APPLES AND ORANGES
BASEL III plans to apply the leverage ratio to all collateral, including initial margin on cleared derivatives transactions (cleared at a clearinghouse). Swap Market participants are looking at cost increases estimated at 60%. A decision such as this will mean sell side will have to add on that capital charge, giving sell side the momentum to seek other venues on which to trade derivatives.
The leverage ratio is one step in calculating the capital requirement of an institution. Generally capital requirements mitigate institutional risk. Capital charges aren’t necessarily good or bad. But there are applications that are unnecessary. Including initial margin posted to a CCP is one of them. The FIA did a great piece on this same topic
Allow me to explain 3 reasons why initial margin should not be included in the leverage ratio calculation.
REASON ONE: INCLUDING INITIAL MARGIN IN LEVERAGE RATIO BLURS SEGREGATION
- After MF Global this much is clear: SEGREGATED means SEPARATE. Yes, PHYSICALLY separate.
- As in it’s not your money. You can’t use it. You can’t leverage it. Don’t touch it and leave it alone.
CONCLUSION: Being Segregated funds, they add no risk to an institution
N.B. I remember having dimes in my loafers “just in case”. The dimes were SEGREGATED from the vast fortune in my pockets (probably 15 cents. If the dimes were intermingled with my other money, Lord knows what kind of chaos could result. So risk was mitigated because I had 2 dimes (the cost of two phone calls). At least my parents thought so. I just thought dimes looked cooler than pennies.
REASON TWO: INCLUDING INITIAL MARGIN MAY SKEW RISK REPORTS
- Initial margin is a good faith deposit posted by both counterparties in a derivatives transaction. By regulation (CEA of 1850?) those funds are SEGREGATED.
- Any profit or loss beyond the initial margin is paid by the losing side of the trade and received by the winning side of the trade.
- VARIATION MARGIN is NOT segregated.
- Variation margin (daily P&L on the derivative contract) stays within the institution’s risk pool and should be a part of the Leverage Ratio calculation.
- Institutions rely on risk reports to properly hedge their risk/s.
- A risk number which includes Initial Margin could skew hedge activities unless a process to back out the initial margin is done (which is rife with risk potential all it’s own).
CONCLUSION: Including Initial Margin could throw off hedges at the desk or business unit level.
REASON THREE: RISK MANAGEMENT IS A BALANCING ACT
- Initial margin can be changed by the CCP at any time (intraday or overnight)
- This means initial margin has TWO locus of review:
- As stated in #2, the value of the collateral posted as Initial margin changes, more collateral will be collected from both parties. Bringing them back to the original initial margin amount.
- As the market on which the derivative is based becomes more volatile, the CCP will require more initial margin.
- Typically exchanges attempt to cover 3 times the average daily move of the underlying asset.
CONCLUSION: Including Initial Margin will double count the risk due to the risk mitigation process already applied to initial margin.
SUMMARY & CONCLUSION
We get it! Regulators want to mitigate as much risk as possible. But Black Swans happen (I know cos I read the book..lol)
Normal distribution says a 5 standard deviation event is expected once every 1.8 million occurrences. The Dow Jones chart below shows 10 events greater than 5 STDEV since 1997. That’s 10 daily events in roughly 3900 occurrences.
No big secret here, the distribution of asset prices are NOT normal but skewed. But, here’s the thing:
We can’t request collateral against every possible outcome or the amount will be too large. Inasmuch as we don’t want systemic risk, we are risk institutions. To make money, you must risk money.
So how do we handle that “black hole of swan”? Institutions size their risk properly and stress test to ensure they’re not under estimated risk more often than their confidence interval allows.
That’s my 3 reasons why Initial Margin should not be included in BASEL III’s leverage ratio data set. J.P. Morgan did a good business unit level piece. It’s perhaps more than one wants to know about this issue, at 20 pages it’s a long piece. But it includes the calculations and some great matrices (cheat sheets) in the appendices.
At east that’s how I see it. I look forward to hearing other points of view. Please leave them in the comment section below.