FX debate (part two of two): The changing face of overlay

FX debate: Alpha quest drives FX market growth (Part one)

FX debate: Participants

Euromoney Has the argument about the value of active currency management been won?

HDH, Overlay: Yes. The equity managers are focusing on their core business, which is selecting the best stocks, and the asset management company or their clients are selecting a currency overlay manager to manage the risk and hence the return. We’ve even seen over the past couple of years a combination of passive mandates and portable alpha and currency, which is quite interesting, because that means the investors can have a strategic passive hedging ratio and, on the other hand, it can benefit from a portable alpha, derived from absolute-return currency programmes, of which he can calibrate the risk and return he wants to get and to be added on to the hedging process.

RG, UBS: We are also seeing our clients getting those types of mandates, where the passive portion is being separated from the alpha. We also see that this alpha is being packaged in both managed accounts and funds.

HS, RBC: Exactly. As there has been more international diversification there has also been a realization that there exists an inherent continuing currency risk that needs to be managed. Corporate foreign exchange risk is a by-product of what a company does. It’s not a risk in terms of an investment decision that a company has made, unless of course it’s hedging capital in subsidiaries.

AE, Millennium: I agree. We have now won the argument about the value of active currency management. The question has moved to what sort of currency manager suits your objectives and whether you want a systematic or discretionary manager or a combination of both? Historically, most clients that want to allocate money will go for a combination of styles, as this gives better diversification of risk and returns. For example, discretionary managers outperformed in April and May and carry strategies during the summer when volatility was low.

HDH, Overlay: We have also seen more clients relying on banks’ research to build their multi-manager programmes, classifying managers by style and track record, of course, but combining them, creating a coalition of managers, and that has been very useful for the clients and for us.

GK, Bank of America: The work that my colleague Amy Middleton has done illustrates that the behaviour of discretionary managers is more difficult to quantify. Yet we also found that distinguishing behaviour between trend-following and fundamental managers is critical to the asset allocation decision. Given the recent market conditions we expect to see further capital allocated to fundamental strategies, and we track this closely.

HB, JPMorgan: There’s a premise in asset management that if you can increase the breadth of the decisions you make you will get better outcomes. So people have removed constraints from their original currency overlay mandates so that you can go long and short the full range of currencies. The logical next step is to put that in a commingled fund structure, because then it doesn’t matter what your assets are or what your base currency is. So more people are opting to go down a fund route, which makes the accounting easier. Once you go down the fund route, you want to put in as little cash as possible to get as much return as possible, so it then becomes a question of how much leverage you’re prepared to take.

RG, UBS: In fact we come across more and more high-levered funds (leverage higher than the usual maximum of five). We’ve seen a considerable number of funds like this being planned, and some of the existing ones are already closed to new investors.

Ubiquitous alpha-seeking

Euromoney The real money manager has historically used currency overlay risk management, but more recently has moved towards alpha total return, and basically created internal hedge funds. How much of the institutional money management growth has been in the overlay side, compared with the total-return product?

MS, Pareto: Well it’s difficult to distinguish between them sometimes, because the overlay product can be either active or passive in itself. You could be managing a pre-existing risk, which is what we think of as overlay, which comes from the international investment. If you do that actively it’s also an alpha product. Or you can be disregarding a pre-existing risk, taking new risk in the currency market, in which case you’ve got an absolute-return product. So we would like to distinguish between risk-reducing activities and risk-taking activities, and we’ve seen both expand very significantly over the past two or three years.

JS, Deutsche Bank: From the banks’ perspective the risk-reducing trades look like index-tracking business. The value added to this business is low transaction costs and high operational efficiency. These are not the only skills required for risk-taking or alpha production.

MS, Pareto: No, the risk-reducing product is defined by what it is overlaying. An institutional investor has international investments, and therefore a pre-existing currency risk. I’d consider managing that risk to be a risk-reducing activity. It can be active, so it can be adding value, generating alpha, but it’s based on what was there in the first place, as opposed to trying to make money in the currency market, regardless of other investments.

AE, Millennium: In effect you have the dual aim of reducing risk while increasing returns.

HB, JPMorgan: You can’t really have a successful business model just based on reducing risk. You’ve got to be able to provide value added as well.

HS, RBC: Otherwise you’re just a consultant.

MS, Pareto: You’ve got to have some value added, but I think the big difference is the nature of that value added, and it’s based on the exposure to, let’s say, the client’s base currency. If you have a pure absolute-return strategy you want to take lots of bets in different currencies to try to produce a very steady stream of alpha. If you can make 300 basis points a year every year, that would be a great track record to have. Whereas if you’re managing pre-existing risk, you still want to generate alpha but you want to generate more alpha during periods when somebody’s base currency is strong, because that’s when they’re suffering in their international investment. It’s less important to generate it when the base currency is weak, because they’re already making money. They are different types of strategy. It’s still a value-added strategy, it might even still have the same value added over a complete cycle, but it has a different shape to the distribution of returns, and that’s because it’s targeting a pre-existing risk. So that’s why we like to separate those two activities, because they have different objectives.

Leverage creep

Euromoney What about leverage? It’s presumably not used in overlay but when it comes to hedge funds it’s an essential investment strategy if expected returns are to be achieved.

HB, JPMorgan: The definition gets a bit murky because some overlay mandates will allow you to sell more Canadian dollars, for example, than existing underlying assets. Is that leverage?

RG, UBS: Yes, and if you consider that collateral requirements to run some of these strategies on managed accounts are sometimes in the order of 5% or below, for alpha strategies, then leverage can be 20% or higher.

MS, Pareto: If we were to be really pure about it, you could say an overlay activity would have no leverage at all. All you could ever do would be to vary the hedge ratio between nought and 100 and either hedge back to the base currency or not, and that would be your entire opportunity set for adding value. But what’s happened over the years is that managers have recognized that they can make more money if they’re allowed to sell more foreign currency than they’ve got invested or allowed to go short the base currency, or if allowed to take cross-rate positions, and it’s still being called an overlay activity. But we try to keep the overlay activity pure and have a separate activity, which is the absolute-return activity, because then we can see whether each is fulfilling its objective or not. As soon as you mix them up you don’t know which one is making the right contribution.

HS, RBC: The decision to employ overlay is in and of itself a leveraged activity, because you are employing FX as an asset class separate and distinct from what you’re investing in. So if you look at it in that sense you’re producing another possible source of return or loss in your portfolio, without funding it, and that is leverage.

MS, Pareto: Generally speaking, these strategies are not funded at all, so you could say they’re infinitely leveraged!

Euromoney The other way to introduce leverage is through derivatives. Are you seeing greater client sophistication here?

GK, Bank of America: We have found those who have been comfortable taking risk with vanilla options have become more comfortable with exotic products. It’s a stepping-stone process. Education plays a big part. We spend a lot of time explaining products, but the objective for most total-return managers is to work out whether a market is going up or down. The trade is really the best risk/reward derivative for the directional view, as opposed to the derivative being an end in itself.

HB, JPMorgan: One structural barrier to greater use of derivatives is the custodians of the real money who are struggling with the volumes of derivatives that asset managers want to send their way. And the other barrier is just the willingness of some trustees to allow you to do things that are new to them. So that goes back to the education.

RG, UBS: It is true that there has been a block in the past and probably will continue to be for the foreseeable future, but the fact that Ucits III, for instance, will allow funds to go short, to leverage and, importantly, to use derivatives, is an important signal for the market that we can now probably broaden the derivative space further. Even at the vanilla end we’re seeing a wider interest in using options, even with volatility at these levels. If you then break the asset management segment down into sub-segments – insurance, pension funds and real money – you can now see opportunities where it is possible to incorporate more sophisticated hedging solutions, including options. So you can see that flexibility helping asset managers to hedge their risks more appropriately.

AE, Millennium: The bottom line is that there is a huge future for the use of options in managing currency risk as well as making money. However, knowledge is still low among the majority of the customer base so until that is increased the main users will be the hedge funds. Nevertheless, I do expect this to change more quickly than most of us would expect.

A lack of understanding

Euromoney Do clients fully understand the leveraged nature of foreign exchange overlay and alpha-seeking strategies?

HB, JPMorgan: We recently came second in a beauty parade with a 20% tracking-error product, to an asset manager who was offering a 40% tracking-error product. In my view, a 40% tracking-error product has got a very high probability of going to zero at some stage, and I’m not sure the client understood the nature of the risk they were taking. It worries me that the flip side of this low-volatility environment is that people are taking much more currency risk than they were before.

MS, Pareto: This is a big problem with currency, and it’s because it’s not a funded strategy. A 20% tracking-error product could be exactly the same as a 40% tracking-error product, because it all depends what you put down as the notional amount that you’re referring to, your denominator.

HS, RBC: If 20% or 40% is what you’re targeting to track, the outcome could be very different.

HB, JPMorgan: Let’s assume a constant denominator, so that you are looking at something that is linearly twice as risky.

MS, Pareto: In that case, the comparison between the two should be information ratios or something like that.

HB, JPMorgan: I think the comparison should be the amount you can lose when you’re drawing down. That’s what you really want to look at, because that’s where the prime broker is going to take the margin away.

MS, Pareto: If your 40% product has twice the return of the 20% the outcome could be exactly the same. You can’t just look at either the tracking error or the return. You have to look at them both together.

HB, JPMorgan: You have to look at the down side. They shouldn’t choose tracking error as their only measure of risk.

MS, Pareto: But if they have, it’s only meaningful if they’re putting it in the context of return as well, because there is nothing else that is common. They’re both defined by the notional amount, by the denominator.

HB, JPMorgan: Well, I’m assuming in this case that the information ratios were similar and that the client just wanted to get as much bang as possible in this fund for their buck, but they weren’t looking closely enough at the potential for periods of underperformance to wipe them out.

HDH, Overlay: Yes, we’ve had some clients asking us for a target return of between 20% and 30%, but at the level of the global portfolio in fact the allocation was smaller. It’s roughly the same. Instead of allocating 10% of their global portfolio to this kind of programme, they were allocating 5%.

RG, UBS: Recent experience tells us that demand is building for diversification into higher-leverage mandates but what sort of underlying client is asking for such leverage? If it is the pension fund market or indeed the local authority market then earlier concerns about the ability or desire of these historically risk-averse investors to transgress their portfolios through risk diversification is a new paradigm and certainly a breakthrough.

HB, JPMorgan: They are the sort of investors that you wouldn’t expect to be buying those kinds of products.

RG, UBS: Well maybe, but it is almost certainly with the backing of the consultants.

HDH, Overlay: If they calculate the percentage in their total portfolio correctly, it’s not a problem. There may be a lower fee but, at the level of the portfolio, the risk is still acceptable.

HB, JPMorgan: Right, but you have to agree that the risk at 40% of your losing your full investment is higher than the risk at 20%.

MS, Pareto: The concept that we probably want, but which hasn’t been picked up by the market yet, is one of contingent capital. Because there’s no capital being invested, you’re giving a manager the facility to take risk. Even in the funded product it depends on leverage, and in an unfunded product it depends on the number that you put down. But in every case there is an underlying concept that asks how much contingent capital is being put at risk by the fund. And that seems to me to be the denominator.

AB, HSBC: People don’t enter into an investment decision on the assumption that they will lose. Everybody looks for return above a reasonable benchmark.

GK, Bank of America: The concept of risk transfer is interesting, because the ultimate responsibility for closing out that transaction lies with the prime broker.

HB, JPMorgan: Yes, and they will do it long before you get to zero in the fund.

GK, Bank of America: Our prime brokerage team are finding there is appetite for products to guarantee a stop loss, but it’s generally associated with a portfolio of FX allocations. So it then becomes a question of whether you get an acceptable level of diversification in a low-volatility environment, or if we all just wait for the high-volatility environment and hope that the portfolio of managers are on the up side rather than the down side.

HB, JPMorgan: I would expect to see some fabulous performance. You’d have the up side of the distribution. Masses of new money would pile in at that peak and, in my experience, when people lose more than about 15% of their investment they start to feel acute pain.

AB, HSBC: Perhaps that’s where the FX market is at the moment. We’ve seen fairly significant growth over the past few years, which has attracted a huge amount of resources. One of the reasons for the lack of volatility in the market is that a lot of people subscribe to the dollar-lower view. There have been other asset markets that haven’t performed as well and people have become slightly risk averse, so central banks have more influence on the market. As we go into the New Year I think the scenario will change completely, but it really depends on whether people continue to commit themselves to FX as an asset class. I worry that the growth of the hedge fund community might not be sustainable if we see returns lower than those from alternative forms of risk-free investment.

AE, Millennium: I disagree. There is so much money looking to enter the alternative investment arena in a variety of products that the inflow will find its way into the business. In addition, I see very little money flowing out as even the more conservative investor base is increasing its allocation to the alternative space.

HB, JPMorgan: It’s a three-year cycle for most of this money. Look at the flow that went into commodities, and the implied carry on commodities now is minus 38% or something like that.

AB, HSBC: I’m certainly not bearish for the industry. I’m cautiously optimistic.

JS, Deutsche Bank: I’m still a little bullish, for two reasons. One, the market was tough to trade with low volatility and no trend. Secondly, a poor year for the FX market has been kind to those invested in a broad collection of managers. And 2006 was down only 2% for many of the indices.

Custodian versus prime broker

AE, Millennium: What about prime brokerage? Are you seeing any movement from investors that historically used custodians into your prime brokerage product?

RG, UBS: I don’t think real money asset managers will break free of the custodian relationship, and might not necessarily be trying to. I do think, though, that the product that prime brokerage offers is interesting, given what custody has always offered in the past. Prime brokerage today has witnessed significant sums being invested in IT in order to make the product truly efficient in processing, credit intermediation and netting. In addition, and most important, prime brokerage does not limit the client to the choice of banks that he would like to deal with. In a “best execution” environment this is essential. I therefore believe that prime brokerage will prove attractive to more than just the hedge fund area of our business. It is a product of the future for the real money side of the business too. It has started already in Asia and in the US and although it is relatively new to the UK I see no reason why real money can’t adopt prime brokerage wholesale.

GK, Bank of America: Within the prime brokerage space there’s no longer such a thing as a cookie-cutter product – every prime brokerage account is different. We focus on staff retention within our prime brokerage group because if every client set-up is different, losing staff can kill a client relationship. We’ve only lost one staff member from the prime brokerage team in the last seven years. We feel this is why we clear more FX PB volume than any other bank. That’s a big differentiator. As the business is changing, we are also working with more managed accounts for real money, as opposed to just a traditional hedge fund relationship. So having the staff in place that can manage that growth is a key part of the prime brokerage business.

HS, RBC: Some traditional money managers are setting up hedge fund products, and that’s exactly what’s happening in our prime brokerage unit. So with the quality service they’re receiving with respect to their hedge fund product, they’re starting to use their prime broker for some of their long-only funds as well. Prime brokers can offer more sophisticated custodial arrangements and they understand how to independently value the instruments in these funds.

HB, JPMorgan: There is an economic barrier to doing that though. When you buy that prime brokerage product, it’s great on the derivatives but it is more expensive. You are paying for the credit in a way that you don’t have to with a custodian.

RG, UBS: But if the prime brokerage relationship is such that it enables your underlying clients to receive a cheaper service, simply because there are fewer trades as a result of the netting impact, then there’s clear benefit for your underlying clients. Efficiency, cost control and better risk management – that’s hard to beat!

HS, RBC: And the intellectual advantage at a prime broker is often higher than with a custodian. Recognizing that FX is the lowest margin/highest volume market in the world, it’s increasingly difficult to differentiate by product or by price. So at RBC, we’re putting a big emphasis now on human resource innovation. We want to work more closely with clients, identify their needs and differentiate ourselves on client service and on intellectual advantage.

Best execution

Euromoney In a lot of investment arenas these days there’s an emphasis on benchmarking and best execution – doing a good job and measuring how you stand. What’s to be said about these issues with regard to foreign exchange? How do you know you’re getting best execution? How do you measure where you are?

HS, RBC: The custodial clients are becoming more aware that there is slippage and they require their custodians to demonstrate best FX execution. So benchmarking is becoming an increasingly important factor for them.

Euromoney And how well equipped are people to deal with the EU’s Markets in Financial Instruments Directive (Mifid) this year? Do you know best execution when you see it?

AB, HSBC: The FX spot specifically is exempt and I think there’s a question over whether FX swaps and NDFs are also going to be exempt.

RG, UBS: I think it depends whether you’re an eligible counterparty or a professional as to whether you’re going to be exempt from best execution or not. If a counterparty confirms that it should be classified as eligible, for instance, then the best execution obligation does not exist.

JS, Deutsche Bank: One technique is obviously using the fixings, which is important for a lot of custodians that we deal with. But it’s a subjective thing.

HB, JPMorgan: Some of the technology now does let you take a snapshot of where the screens were at a specific time.

MS, Pareto: You can look back and see where you’ve traded but you can’t get best execution at the time, because by the time you’d looked at 100 quotes how do you trade? It’s just impossible to demonstrate best execution in the currency markets because there are too many participants at any one time and the prices are constantly changing.

RG, UBS: An added complication with Mifid is that when they mention best execution, it’s not just best price, it’s the best result. This means taking into account not just the price and cost but many other factors, including the entire service offering.

GK, Bank of America: We have seen more interest in Twap (time weighted average price) execution partly as a response to this as well as to minimize market footprint at a point in time.

JS, Deutsche Bank: We expect that clients that manage particular assets might wish to outsource much of the note execution portions of their business to banks.

HDH, Overlay: The use of an FX prime broker helps best execution, because you can have access to the liquidity of many providers with just one line.

Euromoney And what about the sort of benchmarking, measurement and indexing you are doing?

JS, Deutsche Bank: There are two benchmarks that DB proposed that FX managers should be compared with. The first is a collection of transparent strategies (carry model, moving average, ppp etc). This could be considered the beta of the FX markets. Benchmark number two is DB’s investable FXSelect universe of FX-only managers. So there are two ways an investor can see if their manager has performed.

HDH, Overlay: It’s easy to compare managers with the FX Select benchmark. It really depends on how you have designed your trend-following and strategy. I think you probably have three moving averages, but some people have different trend-following strategies. With the carry trade, it really depends on whether the manager is able to beat it or not. In the absolute performance world, FX Select is a good benchmark.

JS, Deutsche Bank: Our goal is to build commonality among all FX players. I think the benchmark benefits the FX market, as investors can get more comfortable with FX as an asset class. That’s DB’s goal.

HS, RBC: But that’s no different from building the S&P 500 index. When you’re making a decision in overlay, either you’re going to do nothing and just invest in the index or you’re going to take select strategies to pursue alpha.

AB, HSBC: The difficulty is comparing apples with oranges, because different investors measure their performance in different ways. Everyone can say the return is 10%. What does that mean compared with someone who’s made 11% but against differing benchmarks? Is that a direct comparison? What’s the starting point? Having something that gives you equality across different client tiers seems very difficult to achieve.

AE, Millennium: To be fair, there are various measurements that allow reasonable equality when comparing one manager against another. The great advantage foreign exchange managers have is that it is very easy to either tailor-make mandates to suit each and every one of their clients’ needs, which they can fund accordingly, or just offer a generic currency fund that is fully funded with a single aim to extract alpha out of the currency market.

AB, HSBC: There are people we all interact with who want to benchmark individual bank performances against their closest peer group. That’s a great idea in theory, but what is retail P&L at one bank may be spot P&L at another. How you categorize business at one bank may be different at another. So you get very different revenue per head and that’s misleading. You’re building a business on a set of assumptions that are probably inaccurate to start with.

Risk management in FX

Euromoney We talked earlier about risks in the FX market, but what about risk management?

AE, Millennium: Risk management is absolutely crucial and proving that it is effective over a long period of time is crucial in obtaining new mandates and keeping existing ones. It is also crucial that you are able to explain clearly how you control your risk. Our tools have been very good and consistent. We look at value at risk (VaR), but only in combination with something we call Rats, which stands for risk-adjusted trade size. In other words, we monitor the risk of our total portfolio at the same time as allocating each individual trade a risk budget, so that at any time we can calculate our total basis points at risk at portfolio level. The two combined have proved very powerful.

JS, Deutsche Bank: Risk management is an important value-added for our clients. Every year a big portion of our technology and people spend is on making our risk management better than it was the previous year. Banks in the foreign exchange business must continually try to improve.

HS, RBC: We find increasingly that new clients setting up look to the banks to advise them or to assist them with their risk management capabilities, because it is a primary risk.

GK, Bank of America: Clients expect the banks to take risk off their hands. As the business has become more electronic, firstly risk is transferred quickly and you also have to know it’s there. We all need tools in place to be able to handle that risk, particularly when a market’s moving. We spend a lot of resources on making sure that if something’s coming down the electronic pipe from a client we know what to do with it.

HS, RBC: And certainly in prime brokerage the clients rely on your risk metrics to keep their margin intact.

GK, Bank of America: We’ve had to build the prime brokerage technology internally over the last few years because clients don’t want us to outsource it.

Missing risks

HB, JPMorgan: As an industry I still think we’re missing some risks. The RiskMetrics-type approach puts so much weight on recent behaviour in the markets and forgets about what the Indonesian rupiah did back in the Asian crisis. That means the risk can be underestimated. We don’t yet have good tools for measuring liquidity risk and how to get out of the trades. You can have line-by-line security on each of your trades, but if the door is not there to get out of it then you can’t really measure those risks.

HS, RBC: Within a bank and within a treasury operation, liquidity risk is considered one of the primary risks, along with credit risk, operational risk and market risk. I agree with you that there is a disproportionate emphasis in the fund sector on managing market risk.

AE, Millennium: I totally agree and what is a concern is how short term some of these VAR models are. Using the example of Turkey this May and June, memories of the Asian crisis were long forgotten and extremely large amounts of risk were allocated, as recent levels of volatility made the positions seem viable. However, potential illiquidity had not been taken into account and as a result the US dollar against the Turkish lira moved 34% in just over a month as the market tried to exit risk through a very small window of liquidity.

RG, UBS: Pro-active risk management has become increasingly important for our clients. As a result we have dedicated resources to building a system that offers them a very similar framework to the one used by our traders. Prime brokers’ clients find this particularly useful as they can manage their whole portfolio on our web site.

Electronic trading

Euromoney What about electronic trading in currency? Will that be the only medium in which we trade going forward?

HS, RBC: In effect the intra-dealer market is almost entirely electronic. The debate is really about the percentage of customer distribution that will go electronic compared with telephone distribution. The equity model and its development are relevant but the FX market isn’t an exchange. It’s an over-the-counter marketplace, so in my opinion there will always be some information asymmetries and execution asymmetries. But I think for the inter-dealer community that has pretty much run its course.

JS, Deutsche Bank: A dominant FX market exchange is a threat to banks. But that’s been what’s inspired the FX banks to relentlessly improve their product. Tighter pricing, better technology, more flexible credit have all come from this competition. Clients are big winners.

Euromoney Why are you worried about an exchange? If there were a demand for one it would have appeared by now!

HS, RBC: We’ll see how FX Marketspace develops, the CME/Reuters exchange. But in my opinion if the market felt that was the right way to go, like with equities, it would go there. Something is stopping that from happening.

GK, Bank of America: If we get to the point where post-trade processing is homogenous, then we can move towards an exchange model, but this is some time away yet. EFX is going to the back room. Error counts can be reduced through post-trade processing. To keep up net revenues you need to be a low-cost provider and have a really tight post-trade processing operation. Then it doesn’t matter if someone picks up the phone or they deal electronically. In addition you have to be able to handle the transaction volume. Headcount is expensive, particularly as the spreads are narrowing, and we need to provide electronic solutions.

RG, UBS: The transition to E-trading will continue as efficiency and productivity gains become increasingly important to all client segments. While e-commerce itself is not new, API feeds are still in their infancy but growing rapidly. This is an area of increased focus for us at UBS. At the same time it is important to recognize that clients demand the appropriate balance of “high touch” and “low touch” interaction with our client service teams and these needs do vary from one segment to another.

AB, HSBC: Lowering the unit cost of production is clearly important, but perhaps only at certain types of institutions that need direct access. I don’t think the exchange works for a lot of other clients.

Euromoney Well, we could go on for longer, but that’s all we have time for. Thank you.