How does a fund manager make sure it matters to an investment bank? Before the crisis, the trick was to maximise the revenue generated through execution, stock borrowing, financing, clearing of futures, bi-lateral clearing of swaps. It helped to maximise the internalisation of transactions by parking entire portfolios in custody with the prime broker, and leaving ample excess collateral with the clearing arm of the same investment bank.

What matters now is rather different: it is the proportion of the balance sheet eaten by the fund manager that counts. With the cost of capital to a bank running at, say, 12 per cent, an investment bank needs to make a return of at least 15 percent on risk- weighted capital to keep its shareholders engaged. An alarmingly high proportion of the business hedge fund managers transact with investment banks falls well short of that threshold.

Swap clearing is the latest line of business to be examined in the light of capital consumption. It is probably returning somewhere between 1½ and 3 per cent to the banks at the moment. The supplementary leverage ratio presents investment banks with an unwelcome challenge to their longstanding ability to net OTC derivative credits and debits before declaring a number on their balance sheet. They can still net the positions, but not in as generous a way as before. This implies that they will have to charge fund managers more for the OTC derivatives business they intermediate.

Since the SwapClear service provided by LCH.Clearnet admitted buy-side counterparties in 2009, fund managers have been able to clear credit and rates swaps through a central counterparty clearing house (CCP) and they could count on strong interest from clearing brokers in supporting them. For the clearing brokers, clearing buy-side trades through CCPs such as SwapClear, CME, Eurex and ICEClear initially seemed necessary, if not highly attractive.

The terms of clearing agreements were generous to fund managers, chiefly because clearing brokers expected clearing to generate additional revenues from collateral transformation trades, foreign exchange and other ancillary business. Above all, clearing brokers were fearful of losing their exchange-traded and swap execution business if they did not add a swap clearing service to their repertoire.

History has turned out differently, because mandatory clearing has interacted with Basel III in a malign way. The buy-side has cleared some of its swap business through CCPs on a mandatory basis in the United States since 2013, and mandatory clearing under the European Market Infrastructure Regulation (EMIR) is scheduled to start for larger alternative fund managers in the first quarter of 2016 and smaller managers toward the end of next year (corporates and pension funds will follow in early 2018).

"There is now pressure on the clearing brokers to turn a profit," says Tom Lodge, a partner at Catalyst in London. "They are reassessing their business models in the light of the leverage ratio." Historically, clearing have not had to worry about return on capital and, even after regulators introduced mandatory clearing, they were anticipating incentives to clear business on behalf of clients. Thanks to the Basel III capital adequacy regime, what that have actually got is the opposite: concrete disincentives to clear client business.

First, mandatory clearing has encouraged fund managers to appoint multiple clearing brokers to cover the risk of a counterparty default. As a result, far from protecting their existing execution franchises, mandatory clearing has cost clearing brokers execution business, while forcing them to maintain a swap clearing capability.

With electronic swap trading platforms already available to the buy-side in the United States, and scheduled to open for business in Europe in 2017, the execution business of fund managers is increasingly likely to bypass execution brokers and go direct.

It follows that clearing swaps in the hope of retaining execution business is a foolhardy strategy for a bank. Several - Bank of New York Mellon, RBS and State Street - have already announced that they have abandoned their earlier plans to clear swaps for fund management clients.

Secondly, while clearing through a CCP is attractive to indirect members such as fund managers, Basel III imposes a triple capital burden on clearing brokers who intermediate their business with CCPs. They have to contribute to the default management fund; allocate capital to the exposure of their clients to the CCPs; and indemnify clients against any losses incurred from using a CCP.

"It costs clearing brokers a lot of money to write the business, and they do not make an economic return on it," explains Stephen Loosley, also a partner at Catalyst. "In the new world, clearing brokers are going to have to charge economic fees to make their clearing business profitable – execution subsidy alone will not be worthwhile.”

For fund managers with swap portfolios, those fees will be non-trivial. One clearing broker told a Bloomberg webinar recently that he expected to increase his clearing fees eight-fold this year. Estimates of the possible increase in fees for clearing exchange-traded derivatives range even higher, at up to 20 times their current level. An eight-fold increase in clearing fees for a hedge fund with a large portfolio of interest rate swaps paying $4 million a year today is looking at up to $40 million.

The costs might not end there either. Once electronic trading begins, clearing brokers will have to declare their credit limits to the trading platforms, to ensure that every trade is certain to clear. This will formalise existing practice. Every investment bank already caps its overall exposure to clients. Any fund manager that eats more of the balance sheet in swap clearing will be able to do less business with a bank on the short side and in financing.

The way to escape such a squeeze, and a potentially massive escalation in clearing costs alongside it, says Tom Lodge, is portfolio compression. "Fund managers need to optimise their rates portfolios to minimise the use of their clearing brokers' financial resources," he explains. "At the moment, buy-side swap traders do not have the same incentive as their sell-side counterparts - a direct hit on their capital resources - to economise on their consumption of the balance sheet. They need to get a lot more sophisticated. There are huge opportunities for them to reduce their gross usage of the capital of the investment banks on their rates portfolios alone."

Netting is not a sophisticated solution. Although it can be straightforward (two longs of $5 million for five years plus one short of $5 million for five years amounts to a net long position of $5 million for five years – it is inevitably complicated by mismatched investment strategies, coupons, maturities and currencies. The solutions, such as blended coupons, are not popular with fund managers.

Even where swap positions can be netted in a simple manner, because the positions are well-matched, the emphasis of the leverage ratio on gross exposures is forcing investment banks to maintain multiple gross exposures to fund managers. This is reducing the balance sheet clearing brokers can make available to each client.

A more effective solution, says Tom Lodge, is multi-lateral portfolio compression. It is a more complex calculation than netting, and entails looking at every swaps position maintained by a fund manager rather than those of any one portfolio, but has a much higher tolerance of mismatches.

"True portfolio compression looks at the overall exposure a trader wants to express, and takes the interest rate swaps that trader has already, and minimises them to express the same degree of risk,” explains Lodge. “It can make a very material difference to the amount if fees a manager had to pay. We estimate that compression can save managers 80 per cent of the cost of increases in clearing fees."

If fees have to rise from $4 million a year to $40 million for a clearing broker to generate a return of more than 12 per cent on risk-weighted capital, that amounts to a saving of $32 million a year. True, $8 million a year still amounts to twice as much as the manager is paying today. If these figures are close to reality for some fund managers – and they probably are - it is surprising that this issue has not attracted more attention.

One reason for the lack of urgency is that the providers of multi-lateral portfolio compression services - once restricted to the TriResolve service offered by TriOptima, but now expanded to include in-house compression offerings from both SwapClear and CME Clearing – have so far concentrated their efforts on the sell-side.

More importantly, it is only in the last 12 months or so that the investment banks have started to deal with the consequences of Basel III – notably by inviting less remunerative clients to take their business elsewhere.

In the swaps business, even this effect is being delayed. The electronic trading platforms have yet to eat into the execution revenues of the brokers (in the United States, SEFs are as yet transacting low volumes of business), so execution is still a factor in maintaining clearing relationships. “Once clearing becomes a stand-alone business, it will have to pay for itself,” warns Loosley.

There is a growing need for urgency. Clearing brokers are already looking to renegotiate the clearing agreements they drew up with fund managers in the headier atmosphere of 2009-10. The prudent fund manager, say Loosley and Lodge, will put in place the operational infrastructure to reduce the potentially massive negative impact on clearing costs by compressing their portfolios.

Dominic Hobson