As Valentines Day draws near, it’s time once again for me to send those roses, to write that drippy love-sick poem, and to make reservations for a nice romantic candlelit dinner. Readers of my old market-geeks blog or my posts on Forex Factory over the years may already know what I’m talking about. Since 2006, I’ve been head-over-heels in love with..sigh…the carry trade.
But it’s been a stormy relationship. After a multi-year dalliance beginning in 2010, when I used the USD as my funding currency, the pain finally became too great, so we had to go our separate ways.
I’m happy to say that we’ve gotten back together recently though, this time with the Yen as our funding currency. At the moment, we’re happily walking hand in hand down the TRY/JPY path, and things are going pretty well.
But the carry trade is a high-maintenance companion with quite a few issues. So enough of this sentimental claptrap; let’s get down to the nitty-gritty of the most recent updates to the strategy.
As I mentioned in my last post, I try to spice up the carry trade relationship by trading around the core position, attempting to take advantage of volatility by extracting profits above and beyond the interest payments. In this way, I hope to lower the average unit cost of my core position and juice the returns.
The issue with this, of course, is the volatility itself combined with the high leverage, and the fact that there’s no ultimate bottom in a relative value market like forex. So the contrarian strategy of “buying from the scared and selling to the greedy” is fine while selling on the way up (I’m assuming a long position in the carry pair throughout this post). But buying more and more on the way down is like skating on thin ice while filling your pockets with rocks. It generally doesn’t end well.
So there’s this sort of tug-of-war between two competing trading goals that I’ll call the leverage goal on the one hand, and the P&L goal on the other. As price moves in my favor (up in the case of our assumed long position), the leverage side of the equation is saying, “Hey, leverage is way too low here! That’s inefficient. Buy more!!” But the P&L side is countering with, “No, sell to the greedy! Take some profits I tell ya!” When price drops, the leverage side wants to sell in order to reduce the risk of a margin call, while the P&L side wants to buy from the scared at bargain prices.
It’s similar to the game of Go, where in order to win, you have to push aggressively, but not so hard that you get overextended and risk ruin on all fronts. So in my latest solution to the quandary, I acknowledge both sides and pander to each in different ways, sort of like most politicians I suppose. Here’s how it works, along with all the yummy mathematical details!
First of all, I developed a quick and dirty goal to use as a guideline for my desired leverage level. Oanda allows only 20:1 on exotic crosses like TRY/JPY, which has about a 6% annual yield. If I leverage that at 12:1, that’s 6% a month, which (by a quick use of the rule of 72) should compound to about 100% per year. That’s easy to remember, so I set my central leverage target to 12.
Ok fine, but now let’s pander a bit to the P&L crowd by adjusting this over a long term timeframe. I set upper and lower bounds 25% away from the central value, so I’m willing to let my target leverage vary from 9% to 15% as a function of where price is in comparison to it’s 52-week range (easily seen with the 52-bar %R indicator on the W1 chart). At the midpoint, the target is 12. As price moves into overbought territory, we want to sell to the greedy, so we dial back the target leverage until it hits 9 at the 52-week high. Similarly, as price drops, we’re willing to buy from the scared up to a leverage of 15:1.
For example, if %R is -75, I want my target leverage to be 75% of the distance from 9 to 15, which is:
9 + 6(.75) = 9 + 4.5 = 13.5
Ok, so that’s my target leverage, which, oddly enough, is actually set using the P&L philosophy. But given that target, it’s now the job of the “Leverage Department” to recommend trades that will bring my actual leverage in line with that target. So in the short term, as prices fluctuate along with my position size, the Leverage Department is prompting me to buy more when price goes up and to sell when they go down. That’s because when price is going my way, my account value grows, lowering my actual leverage and vice-versa.
So how does our Leverage Department come up with a trade recommendation? First, if actual leverage is higher than the target leverage, the advice will be to sell and vice-versa. The amount of the trade is a percentage of the existing position given by this formula:
1 – (lower lev)/(higher lev) where:
the numerator is the lower of either the actual leverage or the target leverage, and the denominator is the higher of the two values.
For example, if I currently have 2000 units, the target leverage is 13.5, and the actual leverage is 8.1, then the Leverage Department would flash a buy recommendation of:
2000(1 – 8.1/13.5) =
2000(1 – 0.6) = 2000(0.4) = 800 units.
So our Leverage Department pretty much has a trend following philosophy. This is in stark contrast with the contrarian philosophy of the P&L Department across the hall. Remember, these are the folks who want to buy from the scared when price is falling and sell to the greedy when it’s rising. We’ve catered to this idea slightly with our long term target leverage adjustment as I mentioned above. But how do we satisfy these wild and crazy guys in the short term?
Well, the P&L Department comes up with it’s trade advice in pretty much the same way as the Leverage Department; they just use different inputs. In this case, we use the cash balance and the net asset value of the account, the latter of which includes our open profit or loss on the position. If the NAV is higher than the CB we have a profit, and the P&L side of the hall is screaming to sell. When the NAV is lower than the CB we’re at a loss and their advice is to buy at bargain prices. Just as with the Leverage Department, the P&L Department multiplies the current position by one minus the lower of CB or NAV over the higher of the two values.
So continuing our example above where we seemed to be profitable (thus driving our actual leverage below our target), let’s say our cash balance is $75 and our NAV is $100. Using our position of 2000 units from above again, the P&L side of the hall would recommend a sale of:
2000(1 – 75/100) =
2000(1 – 0.75) = 2000(0.25) = 500 units.
So what do we have? We have the Leverage Department advising a buy of 800 units and the P&L Department across the hall recommending a sale of 500 units. In general, these two recommendations will be in opposite directions. What’s a trader to do?
What this Capitalist Trader does is to place both trades in the form of limit orders, with the sell order above the current price and the buy order below (which makes the P&L Department happy of course). I generally use about twice the H3 8-bar ATR as a guide for how far away from the price to place the orders. For instance if the ATR is 22, I might place the orders at round number levels about 50 pips above and below the current price.
When one of the orders is hit, I use the new position, leverage, NAV, and cash balance to calculate the next set of orders, and so it goes. In some cases, when the position is at breakeven or when the leverage is right on target, one of the order recommendations will be for zero units. In this case, I just throw a horizontal line onto the chart in place of an actual order, and recalculate if and when that line is pierced.
Finally, note that in cases where the leverage is at the upper boundary, the Leverage Department always has the final say. It doesn’t matter if the P&L Department is screaming for a buy at the top of their lungs; the Leverage Department is in charge of risk control, and they just basically tell the folks across the hall to stick a sock in their pie hole.
So those are the details of my latest and greatest carry trade strategy upgrade. If you’re still awake after all that, then I salute you! Next time, I’ll continue my series of rants on economic and fundamental analysis.
So until then…keep pipping up!