The impact of economic blockade on the performance of Qatari Islamic and conventional banks: a period-and-group-wise comparison

Qatar’s banking system and the blockade

In the midst of 2017, a number of GCC (Saudi Arabia, Bahrain and the United Arab Emirates) and non-GCC (Egypt) states cut all relations with Qatar and imposed an economic blockade (hereby referred to as “the GCC blockade” or “the blockade”). One of the sectors hit hard by the economic embargo was the banking sector. In the aftermath of the blockade till the end of 2017, the commercial banks in Qatar witnessed bank runs whereby the percentage of non-resident deposits to total customer deposits dropped from 25% to 17%. Deposits from the blocking countries were entirely withdrawn, and other non-blockade depositors demanded higher interest rates. Moreover, the foreign interbank funding to Qatari banks plunged to nearly US$30bn within six months of the crisis (Wheatley, 2018). Such withdrawals left the Qatari banking sector in a liquidity limbo, at least in the short run. The rating agency Moody’s lowered the investment status of Qatari banks from stable to a negative score, citing weaker operational conditions in the banks and continued foreign fund withdrawals amounting to 36% of total liabilities (Shaffer, 2017). Reliance on confidence-sensitive external funds left the Qatari banking sector severely susceptible to liquidity issues in cases where investor sentiments shift. The impact of such large foreign money outflows on the economy was substantial such that Qatar’s gross domestic product (GDP) witnessed a decline of 15% (IMF, 2019). Unsurprisingly, the S&P Global marked Qatari banks as the “most vulnerable” in the region because of the ongoing deterioration fuelled by the boycott, decline in real estate prices and hotel occupancy rates along with serious performance decline over the years before the blockade (Spong, 2018). Despite their unique banking operations, Qatari IBs are not expected to be immune from the liquidity conundrum in the state. Belton and Mehmood (2017) stated that this sector would face similar complications, as nearly a quarter of their borrowings come from the Gulf states.

Nonetheless, in spite of the funding crunch, within a short period of time, Qatari banks and the economy, in general, emerged as a survivor or even in better shape than before. This was because of funds injected by the state into the economy and the pouring of deposits from the public sector, outweighing non-resident withdrawals. Within a month of the embargo, in July 2017, the Qatari Government injected more than US$10bn into the domestic banks so as to offset the impact of foreigners’ withdrawals from the sector (Torchia and Evans, 2017). Kerr (2017), citing Moody’s and Fitch, stated that two months after the blockade the money injected into the economy by the government amounted to about US$40bn from its US$340bn sovereign fund to support the economy and the financial system. Also, the public sector and central bank deposited around US$19bn and US$9bn, respectively. A year after the crisis, non-resident deposits amounting to US$9bn returned to the country, further easing the pressure as Wheatley (2018) said. Quite remarkably, the banks witnessed a positive trend a few months after the blockade. Qatari commercial banks’ assets grew by about 11% on a year-on-year basis, credit portfolio by 13%, domestic loans by 15%, service sector loans by around 3%, consumption loans by near 8% and bank cash and precious metals grew by a whopping 42% (Perumal, 2017). All these developments resulted in growing confidence in Qatar’s financial system, leading Moody’s to upgrade the rating of the state to a stable Aa3 notch, accentuating that the blockade will have an insignificant impact on Qatar’s credit status (Jivrai, 2018).

Research design

The GCC blockade was imposed upon Qatar in June 2017. Periods before the announcement of the blockade will be referred to as “pre-blockade”. The blockade happened to coincide with the end of the first half of the financial year when banks prepare their consolidated interim financial statements. This study starts from the first half of 2015 (1H15) and continues to the end of the second half of 2019 (2H19), representing a semi-annual study. Periods (1H15–1H17) and (2H17–2H19) refer to pre-blockade and post-blockade periods, respectively. It should be carefully noted that “post-blockade” does not imply that the blockade is over; it is still ongoing as this paper is being written. It simply refers to the periods following the imposition of the embargo on Qatar.

Firstly, an FRA is used to measure, describe and analyze the performance of the examined Qatari banks in terms of solvency, profitability, efficiency, liquidity and asset quality from 1H15 to 2H19. FRA is hailed for its effectiveness in distinguishing poorly performing banks from well-performing banks while it also accounts for and controls the effect of bank size on the examined variables of the study (Islam, 2014). FRA is performed on the averages of each ratio of both IBs and CBs in every period.

Secondly, to assess the magnitude of the blockade on the performance of the banks, a t-test is used to test whether differences between pre-blockade and post-blockade means are statistically significant or not. T-tests are widely used to compare the means of a particular variable generated from different sample groups. In this study, the paired t-test is used on a sample of data to measure the means of a variable before the blockade and after the blockade. Such a method produces more precise data as stated by Al-achi (2019). To approve or disapprove the t-test outcome, a critical p-value associated with the means is used. The null hypothesis indicates that the pre-blockade means and post-blockade means are equal (μ₁ = μ₂); i.e. the impact of the blockade on bank performance is insignificant.

Study population

Operating banks in Qatar are 18, including seven domestic CBs, four domestic IBs and seven foreign CBs. However, the foreign banks are excluded from this study as it is believed that their exposure to the blockade is minimal, as their parent banks offer them rescue packages. From the seven domestic CBs, Qatar Development Bank (QDB) and International Bank of Qatar (IBQ) are also excluded. This is because QDB is a development bank rather than a commercial bank, and IBQ has been dissolved to merge into another bank – Barwa Bank (BBK) – to form an IB and, IBQ reports are unavailable. Hence, banks selected for this paper are five domestic CBs and four domestic IBs operating in Qatar throughout the study period.

Data collection

To measure the performance of the banks, this study gathers financial data from the banks’ interim financial statements (secondary source) and then converts them into ratios so as to produce a comparable data set. The data of the banks, which cover 1H15 to 2H19, are accessed through the Qatar Stock Exchange (QSE) database, except for BBK which is not listed on the QSE. For BBK data, the bank’s own website is accessed.

Variables

For the purpose of this study, nine financial ratios are used, which are categorized into five sub-categories:

1. solvency;

2. profitability;

3. efficiency;

4. liquidity; and

5. asset quality.

The ratios are in percentage terms.

Description of the data

To provide a general display of our data set, descriptive statistics are performed. As shown in Table 1 the observation number (Obs) of IBs totaled 40 on all ratios. However, due to missing values, the numbers changed, particularly CAR with 9 missing values. This is mainly due to the lack of 1H16 period related data from IBs. The total Obs of CBs is 50 with one missing value. The mean CAR and EMR of IBs are slightly higher than that of CBs, implying a better solvency approach being adopted by IBs with 17.49% CAR and 13.98% EMR. The average ROA and ROE of IBs show slightly higher profitability, but CBs have the highest ROE of 18.26%. The average expenditure of IBs is substantially lower than that of CBs with a 26.5% OEOI. However, a standard deviation (St.D) of 16.5% shows a broad difference among CBs. The average LDR and LAR of both types of banks are quite similar at 104.28% and 107.42% and 67.30% and 64.33%, respectively, but CBs’ lending approaches vary widely (St.D 16.01% and 9.99%). The average NPL indicates that CBs have on average more bad loans than IBs with 2.84% and 1.39%, respectively. Yet, CBs have a larger LLP ratio than IBs, indicating a larger cushion for absorbing default risks.

Performance of Islamic and conventional banks in Qatar

As seen in Table 2, before the blockade IBs were having slightly fluctuating CAR whereas the CAR of CBs was improving steadily. The EMRs of both types of banks were somewhat close and less fluctuating. After the onset of the blockade, IBs continued maintaining slightly fluctuating CAR and EMR, finishing with 19.0% and 14.0%, respectively in 1H19. On the other hand, CBs continued improving their CAR and had floating EMR, finishing with 18.0% and 12.7%, respectively.

This finding contradicts those of Elsiefy (2013), who found that IBs in Qatar faced declining CAR due to declining profitability and rising risk-weighted assets after the 2007–2008 GFC crisis. Yet, our result confirms that the equity-based financing nature of IBs where investors and banks share risks and rewards provides IBs with additional capital buffers during crises (Rod Erfani and Vasigh, 2018). Profit and loss sharing investors provide a certain proportion of capital coverage to their investments, hence lowering the burden on banks during the crisis.

The EMR of both sectors is quite in the same range, indicating that around 12% to 15% of their assets are financed with shareholders’ money; implying greater reliance on external funds and increased risks. Yet, from a banking perspective, financing with external funds like deposits, which are cheaper than equity, is considered a better cost management approach (Miah and Uddin, 2017). The outcome contradicts Elsiefy’s (2013) finding that CBs increase internal funding of assets after a financial crisis whereby IBs increase external financing in the same period.

Our result also disproves the notion that IBs generally maintain a higher asset-to-equity ratio (low EM) than CBs as shown in prior studies (Babatunde and Olaitan, 2013; Istaiteyeh and Milhem, 2015; Rod Erfani and Vasigh, 2018). The overall values – a cumulated average of the means throughout all the periods – indicate that IBs are slightly more solvent than CBs in Qatar.

As shown in Table 3, both IBs and CBs maintained quite similar profitability trends before and after the imposition of the blockade, with IBs reporting higher ROA ratios than CBs in most periods. In terms of ROE, the mid-year ratios of IBs were substantially lower than year-end ROE ratios, indicating a different profit generation approach than CBs. The mid-year ROE fluctuations were only visible during the pre-blockade period for CBs, yet the banks kept a steady ROE ratio after the blockade. Despite the variations, IBs outpaced CBs in terms of profitability.

Our results confirm Elsiefy’s (2013) finding that CBs in Qatar maintained steadier and more sustainable profit growth than IBs before and after the GFC even though IBs’ profitability was higher. However, they contradict Babatunde and Olaitan’s (2013) conclusion that IBs’ generate lower profits than CBs due to their larger operational costs and bad loan expenses. In 2018, IBs in Qatar saw an 8.5% profit growth as they moved to the application of FAS 30 at the discretion of the Qatar central bank’s (QCB) governor (Gulf Times, 2019). FAS 30 better measures expected credit risks. Qatar’s IBs have also registered a significant growth in assets and revenues that was mainly due to the 37.7% growth in Islamic bonds despite the imposition of the embargo (Gulf Times, 2018). The overall profitability of IBs stood at 1.3% ROA and 9.2% ROE while those of CBs stood at 0.9% ROA and 7.5% ROE, implying better profitability for Islamic banks.

Table 4 provides the results of bank efficiency ratios. Operating efficiency of both banking sectors witnessed fairly steady trends during both periods – particularly IBs – with IBs incurring significantly less operating expenses (24.7%) than CBs (40.5%). This shows that at least one-third of IBs’ operating income went to operating expenses they spent to generate such income while CBs allocate minimally two-thirds of their income on similar expenses. This shows that IBs are more efficient than CBs in Qatar, supporting the findings of many studies that used similar ratios (Ansari and Rehman, 2015; Istaiteyeh and Milhem, 2015; Rod Erfani and Vasigh, 2018). The view of Elsiefy (2013) − that IBs tend to reduce staff costs during a crisis and incur higher costs afterwards to prioritize gaining larger market share as compared to CBs, which tend to lower staff costs over the course − is not supported. The higher profitability of Qatari IBs is the key reason for their substantial efficiency, backed by less funding pressures as they mainly rely on retail deposits (Perumal, 2019a). However, Miah and Uddin (2017) found substantial inefficiency with large variations among IBs in the GCC as compared to CBs’ efficient cost-to-income ratios; the authors linked a larger capital base to the inefficiency of the banks. The overall efficiency (OEOI) averages are 26.5% and 50.0% for IBs and CBs, respectively, implying better cost management by IBs. IBs in Qatar are labelled as the most efficient banks in the entire Islamic finance industry (Perumal, 2015), as well as the most efficient banks in Qatar (Statista, 2019).

The LDR and LAR of the banks are almost equal in most periods but differed in a few periods (Table 5). Both bank groups experienced their least lending activities (lowest liquidity) levels in 2H15 – a period of recovery from the oil price crash in the region. LDR indicates that Qatari IBs and CBs face similar liquidity risks as opposed to other studies, which expressed greater liquidity risk exposures to CBs than IBs as the latter rely less on borrowed funds (Abdulle and Kassim, 2012; Ansari and Rehman, 2015; Rod Erfani and Vasigh, 2018).

Elsiefy (2013) found contrasting results, stating that IBs in Qatar held less liquid assets than CBs during the GFC, and thus were exposed to greater liquidity risk and bank runs. IBs had slightly higher and stable LAR than CBs, which suffered from a declining lending capacity from 66.7% to 62.8% after the onset of the blockade.

The liquidity outcome contradicts Elsiefy’s (2013) findings that IBs consistently violated the QCB’s 90% LDR limit and were more heavily engaged in lending than CBs in Qatar. Such behaviour would pose significant liquidity risks on banks on one end but larger returns on the other. On the other hand, Abdulle and Kassim (2012) and Istaiteyeh and Milhem (2015) revealed that IBs had lower LAR than CBs during the GFC, which lowered their exposure to liquidity risks. The overall average LDR and LAR of IBs are 103.5% and 67.2%, respectively, while those of CBs are 107.3% and 64.5%, respectively. This implies broader liquidity issues in the banking industry and infringement of the QCB’s guidance on a 100% limited LDR (IMF, 2019).

In regard to asset quality depicted in Table 6, it is evident that the bad loans (NPL) backed by loan reserves (LLP) of IBs are less than those of CBs, implying that IBs have less risky and better asset quality. These results are similar to Abdulle and Kassim’s (2012) findings. IBs’ NPL grew steadily, reaching 2.1% at the end of 2019. Provisions for such loans fluctuated disproportionately over the periods, despite the fact that they grew heavily post-blockade. Similar to IBs, bad loans of CBs grew constantly over the period, supported by the similar accumulation of LLP. Both banking sectors experienced significant LLP expansion from 1H18 to 2H19, indicating prudent approaches taken by the banks.

These trends contradict Elsiefy’s (2013) discovery that IBs in Qatar managed to decrease bad loans during and after the GFC as compared to CBs which experienced higher levels of bad loans (NPL). However, this research supports the author’s revelation that CBs maintained more adequate coverage for impaired loans as compared to IBs. One attributable motive can be the debt-based business structure of CBs.

In terms of LLP, Miah and Uddin (2017) discovered similar results to our findings and attributed higher LLP ratios to the lower profitability of CBs in the GCC. The overall NPL of IBs is 1.3%, less than half the 2.8% NPL of CBs. On the other hand, the overall LLP of CBs is 126.7%, being substantially higher than IBs’ 90.1% LLP.

Qatar’s banking industry, thus, maintained high asset quality (determined by low NPL and adequate LLP) and was capable of withstanding larger bad loans (IMF, 2019).

Significance of the blockade

To test the significance of the blockade on the Qatari banking sectors, a statistical t-test is performed. As shown in Table 7, the impact of the crisis on IBs’ stability was insignificant as the p-value does not meet the significance levels on both CAR and EMR. This can be attributed to IBs’ slightly higher solvency levels in both periods. As for CBs, the impact was insignificant in terms of EMR but significant in terms of CAR at 0.000% p-value, evidenced by the CAR jump after the blockade. This could be due to the QCB’s imposition of new rules on CBs. From 2017 onwards, CBs in Qatar were required to switch from IAS 39 guidelines to IFRS 9, which determines CBs’ classification and measurement of assets, liabilities and derivatives (QCB, 2018). Due to a possible rise in impairment volatility stemming from IFRS 9, banks tend to raise capital buffers (Deloitte, 2016). Regardless of the stability trends of both sectors, the overall Qatari banking system is stable. Only a very severe economic downturn would be expected to put immense pressure on the capital of banks (IMF, 2018).

In terms of profitability, the impact of the crisis on IBs and CBs is insignificant, implying that these banks managed to maintain their ability to generate profits despite the rift. Ibrahim (2015) found significant profitability for IBs in Qatar in terms of ROA against CBs during 2010–2014, possibly due to high net financing on the IBs’ side. However, our finding is in line with other studies stating differences in profitability between banks is insignificant during financial meltdowns (Ansari and Rehman, 2015; Istaiteyeh and Milhem, 2015). One way that the banks limited the effect of the crisis was through the domestic interbank market. In the aftermath of the crisis, the volume and depth of liquidity increased substantially, offering banks liquidity to optimize returns and enhance profitability (QCB, 2018). Qatari IBs – a sector considered by analysts as the most vulnerable to the crisis – focussed on the domestic market to limit exposure from the diplomatic rift and achieved a 14.5% profit growth one year after the embargo (Group Oxford Business, 2020). Nonetheless, the profitability trends of the overall Qatari banking industry remained strong amidst the crisis, enjoying constant profit growth post-blockade (IMF, 2019).

As per the t-test, the blockade failed to significantly impact both banking groups’ efficiency levels (OEOI), thus accepting the null hypothesis of no impact. The outcome is supported by previous studies that found similar results (Ansari and Rehman, 2015; Istaiteyeh and Milhem, 2015), but it contradicts the notion that financial crises influence the operating costs of banks as presented by others (Babatunde and Olaitan, 2013; Elsiefy, 2013). Rod Erfani and Vasigh (2018) found fluctuations in CBs’ efficiencies pre-and-post GFC, but stable efficiency in IBs in the same periods – which is quite the same in our case. Remarkably, the efficiency of the banking sector was higher (i.e. lower inefficiency index) in 2018 compared to the two previous years, due to reduced administrative costs and improved profitability (QCB, 2018). The sector further displayed superior efficiency in 2019 to that of GCC competitors because of core earnings growth and better cost management (Perumal, 2019b).

The liquidity t-test reveals, with insignificant p-values, that the blockade had no tangible effect on both bank groups’ LDR and LAR. Our finding is contrary to Elsiefy’s (2013) findings, which showed that the GFC impacted the credit growth of banks in Qatar, it is lower after the crisis than before it. IBs generally face more liquidity challenges than CBs as they are mostly exposed to scarce liquidity risk management instruments along with uncertain returns offered to profit-and-loss sharing investors, thus limiting investor preferences (Hasan and Dridi, 2011). The credit growth of Qatari IBs took a hit in 2017 with a decline in assets, deposits and net financing (loans), pushing the sector to switch credit channels (Gulf Times, 2019). Remarkably, however, IBs outpaced their conventional peers with credit growth of 13% in mid-2019 against 4% for the latter (Perumal, 2019c). The impact on the lending capacity (LAR) of Qatar banks was very minor given the swift injection of funds into the system by the State. In 2018, Qatari banks raised a combined amount of US$1.8bn in bonds, of which US$800m was raised by IBs from private and public bonds (ṣukūk) and US$1bn by CBs from private sources (Barbuscia and Arnold, 2018).

The tests of NPL and LLP show that the blockade significantly affected the quality of both bank groups’ assets, despite the dissimilar operational nature of IBs and CBs. According to Hasan and Dridi (2011), IBs’ and CBs’ exposures to various economic segments are similar with few exceptions. Our significant outcome is supported by Elsiefy (2013), who presented a significant change in the asset quality of both bank groups in Qatar before and after the GFC in 2007–2008. The asset quality of IBs (in the entire GCC) is slightly weaker than CBs as they are more prone to risks from the real estate sector due to the asset-backing principle of Islamic finance (Augustine, 2018). According to Abdalla et al. (2019), an oversupply of real estate in Qatar, accompanied by reduced tourism and occupancy rates due to the blockade, created real pressure on the assets of banks, as property prices plummeted and contractors increasingly delayed their payments to banks. As per the authors, banks (largely CBs) with the highest exposures to real estate reported the highest NPL ratios. Hasan and Dridi (2011) found similar significant exposures of Qatari banks to the real estate and construction sectors during the GFC, unlike other regional neighbours. Unlike in the case of the GFC, however, the real sector underwent swift recovery as Ataullah (2017) asserted that less than 100 days after the blockade, this sector was performing stronger than before. Other factors that worsened bank asset quality (higher NPLs) include: deteriorated debtor creditworthiness and tightened liquidity conditions, partly due to an outflow of non-resident funds, escalating investors’ concerns over the financial sector’s health (IMF, 2018). But the return of non-resident and foreign bank funds enabled the banks to boost private-sector lending and geographical diversification of foreign deposits (IMF, 2019).