adrian2 wrote:Shakespeare wrote:That jogs my memory. I think the back-to-back needs non-registered funds for tax arbitrage on a prescribed annuity. IIRC, somebody linked a paper by Milevsky.
Correct
How nice. One word, a link, no hand-waving, no bullshit. Sometimes, not always but sometimes, it's very easy to tell the knowledgable from the ... er ... not so knowledgable.
I've just finished reading the paper - very interesting indeed. It seems that yes, there is [or was in 1999 when the paper was written] a good opportunity to squeeze anywhere between basically 0 [50-year-old, tax rate 25%] to 450bp [80-year-old, tax rate 50%] on an investment in such a product combination relative to a corporate bond with a term equal to the investor's estimated term (more, of course, relative to a bank deposit, which they do most of their benchmarking against).
All tax driven. It depends entirely on Canadian tax rules regarding distribution of expected mortality (far more spread out than what the insurance companies use) and the value of the expected mortality (the CCRA uses [used?] tables prepared in 1971 ... insurance companies are a bit more up to date). It is indeed nice to know that the life insurance industry is getting something for its political contributions beyond simple protection from the banks.
Mind you, I am a little bit skeptical of the authors' use of the term "arbitrage" to describe this strategy. As they say (Section 3.5)
Charupat & Milevsky wrote:Note also that this tax arbitrage exists only in an "expected" sense. It is possible that a posteriori, a mortality swap turns out to be worse than a bank deposit. This will happen if a dramatic rise in interest rates occurs shortly after the swap was constructed.
Even if the 'fairest' comparator is used to benchmark this strategy, a corporate bond maturing in the expected year of death, there is still significant interest rate risk if the annuitant should exceed his life expectancy in a time of higher interest rates. Better if rates are lower, of course, but it makes the hedge less perfect.
The authors point out that if tax rules become more reflective of reality and existing investments are not grandfathered, you've got some problems. So there's a certain amount of political risk.
Another risk not priced in by the authors is liquidity risk. Once you're in the strategy, you're IN ... you can't sell it off as easily as a corporate bond. The term of the strategy is, as far as the investor is concerned, infinite.
And as well, the numerical calculations performed in the article assume that the annuitant/insured will get the lowest rate possible on the insurance part ... pass the medical exam or forget it!
And ... oh yeah ... the funds have to be after tax. Doing this out of RRSP funds means everything's taxed and you don't get the benefit of the funny tax treatment.
Anyway, Adrian2, thanks again. Nice to find one nugget of information amidst 80 posts.