Country CAPE Offered Little Predictive Value For 2015 Performance

| About: Cambria Global (GVAL)
This article is now exclusive for PRO subscribers.


Country CAPE predicted stock market performance well in 2013, but poorly in 2014.

It had little predictive value in 2015.

GVAL might be hurt by the negative economic consequences of Brexit.

The cyclically-adjusted price-earnings (CAPE) ratio, also known as the Shiller P/E ratio, is a popular tool used to value markets. This metric was developed by U.S. economist and Nobel prize winner Robert Shiller, and is defined as price over 10-year average earnings adjusted for inflation. Using 10-year average earnings rather than one-year trailing (or forward) earnings smoothes out fluctuations in net income caused by variations in profit margins over a typical business cycle.

The use of CAPE as a valuation indicator does have its limitations. While high CAPE values have been associated with lower future long-term returns, and lower CAPE values with higher future long-term returns, the CAPE ratio does not reliably predict near-term tops and bottoms. As a case in point, the CAPE ratio has been saying that the U.S. market has been overvalued for years, but despite this, equities have continued to grind higher. Conversely, emerging market and peripheral European countries, despite having persistently low CAPE ratios for years, have seen continual years of stock market losses as news just seems to go from bad to worse.

The CAPE is a personal favorite metric of Mebane Faber, investor, author and co-founder and Chief Investment Officer of Cambria Investment Management. Before launching the Cambria Global Value ETF (NYSEARCA:GVAL) in early 2014, Faber enthusiastically made the case for using country CAPE ratios to select undervalued stock markets. In a Jan. 2nd, 2014, posting on his "Meb Faber Research" website, Faber points out that the countries with the cheapest CAPE ratios massively outperformed those with the highest CAPE ratios in 2013. The bottom-5 and bottom-10 CAPE countries averaged 20.74% and 21.11% returns, respectively, in 2013, while the top-5 and top-10 CAPE countries averaged -17.81% and -5.39%, respectively. This represents a differential of 38.59% for the top-5 versus bottom-5, and 26.5% for the top-10 versus bottom-10, a remarkable outperformance.

However, as described in "Country CAPE Ratios: Wizard In 2013, Dunce In 2014?" and "Country CAPE Ratios (Part 2): Local Currency Returns", CAPE either predicted the reverse trend (USD returns), or had little predictive power at all (local currency returns). Faber himself admitted as much, writing:

[CAPE] worked great in 2013, but in 2014 that worked fairly poorly as many of the cheap markets have gotten even cheaper. I joked with a friend the other day that I was going to write a new edition of my book called Global More Value.

CAPE in 2015

How did country CAPE fare in 2015? On New Year's Day 2016, Faber wrote:

Using a value approach worked well however. The cheapest third of global stock markets outperformed the most expensive by about 5 percentage points.

As is custom, Faber quoted only country ETF performance numbers, which are unhedged and thus represent USD returns. However, several readers in my past articles commented that local currency returns should really be used, given that USD returns are influenced by exchange ratio fluctuations that are independent of actual country stock market performance.

Additionally, I noticed that Faber had moved away from ranking solely using CAPE, and has instead used a "valuation composite of long term ratios like CAPE". However, given that my past articles had used only CAPE, I still wanted to perform the analysis for 2015 using this single metric.

The following table shows a list of countries ranked in ascending order of CAPE ratio at the end of 2014. Also shown are their 2015 USD returns, based on the respective country ETF performance, and their 2015 local currency (LC) returns (source: Bespoke). Note that two countries - Hungary and Czech Republic - do not have ETFs, hence their USD returns are not shown.

Country CAPE end-2014 ETF 2015 USD returns / % 2015 LC return / %
Greece 2.8 (NYSEARCA:GREK) -39.70 -25.25
Russia 5.2 (NYSEARCA:RSX) 3.72 -2.66
Hungary 5.9 43.68
Austria 7.3 (NYSEARCA:EWO) 4.60 9.85
Portugal 7.7 (NYSEARCA:PGAL) 2.35 18.24
Italy 9.6 (NYSEARCA:EWI) 3.14 13.32
Brazil 10.0 (NYSEARCA:EWZ) -41.80 -12.31
Czech Republic 10.3 0.95
Poland 10.8 (NYSEARCA:EPOL) -23.20 -18.25
Ireland 11.0 (NYSEARCA:EIRL) 22.82 30.76
Turkey 11.3 (NYSEARCA:TUR) -31.40 -13.78
Spain 11.6 (NYSEARCA:EWP) -15.50 -6.34
Norway 12.1 (NYSEARCA:NORW) -15.70 2.95
U.K. 12.1 (NYSEARCA:EWU) -7.02 -4.25
South Korea 12.4 (NYSEARCA:EWY) -8.00 2.65
Egypt 13.2 (NYSEARCA:EGPT) -33.40 -22.15
France 13.8 (NYSEARCA:EWQ) 0.24 9.51
Singapore 13.8 (NYSEARCA:EWS) -18.30 -17.63
China 14.3 (NYSEARCA:ASHR) -2.06 10.17
Peru 14.3 (NYSEARCA:EPU) -35.90 -34.18
Finland 14.5 (BATS:EFNL) 3.52 11.01
New Zealand 14.6 (NYSEARCA:ENZL) -1.99 13.01
Belgium 14.6 (NYSEARCA:EWK) 13.77 13.80
Israel 14.8 (NYSEARCA:EIS) 7.79 2.48
Netherlands 15.5 (NYSEARCA:EWN) 1.53 5.04
Australia 15.7 (NYSEARCA:EWA) -9.94 -2.66
Germany 15.8 (NYSEARCA:EWG) -2.79 10.28
Chile 17.4 (BATS:ECH) -18.50 -4.47
Thailand 17.8 (NYSEARCA:THD) -22.00 -14.28
Hong Kong 18.2 (NYSEARCA:EWH) -1.17 -6.80
Malaysia 19.0 (NYSEARCA:EWM) -20.60 -4.31
Sweden 19.1 (NYSEARCA:EWD) -4.54 -1.07
Canada 19.2 (NYSEARCA:EWC) -23.90 -9.61
Taiwan 19.7 (NYSEARCA:EWT) -12.90 -10.89
South Africa 20.9 (NYSEARCA:EZA) -26.00 2.40
Indonesia 20.9 (NYSEARCA:EIDO) -22.80 -12.58
Switzerland 22.4 (NYSEARCA:EWL) 0.27 -1.20
Mexico 22.6 (NYSEARCA:EWW) -14.20 0.98
Colombia 23.1 (NYSEARCA:ICOL) -40.10 -26.34
Japan 23.4 (NYSEARCA:EWJ) 9.17 8.78
U.S. 23.6 (NYSEARCA:SPY) 1.23 0.76
Philippines 26.1 (NYSEARCA:EPHE) -10.80 -3.85
Denmark 30.0 (BATS:EDEN) 20.63 36.22

Let's first plot 2015 country performance against end-2014 CAPE in USD terms (note that data labels for some smaller countries have been omitted for clarity):

We can see from the chart above that there is a negligibly weak positive correlation (R^2 = 0.0113) between 2015 USD returns and CAPE, indicating that CAPE had little predictive value for 2015.

However, the strong U.S. dollar in 2015 meant that a number of countries' stock market performances were artificially depressed. The following chart shows the correlation between 2015 local currency returns and end-2014 CAPE:

We can see that with local currency returns, there is a nominal negative correlation, but it is practically non-existent (R^2 = 0.0005). This indicates that there was no correlation between CAPE and country stock market performance in 2015.

Thus, I'm inclined to suspect that Faber's claim that the cheapest third of countries outperformed the most expensive third by 5% involved a bit of data mining.


While CAPE did not have any predictive value in 2015, the ETF based largely on this metric, GVAL, didn't perform all that badly in 2015 (-8.21%) compared to global shares (NASDAQ:ACWI) (-2.21%).

GVAL Total Return Price Chart

In 2016, GVAL has been keeping pace thus far with global shares. GVAL currently holds these countries in its portfolio: Austria, Italy, Russia, Portugal, Spain, Greece, Poland, Hungary, Brazil, Czech Republic, and Norway. Unfortunately, many of these countries are susceptible to the negative economic consequences of Brexit, which does not bode well for this ETF for the rest of this year.

GVAL Total Return Price Chart

Disclosure: I am/we are long GVAL.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.