The Islamic Interbank Money Market An Analysis of

The Islamic Interbank Money Market : An Analysis of Risks Obiyathulla Ismath Bacha Management Center, IIUM 1

l The basic strategy in developing the Islamic Banking and Finance Sector has been replication. l In introducing Islamic Banking, for example the chosen method, has been to transform the sources and applications of funds of conventional banks into Islamically acceptable ones. l Similarly, the Islamic Interbank Money Market (IIMM), has been based on the introduction of Islamic Money Market instruments that mimic conventional ones. l Overall, such a strategy of choosing the path of least resistence; has been successful. 2

l Both Islamic Banks (Ibs) and the IIMM, have grown in tandem with the overall growth of the Banking sector in Malaysia. l From about a tenth of one percent (0. 1%) in January 1994, deposits in IBs have grown to 7. 5% of Total Deposits in July 2003. l This constitutes a highly impressive average annual growth rate of 62. 5%. [1] l Thus, Malaysia today has a truly dual banking system with both conventional and Islamic banks coexisting side by side. [1] From Jan 1994 to July 2003 3

F ig u re 1 Is la m ic B a n k T o t a l D e p o s it a s p e rc e n ta g e o f C o n v e n tio n a l B a n k T o ta l D e p o s it (J a n -1 9 9 4 --J u ly -2 0 0 3 ) 8 7 6 5 4 3 2 1 0 94 95 96 97 98 99 00 01 02 03 D ate 4

l Though each system operates theoretically within its own sphere, it is inevitable that given a common macro environment, the two systems interact. l A large non-muslim customer base connects both systems. Diagram 1 Financial System Islamic Banking Conventional Banking Islamic Interbank Money Market Conventional Interbank Money Market Takaful Conventional Insurance 5

l The Islamic Interbank Money Market (IIMM) was established at BNM’s initiative and became operational in Jan. 1994. l As with conventional ones, the IIMM enables Islamic Financial Institutions to manage liquidity aside from other roles such as price discovery. l Over the 10 years of its existence, the IIMM has seen the introduction of several new instruments. Key among these are: • Mudarabah Interbank Investment (MII) • Government Investment Issues (GII) • Bank Negara Negotiable Notes (BNNN) • Nego. Islamic Debt Certificates (NIDC) • Islamic Accepted Bills • Other Islamic PDS. l Since the IIMM was intended to play the same role, the money market plays for conventional banking, the instruments were necessarily a replication of conventional money market instruments. 6

Trading and Pricing of IIMM instruments l Taking trading volume as a measure, some of the instruments have been successful while others have seen ‘patchy’ volume. l Instruments such as MII and GII have seen tremendous volume growth while others less so. For example, MII volume has gone from RM 0. 5 bil. In 1994 to RM 283. 8 bil. in 2003. l Yet others like IABs, NIDCs etc. , have not shown similar growth trends. 7

Figure 2 8

Figure 3 9

Figure 4 10

Figure 5 11

Figure 6 12

l There appears to be a shallowness with regards to secondary-market trading. l That volumes are not more substantial despite the huge excess liquidity is worth noting. Pricing l The underlying logic of pricing the IIMM instruments appears to be the same as that of conventional money market instruments. l The same principle of discounting. l Whereas in conventional markets; discount rates – market yields. l Here, the discount rates are derived from required profit rates, mark-ups and indicative “dividends”. 13

Pricing MII: Determining the Profit rate to Fund providers Where: X = amount of profit (in Ringgit) to be paid to the provider of funds P = principal investment r = rate of gross profit (in % p. a) before distribution for investments of one year of the “receiving” bank t = number of days invested k = profit sharing ration 14

Pricing of Islamic Accepted Bills (IABs) Formula : Price of IAB is calculated as follow: Market price = where; F = r = t = Face Value Annual rate of profit Number of days remaining to maturity 15

Pricing of Govt. Invest. Issues (GIIs) Where; r = rate of return per annum t = number of days from the issuing or anniversary date 16

Potential Risks in IIMM l Given the instruments, their pricing mechanism and trading methods, a number of potential risks can arise. Most are common to conventional money markets. 1. 2. 3. 4. 5. 6. Counterparty risk Liquidity risks Interest rate risk Regulatory risk Sharia risk Accounting risks Counterparty risk: refers to potential default/default risk of counterparty. – Given the short-term nature of money market instruments; default risk generally small. (Penn Central; defaulted with CPs aprox. $0. 5 bil) 17

Liquidity risk: refers to risk that arises from Illiquid/inactive trading. l Illiquidity volatility; b – a spreads. l Illiquidity also means price discovery is not well functioning pricing efficiency? l The presence of market makers helps, as in the case of GIIs and BNNNs. l For other instruments, the lack of an active secondary market means liquidity can be a problem. 18

Sharia Risk: A counterpart of regulatory risk? Refers to the possibility that transactions/instruments currently deemed acceptable could subsequently be prohibited. l The issue of mark-ups – Pakistan – Halal Haram Halal again. – Sharia listed stock in Malaysia. Accounting Risk: Refers to the risk that a borrower in a profit-sharing relationship might understate profits. – Though this risk is substantially less with money market instruments, given their short-term nature, accounting risk remains for the external financier. 19

Interest Rate Risk: Is probably the most critical risk where money market instruments are concerned. - The key “price discovery” in money markets are short-term interest rates. - The short end of the yield curve is derived from interbank market trading. - Trading in money markets are designed to be reflective of rate movements, thus money market instruments are highly sensitive. 20

1. Interest rate risk manifests itself in several ways. The key forms being: • Prepayment risk • Reinvestment risk • Repricing risk - Given the short-term nature of money market instruments, repricing risk is by far the most important. - One might be tempted to ask what these interest rate dynamics have to do with Islamic Money Markets? - In dual banking systems, there is indeed a substantial link. 21

l Given extensive linkage between the two systems, when interest rates change in the conventional systems, deposit rates must change within the Islamic banking system. l In the IIMM, even if BNM is the key “counter party”, it will not be in a position to maintain different returns in the Islamic vs. conventional market. l This is inevitable since, in the absence of corresponding changes in Islamic bank deposit rates, rate differentials will prevail leading to easy arbitrage opportunity. 22

l In analyzing the extent of potential interest rate risk for Islamic Banks, two key variables; rates of return and total deposit amounts are examined. l In empirically examining these two variables for both the Islamic and conventional banking sectors, aggregate monthly data sourced from Bank Negara Malaysia (BNM) is used. l The period of study is covers a total of 113 months from January 1994 to July 2003[2]. A total of four variables, two each for each sector is examined. [2] A total of 113 months – data for 2 months Nov, Dec. 1996 were not available. 23

l Consequently, two hypotheses are tested, the first that, that there is no link between 3 month interest rates and Islamic bank 3 month rate of return. The second that there is no link in deposit formation. • To see if a causal relationship might exist, the Granger Causality (with 4 lags) is used to test for causality both ways. 24

Figure 7 25

Results Table 1 Panel A: Correlation between 3 -Month Rate of Return in IB and 3 Month Interest Rate Overall IB-3 MTH-ROR CB-3 -MTHINTR 0. 9214 First Segment IB-3 MTH-ROR 0. 9591 Second Segment IB-3 MTH-ROR 0. 9740 26

Panel B Regression results of Hypotheses (1) Period covered Coefficient (β Probability Value R. Squared Remark Overall (Eq. 1) 0. 660683 0. 0000 0. 849053 Sig. At 0. 0 5% level With lag variable (Eq. 2) 0. 6806059 2. 43125656 E-5 0. 8957719 Sig. At 0. 0 5% level First Segment (Eq. 1) 0. 739860 0. 0000 0. 919925 Sig. At 0. 0 5% level With lag variable (Eq. 2) 0. 749074 0. 0000 0. 922381 Sig. At 0. 0 5% level Second Segment (Eq. 1) 1. 367733 0. 0000 0. 948811 Sig. At 0. 0 5% level With lag variable (Eq. 2) 1. 235162 0. 0000 0. 965823 Sig. At 0. 0 5% level Overall is for the period : January, 1994 to July 2003. First Segment is for the period January 1994 to August 1998. Second Segment is for the period September 1998 to July 2003. 27

Table 2 Panel A Correlation between Total deposit of Commercial & Islamic Banks CB-Tot. Deposit Overall First Segment Second Segment IB. Tot. Deposit IB-Tot. Deposit 0. 8061 0. 9729 0. 9389 29

Panel B Regression results of Hypotheses (2) Period covered Coefficient (β Probability Value R. Squared Remark Overall (Eq. 3) 0. 115048 0. 0000 0. 649945 Sig. At 0. 0 5% level With lag variable (Eq. 4) 0. 115382 0. 0000 0. 650429 Sig. At 0. 0 5% level First Segment (Eq. 3) 0. 19135 0. 0000 0. 946651 Sig. At 0. 0 5% level With lag variable (Eq. 4) 0. 018901 0. 0000 0. 949130 Sig. At 0. 0 5% level Second Segment 0. 372424 0. 0000 0. 881698 Sig. At 0. 0 5% level With lag variable (Eq. 4) 0. 378883 0. 0000 0. 888630 Sig. At 0. 0 5% level Overall is for the period : January, 1994 to July 2003. First Segment is for the period January 1994 to August 1998. Second Segment is for the period September 1998 to July 2003. 30

Panel C Result of Granger Causality Test of Hypotheses 2 (4 lags) Overall Null Hypotheses Obs F-Statistic Probability IBTDEP does not Granger Cause CBTDEP 105 0. 84711 0. 49879 CBTDEP does not Granger Cause IBTDEP 105 2. 90509 0. 02566** First Segment Null Hypotheses Obs F-Statistic Probability IBTDEP does not Granger Cause CBTDEP 46 1. 32014 0. 28056 CBTDEP does not Granger Cause IBTDEP 46 2. 64780 0. 04859** Second Segment Null Hypotheses Obs F-Statistic Probability IBTDEP does not Granger Cause CBTDEP 55 0. 89679 0. 47364 CBTDEP does not Granger Cause IBTDEP 55 1. 82492 0. 14022 **Significant at 5% 31

What these statistical analyses show are two things: l i. Deposit rates of Islamic financial institutions are closely correlated with those conventional banks. ii. Changes in conventional rates granger cause changes in Islamic bank rates of return. ● The key implication of the results however is that, Islamic financial institutions have just as much exposure to rate movements as their conventional counterparts. ● Ironical as it may be; given the “fixed-rate” financing that most Islamic banks have, their asset-liability duration mismatch is likely to be higher and thereby have higher interest rate risk. 32

l In the case of IIMM players, the fact that instruments are priced based on discounting makes them equally susceptible to rate movements. l Note that given the extensive linkage with the conventional sectors, arbitrage flows will ensure that rates in both the Islamic and conventional sectors move in tandem. l Thus, changing rates in the conventional money market must cause rate changes in IIMM. 33

l The fact that duration mismatchs may be worse for Islamic banks has not been lost on authorities. l Recognizing that rate risk is substantial for Islamic banks, BNM last year (2003) introduced a Variable Rate Mechanism for Islamic Banks. l In essence, this instrument allows IBs to charge variable rates on their loans. l Such floating rate loans reduce durations on the asset side and thereby the exposure to rate movements. l Beginning with a higher “selling” price than normal BBAs, the instrument involves a series of rebates if rates fall. l Though this helps IBs with their funding mismatch, the instrument essentially shifts rate risk on to the customer. (though upside is capped). 34

Conclusion l One might argue that since IBs need to share profits only as and when profits are made, relying on IIMM has two advantages. i. ii. IBs are not hurt, since they don’t have the same profit pressure as conventional banks. Duration mismatchs are substantially reduced. • While, duration mismatch is indeed reduced, an Islamic banking system heavily reliant on the money market for placement of funds would not just be less profitable but less competitive. • Furthermore, given the strong correlation we saw between rates and deposit formation in the Islamic and conventional systems, rate and deposit flows will be equalized. 35

l Since we cannot completely isolate the two markets, a whole new set of challenges have to be met. l To avoid inconsistencies and other misalignment that could be damaging (e. g. an S & L style crisis in Islamic Banking) regulators have to be not just watchful but pre emptive. l Two key fronts to work on: – Product innovation alternatives to sophisticated risks. to come up with halal managing the increasingly and – Shariah “flexibility”? 36

37