supply of the largest banks remained stable from at least 2015 to 2020. State‑owned institutions collectively account for around 44% of PCAs at a national level. Shares of supply vary among local markets and the competitive strength of banks differs by region, likely as a result of historical branch locations and strategies. Shares of supply broken down by geographical and demographic characteristics show that providers may target different market segments. For example, consumers opening accounts at La Poste are more likely to have lower income and live in rural areas.
4.4. Market outcomes
This section describes three market outcomes: price, quality and quantity. Analysis of market outcomes provides information about the how competition works and may offer insights on harm to consumers.
4.4.1. Pricing outcomes
Analysis of fees may provide information on competition in the market. For example, significant price dispersion combined with market shares that are stable over time and low rates of consumer switching are consistent with weak consumer engagement. Given that current accounts are complex products involving many services, each with an associated fee, comparisons are challenging. This section describes the trend towards non-interest revenues from current accounts, an index of current account fees, and comparisons of account maintenance fees (the most common fees) for the main account types.
As explained in Section 2.2.1, competition authorities typically assess firms’ profitability in their relevant markets. High and sustained levels of profit indicate that firms are able to exercise market power and increase prices above costs. Unfortunately, granular data on profitability specific to the current account segment was not available to the OECD. Data presented in Section 2.2.1 shows that Tunisian banks have increased overall profits in recent years and have been more profitable than the OECD and EU average. The profitability of Tunisian banks is comparable with those in other Middle East and North Africa (MENA) region countries for which data are available. Given the limited data, it was not possible to compare fees on current accounts across countries.
Recent trends: fees
The BCT provided the OECD with data on total non-interest revenues earned on current accounts and the number of accounts open at each current account provider (excluding La Poste) for the period between 2015 and 2020. Although the data received could not be broken down by PCAs versus BCAs, total revenues divided by the total number of accounts were a proxy for the average all-in price that customers paid for their current accounts.17 Figure 4.4 shows that non-interest revenues per current account adjusted for inflation increased in three of the five years observed. In 2016 non-interest revenues per account increased 4.6% above the rate of inflation. In 2018 they increased 8.5% above inflation. In 2019 they increased 2.8% above inflation. In 2017 they were in line with inflation, and they fell 11% below the inflation rate in 2020, almost certainly due to the COVID‑19 pandemic. Ideally, prices should be compared with costs, however, given that information on cost was not available, prices were adjusted for inflation.
Figure 4.4. Non-interest revenues per current account adjusted for inflation
Note: Weighted average of non-interest revenues per account at 18 banks.
Source: BCT; OECD calculations.
Higher revenues per account may reflect more frequent usage of current accounts. In fact, banks typically charge fees for single transactions, such as making bank transfers, receiving bank transfers, making payments by cheque, cashing in cheques, resetting PIN numbers on cards, making cash payments into accounts at branches and withdrawing cash. Thus, revenues per account could be driven by consumers making more payments over time.
However, evidence shows that listed fees on current accounts have increased significantly. Between 2010 and 2017 the Observatoire de l’Inclusion Financière (OIF) calculated the Indice de Prix des Services Bancaires (IPSB), or Banking Services Price Index, which tracks current account fees. Monthly fees, online banking fees, fees on payments and fees on bank transfers have the largest impact on the index.18 The OIF calculated that the index increased by 65.8% between 2010 and 2017. Several stakeholders said that banks increased fees to cover 70% of their fixed costs. The same stakeholders said the cost of personnel had increased in recent years, and that this justified the increase in fees for consumers.
To counter the trend of increasing fees, in January 2020, the BCT announced a list of 14 banking services, including the opening and closure of bank accounts, as well as deposits and withdrawals, should be provided free of charge (OIF, 2023[3]). This initiative seems not to be legally binding, as stakeholders noted that banks continue to charge fees for the provision of some of these services. The OECD understands that, as part of ongoing financial inclusion reform, the government is considering legislation forcing banks to offer certain services free of charge, at least to a certain category of customers, namely those deemed more vulnerable. As of June 2023, the exact scope of the proposal, both in terms of services and customers, is currently under discussion.
Price dispersion in PCAs
This section shows the distribution of the maximum fees on selected services for current accounts listed by banks and La Poste on their websites. The data was provided by the BCT.
At one end of the scale, price dispersion in a market may indicate that consumers are not shopping around and that providers offering higher fees are not constrained by providers offering lower fees. At the other end, little price dispersion may be consistent with weak competition. This could be particularly concerning, given the frequent interactions among banks and the process for communicating price changes to the BCT via the banking association, as discussed in Section 3.4.
Data available from the OIF include the minimum and maximum fees listed on each bank’s website for various operations associated with PCAs. Figure 4.5 shows the minimum, median and maximum values of the distribution of the maximum annual maintenance fees on various account types. This information does not include the proportion of consumers paying those fees. However, it is also the only information available to consumers when shopping around, so it serves as a proxy for what consumers expect to pay.
Figure 4.5 shows that in October 2021:
The maximum annual management fees listed on banks’ websites for current accounts and cheques are similar across banks. The maximum fees on current accounts are typically around TND 120 and the maximum fees for cheques are typically between TND 60 and TND 70. There is a large dispersion for maximum listed fees on cards.
The maximum fees listed by La Poste are significantly lower than those advertised by banks for current accounts, cards and cheques.
The low degree of dispersion of listed fees for current accounts is consistent with concerns expressed by stakeholders relating to market practices involving exchanges of information on fees.
Figure 4.5. Distribution of maximum management fees for individuals, 2021
Note: The chart shows the distribution of the maximum listed fees for selected services for 12 listed banks and La Poste.
Source: OIF; OECD calculations.
4.4.2. Quality outcomes
A comparison of quality outcomes among providers can provide information on competition in the market. Significant differences in quality and low switching rates are indicative that competition is not working well.
The proportion of individuals in the OECD consumer survey saying they were satisfied or very satisfied can be used as a proxy to assess and compare quality among providers. The limitations of this analysis, however, mean the results should be interpreted with caution. For example, the stated level of satisfaction may not be an accurate proxy for quality because of consumers’ poor understanding of services. In fact, the results of the survey suggest that most consumers do not shop around when they open accounts, and that they are unaware of the fees they are paying.
Level of satisfaction and dispersion across providers
More than one in five consumers in the OECD consumer survey was unsatisfied or very unsatisfied with their PCAs. Customers of La Poste were typically happier, with only 8% saying they were unsatisfied or very unsatisfied, while the proportion of bank customers varied between 6% and 44%, depending on the bank.19
The responses on satisfaction levels may reflect the low expectations that consumers have of services. Responses may also take into account the basic and limited use that consumers make of PCAs. In fact, only 12% of respondents said they used current accounts to make or receive payments, while 37% said they used current accounts to receive salaries from their employer and 53% said they used them to withdraw cash.20 There was no statistically significant correlation between market share and the proportion of satisfied respondents.
According to the OECD’s MSME survey, almost one in four MSMEs was unsatisfied or very unsatisfied with their BCA and more than one in four was dissatisfied with the cross-border functionality of their account.
4.4.3. Quantity outcomes (proportion of consumers and MSMEs with current accounts)
This section presents the proportion of consumers and MSMEs with current accounts. This is an important competitive outcome, as one in five individuals said they did not have a PCA because fees were too high. The section also presents an international comparison using data from the World Bank.
Two thirds of Tunisians do not have a PCA
In 2022, around two‑thirds of Tunisian consumers did not have PCAs. Of the 36% of individuals with current accounts, 29% used banks and 10% used La Poste. Proportions varied according to demographic characteristics. Table 4.5 shows that men were more likely to have current accounts than women and Table 4.6 shows that current accounts were more common among higher-income individuals. Among consumers who said their household had a monthly income of less than TND 1 000, 76% did not have a current account, compared to 36% of those whose households had a monthly income of more than TND 2000.
Table 4.5. Banking penetration by gender
Note: Around 2% of respondents had current accounts at both La Poste and a bank. Two respondents refused to disclose their gender.
Source: OECD consumer survey (Q1 and D3, N=2 995).
Table 4.6. Banking penetration by income brackets
Note: Around 2% of respondents had current accounts at both La Poste and a bank. The sample in the consumer survey was representative of the population according to key demographics such as gender, geography and age, but respondents to the survey may have underreported data on income.
Source: OECD consumer survey (Q1 and D6, N=2 995).
While not perfectly comparable, these estimates are consistent with the estimates of the 2021 World Bank Global Findex Database. According to that database, 37% of Tunisians had an account (by themselves or together with someone else) at a bank or another type of financial institution, or had used a mobile money service in the past year. The proportion was stable compared to 2017 and increased by 10 percentage points from 2014. Other MENA countries for which data was available had comparable banking penetration rates in 2021 and some, such as Morocco, registered an increase compared to previous years.
Figure 4.6. Banking penetration in Tunisia and other MENA countries
Source: 2021 World Bank Global Findex Database.
The OECD consumer survey also asked consumers without PCAs an open question on the reasons they did not have an account. Among the respondents, 55% said they did not need an account, 19% said the fees were too high, and 8% said they did not have the necessary documents. Around 3% said a family member already had an account.21 According to the 2021 Global Findex Database, the most common reasons in Tunisia for not having an account was a lack of funds (71%), followed by the fact that another family member had an account (30%), high fees (18%), a lack of trust in financial institutions (16%) and a lack of necessary documents (14%). The responses were comparable to the average for other MENA countries for which data was available. 22
The large proportion of respondents who said they did not need an account could be explained by the lack of funds (as shown by the Global Findex Database). Another reason could be a preference for using cash and Tunisia’s large informal sector.
Most MSMEs have BCAs
Unlike the market for PCAs, the proportion of MSMEs with BCAs was high. In the survey, 77% of MSMEs said they had BCAs, but results varied significantly according to enterprises’ size.23 Around 70% of micro businesses used BCAs, compared to 94% of enterprises with between 50 and 199 employees.24 The survey showed that 78% of MSMEs without BCAs used PCAs as current accounts for their enterprises.25 This is consistent with a World Bank Enterprise Survey showing that banking penetration rates among MSMEs in Tunisia were higher than in other MENA countries (see Figure 4.7).
As discussed in Annex C, the contact details of respondents to the MSMEs survey come from the registry of the Registre National d’Enterprises (RNE). This means that informal companies were not represented in the sample, which thus likely overestimated the true proportion of businesses with current accounts.
Figure 4.7. Proportion of SMEs with BCAs or savings accounts in Tunisia and MENA countries
Note: Data for Tunisia were from 2020. Data for Morocco and Lebanon data were from 2019 and Egypt data were from 2016.
Source: World Bank Enterprise Surveys.
Among the MSMEs without BCAs, 57% said they preferred to use PCAs and 41% said they did not need a formal financial institution. Around 6% said fees were too complex and another 6% said fees were too high.26
4.4.4. Conclusions on market outcomes
Despite the limited availability of data, evidence suggests that between 2010 and 2020, the overall profitability of banks increased both in absolute terms and as a share of GDP (see Section 2.2.1), that revenues per current account increased above inflation between 2016 and 2020, and that fees increased in nominal terms between 2010 and 2017. The consumer survey found that around one in five Tunisians considered high fees the reason for not having PCAs. Among those with accounts, one in five consumers and one in four MSMEs were unsatisfied with their current accounts. Finally, higher fees were a reason for not having personal current accounts for almost 19% of the two‑thirds of Tunisian consumers without accounts.27
The next section describes the analysis of customer behaviour, which is a key part of the assessment of competition. This includes an assessment of how customers engage with their current accounts, the competitive pressure exerted by the demand side, and companies’ practices that prevent consumers from accessing, assessing and acting on information.
Box 4.1. La Poste Tunisienne
La Poste Tunisienne plays an important role in Tunisia’s retail banking sector. Current accounts offered by La Poste are typically cheaper than those offered by banks and are particularly popular among lower-income customers often excluded by traditional banks, which prefer to focus on consumers with regular and higher incomes. La Poste is also a major player in the savings market, with around 4.1 million savings accounts, representing around 45% of all savings accounts in Tunisia (see Table 4.7). In 2020, deposits held in savings accounts at La Poste were worth around TND 6.7 billion and accounted for around 16% of the total value of deposits in Tunisia. Deposits held in postal savings account are managed by the Ministry of Finance, which remunerates La Poste’s customers at a fixed rate of 2% per annum.
La Poste cannot provide loan products such as overdrafts, personal loans or mortgages because it does not have a banking licence (the OECD understands that La Poste unsuccessfully applied for a banking licence in December 2018). As a result, Tunisians are likely to open current accounts with La Poste after they turn 18, but they typically switch to banks when they need credit.
Table 4.7. Proportion of individuals with current and savings accounts
Note: Proportions do not add up to 100% because some individuals said they held accounts at both a bank and La Poste.
Source: OECD consumer survey (Q1, N=2 995).
National postal operators have played an important role in increasing financial inclusion in many countries. Using the World Bank Global Findex database, Anson et al. (2013[4]) found that postal operators were more likely than traditional financial institutions to serve clients with lower incomes, lower levels of education, and a higher likelihood of being unemployed. Unsurprisingly, they found that the larger the postal branch network was (both in absolute terms and relative to the country’s population), the greater the number of individuals with postal accounts was.
Given La Poste’s extensive branch network, its low fees compared to those of banks, and its large share of supply in the current and savings account segments among young people, La Poste has the potential to exert significant competitive pressure on traditional banks. However, this potential is hindered by its inability to offer lending products.
4.5. Customer behaviour
Effective competition requires well-functioning demand and supply sides. In particular, the ability of consumers to choose products and providers is a key driver of effective competition. Consumers need to be well informed to buy products and services that offer the best value for money, otherwise competition may not work well. Indeed, when consumers are relatively inactive and only poorly informed, and do not switch when better products or services appear on the market, suppliers have few incentives to offer cheaper or higher-quality products and services.
Several aspects of consumer behaviour are important drivers of competition in Tunisia’s banking sector. Consumers need to have sufficient awareness to consider shopping around for current accounts. They need to be able and willing to access and understand information on the features of current accounts and assess available products to identify the most suitable ones. Consumers also need to be able and willing to choose, and switch to, their preferred products. And finally, firms need to perceive consumers as likely to act on better offers in order to have incentives be incentivised 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 lower quality. As discussed at the beginning of this chapter, high prices are one of the main reasons why consumers in Tunisia do not open PCAs. Firms have incentives to weaken competition and create user experiences that make consumers less active. For example, firms may attempt to create barriers limiting consumers’ ability to understand information or to act on that information.
This section considers evidence on customer behaviour in the markets for PCAs and BCAs, and the potential effects on competition. The results of both of the OECD’s surveys suggest that the level of engagement among both consumers and MSMEs is low. Potential reasons for low engagement include a range of barriers to accessing and assessing information to use in searching and shopping around, and barriers to closing and switching accounts.
4.5.1. Low levels of consumer engagement
Consumers engage in the PCA market by understanding the features of the products available, accessing information, comparing and shopping around for products, making conscious decisions on which providers to use, and whether to stay with existing providers or switch to new ones. This section considers the length of time consumers stay with their PCA providers, searching and switching rates, and awareness of fees.
Tunisian adults tend to stay with their PCA providers for a long time. The OECD’s consumer survey showed that 25% of consumers had stayed with their current providers for 5‑10 years and that 40% had stayed with their providers for at least ten years. Breaking down the results by age, the survey results showed that almost 68% of consumers aged 44 or older had been with their banks for more than ten years.28
The level of searching and shopping around is low, with 80% of consumers saying that they had not compared fees and conditions with those offered by other providers when they opened accounts. This means that prices and other conditions are not a key competitive dimension for many consumers.29
The level of awareness of fees is low, with 62% of respondents to the survey saying they did not know anything about the fees associated with their current accounts. Just 31% said they had a vague idea and 7% said that they knew precisely what fees were charged.30
Switching rates are low. Around 3% of account holders had switched current accounts over the previous year (and around 10% had switched over the previous three years). Around 1.5% had closed an old account and around 1.5% had kept an old account open.31 The annual switching rate was comparable to the UK PCA market in the 2010s (Competition and Markets Authority, 2017[5]). Low banking penetration combined with a low switching rate create a significant barrier to entry and expansion in the PCA market. This is discussed in Chapter 7.
4.5.2. Low levels of MSME engagement
MSME engagement is low and switching rates are low, with MSMEs tending to stay with the same providers for a long time. MSMEs are unlikely to search for information on fees and conditions, and they typically do not know how much they are paying.
Around 33% of respondents to the OECD’s MSME survey had stayed with their existing BCA providers for 5‑10 years and 40% had stayed with their BCA providers for at least 10 years.32 Breaking down the results by business age, the survey showed that 91% of enterprises established more than five years ago had been with their account providers for at least five years.33
The level of searching and shopping around is low. In the MSME survey, 67% of respondents said they had not compared fees or other conditions among providers when they opened their accounts. Just 12% had compared fees or other conditions with only one other provider.34 This suggests that fees are not a key competition dimension for many enterprises.
Around one in four MSMEs said they knew nothing about the fees they paid.35
Around 10% of MSMEs had switched current accounts over the previous year and around 22% had switched over the previous three years. Among the 22% that had switched accounts in the previous three years, 44% had closed an old account and 56% had kept an old account open.36 Among the respondents to the survey that said they had switched their BCAs, 50% said they had done so because the new provider offered better service and 14% because it offered lower fees.37 Among the respondents that had not switched, around 73% said they had not done so because they were satisfied with their current accounts and 10% said they were not aware of alternatives.38 However, although the results of the survey indicate that MSMEs have a better understanding of fees than individuals, 22% of MSMEs said they knew nothing about the fees on their BCAs.39
One in three MSMEs had accounts at more than one bank and, among these, 4% had more than one account at the same bank.40 Of the MSMEs that “multi-home” in this way, 81% did so because they wanted access to financial products from different banks, and 21% did so because they needed accounts for cross-border transactions.41
4.5.3. Conclusions on customer engagement
The evidence suggests that engagement among both consumers and MSMEs is low. This weakens the incentives for providers to compete and offer better-value products. This is consistent with high profitability and the trend towards higher fees. Customers’ inactivity may be exacerbated by regulation and market practices. The next section presents several of these practices.
4.5.4. Banks’ practices exacerbate customer inaction
Banks’ practices make it difficult for consumers and MSMEs to engage in the current account market. Evidence suggests that banks’ practices raise barriers to accessing information, barriers to assessing information, barriers to closing and switching accounts, and barriers to opening accounts. In addition, linkages between BCAs and PCAs, alongside linkages between BCAs and financing, may affect competitive dynamics in the BCA market. The remainder of this section considers each of these barriers.
Barriers to accessing information
Article 3 of Circular No. 2006‑12 requires banks to adopt communication policies based on principles of transparency. However, Section 4.5.1 shows that 62% of respondents in the consumer survey said they did not know anything about the fees associated with their current accounts. Only 31% said they had a vague idea, and 7% said they knew precisely what fees were charged. Among the enterprise respondents in the MSME survey, 22% said they knew nothing about the fees associated with their current accounts, 45% said they had a vague idea, and 43% said they knew precisely what fees were charged.42
Article 37 of Circular No. 91‑22 requires banks to publish fees and terms in brochures made available to customers. However, this information is typically not available at bank branches, and websites feature only vague information. A review of current account providers’ websites was carried out to check the availability of information on fees. The review included the websites of Tunisia’s seven largest listed banks. Information on fees and interest rates was typically available only in an aggregate form and was not specific to the accounts offered. When information on fees for a given transaction was available, it was often presented as a range and it was not clear what the exact sum of the fee depended on. In addition, the time validity of the fees was often not specified. In some instances, it was not clear even whether tariffs related to PCAs or BCAs. Current bank practices showing minimum and maximum fees for the entire range of products do not allow customers to access meaningful information on specific products.
Article 2 (d) of Circular No. 2006-11 stipulates that banks are required to notify account holders at least 45 days in advance of any changes to fees or other conditions . However, in practice, banks appear not to comply with this obligation. The OECD understands that the BCT is working on a circular with the objective of increasing banks’ transparency vis-à-vis their customers.
Barriers to assessing information
Low levels of searching may be caused by a lack of tools with which to compare information and barriers to shopping around. Information on fees is complex, as banks typically charge monthly fees plus a fee for each banking operation. This makes comparisons across products and providers difficult. To make a realistic comparison of fees (without taking into account products’ other dimensions), consumers would need to combine information on each charge with considerations of their expected use of accounts.
In addition to that, tools to compare prices and other features of current accounts are lacking. Although price comparison websites exist in Tunisia for certain products, none compares the fees and conditions associated with current accounts.
Barriers to closing and switching accounts
Monetary and non-monetary barriers exist to closing and switching accounts.
Regarding monetary costs, the OECD understands that the BCT published in January 2020 a note to prescribe that closing current accounts should be free of charge (see Section 4.4.1) (OIF, 2023[3]). However, some banks continue to charge fees (directly or indirectly) related to account closures.
Based on data provided by the BCT, at least two banks charge fees of up to TND 60 for closing PCAs. Banks also typically charge consumers fees of approximately TND 100‑200 to provide “indemnity letters” showing that customers closing accounts have no outstanding loans with the bank. In addition to representing an extra monetary cost for customers, indemnity letters further delay the closure of accounts because banks’ internal procedures means it may take up to three months for them to issue such letters. This practice also conflicts with Article 2(1) of Circular No. 2006‑12, which prescribes that banks must set maximum time limits for granting indemnity letters and strictly adhere to these time limits.
In addition, banks typically require customers closing accounts to repay loans outstanding to them. To do this, customers must have sufficient funds to repay their loans and associated early repayment charges, which stakeholders say amount to approximately 5% of outstanding capital and represent a significant obstacle to switching.
In terms of non-monetary costs, several practices make the process of closing and switching PCAs burdensome for consumers. Several stakeholders interviewed by the OECD said banks may not close accounts after they receive closure requests from consumers. One reason may be that the remuneration of branch managers is linked to the number of existing current accounts associated with each branch, making branch managers reluctant to close them. Regulation does not seem to require banks to notify customers in a timely fashion when their accounts are closed. Banks do not typically provide documents stating that accounts have been closed, and as a result consumers are charged maintenance fees even after they have requested that their accounts be closed.
Such practices may be in breach of Article 2(1) of Circular No. 2006‑12, prescribing that banks set maximum time limits for closures of accounts and strictly adhere to them. They also clearly conflict with the 2018 code of conduct regulating the relationships between financial institutions and their clients, in particular with what is prescribed for accounts that have remained dormant for at least 12 months. Indeed, Article 33 of the 2018 code of conduct states that banks are required, among other things, to refrain from charging fees and from demanding fee payment, and to inform customers of their year-end balances. The OECD understands that the BCT is considering introducing an obligation for the timely closure of accounts and reinforcing its surveillance powers to ensure compliance and protect customers.
Finally, switching accounts is burdensome and the expected benefits of doing so are low (for example, because of the low degree of price dispersion observed). If consumers want to switch to new accounts, they must close their existing accounts and open new ones with a new provider. Apart from going through the process of opening and closing accounts, and the related barriers discussed above, consumers need to transfer existing direct debits and standing orders and notify their employers and other parties to make future payments to their new accounts. Unlike banks in many countries, Tunisian banks offer no automated switching services.
Although these barriers exist for both individuals and enterprises, MSMEs may find them harder to overcome for several reasons. First, switching accounts may be particularly risky for MSMEs, as they are likely to make and receive many payments, and they need to ensure that these originate from and arrive in their new accounts. Second, given the reliance of MSMEs on financing, some banks’ requests for the early repayment of outstanding loans can be particularly expensive.
MSMEs also face additional barriers to switching accounts. The linkages between BCAs and finance represent an additional barrier to switching. Given that banks rely on in-house information on payment data relating to existing customers to make lending decisions, an MSME switching BCA providers may lose its existing banking relationship and need to establish a new one with a provider that has limited credit information. Thus, the risk of losing access to financing may deter MSMEs from switching BCA providers. The lack of a fully functional credit information market exacerbates this problem, increasing the cost of switching.
Barriers to opening accounts
Two banks interviewed by the OECD said they charged fees to open current accounts online. Banks typically require an electronic signature certified by the Agence Nationale de Certification Electronique (Tuntrust), or National Electronic Certification Agency. This certification costs banks approximately TND 100 and the OECD understands that it is required by law. Banks typically pass this cost on to account holders. Consistent with this, around 95% of Tunisians and 98% of enterprises have opened their current accounts at branches (see Table 7.2). This limits choice for consumers and may represent a significant barrier to access to current accounts, especially for individuals and enterprises in rural areas.
Although there is no legal requirement in this regard, in practice, banks often require new PCA customers to prove they have a regular income and to have it credited to their new account. This has the effect of excluding potential customers that do not have regular incomes. The banking association and several banks argued that this practice is justified by the delays in crediting transactions such as cash withdrawals to accounts, concerns that consumers may withdraw sums larger than their balances, and the limitations of the public credit information bureau, which make checking the credit histories of a large segment of the population – those who have never accessed credit – impossible. As a result, customers without regular incomes are often redirected towards savings accounts, which typically have limited functionality, lacking such facilities as overdrafts or payment cards. This is consistent with the results of the consumer survey, showing that banking penetration is significantly higher among employed individuals (see Table 4.8).
Table 4.8. Banking penetration by job status (selected categories)
Note: Around 2% of respondents had current accounts at both La Poste and a bank.
Source: OECD consumer survey (Q1 and D5, N=2 995).
4.5.5. Conclusions on customer behaviour
Section 4.2 shows that Tunisian bank customers’ high degree of reliance on branches and their limited use of online banking services limit their choices of providers, especially in rural areas, where branches are scarcer. Choosing providers is further constrained by several market practices that prevent customers from accessing meaningful product information, shopping around, closing and switching accounts.
For example, analysis of the websites of Tunisia’s major banks showed that detailed and meaningful information on fees is not available. The only information available includes wide ranges of fees, and they are not product specific. In addition, a lack of tools for comparing fees (such as a price comparison website for current accounts) increases barriers to shopping around.
Significant barriers also exist to closing and switching accounts. Acting on information and the ability to choose providers is essential to well-functioning markets. However, several monetary and non-monetary barriers exist to closing and switching accounts. Several banks charge direct or indirect fees associated with account closures (such as fees for early repayments of outstanding loans or fees for indemnity letters showing a lack of outstanding loans). The OECD also understands that the process of closing accounts is lengthy and uncertain, resulting in customers continuing to pay account fees even after requesting the closure of accounts.
4.6. Payment cards
Several stakeholders interviewed by the OECD shared concerns that the high cost of using payment cards makes consumers reluctant to use them. This section presents the proportion of consumers with payment cards in Tunisia and discusses potential reasons for their low take‑up.
In Tunisia, PCAs do not typically come with a card to make payments. Cards are considered an additional service that consumers can request, and they come at an additional cost. According to the Global Findex Database, the ownership of debit and credit cards in Tunisia is lower than the average in other MENA countries (see Figure 4.8).
Figure 4.8. Debit and credit card ownership in Tunisia compared to selected MENA countries
Note: The MENA countries included are Algeria, Egypt, Iran, Iraq, Jordan, Lebanon, Libya and Morocco. The minimum, maximum and average of MENA countries does not take Tunisia into account.
Source: 2021 World Bank Global Findex Database.
According to the OECD consumer survey, among the one‑third of consumers with PCAs, 47% did not have a card to make payments.43 Among consumers with cards, only 6% used their cards daily, 35% used them at least once a week, 43% used them at least once a month, and 16% used them less than once a month.44 In the 12 months leading up to the survey, 35% used their cards to make purchases and to withdraw money, 43% used them to withdraw money but did not make any purchases, 1% made purchases but no money withdrawals, and 20% neither purchased anything nor withdrew money.45
The consumer survey also asked PCA customers without cards an open question on the reasons they did not have cards, to which 78% of respondents replied that they did not need one and 13% said card fees were too high. Among the respondents, 1% said shops did not accept cards.46 The survey asked the 78% of consumers who said they did not need cards whether they were likely to change their minds in the following certain hypothetical scenarios: 1) a substantial reduction in the cost of cards; 2) closer or more easily accessible bank branches; 3) a substantial increase in the number of shops accepting cards; and 4) if they found a new job. Figure 4.9 shows that one in three respondents said they were likely or very likely to obtain cards if there were a substantial reduction in fees.
Figure 4.9. Proportion of consumers likely to use a card in certain hypothetical scenarios
Source: OECD consumer survey (Q23, N=140).
The evidence described above suggests that high fees discourage a significant proportion of consumers from using cards as a means of payment. Although fees for using cards are set by banks, stakeholders interviewed by the OECD said that high fees were caused by certain market practices in the payment services industry.
Stakeholders indicated that a potential reason for the high cost of cards was that card issuers such as banks sourced cards from a single provider (MS Solution) via the Societé Monétique Tunisie (SMT), which processes card payments. SMT is owned by banks and was set up in 1989 to process interbank electronic payments. Stakeholders told the OECD that the prices charged by MS Solution were high, in the range of EUR 4‑6 per card, which banks typically passed on to consumers. Anecdotal evidence from stakeholders suggests that this cost is significantly higher than in other countries (for example, the same kinds of cards cost around EUR 0.50 in France).
Stakeholders also indicated that the cost of point-of-sale (POS) terminals and readers to process card payments was high. This has an impact on the usage of payment cards, as the high costs of terminals make merchants reluctant to accept cards or more likely to pass on costs. The OECD understands that the largest provider of POS terminals is MS Solution, although other providers have recently been authorised to provide such devices. Stakeholders told the OECD that POS terminals cost around TND 1 000.
4.6.1. Conclusions
This section shows that few Tunisians have payment cards and that even fewer use them regularly. A number of factors contribute to this. First, high costs for consumers, which may be the result of market practices such as using a single provider for some inputs and by the factors identified in Chapter 3 that weaken incentives to compete and increase the extent to which banks pass costs on to consumers. Second, few merchants accept cards, partly due to the high cost of renting the equipment to process card payments. Third, many consumers may have a preference for cash payments.
4.7. Conclusions
Competition in the market for current accounts does not work as well as it could. The choice of providers is limited and determined by availability of branches near customers, which in rural areas may be limited. The analysis found that demand-side engagement with current accounts is low. Very few individuals and businesses shop around and compare fees and conditions when opening current accounts, and very few are aware of how much they are paying in fees once they use them. Switching rates are also low. Limited consumer choice and low demand-side engagement weaken incentives for banks to compete.
The analysis finds that revenues per current account – a proxy for fees – increased at above‑inflation rates between 2016 and 2020, and that fees advertised on banks’ websites increased in nominal terms between 2010 and 2017. This is consistent with an increase in the overall profitability of banks observed in recent years and discussed in Chapter 2. Two-thirds of Tunisians do not have PCAs and one in five said the high fees were a reason why. The shares of supply of current account providers have been stable since at least 2015, which is consistent with a market in which competition is not working well.
The chapter also identifies bank practices limiting the ability of current account customers to make informed decisions. In particular, the limited availability of meaningful information on fees and other conditions, and the barriers to switching accounts when better offers are available blunt banks’ incentives to compete. This is exacerbated by the ineffective mediation mechanism and the risk of co‑ordinated behaviour described in Chapter 3.
References
[4] Anson, J. et al. (2013), “Financial Inclusion and the Role of the Post Office”, http://econ.worldbank.org. (accessed on 20 October 2022).
[2] Banque Centrale de Tunisie (2022), Rapport Annuel sur la Supervision Bancaire - Exercice 2020.
[5] Competition and Markets Authority (2017), Retail banking market investigation, https://www.gov.uk/cma-cases/review-of-banking-for-small-and-medium-sized-businesses-smes-in-the-uk#final-report (accessed on 22 May 2023).
[1] La Poste Tunisienne (2021), Rapport Annuel 2020, https://www.poste.tn/upload/telechargement/fr/Rapport_annuel_2020_FR_ANG.pdf (accessed on 22 May 2023).
[3] OIF (2023), Liste des services bancaires gratuits, https://oif.bct.gov.tn/blog/les-droits-financiers%20/liste-des-services-bancaires-gratuits (accessed on 22 April 2023).
Notes
← 1. OECD consumer survey (Q6, N=1 086).
← 2. OECD MSME survey (Q5, N=804).
← 3. OECD consumer survey (Q20, N=1 086).
← 4. OECD consumer survey (Q10, N=1 086) and OECD MSME survey (Q9, N=804).
← 5. OECD MSME survey (Q9, N=804).
← 6. OECD consumer survey (Q16, N=1 086).
← 7. Meetings with banks.
← 8. OECD MSME survey (Q13, N=804).
← 9. OECD MSME survey (Q27, N=804 and Q28, N=1 005).
← 10. OECD MSME survey (Q27, N=804 and Q28, N=1 005).
← 11. OECD MSME survey (Q27, N=804 and Q28, N=1 005).
← 12. OECD MSME survey (Q27, N=804 and Q28, N=1 005).
← 13. OECD MSME survey (Q27, N=804 and Q28, N=1 005).
← 14. Note that this analysis does not consider that consumers may be using two personal current accounts, however the results of the survey suggest that “multi-homing” is not very prevalent.
← 15. Shares of supply based on the data submitted by the BCT are broadly consistent. Note that the BCT data cannot distinguish between primary and secondary accounts.
← 16. OECD MSME survey (Q3, N=804).
← 17. Note that the number of accounts includes dormant accounts.
← 18. The IPSB measures the change in the price of banking services. It is based on a set of services comprising payment methods (cash, cheques, withdrawals, domestic and international transfers, and bank cards), account maintenance fees, and other services such as online banking.
← 19. OECD consumer survey (Q12, N=1 086). Banks with fewer than 22 observations were excluded from this calculation.
← 20. OECD consumer survey (Q20, N=1 086).
← 21. OECD consumer survey (Q42, N= 1 086).
← 22. The average in MENA countries does not include Tunisia.
← 23. OECD MSME survey (Q1, N=1 005).
← 24. OECD MSME survey (Q1 and S1, N=1 005).
← 25. OECD MSME survey (Q2, N=201).
← 26. OECD MSME survey (Q24, N=201).
← 27. OECD consumer survey (Q42 and Q1, N=1 899).
← 28. OECD consumer survey (Q5 and D1, N=1 086).
← 29. OECD consumer survey (Q7, N=1 086).
← 30. OECD consumer survey (Q15, N=1 086).
← 31. OECD consumer survey (Q33 and Q34, N=1 086).
← 32. OECD MSME survey (Q1 and Q4, N=804).
← 33. OECD MSME survey (Q4 and S4, N=804).
← 34. OECD MSME survey (Q1 and Q6, N=804).
← 35. OECD MSME survey (Q12, N=804).
← 36. OECD MSME survey (Q17 and Q18, N=202).
← 37. OECD MSME survey (Q17 and Q20, N=202).
← 38. OECD MSME survey (Q17 and Q22, N=602).
← 39. OECD MSME survey (Q1 and Q12, N=804).
← 40. OECD MSME survey (Q15, N=804).
← 41. OECD MSME survey (Q16, N=203).
← 42. OECD MSME survey (Q12, N=804).
← 43. OECD consumer survey (Q23, N=1086). The results of the consumer survey are consistent with the results of the World Bank survey. The OECD estimates that 52% of individuals with PCAs also hold either a debit card or a credit card (i.e. 36%*52%=19%) and the World Bank estimates that 22% of Tunisians gold either a debit card or a credit card (i.e. 20%+2%=22%).
← 44. OECD consumer survey (Q28, N=561).
← 45. OECD consumer survey (Q26 and Q27, N=561).
← 46. OECD consumer survey (Q24, N=525).