October 23, 2006	Monday	Ramazan 29, 1427

November 2006 is going to be an eventful month for the stock market. It should see implementation of a number of critical directives by SECP.

The regulator is increasing cap on ‘badla’ financing of speculative trades from Rs25 billion to Rs55 billion and, perhaps as a balancing act, strengthening management of settlement risk. Within risk management measures, the one that is provoking the most debate is something which till recently was largely unknown – the netting.

In the local context, netting means offsetting sales and purchases by a broker in each security to reduce his cumulative unsettled trades called “exposure.”

Netting methodology used to calculate exposure (called netting regime) is important because all margins that exchange collects from a broker to manage risk of broker default are a function of his exposure.

Lesser netting means higher exposure which in turns means higher margins that reduce financial capacity of stock brokers to carry out speculative trading. Since brokers earn most of their brokerage commission and trading profits from speculative activity, they are highly sensitive to issues pertaining to exposures and margins.

Let’s take a simple example. If on trade day T a broker at KSE buys 10,000 shares of OGDC in regular segment at Rs135, then his exposure is Rs1.35 million. Against this exposure, he submits to exchange initial margin, say 10 per cent of his exposure or Rs135,000. If market price of OGDC falls by a rupee to Rs134, the broker also submits mark to market margin of Rs10,000, which, like the initial margin, is submitted in the form of shares rather than cash.

However, if after a little while on the same day, broker sells 10,000 OGDC shares, then his purchase is offset by his sale, that is, his exposure is netted down to nil and both his initial margin and mark to market margin are freed for further trading.

If the shares are sold at a price higher than Rs135, broker makes a profit. In case of a loss, broker submits the loss, again in the form of shares, which the exchange retains till he actually settles it in cash on settlement day T+3.

KSE does not settle netted trades in full but only their profits and losses, the size and therefore risk of settlement it assumes is greatly reduced. Since realised losses are collected in full, exchange frees up margins on netted trades.

Netting, however, turns every brokerage house into a mini clearing-house settling netted trades of its clients and bearing the risk of client defaults in internal settlements.

Brokers are supposed to take adequate margin from clients so that losses incurred by clients do not lead to broker default. Not everyone complies. To compete for commissions, some brokers allow trading on thin margins, knowing that due to netting the margins they have to submit to exchange are going to be far less than what they should be collecting from their clients.

As a result, speculators trade beyond their financial capacity, which, in volatile times, threatens to create settlement crises. In the past, some attempts were made to deal with this problem through regulations and monitoring but to no avail.

It is in this context that SECP has directed the exchanges to implement a new netting regime, eliminating a number of nettings involved in calculation of exposure. These eliminations can be divided into six types, out of which first four are to be implemented by Nov 6, 2006 and the other two by Feb 1, 2007.

These are: (i) sales by a broker in regular segment shall not be netted with the financing he provides in ‘badla’ segment (ii) where a broker acts both as a ‘badla’ financier and a financee in the same security but the financee is not his client, the two financing contracts shall not be netted (iii) in single stock futures, long (or short) positions in one contract shall not be netted with short (or long) positions in the other contract (iv) purchases/sales in a security, profits and losses that are to be settled on different settlement days shall not be netted with each other or with any other outstanding position in regular or ‘badla’ segment (v) if one client of a broker buys (or sells) a security and another client sells (or buys) it, the two trades shall not be netted in regular or futures segment and (vi) if a broker buys (or sells) a security on his own account (proprietary dealing) and then sells it on account of a client (client dealings), the two shall not be netted in either regular or futures segment.

Elimination of last two types of netting shall have the greatest impact on exposures. Probably this is why their elimination date has been scheduled three months after the date for the earlier four.

Elimination of nettings pertaining to ‘badla’ would complement other measures by SECP, including blocking of ‘badla’ financed shares and ban on in-house ‘badla.’

Together, these measures would help stop ‘badla’ financiers from pledging financed shares to raise further cash for re-lending in ‘badla’ and also address the uncertainty and risk associated with off-system financing.

Informal estimates suggest that after full implementation of new netting regime, margin requirements from brokers as a whole could rise two to four times, with ‘badla’ being hit the hardest.

Any relief provided by Value at Risk (VAR) based initial margins would probably be offset by the new special margin. Brokers who are already collecting adequate margins from their clients would not be affected much, as they would simply pass over client margins to the exchange. Others used to under-margining would have to make their adjustments.

Many brokers are unhappy because new netting regime is likely to reduce speculative trading in the market at a time when turnover has already dropped to half its level in 2005. They probably have only themselves to blame. Had they enforced adequate client level margins through self-regulation, SECP would not have to intervene in this matter.

Indeed, experience has shown that in volatile times risk of broker-level settlement and off-system ‘badla’ financing comes back to the exchange because under margined netted positions are mostly liquidated on and not off the exchange.

We should also recall that in their reports, investigators have identified excessive netting as a cause behind both May 2000 and March 2005 crises. Recently, in June 2006 crisis, it was market participants who were saying that fall in share prices was being accelerated due to under-margined netted positions and in-house ‘badla.’

In the face of such evidence, it would be irresponsible to not act against excessive netting and let the market face a new crisis due to the same old reason.

Given the political realities of stock market, particularly the influence of ‘badla’ lobby, getting the new netting regime implemented is anything but easy. While the first four types of netting seem to be heading for elimination as scheduled, we are likely to see some fireworks and re-negotiations before the last two are done away.

But regardless of the implementation pains of the new netting regime, bottom line is that it is good for the stability of the market even if it comes at the cost of a drop in speculative trading.