Carry Trades, Simply or Not

Greetings folks,

Hope you are well (and safe!). Curious to know if anyone here does Carry Trades and if so; how you setup for them, your successes and/or failures.

Don’t currently trade for Wednesday rollovers but want to explore the possibility. Not sure if it is as straightforward as it appears.

If the we assume random markets (ignoring trade costs) then any trade entered over the rollover period should have net Alpha (edge) of zero in the long term, albeit with periodic trading balance increases and losses for the incidents of winning and losing weekly trades respectively.

So,…the carry interest should accumulate over a period of time separate from the trading PnL fluctuations each week. (fluctuations to zero PnL).

But the spreads increase quite drastically during the rollover periods, so to what extent does that spread increase affect the long term net PnL of the weekly trades, if at all. i.e. can we continue to assume random market behavior and that the trading PnL will tend toward zero in the long run.

1 Like

Hi - I had to remind myself what a Carry Trade was, so I googled it.
A carry trade is when you buy a high-interest currency against a low-interest currency. … For example , if the pound (GBP) has a 5% interest rate and the US dollar (USD) has a 2% interest rate, and you buy or go long on the GBP/USD, you are making a carry trade .

A long time ago (more than 25 years ago) I used to do some carry trades before the invention of the Euro, and recall with delight some of the interest rate differentials. At that time, the intended holding period was “weeks to months”, and I often had difficulty in my mind as to whether I considered this as “trading” or “investment”.

I may be mistaken, but interest rate changes in 2021 seem to be far less frequent, and rates far less volatile, than 25 years ago. So are you relating this activity to Forex trading, or in which market do you conduct carry trades? Sorry for stupid question.

I was thinking about this too.

Carry trades don’t often work very well at a retail level. You need a fairly large position to make it worthwhile, plus you need to make sure price does not go against you long-term. This means you are tying up a good chunk of capital while still hoping to stay in the black, otherwise any profits you make on the interest rates can be wiped out, along with your account.

Greetings, thank you for the reply. I just checked the Carry interest rates after a long while and it seems that they are indeed much lower than they used to be.

USDMX & USDZAR used to be the most likely potentials. Currently USDMXN offers £1.07 per lot on shorts with my broker, so a Wednesday rollover would be £3.21 per lot.

whereas it used to be 7 - 14 multiples higher… With the current rates I agree that it’s not worth proceeding.

The initial idea for how to trade it prospectively is as below:

Example
Short USDMXN
from 10 - 20 minutes before triple rollover time
10m - 240m holding time (depending on when spreads revert back to normal for pair)
1:1 R:R i.e. equidistant TP and SL (or much shorter TP distance to have higher initial incidents of winning trades)
Receive the Carry interest - and accumulate every week
Allow the trade to win or lose.

Assuming random market, then probability of win or lose on 1:1 R:R is 50%

Trivial assumptions: 52 Trades, Risk = 500, Profit = 500, Carry Interest ~ 65, trade cost ~ 25, win rate = 50%

PnL Long run?

  • 52 x [(26 wins x 500) - (26 losses x 500) + (65 - 25)] (wrong)

*Correction: (26 wins x 500) - (26 losses x 500) + [52 weeks x (65 - 25)]

This only works if the market movement for the short (in duration) holding period is sufficiently random to assume a 50% win rate.