My CAPE10 based IPS

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Park
Posts: 690
Joined: Sat Nov 06, 2010 4:56 pm

Re: My CAPE10 based IPS

Post by Park » Sun Sep 30, 2018 11:31 am

The following summarizes previous posts by me on this thread.

https://www.starcapital.de/fileadmin/us ... imling.pdf

See Fig. 6. When CAPE less than 10, historical real returns over the next 10-15 years have been a minimum of 5%. When CAPE greater than 50, returns have been negative on average. If you exclude the Danish and Swedish stock markets, returns have been modest at best, when CAPE greater than 40. For a well diversified investor, I question how relevant the data from the Danish and Swedish stock markets is. For them, the S&P500 data is more appropriate. When S&P500 CAPE is more than 40, there hasn't been a positive return.

The following is from Figure 8.

All data is in local currency, adjusted for inflation and takes into account dividends.

The following is maximum drawdown in relationship to CAPE, using CAPE data from all countries.
CAPE 0-10, average maximum drawdown over the next 3 years is -5.7%, average maximum drawdown over the next 15 years is -5.2%
10-15, -8.8%, -11.1%
15-20, -12.4%, -13.6%
20-25, -18.1%, -23.1%
25-30, -22.3%, -27.5%
30+, -28.8%, -39.5%

More emphasis is given on the returns associated with high valuations, than on the risk associated with such valuations. The disparity in risk adjusted returns, at extremes of valuation, is not small.



viewtopic.php?f=10&t=251522

" Stocks have had remarkably consistent 5%-6% real returns over four centuries."

The above is from a Bogleheads thread started by SimpleGift.

Assume a PE10 of 40. Based on the CAPE model, the predicted real return is 2.69%, or about half the historical average. Stocks have to do about double what the model predicts to get the historical average. And the risk of exposure to a bear market has gone up considerably.




The criticism of moving goal posts has been raised, and it's a legitimate one. My response is as follows. Investors expect an equity risk premium. When you own an asset that could lose half of its value in 2 years, you expect one. Goal posts may change, but I doubt that the equity risk premium will disappear. How much of an equity risk premium is necessary for investors to put money in stocks. My bet is that an expected real return of less than 2.7% is not enough. In part, that conclusion is based on historical data.


A reasonable criticism is how accurate is return based on CAPE? When you look at CAPE in general, not very accurate. But when you look at CAPE at extremes, not too bad. However, that is based completely on historical data :-).



The following is a digression. CAPE has its greatest ability to predict over 10-15 year time horizons. And that's a similar horizon that someone using leverage should have. Present US CAPE is around 33.4, for an expected return of 3.21%. With the cost of leverage subtracted from that, returns using leverage might be low at best.

User avatar
siamond
Posts: 5135
Joined: Mon May 28, 2012 5:50 am

Re: My CAPE10 based IPS

Post by siamond » Sun Sep 30, 2018 6:32 pm

staythecourse wrote:
Sun Sep 30, 2018 10:27 am
Now here comes the problem. Those tails are derived from knowing the results of the ENTIRE data set to now. The guy in late 1990's only had the data UP to his time period. He did not have the resulting fall out AFTER the late 1990's showing PE>30's was a disaster.
Totally agreed, way too many CAPE studies fall in this trap of using ex post data to compute thresholds, and then backtest with such thresholds determined from the entire data set. Please note that the simulations I ran in this thread did not use *any* ex post data, all thresholds are recomputed year over year, solely based on ex ante data.

Park
Posts: 690
Joined: Sat Nov 06, 2010 4:56 pm

Re: My CAPE10 based IPS

Post by Park » Sun Sep 30, 2018 7:55 pm

siamond wrote:
Sun Sep 30, 2018 6:32 pm
staythecourse wrote:
Sun Sep 30, 2018 10:27 am
Now here comes the problem. Those tails are derived from knowing the results of the ENTIRE data set to now. The guy in late 1990's only had the data UP to his time period. He did not have the resulting fall out AFTER the late 1990's showing PE>30's was a disaster.
Totally agreed, way too many CAPE studies fall in this trap of using ex post data to compute thresholds, and then backtest with such thresholds determined from the entire data set. Please note that the simulations I ran in this thread did not use *any* ex post data, all thresholds are recomputed year over year, solely based on ex ante data.
The question is how good does CAPE predict real return, and that is based on historical data. The study I cite from Star Capital is based on 17 countries For 12, the data is based on 34 year market histories, with a starting date of Dec 1979. For 5 countries, the time period is less, with a minimum of around 20 years. We have historical data from CAPE of 4.0 to 91.5. I doubt that the range of CAPE will appreciably increase with time. If markets are becoming more efficient with time, the range might lessen. And at extreme valuations, CAPE has shown historically some ability to predict 10-15 year real return.

Park
Posts: 690
Joined: Sat Nov 06, 2010 4:56 pm

Re: My CAPE10 based IPS

Post by Park » Mon Oct 01, 2018 2:22 pm

https://www.etf.com/sections/swedroe-un ... nopaging=1

"the Nikkei Index peaked at 38,916 on Dec. 29, 1989. It finally reached its low of 7,054 on March 10, 2009. And at about 23,400 (its level on Sept. 18, 2018), it is still down about 40% from its peak almost 29 years later."

The above concisely states why I am interested in valuation based market timing.

Post Reply