Islamic Finance in Movement - Public Opinion in the Arab Region

Transcription

1. Islamic Finance in Movement Public Opinion in the Arab Region ARAB BAROMETER WORKING PAPER NO . 4 July 11, 2016 Clement M. Henry
2. Introduction Islamic banking has been developing since the mid-1970s in peaceful competition with conventional banking . It has made significant progress in the Middle East and North Africa (MENA), where Arab Barometer surveys indicate widespread disapproval of interest-based lending and conventional banks in general. This paper will examine these data in order better to understand the characteristics of Islamic finance’s potential clientele. This study is undertaken with the hope that this financial movement may mobilize and educate people so as gradually to eliminate the informal economies that exploit them. It may encourage responsible entrepreneurship by bringing informal actors into a renovated formal banking system. Others have noted in cross-national studies comparing Muslim and non-Muslim financial behavior that religion is a significant impediment to inclusion in conventional banking systems (Demirguc-Kunt et al, 2013). In societies distrustful of or rejecting conventional interest-based banking Islamic finance can perhaps lure the informal economic actors out into the open by financing small and medium enterprises and even engaging in microfinance, which is much neglected in the region (El-Gamal et al, 2011; El-Komi and Gerson, 2013).1 Evidently sharia-compliant financial assets are steadily increasing as a percentage of commercial banking assets across most of the countries in the Muslim world where data are available. Driven by demand from wealthy Gulf investors, Islamic finance has steadily gained market share in many Arab countries, including Syria. Led by Malaysia, it is also gaining traction in Indonesia and other non-Arab countries having Muslim majorities. Even states with Muslim minorities, such as the United Kingdom, have Islamic banks catering to them. The grand total in the world of sharia-compliant financial assets, approaching $2 trillion, still, however, does not add up to those of a major American or Chinese bank.2 And although Islamic banking continues to grow more rapidly than conventional banks in most Arab countries, the skeptic may ask whether it really mobilizes new clienteles or simply diversifies the portfolios, as in the Gulf, of wealthy investors. This paper will examine the evidence, focusing on selected under-banked countries where Islamic banking may be attracting new participants. But before analyzing the market share data, it is useful to examine Arab survey data to discover the social and political characteristics of the people who might be most attracted to Islamic finance. Potential clients Arab Barometer surveys conducted in 2007-08, 2011, and 2013 indicate that large majorities of representative national samples across the Arab world believe that conventional bank interest is contrary to the teachings of Islam.3 Table 1 summarizes the information, including only those who agreed or disagreed with the proposition that banks charging interest contradict the teachings of Islam. In the second wave of surveys, conducted in 2011, respondents were also asked whether, “in order to meet the demands of the modern economy, banks should be allowed to charge interest.” While the responses were significantly correlated with the question about the teachings of Islam, the relationship was weak (r= -.162, p<.000): a fair number of respondents who had agreed that interest was forbidden in Islam also thought that banks should be allowed to charge interest regardless of the religious teachings.4 The final two columns of Table 1 therefore report the percentages of those really interest-averse people who both agreed that interest was haram and also that banks should not charge interest.
3. Table 1 : Arab Attitudes toward Bank Interest Interest is: Algeria Egypt Iraq Jordan Kuwait Lebanon contrarytoteachingsof Islam 2006-2008 (n=) 2011 (n=) 88.50% 1123 86.30% 980 72.30% 1126 80.00% 1059 85.60% 1038 87.20% 1130 75.80% 650 64.80% 938 68.90% 1202 Morocco Palestine SaudiArabia Sudan Tunisia Yemen 86.00% 1193 85.10% 1221 85.50% 76.70% 64.20% 85.10% 81.30% 966 76.80% 1152 1077 1299 957 1037 Average 81.90% 7129 77.80% 11019 2013 89.90% 70.80% 64.70% 88.90% 77.40% 71.70% 82.60% 90.50% 84.10% (n=) 1124 937 994 1737 983 1038 1149 1000 1154 60.20% 79.50% 80.70% 79.10% never OK 2011 (n=) 51.60% 854 33.50% 1086 60.10% 939 47.10% 1109 27.40% 1115 1126 1096 1087 61.30% 63.60% 41.60% 42.40% 54.10% 1115 992 1224 817 938 13425 47.80% 10189 This paper assumes that these consistently interest-averse respondents would also be the most likely potential participants in Islamic forms of financial intermediation. They constitute sizable blocks of public opinion in what are supposed to be reasonably representative samples across the Arab world. Accordingly the present analysis will focus on the findings of this second wave of public opinion surveys, since the other waves unfortunately had no independent check on the behavioral implications of the very widespread belief that bank interest is a religiously unacceptable practice. Here the second wave locates the potential clients as well as practitioners of Islamic finance from surveys conducted in 2011 in Algeria, Egypt, Iraq, Jordan, Lebanon, Palestine, Saudi Arabia, Sudan, Tunisia, and Yemen. Correlates of Interest Rejection Who then might these potential clients be? As a first step are they more likely to be rich or poor, old or young, male or female, live in a city or in the countryside, be relatively well educated or not? And how significant an influence is the national environment, which may also shape the culture and financial perceptions? Do national education systems have different impacts upon people’s attitudes toward bank interest? And what about religion? Presumably Christians, although welcome in the community of Islamic finance, may not share the same degree of commitment to interest-free banking as Muslims. And among the latter, does one’s religiosity or degree of religious commitment make a difference? And what about one’s politics? In 2011 the Arab world was exceptionally open to surveys of public opinion that featured political questions. Did it make a difference in Egypt, for instance, if one was a Muslim Brother or in other ways committed to political Islam, or were attitudes toward interest-based banking independent of
4. political commitment and consequently more resistant , one might suppose, to the political changes occurring in 2013? The correlates of unconditionally rejecting conventional interest-based banking (“Haram Interest-based banks”) are reported in Table 2. Among the Palestinians, as in the other Table 2: Correlates of "Haram Interest-based banks" rural N= -.028** 9074 rural-Palestine N= age N= gender N= educationlevel -.061* 1115 0.011 10150 -.027** 10189 .054** Politicalinterest N= MonthlyUSD income N= Westernexposure N= Islamism N= Christian -0.009 10149 .071** 7951 -.061** 10021 .093** 10033 -.170** N= N= 9327 8229 educationlevel Religiosity -0.063 .078** (Tunisia) N= N= 814 9310 Womenathome Religiosity .136** .136** N= (Tunisia)N= 10167 808 **.Correlationissignificantatthe0.01level(2-tailed). *.Correlationissignificantatthe0.05level(2-tailed). SaudisandYemeniswerenotaskedwhetherMuslim,Christian,etc. population samples those rejecting conventional interest-based banking tended to come from cities rather than villages (including refugee camps in Palestine). Age did not make a significant difference, but males seemed slightly but significantly (p< 0.01) more likely to reject the conventional interest-based banks than females, as did those having higher levels of education and income. In Tunisia, however, education was negatively correlated, perhaps reflecting Bourguiba’s legacy of modernization. Not surprisingly, the Christians, for the most part in Lebanon and Egypt, were also less likely to reject conventional banking, as were people who had traveled abroad in the past five years. Those who harbored conservative values about women’s social roles also tended to reject interest-based banking,5 as did those whose responses to a series of questions about their religious practices ranked them high on a religiosity scale.6 Interest in politics7 hardly made any difference but what is taken from the survey to indicate political Islamism did make a difference: Islamism, which correlated positively with religiosity (r= 0.215, p< 0.001) and with conservative values about women (r=0.339, p< 0.001), was also significantly associated with opposition to interest-based banking. “Islamism” was derived from correlated responses to six items designed to tap an “Islamist-secular” dimension.8 However, interest in politics was negatively rather than positively correlated with Islamism (r= -0.033, p< 0.001) whereas it was significantly associated with religiosity (r=0.135, p< 0.001) and even more so in
5. Tunisia (r=0.212, p< 0.001), albeit not with “Women at home.” Interest in politics, in other words, was weakly associated with the more secular perspectives connoted by the items detailed in footnote 5. These “secular” attitudes, however, were by no means incompatible with religious commitment; possibly they reinforce one another. To disentangle the possible effects of these correlates, some multivariate analysis is attempted here. A logistic regression is performed in which “Haram Interest-based Banks,” the probability of rejecting conventional banking, is regressed on the entire set of correlates except religious affiliation (so as not to exclude the Saudi and Yemeni data). By including education and religiosity, however, it was necessary to exclude Tunisia, where these variables were coded differently. The Tunisian data are therefore analyzed separately (see below, Table 4). The respondents’ country of residence (except Tunisia, and the baseline case of Jordan) is added to the regression as a dummy variable, and the results are reported in Table 3. Table 3: Logistic regression of “Haram Interest-based Banks” Urban Age Gender Educationlevel Income$100s TimeinWest Religiosity Womenhome Islamism Politicalinterest Algeria Egypt Iraq Lebanon Palestine Sudan Yemen SaudiArabia Constant B 0.203 0.006 0.018 0.078 0.009 0.174 0.263 0.39 0.095 -0.05 0.294 0.593 0.56 0.683 0.689 0.093 0.245 0.603 2.061 S.E. 0.06 0.002 0.054 0.018 0.003 Wald 11.325 7.237 0.106 19.426 7.636 0.034 0.043 0.054 df 1 1 1 1 1 Sig. 0.001 0.007 0.744 0 0.006 Exp(B) 1.225 1.006 1.018 1.081 1.009 26.02 36.867 51.888 1 1 1 0 0 0 0.84 1.301 1.477 0.053 0.03 0.133 3.244 2.772 4.861 1 1 1 0.072 0.096 0.027 0.909 0.951 1.342 0.102 0.099 33.617 32.062 1 1 0 0 0.553 1.75 0.111 0.092 37.557 56.607 1 1 0 0 0.505 1.992 0.11 0.106 0.126 0.708 5.316 22.839 1 1 1 0.4 0.021 0 0.912 1.278 1.827 0.267 59.602 1 0 0.127 This summary view presents the odds in the final column of categorically rejecting interest-based banking. The first row, for example, indicates that the odds of a respondent living in an urban
6. environment (compared to a village or, in the case of the Palestinians, refugee camps as well), are 1.225 to 1, holding all the other variables constant. In other words living in a city increases the odds by 22.5 percent. Conversely a villager is more likely not to reject interest-based banking. The finding is statistically significant: r< 0.001. Age also seems to make a difference. Although one added year of age increases the odds of rejecting interest-based banking by only 0.6 of a percentage point, ten years would increase them to 6 percent, statistically significant at the .01 level (p= 0.007). Gender, however, is no longer even statistically significant (as in Table 2) when all the other variables are taken into account. Education and monthly income, however, retain their statistical significance and deserve further discussion below. Time spent over the past five years in the West is also associated, albeit negatively, with rejection of interest-based banking. One more point of time spent in the West on a four point scale decreases the odds by 16 percent that the traveler will reject this Western form of financing. Religiosity and conservative attitudes about women’s place in the world, on the other hand, are both strongly associated with the rejection of interest-based banking. When all the other variables are taken into account, however, political Islamism loses the positive correlation with rejecting interest-based banking that was reported in Table 2. The correlation reported in Table 2 is reversed because religiosity and conservative values connoted by “Women home” have the critical independent effects. Pious conservative Muslims who also believe in a separation of religion from politics will be even more opposed to interest-based banking than pious conservatives who also conflate religion and politics. This may raise the prospects for Islamic finance in countries like Egypt where political Islam has suffered losses but where the public remains conservative and highly religious. Whatever the fortunes of political Islamism, sympathy for Islamic finance may endure, as it seems protected by these traditional values. Do they vary, however, depending on other country level variables, including the local culture and politics? The most striking finding in Table 3 is that after taking account of the other variables there are still enormous differences in banking preferences among residents of the different countries. With the exception of Sudan, all of the dummy variables are statistically significant. That is to say that the country of residence, presumably its culture and institutions, apparently help to shape these preferences. Compared to the baseline case of Jordan (selected because it is closest to the general average reported in Table 1), even those living in the neighboring territories of occupied Palestine display remarkable differences. The odds become almost two to one that the Palestinian living under occupation will reject conventional interest-based banking (after taking all the other variables into account). Table 3 also shows that living in Algeria makes one 34 per cent more likely to oppose this Western practice than in the base line case of Jordan. And, as in Palestine, the odds soar to almost two to 1 in Saudi Arabia, where virtually all retail banking is sharia-compliant. In Egypt and Lebanon, on the other hand, with their Christian minorities, the odds decline to about 1 to 2.
7. Country specific effects of income and education Interactions between country of residence , on the one hand, and education and income, on the other, may offer a better understanding of the national variations in attitudes toward interestbased banking. It has already been noted in Table 1 that the correlation between these attitudes and the respondent’s educational level changed sign in Tunisia, where more education was associated with less distrust of interest-based banking. Table 4 applies the logistic model developed in Table 3 to Tunisia. Table 4: Logistic regression of “Haram Interest-based Banks” (Tunisia only) Urban age gender educationlevel Income$100s TimeinWest Religiosity womenhome Islamism Politicalinterest Constant B -0.193 0 0.061 -0.015 -0.009 0 0.319 0.186 -0.025 -0.038 -1.526 S.E. 0.192 0.007 0.184 0.065 0.013 0.109 0.114 0.208 0.168 0.105 0.821 Wald 1.014 0 0.11 0.05 0.452 0 7.798 0.796 0.022 0.131 3.456 df 1 1 1 1 1 1 1 1 1 1 1 Sig. 0.314 0.996 0.74 0.823 0.501 0.996 0.005 0.372 0.883 0.718 0.063 Exp(B) 0.824 1 1.063 0.985 0.991 1 1.375 1.204 0.976 0.963 0.217 There are other differences between Tunisia and the other Arab countries. In Tunisia urban residence is associated with less, not more opposition to interest-based banking (although the result is not statistically significant), and time spent in the West makes no difference, perhaps given the exposure to Western values inside the country, a legacy of Habib Bourguiba built on the country’s Mediterranean identity. In Tunisia the only variable that is statistically significant is religiosity. But changes in sign of both household income and education suggest the need to examine the interactions of these variables with the other national environments as well. Table 5 shows the additions to Table 3 that specify country interactions with education and income.
8. Table 5 : Logistic regression with country specific correlates of education and income Urban Age gender education(Jordan) Income$100s(Jordan) TimeinWest Religiosity womenhome Islamism Politicalinterest Education_Algeria Educ_Egypt Educ_Iraq Educ_Palestine Educ_Lebanon Educ_Yemen Educ_Saudi Educ_Sudan Income$100s_Algeria Income$100s_Egypt Income$100s_Iraq Income$100s_Palestine Income$100s_Lebanon Income$100s_Yemen Income_$100sSaudi Income$100s_Sudan Algeria Egypt Iraq B 0.22 0.005 0.001 0.133 0.004 0.172 0.258 0.381 0.095 0.054 0.015 0.089 0.226 0.082 0.161 0.087 0.115 0.116 0.015 0.054 0.025 0.005 0.032 0.022 0.032 0.029 S.E. Wald df Sig. Exp(B) 0.062 12.741 1 0 1.246 0.002 5.261 1 0.022 1.005 0.463 -0.39 1.211 0.342 1.828 0.261 2.224 0.283 18.357 0.055 0.048 0 7.572 1 0.989 1 0.006 0.999 1.142 0.01 0.159 1 0.69 0.996 0.035 24.597 0.044 34.337 0.055 48.442 1 1 1 0 0 0 0.842 1.294 1.463 0.053 3.212 1 0.073 0.909 0.03 3.153 1 0.076 0.948 0.08 0.038 1 0.846 0.985 0.061 2.103 1 0.147 0.915 0.068 10.949 1 0.001 0.798 0.068 1.463 1 0.226 0.921 0.074 0.069 4.724 1.561 1 0.03 1 0.212 0.851 1.091 0.072 0.069 2.515 2.77 1 0.113 1 0.096 0.892 1.122 0.033 0.041 0.016 0.015 0.015 0.027 0.011 0.026 0.209 1.694 2.357 0.139 4.599 0.653 7.843 1.216 1 1 1 1 1 1 1 1 0.648 0.193 0.125 0.71 0.032 0.419 0.005 0.27 0.985 1.055 1.025 0.995 0.968 0.978 1.033 0.972 1 0.176 1 0.136 1 0 1.589 0.677 3.357
9. Lebanon Palestine Sudan Yemen SaudiArabia Constant 0 .407 1.064 0.492 0.123 0.429 2.116 0.324 1.574 0.286 13.877 0.313 2.468 0.319 0.15 0.337 1.621 1 0.21 1 0 1 0.116 1 0.699 1 0.203 0.329 41.305 1 0 1.502 2.898 0.612 0.884 1.536 0.12 Education for the baseline case of Jordan remains positively correlated with opposition to interest-based banking, whereas monthly income loses any statistical significance there. Turning to other national environments, however, education reduces the potential commitment to Islamic finance except in Sudan and Yemen: in Iraq a one level increase in education reduces the odds of rejecting conventional interest-based banking by about 20 percent. In Lebanon, too, education reduces the commitment significantly. By contrast, education in Sudan and Yemen is associated with slight, albeit statistically not very significant (p< 0.1 in Sudan) increases in opposition to conventional banking practices. As for household income, where there are wide average per capita differences across Arab countries, two of the wealthier ones show statistically significant associations between family income and the potential commitment to Islamic finance. They work, however, in opposed directions. In Saudi Arabia a one-hundred dollar increase in monthly income increases one’s odds by about 3 percent, and the effects appear to be statistically significant (p=0.005), suggesting that portfolio diversification might be driving Islamic finance.9 In Lebanon, by contrast, the odds decrease by small but statistically significant amounts. Possibly wealth has different effects in Lebanon because much of it is held in Christian hands. In the sample the average monthly income of the 478 Christians was over $1500 whereas the 633 Muslims averaged $1143. What impact, then, does being a Christian have on the odds of rejecting conventional banking? Table 6 examines the available data, adding religion to our list of variables at the expense of losing the data from Yemen and Saudi Arabia, where the question was not asked.
10. Table 6 : Logistic Regression of “Haram” adding Christian variable Urban Age Gender Educationlevel (Jordan) Income$100s(Jordan) TimeinWest Christian Religiosity womenhome Islamism Politicalinterest B S.E. 0.278 0.068 0.004 0.002 0.006 0.062 0.12 0.049 0.003 0.01 0.203 0.041 1.492 0.183 0.269 0.467 0.037 0.006 Education_Algeria 0.003 Educ_Egypt 0.094 Educ_Iraq 0.218 Educ_Palestine 0.077 Educ_Lebanon 0.183 Educ_Sudan 0.124 Income$100s_Algeria 0.016 Income$100s_Egypt 0.075 Income$100s_Iraq 0.024 Income$100s_Palestine 0.003 Income$100s_Lebanon 0.018 Income$100s_Sudan 0.028 Algeria 0.446 Egypt -0.42 Iraq 1.146 Lebanon 0.757 Palestine 1.021 Sudan 0.569 Constant 1.025 Wald df Sig. Exp(B) 16.815 1 0 1.321 2.802 1 0.094 1.004 0.009 1 0.926 0.994 6.118 0.07 1 0.013 1 0.791 1.128 0.997 24.939 1 0 0.816 66.831 1 0 0.225 0.049 0.063 0.062 0.035 29.87 54.803 0.354 0.031 1 0 1 0 1 0.552 1 0.86 1.308 1.596 0.964 1.006 0.081 0.062 0.069 0.069 0.076 0.07 0.001 2.29 10.056 1.271 5.767 3.173 1 1 1 1 1 1 0.971 0.13 0.002 0.259 0.016 0.075 0.997 0.911 0.804 0.926 0.833 1.132 0.034 0.041 0.016 0.015 0.016 0.026 0.241 3.279 2.158 0.05 1.309 1.123 1 1 1 1 1 1 0.623 0.07 0.142 0.822 0.253 0.289 0.984 1.078 1.024 0.997 0.982 0.973 0.345 0.264 0.285 0.336 0.288 0.315 1.671 2.535 16.183 5.092 12.576 3.257 1 1 1 1 1 1 0.196 0.111 0 0.024 0 0.071 1.562 0.657 3.147 2.132 2.777 0.566 0.413 6.161 1 0.013 0.359
11. Wealth still seemed in Lebanon to diminish rather than increase opposition to conventional banking as in Saudi Arabia , but the negative correlation was weak and statistically insignificant (p< 0.253). In general, Christians are obviously more likely to distinguish between bank interest and usury and to tolerate conventional banking practices. Controlling for the other variables in our model, the odds of the Christian opposing the conventional practices are reduced to .225 to 1, in other words to less than 1 to 4. Still it may be worth mentioning that 11.6 percent of the 490 Christians in the sample did in fact oppose interest-based banking (compared to 47.3 percent of the Muslims, excluding Saudi Arabia and Yemen). Education in Lebanon, where most of the Christians were concentrated, also independently reduced opposition – lowering the odds by almost 13 percent per level of education. Yet, once all these variables are taken into account, the residual effects of residing in Lebanon work in the other direction, increasing the odds of opposing conventional practices by more than 2 to 1 compared to Jordan. Only Iraq and Palestine presented higher odds of rejecting conventional banking when all the other variables in the model are taken into account. Country specific interactions with ideational variables How do our ideational variables interact in the different national contexts? So far it seems that religiosity and conservative attitudes toward women are the two ideational variables most closely associated with rejection of conventional financial practices. But these spectra of opinion, along with their correlate of “Islamism,” also display significant national variations. Might it follow that religiosity and conservative values have different effects, depending on their national cultural contexts? Or do they interact in more or less the same way, despite national differences? We have already seen in Tunisia (Table 4) that religiosity and “women at home” increase the odds of rejecting interest-based banking pretty much as they do elsewhere in the aggregate (Table 3) despite the cultural transformations of the country over the past century. To assess the other country specific interactions in the sample, while keeping the Muslim-Christian variable where it is relevant, we subdivide the sample by country. Table 7 presents the findings (abbreviated just to present the statistical significance and the odds offered by each variable, controlling for the others) for each country. The results suggest that the interactions between religiosity, conservative attitudes toward women, and Islamism-secularism do indeed vary across national contexts. As noted earlier, when controlling for the other variables, “Islamism” correlates negatively (and therefore secularism positively) with rejecting interest-based banking in Tunisia and in the region as a whole. In Algeria, Egypt, Palestine, and Yemen, however, the correlations are reversed. In Algeria an increase of one point on the four point scale ranging from secularism to Islamism is correlated with an increase of over 50 percent in the odds of rejecting these banks, controlling for the positive effects of religiosity and conservative values about women. While perhaps due to sample size the result was not statistically significant, Algeria also strongly contradicted another general finding, about the weak negative relationship between interest in politics and rejection of conventional interest-based banking. Elsewhere, except in Egypt under a Muslim Brotherhood government, an interest in politics had virtually no relationship to attitudes about finance. The potential clientele for Islamic finance generally appears to be politically apathetic. In Algeria, however, the potential clientele can be mobilized. Conservative values, religiosity, and Islamism march hand in hand with political interest in rejecting conventional banking. In Egypt, too, at
12. least in 2011 , before rise and the fall of President Mursi, the Muslim Brothers seemed to be cultivating a political audience for Islamic finance.10 Table 7: Logistic Regressions by country Urban Age Female Education Income$100s TimeinWest Religiosity womenhome Islamism Political interest Christian Constant Algeria Egypt Iraq Jordan Sig. Exp(B) Sig. Exp(B) Sig. Exp(B) Sig. Exp(B) 0.488 1.248 0 1.733 0.838 1.036 0.706 1.068 0.39 1.01 0.086 0.99 0.709 1.002 0.351 1.005 0.927 1.027 0.674 1.073 0.441 0.889 0.017 0.717 0.103 1.16 0.35 0.958 0.178 0.93 0.029 1.121 0.46 0.972 0.085 1.078 0.059 1.027 0.813 0.997 0.008 0.632 0.001 0.418 0.056 0.71 0.592 0.961 0 3.246 0.2 1.231 0.01 0.682 0 1.472 0.17 1.433 0.539 0.904 0.001 1.802 0 2.236 0.189 1.528 0.493 1.105 0.273 0.838 0.705 0.946 0.008 1.654 0.027 1.245 0 0.684 0.908 1.009 0.004 0.159 0.387 0.319 0.216 0.554 0 0.003 0.388 2.705 0.06 20.608 0.003 0.073 Lebanon SaudiArabia Sudan Yemen Sig. Exp(B) Sig. Exp(B) Sig. Exp(B) Sig. Exp(B) Urban 0.41 1.438 0.361 0.717 0.029 1.497 0 1.258 Age 0.51 0.996 0 1.029 0.032 1.021 0.039 1.004 Female 0.01 1.585 0.448 0.857 0.076 0.708 0.185 1.072 Education 0.21 0.925 0.591 1.036 0 1.31 0 1.083 Income$100s 0.17 0.983 0 1.02 0.256 0.973 0 1.019 TimeinWest 0.10 0.848 0.961 0.996 0.098 0.822 0 0.827 Religiosity 0.01 1.289 0.013 1.559 0.726 1.066 0 1.22 womenhome 0.14 1.297 0.001 0.535 0.011 1.699 0 1.596 Islamism 0.01 0.668 0.071 0.749 0.27 0.794 0.002 1.16 Political interest 0.87 1.015 0 0.646 0.803 0.973 0.035 0.942 Christian 0.00 0.163 Constant 0.52 1.744 0.151 3.861 0.031 0.09 0 0.079 The Egyptian survey also included questions about party affiliation. Of the 1219 Egyptians only eight were actual members of a political party but an additional 95 intended to join one. Of these only 46 expressed opposition to banks. Among future recruits to the Freedom and Justice Party (Muslim Brotherhood) opinion was divided, with 15 opposed, 11 in favor, and two on the fence, not expressing consistent responses to the questions concerning interest-based banks. Prospective members of the Tunisian Nahda also expressed mixed attitudes, as did ostensibly secular party sympathizers in both countries. Palestine and Yemen were the only other national
13. environments in which Islamism increased the odds against interest-based banking ; but they were not accompanied by any interest in politics – in Yemen an interest in politics is significantly associated with diminished odds! In Saudi Arabia interest in politics is associated with even lower odds for rejecting conventional banking than in Yemen. Age, wealth and religiosity seem to be the only strong correlates, whereas conservative attitudes toward women and “Islamism” work in the other direction. Given the pervasive religious culture and high penetration of Islamic finance in the kingdom, politics and “Islamism” actually lower the odds against conventional banking. The more significant finding is that age, wealth, and religiosity are all strongly and significantly correlated with the rejection of interest-based banking. Wealthy, pious senior citizens seem most ready to embrace Islamic finance, driving their portfolio diversification. Such preferences, possibly shared by other wealthy senior citizens of the GCC states, nay indeed be driving the expansion of Islamic finance in the region and beyond. The conclusion, then, is that there is indeed a significant potential clientele in the Arab world for Islamic banking, and that it is largely the conservative, pious segments of the populations that are most favorably inclined. The survey confirms earlier impressions that Islamic finance represents an alliance between mainstream ulama, coopted to monitor compliance with Islamic jurisprudence, and conservative Muslim investors (Kahf, 2005). It also draws attention to the special opportunities for Islamic finance in countries where conventional banking is soundly rejected and where there may be possibilities for political mobilization. Countries that are generally underbanked and little penetrated by Islamic finance may offer the greatest market opportunities. Rising Market Shares of Islamic Banks The market share data reported in Table 8 trace the rise of Islamic banks in the MENA region and parts of Southeast Asia. In the forty years since the Dubai Islamic Bank opened for business the Islamic finance movement has obviously made great strides across the Muslim world -- and even into parts of Europe and the United States that are not reported here. Table 8 is a work in progress.11 It includes all the countries surveyed in the Arab Barometer with the exception of Palestine.
14. Table 8 : Sharia-compliant market shares of commercial banking systems Year 1986 1996 1997 1998 2000 0.4% 0.5% 0.8% Bahrain 6.7% 9.8% Egypt 9.7% 5.1% including Banque Misr's Islamic 8.1%branches' deposits 1.0% 2001 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 first established Algeria \ 1991 1979 1977 15.7% 5.7% 8.6% 10.1% 11.5% 2.3% 1.7% 1.6% 5.7% 5.8% 9.5% 9.7% 10.3% 46% Iraq 6.2% 25.0% Jordan (JIB) 1978 7.0% 8.4% 8.2% 8.0% including Islamic International Arab Bank 8.9% Kuwait 1977 18.0% 16.2% 16.3% 15.5% Lebanon 1991 0.1% 0.0% 0.1% 6.9% 9.4% 7.1% 8.8% 10.7% 18.1% 22.6% 23.7% 24.4% 7.7% 8.2% 8.3% 8.3% 8.2% 10.8% 11.3% 11.3% 11.3% 11.4% 31% 0.4% Libya Morocco Qatar 1982 10.4% 17.8% 18.1% 13.7% 14.5% 14.9% 13.2% Saudi Arabia 1988 11.3% 11.1% 11.5% 13.9% 14.0% 14.1% 13.7% 38.4% including sharia-compliant windows of conventional banks 17.0% Sudan Turkey UAE Yemen 1983 1985 1975 1996 0.2% 0.8% 3.2% 0.6% 3.6% 0.6% 3.6% 0.8% 3.5% 3.6% 5.5% 7.0% 7.3% 3.6% 1.8% 7.9% 3.5% 3.8% 4.3% 10.9% 14.4% 16.3% 16.1% 4% 39.2% 41.8% Bangladesh 1992 Malaysia 1983 35% 53.0% 27.9% Syria (percentage of private resident deposits in local currency) Tunisia 24.0% 30.0% including sharia-compliant windows of conventional banks Indonesia 1.6% 1.6% including Islamic windows of 2% conventional 2% banks (rough estimate) 4.4% 4.7% 5% 5.6% 17.0% 17.0% 37.1% 41.7% 39.1% 31.9% 32.6% 42.1% 65% 1.2% 2.0% 4% 14.9% 10.5% 11.8% 4.6% 16.5% 18.4% 19.6% 21.1% 22.3% 23.3% 25% Pakistan Meezan Bank 1997 0.5% 0.8% 9% Evidently Algeria, Lebanon, and Tunisia lag behind the other countries. Algeria deserves particular attention, given its large population (approaching 40 million) and oil wealth. The survey indicated, however, that over half the Algerians who answered both questions about bank interest opposed conventional banking, and the odds significantly increase among respondents expressing political interest and opposition to secularism. There is obvious dissatisfaction and a significant potential clientele, yet Islamic banking has apparently stagnated since 1991, when it was introduced to Algeria. The Al Baraka Group created a joint venture at that time with the Banque d’Agriculture et Développement Rurale, one of Algeria’s four dominant public sector banks, and despite the former’s majority ownership Al Baraka Algeria seems to be managed like a public sector bank. It opened its doors for business just as Algeria’s decade of civil war was underway. Efforts to privatize banking in Algeria then went astray in 2003 with a series of spectacular scandals and bankruptcies. Consequently the government has not encouraged private sector banking, other than foreign banks, to penetrate what is still predominantly a state-run commercial banking system. The looming international financial crisis, notably trouble with housing loans at Citibank, deterred the Algeria government from privatizing one of its large public sector banks in 2007. Algeria, however, is only one example of the seriously underdeveloped financing that has retarded investment and growth in much of the MENA.12 One indicator of the impoverished
15. nature of banking systems is the amount of loose currency outside them . People prefer to hold their cash, at least partly out of widespread distrust for conventional banking as indicated in the Arab Barometer surveys. In fact the Arab countries can be compared along this dimension of “contract-intensive money,” or the proportion of the broad (M2) money supply serviced by the commercial banking systems.13 Within the region, as Table 9 indicates, Algeria, Iraq, Libya, Sudan, Syria, and Yemen stand out for their relatively low CIM ratios, together with Morocco. The only comparable standouts seem to be some of those other Muslim countries that were part of the Soviet Union. They are not so different, however, from other former components of the USSR such as Armenia, Georgia, and the Russian heartland. The other Muslim comparators in Table 9, Indonesia and Malaysia, support substantially higher CIM ratios and also, especially Malaysia, reflect substantial penetration of sharia-compliant instruments into the commercial banking system. Table 9: Contract-Intensive Money in the Arab, Post-Soviet and other Comparators 2001 2010 2011 Algeria 76.4% 77.6% Iraq 30.2% Libya 74.0% 79.9% Sudan 61.4% 70.4% Syria 66.0% 62.9% SyrianCentralBankdata 68.0% Yemen 74.3% 54.4% 83.1% 69.3% 71.7% 73.5% 75.9% 73.7% 55.4% 73.5% 67.3% 72.8% 72.8% 52.6% 78.7% 79.9% 69.5% 69.3% 71.2% 65.7% 70.9% 74.7% Morocco Jordan Egypt Tunisia 76.4% 87.3% 84.3% 86.3% 75.9% 87.5% 82.0% 84.6% 76.2% 87.1% 81.5% 86.2% 75.6% 86.8% 80.9% 85.8% SaudiArabia 81.8% 85.7% 89.7% Bahrain 94.6% 95.6% Kuwait 95.0% 94.8% 96.1% Oman 87.1% 89.6% Qatar 92.9% 94.5% 97.0% 88.8% 95.1% 95.7% 89.5% 97.1% 86.6% 95.0% 95.5% 89.2% 97.1% 89.4% 95.0% 95.4% 88.7% 97.3% Lebanon Turkey 95.1% 91.8% 95.1% 95.0% 91.9% 91.8% 72.8% 84.7% 85.7% 85.4% 2005 75.9% 86.6% 86.0% 84.5% 92.1% 93.7% 95.4% 90.6% 91.8% 91.7% 2012 2013 2001 2005 Afganistan Albania Azerbaijan Armenia Bulgaria Georgia Ukraine Kazakhstan Russia Tajikistan 2010 2011 2012 2013 52.1% 52.4% 55.7% 54.5% 64.0% 25.4% 30.9% 76.8% 26.5% 66.1% 69.8% 61.3% 63.7% 24.9% 11.9% 12.6% 20.3% 73.7% 9.8% 55.2% 66.5% 30.2% 33.3% 83.6% 45.3% 66.5% 80.1% 71.1% 18.5% 68.6% 30.4% 39.5% 84.6% 53.6% 69.3% 80.6% 71.8% 22.7% 70.0% 29.2% 40.4% 84.5% 52.9% 71.1% 79.7% 72.0% 22.7% 70.0% 32.9% 47.3% 84.7% 56.6% 71.1% 79.7% 73.5% Indonesia Malaysia 89.2% 88.0% 87.1% 87.0% 86.7% 86.6% 93.0% 94.4% 94.7% 94.9% 95.0% Thailand 91.8% 90.5% 90.5% 90.8% 90.9% 91.1% Argentina Brazil Venzuela 76.4% 61.0% 64.2% 61.4% 60.4% 61.1% 88.3% 88.0% 88.9% 89.9% 89.4% 89.6% 86.5% 87.4% 87.7% 89.7% 90.1% 90.4% UK 97.7% 98.1% 98.3% 98.1% 98.0% 98.0% Given these data, are there grounds for expecting that Islamic finance could become more inclusive than conventional banking and bring little investors and entrepreneurs out of the informal economy into a better protected environment of sharia-compliant financial intermediation? Islamic finance in Iraq, Sudan, and Yemen already have substantial market shares comparable to those in the GCC countries, yet they remain underbanked. The survey data showed majorities still rejecting conventional banks in Iraq and Yemen, apparently driven by conservative values rather than politics, but in Sudan, where Islamic banking has a longer history linked to Hassan Turabi’s rise to power, there is no longer majority support, and “Islamism”
16. works against it . In Iraq and Yemen political conditions may be too unstable to encourage entry into the formal financial sector, whether conventional or Islamic. Curiously, however, there was some evidence in Syria, before the eruption of civil strife, that Islamic finance was indeed attracting more popular participation in the banking system. After 2006, when Syria permitted three Islamic banks to begin operations, they quickly penetrated Syria’s rudimentary financial markets. By 2011 they had gained 7.3 per cent of Syria’s total resident deposits in state and private sector banks.14 During the period Syria’s CIM ratio steadily rose until 2011, when it fell under the pressures of political events, terminating chances to test whether Islamic finance might be attracting new clienteles. Had violent jihadist factions not hijacked the opposition, both sides could have shared an interest in continuing experimentation with Islamic finance, a source of legitimation for the regime yet also, apart from dress codes and family status laws, one of the very few nonviolent “Islamist” identity markers. The countries with the greatest potential for Islamic finance may be those that are both underbanked and so far deprived of Islamic finance. Algeria and Morocco stand out as the most promising prospects. The survey further confirmed popular political support in the Algeria for reinforcing the public sector efforts of Al-Baraka Algeria with more private management and initiative. In the midst of a prolonged succession crisis, the necessary reforms in financial structure remain on hold but meanwhile Morocco, unfortunately not included in the second wavc of surveys, is forging ahead, with legislation passed in November 2014 enabling Islamic banks and sukuk finally to get licenses.15 The countries that are less underbanked, such as Egypt and Jordan, also appear more open to Islamic finance. Far less penetrated by Islamic finance, Tunisia may offer interesting opportunities. Table 4 indicated that it was not political interest or Islamism that was associated with the rejection of conventional banking; rather it was religiosity and conservative attitudes toward women. In other words Islamic finance may progress in Tunisia quite independently of the country’s politics. Although the Nahda Party lost the legislative elections held in October 2014, private business will continue to advocate sukuk and more Islamic banking to attract Gulf investors, if only to keep up with Morocco. In Egypt as well, sukuk legislation prepared under President Mursi is still likely to materialize, given the importance of wealthy private investors from the GCC countries. Finally the Gulf countries and Lebanon are highly banked, but Lebanon also stands out in the region as being averse to Islamic banking. Opinion about interest-based banking in Lebanon was apolitical and rooted in conservative Muslim values, but Beirut’s bank culture has stubborn roots as well, among both Christian and Muslim bankers. By contrast Saudi Arabia, propelled in part by rapid increases in sharia-compliant market shares (mainly within preexisting conventional banks) associated with clients’ age, religiosity and wealth, was reaching the higher CIM levels of its city state neighbors. Conclusion This paper has triangulated the results of the Arab Barometer surveys with the penetration of Islamic finance into the conventional banking across much of the MENA and parts of Asia, and with the inclusiveness of their respective banking systems. Interesting patterns of potential
17. development emerge . Among the underbanked (low CIM) states, notably Algeria and Morocco, survey data offer hope, as stated at the outset of this paper, that Islamic finance may mobilize new constituencies of economic actors that will broaden the social bases of financial intermediation. These two countries are not only the most populous, second only to Egypt, in the Arab world. Representative samples of their respective populations also registered among the highest percentages of people believing that bank interest is forbidden in Islam. In Algeria, where data were also available about rejection rates of conventional banking, the analysis of the survey results reveals a politically sensitized potential constituency for Islamic finance. Promoting Islamic finance more vigorously might serve the interests of both the government and Islamist oppositions. More inclusive banking could contribute to the diversification of the Algerian economy as well as giving the government added legitimacy and the opposition some substantive “Islamist” achievement. Saudi Arabia exemplifies another pattern of development that the other members of the GCC probably shared. These wealthy states have graduated, so to speak, from cash-based economies with large informal sectors. Islamic finance did not serve so much to bring more people into the banking system as to be a creative means of portfolio diversification. Since the sum of private wealth in Saudi Arabia far outstrips that of the government (Luciani 2005), such diversification is likely to continue to be an engine for the development of Islamic finance and investment across the Muslim world. The growth of Islamic finance is also likely to have broader implications for economic development in the region. Islamic finance underlines the need for greater transparency and accountability because much of the financing is equity rather than debt-based. Whether mudaraba or musharaka or even sukuk, these distinctively Islamic instruments are more akin to venture capital or shareholdings than to conventional debt. Corporate sukuk, for instance, while generating fixed sources of income, may be more like preferred stock than debentures because in the event of bankruptcy the claims of conventional bondholders have priority.16 Islamic banks at least in theory have a greater stake in transparency and accountability than conventional banks. Since their distinctive form of financing is equity rather than debt based, the investors require more extensive information about their entrepreneurs than the standard credit checklist for conventional borrowers. These theoretical considerations may be quite irrelevant to war zones in North Africa or the Levant but in the long run Islamic finance, by sapping the foundations of the MENA’s informal economies and responding to popular desires for interestfree banking, may build up civil societies supporting civil states. Even if viewed as illiberal, such states may be more liberal and inclusive than military dictatorships.
18. Bibliography Ali , S. Nazim (2012), Shari’a-compliant Microfinance, Islamic Studies Series, London and New York: Routledge. Arab Barometer 2014: http://www.arabbarometer.org/content/about-center (Accessed 21 May 2015). Clague, Christopher, Philip Keefer, Stephen Knack, and Mancur Olson (1999), ContractIntensive Money: Contract Enforcement, Property Rights, and Economic Performance, World Bank: MPRA Paper No. 25717, posted 2010 http://mpra.ub.unimuenchen.de/25717/ (Accessed 22 May 2015). Demirguc-Kunt, Asli, and Leora Klapper (2012), Measuring Financial Inclusion: the Global Findex Database, Policy Research Working Paper 6025, Washington, DC: World Bank. Demirguc-Kunt, Asli, Leora Klapper, and Douglas Randall, 2013. Islamic Finance and Financial Inclusion: Measuring Use of and Demand for Formal Financial Services among Muslim Adults, Policy Research Paper Working Paper 6642, Washington, DC: World Bank. El-Gamal, Mahmoud, Mohamed El-Komi, Dean Karlan, and Adam Osman (2014), “Bankinsured RoSCA for microfinance: Experimental evidence in poor Egyptian villages.” Journal of Economic Behavior & Organization, Volume 103, Supplement, July 2014, pp. S56-S73: http://karlan.yale.edu/p/EMF-05-12.pdf (Accessed 21 May 2015). El-Komi, Mohamed, Rachel Croson 2013. “Experiments in Islamic Microfinance,” Journal of Economic Behavior & Organization, Volume 95: pp. 252-269: http://www.sciencedirect.com/science/article/pii/S0167268112001655 (Accessed 22 May 2015). Ernst and Young (2013), World Islamic Banking Competitiveness Report 2013-2014: http://www.ey.com/Publication/vwLUAssets/World_Islamic_Banking_Competitiveness_ Report_201314/$FILE/World%20Islamic%20Banking%20Competitiveness%20Report%20201314.pdf (Accessed 21 May 2015) Henry, C. M., and R. Springborg (2010) Globalization and the Politics of Development in the Middle East. 2nd ed. Cambridge UK: Cambridge University Press. Hrnjic, Emir, Harun Kapetanovic, and David Reeb (2014), “Emirates Airline: A Billion-Dollar Sukuk-Bond issue, Richard Ivey School of Business Foundation, National University of Singapore. Kahf, Monzer (2005), Islamic Banks: The Rise of a New Power Alliance of Wealth and Shari’a Scholarship, in Clement M. Henry and Rodney Wilson, eds., The Politics of Islamic Finance, Edinburgh University Press. Khan, M.M., and M.I. Bhatti (2008), Islamic Banking and Finance: on its way to globalization, Managerial Finance 34:10. http://www.perpustakaan.depkeu.go.id/FOLDERJURNAL/Islamic%20Banking%20and %20Finance.pdf (Accessed 22 May 2015). Luciani, Giacomo (2005), “From Private Sector to National Bourgeoisie: Saudi Arabian Business.” In Paul Aarts and Gerd Nonneman, eds. Saudi Arabia in the Balance: Political Economy, Society, Foreign Affairs, London: Hurst, pp. 105–123. Tessler, Mark, Amaney Jamal, and Michael Robbins (2014), “New Findings on Arabs and Democracy,” in Larry Diamond and Marc F. Plattner (eds), Democratization and Authoritarianism in the Arab World. Baltimore, MD: Johns Hopkins University Press,
19. pp . 54-68: http://www.aihr-resourcescenter.org/administrator/upload/documents/new.pdf (Accessed 22 May 2015). World Bank (2011), Financial Access and Stability: A Roadmap for the Middle East and North Africa. World Bank Flagship Report: http://siteresources.worldbank.org/INTMENA/Resources/Financial_Flagship_Report_Mi ddle_East_North_Africa_2011_Full_Report.pdf (Accessed 22 May 2015). 1 Converted to sharia-compliant financing by the Abu Dhabi Islamic Bank, the former National Development Bank of Egypt has an infrastructure for microfinance that may be revived. Also in Yemen Al-Kuraimi Bank has converted its microfinance services to fit Islamic norms: http://www.alkuraimi.com/en/sections.aspx?id=19 (accessed 21 May 2015). The growing interest in Islamic microfinance is documented in S. Nazim Ali, 2012. 2 In 2014 twelve international banks, led by the Industrial and Commercial Bank of China, had total assets exceeding $2 trillion dollars. See Top Banks in the World 2014: http://www.relbanks.com/worlds-top-banks/assets (accessed 21 May 2015). 3 The Arab Barometer (http://www.arabbarometer.org/), originally founded at the University of Michigan in 2005 in consultation with the Global Barometer (http://www.globalbarometer.net ), partners with the Arab Reform Initiative (http://www.arab-reform.net/) in association with other universities in the Arab world to carry out periodic surveys of public opinion. The three waves surveyed representative national samples, each including over 1000 respondents). The countries covered are reported in Table 1. 4 Demirguc-Kunt et al (2013: 5) found in a 2012 survey of over 5000 adults focused on banking in Algeria, Egypt, Morocco, Tunisia, and Yemen that 45 per cent preferred a sharia-compliant product over a less expensive conventional bank product, but that 37 percent preferred the conventional product or did not have a preference. 5 Respondents were asked on a four-point scale whether 1) A woman can become the prime minister or president of a Muslim state; 2) A married woman can work outside the home; 3) In general, men are better at political leadership than women; 4) University education for males is more important than university education for females; 5) Men and women should have equal work opportunities; 6) It is permissible for a woman to travel abroad by herself; 7) A woman should obtain her inheritance (she should not be denied her inheritance); 8) Women’s share of inheritance should be equal to that of men; 9) Women can assume judicial positions; 10) Women can become ministers. Reversing the scoring for items 3 and 4, the scores were then averaged to form a new variable, “woman at home,” since the conservative responses received the highest scores. Alpha (coefficient of reliability) = 0.80. 6 Like “women at home,” religiosity was derived from a set of interrelated questions. This analysis assumes that “religiosity” may be understood as a continuum, ranging in intensity, rather than a special state of mind reflected by the respondents’ answers to one or two questions (depending on the analyst’s definition of what “real” religiosity might be). Respondents were asked on a four point scale ranging from always to rarely (Tunisians were given “never” as a fifth point) whether they prayed, fasted during Ramadan or Lent, watched or listened to religious programs on TV or radio, attended religious classes in mosque or church, attended Friday or Sunday prayer services, listened to or read the Quran or Bible, and read religious books. The scores were then averaged to form a new continuous variable, “religiosity,” with the highest scores awarded those who engaged the most in these activities. Alpha (coefficient of reliability) = 0.78. 7 Political interest was indicated by the respondent’s self-evaluation and by how much he or she followed the news in their country. Alpha (coefficient of reliability) = 0.86 8 Respondents were asked on a four-point scale whether 1) Religious leaders (imams, preachers, priests) should not interfere in voters’ decisions in elections; 2) Your country is better off if religious people hold public positions in the state; 3) Religious leaders (imams, preachers, priests) should have influence over government decisions; 4)
20. Religious practices are private and should be separated from social and political life ; 5) Religious associations and institutions (excluding political parties) should not influence voters’ decisions in elections; 6) Mosques and churches should not be used for election campaigning. The coding on items (2) and (3) were reversed, so that the responses indicating more “Islamism” had higher scores while more secular responses had lower scores. Alpha (coefficient of reliability) = 0.69 or borderline reliability of their respective inter-correlations. 9 The Saudi sample was considerably wealthier than the others. Mean monthly income converted to US dollars varied from $175 and $344 in Egypt and Sudan, respectively, up to $1306 and $3370 in Lebanon and Saudi Arabia. The overall mean was $913, with a huge standard deviation of $1634. 10 The Arab Barometer second wave of interviews in Egypt were carried out in 2011 after the fall of Mubarak (Tessler et al 2014: 55); President Mursi was elected in June 2012 and overthrown on July 3, 2013. 11 Much of the data were compiled by adding up the relevant data from annual reports of Islamic banks and comparing them to combined commercial bank data offered by the respective central banks. Other sources of market shares are also included in Table 8. In addition to the author’s data set, also available in Henry and Springborg (2010), Hong Kong Shanghai Banking Corporation sponsored a special supplement to The Banker (Nov 2009) giving total sharia-compliant assets of commercial banks in the world. A McKinsey 2006 report has data presented here in boldface. Other sources were Kuwait Bankers Institute, Ernst and Young, the Organization of the Islamic Conference (Istanbul), and a White Paper by M.M. Khan and M.I. Bhatti in Managerial Finance 34:10 (2008): http://www.perpustakaan.depkeu.go.id/FOLDERJURNAL/Islamic%20Banking%20and%20Finance.pdf (Accessed 21 May 2015). The data for Yemen 2006-2013, drawn from the Central Bank of Yemen, are the ratio of Islamic to total advances of commercial banks (including loans of conventional interest-based banking. The total sharia compliant assets in 2011 and 2012 were also stated to be 31 per cent, so that financing was a fair proxy for market penetration although deposits data were not available. The source for the Tunisian sharia-compliant share of commercial bank deposits is the 2012 Annual Report of the Association Professionnelle des Banques Tunisiennes. The data for Syria were retrieved from the Central Bank https://www.banquecentrale.gov.sy/main-eg.htm where the consolidated balance sheets of all commercial banks, private commercial banks, and Islamic banks were available through March 2011. The Syrian data reported in Table 9 also stops in March 2011. The 2013 data for deposits as a % total banking deposits for Bahrain, Kuwait, Qatar, UAE, Malaysia, Pakistan, Turkey, and Indonesia were drawn from EY’s World Islamic Banking Competitiveness Report of 2014-15. Saudi deposit breakdowns were not available. 12 See World Bank 2011. See also Demirguc-Kunt and Klapper, 2012: 56, where they note that 12 per cent of MENA respondents claimed religion to be the reason they did not have bank accounts, compared to 8 per cent in South Asia and no more than 4 per cent in other (mostly non-Muslim) regions of the world. 13 Clague et al 1999 suggested that this ratio was a proxy for property rights and perhaps even for confidence in public institutions. 14 The Islamic banks by March 2011 held 15.4 percent of the total assets of Syria’s burgeoning private sector, which had risen to contain 27.9% of the Syria’s total commercial bank assets. 15 Aziz El Yaakoubi, “Update 2: Moroccan parliament approves Islamic finance legislation,” Reuters, 25 November 2014. 16 See, for example, the National University of Singapore Business School case study by Hrnjic et al, 2014, of “Emirates Airline: A Billion-Dollar Sukuk-Bond issue,” which compared the pricing of the sukuk with that of slightly smaller conventional bond issue. In the event of bankruptcy, the conventional bondholders would have priority over the sukuk holders. The latter also was a cheaper source of funds for the airline.