Saturday, November 16, 2013

Hypocrisy or synthesis? -- From Islamic Finance to Religion and Society

I have noticed in my sermons over the past few months a clear pattern. Although I hadn't thought about them in those terms when I wrote them, sermons at one mosque have been neo-Mu`tazilite in orientation, while sermons at another have been Ash`arite, and both had a dose of Sufism in the mix.

Professionally, my approach to Islamic finance, which has won me too many enemies (and, at best, very few and very quiet friends) in the industry's professional and even academic circles, has been solidly new-Mu`tazilite. By the latter I mean the approach of Muhammad `Abduh and his students, which is becoming a minority approach even in Al-Azhar, as it has fallen under traditionalist influence with gulf funding (not surprisingly, even Rashid Rida, `Abduh's main student, had deviated from the latter's approach towards a more traditionalist one).

In the arena of Islamic finance, it was easy to come to my conclusions and to stick to them against the onslaught of attacks from industry consultants ("Shari`a Scholars" in industry parlance), because even the most revered scholars in the Hanbali GCC -- Ibn Taymiya and his student Ibn Qayim -- had strongly condemned the use of legal stratagems (such as today's murabaha, tawarruq, etc. to synthesize interest based loans -- see for example this previous posting on my blog for a clearly rationalist quote).

Historically, Hanbali jurists were dismayed at classical-period Hanafis using legal stratagems to allow circumvention of Islamic law of riba -- as the latter must have believed that the intent of the law didn't apply for some transactions or in cases of need, and therefore found the hiyal (legal stratagems) that allowed it. The master of this game in recent times has been Justice Taqi Usmani, a Pakistani Hanafi jurist and son of a former Pakistani Grand-Mufti, who himself lamented the state of the industry at a panel in which I participated circa 2005, when he said that they have replaced fiqh al-mu`amalat (jurisprudence of financial transactions) with fiqh al-hiyal (jurisprudence of legal stragems).

The interesting thing, however, is that the Hanafi legal code, Majallat Al-Ahkam Al-`Adliya, imposed by the Ottoman Empire, had accepted the fundamental rule expounded by Ibn Qayim in I`lam Al-Muwaqqi`n that "What Matters in Contracts Is Their Substance, Not Their Wording and Constructions" (العبرة في العقود بالمعاني و ليست بالألفاظ و المباني). Thus, modern Hanafis and Hanbalis would have agreed on the incoherence of the financial-engineering driven form-over-substance Islamic finance. Except that, in practice, those who built the current industry have adopted an incoherent traditionalist approach, even as they paid lip service to the juristic role of substance-based analysis of contracts.

I do not regret my adoption of a rationalist approach on the issues of Islamic jurisprudence of financial transactions, even if it meant being ostracized by conference organizers for fear of upsetting their main industry-participant sponsors, because traditionalists (from the Hanbali Ibn Qayim to the Hanafi Ibn `Abidin) had themselves looked at these matters rationally. It is impossible for me to know whether modern "Islamic finance" quasi-scholarship did not reach the same opinions reached by Muhammad `Abduh and scholars in his tradition for religious or financial reasons, whether personal or social (some claim that they help Muslims who otherwise would be isolated financially; a claim that I have not seen verified empirically), or for truly religious reasons. Regardless, I have chosen not to be hypocritical, so I spoke my mind on that industry, and have no regrets about that.

Moving beyond pseudo-Islamic Finance, however, I am now very interested in issues of "Religion and Economics," especially as they pertain to my native Egypt, and, to a lesser extent, other majority-Muslim countries. Many have argued that the major problem of this part of the world is the continued failure to find a synthesis between rationalist and traditionalist approaches to modernity. Even within the economic sphere, to which I limit my attention professionally, this has meant that a revolutionary (sometimes militant, as Bruce Lawrence has shown) reaction to modernity, as represented today by the post-Washington-Consensus Washington Consensus (or WC-II; neo-libaralism with social safety nets and a focus on building proper institutions), cannot be channeled productively without reaching a synthesis that will be acceptable not only to intellectual modernists, secular and religious, which is already a daunting task, but also to traditionalists and the public at large, which carries an incoherently compartmentalized religious approach -- in large part due to deficits in education, as students overspecialize early in life into technical fields, or receive an "illiberal arts" education where freedom of thought is stifled.

As I struggle for an approach to study "Religion and Economics" for economic development in that part of the world, I have noticed my own compartmentalization of religious styles (traditionalist in some mosques, and rationalist in others; selectively choosing elements of each tradition that suit my temperament and would not be too uncomfortable for the congregants). I wonder if this compartmentalization is a sign of the same unproductive hypocrisy that has plagued my native part of the world for at least a century and a half. It is easiest for me to adopt the rationalist modernist approach professionally (which will help me to publish again more frequently in mainstream economics venues), but that is of no value to anyone but perhaps my own academic career. That defeats the purpose of having become an economist in the first place; being a mainstream engineer like my brothers would have produced a lot more value to society.

Trying to force-feed rationalist modernist economic approaches to my native Egypt, as many have tried and failed for decades, is also a waste of time, and potentially a dangerous one at that. In the meantime, the traditionalist approach is anachronistic and often irrational, and my experience with pseudo-Islamic finance tells me that even if I can cynically play the traditionalist game by its rules, my constitution will not allow me to do so convincingly. It is, of course, delusional to think that I can contribute substantially to finding that elusive synthesis between rationalist and traditionalist approaches (which has eluded well-intentioned people at least from the time of Al-Afghani and `Abduh to our day). However, in a Kantian universalizability sense, if many like-minded students work together, the task may not be impossible. Whether that is itself a form of self-delusional hypocrisy or genuine desire to conduct useful social science remains to be seen.

Monday, November 04, 2013

Attitudes towards Democracy and Economics

This should be my last post in this string. Looking again through the estimated Pew data structure, the question Q21 on attitudes towards democracy (always best, I don't care, or sometimes not best) is nested in the Bayesian network between two questions Q72 (do you prefer a good democracy or a strong economy) and Q71 (do you prefer a strong democracy or a strong leader).

The substructure is Q72 -> Q21 -> Q71, and the antecedent Q72 is driven by views on the importance of improving economic conditions Q115G.

The joint distributions of Q21 with its antecedent, Q72, and its child Q71 is self explanatory and shown below.

Joint Distribution of Q72 and Q21, Pew Spring 2012 Egypt survey
Almost 50-50 split between prioritizing economy and democracy, with very positive views on democracy.

Joint Distribution of Q21 and Q71, Pew Spring 2012 Egypt Survey
Very strong pro-democracy sentiments in Spring 2012. This is where a lot of public opinion might have shifted, and may, of course, shift again.

Secular/Anti-Secular and Left/Right in Egypt; reexamining the 2012 presidential elections

The rabbit hole goes deeper still, but also yields a better explanation of what happened during summer 2012. Following on the earlier post today, I decided to look at the breakdown of attitudes toward free markets by perception of religious role in politics and views of whether the respondents wanted a larger or smaller role.

Below is an association plot, which shows for each cross cell between these two variables deviations from an independence model.

Association table for attitudes toward religion in politics and the role of free markets
Two big sets of deviations are apparent:

  • The anti-secular left is represented by the two positive big boxes near the bottom left (role of religion is large, that is a good thing, and disagree that free markets help most people)
  • The anti-secular right is represented by two positive big boxes near the top right (role of religion is small, that is a bad thing, and agree strongly that free markets help most people)
That explains why the religious "right" (socially speaking, I refer to this as anti-secular) is fragmented. Those with more Salafi leanings, I suppose, who think that more religion in politics would be a good thing, and lament its absence, are pro-market! Is this the Khairat El-Shater types? In the meantime, those who believe that there is already a large role of religion in politics and that this is a good thing are anti-market. Is this the Abdul-Monem Abul-Fotouh types?

Of course, there are also the secular left (top two large negative boxes in the second column from right, those who believe that role of religion is small and that is a good thing, but want less free market), and the secular right (bottom two negative boxes in second column from right, role of religion is small, and that is a good thing, but they would like more free market).

Current polarization of Egyptian society is along the secular anti-secular dimension, but because both have right-left divisions on economic grounds, neither can get their act together.

As argued in a much earlier post (and lecture at AUC) in May 2012 before the presidential elections, I argued that the political center was split between the anti-secular left (Abou El Fotouh) and the secular left (Sabbahi). Sure enough, these are the four large positive boxes near the bottom left and top right shown here. As I argued then, a coalition of the two would have easily won the election, but in the event, votes were divided and the two extremes of the anti-secular right (Morsi) and the secular right (Shafiq) made it to the second round, with low enthusiasm for both.

More on Egypt's Anti-Secularist Left

Further to some of the summary of evidence that I have showed in the previous post, some of which using World Values Survey 2008 (data collected 2005-7), I show below some results from the latest available Pew Spring 2012 survey. It shows the same pattern of a strong Islamist left contingency.

The Bayesian network analysis for this survey (using all 160 variables) identifies that the question Q26:
"Please tell me whether you completely agree, mostly agree, mostly disagree or completely disagree with the following statement - most people are better off in a free market economy, even though some people are rich and some are poor?"
is independent of all other variables in the survey conditional on the question Q62:
"How much of a role do you think Islam plays in the political life of our country – a very large role, a fairly large role, a fairly small role, or a very small role?"
In other words, religious attitudes are still the best predictor of views toward neo-liberal pro-market economic policies. The joint distribution of the two variables is shown here:
Joint Distribution of Q62 and Q26 responses from Pew 2012 Egypt survey
 So, we can see that those who think that Islam plays a fairly large role in politics is significantly tilted to disagreeing with the assumption that free market policies benefit most people, and those who think that it plays a fairly small role are more likely to think that free market policies do in fact benefit most people.

This is the same relationship between Islamist and leftist tendencies, but it appears less dramatic in the Pew data because they didn't ask whether religion should play a bigger role; rather, Q62 only asked if it does.

To establish that this framing of the question diluted the result somewhat, we show the relationship between Q62 and the follow up question Q63 on whether or not the current role of Islam, as perceived by the respondent, is a good or bad thing (the Bayesian network structure shows that Q62 is independent of all other variables conditional on Q63 and the respondent's religion).

Joint Distribution of Q63 and Q62 responses from Pew 2012 Egypt survey
Note that 31% of the respondents think that the role of Islam in politics is fairly large and that that's a good thing, but that an additional 23% think that the role of Islam is fairly small and that is a bad thing. Adding the two, we get 54% of the population being decidedly anti-secular. If Pew had asked if Islam should play a large role in politics, they would have been in the same group, and would have a shown a stronger anti-secular leftist contingency, as shown in the WVS data.

Finally, Pew only asked about "Islam" playing a role in politics, rather than "religion," which biases Christians' responses to the question. The WVS data shows that even among Christians, a very significant percentage opined that "it is an essential feature of democracy that religious authorities should interpret the law."

One potential objection that a friend mentioned regarding the usefulness of these findings is that public opinion is believed by some to have moved considerably after the abject failure of the Muslim Brotherhood regime to rule during their year in power (July 2012 -- June 2013). I am sure that some people who were willing to give MB the benefit of the doubt, and hopeful that they will moderate once in power, as examples of Islamists in some other countries have suggested, have been badly disappointed in their authoritarianism and incompetence.

Nonetheless, I find it hard to believe that such strong public sentiments, built over nearly a century, during which Islamist and leftist views commingled, could really have dissipated so quickly. The underlying dynamics are often veiled behind identity political discourse, but I would argue (not surprisingly, because I am an economist, after all) that -- as in that old Clinton campaign slogan from two decades ago -- "it's the economy, stupid."

There is much more evidence to go with this story, but it's probably best to keep some material for the formal paper.


Friday, November 01, 2013

Income Distribution Dynamics and Leftist Islamism in Egypt

I am working with a former Korean student of mine (Deockhyun Ryu, who went back to Korea but is currently visiting Rice) on extending our work on income distribution dynamics to consider within as well as between country income dynamics. So, out of curiosity, and as input into my other ongoing work on understanding the Arab Spring and the role of leftist religion therein, I plotted this graph for Egypt, showing the evolution of per capita income by decile. Divergence between the top decile and the rest is very clear starting with the open door policies in the 1970s. This is in contrast to the pattern in Korea, whose growth was much more inclusive (less divergence between deciles), but somewhat similar to Turkey's. 


Egypt Per Capita Income by Decile (2005 PPP Dollars; data: PWT 7.1 and WIID2C) 
(Note: an earlier version of this plot had used the wrong deflator).

Korea Per Capita Income by Decile
Turkey Per Capita Income by decile

We do not see the same strong dynamic between religious and class discourses in Egypt vs. Turkey. Inequality in Turkey remains high, but it has been declining in Turkey, even as it increased in Egypt. This figure compares the ratio between the top decile per capita income to the 5th decile (middle class):

Ratio of top decile to 5th decile per capita income


One way to look at the difference between Egyptian leftist Islamism and the Turkish mix is to look at the relationships between variables in the WVS, which I show below (the strong link between these variables are detected by Bayesian Network analysis of 200+ variables from the WVS; the causal link for Egypt goes from leftist tendencies to Islamism). 

Note how for Egypt, the majority are both leftist (democracy for them means redistribution from the rich to the poor) and anti-secular (democracy for them means that religious authorities should interpret the law):

Egypt's leftist Islamism (data: WVS 2008)

By contrast, for Turkey, the relationship is much more mixed, with smaller centers of gravity at all corners (they have their leftist Islamists, but also their leftist secularists, and their rightist secularists, etc.):

Turkey's Split Personality (data: WVS 2008)