DynexCorp - from systems to market pulse: a life long performance in currencies

e-Forex talks with Peter Panholzer of DynexCorp, one of the oldest Currency Management firms in the world.

Peter how long have you been in the Currency Management business and what attracted you to the industry? 

I made my first currency trade in October 1973. After Bretton Woods, spot settlement rules were still in their infancy. At that time you had a choice of forward swaps, such as 30 days, 60 days, etc which could only be realised at expiry. At Ross Perot’s Dupont Glore Forgan office in Toronto arbitraging between 30-60 and 60-90 day swap spreads was a popular strategy. After 30 days your swaps mutated to spot-30 and 30-60. If an expiry fell on a weekend and the invariably mean-reverting arbitrage happened to be in your favour, you could either earn settlement interest or roll the swap positions forward.  It was a win win – now or later – situation.

That was two years after the gold standard had been abandoned – causing the big inflation of the 1970s – and one year after the creation of currency futures contracts at the IMM in Chicago. I am still in touch with Mark J. Powers, the father of currency futures (together with Leo Melamed). Mark allocated pension fund money to us in 1990. In their due diligence, he and his partner Mike Dubin put emphasis on knowing the manager personally by asking him tricky questions over an extended lunch. I was impressed by this pleasant approach. 

I had actually abandoned my successful career as an architect to become a commodity trader and by 1978 a pure currency trader. A highly diversified trend-following system in commodities during the inflationary 1970s was virtually assured to be profitable.  When I started a currencies-only system with the five currency pairs then available at the IMM, everyone warned me that this wouldn’t be enough diversification. They were proven wrong. While I was an account executive at ContiCommodity in Toronto, the Magnum Program became the company’s star performer, between 1979 and 1983. I presented the program in the United States to 14 branch offices in 17 days. Conti had two thousand account executives world-wide, and the Magnum Program was offered on four continents. 

One program unit of $25,000 traded a total of five contracts, one for each currency (DEM, JPY, CHF, USD and CAD). We had a specialist IMM floor trader assigned (Andy Good) who worked our orders on occasion as high as 1,000 contracts.  In 1981 that was a good size. During the first three years the (widely dispersed) average annual profit was 54%. Checking this out today, the Magnum Program used ten times leverage. No wonder the system eventually suffered the contractual drawdown limit.  These were the early years of managed futures. A maximum drawdown not exceeding the average annual profit was quite acceptable for a trading system. Now we trade 0.5 to 1.5 x leverage (e.g. with no leverage, on balance). 

If you ask me what made me switch from architecture to the financial markets, it was the sad fact that architects have to fight for their fees, and fund managers don’t. That’s too bad. There are many instances where architects have turned to all kinds of other imaginable endeavours: most curiously one went into the funeral business, another one became a clown. Among these former architects are celebrities such as Paco Rabanne and FX trader Bill Lipschutz.

Peter Panholzer

Peter Panholzer

When did DynexCorp commence operations and what services does the company provide?

In 1981 Conti moved me to their branch in Lugano, Switzerland. After the sale of ContiCommodity in 1984 to Refco who had no office in Lugano, I was looking for a new home – and found it with E.F. Hutton in London, at the firm’s Mayfair office. I was hired by Bob Scheiner, an exacting Navy man, who generously offered a mews house for three months behind Cadogan Square until I had found my own place, a small town house on Paradise Walk near the Royal Hospital. I actually got mail posted to Peter Pan, Paradise Walk, London SW3!

After a brief stint at Refco London, I became independent in 1988 and formed my own firm, Dynex Corporation (DynexCorp today), with offices in Geneva. Our first corporate clients were the successful nostro accounts of two private Swiss banks. In their wake followed a slew of pension fund money from Japan and the United Kingdom. We have managed institutional funds since 1990.

Who are the key people in the firm and what are their main day to day responsibilities?

I am the chief trader concerned with portfolio and risk construction, a task I have performed for the last 32 years. Risk levels are monitored via audio alarms and also checked by other members of the team. 

Marlene Harrison looks after monitoring and distribution. She has been in the investment business in Toronto since the early 1980s and also managed accounts in 1985/86 with a successful S&P futures trading system. She is invaluable with regard to client communications, overseeing performance statistics and organising promotional activities. A superb motorist, she likes driving super-fast cars, not only where there is no speed limit but also on the winding roads of the Riviera.

Jeff Boyko, with an engineering degree from his home town of Winnipeg in Canada, has been at our European trading desk for a good number of years. 

He developed an outstanding record in Canadian dollar hedging for commercial clients. After acquiring his CFA he took charge of product design, client relations and marketing. He recently created a Canadian approved fund awaiting seeding. Jeff’s dedication to the promotion of currency markets as an asset class is exemplary. 

Paul-André Jacot, doyen of the Swiss Futures and Options Association (SFOA), has been with us as Director and Advisor to the Board since 2001. 

He is the organiser of the annual SFOA Bürgenstock Conference convening representatives from futures exchanges, regulators and self-regulatory organisations from all over the world. In 2009 Paul was inducted into the Futures Hall of Fame established to commemorate outstanding commitment to the global futures and options community. We have the same core team since many years, complemented by marketing, legal, accounting and IT.

Boyko, Peter Panholzer and Marlene 

Jeff Boyko, Peter Panholzer and Marlene Harrison

What do you consider to be the key strengths and operational advantages of DynexCorp’s currency portfolio management services?

Having been hooked on currency trading for almost 40 years, one could say I wear currencies like I am wearing my clothes, even like my own skin. 

My reaction to familiar market situations is instinctive, acquired through life-long conditioning. The new kids on the block will always come up with new-fangled ideas, but looking back, not much has changed in what really drives the currency markets. Supply and demand fundamentals do not work as easily as they do in commodities, but positioning, recursive feedback and emotions do.  

I have to thank a brilliant trader, Albert Friedberg, President of Friedberg Mercantile Group in Toronto, for introducing me to the contrarian way of looking at markets as early as 1973. Ever since, I have used sentiment and contrary opinion to guide the non-systematic part of our trading approach. Another highly successful proponent of analysing market sentiment is John Percival, with whom we have been closely associated since the mid-1990s. A former Lex Columnist of the Financial Times, JP writes the weekly Currency Bulletin (since 1984) which must be one of the oldest currency advisory letters in existence. He is also the author of the book The Way of the Dollar (out-of-print, but available as e-book). 

This is quintessential reading! For me this book became the bible of currency trading.

Paul-André Jacot

Paul-André Jacot

Our operational advantage is our longevity, survival power and persistence. Starting with the Magnum Program in 1979 we have a continuous track record of 32 years, 90% of which are certified or audited. There is no failure, pause or interruption. Surely this is unique? We cannot think of any other currency manager still in business who can match this record, other than Albert Friedberg in Toronto. Reliability and continuity is often the most sought after requirement by institutional clients. 

Our other operational advantage is our compact size. Small is beautiful, said Leopold Kohr, the Austrian economist, who wrote in 1957 “There seems to be only one cause behind all forms of social misery: bigness“, in his visionary book “The breakdown of Nations”.  Because of the compactness of the underlying instruments, currency hedge fund managers succeed in operating efficiently with a limited number of staff. Computerisation, electronic trading and Straight Through Processing have made this even more feasible. 

How would you describe your investment philosophy and what are the main strategies which DynexCorp has traditionally employed?

Traditionally we have been trading-system oriented.  But trading-systems seem to have a half life and start to decay after a handful of years. Caxton Corporation’s Bruce Kovner pointed this out when I had the chance to converse with him for a full evening at his Manhattan apartment (his family had gone skiing). We shared the opinion that systems are tweaked every so often, mostly of course after a draw-down. I postulated in the late 1980s that baby markets lend themselves better to systematic trading than mature markets. In effect, the Magnum Program which ran from 1979 to 1984 had a good chance since currencies in those years were indeed baby markets. The Magnum trading rules which worked fine for currencies would not have worked with the stock market (e.g. the S&P500 Index), the most mature market of all. 

Today’s currency markets have become mature by their sheer size and liquidity. For that reason we have abandoned purely systematic trading and are putting more weight on market sentiment, positioning research and contrarian opinion, particularly in our Tetra FX Macro strategy which has done extremely well since 2009 and puts us currently in the top echelon of the currency alpha manager universe, as surveyed over the last three years via Deutsche Bank’s FX Select platform. 

Another project we have pursued since 2007 is designing and managing Dynex Dynamic Currency Alpha, a currency manager talent basket with dynamic allocation and daily liquidity.  This project was made possible through Deutsche Bank, who pioneered the FX Select Platform in 2004. Today $5 billion are managed through this platform.

DynexCorp - from systems to market pulse: a life long performance in currencies

What style of trading do you undertake to execute your strategies? Do you have any particular preference for specific trading time horizons, such as medium to long term?

Our trading campaigns reach for intermediate targets within a week, but our trading horizon is also a function of volatility. Therefore risk measures may vary with the currency pair in question. We have extended our variable scale of risk windows to 8-hourly intervals, using proprietary statistical models surveying systematically 35 currency pairs, fully aware of the auto-correlation that this may entail. It is a common misconception that the behaviour of one currency pair might be useful in forecasting the course of another. 

How do you go about back-testing your proprietary currency trading systems and signals to confirm that your strategies will perform and remain relevant to the performance criteria of the firm?

I had my first computer in 1977; the biggest storage medium was a single 8 inch floppy drive! By 1983, while commuting between Europe and Canada, 

I owned four portable Otrona Attache, each of them costing an arm and a leg by today’s standards, the first truly portable computer with a small CRT screen. NASA’s Jet Propulsion Laboratories loved these machines. Computer lore has it that an Otrona Attache even flew on the space shuttle and also appeared in a Bond movie.  

We have done back testing galore. Over three decades. In the early years I did my own programming. It is obviously of great advantage to have programming ideas and trading experience housed in the same brain. The trick with back testing is to have a perfect underlying data set (an impossibility).  Instead of using early pre-Microsoft spreadsheets such as SuperCalc, I coded each application including tables in C-Basic Compiler by Digital Research. This was tedious but the output was faster than Excel today. Since the late 1980s we outsourced a lot to SnapDragon Systems in Oxford, England, whose principal, Adam Hartley, holds a PhD in Theoretical Atomic Physics at Oxford’s Balliol College and keeps reminding us that “every strategy has its Achilles heel”.

“It is a common misconception that the behaviour of one currency pair might be useful in forecasting the course of another.”

“It is a common misconception that the behaviour of one currency pair might be useful in forecasting the course of another.”

What Risk Management frameworks have you developed and how do you apply them with respect to different strategies?

Our STP-linked in-house software refreshes risk limits in time slices after each transaction, in or out. Our risk tolerance is low. This means that we use no leverage or operate at a marginal distance around the funded capital amount exposure. We also tend to step aside completely after our risk tolerance has been challenged in order to continue with a completely clean slate.  Any account operated via a TRS is protected by a stop-loss amount monitored by Deutsche Bank, the counter-party to the TRS. A stop-loss protected Total Return Swap is more cost effective than Capital Guaranteed structures. 

The trusty Otrona Attache

The trusty Otrona Attache

To what extent have you developed research agendas and analytical programmes to enhance your existing strategies and find new ones?

We use statistical research as well as subjective sentiment-driven techniques. In 1988 we were first in performing Rescaled Range Analysis and plotting the Hurst exponent of five IMM currencies, relying primarily on Jens Feder’s just published textbook Fractals. An ominous quirk of fate would have it that 23 years later Hans Feder, the author’s son, is leading the FX Select team at Deutsche Bank. We demonstrated the declining memory effect in currencies since 1973 – and thus the decreasing efficiency of trading systems, leading to our concept of baby-markets being more useful for trading systems than mature markets (for example, emerging markets currencies are today’s baby markets). During that time we were also first to point to the odd fact that currencies do not scale logarithmically like commodities or stocks. Think about that for a while: it makes risk-scaling easier.

Today we concentrate on obtaining intelligence about market flows, positioning changes, contrary opinion reports and sentiment analyses. We pay particular attention to information offered gratuitously. News media are often an excellent source for contrarian positioning. Their gratuitous trading recommendations remain unpunished if they fail, since they escape the negative feedback from speculators who have followed them. When a particular market occupies a large part of public perception and becomes one-sided, alarm bells start to ring in tune with Bob Dylan’s lyrics that you don’t have to be a weatherman to know which way the wind’s blowin’.

DynexCorp - from systems to market pulse: a life long performance in currencies

What sort of trading platforms does DynexCorp find most appropriate and what factors influenced that choice?

We developed a fruitful relationship with Deutsche Bank, worldwide leader in FX liquidity, as early as 2002. Our preferred dealing platform is Deutsche Bank’s Autobahn FX. Not only has it become a benchmark for liquidity and pricing, it also has the most down-to-earth interface design. We consider the learning curve for most electronic retail platforms too onerous. Since the start of eFX, bid-ask spreads have collapsed, for the benefit our clients. We favour single click-and-deal transactions or through our API, not revealing our intentions. The advantages are obvious: tighter spreads, faster fills, tighter risks. Our trading room used to be a lot louder during the 1980s.

How did you go about building your trading desk IT infrastructure and was the trading software and connectivity technology provided by third party vendors?

With regards to operational computing we always have an IT consultant at hand. Everyone in the company works with the same computer or laptop model. We have commissioned SnapDragon Systems to design custom applications for charting, APIs and robotic trading. The problem with robotic trading is its extreme reliance on connectivity and monitoring. Once you have to look after both, the robot is not so much a robot any more, or is it? As in the kitchen, a robot is just a friendly beast that has the knack of breaking down when you most need it.

Trading technology has come a long way during the last 20 years. How much reliance does DynexCorp place on the latest automated trading toolsets and execution algorithms? 

Technology can only improve our lives. An e-platform is the difference between having a car and having to travel by foot. I do not believe we could trade the volume we do, without eFX platforms. Trading electronically has greatly reduced the chance of human errors. Online trading is linked to Straight Through Processing. That’s an immediate relief for the back office, with a 70% reduction in time. eFX technology allows systematic trading to be executed more accurately via an API and also generates a data stream that is an important research material for us. We store it for future research.

As you know we also publish Currency Investor magazine and so are great champions of currencies as a tactical asset class. How difficult have you found it to persuade investors that currencies can be leveraged not just for risk mitigation by institutional asset managers but are also as an excellent means for achieving more balanced and diversified personal investment portfolios?

In the world of alternative investments, pure currency alpha should have moved much sooner into the limelight as an asset class. It compares favourably in terms of a) return, b) risk and c) correlation statistics. The currency market is after all bigger and more liquid than any other. It has been difficult for a reason, to persuade institutional investors to diversify their portfolios by adding uncorrelated alpha returns from currencies. 

Pension and endowment funds giving pure currency alpha a try may have typically made the mistake of turning to currency overlay managers they are familiar with and basing their decisions in large part on the size of the manager.  Unfortunately asset size is not a good criterion for handing out mandates for pure currency alpha. 

Overlay managers as well as banks have delivered disappointing currency alpha results, thus creating misleading market lore among institutions about unhappy currency alpha experiences.  Without applying the selection criteria described below, naively allocated currency alpha would actually have been flat over the last five years and might as well be equivocated with cash, e.g.  zero return. 

Yet there is a really very good reason why institutional investors should move into currencies as a non-correlated asset class – if they pick the top performers, an elitist group with a long historical background and a remarkable knack for persistence, rotating inside the top 1-, 2- and 3-year rankings over extended time spans. They do not include – nor does their asset size compare with – overlay managers or banks. Trading systems, unless applied to emerging (“baby”) markets should also be avoided, for the reasons cited earlier. 

Over the last five years, a portfolio composed of these inherently superior managers returned much higher rewards than the currency manager universe. Their selection does not rely on the benefit of hindsight: 

it is based purely on their pedigree and the intelligence gathered about their investment teams. DynexCorp is currently among the top five performers for the last 12, 24 and 36 months. 

What new strategies and products has DynexCorp been exploring as part of your continuing efforts to widen the range of investment solutions and tailored services you make available to clients?

We consider our investable Dynamic Currency Alpha Index a very promising idea. Cost-effective currency alpha may be obtained via dynamically managed currency manager talent baskets through the FXSelect Platform (pioneered by Deutsche Bank in 2004) which also allows notional funding. Deutsche Bank is the Prime Broker and the product is available via a Total Return Swap. FXSelect manager performance data, accessible on demand and reported at the end of each day, are hard performance data, e.g. not performance communicated by individual managers “on a reported basis”, which is often curiously different from their “real” data on FXSelect.  

Looking ahead, what are the main challenges facing DynexCorp as you seek new ways for capturing and exploiting investment opportunities with currencies?

My guess is that a tremendous challenge lies ahead: just look back at the huge inflationary effect of abandoning the Gold Standard and Bretton Woods in 1971 – with the main purpose of financing the Welfare State. During the following ten years the US dollar fell from 4.3 to 1.6 Swiss francs and gold rose from 42 to 460$/oz. 

We might eventually face violent consequences of the recent seemingly bottomless bailout barrel of the EU and the threat of sovereign default in the United States. Volatility will invariably increase in the aftermath of such critical events. These are the seeds of hyper-inflation further down the road, if you care to follow the writings of Steve Hanke, Professor of Applied Economics at The Johns Hopkins University in Baltimore.  During such upheavals exchange rate volatility typically increases. Volatility is good as long as it is matched by liquidity. If that happens then eFX technology has put market liquidity on a winning road for us all.

DynexCorp - from systems to market pulse: a life long performance in currencies