Skip to main content
Track Carry trading
4 / 5
Library

2007issue C101-4

Turtle forex breakouts and the daily carry ledger

A classroom check keeps turtle-trading blind to rates, then replays the same forex-pair positions as a daily funding-overlay. The 2007 dollar-yen and New Zealand dollar maps show when trend-following and a carry-trade pointed the same way.

  • A forex-pair always owns one currency and shorts another, so interest is received on one leg and paid on the other for each day the position stays open.
  • Turtle-trading hunts longer-term breakouts and ignores national interest-rate levels, while a carry-trade buys the higher-yielding currency and tries to hold while the interest-rate-differential stays favorable.
  • A 2007 dollar-yen illustration, and a New Zealand dollar advance against the dollar, were cases where the trend position and the positive-carry overlay pointed the same way.
  • If being long the higher-rate currency were statistically random, carry would be expected to net near a wash, except that brokers were described as charging more on negative-carry than they credit on positive-carry.
Entries in this reading3 entries

Two ledgers for one forex-pair

Editorial frame: treat the 2007 illustration as a classroom check with two ledgers. First, take turtle-trading entries, exits, and holds from price breakouts only. Second, replay those same positions as a funding-overlay of daily credits and debits.

A forex-pair always means owning one country's money and shorting another's. The holder is entitled to interest on the long currency and obligated to pay interest on the short currency. The brokerage applies that interest as a daily credit or debit for each day the position remains open, so every forex trade has some carry, whether small or large.

Turtle-trading does not read rates

The turtle-trading procedure described here hunts longer-term breakouts and ignores national interest-rate levels. Entry, exit, and hold decisions come from price structure, not from monetary fundamentals. Trend-following is treated the same way: continued directional movement is taken from chart breakouts.

A carry-trade is the opposite habit. Carry-only desks buy the higher-yielding currency, sell the lower-yielding one, and try to hold while the interest-rate-differential stays favorable. They may accept a limited adverse price move if the pip loss does not fully cancel the interest they are collecting.

Replay the same trades as a funding-overlay

Historical interest-rate series usable for pair differentials were described as available from the start of 2005. That window was treated as long enough to isolate the funding-overlay, the daily credit or debit that accumulates whether price is moving with or against the trade.

A 2007 dollar-yen illustration used U.S. rates near 5 percent and Japanese rates near 1 percent. Long the pair implied about a 4 percent positive-carry. Short the pair implied the reverse daily cost, which is negative-carry.

When trend-following and carry-trading pointed the same way

The 2007 case tied a multi-year decline in the U.S. dollar and a still-weaker yen to a stretch of long-foreign, short-dollar trend trades. Those holdings more often collected the higher rate than paid it.

A contemporaneous New Zealand dollar advance versus the U.S. dollar was cited with local rates around 7 to 8 percent. There the trend position and the positive-carry overlay pointed the same way.

Why random carry would not stay a wash

If being long the higher-rate currency were statistically random, carry would be expected to net near a wash over time. Brokers were described as charging more on negative-carry holdings than they credit on positive-carry holdings.

Editorial reading: that wider debit than credit is how an uncorrelated funding-overlay becomes a slow leak.

What had changed

Trend-following, turtle-trading, and daily interest settlement were treated as already present in forex. What had changed was the ability to separate and measure the carry overlay.

Educational research material, not investment advice. Historical source context does not establish present-day performance.
4 of 5 in the Carry trading track
201034-39 pp.Next on Carry tradingCross-pair correlation regimes in uncertain marketsCorrelation analysis is applied to the two dollar legs first. Pearson’s product-moment coefficient is the standard linear measure of how those pairs move together, scaled from -1 to +1.
All readings on this track · 5 readings
  1. 2006Three-bank carry calendar when dollar spreads stop widening
  2. 2006Selecting a currency pair by policy, carry, and oil translation
  3. 2007When dollar rebounds meet carry and reserve outflows
  4. 2007Turtle forex breakouts and the daily carry ledger
  5. 2010Cross-pair correlation regimes in uncertain markets
All 5 readings tagged Carry trading
Also on Carry trading5 readings