USD/TRY/CAD Hedged Carry Trades: Keeping Some Dry Powder Handy

Last month I reported on the status of my USD/TRY/CAD hedge carry trade method, which included mention of the young account surviving the Dubai debt debacle. As I said last month, at its worst the account hit about -54% in unrealized losses, but survived and rebounded; as I type this, the account is sitting at -9% unrealized loss, it has been better than -7% today (during the London/New York session), and the income it is generating monthly is presently up 22% from the December withdrawal amount. Not bad at all for having taken such a beating.

Part of the dramatic recovery of the account is due to a rebound in risk appetite, which benefits TRY, but possibly the biggest reason at the present time is my adoption of a new approach to running this type of account: keeping cash on hand deliberately, the "dry powder" mentioned in the title of this post.

One of the things I've realized since launching this account is that my previous approach - get all of the cash in-play according to the ratio that I use right away - may have been screwing up the averaging of the USD/TRY and USD/CAD positions. That would have the effect of amplifying divergence events, the times when the pairs are not moving in sync with each other, which I am trying to avoid.

What I've been doing then is taking the free cash in the account at the beginning of each month and dividing that sum by 30. The result is the amount I buy in with each morning that spreads are at their narrowest (in other words, Monday through Friday only), and I don't worry about the charting. The reason I'm using 30 as a divisor is that it is intended to approximate a calendar month, which is how long I would want it to take to get new cash plugged in and earning. Because I do not buy new positions on weekends, it would take longer than 30 days to get a given amount of cash in-play, but I want that to happen since free cash helps to protect the account against sharp price movements (and it all gets "recycled" into the working total of the following month anyway). Depending on how much free cash is on hand at the beginning of a particular month, I may increase that divisor though, perhaps up to 365 for huge amounts. In general I want subsequent buys to average more smoothly with the aggregate of positions "behind" each one, to avoid having a few relatively big positions distorting the average of lots of smaller positions.

My second hedge carry account was started from day one using this approach, and it is doing very well. As of right now it has an unrealized gain of 2.11% and is earning at 57.22% annually. (This is actually a bit lower than what it will ultimately earn - buying in daily distorts the interest rate downward somewhat because of the reduced income a new position generates on day 1. You have to wait for a full 24 hour cycle to pass before the "true" income will be generated, so in a hedge carry that is constantly being added on to like this you won't really know the rate of the account until Saturday rolls around.)

Another benefit of this method is that during the time I am growing any such account, I can "capitalize" the portion of the interest that I do not draw out each month. I simply add it to the cash that is not in-play and proceed as I've described above. I'm comfortable doing this because the left over interest is not required to pad and protect the account as much since the cash being used to build up the account, plus the daily accrued interest, serves this purpose from month to month as the account grows. When each of my accounts reach my target (in terms of monthly income that I extract), then I will revert back to allowing the left over interest to accumulate and lower the leverage.