The Foreign Exchange Matrix. Barbara Rockefeller

market currencies, to advanced country fixed income securities, to mutual fund valuations, to the new-technology energy names on the NASDAQ, to the Shanghai and Nikkei indices, to the dollar/yen… and so on. Transactions are conducted at such speed and the world is so interconnected today that we can’t explain everything, and often, ex-post explanations are unconvincing. Explanations do exist, but we may never know them. This is a wonderful and frightening thing. ^[3]

      The actors

      Let’s take the actors, or agents as academics might call them. How you interpret FX market developments depends on what sort of trader you are, so it is important that we consider how various actors think.

      Let’s say, for example, a key technical level is being hit – but it’s Friday at 3 pm. The FX market’s reaction to the key level is entirely different from when a key level is hit at 8:30 am on a Monday morning. More nuanced than this, a key level being hit at 3 pm in London on any day is different from the level being hit at 3 pm in New York. In the London case, traders still have several hours to see what the New York market will make of the development. If New York responds predictably to the key level, London traders have a new profit opportunity, although it means working late.

      The New York traders do have an overlapping time zone – New Zealand opens around 3 pm New York time, Australia opens around 5 pm and Tokyo opens around 7 pm. But these are much smaller markets than London and New York, and New York traders have different work habits. They arrive at work early to partake of the London and European action, rather than staying late to join the Asian markets. The importance of the key level is far lower in New York at 3 pm on any day than at 8:30 am on any day.

      Now consider which group of actors cares the most about a key level at any time of day – technical traders. After all, with algorithm-trading and pre-set electronic entries and exits, the trader doesn’t have to be physically at the trading terminal to respond to a technical event like a key level. If the key level is a match of a past benchmark high or low, or a round number, or widely publicised (like a Fibonacci level), then non-technical traders know about it, too. If the technical group responds as expected, the non-technical groups feel compelled to react, whether defensively or opportunistically.

      In other words, the sets of players are interactive. Because we have so much news and chatter in the electronic age, we are all getting a great deal of information about the other players, and we are getting it 24 hours a day. A good case is when the FX newswires have reported where a big buy-side client has an option strike, often a round number. The existence of the option and its strike price is a mechanical aspect of the market and not strictly speaking technical, but the strike level is almost always set by technical considerations. It may be just past a moving average, a historical benchmark level, or some other calculation.

      The financial institution that wrote the option will have hedged at least part of its position (although not all, or it wouldn’t make a profit on the transaction), but still probably prefers not to have to pay up, while other market players know that a failed test of the level opens the door for a big move in the opposite direction of the strike. This is a case in which the widespread knowledge of the option strike is the top factor and it would take a big event in the macroeconomic world to overwhelm it.

      Positions

      The option strike price case illustrates the most important aspect of the player groups – their positions. When the vast majority of the market holds a particular belief, their positions reflect those beliefs. If a set of factors lines up against the pound sterling, for example, a short sterling position is rewarded with every news release reinforcing the negative tone. Good news is dismissed and disregarded. Every trader is gunning for sterling. It doesn’t matter that by any objective measure on a fundamentals matrix, five factors are negative and two factors are positive.

      But there is always a tipping point at which the market gets oversold sterling – something experienced traders can smell and that technical analysts can measure – whereupon a key level gets hit and a cascade of short-covering ensues. The tipping point may appear with or without a news announcement pertaining to a factor. Sometimes the factor is directly relevant, whereupon the press says “the release of the XYZ data caused the pound to firm.” In other instances, and we say it’s the majority of the time, the release of the XYZ data is just an excuse for traders to get out of Dodge. If you can’t sell it (anymore), buy it.

      Hence the seeming perversity of the FX market. The objective measure would still have four negative factors versus the new count of three positive ones; it would still be net negative. Logically, we say the pound should remain in a falling trend and the factor weights do not line up to support a rise, but the market is not responding only to the factors, but to the quickly disappearing probability of making a profit from the existing short position.

      This is the sense in which the market is not always right, as the old law has it. The market is always right in the sense that every trader has to accept the prices offered and a single trader cannot change the market’s mind about what is the correct price. The market can be dead wrong about factor analysis but it doesn’t matter. You have to trade the prices in front of you, not what you think prices should be.

      Making sense of the information

      Now enter the poor retail trader, who often has no idea that FX trading is not easy, as the TV and website ads proclaim. Retail FX trading is rapidly expanding, hot on the heels of professional asset managers diversifying their base. How do these new traders make sense of the market? FX commentary is keeping pace, but much of it is of dubious quality.

      One book author asserts the Canadian dollar is 40% correlated with the price of oil – without naming the time period over which the statistic was calculated, rendering it literally useless. In practice, you can manipulate the CAD to be correlated with the price of oil to any number that suits your argument by changing the correlation timeframe. In the funniest case, a publisher was so eager to get a book in print and capitalise on the demand for market commentary that it failed to notice a whole chapter and multiple other references devoted to “rouge” traders (when “rogue” was the intent).

      The lesson here is to be careful what conclusions you draw from what you read. And also, to come back to our matrix, that single-factor explanations of FX market behaviour are always wrong. To explain a move, you need:

      1 fundamental factors;

      2 technical measurement of the sentiment derived from those factors;

      3 technical dynamics, and

      4 existing positions of the key players.

      Once you have these four things and decide which weight each should have, you can begin filling out your matrix and deciding how best to position yourself for a move. In the chapters ahead, we will go deeper into the specific factors that drive currencies.

      It’s probably fair to say that very few, if any, highly sophisticated FX market players have programmed a matrix to include all the factors or succeeded in figuring out how a change in one fundamental feeds the technicals and then how the resulting positions create a feedback effect on other technicals and sometimes the fundamentals themselves. A computer program that is capable of taking new data and have it light up the relevant variables in some kind of logical order and have it all result in a price deduction would be a splendid machine, indeed. If some advanced hedge fund or sovereign fund has created a working matrix, they are keeping it a deep, dark secret.

      Until such a thing can be devised, we mere mortals must struggle with the sets of variables the old-fashioned way, using our brains. It should come as no surprise that some of the best traders are fairly ignorant of economics and historical context, but are cracker-jack at game theory. In the chapters that follow, we do not actually mention the matrix very often. Instead our goal is to describe the FX market in ways that will be useful to the reader in creating his own matrix in his imagination, if not in programming code.

      Endnotes

      1