I am so tempted to short 30-year bonds right now. I might soon.
There are a few ways to represent this position:
1. Short bond futures (CME) – this is the world’s most liquid proxy to treasury bonds (other than dealing with the underlying product directly!). It has the advantages of liquidity and dealing strictly with the capital and not income component of the bond.
2. Purchase/short a liquid ETF that deals with long-term treasuries. There are a few to choose from:
– iShares 20-year+ government bond fund (TLT) – MER is 0.15%; average term of bond is 27.8 years – fund is highly liquid and shortable;
– Proshares Ultra/short (2x) 20-year+ (UBT/TBT) – MER is 0.95%; linked to TLT performance above as basis index. UBT is not very liquid, while TBT is very liquid.
ETFs have the advantage of being tradable in smaller amounts than futures (Future contracts are for $100,000 face value of product, which currently trade around 142% of par for the June contract). Futures typically have a spread advantage ($31.25 per $142,000 notional value), but liquid ETFs such as TLT have typically had penny spreads, resulting in comparable slippage. As previously mentioned the futures have an advantage with stripping the income-related aspects of the bond, and also tax advantages (both in the USA and Canada).
With long term bonds futures, you will have the “roll” to contend with.
ETF’s on the other hand are synthetic, and are usually composed of options and futures, so you not only have the roll, but you have option decay. I have found that it is usually better to short the opposite direction. So if you want to short bonds, short the bull fund as opposed to longing the bear fund. This generally puts you on the right side of the decay.
Something to look at anyway.
My humble opinion about the implementation of this widow-maker trade is that one should never, ever be long levered inverse ETFs for holding periods of any length. The converse is that — at general collateral — one should be short levered inverse ETFs at every opportunity.
In a hypothetical dream world in which your broker lends you UBT and TBT at GC, the best way to bet against Treasuries is shorting UBT. But in that dream world one can simply run a relatively low risk, unrebalanced arbitrage position of being short both UBT and TBT.
fsYard — Didn’t notice your post. Can you get me a general collateral or anything close to that locate on, say, QID/QLD pair or SDS/SSO pair? Rainbows and Unicorns, at least from my share lending contacts.
Sorry Ptuomov, I can’t help you with that.
Ptuomov,
In your hypothetical dream world, how would you handle the rebalancing?
Thanks!
FsYard
Even the NYTImes is reporting this play:
http://www.nytimes.com/2012/02/28/business/era-of-low-cost-borrowing-benefits-federal-government.htm
I would not rebalance, except when harvesting tax losses from the losing leg. For tax purposes, I would sometimes close the losing leg and replace it temporarily with an alternative ETF for 60 days. During this tax loss harvesting, I’d realign the losing leg to match the (unrealized) winning leg.
If you start dynamically hedging that long-short position, the joke’s on you and you’ve just lost all your edge (created by the ETF’s trading costs) with your own trading costs. Instead, the right thing to do here would be to just put the position on _in moderation_ and then not rebalance, with the exception of tax losses.