UNG (Nat Gas), USO (Crude Oil), and the Term-Structure Play
Both ETFs (but mostly UNG) are popular term-structure plays as of late (by shorting the ETF to capture the benefit of contango in the futures on which they’re based).
But how well would that approach have performed historically? In this post I’ll try to answer that using our expanded dataset.
The graphs above show the same super simple strategy applied to both UNG and USO (see end of post for a discussion of more sophisticated approaches):
Go long at today’s close when the nearest month futures contract closes above the second month (backwardation), or short when it closes below (contango). Hold the position until the opposite condition is met.
The strategy is in red and buy & hold is in grey. Results do not include transaction costs, slippage, taxes, or return on cash.
This simple strategy would have performed well over the last few years, and that’s surely why traders have been lulled in to thinking UNG and (to a lesser degree) USO are akin to VXX (i.e. that persistent contango is the primary driver of returns, which is itself not always true).
But our expanded dataset shows that, over a long enough horizon, this approach has been a dud, and will most likely continue to be a dud in the future.
Too Simple by Half
Obviously, this is an oversimplified example.
But I reach a similar conclusion when I look at other variations such as: (a) using all months (rather than just the first/second months) to measure the futures term-structure, (b) only trading very strong contango or backwardation, (c) only considering the term-structure around the 4 days each month when each ETF rolls contracts, (d) only considering longs or shorts, and (e) using moving averages to smooth the term-structure measurement.
In short, whereas other futures-based ETPs like VXX/XIV are (somewhat) tradeable simply by following the futures term-structure (read more), that’s not the case for UNG and USO.
Long-term trading success here might include taking the term-structure into consideration, but will also require timing the underlying commodity as well.
. . . . .
Filed under: Trading Strategies | 2 Comments