IAS39 problems and solutions
Treasurers might be familiar with the basics of IAS39, but it's the details that will force them to change their funding and hedging strategies. Euromoney's panel of experts discusses the profound implications of this complex piece of regulation.
SD, PricewaterhouseCoopers Given the ongoing changes to IAS39 and their implications, perhaps we could start by getting Pauline to explain the latest situation.
PW, PricewaterhouseCoopers Well, the main point to make is that IAS39 is a work in progress. The standard was revised significantly at the end of December. And we have learnt that another exposure draft is expected in April that will look at whether people have the right to choose to fair value everything or not. The latest estimate of when we might see a final version of the standard is November or December 2004.
The area that's been the most controversial throughout the development of IAS39 is hedging. In that area, there were a few changes made in December to make the standard more comparable with US GAAP, but the key areas that we had hoped for change on were not changed at all.
A key factor in these delays is Europe. EU internal markets commissioner Frits Bolkestein has said that he does not intend to recommend endorsement of IAS39 unless there is clear evidence that the issues being raised by the banks and insurance companies are being resolved and until the European Central Bank is happy with the standard. So the whole thing has become hugely political.
To adopt or not to adopt? SD, PricewaterhouseCoopers Peter, perhaps you could give us a corporate view – particularly as you report under US GAAP as well.
PZ, Unilever Well, the easiest approach to IAS39 is to say: "Let's wait until the standard is final and then implement it." However, for anyone already starting to implement IFRS, it would be odd to do that and not implement IAS32 and 39. The SEC just made that decision clearer by ruling that foreign registrants in the US, like Unilever, can get an exception to the rule that they have to show two years' comparative data, as long as they implement full IAS. So the pros of doing a full IAS are getting stronger.
PW, PricewaterhouseCoopers Even for those who aren't SEC registrants there are implications: the IASB exemptions from having to go back and apply everything retrospectively with first-time adoption of IFRS1 only apply if you apply IFRS in total.
MB, Barclays Capital A lot of corporate clients I've talked to also see working towards IAS39 as part of good corporate governance.
MA, BNP Paribas In addition, even though IAS 32 and 39 may not be required by 2005, we are finding that our clients still need to consider the impacts for longer-term hedges.
SD, PricewaterhouseCoopers François, you've also got another perspective on this, being an existing IFRS reporter.
FM, RTL Yes. We were early adopters of IAS32 and 39 and we have reported under IAS for three years in a row. So, if the final decision from the EU postpones 32 and 39, it will create competitive disadvantage for the pioneers. And, in my role as head of the EACT [European Association of Corporate Treasurers], I can say that our members still have a lot of issues with the standard and that, although we have some success in lobbying the board, some fundamental parts of IAS39 – such as those regarding netting – still remain unacceptable.
SD, PricewaterhouseCoopers Nick, what's your view? Is it this a competitive issue that is important to British Airways?
NG, British Airways Yes. For us the overriding question is: "What is the rest of the sector doing?", because our results are compared with theirs. So if our competitors adopt, and Lufthansa has, the temptation will be for us to adopt. That said, adopting is very expensive and so we have to ask whether we want to take the additional risk of implementing a standard that has not been finalized.
SD, PricewaterhouseCoopers Maybe we can just touch briefly on the discussion that's gone on behind the scenes, between the corporates and the IASB. François, give us a flavour of the EACT's lobbying.
FM, RTL The ACT [UK Association of Corporate Treasurers] and the EACT, representing the majority of corporate treasury in Europe, have tried with Sebastian to lobby the board, particularly on netting and effectiveness testing [see below].
It was difficult to lobby the EU, although they were responsive. But the discussions with the IASB were very difficult because they are pretty rigid and inflexible. I got the impression that they didn't want to listen to our arguments and were surprised by the way we were handling hedging and netting positions, even though these were industry standard procedures. Also, I felt that the banks were being treated differently to the corporates even though with 7,000 corporates affected, we are in the majority.
SD, PricewaterhouseCoopers Peter, you were also at the roundtables to which François was partly referring.
Corporate issues PZ, Unilever Yes, we participated in one of the roundtable sessions and subsequently had a private session with the IASB together with Vodafone and Reed Elsevier, and I agree with François: it was quite frustrating. We all explained our risk management policies and why we have them and we explained that most corporates would share these issues with us. And basically their reply was: "Sorry, your risk management policies unfortunately do not fit into our rules". And that was that.
PW, PricewaterhouseCoopers In defence of the IASB, we should welcome the establishment of the high-level committee that has recently been set up under the auspices of the EU, with the aim of bringing people together. I suspect there aren't enough corporates represented on it – I think it's primarily banking and insurance – and that should be addressed; but at least the board has recognized that it should be talking to people at an earlier stage. Remember that the board inherited a flawed standard and they know it's flawed. That's why they've been trying to do quick fixes all the way through – which, I agree, is no way to set standards.
FM, RTL I agree. It's a first step and it was the first time to my knowledge that we had an opportunity to discuss the issues in a public hearing. But I think they need to listen harder to the corporates because we were the early adopters, not the banks.
GD, Dresdner Kleinwort Wasserstein And of course this makes a big difference. IAS39 is, in its philosophy, fair-value based, which is much more appropriate for a banking environment, with a dynamic hedging approach, as opposed to the corporate environment where you hedge cashflows, where you're doing static hedging, and where you hedge according to a benchmark. These are two different worlds and it's because of this that we have problems fitting IAS39 into the corporate world and why we have all the exceptions and difficulties with hedge accounting.
SD, PricewaterhouseCoopers Indeed and it's some of those technical issues that I'd like to turn to now after that review of developments. First, since companies will have to start the process of adopting, what are the strategic issues they will have to consider?
Strategic issues MB, Barclays Capital Well, the key point is that we are moving into a totally different environment – the fair-value environment. This will significantly change how accounts look; it will introduce much more volatility into accounts and that's a hard thing for companies to embrace. So even before making any decisions about systems and processes, companies have to review their hedge accounting policy; ask themselves: "If I continue what I'm doing at the minute, what will be the impact of IAS39 on my accounts?"; "Given that I'm going to have some volatility, how much am I prepared to accept?"; "Do I look at my peers and take my lead from them?".
The other core issue is: given its existing hedging strategies, does a company go for hedge accounting and to what extent? And to what extent should they change their hedging policies to do this? The most important thing I tell people is that the economics still have to be the driver, not the accounting.
The more sophisticated corporate clients say that if they have identified a risk or a hedging policy that doesn't qualify for hedge accounting and causes volatility, then their CFOs are comfortable with that but they want there to be a good economic reason for it.
PZ, Unilever Yes. The core problem is where to change your existing risk management policies. The problem with the volatility is twofold: the first is communication. Giving an explanation for the numbers as such should not be a huge problem. However, the downside of that may be that your financial reports will increasingly become explanations of financial structures and risk management, rather than an explanation of your underlying business performance. And that brings me to the second issue. If you have a P&L which reflects significant multi-year volatility because of derivatives positions related to, say, the pension fund or risk management activities, then your performance statement is increasingly less about this year's performance and less about actual cashflow, which in our opinion is the ultimate performance measurement.
FM, RTL Let me also try to be positive about it. IAS39 forces companies to revisit their risk management policies and to look at them in a new way. So it is not all negative.
NC, Logica CMG Absolutely. It may be an accounting standard but it's also trying to introduce a quality standard to treasury, a sort of ISO9002 for treasury. My major criticism is that there are an awful lot of words, but very few examples. That it introduces volatility is not necessarily something we should shy away from. But I'm not sure that fair-value accounting needs to go through the P&L. I think it can be handled by adequate disclosure elsewhere.
PZ, Unilever I largely agree with fair-value accounting; I don't agree with some of the hedge accounting rules, where structures designed to manage risk 10 years out affect our current year's P&L. Where hedge activities relate to future years and future risks, don't have the fair value movement booked in today's P&L, but in reserves, so that the shareholder can say: "Ah, this company uses derivatives, but when I look at the P&L I can still distil this year's performance".
MA, BNP Paribas My view is that financial statements are a communication tool. As long as all the information is in them and everyone fully understands it, it is irrelevant really where things are placed. However, I also agree that the financial statements should be an accurate reflection of the true performance of a business, and that the performance needs to be explained as clearly as possible. It is equally important that investors be educated in how to read and interpret the information.
PZ, Unilever Well, we have investors who come to us and say: "Procter [& Gamble]'s balance sheet is much stronger than yours"; and I say: "Have you also considered which accounting rules they're under?", and often they haven't.
GD, Dresdner Kleinwort Wasserstein That's interesting because I have discovered from our customers that these new standards increase communications – not just internally between treasury, accounting and auditing, but also between corporates and their bankers.
FM, RTL And corporates recognize this. An EACT survey on IAS39 adoption concluded that the principles behind the standard are consistent with treasury best practices, notably on the internal control side. Corporates agree that the documentation identifies hedges and calculates fair values for derivatives, all consistently with treasury best practice.
Hedge accounting and risk policy SD, PricewaterhouseCoopers Can we perhaps move on and get into more of the detail on hedge accounting? Melissa, how are your clients dealing with the questions: "Should we do hedge accounting or not?" and "When should we re-engineer policies in order to get hedge accounting?"
MA, BNP Paribas Well, in many situations, there are definite advantages to hedge accounting, the most obvious being that the easiest way to demonstrate to investors that you have an effective hedging policy is to have it reflected as such in your financial statements. If hedge accounting aligns with the company's management policies and treasury best practices then it is often a justifiable cost. Where it gets much more costly is when it doesn't do that and changes to hedging strategies must be made in order to comply.
Hedge accounting is a set of rules and regulations that dictate the types of risks that can be hedged and the types of products that can be used for hedging. If a company makes a blanket decision that all hedges must qualify for hedge accounting, it can lead to situations in which a treasurer may be forced to change the hedging strategy or forgo hedging. In that case, the company's financial statements may appear stable under the accounting rules but in reality there is true, underlying economic risk which is unhedged.
At the other end of the spectrum, a company could opt to forgo hedge accounting completely and accept the P&L volatility, on the basis that the accounting tail should not wag the commercial dog. This decision will lead to accounting results which will need to be explained, and the onus will certainly be on the company to ensure that the investment community understands the mismatch between accounting and economics, and that has a significant cost as well.
In our experience, best practice is not to make blanket decisions but instead is to manage risks on a much more granular level to ensure the optimal balance between economics and accounting. Where the accounting results don't line up with the underlying economics, decisions have to be taken about whether hedge accounting is appropriate, whether policies need to be changed and, if so, how. The hard question is simply this: "How much of an economic sacrifice should I make in order to be able to have the reporting that I want in my financial statements?"
MB, Barclays Capital I agree with Melissa. It's a bit like the 80/20 rule. You'll probably find that a corporate can remove 80% of the earnings volatility by applying hedge accounting to 20% of its hedging transactions. After the easy hits, where there are transactions that are not vanilla hedges but more value-added risk management solutions, then you must make a cost/benefit decision and assess whether it's worth doing all the documentation and the hedge effectiveness testing and so on.
PZ, Unilever But that kind of decision process is difficult to manage across a large company. A company is not a static organization and exposures constantly change, also across units. How do you see this working in practice?
MB, Barclays Capital In practice you identify hedges by type – generally compliant or non-compliant. A simple vanilla hedge is, say, a newly issued bond swapped into floating. These are the easy hits in terms of hedge accounting. Then, where you're hedging, say, foreign currency sales or purchases in the future where you have "highly probable" issues, you may run into difficulties getting hedge accounting, but you realize from an economic point of view that not having hedge accounting is justified. Then you must monitor the volatility, and that's where the banks can help treasurers – in quantifying what that volatility can be. And the same goes for structured derivatives, or even vanilla options, where you still have over- or under-effectiveness even if you do get between the 80–125 [percent effectiveness for a hedge required under hedge accounting]. The key thing is to realize and accept that you almost never get zero earnings volatility, there's always some there.
MA, BNP Paribas I agree. Under IAS, a company must test every hedge and report every penny of ineffectiveness in the P&L. This is very different from the US hedge accounting rules which include a 'shortcut' method of testing effectiveness and companies are not required to report ineffectiveness for many commonly used hedges.
GD, Dresdner Kleinwort Wasserstein I think I can add two things from my experience. First, there is a distinction between interest-rate and FX risk. Hedge accounting is often applied to interest-rate risk because it's pretty long term and so it creates significant accounting volatility. FX risk on the other hand is often very short term and is the result of a large number of transactions, so getting hedge accounting requires a lot of work. That's why many of our customers do not apply hedge accounting to FX.
Secondly, many of our clients over-estimate the hedge accounting issue: they are first confronted with IAS – with accounting volatility – and they think it's huge and could overwhelm annual revenues. Then, when you go into more detail with them, it's not so. So the quantification stage is very important before going ahead and applying or dismissing hedge accounting. If the financial exposures are small, then don't apply hedge accounting.
SD, PricewaterhouseCoopers You're saying that the company's earnings volatility tolerance is higher than they might have expected before they did the analysis?
GD, Dresdner Kleinwort Wasserstein Yes.
PZ, Unilever But sometimes you need to be careful. In 1999 we ran a check on the potential consolidated earnings volatility related to our FX contracts and found it was less than e25 million, which was acceptable. The many FX exposures within the company offset each other to a large extent. One could conclude therefore that hedge accounting was not needed. However, quite a few individual operating units have substantial volatility and not applying hedge accounting would distort their results. And if we tried to hedge these selectively, the portfolio offsets disappeared, creating an overall position of plus or minus e100 million. So, one cannot limit oneself to the consolidated picture only.
SD, PricewaterhouseCoopers The question here is: "Where is the volatility?" And there are two kinds. First, there's the volatility that comes from the derivatives that you're not getting hedge accounting on. Second, and nastier because it's more difficult to explain away, is the one that comes from the unhedged exposure, or the hedged exposure which isn't getting the benefit of an offset because some of the derivatives are not getting hedge accounting.
NG, British Airways For our fuel hedging, we started with the aim of continuing to do the right thing from an economic and a cashflow perspective. Then we decided we needed to be in a position to moderate the volatility, because if in a year's time we were the only guys out there with volatility in our accounts, we would want to change pretty quickly. Now, that means either changing the markets we hedge in, which would move us from a very liquid crude oil market to an illiquid jet fuel market, or building a lot more complexity into the way we hedge. That does makes our life more complicated, but it does give us the opportunity to moderate the volatility.
MA, BNP Paribas A lot of corporates are going through this analysis now. One thing to remember is that this standard is a set of rules and so form often matters; many times you can restructure your position so that although you don't achieve full compliance you're in a much better position than when you started. So it's important to consider the economic consequences of that kind of policy: is it a little economic consequence for a big benefit or is there a big economic consequence, like moving from crude to jet?
PW, PricewaterhouseCoopers Nick's point is a very valid one. Commodity hedging creates all sorts of problems, because the rules weren't designed to address it. It's here that companies have had to invent new systems which don't reflect what they want to do, but simply eliminate volatility.
SD, PricewaterhouseCoopers Yes, and such workarounds create operational risk because a layer of documentation and management and control now exists solely for the purpose of achieving a particular accounting objective that doesn't flow naturally from treasury's risk management policy.
Operational risk NC, Logica CMG Just for clarification, is the issue with your commodity hedging that though you actually purchase jet fuel, crude oil is a surrogate hedge for that?
NG, British Airways Yes. Three-quarters of the price of a gallon of jet fuel is the underlying crude, so it is a natural and reasonable hedge and we've been very successful in smoothing our P&L and cashflow using it. But if you try and do an 80–125 effectiveness test on any individual deal, it doesn't necessarily work. Now we could try to hedge directly in the jet market, but it's not liquid and our volumes would move the market. We can't afford to do that.
SD, PricewaterhouseCoopers The other common example of accounting driving policy rather than economics is when treasurers use micro-hedge strategies rather than aggregating risks and hedging them on a combined basis. If they do that, then they have to use a larger number of derivatives with characteristics more finely tuned to individual exposures.
PW, PricewaterhouseCoopers Which is precisely what we've seen happen in the US in response to FAS133. But I think it's worth saying that a lot of hedging transactions do contain real volatility and IAS39 does move you towards getting that volatility recognized. I think in many instances it actually gives you a much better view than you've got already. I think half the problem is that people don't like the view it presents, rather than that they don't think it shows the economic reality.
NC, Logica CMG But Nick's problem is that he gets his hedge effectiveness by going into a thinner market. You get the hedge accounting but you lose the economics of a good hedge.
NG, British Airways Yes.
SD, PricewaterhouseCoopers The issue is that risk management is often about risk reduction. So hedge policies are designed to achieve the level of risk that is tolerable. However, the standard is imposing a level of risk reduction – 80% – that is higher than some strategies would aim for.
MB, Barclays Capital We have focused on the pros and cons of hedge accounting and how it affects the P&L and balance sheet but what about the fair value election alternative?
MA, BNP Paribas I think that if the fair value option enables companies to avoid the challenges of hedge accounting, it's an excellent alternative. However, the new exposure draft will restrict its use. One of the requirements will be that if you want to use it you need to show that the risks are "substantially offset". So if you have a bond and you've hedged only your interest rate and not your credit spread, have you "substantially offset" the risks? If you have to go through a kind of hedge accounting test to prove it, it may end up that the fair-value option is no easier than hedge accounting.
SD, PricewaterhouseCoopers Before we move on, we have been debating how well the standard aligns with existing policies and what its effects might be on financial reporting, but has anyone seen examples of companies choosing not to hedge risks that they previously used to because of the new accounting rules?
MB, Barclays Capital We are seeing a number of clients putting their hedging on hold until they've worked out the IAS implications. Certainly the amount of hedging being done in controversial categories has dropped dramatically. One example is in hedging foreign currency earnings from subsidiaries. Here some companies have suffered from dollar weakness and believe that the accounting rules have created a poor economic result.
GD, Dresdner Kleinwort Wasserstein One thing we have seen, which was unexpected, is that options use has increased. People are using options instead of, for example, forwards, because they know that with options the P&L volatility is lower than with forwards (as options don't have the 100% delta or 100% sensitivity). People are using out-of-the money options as a sort of compromise between not hedging (because of IAS regulations) and applying hedge accounting, as this is smoother in terms of accounting volatility.
SD, PricewaterhouseCoopers We've seen in many instances companies going back to simpler products and then getting hedge accounting for them. But this idea that companies would stop hedging until they got used to the rules or until the rules stabilized or until they figured out how they were going to deal with them is very worrying.
MA, BNP Paribas It's certainly a trend. Some companies are either delaying hedging decisions, sometimes in quite volatile areas where they really are taking some significant short-term risk, or they are putting on shorter-tenor hedges than they would have in order to give themselves the chance to restructure their hedges over the longer term if necessary. Then of course they're taking risk on the back end, even though they may be hedged in the short term.
Hedge effectiveness testing SD, PricewaterhouseCoopers Let's move on now to some of the difficult technical issues that companies are struggling with. Why don't we start with hedge effectiveness testing? Gerald what are you seeing on this.
GD, Dresdner Kleinwort Wasserstein Well, the first question is usually: "Can we get rid of effectiveness testing by applying any kind of short-cut?". If not then we have to decide which effectiveness test to apply – dollar offset, variance reduction, or some form of regression test. Finally, companies must look at how the tests and existing risk management strategies fit together. Does the 80-125 effectiveness band match existing risk reduction objectives or is a narrower band more appropriate? What is the difference between the prospective tests and the retrospective tests? And so on.
SD, PricewaterhouseCoopers Nick, how have you found this working with fuel hedging?
NG, British Airways Well, within British Airways our largest mark-to-market position is on our fuel, so we have to try to crack that one first before we tackle smaller positions in interest rates and forex. For me the biggest problem is lack of guidance: we have to use effectiveness testing in an appropriate way, but what is an appropriate regression test? There's not much guidance in the standard and there is no market standard either. Comparability across the market sector is essential so we will work with the other European airlines towards a common effectiveness testing methodology but at this stage I cannot guarantee success.
SD, PricewaterhouseCoopers Can we benefit from the US experience here? What are the banks able to offer clients here?
MA, BNP Paribas This is a key area where the banks can lend their expertise and the sophistication of their own models to help to develop a hedge accounting model that the corporate can use. Now, it's hard to get one model that covers every situation – for example, Nick's problem is infrequent – but we can still assist our clients with understanding and designing the appropriate model for their hedging relationships.
MB, Barclays Capital I think we can lend corporates our expertise in analyzing the complexities of hedged positions, which is not something they have historically had to do in this detail. It's only when you try it that you realize how difficult hedge effectiveness testing is. The standards and surrounding guidance talk about "the perfect hedge", but there are very few instances of one. There are numerous potential causes of ineffectiveness even with a vanilla hedge that are not obvious until you start the effectiveness testing.
One key thing is to recognize that in many transactions you don't really need the prospective test. If the terms of the hedge and underlying more or less align you may have enough grounds to conclude that you expect the hedge to be highly effective. For retrospective testing, regression has advantages. Of course, regression is easier for existing positions and harder for new hedges because in the latter you have to answer the question: "Would it have worked on a regression basis?". But for these types of tests, regression does seem to be the winner.
SD, PricewaterhouseCoopers What about alternatives to regression such as the dollar offset test? That's the one the standard refers to and which people have tried to use initially, but have often found flawed in practice.
MA, BNP Paribas With dollar offset our clients have an issue with small variances on large notionals. The test looks at how much a hedge moves versus how much a hedge item moves and then evaluates whether they are within 80–125 of each other. If, for example, a hedge moves by one dollar, and the hedged item moves by two dollars, the dollar offset test would show that the hedge is only 50% effective, and hedge accounting would not be allowed. But, if the underlying notional is $100 million, then in reality the economic hedging relationship is almost perfect even though the test shows otherwise.
With regression analysis your test is only as good as the volume of data points that you have. If you only have a couple of days' experience, that's not necessarily going to be useful if you're looking at a 15-year hedge. People are also finding that for option-based strategies or products, tests using VAR or volatility measures are more practical.
GD, Dresdner Kleinwort Wasserstein For me it is less about the models and more about communication. We can certainly provide sophisticated models for our clients, but if the auditor won't sign them off because they come out of a black box, then they are useless. We advise clients to involve people from audit, treasury, corporate accounting and their bank from the beginning.
SD, PricewaterhouseCoopers What about software providers? What do you think of the tools they've developed in this area?
FM, RTL It is still one of the weakest points of the TMS [treasury management system] providers. We run SAP's treasury module and it's still weak, even though SAP is a strong provider. However, it is genuinely difficult for them because it is hard to define a set of standard operations that apply to all their customers. Perhaps we will end up with a mixture of systems, some supplied by traditional TMS providers and others supplied by the banks.
PW, PricewaterhouseCoopers I'm disappointed to hear that. I accept the US short-cut method helps corporates avoid these issues, but there are many interest-rate strategies that do not achieve hedge accounting under the short-cut method, even under FAS133. I thought that US systems providers must have developed systems for effectiveness testing, particularly because this requirement has been in place for a number of years.
MB, Barclays Capital We have a US GAAP specialist in our US office and I've asked them that question. The answer is that first, using the shortcut is a big thing in the US; second, which is a corollary of that, US corporates are much less frequent users of structured derivatives; and third, outside the shortcut there's the 'easy pass' methods, like G9, and G20. So in the US most corporates are trying to get into one of those three boxes as they can eliminate the requirement to use a numerical effectiveness test. It's G20 for options used in a cashflow hedge, the short-cut method for interest rate swaps and G9 for other types of cashflow hedge provided the terms of the derivative and underlying have been matched. So there's actually only a small proportion of positions going through what they call the 'long-haul' effectiveness testing method in the US, which is why systems suppliers have few examples to offer.
SD, PricewaterhouseCoopers The other aspect to this is that vendors prefer to develop solutions for specific customers rather than invest in development themselves, so they need corporates to define their requirements first. Let's move on from effectiveness testing to another topic, which I know is dear to the heart of François – the treatment of affiliate and consolidated positions.
FX hedging problems FM, RTL Yes we have a big issue with this. Corporates forecast foreign currency exposures at affiliate level, hedge them using internal derivative contracts into a central treasury function and then lay off that net position into the external market. Right now, there is a difference between IAS and US GAAP. US GAAP allows you to maintain the hedge accounting you've created at the affiliate level with the internal derivatives, provided that that position has been completely offset externally. This treatment is consistent with what corporates do in terms of managing FX risk through their central treasury, but it is not allowed under IAS. Now this is a real issue for some of the biggest corporates in Europe and as the EACT we have said that if the board does not change its position, we should definitely not adopt IAS39 as a standard. I am afraid that we haven't obtained satisfaction on this point and we will not, in my opinion. This means that again we create a competitive disadvantage for early adopters who have to comply versus those who adopt IAS maybe without 32 and 39. Nokia, Nestlé, RTL and some others will be penalized. And we have no adequate explanation for why they will not accept this netting.
PW, PricewaterhouseCoopers This is one that has puzzled me and we've supported netting because it is a practical method of approaching the management of a central treasury. I know why the board says it cannot do it, but I don't think it has ever satisfactorily proved that its explanation is relevant. It says that this netting effectively means that you defer internal gains and losses, and under normal accounting principles internal gains and losses are eliminated on consolidation.
PZ, Unilever And it's even worse, because the latest version of IAS39 does not allow hedge accounting for hedges of foreign exchange risk in forecasted intra-group transactions.
To me it seems as though it is an ideological position of the IASB that internal contracts can never be a hedged item and the board is very dogmatic about this. A large part of the funding of our subsidiaries is financed from centrally available cash. For instance, our US subsidiary is largely funded with floating dollars out of our central euro excess cash positions and it manages its interest-rate profile via interest-rate swaps. Under IAS we will not get hedge accounting for these swaps because there is no external debt. If we instead deposit our central cash with banks and our US subsidiary takes up a third-party loan, we would be able to apply hedge accounting to the swaps. In order to avoid P&L volatility you are almost forced to use third-party funding even where it is not needed.
SD, PricewaterhouseCoopers I think we're back to the point made earlier: that the economics should drive the accounting and not the other way around. One of the solutions to the central treasury issue is to execute the netting externally by creating a special purpose entity that no-one would consolidate in which the netting would take place. Now, that cannot be the sort of thing that standard-setters want to encourage, but it's the sort of structure that people may put in place to get round the rule.
FM, RTL Or you could centralize all hedging in the same bank – though this creates huge problems for both sides.
GD, Dresdner Kleinwort Wasserstein This is a really big issue. Corporates have spent all this time and money on netting and treasury centralization, and suddenly they are looking at having to decentralize it all. The result is that people just do not apply hedge accounting, they take the cost of higher accounting volatility and will wait until this issue is solved.
PZ, Unilever But if you don't apply hedge accounting, then what happens to your internal performance measurement?
SD, PricewaterhouseCoopers You're then almost forced to use a different set of books where there is hedge accounting...
PZ, Unilever Exactly, and that would definitely be the wrong thing to do.
SD, PricewaterhouseCoopers ...and is contrary to one objective of external financial reporting, which is to try to provide users with the same picture of the company as management is looking at.
PW, PricewaterhouseCoopers It's also another level of operational risk.
NC, Logica CMG Can I just ask, does this only apply to hedging that's related to those assets or liabilities you want to keep off balance sheet? It wouldn't apply to, say, sales ledger balances which you wanted to hedge centrally?
SD, PricewaterhouseCoopers The problem is with hedges of forecasted transactions because there's no natural offset. Monetary assets and liabilities, like payables and receivables, are not an issue to the same extent, because there is a natural offset.
Now I'm aware that one or more of the US banks are offering corporates the opportunity of hedging their exposures on a gross basis but getting a rebate on the spread.
MB, Barclays Capital Yes, we've had that enquiry – to execute several derivative contracts that provide a one-for-one hedge for each gross exposure, with the pricing for the package based on an equivalent net contract.
SD, PricewaterhouseCoopers What about other issues on FX hedging? Gerald, what are you seeing here?
Translation hedging GD, Dresdner Kleinwort Wasserstein In FX there are two things that are a little different from the general hedge accounting framework. One is forecasted transactions, the other is net investment hedging – translation risk hedging: first, is it possible? secondly should corporates do it? What does it mean for accounting volatility? What does it mean for equity volatility?
SD, PricewaterhouseCoopers Can we pick up on translation hedging first? What are the issues?
MB, Barclays Capital To be clear, this is the retranslation of, say, dollar earnings from dollar subsidiaries brought into the consolidated P&L. The debate is: does the group have a risk? Now you can say that this FX risk will affect the consolidated P&L so there is a risk that can be hedged at the consolidated level. But can it be done?
PW, PricewaterhouseCoopers Unfortunately the position is clear: the normal practice of trying to hedge your net income, or even trying to link that back to your dollar sales and your dollar sub and hedging that, is without doubt not permitted under either IAS39 or FAS133.
MA, BNP Paribas That said, for those who do want to hedge this type of exposure, solutions revolve around finding exposures that achieve an offsetting result, without applying hedge accounting – effectively providing a natural hedge – because direct hedge accounting with the derivative doesn't work.
SD, PricewaterhouseCoopers Often, though, you've still got the problem of P&L geography, because you're not going to get the effects of the hedge into the right line.
MA, BNP Paribas Quite often that's the case, and if you do get it in the right line, it's because you've used a complex transaction and that has a management cost.
SD, PricewaterhouseCoopers And again it's a layer of complexity which is only there to get the accounting result, rather than reflecting an economic risk management requirement.
MA, BNP Paribas That's right.
GD, Dresdner Kleinwort Wasserstein In this area, our clients are less concerned with whether hedge accounting can be applied and more concerned with whether they need to hedge the risk in the first place. If they decide to hedge, which is a growing trend, they try to use instruments that reduce the accounting-induced volatility as much as possible, such as average-rate or double average-rate structures. But what drives the decision is, say, looking at euro/dollar volatility of 15% and a 20% weaker dollar, not the much smaller volatility created by hedging.
PW, PricewaterhouseCoopers Another related issue is that of managing the FX risk of inter-company sales between subsidiaries with different functional currencies. Clearly, there is an exposure in terms of risk on the consolidated gross margin and no apparent means of hedging it. Traditionally, under IAS39 before, supported by an IGC [Implementation Guidance Committee ruling], instead of hedging the external transactions, companies hedged the internal forecasted sales, because obviously an internal transaction between two entities with different functional currencies gives rise to an exposure in the income statement. So, as far as I'm aware, it's being done relatively frequently by existing IAS preparers, and it's explicitly permitted under FAS133. In the process of finalizing IAS39, the IGC in question was removed (IGC 137-14) and a paragraph introduced (paragraph 80) which explicitly says you cannot hedge inter-company forecasted transactions. The only exception is where a monetary asset or liability arises, creating an FX exposure. But it explicitly prohibits all the normal practices used for managing this exposure. I can only speculate as to why that happened and my guess is it was an oversight. We believe this has a huge effect, not only on existing IAS preparers, but also on first-time adopters. And it's an instance of non-convergence with US GAAP.
PZ, Unilever This is a big issue for us as a foreign US registrant.
MA, BNP Paribas We have a list of clients struggling with this, some of whom have written comment letters themselves very recently.
SD, PricewaterhouseCoopers Gerald, you also mentioned the difficulties your customers are having with hedges of forecasted transactions.
GD, Dresdner Kleinwort Wasserstein Yes. The first problem is deciding the likelihood that a transaction in the future will occur. Is it enough to refer to historical data? How can we specify that it's likely enough for the standard? Then there is the problem of what products to use to hedge it. Do we use symmetric products like forwards, or do we use options ?
SD, PricewaterhouseCoopers The complication is that companies often use options when there's uncertainty in the forecast but the standard demands that you first look at the forecast and prove that an exposure is highly probable and only after that allows you to determine which instrument you can use..
GD, Dresdner Kleinwort Wasserstein And in FX the more forecasted transactions a corporate has, the higher the accounting volatility.
SD, PricewaterhouseCoopers There are two issues aren't there? First, the specific issue of tender hedging and second, the more general point that forecasted transactions normally need hedge accounting because there's no natural offset.
NC, Logica CMG I have a tender hedging problem: at any one time we have, say, 20 tenders out and we win about a third of them. So what I need is a single option on all the contracts. Otherwise I'd end up with 20 or 30 option contracts which, because tender periods can be extended, I'd be likely to have to extend and alter. And there is one insurance company which will provide a hedge such that if you win a contract you get a guaranteed exchange rate, if you don't win the contract then you can walk away from the contract.
MA, BNP Paribas The difficulty will be in making sure that that is treated as an insurance contract and not a derivative.
MB, Barclays Capital One could use the "hypothetical derivative" argument here, because whatever the outcome, you have the hedge if you need it. So it short-circuits the need to be "highly probable", because if it doesn't happen, you don't have the pay-off of the hedge. It's quite an interesting concept.
MA, BNP Paribas Exactly. The other solutions I've seen are very similar – having two outcomes and making sure that the derivative is only there to the extent that you need it, and even making sure the cost of that derivative, to some extent, is only incurred if you need the hedge.
GD, Dresdner Kleinwort Wasserstein Along those lines, an old product is now gaining popularity because of IAS: compound options. These have a very low initial cost and are not very sensitive to spot fluctuations so you eliminate the need for hedge accounting and all the discussions. You just choose the products, knowing that you would never use hedge accounting with them.
NC, Logica CMG I would be a bit against compound options because I think it might be difficult to extend the maturity on a competitive basis.
NG, British Airways I think some of those structures might be accounting friendly, but they're not very accounting system friendly. Somebody somewhere has got to mark the derivative to market every month and they're going to be on the phone to the bank to do it every time.
MB, Barclays Capital I'd like to get other people's views as well. For FX cashflows with timing uncertainty, I see a general consensus that the spot-spot method seems best, because even if you do have a change in the timing, because you're not assessing effectiveness on the forward price basis, you end up still being highly effective. Is there any general consensus for forward-forward or spot-spot in terms of cashflow hedging?
FM, RTL It depends on the underlying transactions. We use both methods depending on our cashflow forecast (under the cashflow hedge model). If it is a firm commitment as defined by IAS 39 (with a defined and precise date for payment) you can apply the forward-forward method. If not, you should use spot-spot which is of course always 100% effective. It eases the effectiveness testing while leaving a potential volatility on swap points (the swap points will be considered as ineffective and will directly impact the P&L account). If you decide to opt for the other method, you will reduce the potential P&L volatility that CFO's and treasurers usually try to avoid.
SD, PricewaterhouseCoopers The advantage you have over many companies is that you identify the hedged item as a specific contract, as opposed to being 'my forecasted sales for a particular time bucket'...
FM, RTL Yes.
SD, PricewaterhouseCoopers So you're able to objectively identify the hedged item, which I think is the challenge here.
MA, BNP Paribas I think for most of our clients, to the extent that the forward point volatility is tolerable, the spot-spot method provides a much easier way to track and ensure that you're going to continue to get hedge accounting throughout the life of your hedges. But when you're talking about currency pairs, where you have huge forward point differentials, the consideration could be significantly different.
SD, PricewaterhouseCoopers This perhaps brings us on to the other area of complexity with hedging forecasted transactions, which is the processing, the generation of the accounting entries and, in particular, the release of the hedging reserve.
PZ, Unilever I think this is typically an area where you need advice from your auditors, in order to agree on practical solutions for issues such as the release of cumulative gains & losses from reserves to P&L related to intra-group deliveries. By the way, many of the discussions centre around foreign exchange, whereas much of the increased volatility may actually come from interest rate derivatives. We have decoupled our interest rate management from our funding plan. All subsidiaries are funded floating in their local currency, mainly sourced from central treasury funds. When these subsidiaries want to fix their interest costs they buy interest-rate swaps in the market. Here IAS creates another problem. When the currency of the original source of funding is not the same as the currency of the ultimate need, you will no longer be able to hedge account the swap against the external debt, simply because the currencies differ. This may lead corporates to source directly in the ultimate currency and not to look which market offers the cheapest funding. We have explained this to IASB. They understand our issue but cannot offer a solution that works within their rules.
Options and structured products SD, PricewaterhouseCoopers Let's move on to the issues created by hedging with options and structured products. Melissa, I think this is an area where you have done a lot of work with your clients?
MA, BNP Paribas Yes we have. Clearly, from an economic perspective, there are many situations where using an option-based or structured hedge will provide an economically superior position to a plain vanilla forward or swap. In fact, in some situations, using a forward or swap may actually increase risk instead of reducing it.
From an accounting perspective, there is very little guidance in IAS39 on how to assess more complex products. However, in our experience, we've found that many of these types of hedges will qualify for hedge accounting in full or in part, especially if some of the US guidance is applied. It's sometimes a straightforward assessment, but at other times it requires ensuring that the transaction is structured appropriately. As such, the cost/benefit of hedge accounting is something that should be considered on a case by case basis to determine whether achieving hedge accounting is critical enough to influence the economic hedging decisions.
MB, Barclays Capital Just one more point. You find quite often that structured products only contain one small feature that makes them structured – such as a knock-in forward – where just one thing makes them imperfect. In these cases, it's often enough to separately transact the two parts – in this example, do the basic forward – and you capture 95% of the volatility of the transaction, therefore qualifying for hedge accounting. The remainder, in this case a barrier option, will then fit within the company's appetite for P&L volatility.
SD, PricewaterhouseCoopers Are people not finding that in spite of that, companies are shying away from using more structured products?
FM, RTL We at the EACT surveyed corporates and found initially that that was true. Companies tested their systems and processes with simple products. I am sure that we will enter a second phase in which the banks will be able to provide more sophisticated solutions as well as helping with advice on the way to handle them.
SD, PricewaterhouseCoopers At that point I would like to thank everyone for an interesting and topical discussion, one that will continue even after IAS39 is finalized and will become an integral part of any discussion on financial risk management.
Melissa Allen (MA) Structurer, Global Risk Solutions Group, BNP Paribas, resident accounting specialist for fixed-income products. Also Chair of the European Accounting Committee for ISDA (the International Swaps and Derivatives Association).
Mark Baillie (MB) Associate Director, Financing Solutions Group, Barclays Capital, specializes in the accounting treatment of structured fixed-income and equity products.
Neil Cotter (NC) Group Treasurer, Logica CMG, the largest UK-based IT services company.
Gerald Dannhäuser (GD) Head of Frankfurt Foreign Exchange Sales and Advisory, Dresdner Kleinwort Wasserstein, responsible for optimizing risk management strategies including complying with IAS.
Nick Goddard (NG) Treasury Controller, British Airways, currently analyzing the impact of IAS39 on the company's interest rate, foreign exchange and fuel hedging.
François Masquelier (FM) Head of Corporate Finance & Treasury at media company RTL, and Chairman of the European Association of Corporate Treasurers.
Sebastian di Paola (SD) Partner, Corporate Treasury Solutions Group, at PricewaterhouseCoopers. Sebastian is based in Brussels and has spent much of the past five years helping companies to get their treasury operations compliant with IAS39 and its US counterpart, FAS133.
Pauline Wallace (PW) Senior Partner, Global Corporate Reporting Group, PwC with responsibility for IFRS accounting policy for financial instruments and financial institutions. Previously a member of the implementation guidance committee for the original IAS39.
Peter Zegger (PZ) Corporate Centre Controller, Unilever, responsible for Control in the Parent companies and the treasury back office, with previous positions in Treasury and finance director of Unilever subsidiaries.