Credit pricing principles

helen-kane-hedge-trackers

A new respect for counterparty risk is reshaping the commercial landscape in the derivatives market. Growing numbers of banks are applying additional charges to their transactions, reflecting the cost they would bear if a counterparty collapsed while owing them money. These charges are calculated and applied differently from bank to bank, resulting in quotes that can vary by millions of dollars for the same transaction. Unsurprisingly, derivatives users are shopping around – and, for many, price remains the most important criterion, meaning those with the most conservative credit pricing risk losing out.

Japanese car manufacturer Toyota first encountered this new reality in February this year, after raising €3.75 billion in the European medium-term note market. The company has a strict policy of avoiding foreign exchange risk, so had to swap the euros into US dollars. The quotes it received from its banks were higher – and more divergent – than the firm had been expecting.

“Prior to the crisis, times were good, credit was easy, and we had enough relationships with bankers that no-one would dare charge us additional credit. But when we went into the market with these cross-currency swaps – a transaction with a bit more risk and a big size – the banks suddenly started discussing credit charges with us,” says Jonathan Cowan, national manager for the rate desk and derivatives at Toyota Financial Services in Torrance, California.

The gap between the widest quote and the tightest was 5 basis points, says Cowan, which could have added up to a $5 million difference in cost over the life of the swaps. Toyota went with the bank offering the tightest price.

Since then, the company has encountered the same thing on a regular basis. With simpler, plain vanilla trades, the credit charge will vary from zero to a single basis point, says Cowan. For something longer dated, more volatile or just bigger, Toyota could be getting quotes that include as much as a 15bp or 20bp credit component from some banks – and the difference in these quotes from one dealer to the next can be as much as 10bp, he adds.

Cowan has taken time to speak to his bankers, so he understands why this is happening: “Before the crisis, credit charges for a company as strong as ours might have been a fraction of a basis point. Now, depending on the complexity of the transaction, you are talking about a handful of basis points. And that is no longer something a bank can just eat. The banks have changed a little bit – there is more pressure on their trading desks. They basically get charged for whatever credit they extend to Toyota, so unless they are making more profit somewhere else, they will want to make sure they are covered.”

There is a fair amount of wiggle room in a system like this. As long as the relationship as a whole is compensating the bank for the risk it takes on a given trade, then it has the flexibility to reduce the credit charge – and Toyota has learned how to make this work in its favour. So, for example, if the company has a portfolio of trades with two different banks, one of which has a net mark-to-market value in Toyota’s favour while the other does not, Cowan says the company knows it will get a lower credit charge from the former. “My exposure to Bank A will be different from my exposure to Bank B, so we can be strategic about where we add that next trade to make sure we are paying the lowest costs. It’s not impossible that one bank would add a 3bp charge and another would add 13bp,” he says.

But over and above these idiosyncratic portfolio and relationship issues, there are simply some banks that are being more conservative and charging more for credit. Cowan refuses to name names but says these banks tend to be the ones that were in most trouble during the crisis and received larger chunks of taxpayer money. And they are the ones now seeing less of Toyota’s business.

“Toyota is still a very strong, highly rated company, but there are banks that price wider than others for Toyota credit, so those banks don’t see a lot of our business – and we certainly tell them the reason why,” says Cowan.

Not every company has had exactly the same experience. The treasurer at one of the 50 largest US corporates says there was a severe dislocation following the collapse of Lehman Brothers and the bail-out of American International Group, but pricing terms have now returned to normal: “Nobody has said they won’t trade with us or told us it is going to cost us a few basis points more. In the fourth quarter of last year, everyone was very guarded about using their balance sheet for anything, but subsequent to that we haven’t seen it too much and prices have gone back to where we were seeing them before the crisis.”

He does concede, however, that the company could have been unaffected because of its credit strength and its relatively simple financing and hedging needs – it rarely enters the capital markets and tends to use options to hedge its foreign exchange exposures, rather than long-dated cross-currency swaps, which represent a greater credit risk for banks.

Outside the corporate elite, smaller derivatives users are seeing higher prices for counterparty credit risk, says Helen Kane, founder and president of Hedge Trackers, a California-based derivatives consulting firm that works with middle-market customers. “Prices are higher, but there has always been a very mushy line between what is profit and what is credit. It’s not clear to these companies how that calculation works and I don’t know, on the banking side, how that is being determined these days,” she says.

The principle of what banks call a credit valuation adjustment (CVA) is fairly straightforward. “The idea is that if there is a transaction where the client owes you $1 million and the client defaults, you would want to be able to replace the transaction, and the question is how much that would cost you. Essentially, you want to buy the option to enter into the same transaction on the same terms at the time the counterparty defaults – and that is what you ask the customer to pay upfront,” explains Athanassios Diplas, global head of counterparty portfolio management at Deutsche Bank in New York.

Banks approach the actual CVA calculation by estimating two things: the chances of a customer defaulting, and the size of the exposure at default. The probability of default is typically derived from spreads in the credit default swap (CDS) market. Many banks then use the money raised by the credit charge to buy CDSs covering some or all the counterparty exposure, with the whole process of analysing risk, pricing and buying cover managed for the derivatives business by a centralised CVA desk.

Not everyone does it the same way, but a growing number of banks have the same ambition – to include some kind of margin to cover counterparty risk. At Skandinaviska Enskilda Banken (SEB), there is no CVA desk. Philip Winckle, Stockholm-based head of credit risk control, says the bank wanted to avoid the time and expense involved in setting one up, while still making sure the derivatives business was earning a high enough return to cover the counterparty exposure. So, instead of calculating a specific component to be added on to the price upfront, SEB ensures counterparty risk is bundled into its assessment of the capital required to support a transaction. The traders then need to work out a price that ensures an appropriate return on the capital they’re consuming.

A London-based senior credit risk manager at one large European bank says the market has definitely changed. “The banks that were most disciplined about this – such as JP Morgan and Goldman Sachs – were the ones least hit by the crisis. Now everyone, every bank, wants to do the same. If you are a corporate client, whichever bank you go to is going to charge you for the counterparty risk,” he notes.

In practice, this can be complicated. Each time a transaction is conducted, the bank has to calculate how the new trade affects net exposure to that counterparty. A swap that is projected to go the bank’s way may not attract a hefty CVA if existing trades are in the customer’s favour, but trying to work out these offsetting effects across a range of different trades and asset classes isn’t easy. “If it was just one transaction in isolation, then every bank would probably come up with a similar number. Where it gets more complicated is when I am owed money on an interest rate swap transaction where I am receiving fixed, but offsetting that I owe money to the same counterparty because I have done a CDS trade. People have ways of working out the correlation between those positions, but this is where differences creep in,” says Deutsche Bank’s Diplas.

In addition, the trading desks are free to decide for themselves how much of the CVA to pass on to the client. What is not at their discretion is the need to earn a return to cover the credit cost – but that could be achieved through a separate trade or even by another business area. “A lot of banks don’t mind swallowing some of the cost on a certain product if they can make it up somewhere else. So if they’re providing us with a full range of services – master credit facilities, capital market underwriting, derivatives, cash management – and are looking at Toyota as a whole, then they may decide to price a little more aggressively than they would like to in a certain area if they feel they can make it up in another,” says Cowan of Toyota Financial Services.

These differences in methodology and application account for the pricing disparities Cowan and others have encountered this year – and they are already influencing customer choices in a way that could have longer-term significance, says Aaron Brown, a risk manager at Greenwich-based hedge fund AQR Capital Management. While working at Morgan Stanley, Brown introduced the bank’s CVA regime in 2001 and 2002, in the aftermath of the last credit downturn, and he claims the firm was one of the first to use credit spread-based charges systematically. At the time, the industry tended to make rough adjustments to the price by looking at the counterparty’s credit rating and adding a spread, with the trading desks themselves often managing the process. When the approach was introduced at Morgan Stanley, the bank found itself attracting customers with a lower spread than their rating implied, while those with a higher spread would trade elsewhere, says Brown: “It did us a lot of good and we ended up cherry-picking a lot of strong business. We took a certain amount of quiet pride in that.”

The flip side, of course, is that other banks ended up with dicier credits, says Brown – and the same could be happening now as a result of the divergent prices banks are quoting. Kristian Liiv, counterparty credit risk control manager at SEB in Stockholm, says: “Banks that don’t price accurately for counterparty risk are at risk of adverse selection, and my guess is this is happening already.”

Equally, though, counterparties are under no obligation to trade with the bank offering a price that is technically the most accurate if another is offering a lower-cost transaction – so banks using a CVA approach may find themselves losing business. Deutsche’s Diplas says this was a constant headache prior to the crisis: “Some guys would just use credit lines instead and wouldn’t charge for the counterparty risk if the line wasn’t full. Effectively, you had credit risk masquerading as profits, but they would win that business. It put some of us, who were actually trying to take these things into account, in a worse commercial situation.”

His hope is the playing field will now be more level. But with methodologies varying from bank to bank – as well as the portfolios and relationships that make an institution more or less willing to soak up the credit charge elsewhere – pricing disparities are likely to remain. And the London-based senior credit risk manager warns that commercial pressures could erode the industry’s good intentions: “If you have a model that charges extensively for CVA, you won’t be in the market for that long. You’ll kill your business. So, it is an exercise in calibration, I would say.”

BOX

Collateral limitations spur credit pricing

Pricing for counterparty risk is essential because collateral – a more widely used risk mitigant – has serious limitations, some bankers argue.

If each transaction, or portfolio of transactions, was covered by an agreement to adjust collateral in real time, it would be the ideal way of mitigating risk, says Kristian Liiv, counterparty credit risk control manager at Skandinaviska Enskilda Banken (SEB) in Stockholm. But such an approach would be capital-intensive and require sophisticated back offices, which some hedge funds and most banks possess but other counterparties lack. Instead, most collateral agreements with financial institutions calculate posting requirements on a daily basis, while corporate customers might post collateral against their positions only once a month. In both situations, sudden market movements can result in collateral being insufficient to cover the position if the counterparty should default.

Even daily collateral agreements can leave a considerable amount of counterparty risk. First, no collateral is typically required until a threshold exposure level is breached, says Liiv – two banks might agree to accept the first €5 million of exposure on an uncollateralised basis, for example. In addition, collateral agreements specify a minimum transfer amount to prevent unnecessary strain on the back office. The minimum transfer amount could be as much as €5 million, meaning the default of one counterparty could leave the other with close to €10 million in uncollateralised exposure, he says.

On top of that, if a distressed counterparty fails to post collateral, the other bank has an option to call it a technical default and settle the transaction within one day of the missed transfer. In reality, says Liiv, the two banks would argue about it for a few days before any action was taken. In the meantime, the exposure could continue to grow, without collateral coming in to offset it.

One final problem is the value of the collateral itself can change dramatically – as the treasurer of one large US corporate can testify. As liquidity evaporated from credit markets in the third quarter of 2007, he noticed two securities lending counterparties had been posting structured credit to the company as collateral. The collateral agreements specified the paper was acceptable, but the company’s policies said the opposite, so the treasurer went back to both banks and asked them to rewrite the terms of the agreements and take the collateral back. One agreed to do so; the other refused.

“They said: ‘Well, we don’t want it back.’ We said that was kind of silly – they were probably making a couple of million dollars a year from this business with us, and selling the collateral in the market would have resulted in a $117,000 loss. It seemed irrational, but it became clear to us the reason they didn’t want to take it back was because they had a mountain of the stuff and they were trying to shove it everywhere,” he says. The company ended its relationship with the bank, which went out of business later in the crisis.

The bottom line, says Aaron Brown, a risk manager at hedge fund AQR Capital Management in Greenwich, is that collateral needs to be supplemented by pricing for credit risk upfront: “This is a common misconception. Collateralising a trade dramatically reduces your exposure, but it doesn’t remove it – you can definitely lose money on collateralised trades.”

Only users who have a paid subscription or are part of a corporate subscription are able to print or copy content.

To access these options, along with all other subscription benefits, please contact info@risk.net or view our subscription options here: http://subscriptions.risk.net/subscribe

You are currently unable to copy this content. Please contact info@risk.net to find out more.

Credit risk & modelling – Special report 2021

This Risk special report provides an insight on the challenges facing banks in measuring and mitigating credit risk in the current environment, and the strategies they are deploying to adapt to a more stringent regulatory approach.

The wild world of credit models

The Covid-19 pandemic has induced a kind of schizophrenia in loan-loss models. When the pandemic hit, banks overprovisioned for credit losses on the assumption that the economy would head south. But when government stimulus packages put wads of cash in…

You need to sign in to use this feature. If you don’t have a Risk.net account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here