Risk premium measure

[kid hustlr]

Just a quick one.

Have been thinking recently about risk premium and the return on the ASX200 vs the aussie 10 year bond. As mentioned in this thread I think it's an interesting time moving forward.

What I'm trying to derive is a quick measure of risk premium. I know a fairly 'standard' one is dividend yield vs 10 year yield but to me that's not ideal.

Link below shows the historical relationship and it's not very mean reverting.

http://asxiq.com/blog/all-ords-dividend-yield-minus-10-yr-bond-yield-spread-at-2/

Effectively dividend yield isn't a great measure because companies can retain profits/reinvest/etc so it doesn't paint the whole picture.

I'm thinking a better comparison is forecast ROE vs 10 year yield? Any other thoughts/suggestions?

Edit: perhaps I should just inverse the current p/e ratio and compare to the 10 year yield?

EDIT 2: Investopedia agrees http://www.investopedia.com/terms/e/earningsyield.asp

 

[DeepState]

Just a quick one.

Have been thinking recently about risk premium and the return on the ASX200 vs the aussie 10 year bond. As mentioned in this thread I think it's an interesting time moving forward.

What I'm trying to derive is a quick measure of risk premium. I know a fairly 'standard' one is dividend yield vs 10 year yield but to me that's not ideal.

Link below shows the historical relationship and it's not very mean reverting.

http://asxiq.com/blog/all-ords-dividend-yield-minus-10-yr-bond-yield-spread-at-2/

Effectively dividend yield isn't a great measure because companies can retain profits/reinvest/etc so it doesn't paint the whole picture.

I'm thinking a better comparison is forecast ROE vs 10 year yield? Any other thoughts/suggestions?

Edit: perhaps I should just inverse the current p/e ratio and compare to the 10 year yield?

EDIT 2: Investopedia agrees http://www.investopedia.com/terms/e/earningsyield.asp

You are looking at the Fed model when you compare EY to Bond yield. It is technically not correct as the 'right' equilibrium rate differential will differ depending on the inflation level which is not stated in the equation.

Other measures worth looking at are DCF/DDM based, but these are more assumption heavy that purely historical measures. If you are looking at historically oriented measures only then the cyclically adjusted Shiller measures are well favoured. Check out CAPE and its variants. Useful sites include Hussman and GMO. Shiller has his own site. Interestingly, these measures have worked better when bond yields are not even considered.

My view is that these are very useful starting points although, like any single metric, the underlying data is worth considering also if a lot is riding on this.

At this time, you will likely find that the US market is trading expensive and the Australian market is fair to slightly expensive. These calls do not take into account the impact of LSAP programs and ultra low interest rates at the short end causing search for yield and risk taking activity.

 

[kid hustlr]

Yeah thx RY was hoping u would chime in. I put a note in my work about the impact of inflation but perhaps I underestimated it's importance.

Effectively I just wanted some type of rough guide which identified when a market may be cheap or expensive. I think I've got that with the earnings yield vs bond yield measure but clearly I have a lot more to think about if trying to develop some kind of long term passive style strategy. I'll have a look at your suggestions and report back with questions.

 

[craft]

Just a quick one.

Have been thinking recently about risk premium and the return on the ASX200 vs the aussie 10 year bond. As mentioned in this thread I think it's an interesting time moving forward.

What I'm trying to derive is a quick measure of risk premium. I know a fairly 'standard' one is dividend yield vs 10 year yield but to me that's not ideal.

Link below shows the historical relationship and it's not very mean reverting.

http://asxiq.com/blog/all-ords-dividend-yield-minus-10-yr-bond-yield-spread-at-2/

Effectively dividend yield isn't a great measure because companies can retain profits/reinvest/etc so it doesn't paint the whole picture.

I'm thinking a better comparison is forecast ROE vs 10 year yield? Any other thoughts/suggestions?

Edit: perhaps I should just inverse the current p/e ratio and compare to the 10 year yield?

EDIT 2: Investopedia agrees http://www.investopedia.com/terms/e/earningsyield.asp

Is this the sort of thing you are thinking of?

https://www.aussiestockforums.com/forums/showthread.php?t=23385&page=18&p=710550&viewfull=1#post710550

Chart updated and discussed a bit further here.

https://www.aussiestockforums.com/forums/showthread.php?t=28384&p=824016&viewfull=1#post824016

 

[kid hustlr]

thx craft will have a look. At a glance yes that is although my knowledge won't extent as far you guys so will take some time to understand it - will report back in time.

 

[kid hustlr]

Just thought I'd bump this. I have read through those post's several times craft and they are very very good.

I mentioned before I'm not of the 'fundamental mould' but having an understanding of this type of stuff I think will only benefit me moving forward.

I really liked your trend earnings calc as well it smooths the data - my data is very spiky around the GFC for example so it skews it pretty strongly.

Finally, I also note your confidence around June 2012 - I suggest other members check ASX performance since June 2012.

Finally part 2 - interested to hear your thoughts now, with another wave of cuts starting (or on the horizon?) my premium measure is starting to look bullish again.

 

[valuepack]

Not sure if it this really relates to your thread on risk premium as this is for US stocks but I just wanted to share this a graph I created of S&P 500 Index vs dividend yield vs earnings % (1 ÷ PE Ratio) vs inflation vs T-Bond 10 year rate:



Based on data from:

http://pages.stern.nyu.edu/~adamodar/New_Home_Page/datafile/histretSP.html

Data for Australia seems a little bit harder to find.