Sector rotation, thoughts for long only traders

[sinner]

Hi,

I've been doing research on sector rotation recently, most of you who have researched the subject are probably familiar with this image or one much like it



I have been thinking a lot about this chart over the last couple of days.

To my mind it seems apparent that if you are a long only investor who likes to follow the market trend then sectors like Industrials, Materials, Energy will perform the best against a timed market.

e.g. if you tried the same long only strategies with Healthcare or Finance you would be going against the larger stock market trend and while returns may outperform the index for such sectors at that point in the cycle they might not be profitable. Therefore positions in sectors which go against the market trend might require hedging against the market (e.g. long Financials short XJO) to return a profit and this is out of the scope of a long only investor.

So traders who follow the "market trend" when its up to buy spec stocks should be focusing in the sectors associated with a bull market rather than all sectors, and hopefully scans should confirm the rotation into and out of each sector to give extra confidence in the model.

I also note, that the energy sector sits at the top of the stock market cycle, so when energy stocks are outperforming the index and then stop outperforming that is indicative a cyclical market top is being put in place. This seems confirmed when looking at the XLE/SPY or XEJ/XJO for 2011.

What do you guys think?

 

[skc]

I think sector rotation has good merits but there are a few factors in play that means you need to tailor the sector rotation approach slightly differently for the Australian market...

- Most ASX sectors are dominated by 2-3 companies... variation in the performance of an individual company can throw the whole sector's performance out of whack.

- Australia is export driven, esp for base materials. So if our customers (i.e. China) are not experiencing the same cycle phases as we are, the sector cycles would not conform to the textbook.

- Rotating out of a sector incurs capital gains tax which may erode a large part of the edge.

There are probably ways to mitigate some of these, but I don't have time to think it through at the moment.

 

[sinner]

I think sector rotation has good merits but there are a few factors in play that means you need to tailor the sector rotation approach slightly differently for the Australian market...

- Most ASX sectors are dominated by 2-3 companies... variation in the performance of an individual company can throw the whole sector's performance out of whack.

Well I was thinking that you could use a simpler trend following system and instead of taking exposure to the S&P market cap weighted index, you simply use your trend following system in the sectors matching the current rotation.

- Australia is export driven, esp for base materials. So if our customers (i.e. China) are not experiencing the same cycle phases as we are, the sector cycles would not conform to the textbook.

This is a chicken and the egg problem, but if you think about it this shouldn't be a problem.

- Rotating out of a sector incurs capital gains tax which may erode a large part of the edge.

Yeah that is a problem.

 

[Eager]

Yeah that is a problem.

Ahhh, I get it now.

Buy shares with the aim of making a loss, because it is better to lose money rather than make a profit, and pay a portion of that profit in tax, but you get to keep the rest - whether it be enough to buy a burger, a bird bath, a holiday or a house.

Right.

 

[Moriarty]

Ahhh, I get it now.

Buy shares with the aim of making a loss, because it is better to lose money rather than make a profit, and pay a portion of that profit in tax, but you get to keep the rest - whether it be enough to buy a burger, a bird bath, a holiday or a house.

Right.

I have to admit I don't understand this either. If you're planning on making a capital return on stocks, you can't escape paying CGT, unless you count holding them from more than 12 months so you can access the 50% discount, but then there's no point passing up a gain just because you have to pay tax on it. Perhaps there's something I'm missing here...?

 

[skc]

I have to admit I don't understand this either. If you're planning on making a capital return on stocks, you can't escape paying CGT, unless you count holding them from more than 12 months so you can access the 50% discount, but then there's no point passing up a gain just because you have to pay tax on it. Perhaps there's something I'm missing here...?

We are talking about CGT in the context of sector rotation. Not CGT on its own. The comment on CGT is made with reference to no sector rotation... i.e. if you would just buy and hold a sector.

Let's say a sector's performance for the next 6 years at +5%, +8%, -2%, -6%, +2%, +5%.

If you simply buy and hold you would turn $100 into $111.88.

If you say perform sector rotation and sat out the years with -6% and +2%, your return would instead be $100 into $116.69 (without CGT).

But if you pay 22.5% tax (discounted CGT at max marginal tax rate) when you rotate out of the sector, the return would be $100 into $114.06

So while sector rotation may give you an edge compared to buy-and-hold ($116.69 vs $111.88), that edge is eroded quite significantly by CGT to $114.06 in this example.

Obviously different return distribution, entry and exit timing and tax rates would affect the magnitude differently, but it is something that must be taken into consideration.

Ahhh, I get it now.

Buy shares with the aim of making a loss, because it is better to lose money rather than make a profit, and pay a portion of that profit in tax, but you get to keep the rest - whether it be enough to buy a burger, a bird bath, a holiday or a house.

Right.

Before you make sacarstic comments towards an argument perhaps you should try to understand it first.

 

[Moriarty]

We are talking about CGT in the context of sector rotation. Not CGT on its own. The comment on CGT is made with reference to no sector rotation... i.e. if you would just buy and hold a sector.

Let's say a sector's performance for the next 6 years at +5%, +8%, -2%, -6%, +2%, +5%.

If you simply buy and hold you would turn $100 into $111.88.

If you say perform sector rotation and sat out the years with -6% and +2%, your return would instead be $100 into $116.69 (without CGT).

But if you pay 22.5% tax (discounted CGT at max marginal tax rate) when you rotate out of the sector, the return would be $100 into $114.06

So while sector rotation may give you an edge compared to buy-and-hold ($116.69 vs $111.88), that edge is eroded quite significantly by CGT to $114.06 in this example.

Obviously different return distribution, entry and exit timing and tax rates would affect the magnitude differently, but it is something that must be taken into consideration.
:

Ok, so it's more a turnover vs buy-and-hold decision. That makes sense from the point of view of whether or not you hold long enough to get the discount, but if you assume you receive the discount regardless, you'll end up paying CGT on your gain at the same rate when you eventually sell your buy-and-hold shares later on down the track...

I tend to be medium-long term holder - I try not to turnover prior to 12 months so I can access the discount, but beyond that I tend to ignore CGT as a fact of life (unfortunately).

 

[Eager]

We are talking about CGT in the context of sector rotation. Not CGT on its own. The comment on CGT is made with reference to no sector rotation... i.e. if you would just buy and hold a sector.

Let's say a sector's performance for the next 6 years at +5%, +8%, -2%, -6%, +2%, +5%.

If you simply buy and hold you would turn $100 into $111.88.

If you say perform sector rotation and sat out the years with -6% and +2%, your return would instead be $100 into $116.69 (without CGT).

But if you pay 22.5% tax (discounted CGT at max marginal tax rate) when you rotate out of the sector, the return would be $100 into $114.06

So while sector rotation may give you an edge compared to buy-and-hold ($116.69 vs $111.88), that edge is eroded quite significantly by CGT to $114.06 in this example.

Obviously different return distribution, entry and exit timing and tax rates would affect the magnitude differently, but it is something that must be taken into consideration.



Before you make sacarstic comments towards an argument perhaps you should try to understand it first.

My understanding is that any profit is better than any loss, whether tax is paid on it or not.

 

[Eager]

I tend to be medium-long term holder - I try not to turnover prior to 12 months so I can access the discount, but beyond that I tend to ignore CGT as a fact of life (unfortunately).

Same, just like brokerage.