Pricing equities / Shouldn't I be putting my money in Int'l?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
Posts: 377
Joined: Sun Dec 10, 2017 4:48 am

Pricing equities / Shouldn't I be putting my money in Int'l?

Post by danielc » Thu Jul 05, 2018 11:29 am


This question might get dangerously close to market timing. I'm not interested in or advocating market timing. I am currently reading Bill Bernstein's book Rational Expectations: Asset Allocation for Investing Adults. Specifically, I'm reading the sectiona bout the Gordon Equation which Bernstein writes as:

(expected return) = (dividend yield) + (dividend growth rate)

To this he adds another term to account for the change in valuation. Jack Bogle refers to this as the speculative term. It refers to the fact that P/E ratios change over time as people alternate from exuberance to depression.

(expected return) = (dividend yield) + (dividend growth rate) + (change in valuation)

So I want to apply this formula to the present day. If I look at ITOT (total stock market ETF), I see a current distribution yield of 1.9%. Back in the book, Bernstein suggests a historical real dividend growth rate of around 1.33%. In other words:

dividend yield = 1.90%
dividend growth rate = 1.33%

To estimate the last term, Berstein looks at the CAPE or the Shiller P/E ratio. Right now the CAPE is at 32 and the historical average is 17. Bernstein suggests that the CAPE has been increasing over long periods of time, so perhaps a CAPE of 20 is a better estimate of a "reasonable" valuation at the present day. Alright. So the expectation is that the CAPE will be revised down from 32 to 20. Let's say you have an investment horizon of 25 years and we imagine that the CAPE becomes 20 over that time period. The annualized change in valuation would be:

change in valuation = (20/32)^(1/25) - 1 = - 1.86%

Now we can compute the real expected rate of return: return = 1.90% + 1.33% - 1.86% = 1.37%

That's a bit depressing. Now, Rick Ferri estimates 3% real return. You could get that number assuming no change to the CAPE. Now let's comapre this to international markets. In IXUS the dividend yield is 2.31% and while I do not have a CAPE for international markets, I note that their current P/E ratio is 14. That P/E is close to (slightly below) the historical value for the US. So perhaps it's not unreasonable to assume that valuations will not change a lot. If we assume the same 1.33% real dividend growth rate that Bernstein used for the US, you find that the expected return for international markets is:

Int'l real return = 2.31% + 1.33% + 0.00% = 3.64%

That looks a lot better than than the 1.37% for US stocks right now. Does this not mean that I should probably tilt toward international stocks? More generally, does this not mean that every year when I rebalance I should tilt toward whichever market has the highest expected yield as computed by the above formula?

Posts: 1500
Joined: Sun Mar 02, 2014 8:34 am
Location: Florida

Re: Pricing equities / Shouldn't I be putting my money in Int'l?

Post by asif408 » Thu Jul 05, 2018 1:08 pm


You have a good sense of humor making estimates of future returns to two decimal places, considering the uncertainty around future returns.

But to answer your question, I agree with your general premise, and in that book Bill suggests you will probably do well tilting to lower valuation areas, for example more in emerging and developed Int'l vs US right now.

Just a few notes: at the time that book was written in 2014 US CAPE wasn't 32, so consider what CAPE was in 2014 when making your adjustment. The other thing is country valuation averages may differ, so it may not be an apples to apples comparison. A suggestion would be to look at valuation compared to history for each market along with relative valuation between markets. Finally you could tilt every year, but you have to consider tax implications and momentum. Generally these bouts of outperformance last a few years, so rebalancing or overrebalancing may not be necessary that often, and may hurt more than help if a lot of it is in a taxable account.

User avatar
Posts: 2167
Joined: Thu Nov 20, 2014 8:57 pm
Location: Michigan

Re: Pricing equities / Shouldn't I be putting my money in Int'l?

Post by knpstr » Thu Jul 05, 2018 1:18 pm

You guess is as good as anyone.

No one knows. Perhaps a technological breakthrough is right around the corner that stimulates all world economies. Such events can't be predicted in advance. Such things tend to "take the world by storm".

Just pick an allocation you're comfortable with. Don't over think it. The most important part is to leave your funds alone (don't jump in and out) and add to your account at regular intervals.
Very little is needed to make a happy life; it is all within yourself, in your way of thinking. -Marcus Aurelius

User avatar
Posts: 5756
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: Pricing equities / Shouldn't I be putting my money in Int'l?

Post by triceratop » Thu Jul 05, 2018 1:22 pm

Historical dividend growth numbers draw may be an imperfect future predictor given that a substantial amount of cash returned to shareholders is done through share buybacks. If the data is nonstationary like this, using simple formulas can give incorrect results.
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."

User avatar
Svensk Anga
Posts: 442
Joined: Sun Dec 23, 2012 5:16 pm

Re: Pricing equities / Shouldn't I be putting my money in Int'l?

Post by Svensk Anga » Thu Jul 05, 2018 1:27 pm

There are at least a couple issues muddying the waters here.

The Gordon equation works with dividend yield and comes from a time before share repurchases became a significant component of corporate cash return to shareholders. If you add the net repurchase yield to the true dividend yield, projected returns are higher. From this post ( viewtopic.php?p=4002621#p4002621 ), it looks like net repurchase yield in the US is about 2%. I don't know that foreign firms buy back anywhere near this extent.

CAPE10 is high now in part because the 10-year trailing real earnings include the awful results from late 2008 and all of 2009. These will shortly start aging out of the average. CAPE10 should go down, even if stock prices don't budge. The market knows this and has priced US equity accordingly. Calculate a CAPE8 if it makes you feel better.

The earnings used in the CAPE10 calculation are now (since 2001) calculated on a different basis than the long-term history you are using as a reference. The newer definition of earnings gives lower results, so higher CAPE, than the old definition. The mean going forward might be higher than 20.

It might be that the US earnings growth rate that you used will be higher than the reality for foreign companies.

Post Reply