Should Islamic banks use mudaraba and musharaka?

A while ago, I started a series of posts about how Islamic financial institutions work, looking at their products one by one, and I got distracted from it by other things I was working on.  However, I was looking through the financial statements of a large Islamic bank and I had a few thoughts.  Hopefully I'll get to that series of posts again at some point, but this was still interesting to me.  Here's the breakdown of assets by the types of contracts used (excluding cash and non-financing assets):

	30-Jun-12	30-Jun-11
Murabaha	84.8%	83.1%
Ijara	0.3%	0.3%
Salam	1.0%	0.8%
Istisna'a	0.2%	0.3%
Mudaraba	4.4%	6.2%
Musharaka	3.4%	3.5%
Ijara Muntahia Bittamleek	5.8%	5.8%

The breakdown should be relatively expected to most people, since the Islamic finance industry has used murabaha because it is the easiest to use by banks that are trying to (or forced to by regulations) keep the same business model as in conventional banking.  It also conforms to rough estimates that pop up in articles every now and then that 90% or thereabouts of Islamic finance is based on murabaha.

It shouldn't be surprising either that most of the ijara contracts are ijara muntahia bittamleek (leases that end in ownership transfer from the lessor to the lessee), since it is a bank, not a leasing company (which would have a different breakdown between financing and operating leases). 

What is a little surprising is how much of the non-murabaha contracts are mudaraba and musharaka. To see if the last few years were outliers, I pulled up the total share for east as a percentage of the bank's assets going back to 2008. 

	2008	2009	2010	2011	2012
Mudaraba/Musharaka	8.3%	7.2%	7.0%	9.8%	7.9%

While these seem like small amounts, their absolute amount is larger since this is a large bank with about $17 billion in assets (the latest total mudaraba and musharaka assets were $859 million).  This is still very small relative to the total size of Islamic finance which is itself a small part of the global financial industry, but it is still there as a portion of Islamic banks' balance sheets.  There is also the chance that these contracts are used to replicate conventional financial lending arrangements (with anticipated profit rates that determine periodic payments), but that cannot be gleaned from the financial statements. 

Regardless of whether the share of mudaraba and musharaka is large or small, there remains the question of whether a bank should be engaged in taking equity positions at all (for example, banks in the US are not allowed to make equity investments so musharaka-based products are offered only by non-bank financial institutions).  Within Islamic finance, the practical answer to that question is 'no', and a resounding 'no'.  But the theoretical answer is 'yes' since the first model for an Islamic bank is based on a two-tier mudaraba, where the bank provides capital to businesses and shares in the profits and absorbs any losses. 

This probably is not the best system for banking because when people put money into banks as deposits (rather than investing as shareholders), they expect to be able to use that money for transactions or for savings, and are not expecting to recognize losses of capital (or else they would find a better destination, like a mutual fund, where they would fully get the profits from investments (Islamic banks pay out returns to depositors in a way to compete with conventional banks so returns are based on the prevailing interest rates on deposits). 

This conflicts with the idea that Islamic banks should use a model as close as possible to a two-tiered mudaraba.  Islamic banks do use the mudaraba structure for deposits (this bank had profit-sharing investment accounts nearly 3 times the value of current accounts), but shelter depositors from losses to the greatest extent possible by reserve funds that build up profits in excess of the profits paid out (based on prevailing interest rates on deposits) and shareholders will likely be expected to give up profits to meet the prevailing rate of return if there were ever a shortfall. 

On the issue of banks engaging in mudaraba, Dr. Zubair Hasan at INCEIF wrote a post recently that discussed the profit sharing model and agency problems.  In his post is a diagram which shows the potential to inject mudaraba financing into an operating business where the firm's profits are separated in a way from the mudaraba profits (since the mudaraba is funding a portion of the business and its profit share is dependent only on this portion of the business). 

This is a way that a mudaraba could be incorporated into more business financing in the same way that mudaraba sukuk are used to fund aspects of businesses (for example, an expansion of a bank's Islamic banking business).  The new (post-AAOIFI ruling) way mudaraba sukuk work is that the funds are invested and an anticipated profit-rate is paid periodically with the final redemption based on the market value of the assets (less any shortfall in the actual profits for the rabb ul-maal over the life of the sukuk). 

This is a nice solution for a sukuk issuer that wants debt where its payments are predictable and the redemption value is based on the underlying value of the business being financed, but from the financier's perspective (assuming that is a bank for this discussion) it is not a good way to operate since it gets limited upside and assumes the risk in cases of loss.  There are likely ways to make it more like a conventional loan, but that further muddies the water of the underlying structure in order to make it a banking product. 

This complexity subtracts value because it transforms what is fundamentally supposed to be a relationship that is equity-based not debt-based.  There are much better ways for equity investment that don't require such complicated modification, like, well, equity mutual funds.  If an investor wants to take advantage of a professional manager and is willing to take losses and wants to have daily liquidity, why change how banks work and why not just go invest in a mutual fund that selects Shari'ah-compliant equities where all of the losses are borne by the investor and it takes a proportional share of equity corresponding to the dividends and capital gains paid out. 

It is perfectly possible to design Islamic banks to incorporate mudaraba and musharaka contracts, but it doesn't seem fundamentally useful when there are more functional, simple ways to use those contracts and depositors are generally not prepared to risk principal. Maybe the debate should be more about why the other forms of non-murabaha debt financing like istisna'a and salam (not to mention asset-based or asset-backed financing using ijara) are not used more and murabaha less.