Thursday, December 20, 2012
You might want to read this paper from Vanguard's research department on the utility of major macro approaches to predicting future stock returns. The conclusions are so well stated I’ll reprint a long quote from the article:
Asset allocation and the difficulty
of predicting the future
We’ve shown that forecasting stock returns is a
difficult endeavor, and essentially impossible in the
short term. Even over longer time horizons, many
metrics and rough “rules of thumb” commonly
assumed to have predictive ability have had little
or no power in explaining the long-run equity return
over inflation. Although valuations have been the
most useful measure in this regard, even they
have performed modestly, leaving nearly 60%
of the variation in long-term returns unexplained.
What predictive power valuations do have is
further clouded by our observation that different
valuations, although statistically equivalent, can
produce different “point forecasts” for future
This underscores a key principle in Vanguard’s
approach to investing: The future is difficult to
predict. As such, we encourage investors not to
focus on the “point forecasts” that result from
various forecasting models and instead turn their
attention to the distribution of potential future
Once future prospects are viewed in a distributional
framework, the benefits of strategic asset allocation
become clear. A focus on the distribution of possible
outcomes highlights the benefits and trade-offs of
changing a stock allocation: Stocks have a higher
average expected return than many less-risky asset
classes, but with a much wider distribution, or level
of risk. Diversifying equities with an allocation to
fixed income assets can be an attractive option for
those investors interested in mitigating the “tails” in
this wide distribution, and thereby treating the future
with the humility it deserves
A couple points from the article:
· In the short term markets are basically unpredictable.
· In the longer term p/e ratios offer some predictability high p/e ratios predict lower long term returns as they revert to the mean and vice versa. But even this measure had only correlation of .40 to future returns meaning even in the best case 60% of market returns were unpredictable.
Some Implications for Investors:
1. These relationships between p/e ratios and future returns might argue for a bit of a “value tilt” in a portfolio overweighting low p/e stocks. But such a strategy can only be expected to generate a higher long term return if the investor is willing to stick with the strategy for a long period.
During the tech boom of the 1990s with growth significantly outperforming value many investors sold their value stocks. And since there is a herding (momentum) factor in the shorter term markets can “overshoot” taking prices and p/e ratios higher before they revert to the mean and fall. This makes it even harder for value investors to hold on. In light of human nature and well known factors in behavioral finance one might be well advised not to weight too highly to value stocks.
Professional investors are not immune to the tendency to join the move to high p/e stocks and abandon value investing …even if their basic instincts tell them not to do so. Professional investors face “benchmark risk” which translates into “career risk”. A period of 3,2, or even 1 year of underperformance vs. a benchmark such as the S+P 500 will lose to outflows of funds from the pool of assets under management. Less assets under management = lower compensation for the manager and the fund company in turn raising “career risk” for the manager. Hence a tendency for even the more knowledgeable professional investor to join the herd in a bubble rather than “fight the tape”.
Recognizing these tendencies it’s not hard to see why bubbles and busts can occur in the short term and why virtually all the great (few as they are) long term investors have been value investors and why all the most successful factor based investing strategies (used by passive vehicles such as ETFs and index funds) use some variant of a value tilt even while they may differ on exact methodology.
2. Point forecasts are basically useless but longer term probabilistic approaches can be useful.
I am in the midst of reading Nate Silver’s The Signal and the Noise and he reinforces this approach. While it is extremely difficult to make a point forecast it is far easier to forecast a range of returns and a probability of outcomes. Although people focused on his highly successful forecasts for the results of the last two elections they might have overlooked the fact that he always gave his forecasts in terms of probabilities rather than a point forecast. He has some interesting things to say about financial markets
In this interview (interestingly the motley fool which hosts the interview promotes exactly the approach Silver argues against)
He pretty much lines up with the view the view that one can forecast the long term probability of a range of return with much greater success than trying to make a point forecast
probability of identifying the probability of returns over a ten year period far more useful than a point forecast. While it may disappoint those looking for a point forecast here is how Vanguard views the outlook for future returns:
Turning to real returns, we estimate a
slightly greater than 50% likelihood that over the
2012–2022 period, the broad U.S. stock market
will earn at least a 5% average annualized real
return. As such, we feel our expectation for the
forward real return is quite in line with the historical
average of 6.8% that has been observed since 1926,
and does not represent a drastic change in the
risk–reward characteristics of the equity market.
Match that up with a commonly held approach that current yields represent the best forecast of future nominal bond returns one could get an idea of a portfolios expected return. A strong case has been made that this argues for a negative real return on treasury securities. As a consequence it would argue for a significant weighting in corporate bonds as well avoiding long duration since a low future inflation expectation is built into future corporate and treasury bond yields.
So final thought: skip the end of year forecast articles, instead spend your time enjoying family and friends over the holidays.
Tuesday, December 18, 2012
The US stock market is certainly closing on a great year. + 16,3 %for the SP 500 (ETF SPY) and +16.4% for the US total market (VTI)
But while you might not have been looking emerging markets (etf EEM) have just about caught up. They are up 16.1% ytd.
The biggest surprise for some is the gain for German stocks +30.7.
Growth of $100,000 top chart, Total Return and Volatility Below YTD Also included is AGG the total bond market index etf.
Long term stock market forecasts (unreliable as they are) seem to range from the +5% to +7%. That means that if returns revert to the long term mean forecasts, missing this years 16% + returns will mean a big hit to the long term returns on investors.....
And it seems that many investors did exactly that moving money out of stocks and into bonds. An anecdotal example of investment behavior is here .
Numbers on the flows from Lipper are here. Reflecting the trend seen all through 2012, during November 2012 investors added $25.7 billion into bond funds and $69.6 billion into money market funds while they pulled a net $14.4 billion out of stock funds.
Individual investors are notorious for performance chasing and market timing rather than sticking to a long term asset allocation, It seems investors had their eyes firmly directed into the rearview mirror making their allocation for 2012 based on the 2011 results: +1.9% for the SP 500 (SPY) and + 7.7% for the Aggregate bond index (AGG) and emerging markets -18.8%. Growth of $100,000 and return volatility below.
Academic research has found that it “beta doesn’t always line up” that it is possible to construct portfolios that produce alpha relative to the cap weighted market index : higher risk adjusted returns. Note that the strategy is not expected to outperform the market but rather to offer greater return relative to the risk of the portfolio. Thus the expectation is that they will outperform the market cap index in stable or down markets and outperform in up markets but produce higher return per unit of risk (alpha). You can find a discussion of this strategy by one of its principal academic advocates Robert Haugen here
ETFs replicating this strategy have been in the market a bit over a year and so far have met their expectations. Additionally although they correlate strongly with value stocks (there is considerable overlap in holdings) they produce meaningfully different risk/retun characterisitcs.
Here is the one year total return (top)and volatility ( bottom) a measure of risk) for one year for. VTI the total stock market, VTV large value US, SPLV the S P 500 low volatility strategy and USMV the total US market minimum volatility ETF as well as AGG the total bond market index:
Note that although the low/minimum volatility ETFs have produced lower absolute returns they have delivered on their goal of better risk adjusted returns. They have produced a bit over 80% of the return of the cap weighted index with less than 70% of the risk.
The international minimum value portfolios in emerging markets EEMV and developed markets minimum volatility EFAV. In the developed markets EFAV has given better risk returns than the cap weighted EFA
Both the emerging and developed international one year return (top) and risk (volatility) are below
and in emerging markets EEMV has achieved the holy grail of inveting more return and less risk vs. the cap weighted EEM.
The wide divergence between the 2 emerging markets instruments can be traced to the divergent sector weightings(below). EEMV has significantly lower weighting in the commodity industries : energy and basic materials which have high volatility due to the volatility of commodity prices. Given the short history of these instruments it is not hard to imagine a period in which strong and volatile commodity prices produced higher return and higher risk for EEM. Given the fickle nature of commodity prices on should be hesitant to draw too many conclusions.
EEM Sector Breakdown
EEMV Sector Breakdown
How to use these strategies:
There are two major uses for these instruments:
- 1. Constructing a lower risk/lower prospective return equity strategy. For investors looking for lower potential losses in a down market and willing to forego higher returns in strong bull markets this would be a good substitute for part of their equity holdings.
- “ Risk Budgeting” based on the data a minimum/low volatility strategy could allow an investor to increase his equity allocation relative to lower volatility fixed income and increase portfolio return per unit of risk.
Sunday, December 16, 2012
The WSJ reported on a topic that has been revolving in professional circles for quite awhile. Vanguard will be changing benchmarks for its emerging markets ETF (ticker VWO)going from MSCI to FTSE as its index benchmark. The most important impact, as the article notes is that South Korea will no longer be in the Vanguard ETF:
As a consequence investors have been pulling large amounts out of the Vanguard ETF, the article reports $900 million in outflows in the past month .
Looking at performance over the last 3 years, it is not hard to see the reason for investors desire to retain investments in South Korea.
3 year performance
VWO Vanguard Emerging Markets (which still includes South Korea), 13.2%
SCHE Schwab Emerging Markets ETF (which is based on the MSCI emerging markets index which
excludes South Korea) 6.8%
South Korea country index EWY .33.4%
But investors and the WSJ may have jumped the gun. A strong argument can be made to retain the South Korean exposure but the change will not occur overnight and has not yet begun. Below is the performance of Vanguar
The transition at Vanguard will be done gradually over six months beginning in January, so arguably even those that want to retain South Korea as part of their emerging market holdings might have rushed to the exits too soon. The prospective change has not yet had an imapct on performance.
As can be seen below VWO the Vanguard ETF has had nearly identical performance to EEM the ishares ETF that has no plans to switch away from the MSCI index which includes South Korea. During the period of those $900 million outflows (since November 1) Both VWO and EEM have outperformed Schwab’s emerging markets ETF SCHE which already uses the FTSE index which excludes South Korea.
Performance is shown as growth of $100,000 VWO blue, EEM green, SCHE yellow) click to enlarge
What is the best way for investors to proceed:
- They could switch before January from VWO to EEM. Based on inflows into EEM it seems that some investors are doing exactly that. EEM does however have relatively high management fees of .67%. Investors might want to consider the newer lower fee ishares emerging markets ETF IEMG (.18% management fee) which both include South Korea.
- They could retain the low fee VWO and add holdings of EWY the South Korea ETF.
Some have expressed the view that in response to the VWO rebalancing South Korean stocks may have a selloff that creates along term buying opportunity in the South Korean market. Two ways to implement a timing strategy for buying South Korea would be:
Selling VWO and investing the proceeds into SCHE the Schwab ETF which is based on the FTSE (non South Korea inclusive) index.
After a selloff of South Korean stocks sell the SCHE and then buy one of the ETFs that includes South Korea (EEM or IEMG)....or retain the SCHE position and add a country position in EWY (South Korea)
Hold the VWO position but add a position in the South Korea ETF (symbol EWY) either now, at some point during the transition process or at the time of a possible selloff.
Of course the latter move may just be too esoteric for many investors.
But as the number shows there is a good reason to make portfolio adjustments in the near future in response to the changes in VWOs indexing methodology.
While on the topic of emerging market investing it is worth making note of the relatively new minimum volatitlity emerging markets index which not only has a significant holding in South Korea 10.45% but also has a different mix of companies compared to other emerging market indices thus giving more diversification in emerging market holdings. While EEMV has only been in existence since October 2011 it's track record is impressive (eem vs eemv below returns top volatility below).
The minimum volatility strategy has become available across for both US and international markets and there is a significant amount of research supporting that approach.