Emerging markets 1900-2013

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Topic Author
Robert T
Posts: 2574
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Emerging markets 1900-2013

Post by Robert T » Tue Aug 26, 2014 3:07 am

.
The 2014 issue of the Credit Suisse Global Investment Returns Yearbook, has an interesting section on emerging markets https://publications.credit-suisse.com/ ... B5D14A7818 Dimson, Marsh and Staunton create an emerging market index back to 1900 (using GDP per capita as the annual sort metric to classify whether a country is developed or emerging). Here are some extracts/key findings:

On construction of the index:
  • For the 23 countries in the Dimson, Marsh and Staunton (DMS) database in 1900, seven would have been deemed emerging markets [using the GDP per capita criteria] – China, Finland, Japan, Portugal, Russia, South Africa, and Spain. Three are still emerging today – 114 years later – namely China, Russia, South Africa.

    Countries beyond the DMS database were also added when data became available. Other markets included were Brazil and India since 1955; Korea and Hong Kong since 1963 (the latter moved to developed in 1977); Singapore since 1966 (until it moved to developed in 1980); Malaysia since 1970; Argentina, Chile, Greece, Mexico, Thailand, and Zimbabwe since 1978; Jordan since 1978; Colombia, Pakistan, Philippines and Taiwan since 1985; and Turkey since 1987, and the MSCI Index was used since 1988.
Sub-period EM performance illustrates risks and correlations with developed markets
  • Outperformed developed markets 1900 to 1917

    Hit by October 1917 Revolution in Russia when investors in Russian stocks lost everything

    Returned less than developed markets from early 1920s to 1929 peak, but also declined by less during Great Depression.

    Matched developed market returns from mid 1930s to mid-1940s

    Collapsed in 1945-1949. Largest contributor was Japan equities that lost almost 98% of their market value in US dollar terms. Markets in China closed in 1949 following the communist victory, and where investors in Chinese equities effectively lost everything. Other markets such as Spain and South Africa also performed very poorly in the immediate aftermath of World War II. [Just to note the DMS index only included 6 countries over this period – China, Japan, Portugal, Russia, South Africa, Spain; Finland moved to developed in 1932).

    From 1950 to 2013 emerging markets staged a long fight back, albeit with periodic setbacks. Highest frequency of crises was in 1980s and 1990s – on average, in these two decades, emerging countries averaged more than 3 times more crises per country than developed countries.

    From 2000-2010 the annualized returns on MSCI EM index was 10.9% versus just 1.3% for developed markets.

    Emerging markets underperformed developed markets over 1900-2013 period, but outperformed since 1950
    • Annualized return (%)

      1900-2013
      Emerging markets equities = 7.4%
      Developed market equities = 8.3%

      1950-2013
      Emerging markets equities = 12.5%
      Developed markets equities = 10.8%

      Since 1950 emerging markets outperformed but since 1900 they underperformed developed markets (the risks that showed up in 5 of the 114 years (1945-1949) where significant enough to lead to underperformance over the full 114 year period.
    Correlation with developed markets have increased
    • Correlations of returns with developed market have increases from around 0.55 (correlation coefficient) in 1980s-1990s to about 0.8 in 2000s, and average volatility of EM country returns has declined.
    ‘Value’ has outperformed ‘growth’ both within and across EM countries
    • With the data they have, 2000-2013, value outperformed growth in EM by an average of 4.1%, which was larger than the 3.1% for developed markets (small caps underperformed in EM relative to developed markets). Just to note Jason Hsu's EM database which starts in 1995 also shows a large value premium in EM

      From 1976-2013, quintiles sort of EM countries by dividend yield, annually rebalanced, showed the quintiles with highest yields provided an annualized return of 31%, while those with the lowest start-year yields gave an annualized return of 10%. The highest yielding countries outperformed the lowest yielders by 19% per year.
    Relevant points for investors (at least to me)
    • EM are not guaranteed to outperform developed markets

      EM returns had larger ‘negative tails’, and when these showed up as in 1945-1949 they had significant negative effects

      Diversification matters, with only half a dozen countries in an index, bad performance of a few countries can dramatically effect aggregate performance (perhaps a lesson for the frontier market indexes)

      With higher correlations with developed markets and declining volatility, EM diversification benefits have likely declined
    What does this mean for inclusion of EM in a portfolio (at least to me)
    • If standard deviations and correlations of EM and developed market equities is similar to those since 1990, then (at least by my estimate), EM returns would need to be 2% higher than developed markets to justify inclusion in a portfolio on a mean-variance basis – and only up to about a 10% allocation. Swensen in his 2009 edition of the Pioneering Portfolio Management), expects a 2% premium (as was the case since 1950). I presume his analysis of expected capital market returns, volatilities and correlations informed his suggestion to investors for a 7-14% EM allocation within equities. This is similar to what I also get for an 'optimal' allocation (on a mean-variance efficiency basis) using a 2% expected EM premium, with similar volatility and correlations to the last two decades. If we expected the volatility to be lower, and correlations to be higher, but with a continued 2% premium then the ‘optimal’ allocation would be higher, and if we expected the premium to be lower, but correlations and volatility to remain the same then the optimal allocation would be lower. Bernstein suggests (in his Rational expectations book) a 2% expected EM premium over US stocks, but a higher expected return for Non-US developed market equities than EM equities - suggesting a less than 2% premium on EM above developed market equities in aggregate.

      Some have argued that EM stocks don’t add much beyond developed market small value. And if the latter have an expected premium of 2% over a DM market portfolio, then the argument has some validity on a mean-variance basis. However, if EM value stocks are used instead of an EM market portfolio, and if EM value stocks have a similar 2% premium over developed market value stocks, then it still makes sense, on a mean-variance basis, to include EM stocks.

      This implies that adding an EM market allocation to a small cap and value tilted developed market portfolio may not add much to ‘expected’ mean-variance efficiency, while adding an EM value allocation would.
Just to note that an increasing EM share in the MSCI All Cap World index does not automatically imply this was derived from higher returns in EM. For example, the EM share of the MSCI All Cap World Index (the aggregate world equity market) increased from less than 1% in 1987 to 11% in 2007. We know from the earlier work of Bernstein and Arnott that share dilution (issuance of new shares) in developed markets has been about 2 percent annualized–so overall market cap (price x number of shares) has increased at about 9% percent year in developed markets. For EM to get to an 11% world market share in 2007 from 1% in 1987, the implied market cap return for EM equity over this period is 25.7%, however the actual annualized price return was 13.4% which implies a 12.3% per year share dilution!

I will stick with my 13% equity allocation to EM (no plan to change that), but seems prudent to capture some of my targeted portfolio value exposure in EM.

Robert
.

User avatar
in_reality
Posts: 4529
Joined: Fri Jul 12, 2013 6:13 am

Re: Emerging markets 1900-2013

Post by in_reality » Tue Aug 26, 2014 5:32 am

Interesting stuff ... thanks!

User avatar
SimpleGift
Posts: 3298
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: Emerging markets 1900-2013

Post by SimpleGift » Tue Aug 26, 2014 8:59 am

Thanks, Robert T, for posting the excellent summary.

It's interesting to see how relatively consistent the returns of emerging markets have been over the last century (with the exception of the world war periods!). Nearly all of the other historical returns series we've had just date back to about 1990 (when emerging market funds first appeared), so the outperformance of emerging markets in recent decades has seemed somewhat exceptional — but actually they've had strong returns in many previous periods as well.

Image
Source: Global Investment Returns Yearbook-2014

larryswedroe
Posts: 15941
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Emerging markets 1900-2013

Post by larryswedroe » Tue Aug 26, 2014 10:58 am

Robert
I've been making that point for a very long time, if going to take risk of EM might as well take it in EMS/EMV where premiums should logically be large (weaker financial systems should lead to higher premiums) and the correlations of EMS and EMV to EM are close to 1. That also btw allows you to either capture the higher expected returns OR (and most forget this option) to have lower overall equity exposure, including less to EM if you chose, so tail risks get cut

Larry

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Emerging markets 1900-2013

Post by Rodc » Tue Aug 26, 2014 11:31 am

Hi Robert, very interesting. Thanks.

I do think the writeup would be improved with a more explicit analysis of risks. For example sample CAGR is given without an associated risk metric (say standard deviation).
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

User avatar
Topic Author
Robert T
Posts: 2574
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: Emerging markets 1900-2013

Post by Robert T » Tue Aug 26, 2014 3:45 pm

Rodc wrote:I do think the writeup would be improved with a more explicit analysis of risks. For example sample CAGR is given without an associated risk metric (say standard deviation).
.

I agree, the yearbook did not provide the standard deviation numbers with the aggregate long-term returns, and their message on relative volatility overall is that it has come down significantly. They show a graph comparing relative volatility of EM and Developed markets, where pre-1980 EM volatility was almost double developed markets (1.9 times), but by 2013 it had declined to 1.1 times.

In my view the main risk story with EM equities, briefly touched on, is the large negative tail - this was clearly demonstrated in 1945-49 when EM collapsed. Standard devations doesn't capture this negative skewness. An earlier article by MSCI demonstrates this well - see figures on pg 36 in this link to the article http://www.factset.com/insight/article/ ... ersary.pdf

The implication for me is to hold US treasuries as fixed income, to off-set (inoculate) a bit of the negative skewness effect, as when EM does badly there is usually flight to quality and treasuries do relatively well. Here are the correlation coefficients of various fixed income holdings with EM equities, for the years of negative EM equities returns since 1990 (of which there were 11). Don't have annual data for the previous preiods.
  • +0.67 = HY bonds
    +0.06 = T-bills
    -0.03 = US Aggregate Bond Index
    -0.38 = 5 yr Treasury Notes
    -0.48 = Long-term Government Bonds
So treasuries seems to provide greatest protection, including when 'negative tail' risk shows up as in 1998. Preference is intermediate term (for combination of reasons).

Robert
.

berntson
Posts: 1366
Joined: Mon Oct 29, 2012 12:10 pm

Re: Emerging markets 1900-2013

Post by berntson » Tue Aug 26, 2014 3:50 pm

Robert: Thanks for the thoughtful and careful analysis. I too enjoyed the Credit Suisse article on emerging markets. Have you incorporated an emerging market value fund into your portfolio? I like the RAFI funds in general, and so I want to like PXH. The fund has had remarkably bad tracking error though. I gather that some of this is because the original index was not float-adjusted, a problem which has now been fixed. I would have more confidence that the fund was worth adding to my portfolio if there were a good way to run three-factor regressions for emerging market funds.

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Emerging markets 1900-2013

Post by Rodc » Tue Aug 26, 2014 3:53 pm

Robert T wrote:
Rodc wrote:I do think the writeup would be improved with a more explicit analysis of risks. For example sample CAGR is given without an associated risk metric (say standard deviation).
.

I agree, the yearbook did not provide the standard deviation numbers with the aggregate long-term returns, and their message on relative volatility overall is that it has come down significantly. They show a graph comparing relative volatility of EM and Developed markets, where pre-1980 EM volatility was almost double developed markets (1.9 times), but by 2013 it had declined to 1.1 times.

In my view the main risk story with EM equities, briefly touched on, is the large negative tail - this was clearly demonstrated in 1945-49 when EM collapsed. Standard devations doesn't capture this negative skewness. An earlier article by MSCI demonstrates this well - see figures on pg 36 in this link to the article http://www.factset.com/insight/article/ ... ersary.pdf

The implication for me is to hold US treasuries as fixed income, to off-set (inoculate) a bit of the negative skewness effect, as when EM does badly there is usually flight to quality and treasuries do relatively well. Here are the correlation coefficients of various fixed income holdings with EM equities, for the years of negative EM equities returns since 1990 (of which there were 11). Don't have annual data for the previous preiods.
  • +0.67 = HY bonds
    +0.06 = T-bills
    -0.03 = US Aggregate Bond Index
    -0.38 = 5 yr Treasury Notes
    -0.48 = Long-term Government Bonds
So treasuries seems to provide greatest protection, including when 'negative tail' risk shows up as in 1998. Preference is intermediate term (for combination of reasons).

Robert
.
Thanks. Appreciated.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

User avatar
matjen
Posts: 2087
Joined: Sun Nov 20, 2011 11:30 pm

Re: Emerging markets 1900-2013

Post by matjen » Tue Aug 26, 2014 4:39 pm

Wonderful and interesting post Robert. Thank you for sharing.
A man is rich in proportion to the number of things he can afford to let alone.

golfvestor
Posts: 80
Joined: Fri Aug 22, 2014 11:50 pm

Re: Emerging markets 1900-2013

Post by golfvestor » Tue Aug 26, 2014 4:44 pm

The OP's post is proof that people will read into the data whatever they want to see in the data, textbook confirmation bias. He presents data showing that developed stocks beat EM by 0.9% annualized, but then concludes there should be an EM premium of 2% annualized.

User avatar
matjen
Posts: 2087
Joined: Sun Nov 20, 2011 11:30 pm

Re: Emerging markets 1900-2013

Post by matjen » Tue Aug 26, 2014 4:48 pm

golfvestor wrote:The OP's post is proof that people will read into the data whatever they want to see in the data, textbook confirmation bias. He presents data showing that developed stocks beat EM by 0.9% annualized, but then concludes there should be an EM premium of 2% annualized.
Not since 1950...What do you think the odds of WWIII happening anytime soon are?
A man is rich in proportion to the number of things he can afford to let alone.

User avatar
nisiprius
Advisory Board
Posts: 38470
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Emerging markets 1900-2013

Post by nisiprius » Tue Aug 26, 2014 5:19 pm

The "Emerging markets revisited" chapter of the yearbook is fantasy football. How can there be statistics for "emerging markets" from 1900 when

a) the term "was originally brought into fashion in the 1980s by then World Bank economist Antoine Van Agtmael," https://en.wikipedia.org/wiki/Emerging_markets -- a fact noticed in the Credit Suisse yearbook itself, and

b) the first index, the MSCI Emerging Markets Index, was not created until 1998? Well, actually according to the Credit Suisse yearbook there was an earlier one, created in 1985. They say correctly that "the relative recency of these indexes is unhelpful for investors seeking a longer-term performance record." But it is not so clear that they can fill the gap from their "extensive" database of 23 countries.

One must at least suspect hindsight and survivorship bias, due to the possibility that if an emerging market fails to emerge, it never becomes important enough to make it into their "extensive long-run returns database." There are more than 23 countries with stock exchanges.

Where is Malaysia, whose Kuala Lumpur stock exchange was founded in 1930? Not in their database.
Where is Argentina, which has had a stock market since 1854? Not in their database.
Where is India, whose Bombay stock exchange was founded in 1875? Not in their database.

How sure are we that there has never been a time when these stock markets looked like "emerging markets?"

This sort of thing leads me to doubt the objectivity of Dimson & al. and wonder whether they are not pushing a point of view.

The best available data is not necessarily good data.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

golfvestor
Posts: 80
Joined: Fri Aug 22, 2014 11:50 pm

Re: Emerging markets 1900-2013

Post by golfvestor » Tue Aug 26, 2014 5:48 pm

matjen wrote:
golfvestor wrote:The OP's post is proof that people will read into the data whatever they want to see in the data, textbook confirmation bias. He presents data showing that developed stocks beat EM by 0.9% annualized, but then concludes there should be an EM premium of 2% annualized.
Not since 1950...What do you think the odds of WWIII happening anytime soon are?
EM way under-performed in the 40s, but it is not at all clear how much WWII was the cause. China probably would have gone communist with or without WWII. Some EM countries like Japan were obviously devastated, but so were many developed countries like Germany.

How likely do I think a war between two major EM countries, such as Pakistan and India is? Probably higher than you do. How familiar are you with Pakistani politics? Can you even name their prime minister without using Google?

User avatar
SimpleGift
Posts: 3298
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: Emerging markets 1900-2013

Post by SimpleGift » Tue Aug 26, 2014 6:16 pm

nisiprius wrote:This sort of thing leads me to doubt the objectivity of Dimson & al. and wonder whether they are not pushing a point of view. The best available data is not necessarily good data.
There's no doubt that it can be extremely difficult to reconstruct global stock market returns going back 100 years. For example, in the classic 1999 paper, Global Stock Markets in the Twentieth Century, the researchers were only able to find reliable market data for many countries going back to 1920 — and even these data series had long gaps and other problems:
Jorian and Goetzmann wrote:We have permanent gaps in the series for Chile, Germany, Japan, Peru, Portugal, and Argentina. The gap for Chile is filled using data from publications from the Chilean Central Bank. The gap for Germany is covered using data spliced by Gielen (1994). The gap for Japan is bridged using Bank of Japan (1966) data. The gap for Peru is filled using data received by the Lima stock exchange. To cover the gap for Portugal, we use information from the Portuguese Central Bank. Overall, Argentina is the only remaining country with a permanent break over July 1965 to December 1975, which is the first date for which we have data from the IFC. We have been unable to find data to bridge the gap.
However, this doesn't mean that researchers should just give up and not try to piece together the data that's available to them and then draw reasonable inferences from it. I do feel that Dimson, Marsh and Staunton could have been more transparent about the data gaps they encountered and the data choices they made in reconstructing their emerging markets time series — but it seems too far to ascribe any hidden agenda to their efforts.

User avatar
Topic Author
Robert T
Posts: 2574
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: Emerging markets 1900-2013

Post by Robert T » Tue Aug 26, 2014 7:40 pm

golfvestor wrote:The OP's post is proof that people will read into the data whatever they want to see in the data, textbook confirmation bias. He presents data showing that developed stocks beat EM by 0.9% annualized, but then concludes there should be an EM premium of 2% annualized.

Did not say 'should', your word.
.

User avatar
in_reality
Posts: 4529
Joined: Fri Jul 12, 2013 6:13 am

Re: Emerging markets 1900-2013

Post by in_reality » Tue Aug 26, 2014 9:00 pm

golfvestor wrote: How likely do I think a war between two major EM countries, such as Pakistan and India is? Probably higher than you do. How familiar are you with Pakistani politics? Can you even name their prime minister without using Google?
Ok so that puts me at a 0.7% risk of loss (2.11% of foreign holdings in India X 35% foreign holdings).

More threatening is China - Vietnam, China - Taiwan etc as China is a much bigger holding per market cap.

Which brings out my concern of using only market cap in emerging markets. Market cap will tilt towards having a lot of large state owned firms (China, Russia, Brazil) that may not really respect shareholders. Also, large caps are more export oriented firms and not as much the local economy.
berntson wrote: I like the RAFI funds in general, and so I want to like PXH. The fund has had remarkably bad tracking error though. I gather that some of this is because the original index was not float-adjusted, a problem which has now been fixed. I would have more confidence that the fund was worth adding to my portfolio if there were a good way to run three-factor regressions for emerging market funds.
Wasn't PXH using ADR's (which don't track the index so faithfully) in their early days to reduce transaction costs?

Anyway, I blend DGS (0.63%ER small mid cap), DVYE (0.49% ER all caps) and possibly DEM (0.63% ER large cap). They are all 55-60% value. PXH (0.49% ER) is about 46% value and quite heavily large cap so more into the state owned firms (which may or may not be bad -- I can offer no real educated opinion other to say it's not my preference). I also mix in FRN 0.70% ER which is large and mid cap and only value tilted for the mid cap but is only 6% Asia compared to all those other funds 55-70%. These are in tax sheltered accounts. Maybe tilting to value by selecting dividend funds isn't the best though. Am also 20% VEMAX (Vanguard Emerging Markets Admiral)

Overall my exposure to China is much lower that market cap would dictate, but I figure many of the countries I am in are going to be strongly correlated to the performance of the Chinese economy and given how brazen China is being about fines for "anti-trust" violations, I really don't expect state owned firms to offer exactly real returns. I know I can't know that though, and I am looking at it wrong but ...
Last edited by in_reality on Tue Aug 26, 2014 9:30 pm, edited 1 time in total.

User avatar
Epsilon Delta
Posts: 8090
Joined: Thu Apr 28, 2011 7:00 pm

Re: Emerging markets 1900-2013

Post by Epsilon Delta » Tue Aug 26, 2014 9:27 pm

nisiprius wrote: Where is Argentina, which has had a stock market since 1854? Not in their database.

How sure are we that there has never been a time when these stock markets looked like "emerging markets?"
In 1900 Argentina looked more like a developed country, where it would have taken quite a few basis points off the developed market returns until it was demoted sometime after 1940 when it would have taken a few more basis points off the emerging markets returns.

It's not entirely clear that Japan was an emerging market in 1940. If we attribute that 98% drop to developed markets things look a bit different.

SnowSkier
Posts: 265
Joined: Fri Mar 16, 2012 12:21 pm
Location: Austin

Re: Emerging markets 1900-2013

Post by SnowSkier » Tue Aug 26, 2014 11:28 pm

Robert T wrote:I will stick with my 13% equity allocation to EM (no plan to change that), but seems prudent to capture some of my targeted portfolio value exposure in EM.
Robert, Thanks!...very interesting

Robert/All, Thoughts on how to tilt to Value in Emerging Markets?

RAFI/Fundamental:
- PXH? (PowerShares FTSE RAFI Emerging, expense 0.49%, assets 400M)
- FNDE? (Schwab Fundamental Emerging Large, expense 0.47%, assets 43M), which should match SFENX (Schwab Fundamental Emerging Large, expense 0.51%, assets 380M)
Other?

User avatar
matjen
Posts: 2087
Joined: Sun Nov 20, 2011 11:30 pm

Re: Emerging markets 1900-2013

Post by matjen » Wed Aug 27, 2014 7:26 am

For those with DFA access, DFA World ex US Targeted Val Instl (DWUSX) makes for a nice one stop shop for International and emerging markets SCV.
A man is rich in proportion to the number of things he can afford to let alone.

Valuethinker
Posts: 38176
Joined: Fri May 11, 2007 11:07 am

Re: Emerging markets 1900-2013

Post by Valuethinker » Wed Aug 27, 2014 7:47 am

golfvestor wrote:
EM way under-performed in the 40s, but it is not at all clear how much WWII was the cause. China probably would have gone communist with or without WWII. Some EM countries like Japan were obviously devastated, but so were many developed countries like Germany.
Actually recent scholarship is a lot kinder to Chiang Kai Shek. Had WW2 not happened, he might have held in there. The communists were a minor force until relatively late in the day.

http://www.amazon.com/Chinas-War-Japan- ... B00AZRDP32

Remember the PRC had the Great Leap Forward and the Cultural Revolution, each of which probably set its economic development back by a decade (besides enormous human cost). Now they have evolved into something which looks similar to the KMT's Taiwan (a dictatorship with capitalism) or South Korea at least until democracy set in in the last couple of decades.

As I understand it, it is still very impolite to mention the Great Leap Forward in China. Something like 40m people are estimated to have starved to death.

Maybe Japan was an 'Emerging Market' pre WW2, but it was a major industrialized country and military power-- signatory to the Washingon Naval Treaty (the strategic arms limitation talks of the interwar years). I would rank it if not up there with the first tier (France Germany Britain USA) then with southern tier countries like Italy and Spain, certainly advanced than Poland, perhaps even the USSR itself (both highly industrialized and quite backward).
How likely do I think a war between two major EM countries, such as Pakistan and India is? Probably higher than you do. How familiar are you with Pakistani politics? Can you even name their prime minister without using Google?
We are not likely to know if that is going to happen before stock markets do. What is important though, which I think you are hinting at, is the very substantial event risk in the Indian subcontinent. Another attack like the Parliament or the Mumbai raids, at the wrong political moment, and there could be reprisals (military) which would lead to war. In some ways, India has shown remarkable restraint: perhaps because both sides have nuclear weapons.

User avatar
matjen
Posts: 2087
Joined: Sun Nov 20, 2011 11:30 pm

Re: Emerging markets 1900-2013

Post by matjen » Wed Aug 27, 2014 9:10 am

Valuethinker wrote:We are not likely to know if that is going to happen before stock markets do. What is important though, which I think you are hinting at, is the very substantial event risk in the Indian subcontinent. Another attack like the Parliament or the Mumbai raids, at the wrong political moment, and there could be reprisals (military) which would lead to war. In some ways, India has shown remarkable restraint: perhaps because both sides have nuclear weapons.
It just seems to me that there is always "something" going on and there always will be. Perhaps Pakistan is unstable compared to decades past...let's just be grateful that nothing untoward is going on in the Ukraine /Russia or in Israel/Gaza. ;-)
A man is rich in proportion to the number of things he can afford to let alone.

Deep Thoughts
Posts: 102
Joined: Tue May 13, 2014 1:27 pm

Re: Emerging markets 1900-2013

Post by Deep Thoughts » Wed Aug 27, 2014 9:38 am

matjen wrote:For those with DFA access, DFA World ex US Targeted Val Instl (DWUSX) makes for a nice one stop shop for International and emerging markets SCV.
It's ER is .79. The Morningstar Boxes indicate that its more MidCap Value than SV. I wish we had some good factor analysis for this fund (or any of the DFA international funds).

I'm thinking that DEMSX (EM Small) is pretty valuey (ER .7) Combine that in market weight with DISVX (International SV) (ER .7) and you probably load small more
and maybe the same in value.

User avatar
matjen
Posts: 2087
Joined: Sun Nov 20, 2011 11:30 pm

Re: Emerging markets 1900-2013

Post by matjen » Wed Aug 27, 2014 10:03 am

Deep Thoughts wrote:
matjen wrote:For those with DFA access, DFA World ex US Targeted Val Instl (DWUSX) makes for a nice one stop shop for International and emerging markets SCV.
It's ER is .79. The Morningstar Boxes indicate that its more MidCap Value than SV. I wish we had some good factor analysis for this fund (or any of the DFA international funds).

I'm thinking that DEMSX (EM Small) is pretty valuey (ER .7) Combine that in market weight with DISVX (International SV) (ER .7) and you probably load small more
and maybe the same in value.
Well that is why I said "one stop shop" which probably didn't really clarify what I was trying to say. It is basically a heavily tilted fund toward SCV. Much more than their US Core and Vector it seems to me. If you already have only VXUS or VEU than perhaps a pure SCV fund would be what you are looking for.
A man is rich in proportion to the number of things he can afford to let alone.

denovo
Posts: 4405
Joined: Sun Oct 13, 2013 1:04 pm

Re: Emerging markets 1900-2013

Post by denovo » Wed Aug 27, 2014 10:09 am

Just wondering, how active or liquid was the Chinese Stock Market in 1900, or the Russian Stock Market. Are we confident about the prices we have? Were they even open to foreigners?
"Don't trust everything you read on the Internet"- Abraham Lincoln

User avatar
SimpleGift
Posts: 3298
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: Emerging markets 1900-2013

Post by SimpleGift » Wed Aug 27, 2014 12:08 pm

denovo wrote:Just wondering, how active or liquid was the Chinese Stock Market in 1900, or the Russian Stock Market. Are we confident about the prices we have? Were they even open to foreigners?
Good questions. It appears Dimson, Marsh and Staunton used only seven countries to construct their emerging markets series from 1900 to 1955 — including China and Russia, along with Finland, Japan, Portugal, South Africa and Spain. Then they kept adding countries as more data became available, such as India and Brazil in 1955, then Korea and Hong Kong in 1963, and so on, up to the inception of the MSCI Emerging Markets Index in 1988.

Unfortunately, they don't go into detail about how they constructed their stock market histories for countries such as China and Russia in the early 1900s. This is what they have to say about their methodology:
Dimson, Marsh and Staunton wrote:In the 2010 Yearbook, we pointed out that, despite the complexity of index compilers’ procedures, there was a simple rule that replicated their decisions very accurately. The rule was to categorize countries as developed if they had GDP per capita in excess of USD 25,000.

For the 23 countries in our database in 1900, seven would have been deemed emerging markets: China, Finland, Japan, Portugal, Russia, South Africa, and Spain. Three are still emerging today – 114 years later – namely, China, Russia, and South Africa.

Using the GDP per capita rule, we construct a long-run emerging markets index starting in 1900 initially with seven countries. Rather than restricting this to the emerging countries in the DMS database, we add in further markets once returns data becomes available. Thus, in 1955, we add Brazil and India; in 1963, Korea and Hong Kong (until the latter moved to developed in 1977); in 1966, Singapore (until it moved to developed in 1980); in 1970, Malaysia; in 1976, Argentina, Chile, Greece, Mexico, Thailand, and Zimbabwe; in 1978, Jordan; in 1985, Colombia, Pakistan, Philippines and Taiwan; and in 1987, Turkey. We then link into the MSCI Emerging Markets index from its inception in 1988.

As a comparator, we create a developed markets index, using the same GDP per capita rule. This had 16 constituents in 1900, and was joined by Finland in 1932 and Japan in 1967. We then link into the MSCI Developed World Index from its start date in 1970. Our indexes are computed in US dollars, and include reinvested dividends.

garlandwhizzer
Posts: 2323
Joined: Fri Aug 06, 2010 3:42 pm

Re: Emerging markets 1900-2013

Post by garlandwhizzer » Wed Aug 27, 2014 12:54 pm

Very interesting post, Robert T. The real question is will the future be a repeat of the past, rhyme with it, or be something entirely different? That in my mind at least is unknown at present.

There is now a conjunction of factors in essentially all DMs that have never before existed in history. First of all very high debt levels relative to GDP. Economic history suggests that when debt exceeds 90% of GDP economic growth tends to decline. Essentially all DMs are there now or beyond that point. Second, a demographic time bomb in approaching in all DMs with aging populations, long life expectancies beyond economically productive years, and retirement promises already made to the elderly without adequate funds to fulfill these promises. The only way out is more debt or more savings, both of which reduce final demand and further decrease expected economic growth. Currently there is zero to slow economic growth in all DMs that seems resistant to aggressive central bank policies a fact which suggests that these factors may be active at present and could increase as time goes by.

Japan has been struggling with deflation for 25 years and both stocks are real estate are considerably cheaper now than in 1990. Deflation has a tendency to persist for long periods of time. Why buy anything now when it's going to be cheaper before long. That tends to exert persistent down-pressure on final demand which in turn keeps driving deflation. Central banks can break the back of inflation by raising interest rates but they cannot effectively lower interest rates below zero and tend to struggle long term to overcome deflation as they have in Japan for 25 years.

Europe's economic growth is essentially non-existant for 5 years and actually appears to be declining in Germany, France, and Italy at present. German ten year bonds currently yield less than 1%, a fact that recalls Japan's incredibly low interest rates and reminds us that there is real threat of deflation in the Eurozone. I don't know whether this will happen or not but it is at some level of risk.

On the other hand EMs have younger populations, considerably less debt, no demographic time bomb, and more rapid economic growth. That does not mean that EMs can't screw it up. They can, often do, and they are clearly riskier. But it does bring into question the validity of comparing past performance over 50 - 100 years of EM versus DM and projecting that into the future as if it is a fait accompli.

Robert T quoted Bernstein's positive outlook on DMs relative to EMs going forward, but it's important to realize that both Grantham and Asness predict EM outperformance relative to both US and DM going forward 7 - 10 years. Who is right in this? I don't know and I'm splitting my international allocation 50/50 between EM and DM to cover both bases.

Garland Whizzer

golfvestor
Posts: 80
Joined: Fri Aug 22, 2014 11:50 pm

Re: Emerging markets 1900-2013

Post by golfvestor » Wed Aug 27, 2014 6:09 pm

Simplegift wrote: Using the GDP per capita rule, we construct a long-run emerging markets index starting in 1900 initially with seven countries. Rather than restricting this to the emerging countries in the DMS database, we add in further markets once returns data becomes available. Thus, in 1955, we add Brazil and India; in 1963, Korea and Hong Kong (until the latter moved to developed in 1977); in 1966, Singapore (until it moved to developed in 1980); in 1970, Malaysia; in 1976, Argentina, Chile, Greece, Mexico, Thailand, and Zimbabwe; in 1978, Jordan; in 1985, Colombia, Pakistan, Philippines and Taiwan; and in 1987, Turkey. We then link into the MSCI Emerging Markets index from its inception in 1988.
This definitely makes it look a lot like nisiprius's description of fantasy football. Mexico had a stock exchange since 1886, but only got added to the data in 1976. Taiwan had a stock market since 1962, but only got added in 1985. The Philippines had a stock exchange since 1927, but only got included in 1985. Maybe, we only have three decades of good data on emerging market returns.

User avatar
SimpleGift
Posts: 3298
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: Emerging markets 1900-2013

Post by SimpleGift » Wed Aug 27, 2014 7:20 pm

As noted upthread, just because a country had a stock exchange in 1886, it does not mean that a continuous and useful record of its stock returns over the 128 year period is available to researchers. For various reasons (wars, revolutions, floods, fires, etc.), many stock exchanges in even developed countries do not have an unbroken and entirely useful record of their stock market returns.

If this particular emerging market data series was put together by the equity department at Goldman Sachs (originator and promoter of the BRICs concept), then I'd share the skepticism. And I'd certainly like to know much more about how Dimson, Marsh and Staunton pieced together their historical returns data for the early 1900s. But these fellows are the foremost researchers in the world today when it comes to historical stock market series — so I think we need to cut them a little slack. If there was more useful data sitting out there, I expect they'd be including it in their database and analysis.

User avatar
Topic Author
Robert T
Posts: 2574
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: Emerging markets 1900-2013

Post by Robert T » Wed Aug 27, 2014 8:51 pm

berntson wrote:I like the RAFI funds in general, and so I want to like PXH. The fund has had remarkably bad tracking error though. I gather that some of this is because the original index was not float-adjusted, a problem which has now been fixed. I would have more confidence that the fund was worth adding to my portfolio if there were a good way to run three-factor regressions for emerging market funds.
Berntson,

All emerging market index funds seem to have larger tracking error than developed market index funds – it seems to be the nature of EMs (higher trade costs, higher sampling error risk etc). Here's a quick comparison of tracking error from 2008-2013, where tracking error is measured as the difference in annualized returns of the fund relative to the underlying index/benchmark over this period.

-1.14% = Powershares FTSE RAFI emerging market (PXH)
-1.13% = Wisdomtree Emerging Market Small Cap Dividend (DGS)
-1.20% = Vanguard Emerging Market (VWO)

And if we compare the DFA EM Value fund with the EM Value series in the DFA Matrix book for the same time period, there’s a -2.23% tracking error. So over the last 6 years the annualized tracking error of PXH has been no worse than Wisdomtree, Vanguard or DFA. The big difference is the individual year tracking error, which has been largest for PXH particularly in 2009 when it was huge. Over the last 3 years there has been a much close match of PHX with its benchmark. Not sure if this is because RAFI have sorted out some of the earlier issues, or because EM equities have been less volatile. Perhaps a bit of both.

On factor analysis of the EM funds, you are right, there is less accessible data to do this for EM than for developed markets. I think the closest we have is Jason Hsu’s database which includes EM mkt, SmB, and HmL from 07/1995 to 06/2011, as well as a fundamental EM index from 01/1996 to 06/2011.

Using Hsu’s dataset here are the factor loads on PXH for the overlapping years 10/2007 to 06/2011

alpha = 0.03
mkt = 0.98
size = -0.19
value = 0.39

So larger cap (and some consistency with M* that shows average market cap to be larger than VWO: 28b vs 20bn). The value load is similar to the value loads on the RAFI funds in developed markets. So results seem plausible. I was surprised to see a zero alpha given the tracking error, but then I check the factor loads on the RAFI EM index and it had a positive alpha over this time period. Interestingly, while the value load on PXH was similar the RAFI EM index, the size load was more negative -0.19 vs. -0.09, suggesting the tracking error/style drift was to larger cap. While the EM size and value premium from Hsu’s data were similar for 10/2007 to 06/2011, for the full period 07/1995 to 6/2011 the EM size premium was negative and the value premium was huge.

But, overall, I have some question marks over the quality of Hsu’s data - if I do a factor analysis on VWO it also shows a large cap and value tilt, which seems a little weird. A reasonable assumption IMO for PXH is that it will have a similar value load to the RAFI developed market (large cap funds), but perhaps a larger size load. The fund comprises the top 350 companies with the largest RAFI fundamental values of the large, mid and small company stocks from the FTSE EM index which seems to imply are larger cap fund than if the full universe is used.

I would prefer to see more stocks in the index, as I think sampling error risk in EM is fairly high. The tilted ETF EM index with largest number of stocks I could find is the Flexshares Morningstar Emerging Market factor tilt (TLTE) which has over 2000 stocks, but still a similar 1% tracking errors since inception (although the fund is fairly new) – seems to have about a core 2 tilt.

My preference on EM ETF is PXH (warts and all), DGS also good, but tax-cost will likely be higher given the dividend sorts. TLTE is interesting, but not enough of track record yet, and no availability of back-tested data to see resulting characteristics.

Just a note on funds like DFA World-ex US (targeted) value - a reason for spending some time on EM analysis was to decide on whether to let my EM allocation increase with EM market cap share (as now the EM cap weight share in MSCI ACWI has caught up with my long-term EM fixed allocation holding). My conclusion is to retain fixed weights as long-term target and not let EM allocation increase inline with its share in MSCI ACWI. Main reasons (i) the apparent increase in correlations and lower SDs don't seem to justify, on a mean-variance' efficiency basis, an increased EM allocation, and (ii) main risk in EM, IMO, is negative tail risk - when thing go badly, they can go really badly, and don't want to be giving a higher allocation to this negative tail risk.

Just my take.

Robert
.

golfvestor
Posts: 80
Joined: Fri Aug 22, 2014 11:50 pm

Re: Emerging markets 1900-2013

Post by golfvestor » Wed Aug 27, 2014 10:26 pm

Simplegift wrote:If there was more useful data sitting out there, I expect they'd be including it in their database and analysis.
Even assuming some set represents the best available data doesn't mean it's high quality data or that it can tell us with any reasonable level of statistical confidence whether there was an EM premium from 1900-1985 or what mix of DM/EM has the highest mean-variance efficiency.

User avatar
gordoni2
Posts: 197
Joined: Wed Aug 15, 2007 12:20 am
Contact:

Re: Emerging markets 1900-2013

Post by gordoni2 » Wed Aug 27, 2014 10:55 pm

Very informative.

Even for 50-100 year periods the standard error of the mean will be 2-3% (assuming an annual standard deviation of around 20%). Thus any difference between the two could simply be the result of sampling noise.

golfvestor
Posts: 80
Joined: Fri Aug 22, 2014 11:50 pm

Re: Emerging markets 1900-2013

Post by golfvestor » Wed Aug 27, 2014 11:40 pm

gordoni2 wrote:Very informative.

Even for 50-100 year periods the standard error of the mean will be 2-3%
(assuming an annual standard deviation of around 20%). Thus any difference between the two could simply be the result of sampling noise.
This is really stunning when you think about it's implications for factor tilts. That means with 100 years of data, any out-performance of less than 4% annually could be explained by chance using a 95% confidence level.

edge
Posts: 3388
Joined: Mon Feb 19, 2007 7:44 pm
Location: NY

Re: Emerging markets 1900-2013

Post by edge » Thu Aug 28, 2014 12:04 am

matjen wrote:
Deep Thoughts wrote:
matjen wrote:For those with DFA access, DFA World ex US Targeted Val Instl (DWUSX) makes for a nice one stop shop for International and emerging markets SCV.
It's ER is .79. The Morningstar Boxes indicate that its more MidCap Value than SV. I wish we had some good factor analysis for this fund (or any of the DFA international funds).

I'm thinking that DEMSX (EM Small) is pretty valuey (ER .7) Combine that in market weight with DISVX (International SV) (ER .7) and you probably load small more
and maybe the same in value.
Well that is why I said "one stop shop" which probably didn't really clarify what I was trying to say. It is basically a heavily tilted fund toward SCV. Much more than their US Core and Vector it seems to me. If you already have only VXUS or VEU than perhaps a pure SCV fund would be what you are looking for.
I use DFEVX and DFVQX.

SnowSkier
Posts: 265
Joined: Fri Mar 16, 2012 12:21 pm
Location: Austin

Re: Emerging markets 1900-2013

Post by SnowSkier » Thu Aug 28, 2014 12:24 am

Robert T wrote:
berntson wrote:I like the RAFI funds in general, and so I want to like PXH...
...
My preference on EM ETF is PXH (warts and all), DGS also good, but tax-cost will likely be higher given the dividend sorts. TLTE is interesting, but not enough of track record yet, and no availability of back-tested data to see resulting characteristics.
...
Just my take.
Great info above on PXH and its index (FTSE RAFI Emerging). Thanks!

Any thoughts on the Russell Fundamental Emerging Markets Large Co Index and the funds that use it SFENX and FNDE?

FNDE is smaller than desired (43M in assets), but if SFENX grew to 380M in assets, surely the almost-identical FNDE will grow as well.

denovo
Posts: 4405
Joined: Sun Oct 13, 2013 1:04 pm

Re: Emerging markets 1900-2013

Post by denovo » Thu Aug 28, 2014 12:26 am

garlandwhizzer wrote:
First of all very high debt levels relative to GDP. Economic history suggests that when debt exceeds 90% of GDP economic growth tends to decline.
No, that study was invalidated.

https://en.wikipedia.org/wiki/Growth_in_a_Time_of_Debt

The authors themselves conceded methodological flaws in their study, to whit
"Don't trust everything you read on the Internet"- Abraham Lincoln

denovo
Posts: 4405
Joined: Sun Oct 13, 2013 1:04 pm

Re: Emerging markets 1900-2013

Post by denovo » Thu Aug 28, 2014 12:28 am

Simplegift wrote:
denovo wrote:Just wondering, how active or liquid was the Chinese Stock Market in 1900, or the Russian Stock Market. Are we confident about the prices we have? Were they even open to foreigners?
Good questions. It appears Dimson, Marsh and Staunton used only seven countries to construct their emerging markets series from 1900 to 1955 — including China and Russia, along with Finland, Japan, Portugal, South Africa and Spain. Then they kept adding countries as more data became available, such as India and Brazil in 1955, then Korea and Hong Kong in 1963, and so on, up to the inception of the MSCI Emerging Markets Index in 1988.

Unfortunately, they don't go into detail about how they constructed their stock market histories for countries such as China and Russia in the early 1900s. This is what they have to say about their methodology:
Yes, I don't know if we have any economic historians here, but my concern is that the data isn't so reliable from that time for obvious reasons.
"Don't trust everything you read on the Internet"- Abraham Lincoln

oragne lovre
Posts: 484
Joined: Tue Nov 25, 2008 3:34 pm

Re: Emerging markets 1900-2013

Post by oragne lovre » Thu Aug 28, 2014 12:54 am

OP started an interesting post that has triggered interesting comments.
I, however, have to admit that I don't fully understand how the OP has reached such a specific recommendation of 13% allocation to EM.
It's nevertheless an excellent thread for me to learn more about the history of emerging markets.
The finest, albeit the most difficult, of all human achievements is being reasonable.

User avatar
SimpleGift
Posts: 3298
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: Emerging markets 1900-2013

Post by SimpleGift » Thu Aug 28, 2014 12:57 am

denovo wrote:Yes, I don't know if we have any economic historians here, but my concern is that the data isn't so reliable from that time for obvious reasons.
After a little digging, I was able to find more information about how Dimson, Marsh and Staunton addressed survivorship bias in their stock markets data series, plus how they added Russia and China to their database (despite these markets disappearing in 1917 and 1949, respectively). It's good reading for those interested in the history of financial markets. See pages 9-11 in their Global Investment Returns Yearbook, 2013.
Dimson, Marsh and Staunton wrote:At our base date of 1900, stock exchanges existed in 33 of today’s nations. Until this year, our database contained 19 countries, accounting for some 87% of world market capitalization at end-1899. But, despite this extensive coverage, it is still possible that we are overstating worldwide equity returns by omitting countries that performed poorly or failed to survive.

The two largest missing markets were Austria- Hungary and Russia, which, at end-1899, accounted for 5% and 6% of world market capitalization, respectively. The best-known cases of markets that failed to survive were Russia and China. We have now added these countries to our database. With Austria, we now have 20 countries with continuous histories from 1900 to the present day. Russia and China have discontinuous histories, but we are still able to fully include them in our revised world index.
Note that, by adding these countries, the DMS database accounts for over 98% of global market capitalization in 1900. The 11 stock markets not included in their 1900 database presumably had poor data, or were too small to matter, or both.
Last edited by SimpleGift on Thu Aug 28, 2014 2:41 am, edited 2 times in total.

User avatar
in_reality
Posts: 4529
Joined: Fri Jul 12, 2013 6:13 am

Re: Emerging markets 1900-2013

Post by in_reality » Thu Aug 28, 2014 1:07 am

oragne lovre wrote: I don't fully understand how the OP has reached such a specific recommendation of 13% allocation to EM.
Isn't that simply the market cap allocation? Emerging markets make up 13% of ex-US stocks.

Not exactly sure though, VXUS (Total International Stock) has it at 14.4%.
VT (Vanguard Total World) has it at 7.4% for the world which comes out to be the same 14.4% of international stocks are EM.

User avatar
Topic Author
Robert T
Posts: 2574
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: Emerging markets 1900-2013

Post by Robert T » Thu Aug 28, 2014 3:30 am

SnowSkier wrote:Any thoughts on the Russell Fundamental Emerging Markets Large Co Index and the funds that use it SFENX and FNDE
On the Schwab Emerging Markets Large Company Index (SFENX), since inception 01/31/2008 to end June 2014 (similar period to earlier comparisons), tracking error over this period has been -1.86%, compared to -1.14% for PXH. In addition, SFENX (and the recent ETF version FNDE) has less stock holdings at 254 – so prefer PXH.
oragne lovre wrote:I, however, have to admit that I don't fully understand how the OP has reached such a specific recommendation of 13% allocation to EM.
There’s nothing special about 13%. I was not presenting it as a recommendation, but just stating my current allocation. This is the long-term allocation target I had for EM when I set up my portfolio allocation over 11 years ago – it was essentially just over double the EM share in world equity markets at the time (as I recall). Added for portfolio diversification. Currently the EM share of world equity markets has now increased to 13% (from the Credit Suisse Yearbook – “Today, they [EM] comprise 13% of the free float investible universe of world equities…” which is also consistent (close to) the number at the end of the DFA Matrix book of 12% http://www.robasciotti.com/docs_informa ... 202014.pdf , so I had the question on whether, from now on, I should increase my portfolio allocation share to EM in-line with the world share or perhaps even increase the allocation to my initial 2x market weight, I think not to both.

Robert
.

SnowSkier
Posts: 265
Joined: Fri Mar 16, 2012 12:21 pm
Location: Austin

Re: Emerging markets 1900-2013

Post by SnowSkier » Thu Aug 28, 2014 5:51 pm

Robert T wrote:
SnowSkier wrote:Any thoughts on the Russell Fundamental Emerging Markets Large Co Index and the funds that use it SFENX and FNDE
On the Schwab Emerging Markets Large Company Index (SFENX), since inception 01/31/2008 to end June 2014 (similar period to earlier comparisons), tracking error over this period has been -1.86%, compared to -1.14% for PXH. In addition, SFENX (and the recent ETF version FNDE) has less stock holdings at 254 – so prefer PXH.
Thanks Robert!

berntson
Posts: 1366
Joined: Mon Oct 29, 2012 12:10 pm

Re: Emerging markets 1900-2013

Post by berntson » Fri Aug 29, 2014 12:55 pm

Robert,

Thank you for the exceptionally helpful post. It's good to see how similar PXH's average tracking error has been to other funds. I've been using IEMG for emerging markets. But if it looks like PowerShares is doing a reasonable job managing their fund, perhaps I should consider using PXH for new contributions. I would like to be able to get a more uniform value tilt across domestic, developed, and emerging markets. The tilt towards large caps could be counterbalanced by holding more VSS.

I suppose one question is whether the tilt is strong enough to justify the extra expense of the fund. I typically use a 2% value premium and 1% small premium when estimating expected returns. With a .4 HML and -.2 SMB loading, that leaves PXH with 60bp of additional expected returns. But the fund also costs 35bp more than VWO, so really we only have 25bp of additional expected return. That doesn't leave much margin for error. On the other hand, the premiums I'm using are small by historical standards and perhaps too conservative. Using higher expected premiums obviously make the fund look better. Also, it could be that the premium for emerging markets can be expected to be larger than for developed markets. Arnott makes the point in his book that backtested fundamental indexes do much better in emerging markets than developed markets. But I'm also a bit skeptical of his data.

User avatar
Topic Author
Robert T
Posts: 2574
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: Emerging markets 1900-2013

Post by Robert T » Sat Aug 30, 2014 9:16 am

berntson wrote:Robert,

Thank you for the exceptionally helpful post. It's good to see how similar PXH's average tracking error has been to other funds. I've been using IEMG for emerging markets. But if it looks like PowerShares is doing a reasonable job managing their fund, perhaps I should consider using PXH for new contributions. I would like to be able to get a more uniform value tilt across domestic, developed, and emerging markets. The tilt towards large caps could be counterbalanced by holding more VSS.

I suppose one question is whether the tilt is strong enough to justify the extra expense of the fund. I typically use a 2% value premium and 1% small premium when estimating expected returns. With a .4 HML and -.2 SMB loading, that leaves PXH with 60bp of additional expected returns. But the fund also costs 35bp more than VWO, so really we only have 25bp of additional expected return. That doesn't leave much margin for error. On the other hand, the premiums I'm using are small by historical standards and perhaps too conservative. Using higher expected premiums obviously make the fund look better. Also, it could be that the premium for emerging markets can be expected to be larger than for developed markets. Arnott makes the point in his book that backtested fundamental indexes do much better in emerging markets than developed markets. But I'm also a bit skeptical of his data.
Bernston,

On comparators: I too am a bit skeptical of the emerging market data Arnott presents in his book and that Jason Hsu has on his website – both show fairly dramatic outperformance of the fundamental index series. For example – for the time period that Jason Hsu has available data for the emerging market fundamental series.

1996-2010 annualized returns (%)/standard deviation

..8.2/36.4 = MSCI Emerging Markets Large Value index
10.7/43.2 = MSCI Emerging Markets Small Value index
11.5/41.5 = DFA Emerging Markets Value
11.7/38.2 = MSCI Emerging Markets Value weighted index
18.9/36.9 = RAFI Emerging Markets index

I think the numbers from FTSE are more credible. For example:

10 year annualized returns to end June 2013

12.1 = MSCI Emerging Markets Large Value index
12.6 = MSCI Emerging Markets Value Index (tracked by EVAL)
15.1 = MSCI Emerging Markets Small Value index
13.1 = MSCI Emerging Markets Value weighted index
13.5 = DFA Emerging Markets Value
14.6 = FTSE RAFI Emerging Markets index

Just to note that the last 10 years was a period when smallcaps outperformed large, and that the RAFI series has a negative size load. Since inception 2007, PXH average market cap has been larger than VWO in all years except one, when it was slightly smaller. So, from the above, perhaps returns of PXH will be more in-line with returns of MSCI EM value weighted index and DFA EM value (core), rather than providing the magnitude of outperformance suggested by Arnott and Hsu. Time will tell.

On ‘expected size and value premiums’ in emerging markets – there seems more consensus in the literature on the value premium than the size premium. E.g. two similarly titled papers http://papers.ssrn.com/sol3/papers.cfm? ... id=2351355 , http://papers.ssrn.com/sol3/papers.cfm? ... id=2070832 seem to come to similar conclusions (replicable results) that the value premium in emerging markets has been significant and present in both large and small cap stocks, and that the size premium has ‘weaker statistical significant’ than the value premium.

If we look at the returns of the MSCI emerging markets large cap index and MSCI emerging market small cap index for the full period for which data are available 1995-2013 the annualized return difference was 0.1%, so an extremely small premium for this time period, although it is larger for the more recent period, and we know from US data that there can be long periods (17yrs in the case of US – when large caps can outperform small caps).

Fama-French, in their ‘international evidence’ paper had to use the IFC series to look at the emerging market small cap premium, as …”Unlike the MSCI data, the IFC data covers small stocks…”. So perhaps the reason for the ‘weaker’ small cap premium in the MSCI data is that it excludes many EM smallcap stocks. Interestingly, the best performing DFA EM fund since inception in 1998 is their small cap stock fund whose returns were 5% higher than the market. The MSCI data only show a 2.2% premium of small caps over large since 1998. So perhaps DFA uses a larger universe of EM stocks to select its small cap stocks. A challenge in EMs seems to be the treatment of IPOs which is the source of much of the 12.5% share dilution (earlier estimate), many of these are small caps (higher share than in developed countries), and perhaps this is the reason retail indexes seem to take a wider birth (as per Fama-French assessment). This earlier paper on China highlights some of these issues - http://people.stern.nyu.edu/jmei/b40/ChinaIndexCom.pdf
So following this I am not so sure on the EM size premium, and would not be surprised if it is fairly small in retail indexes going forward. While the average market cap of an EM small cap fund is similar to a US small cap fund, the average company market cap in the US is more than twice the size of EMs - so a smaller relative size differences between small and large in EM.

In contrast, some analyses shows the value premium to be larger in emerging markets than in developed markets e.g. this Brandes article, using data from 1980 finds – when comparing decile 1 and 10 value sorts that the value premium was more than double the premium found in non-US developed markets. While EM value funds are not concentrated in decile 10, they likely include these stocks and pick up some of this differences. http://www.grahamanddoddsville.net/word ... ts0708.pdf . The Credit Suisse yearbook also showed a large difference between country quintle sorts on dividend yield (value) and perhaps some of this is also picked up in a RAFI type methodology.

On what expected size and value loads to use: Not sure. FWIW when I set up my portfolio at start of 2003 I simply used long-term estimates of 2% for size and 4% for value (on the assumption that long-term investor behavior going forward would be similar to their behavior over the last 80 years). I only used a market portfolio for EM so didn’t have to think about differing size and value premiums between markets. Even if the same premiums are used, I think there is still case for PXH, due to its potential diversification benefit - the correlation of EM Value with rest of portfolio likely lower than EM market perhaps with higher volatility of this component, which together likely add to portfolio mean-variance efficiency, beyond factor load implied return differences. Obviously no guarantees, and EM could significantly underperform for long-period, or outperform as it has since 2003 (better than expected so far).

Robert
.

User avatar
packer16
Posts: 1214
Joined: Sat Jan 04, 2014 2:28 pm

Re: Emerging markets 1900-2013

Post by packer16 » Sat Aug 30, 2014 9:29 am

What you may be seeing is an institutional (mutual fund, pension fund, etc.) access premium. The primary stocks available to institutions are large cap developed markets stocks. The stocks not available to these investors due to liquidity, % ownership rules and impact on overall performance are small cap US and EM stocks of all sizes due to the small size of EMs vs. DMs. This would explain why the small stock premium persists in the US and other developed markets but is not present in smaller EMs.

Packer
Buy cheap and something good might happen

User avatar
nedsaid
Posts: 11825
Joined: Fri Nov 23, 2012 12:33 pm

Re: Emerging markets 1900-2013

Post by nedsaid » Sat Aug 30, 2014 3:29 pm

My foggy memory banks recall the term Emerging Markets being used only since the 1980's, so my recollection is similar to Nisiprius. I am a bit skeptical about reliable data going back to 1900 for the US Stock Market and even more skeptical about the data for Emerging Markets. Still the data compiled by Dimson, Marsh, and Staunton are better than nothing.

This research suggests that there are Developed Market/Emerging Market cycles just as there are for Small/Large, Value/Growth, and US/International. It also shows the effects of wars and revolutions on market returns, that is a market really can go to zero. This tells me that very broad diversification all over the world makes sense, the closing of a nation's markets doesn't spell doom for a world-wide investor.

Garland Whizzer's posts are always interesting and good. Thought I would make a couple comments here. I would argue that having a debt ratio of over 90% does not doom a country to subpar economic growth. (Edit: debt levels at least in Japan might be a symptom caused by low birth rates and not in themselves a cause for lower economic growth. I wonder if this is true for certain European nations as well). Indeed, the US experienced this after World War II but we worked the ratio down. Of course, it helped that many of our competitors suffered great damage from the war and our economic base was unscathed. (Edit: note also that war and financial crisis both contribute to higher levels of national debt). I agree that demographic factors in the developed nations (i.e. aging populations) could also slow growth. I agree with him that younger populations, lower debt levels, and faster economic growth are good reasons to invest in Emerging Markets. I suppose the best strategy with all this in mind would be to tilt the International portion of a portfolio towards emerging markets.

There are counter arguments of course. First, there are cycles in birth rates. In the 1930's, birth rates declined only to boom after World War II. Birth rates have risen and fallen again since. Second, economic growth does not always translate into stock market performance. Sometimes slower growing economies have better performing stock markets than faster growing economies. Third, the markets know all this stuff and Developed Market stock markets may perform in similar fashion to the Emerging Market stock markets. In other words, markets are relatively efficient.

I am in the camp of tilting an International Stock portfolio towards Emerging Markets. Time will tell who is right.
A fool and his money are good for business.

Valuethinker
Posts: 38176
Joined: Fri May 11, 2007 11:07 am

Re: Emerging markets 1900-2013

Post by Valuethinker » Sat Aug 30, 2014 4:57 pm

nedsaid wrote:My foggy memory banks recall the term Emerging Markets being used only since the 1980's, so my recollection is similar to Nisiprius. I am a bit skeptical about reliable data going back to 1900 for the US Stock Market and even more skeptical about the data for Emerging Markets. Still the data compiled by Dimson, Marsh, and Staunton are better than nothing.

This research suggests that there are Developed Market/Emerging Market cycles just as there are for Small/Large, Value/Growth, and US/International. It also shows the effects of wars and revolutions on market returns, that is a market really can go to zero. This tells me that very broad diversification all over the world makes sense, the closing of a nation's markets doesn't spell doom for a world-wide investor.
No you need a breakdown of international investing, as well as exchange controls, for that. See the 1930s, or for that matter the 1940s 50s and 60s (exchange controls and other barriers to private capital flow).

Interesting comment from Helene Rey (named by the IMF as one of the '25 most promising young economists in the world'-- teaches at London Business School)-- all this capital mobility has contributed a lot to instability, but not much to economic growth. There's a warning in that analysis.
Garland Whizzer's posts are always interesting and good. Thought I would make a couple comments here. I would argue that having a debt ratio of over 90% does not doom a country to subpar economic growth.
Reinhardt and Rogoff got their regressions wrong-- when they copied over the spreadsheet, they missed some rows of data. So the '90%' is a number plucked out of the air, and erroneously so.
(Edit: debt levels at least in Japan might be a symptom caused by low birth rates and not in themselves a cause for lower economic growth. I wonder if this is true for certain European nations as well).
Demographics certainly has an influence on GDP growth. But the real issue with Japan was the bubble blew, and the economy hasn't grown since. They are stuck in a non growth equilibrium. Thus government revenues are low, and spending relatively high. And GDP does not grow. Negative inflation doesn't help.
Indeed, the US experienced this after World War II but we worked the ratio down.
In fact, the US experienced this during WW2, but no one cared much. When you are repressing domestic consumption and maxxing military consumption, your budget goes to hell. Tant pis (too bad) as the French say. The private sector can't really borrow much money, because the public sector crowds it out.

Post WW2 you have rising GDP, rising government revenues, rising everything really. So the ratio falls. But of course private debt to GDP has been rising pretty steadily since the early 1950s.
Of course, it helped that many of our competitors suffered great damage from the war and our economic base was unscathed. (Edit: note also that war and financial crisis both contribute to higher levels of national debt).
War it is obvious why. Financial crisis you have triple whammy: tax revenues fall, expenditures on unemployment etc. rise, and GDP does not grow so debt/ GDP rises. HOWEVER public debt to GDP rises, whereas private debt to GDP tends to fall-- the deleveraging is why you get sluggish growth.
I agree that demographic factors in the developed nations (i.e. aging populations) could also slow growth. I agree with him that younger populations, lower debt levels, and faster economic growth are good reasons to invest in Emerging Markets. I suppose the best strategy with all this in mind would be to tilt the International portion of a portfolio towards emerging markets.
Or tilt your portfolio towards companies with lots of consumers in Emerging Markets. Like GE, Exxon (EM burn oil and gas and need to extract it), P&G, Unilever, Nestle etc. GM does well in Brazil and China. BMW. How about Rio Tinto and BHP Billiton? Both are basically driven by Chinese demand for copper and iron ore. Standard and Chartered? Almost all of its banking business is in Asia and Africa. HSBC-- world's larget bank makes its big profits in Hong Kong.

Then you get EM companies which are big in developed markets. Tata-- tea, steel (bought British Steel/ Corus), Jaguar Rover (stunning British export success story). Cemex (big cement player in developed countries). Arcelor Mittal arguably (it's actually listed in Europe, and is an EM steel co, but also owns Arcelor which was a Luxembourg-European steel co).
There are counter arguments of course. First, there are cycles in birth rates. In the 1930's, birth rates declined only to boom after World War II. Birth rates have risen and fallen again since.
Careful. The Baby Boom was much weaker in Europe and the UK. It's basically a significant factor in the Anglosphere (US, Canada, Australia). All of those countries had very bad 1930s, and a sharp fall in birth rate in the 30s. And then surprising (compared to the post 1918 situation) prosperity in the 1940s and 50s.

Birth rates have been on a downward trend in developed countries since around the mid 19th century. The Baby Boom really was a localized 'blip' after which fertility rates returned to their downward trend. Particularly so If we think high US birthrates are partly the effect of immigrants and the children of immigrants having children (that is not the whole picture, I don't think, but the latest data on birthrates someone showed me here puts the US below France?). Japan, Italy, Spain seem to suggest it could go lower still. (China also but there is social engineering in that one).
Second, economic growth does not always translate into stock market performance. Sometimes slower growing economies have better performing stock markets than faster growing economies. Third, the markets know all this stuff and Developed Market stock markets may perform in similar fashion to the Emerging Market stock markets. In other words, markets are relatively efficient.
Indeed not. There is no correlation between economic growth and stock market performance, I believe Dimson & Marsh found.
I am in the camp of tilting an International Stock portfolio towards Emerging Markets. Time will tell who is right.
PEs are low but you are buying poor governance. I rather like the DFA approach which is to 'clean' the numbers and the portfolios of companies before investing in EM. Some EM companies are shareholder value friendly, some are not.

Russia looks appetizingly cheap, if you can bear the risks.

berntson
Posts: 1366
Joined: Mon Oct 29, 2012 12:10 pm

Re: Emerging markets 1900-2013

Post by berntson » Sat Aug 30, 2014 5:52 pm

Robert,

Thanks for links and the great summary of the literature. It looks like I have some reading to do! Your point about the market cap spread being smaller in emerging markets, and thus predicting a smaller small premium, is interesting and something I hadn't considered. It would make sense. We know that value/growth valuations spreads somewhat predict future returns for the value premium. It would make sense if smaller large/growth spreads also predicted a smaller small premium.

User avatar
Epsilon Delta
Posts: 8090
Joined: Thu Apr 28, 2011 7:00 pm

Re: Emerging markets 1900-2013

Post by Epsilon Delta » Sat Aug 30, 2014 7:10 pm

packer16 wrote:What you may be seeing is an institutional (mutual fund, pension fund, etc.) access premium. The primary stocks available to institutions are large cap developed markets stocks. The stocks not available to these investors due to liquidity, % ownership rules and impact on overall performance are small cap US and EM stocks of all sizes due to the small size of EMs vs. DMs. This would explain why the small stock premium persists in the US and other developed markets but is not present in smaller EMs.

Packer
The argument that EM and small cap are too small for the big boys isn't plausible. It requires that there are no competent, medium sized investors. If these small markets are big enough for a few thousand individual investors to pick up a premium there are plenty of pension funds, endowments, startup funds, etc., that are the right size to take advantage, small enough that it matters to them but large enough to take most of the premium.

User avatar
packer16
Posts: 1214
Joined: Sat Jan 04, 2014 2:28 pm

Re: Emerging markets 1900-2013

Post by packer16 » Sat Aug 30, 2014 8:01 pm

Epsilon Delta wrote:
packer16 wrote:What you may be seeing is an institutional (mutual fund, pension fund, etc.) access premium. The primary stocks available to institutions are large cap developed markets stocks. The stocks not available to these investors due to liquidity, % ownership rules and impact on overall performance are small cap US and EM stocks of all sizes due to the small size of EMs vs. DMs. This would explain why the small stock premium persists in the US and other developed markets but is not present in smaller EMs.

Packer
The argument that EM and small cap are too small for the big boys isn't plausible. It requires that there are no competent, medium sized investors. If these small markets are big enough for a few thousand individual investors to pick up a premium there are plenty of pension funds, endowments, startup funds, etc., that are the right size to take advantage, small enough that it matters to them but large enough to take most of the premium.
This is based upon discussions with managers. They will not touch anything below $150m in market cap in the US and many EM firms are smaller than this and some managers can't get access to markets (like S. Korea) because they have to get approval from all of there shareholders. For many of the big boys purchasing these firms does not make sense. They can't own more than 5% of the company so that is a $7.5 million investment for a pension fund, a drop in the bucket. The economics just do not make sense here versus playing in the larger cap game where if they do well they will have an impact.

Try to find a mutual fund with an average mkt cap anywhere close to $150m. It is very rare. The average mkt cap of the Vanguard Small Cap Value fund in $2.8 billion. There are some smaller funds but the economic incentives are smaller than in the larger cap space. It easier to play in the larger space and collect your fee than put up capital in this space and collect returns that are subject to losses.

Packer
Buy cheap and something good might happen

User avatar
nedsaid
Posts: 11825
Joined: Fri Nov 23, 2012 12:33 pm

Re: Emerging markets 1900-2013

Post by nedsaid » Sat Aug 30, 2014 8:55 pm

Valuethinker, thank you for commenting on my post.

Excellent point about capital mobility around the world. We are seeing capital flight from Russia right now with the Ukraine crisis. Whenever a crisis like this hits, we could expect money to flow from emerging nations to safe haven nations. So the Russian Stock Market would be hurt and the US Stock Market would be helped. It seems that there are probably trillions in what I call "hot money" moving around the world at the speed of a mouse click. This causes greater volatility in the stock markets around the world, particularly the emerging markets which are most prone to capital flight.

I would agree that Japan is still feeling the effects of the bursting of their stock market and real estate bubbles in the early 1990's. It was the loss of an immense amount of wealth which Japan has never quite recovered from. I also think of how the financial crisis affected Ireland and the United States. In effect, what happened was an exchange of public debt for private debt. Debt is money in the economy and if it is destroyed in the private sector, the public sector has to pick up the slack if you don't want to risk depression. From what I understand, Japan has not written off its bad private sector debts to the degree that we did in the United States. What has helped Japan immensely is its large trade surplus particularly with the United States.

I should have said that World War II was the cause of the build up of public debt and if I remember right it went to over 120 percent of GDP. This is one reason I was very skeptical of the 90% of debt to GDP ratio being sort of a Rubicon if crossed dooms a nation to lower economic growth. As I recall, the United States emerged from the Revolutionary War pretty well broke. But somehow, we succeeded as a nation after that and after WWII.

You also hit on a very important point of credit expansion in the private sector. This pretty much hit its limit just before the 2008-2009 financial crisis. Obviously, we had credit contraction in the private sector since then (though it is now growing again) and that doesn't help economic expansion. The US Government picked up the slack to the tune of trillions of dollars. The post WWII credit expansion as well as population growth were big drivers to economic growth.

Yes, you can benefit from economic growth in the Emerging Markets by investing in US companies that derive a lot of their business from Emerging Markets companies. The American Funds fund family invests in Emerging Markets that way. You also raised the point about large Emerging Market companies that do business here in the United States. Funny that you would mention that, I have a family member who works for CEMEX, the Mexican company.

Yes, there are more risks in Emerging Markets. I saw posted the Standard Deviation of Emerging Markets over different time periods and I was astonished how darned volatile this stuff is. The numbers were higher than for US Small-Cap Value. I invest in Emerging Markets but I don't overdo it for this reason.
A fool and his money are good for business.

Post Reply