Archive for the ‘Allocation’ Category

Risk Asset Allocation Based On Momentum Relative To T-Bills

Friday, September 7th, 2018
  • It’s more about winning by not losing in Bear markets than winning in Bull markets.
  • The approach tends to generate higher returns and lower maximum drawdowns than Buy-and-Hold when a Bear is in the study period.
  • The approach tends to underperform Buy-and-Hold in study periods that do not include Bear markets.
  • The approach with a single risk asset versus T-Bills tends to be more tax efficient in regular accounts than selecting the highest momentum of several risk assets or T-Bills due to longer in-market periods with more long-term taxable gains.
  • The approach is better suited to tax-deferred accounts than regular taxable accounts.
Questions continue to come in about systematic approaches to tactical allocation portfolio changes – methods not based on biases, media frenzy, forecasts, chart patterns, valuation or fundamentals.

Probably the simplest approach, and one that is about as objective and non-judgmental as systems can be, is to hold those risk assets that are doing best, unless none of them are doing better than T-Bills; in which case hold T-Bills until the risk assets start doing better than T-Bills.

The most reliable aspect and driver of returns of the approach is minimizing major drawdowns, as shown by the flat lines in the blue timing portfolio when the red benchmark portfolio experienced major drawdowns.

Secondarily, the approach may increase total return. However, each trade triggers a tax event, so only those cases where the average holding time between trades is over 1 year avoid ordinary income taxes on the trades; and those that have the longest position hold times benefit most by compound growth.

The approach can be used in an outright decision to hold risk assets or risk-free T-Bills (or other “cash”), or it can be used for guidance in a strategic allocation portfolio to make overweight and underweight decisions.

The approach is not tax efficient for regular taxable accounts, but in some cases the outperformance may be worth the tax cost. Trading systems such as this are best used in tax-deferred accounts.

Relative and absolute momentum can be done at an extremely simple level, such as rotating between one specific risk asset and T-Bills (e.g. the S&P 500 and T-Bills).

That method could be used in layers, with more than one pair of two assets (e.g. S&P 500 and T-Bills as one pair, and emerging markets stocks and T-Bills as another).

It is also possible with the method to choose one or more risk assets within a larger group of risk assets or T-Bills (e.g. the two strongest momentum risk assets out of five or T-Bills).

Working with simple pairs of one risk asset versus T-Bills is more likely to create longer position hold times, and therefore be more tax efficient for regular taxable accounts than alternating between risk assets or T-Bills. Alternating between risk assets creates more frequent trades and would be more suitable for in tax deferred accounts.

The method helps avoid the severity of Bear markets, but it also needs Bears to generate enough of a performance difference to make the effort and potential tax cost worthwhile.

The approach does not work with all assets. Testing is required to see if it worked in the past.

Here are some actual 09/03/18 relative return and absolute momentum indicated holdings:

It is possible to operate this approach by visual inspection of charts, but for more precision and to avoid seeing what you may want to see, using quantitative data is preferable.

You would need to find the performance evaluation period (e.g. the number of days, weeks, months quarters, years) that produces the best combination of high return and low maximum drawdown, based on long-term history, and the best frequency of decision making (e.g. daily, weekly, monthly quarterly or longer). The “best fit” evaluation period and decision frequency will probably drift over time, so it needs to be regularly evaluated for adjustment.

The information that follows gets a bit into the weeds of the approach.

The securities we used for the test were selected in part because they had adequate history to capture a sufficiently long study period:

  • VFINX: S&P 500
  • VGTSX: total international stocks
  • VEIEX: emerging markets stocks
  • VGSIX: US real estate
  • VFINX: intermediate-term Treasuries
  • Treasury Bills

ETF alternatives may be more tax efficient for current use, but did not have sufficient history for long-period analysis.  These ETFs are classes that invest in and share the same underlying portfolio as the mutual funds used in the study (except for BIL):

  • VOO: S&P 500
  • VXUS: total international stocks
  • VWO: emerging markets stocks
  • VNQ: US real estate
  • VGIT: intermediate-term Treasuries
  • BIL: Treasury Bills

If the investor wishes to use options (for example, covered Calls) on the S&P 500 position, then SPY would be needed because it has the options and liquidity that would be suitable.

There are other ETFs from other sponsors that could just as well be used instead of the Vanguard ETFs in a forward practice.

First, Portfolio Visualizer is among the sites that provide good tools for experimenting with the approach, and it is free. There are subscription sites that have a number of pre-built models that can be followed as well.

The data that follows is entirely from Portfolio Visualizer.

The tables above and below illustrate how the evaluation period and frequency might be regularly reviewed for possible adjustment, and provide evidence of the efficacy of the approaches from 1997 to 2018 YTD, as well as return, maximum drawdown and several other important metrics for the results for method and their benchmarks.

The illustrative cases are for the S&P 500 as a single risk asset versus T-Bills (above); and the best 2 of 5 risk assets versus T-Bills (below).

The charts and tables that follow are for S&P 500 versus T-Bills, and for selecting the best 2 of 5 assets (S&P 500, DM stocks, EM stocks, US REITs, and Intermediate-term Treasuries) versus T-Bills.

In each case, the backtest period was about 42 years (1997-2018 YTD).

In the best 2 of 5, the result could be 2 risk assets, 1 risk asset and T-Bills, or all T-Bills.

Single Risk Asset vs T-Bills

The timing model produced a 19.67% return over the period 1997-August 2018 versus 8.42% for the equal weight portfolio, which in this case is 100% S&P 500. Note that these results are based on investable assets, not indexes.

While results such as this could have been available to you if you did this on your own, they would not have been effective for a “hedge” fund type investment with their 2% management fees and 20% of gains above a preferred return, because the return difference of 2.22% would have been wiped out by their fees.

The volatility (standard deviation) of the method was lower than buy-and-hold. The best year was better, and the worst year was dramatically better (negative 7.67% versus negative 37.02%). The superior maximum drawdown was critical (down 16.31% versus down 50.97% for buy-and-hold).

The average trade lasted 29.9 months, which tended to create capital gains as opposed to ordinary income when trades were executed.

The Sharpe Ratio (return in excess of T-Bill return, divided by the standard deviation of return – how “bumpy” the ride was) was better for the method (0.79 versus 0.48), as was the Sortino Ratio (like the Sharpe Ratio but only divided by the size of downside volatility) at 1.29 versus 0.70.

Last, as a small form of diversification, the correlation to total US stocks was only 0.74 versus 0.99 for buy-and- hold

The complete trade history from 1997 is in this table:

However, not all backtests are inspiring. The specific time period you test and whether or not a Bear intervened are critical to how attractive the method appears.

It can be years of effort with little if any outperformance to show (and taxes in regular accounts) if a Bear is not in the period.

It takes a Bear to make relative momentum versus T-Bills look good.

Best 2 of 5 Risk Assets vs T-Bills

As with the single risk asset (S&P 500) versus T-Bills, the best 2 of 5 risk assets was superior (excluding any tax considerations) to the equal weight, buy-and-hold approach.

Compound return was better by 4.73%. Standard deviation was almost the same. The best year was better by 16.73%. The worst year was better by 13.68%. The maximum drawdown was better (less severe) by 26.15%. The Sharpe Ratio was higher by 0.37. The Sortino Ratio was better by 0.79. The correlation to total US stocks was lower by 0.30, and the Beta to total US stocks was lower by 0.22. All good.

There were many trades for an average trade holding time of 2.37 months. That created a lot of ordinary income for someone doing this in a regular taxable account. The trades over the past year are in this table.

Selecting the best 2 of 5 risk assets might be expected to be more effective during periods without a Bear market, but the evidence in our test suggests that is not the case. Looking at the period 2010 through August 2018 (a period without a Bear), we were able to find an evaluation period that produced a 2+% higher return (before any tax costs), but it produced a 1+% more severe maximum drawdown — it did not provide the downside protection the method is intended to provide.

The most useful evaluation period we found for the best 2 of 5 model was 12 months. It produced a 9.56% return versus 7.49% for the equal weighted benchmark, but experienced a 16.69% maximum drawdown versus 15.50% for the benchmark.

The following charts show the results for evaluation periods of 1, 11 and 12 months.

1 Month Evaluation Period (2010 –Aug. 2018)


11 Months Evaluation Period (2010 – Aug. 2018)


12 Months Evaluation Period (2010 – Aug 2018)


Because we consider maximum drawdown protection key to the utility of relative momentum versus T-Bills, our conclusion is that the method needs a Bear market to be attractive.

Relative momentum allocation between a risk asset and T-Bills can produce meaningfully higher return and substantially less severe maximum drawdowns, but the fact is that the avoidance of large drawdowns (Bear markets) is the most important contributor to the potential of the method to outperform.

If there is no Bear market during the time the method is used, the potential for outperformance is limited.

Tax costs are a problem in taxable accounts (more when selecting among multiple risk assets, because position holding time is shorter with more ordinary income tax cost).

No universal evaluation period is effective for all risk assets, and the evaluation periods that work best drift over time, making continuous curve fitting part of the challenge and work load.

Not all assets have good potential with the method (basically only those that have occasional major drawdowns appear to work well).

Because of the role of Bear markets in creating advantage for the method, it is important when reviewing reports about the method to distinguish between long-term studies incorporating one or more Bears, and short-term studies of Bull markets.

Winning by not losing is the most important benefit of the method. The benefit derived is greatest when the loss experienced by buy-and-hold is greatest.

As must always be said, past performance is no guarantee of future performance.

I hope that for those of you who have been asking about the popular idea of risk reduction with relative momentum versus T-Bills, that this is helpful.

 

 

Equity Reallocation Between US, Developed And Emerging Markets

Monday, January 8th, 2018
Within the equity allocation of portfolios, with tax awareness, we are taking these actions in discretionary accounts, and are recommending these actions in advice & consent accounts, and in coaching relationships:
  • Underweight US stocks (preferably to less than the 52% world weight)
  • Overweight non-US Developed Mkt stocks (preferably to more than the 39% world weight)
  • Overweight Emerging Mkt stocks (preferably to more than the 9% world weight)

We have been strongly overweight US stocks within the equity allocation for years, and that has been beneficial; but now evidence suggests that greater opportunity lies in other parts of the world.

A significant challenge we face is the pretty much across the board embedded gains in our stock positions.  Reallocation in tax deferred accounts is not a problem, but in regular taxable accounts, we need to understand the taxable gain preferences (and tax loss carry-forwards) of our clients versus their need to reallocate.

The following information supporting our view provides:

  • Comparative valuation measures for the three regions (US, non-US Developed and Emerging markets)
  • What is fair value of stocks today relative to bonds?
  • Recent and multi-year relative total return performance of the three regions versus the world index
  • Multi-period institutional forward return forecasts for the three regions
  • Net flow of funds to US and International/Global mutual funds and ETFs
  • Current short-term technical ratings for the three regions and the world
  • QVM intermediate-term trend ratings for the three regions and the world.

COMPARATIVE VALUATION MEASURES 

  • Price-to-Book Value Ratio
  • Price-to-Sales Ratio
  • Price-to-10yr Av Earnings Ratio (by Robert Shiller, Yale Economist, Nobel laureate)
  • Price-to-Highest Historical Earnings Ratio (by John Hussman, analyst and mutual fund manager)
  • Price-to-Total of Book Value and Debt (by James Tobin, deceased Yale Economist, Nobel laureate)

The short-story here is that by each measure, the US is the most expensive; non-US Developed markets are either somewhat less expensive or mostly significantly less expensive, and Emerging markets are significantly less expensive.  If we are seeking best relative value, and potentially least exposed to valuation contraction, regions with lower Price-to-Whatever ratios are generally more attractive.

The following comparative valuation measures were generated recently by Research Affiliates for their January article, ”CAPE Fear: Why CAPE Naysayers Are Wrong”. Valuations are compared between the regions and for each region relative to its own history.

Each of the following measures presents the current value inside of a “box and whiskers, explained by this image:

Price-to-Book Ratio: 

Price-to-Sales Ratio: 

Price-to-10yr Av Earnings:

This ratio (commonly called CAPE, for cyclically adjusted Price Earnings Ratio, where the adjustment is an inflation adjustment to historical earnings) was developed by Robert Shiller, Yale Economist, based on the idea published by Benjamin Graham many years ago.

Hussman P/E Ratio:

John Hussman divides the current price by the highest prior peak earnings, to present the lowest possible P/E (most favorable) based on actual historical results.

Tobin’s Q Ratio:

Tobin’s Q ratio, is the ratio of the market expressed value of a company (as measured by the market value of its outstanding stock plus debt) divided by the current replacement cost of the company’s assets (actual replacement cost is a complex determination, so for simplicity book value is often used for replacement cost, which typically overstates the true ratio, because book value is after depreciation and is not inflation adjusted — however in a time series for a single security, or between indexes, it is probably useful).

WHAT IS A MEASURE OF FAIR VALUE TODAY RELATIVE TO BONDS?

One expert to consult is Benjamin Graham (1894-1976), the acknowledged farther of value investing (as well as professor to and eventual employer of Warren Buffet).

He wrote the seminal books: “Security Analysis” (1934) with David Dodd, and the “Intelligent Investor” (1949) which Warren Buffet described as “the best book about investing ever written.”  I remember reading “Intelligent Investor” in the summer of 1968 while working trail crew on the Appalachian Trail between my Sophomore and Junior years in college.  It was my first exposure to investing ideas.  It was captivating then.

In “Common Sense Investing: The Papers of Benjamin Graham” (accessible here) he said:

“It seems logical to me that the earnings/price ratio of stocks generally should bear a relationship to bond interest rates.  …I should want the Dow or Standard & Poor’s to return an earnings yield of at least four- thirds that on AAA bonds to give them competitive attractiveness with bond investments.”

Today, AAA credit yields 3.47%.  Four thirds of that suggests an earnings yield is 4.63% (or a P/E ratio of 21.6).  Prospectively, with the Fed on course to raise short-term rates by 0.5% to 1.0% in 2018, it is reasonable to assume that the AAA yield will be higher than 3.47% at the end of 2018, and that an attractive earnings yield would be higher than 4.63% (a P/E lower than 21.6).

Today, according to Vanguard, the trailing P/E on each of the funds respresenting world stocks and three stock regions are:

  • World stocks (VT) P/E 18.9 (earnings yield 5.29%)
  • Total US stocks (VTI) P/E 22.7 (earnings yield 4.41%)
  • Large-Cap non-US Developed markets stocks (VEA) P/E 16.0 (earnings yield 6.25%)
  • Large-Cap Emerging markets stocks (VWO) P/E 14.8 (earnings yield 6.76%)

This is just one generalized way to look at fair value, but given the source, it is worth noting.  The US is somewhat expensive by this Graham metric.  The world and non-US developed markets are attractive.  Emerging markets are very attractive (and much more volatile – forecasted to be in the low 20’s versus mid-teens for US and non-US Developed markets).

COMPARATIVE TOTAL RETURN

Each chart shows cumulative returns for total world stocks (VT) versus the total US stock market (VTI) and the large-cap non-US Developed markets (VEA) and the large-cap Emerging markets (VWO).  The US has been the clear winner since the stock market bottom in March of 2009, but international markets have recently pulled ahead.  Since the US has stretched the valuation rubber band more than the other two regions, that suggests greater forward opportunity internationally.

3 Months cid:image020.png@01D387EF.3F0AC400

 1 Yearcid:image021.png@01D387EF.3F0AC400

 3 Years

cid:image022.png@01D387EF.3F0AC400

 From the 2009 US Stock Market Bottomcid:image023.png@01D387EF.3F0AC400

NET FLOW OF FUNDS

The Net Flow of Funds to mutual funds and ETFs has favored international and global funds over US domestic funds over the past 3 years.  US domestic funds still outperformed international and global funds, but investor preference is clear in this chart.  Until international and global funds show signs of overvaluation, “follow the money” may be a good thing to do.

INSTITUTIONAL FORWARD RETURN FORECASTS

While institutions vary in the magnitude of multi-year forward annualized total returns, they are in essential agreement in forecasting Emerging markets being the highest return opportunity among the three regions ( with significantly higher volatility than in the US or non-US Developed markets).  US stocks are generally seen as offering the lowest returns.  Institutional forecasts are consistent with the pattern of Net Flow of Funds.

 CURRENT SHORT-TERM TECHNICAL CONDITION

These ratings come from BarChart.com.  In their parlance, “short-term” is in the vicinity of 20 days; “medium-term” is in the vicinity of 50 days; and “long-term” is in the vicinity of 100 days – all of which we view as short-term.

QVM INTERMEDIATE TREND CONDITION

Our proprietary trend indicator is measured on a monthly basis over a period in excess of 1 year, and we believe indicates the intermediate trend condition.   A description of the measurement method  can be found here:

Note that while the world and all three regions are in strong positive up trend; but they are “running hot” and are “overbought” — suggesting a likely slowing or Correcting to cease being “overbought”.

 

10-Year Future Returns For Model Portfolios From 5 Thought Leaders

Monday, October 16th, 2017

Let’s look at what 5 important thought leaders had to say about constructing a portfolio for the long-term (10+ years); and how their portfolio models are expected to perform over the next 10 years, based on asset returns, volatility and correlations according to JP Morgan Asset Management and Research Affiliates.

Here’s the spoiler:

  • Ray Dalio’s “All Weather” portfolio has the best risk adjusted expected return
  • David Swensen’s “Reference” portfolio has the highest expected return
  • John Bogle’s traditional “60/40 Balanced” model using global stocks is a contender
  • Harry Browne’s “Permanent” portfolio isn’t a contender
  • Warren Buffet’s “Estate Plan” portfolio has the highest potential, but the worst risk adjusted return
  • For all portfolios, the expected returns are significantly below returns of the last several years.

Here are mug shots of the 5 thought leaders.

(click images to enlarge)

Their portfolio models in our approach, would be for the Broad Core sleeve.  More income intensive securities would go into the Income Core sleeve of the portfolio.   And, when, as and if it makes sense to make some tactical bets, we can add narrowly focused assets we believe will do particularly well in the Tactical Opportunity sleeve of the portfolio.  This discussion is only about portfolio models for the Broad Core sleeve.

  • David Swensen is the long-time CIO of the Yale endowment.  He is acknowledged as one of the best at that job, if not the best.  In 2005, he translated his 2000 book for institutional portfolio managers (“Pioneering Portfolio Management: An Unconventional Approach to Institutional Investment”) to one suitable for individual investors.  In that book (“Unconventional Success: A Fundamental Approach to Personal Investment”), he described a “reference portfolio” for the long-term.  He made it clear that adjustments are needed to fit the specific needs, goals, limitations and psychology of each investor; but that the reference portfolio is a good place to begin the thinking process about a suitable portfolio. In his reference portfolio he stresses the functions which must be provided (capital growth, inflation protection, deflation protection, and protection in times of crisis).  He specified assets for the portfolio and how much of each function each asset provides.
  • Ray Dalio is the founder and CEO of Bridgewater Associates, one of the oldest and largest private funds (about 40 years old and about $160 billion in assets).  He wrote about his All Weather strategy in a white paper in January 2012.  His fund has been very successful with the core All Weather portfolio model, and around which they make certain other tactical investments.  Their core portfolio is equally weighted to deal with 4 scenarios (rising corporate profits, falling corporate profits, rising interest rates, and  falling interest rates).  Then within each of those 4 segments of the All Weather portfolio, he specified suitable asset categories, which are then equal weighted for volatility (“risk parity”).
  • Harry Browne (deceased) was an investment advisor and author, who in his 1999 book “Fail Safe Investing” was focused on limiting losses in a portfolio that would be profitable in any market, which he posited was best accomplished with equal amounts invested in stocks, long-term Treasuries, T-Bills and Gold.  That was dubbed a “permanent” portfolio.
  • John Bogle is the founder and former CEO of Vanguard.  He virtually created the index mutual fund industry, launching the first ever index fund (dubbed “Bogle’s Folly”) on December 31, 1975, to track the S&P 500 index.  He is a firm believer that no amount of effort can predict the future of markets, and that active management is not competitive with index funds over long periods.  He recommends simply owning a broad index stock fund, such as one based on the S&P 500, and a broad bond index fund, such as one based on the aggregate US bond market.  He is an advocate of a 60% stocks, 40% bonds allocation.
  • Warren Buffet is the CEO of Berkshire Hathaway, which has made him one of the richest people in the world.  He is a fierce critic of the expense drag created by actively managed funds and their trading costs due to high turnover.  He believes that low cost index funds are the only way to go, and that a domestic focus is at least good enough, if not best.  In his 2013 letter to Berkshire Hathaway shareholders he gave investment advise which he felt was good enough for most investors by revealing his instructions to his estate trustee:  “… instructions I’ve laid out in my will … cash will be delivered to a trustee for my wife’s benefit … My advice to the trustee could not be more simple: Put 10% of the cash in short-term government bonds and 90% in a very low-cost S&P 500 index fund. (I suggest Vanguard’s.) I believe the trust’s long-term results from this policy will be superior to those attained by most investors – whether pension funds, institutions or individuals – who employ high-fee managers.”

Using the 10-year forecasts for asset returns, volatilities and correlations published by JP Morgan Asset Management (optimistic) and by Research Affiliates (less optimistic), let’s see how the recommended portfolios from those 5 thought leaders would be expected to perform in the future.

It is important to know that any single value projection of portfolio return is not a good idea.  Because assets go up and down in price over time, it is more reasonable to think of future returns in terms of a probability range of return; with the most likely return near the middle of the distribution, and the increasingly less likely, but still possible returns spread out above and below the middle.  That is how our first image is presented.  The probability distribution is developed through Monte Carlo simulation.

For each portfolio, the return probability distribution is represented by a box with whiskers.  The line in the middle of the box is the 50th percentile return, which is equal to or near the mean expected return.  Half of the probable returns expected to be higher than the 50th percentile, and half lower than that value.  The entire box encompasses the middle 80% of probable returns (from the 10th to the 90th percentile).  The whiskers extend out above and below the box so that the entire box and whiskers encompass 98% of probable returns (only 1% are expected to be higher, and only 1% are expected to be lower).  Because of shocks and crises, the lower 1% could be a lot lower.  For that, we present what could be thought of as a worst case (at least a very bad case) year, when the return is 3.5 standard deviations below the expected mean return.

This chart is the main one we used to decide our spoiler alert at the beginning of this letter.

The green colored boxes and whiskers are based on the more optimistic JP Morgan assumptions.  The yellow colored boxes and whiskers are based on the less optimistic Research Affiliates assumptions.

Note, that only the deeper green and deeper yellow colored boxes and whiskers have their 1st percentile above zero. If the bottom whisker is above zero, it means that the forecasts expects a 99% chance that at the end of 10-years, your portfolio will be worth more than at the beginning.

The portfolio models with the bottom whisker above zero when using JP Morgan assumptions are the Swensen Reference, Dalio All Weather, and Bogle 60/40 with global stocks. The models with 99% chance of a positive 10-year outcome when using the Research Affiliates assumptions are only the Dalio All Weather portfolio, and the Simple 2 Asset Risk Parity model (which uses the S&P 500 and intermediate-term Treasuries in a ratio that expects the same risk from each asset).

The lighter colored boxes and whiskers have varying probabilities less than 99% that the portfolio will grow in value over 10 years.  In fact, the simplified Dalio All Weather portfolio (published  by Tony Robbins, based on his discussions with Dalio), has about a 10% chance after 10 years of being worth less than the starting value.  The Warren Buffet “estate plan” portfolio has by far and away the higher upside potential (because it is 90% stocks), but also something near a 10% chance of a cumulative negative return.

The best “worst case” year is with the Dalio All Weather portfolio with both the JP Morgan and Research Affiliates assumptions, but it has an unsatisfying most likely return below 5%.  The highest most likely return is the Swensen Reference portfolio, which has a 99% chance of a positive 10-year return under JP Morgan assumptions, but has a small chance of a negative 10-year cumulative return with assumptions from Research Affiliates.

The Bogle 60/40 with global stocks fares well under JP Morgan assumptions, but has a small chance of a negative 10-year return under Research Affiliates assumptions.

The Swensen portfolio and the Bogle portfolio have the highest current yield, which is somewhat of a hedge against the downside, at 2.35% and 2.36% respectively — essentially the same yield.

Overall, for the more risk averse folks who can tolerate returns below 5%, the Dalio All Weather portfolio is attractive.  For those who require return of 5% or more, with limited risk of a 10-year negative return; the Swensen mode is attractive.  The Bogle 60/40 with global stocks is a close runner-up to the Swensen model, but not quite as attractive based on the two sets of assumptions we used.

The 90% stocks approach from Buffet is just fine if you very young and have more than 10-years for “time in the market” to work for you; or, you are wealthy enough to start, and would continue to be wealthy enough with no retirement lifestyle change even if you had a cumulative 5% annualized 10-year negative return.  Otherwise, 90% in stocks may be a bit to aggressive for most investors over 50 to 55 years old, particularly at this time in the market cycle.

Here is a different way to look at the same data.  These next two tables are color coded to show 10-year annualized returns:

  • greater than 7% in dark green
  • from 5% to 7% in light green
  • below 5%, but positive in light pink
  • negative returns in dark pink.

With JP Morgan Assumptions:

With Research Affiliates Assumptions: 

And here is yet one more way to use color coding to gain perspective on the relative merits and demerits of the portfolio models.  All of the returns taken together are coded red, yellow or green on a sliding scale from lowest in red to highest in green.

With JP Morgan Assumptions:

With Research Affiliates Assumptions:

Now for the grand reveal.  What is inside each of these portfolios?  Focus on the category, not just the proxy ETF. We did the projection without any specific security in mind.  The securities we show here as proxies are not the only ones that fit into those models, but they are representative of those that do.

The ETFs used to label the allocation in this next table are simply short-hand to identify the asset classes from the table above.  The numbers are all percentages.

I’d to ask each of you to think about these projections for these model portfolios as general approaches, and tell me which appeal most and least to you as they relate to your Broad Core sleeve (as distinct from the Income Core sleeve and Tactical Opportunity sleeve of your portfolio) to see if we are still on the right track for your personal situation and needs.

Looking forward to our discussion.

 

[symbols identified in this article: SPY, VXUS, VWO, VGIT, TIP, VNQ, GLD, DBC, VGLT, BND]

What Is Asset Allocation For Everybody Else?

Wednesday, April 19th, 2017

(click images to enlarge)

combinedMF-ETF-MMF

AssetAllocation1500ETFs

Equity Market Conditions Assessment & Portfolio Allocation Intentions 2017-03-17

Monday, March 13th, 2017

This note has three parts:

  1. Short summary of our current market view and portfolio allocation implications
  2. Bullet point outline of details behind our thinking in 5 segments (Trend, Valuation, Sentiment, Breadth, Forecasts)
  3. Supporting graphics for most of the bullet points provided in 24 charts and tables.

The short summary is of our market view and intended allocation actions for discretionary accounts, and recommended actions for coaching or “prior approval” accounts

For those of you who want a feel for why we have our market view and why we believe the allocation changes are appropriate; the bullet points will help.

If you want to see what data is behind most of the bullet points, you will want to look at the 24 supporting graphics.

There is an unfortunate need to use some jargon in the bullet points and graphics which may be unfamiliar to some of you, so please call or write in to have any of them explained; and to discuss their significance to portfolio decisions.

THE SHORT SUMMARY …………

Major world equity markets are in up trends — but there is mounting evidence that the US markets are over-extended and significantly vulnerable to a meaningful downward adjustment based on a combination of valuation, breadth, possible turmoil from key elections in Europe; and as Goldman Sachs puts it “rhetoric meets reality” in Washington.

Downside risk exists, but while the trend remains upward, we are remaining invested.  However, we are not committing additional assets from cash positions to US equity risk positions at this time (except for dollar-cost-averaging programs) due to the elevated vulnerability of the US stocks market.  We will be transferring some of the US equity risk assets in portfolios to some international markets that are in intermediate-term up trends that offer better valuation opportunities.

Portfolio changes or recommendations will be framed within the strategic allocation policy level of each client which varies based on individual needs, goals, stage of financial life, preferences, risk tolerance, and other limits or factors.

Based on valuation and long-term forecasted returns, US stocks exposures will transition from the higher end of individual portfolio policy allocations to the long-term strategic objective levels, or a bit below.  We are currently underweighted non-US developed markets and emerging markets allocations, which we will gradually raise to the long-term strategic allocations levels of each individual portfolio’s allocation policy.

Emerging markets have a more attractive valuation level than other non-US international stock markets (although they pose significantly more volatility), and allocation to them may be raised somewhat above strategic target levels within individual permitted allocation ranges.

For determination of intermediate trend status, we relay on our monthly 4-factor indicator. For more information about our trend following indicator and its performance implications, click here to see our descriptive video.

THE BULLET POINTS …………

  • TREND (see Figures 1-6): The intermediate-term stock trends are:
    • United States – UP
    • Non-US Developed Markets – UP
    • Emerging Markets – UP
  • VALUATION (see figures 7-12): Based on history:
    • United States – Expensive on Price-to-Book and Price-to-10yrAvEarnings and not expensive when earnings yield is compared to Treasury yields.  However, when rates rise the comparison will worsen, making stocks more expensive.
    • Non-US Developed Markets – Moderately expensive on Price-to-Book and moderately inexpensive on Price-to-10yrAvEarnings
    • Emerging Markets – Significantly Inexpensive on Price-to-Book and Price-to-10yrAvEarnings
  • SENTIMENT (see Figures 13-17): for US stocks are:
    • Institutional Investors – are reducing equity allocations (a Bearish indication)
    • Investment Newsletter Writers – Bullish at record high levels (a Bearish contra indication)
    • Individual Retail Investors – strongly Bearish this week but neutral last week
    • Options Market – complacent to mixed (jargon terms defined below):
      • Volatility Index – below 200-day and long-term average (complacent, expects smooth ride next 30 days)
      • Skew Index – above 200 day average and long-term trend line (nervous about possible large downside move, next 30 days)
      • Individual Equities PUT/CALL ratio – is 9% above its 200-day and its 10-year average (more cautious that institutional investors)
      • Index PUT/CALL ratio – is 7 % above 200-day average and 4% below its 10-year average (cautious but mixed signal)
  • BREADTH (see Figures 18-21) the trends are:
    • Percent of S&P 1500 stocks in Correction, Bear or Severe Bear have decidedly turned up (Bearish)
    • Percent of S&P 1500 stocks within 2% of their 12-month highs have decidedly turned down (Bearish)
    • The net flow of money is into S&P 1500 stocks over 3 month, 6 months and 1 years, but the leading edge of those flow has turned down
    • The net flow is explained by the net Buying Pressure declining substantially more than the Selling Pressure; and both measures are at levels below the 12-month average indicating reduced overall force driving the market
  • FORECASTS (see Figures 22-24):
    • “Street” consensus 2017 S&P 500 earnings growth 8.9% on revenue growth of 7.2%
    • “Street” consensus 2018 S&P 500 earnings growth of 12.0& on revenue growth of 5.1%
    • Consensus 3-5 year earnings growth for S&P 500 is 8.89%
    • Consensus 3-5 year earnings growth for MSCI non-US developed markets stocks index is 8.76%
    • Consensus 3-5 year earnings growth for MSCI emerging markets stocks index is 10.37%
    • Consensus 3-5 year earnings growth for MSCI core Europe stocks is 8.11%
    • Consensus 3-5 year earnings growth for MSCI Japan stocks is 9.43%
    • Consensus 3-5 year earnings growth for MSCI China stocks is 7.13%
    • Bank of America/Merrill Lynch just raised its 2017 S&P 500 price target from 2300 to 2450
    • Research Affiliates (leading factor based investor) forecasts 10-year real (after inflation) returns:
      • US large-cap stocks 0.7% (with 14.4% volatility)
      • US small-cap stocks 0.5% (with 19.6% volatility)
      • Non-US Developed Markets stocks 5.4% (with 17.0% volatility)
      • Emerging Markets stocks 7.0% (with 23.3% volatility)
    • GMO Bearish Mgr (lowest min fund investment $10 million available) forecasts 7-year real returns
      • US large-cap stocks – negative 3.4%
      • US small-cap stocks – negative 2.7%
      • Large International stocks – positive 0.2%
      • Emerging Market stocks – positive 4.1%
    • BlackRock (fund manager in the world) forecasts 5-year nominal returns
      • Large US stocks 4.1% (with 15.5% volatility)
      • Small US stocks 4.1% (with 18.7% volatility)
      • Large International stocks 5.5% (with 18.5% volatility)
      • Emerging Markets stocks 5.5% (with 23.3% volatility)

Options Jargon Description:

VIX: VIX is the options pricing implied volatility of the S&P 500 index over the next 30 days, based on at-the-money options

SKEW: SKEW measures the relative options “implied volatility” (essentially price) of S&P 500 out-of-the-money PUTs versus out-of-the-money CALLs with strike prices the same distance from the market price – essentially measuring the perceived “left tail risk” (tail risk is the probability of prices going below the level that is predicted by a normal probability Bell curve).

Portfolio managers are predisposed to buy PUTs for protection and sell CALLs for yield, which tends to increase out-of-the-money PUT premiums and depress out-of-the-money CALL premiums.

SKEW of 100 means the market expects equal implied volatility (essentially prices) for out-of-the money PUTs and CALLS.  SKEW greater than 100 means the market expects higher implied volatility (prices) for PUTs relative to CALLS – more perceived large downside risk.

The record low SKEW was 101.9 on March 21, 1991. The long-term average SKEW is around 115, and the high is around 150. The current 200-day average SKEW is about 130, and the current level is about 140. That means there is a heightened concern about a greater than typical risk of a large downside move in US stocks.

INDIVIDUAL EQUITIES PUT/CALL RATIO: The Equities PUT/CALL ratio is the PUTs volume divided by the CALLs volume on individual stocks. This tends to be reflection of actions by retail investors; and is often a contrary indicator.

INDEX PUT/CALL RATIO: The Index PUT/CALL ratio is also the ratio of the volume of PUTS and CALLS, but tends to be a reflection of the actions of institutional investors; and is not considered a contrary indicator.

THE SUPPORTING GRAPHICS …………

(click images to enlarge)

TREND

FIGURE 1: 
Over the past year, US stocks (VOO), non-US Developed Markets (VEA) and Emerging Markets (VWO) are generally in an up trend, although the US is way out front.

2017-03-13_00
FIGURE 2:
Over 3 years, the three regions are up, but the US is way ahead, and did not have as severe down moves as the other two regions.

As you will see the outperformance by the US is related to its current overvaluation, and the weaker performance of the other markets, is related to their more attractive valuation.
2017-03-13_01
FIGURE 3:
Our 4 factor monthly trend indicators ranks each of the three regions as in an up trend.
2017-03-13_02
FIGURE 4:
This time series of our trend indicator for the US shows the up trend established since March 2016.
2017-03-13_03
FIGURE 5:
The up trend in non-US Developed Markets was established in November 2016.
2017-03-13_04
FIGURE 6:
The up trend in Emerging Markets was established at the end of December 2016.
2017-03-13_05

VALUATION

FIGURE 7:
On price-to-book basis the US is very expensive (at the top of its 10-year range). Non-US Developed Markets are “normally” valued (at just above their median level). Emerging markets are inexpensive (price significantly below their median level).
2017-03-13_06
FIGURE 8:
In terms of the Shiller CAPE Ratio (price vs 10-year inflation adjusted average earnings), the US is very expensive relative to is long-term history. Developed Markets are inexpensive, and Emerging Markets are significantly inexpensive.
2017-03-13_07
FIGURE 9:
Based on a variety of valuation metrics the US is expensive. The Developed and Emerging Markets inexpensive by comparison.  Emerging Market have more attractive valuations than the Developed Markets.

In terms of profitability, the US is tops, which partly explains the higher valuation. Developed Markets are less profitable than Emerging Markets.

Emerging markets have competitive dividend yields and the lowest payout ratios.

Emerging markets seem to be a bit less leveraged than US stocks, and the Developed Markets are the most levered.
2017-03-13_08
VALUATION of US ALONE ….

In addition to the price-to-book and Shiller P/E to their respective histories, several other valuation metrics for the US should be considered; almost all of which suggest the market is very expensive

The “equity risk premium” [(stock earnings / price) – (10-yr Treasury yield)] is the only key valuation metric that suggest that stocks may not be overvalued; and that argument depends of the current historically low Treasury yields.

FIGURE 10:

Current equity risk premium for the 10-year inflation adjusted earnings-to-S&P 500 price is 0.90%. Since 1881, the equity risk premium for the S&P 500 and its large-cap precursors was higher 68% of the time.  That suggest modest overvaluation.

However, since the risk premium first went negative in 1964 (except for 4 months in 1929), the equity risk premium was only higher than now 30% of the time — a Bullish suggestion.

The question is whether the 135 history, or the 52 year history is the more important to consider. If the long history is more important, then the S&P 500 is somewhat expensive relative to the yield on 10-year Treasuries; but if the shorter history is more important, then the S&P 500 is inexpensive relative to Treasury yields.  However, Treasury yields are suppressed, and if they normalize to something in the 3% to 4% range before profits increase a lot, then stocks are expensive.
2017-03-13_09
FIGURE 11:
Current equity risk premium for the 12-month trailing earnings-to-S&P 500 price is 1.17%. Since 1881, the equity risk premium for the S&P 500 and its large-cap precursors was higher 68% of the time.  This also suggests moderate overvaluation.

However, since that risk premium first went negative in 1967 (except for 1 month in 1921), the equity risk premium was only higher than now 29% of the time — a Bullish indication.

The question is whether the 135 history, or the 49 year history is the more important to consider. The same logic applies as it does for the equity risk premium based on the 10-year inflation adjusted earnings yield.
2017-03-13_10
FIGURE 12:

S&P 500 GAAP earnings are not much higher than they were 4 years ago, yet the price of the index is a lot higher. That means much of the rise in the price of the index is merely paying more the what you get, not getting proportionately more for paying more.

10-year Treasury rates in 2013 more than doubled from less than 1.5% to more than 3%, yet the S&P 500 continued to rise in price faster than earnings.

Once again 10-year Treasuries have risen in 2016 from less than 1.5% to more than 2.5% and the price of the S&P 500 has continued to rise, even in the face of flat earnings, with falling earnings close behind in the rear view mirror.

If interest rates should make it to 3% in the near-term (not an unthinkable event), the equity risk premium on trailing 12-month earnings would drop from 1.17% to about 0.75%. Since 1881, the risk premium has been higher than 0.75% more than 70% of the time; and since 1967 it has been higher 64% of the time.  That would be Bearish.

This suggests valuation vulnerability in the face of probable moderate interest rate increases.

2017-03-13_11
SENTIMENT

FIGURE 13
The State Street Investors Confidence index is a behaviorally measured sentiment index — a measure of increases or decreases in public equity allocation in actual institutionally managed portfolios, a real measure of market sentiment by large institutions.

The rate of increase in their public equity risk allocations began a decline in around 2 years ago.  They began actually decreasing their public equity allocations in 2016 and continue to do so.

This is not an endorsement of current stocks markets.
2017-03-13_12
FIGURE 14:
Investors Intelligence monitors 100 leading investment newsletters to gauge the Bullish or Bearish sentiment of those writers. Extreme peaks in sentiment tend to be contrary indicators (not perfectly, of course), but when “everybody” is Bullish or Bearish, a trend is often about to be exhausted; because there are few additional people to join the point of view and bring move more money in the direction of the trend.

The current Bull-Bear spread is among the most extreme Bullishness of the last 10 years.  This suggests that a corrective action is likely nearby.

2017-03-13_13
FIGURE 15:
The American Association of Individual Investors conducts a continuous online survey of it members — essentially the retail investor.  Last week when this chart was created, the Bull-Bear spread was 2.29%, barely on the Bullish side of neutral.  In the subsequent week it turned on a dime dropping to negative 16.5%; strongly Bearish.

The chart suggests that this data is more a coincident indicators than a forward indicator, so the drop in sentiment parallels the recent weakness in the up trend.
2017-03-13_14
FIGURE 16:
TD Ameritrade publishes the Investor Movement Index.  It is Bullish at this time.  Here is what they do to make their index.

Each month Ameritrade calculates a short-term Beta (volatility relative to a benchmark such as the S&P 500) for each security.
Then it takes a sample of hundreds of thousands of customer accounts with at least $2,000 in their account and in which at least 1 trade was done the month, from its approximate 6 million customers.
It measures the total equity allocation and the aggregate short-term Beta (volatility relative to volatility of the S&P 500) of the equities in each portfolio (and other undisclosed factors) to develop the risk level of each portfolio.
Then equal weighting each account without regard to size or number of trades, it finds the median equity risk exposure, and puts that on its index scale (scale parameters not disclosed), and plots that versus the S&P 500.
The level and direction of the index is an indication of actual retail investor behavior instead of what they might say about their sentiment.

2017-03-13_15
FIGURE 17:
The options market reveals the actual risk taking behavior of investors in terms of the pursuit of gain (buying CALLs) or seeking protections (buying PUTs).  Here are 4 measures of options market behavior.

VIX measures the expected volatility of the S&P 500 over the next 30 days.  At under 12, the VIX is well below the 10 year average of about 20, and among the lowest levels of the past 10 years.  This is complacency.  Complacency is probably like everybody being on the same side of a boat, which makes the boat prone to tip over.  Volatility is a mean reverting measure, which suggest more volatility in the relatively near future than in the relatively near past.

The SKEW Index (defined above in the jargon section) is a measure of the concern over the size and probability of an unusually large downside move.  That measure is elevated, which gives reason for caution.

The Equity PUT/CALL index (defined above in the jargon section) is a measure of the relative “protection seeking/opportunity seeking” behavior of mostly retail investors.  That ratio is slightly elevated versus average levels, indicating a mildly increased relative pursuit of protection.

The Index PUT/CALL index (defined above in the jargon section) is a measure of the of the relative “protection seeking/opportunity seeking” behavior of mostly institutional investors.  That ratio is slightly lower than average, indicating a mildly lower than average relative pursuit of protection.
2017-03-13_16
BREADTH
This is one of our favorite measures, and one that we directly measure weekly since the beginning of 2014.  Breadth measures the condition or behavior of the overall membership of the broad S&P 1500 index and compares it to the price level of the market-cap weighted S&P 500 index — in other words, it checks to see if the rank and file members of the market are going in the same direction as the mega-cap leaders of the market.

For example, the price movement of Apple and Exxon have a lot more impact of the price of the S&P 500 or the S&P 1500 than 100’s of smaller members of the S&P 500 and S&P 1500.  If the rank and file are going in the same direction as the leadership, that is Bullish breadth.  If they are going in the opposite direction, that is Bearish breadth.  The leaders can only go so far for so long without the rank and file coming along.

FIGURE 18:
The percentage of S&P 1500 index constituents in a Correction or worse (grey line — down 10% or more from their 12-month high) rose dramatically before the November presidential election, then dropped off just as steeply to among the lowest levels in the past 3 years.  Just recently, however, the percentage in Correction or worse sharply turned up — still in “normal” range, but the direction change is a negative for the current stocks rally.

The same is true, but to a lesser extent for the percentage of S&P 1500 stocks in a Bear or worse (blue line — down 20% or more), or in a severe Bear or worse (red line — down 30% or more).
2017-03-13_17
FIGURE 19:
The percentage of S&P 1500 stocks within 2% of their 12-month high was declining prior to the election, turned up sharply to reach the highest level in the past 3 years immediately after the election, but has since declined to the “normal” range with a current downward direction.  The provides a note of caution about the current rally.
2017-03-13_18
FIGURE 20:
The Net Buying Pressure (Buying Pressure / Sum (Buying Pressure + Selling Pressure), which has been net positive since the bottom of the early 2016 Correction, began to decline before the election; resumed growing strength after the election; but has recently been losing steam.  This is not supportive of the current rally.
2017-03-13_19

BuyingPressureMethod
FIGURE 21:

The separate Buying Pressure and Selling Pressure components of the S&P 1500 stocks Net Buying Pressure in the figure above are shown here.

The rising S&P 500 price in 2016 was not matched by rising Buying or Selling Pressure, showing waning enthusiasm for equities.  After the election both Buying and Selling Pressure rose, but Buying Pressure rose more than Selling Pressure.  Recently however, the decline in Net Buying Pressure noted above, is the result of Buying Pressure declining substantially, while Selling Pressure has declined far less.

These data also suggest the fuel of the rally may be running low.
2017-03-13_20

PressureCompnentsMethod
FORECASTS
Looking way down the road with 5-10 year forecasts, and supporting our view that a shift in allocation more toward international equities, and less in US equities is appropriate, are forecasts by Research Affiliates (a noted factor-based asset manager), by GMO ( a Bearish asset manager for the very wealthy — $10 million and up to invest in their funds); and by BlackRock (the largest fund manager in the world).
FIGURE 22:
Research affiliates studies factors with focus on the current Shiller CAPE Ratio (Price divided by the inflation adjusted 10-year average Earnings) relative to its historical median, and historical highs and lows.  They find that to be a good long-term indicator of opportunity.  They view the CAPE Ratio as mean reverting over the long-term.

Based on CAPE and other factors, this chart shows how they see the real (nominal less inflation) return and the volatility of US, non-US Developed Markets (“EAFE”) and Emerging Markets (“EM”) working out over the next 10 years on an annualized basis.  They definitely see international equities as the place to be — but with more volatility.

2017-03-13_21
FIGURE 23:
GMO does not disclose their forecasting methodology, but they are among the most Bearish institutional manager, so worth noting for that.  Over the next 7 years, they see the real return on US large-cap stocks as negative 3+%; the real return on large international stocks as barely positive; and the real return on Emerging Market as positive 4+%.  They also see negative real returns on US and Dollar hedged international bonds, but positive 1+% real returns on Emerging Markets debt.

2017-03-13_22
FIGURE 24:
Over the next 5 years, BlackRock sees nominal return on US stocks as very low, and much lower than international stocks.  They see non-US Developed Markets stocks as generating nominal return  at about the same level as Emerging Markets, but with volatility similar to that of US small-cap stocks; whereas they see much higher volatility for Emerging Market stocks.
2017-03-13_23

 

 

Target Date Funds As Aid In Retirement Portfolio Design

Sunday, October 4th, 2015
  • Investors in or near retirement should be aware of portfolio design leading fund sponsors suggest as appropriate
  • Leading target date funds appear to generally have less severe worst drawdowns than a US 60/40 balanced fund
  • The funds have slightly higher yields than a US 60/40 balanced fund
  • Target date funds have underperformed a US 60/40 balanced fund in part due to a cash reserve component and non-US stocks
  • Non-US stocks drag on historical performance could become future boost to performance.

INTENDED AUDIENCE

This article is suitable for investors who are in retirement or nearly so, and who are or will rely heavily on their portfolio to support lifestyle.

It is not suitable for those with many years to retirement, or those with a lot more money in their portfolio than they will need to support their lifestyle.

SHORT-TERM and LONG-TERM

We have been writing about the short-term recently (here and here and here), because we are in a Correction, that may become a severe Correction, and possibly a Bear.

For our clients who fit the profile of being in or near retirement and heavily dependent of their portfolio to support lifestyle in retirement, we have tactically increased cash in the build-up to and within this Correction, as breadth and other technical have deteriorated.

However, we don’t want to lose sight of long-term strategic investment.  This article is about asset allocation for investors that fit that retirement, pre-retirement, portfolio dependence profile.

WHERE TO BEGIN ALLOCATION THINKING

We think it is a good idea to begin thinking about allocation by:

  1. reviewing the history of simple risk levels (see our homepage) from very conservative to very aggressive to get a sense of where you would have been comfortable
  2. reviewing what respected teams of professionals at leading fund families believe is appropriate based on years to retirement (they assume generic investor without differentiated circumstances).

This article is about the second of those two important review — basically looking at what are called “target date” funds.

Generally, portfolios should have a long-term strategic core, and may have an additional tactical component.  We think some combination of risk level portfolio selection and/or target date portfolio selection can make a suitable portfolio for many investors.

You may or may not want to follow target date allocations, but you would be well advised to be aware of the portfolio models as you develop your own.

In effect, we would suggest using risk level models and target date models as a starting point from which you may decide to build and deviate according to your needs and preferences, but with the assumption that the target date  models are based on informed attempts at long-term balance of return and risk appropriate for each stage of financial life.

For example, an investor might deviate one way or the other from more aggressive to less aggressive based on the size of their portfolio relative to what they need to support their lifestyle, and the size of non-portfolio related income sources; or merely their emotional comfort level with portfolio volatility.

There no precise allocation that is certain to be best, which is revealed by the variation in models among leading target date fund sponsors.  Their allocations are different, but similar in most respects.

FUND FAMILY SELECTION

For this article, we identified the 7 fund families with high Morningstar analyst ratings for future performance (those ranked Gold and Silver, excluding those ranked Bronze, Neutral or Not Rated).

Those 7 families are:

  • Fidelity
  • Vanguard
  • T. Rowe Price
  • American Funds
  • Black Rock
  • JP Morgan
  • MFS

Fidelity, Vanguard and T. Rowe Price have about 75% of the assets in all target date funds from all sponsoring families combined.

ASSET CATEGORIES CONSIDERED

We then used Morningstar’s consolidated summary of their detailed holdings to present and compare the target date funds from each family.  The holdings were summarized into:

  • Net Cash
  • Net US Stocks
  • Net Non-US Stocks
  • Net Bonds
  • Other

While we have gathered that data for retirement target dates out 30 years.  This article is just about target date funds for those now in retirement or within 5 years of retirement.

PROXY INVESTMENT FUNDS USED

We simulated the hypothetical past performance of those target date funds using these Vanguard funds:

  • VMMXX (money market)
  • VTSAX (total US stocks)
  • VGSTX (total non-US stocks)
  • VBLTX (aggregate US bonds)

Admittedly, this is a gross proxy summary of the holdings of the subject target date funds  The funds may hold individual stocks or bonds, may hold international bonds, may use some derivatives, and may have some short cash or short equities. Nonetheless, we think these Vanguard funds are good enough to serve as a proxy for the average target date funds, and as a baseline model for you to examine target date funds and to plan your own allocation.

THE BENCHMARK

As a benchmark for each allocation, we chose the Vanguard Balanced fund which is 60% US stocks/40% US bonds index fund. Figure 1 shows the best and worst periods over the last 10 years for that fund, as well as its current trailing yield.

FIGURE 1:

20151004_Balanced

So, let’s keep the 2.10% yield in mind as we look at the models, and also the 19.7% 3-month worst drawdown, the 27.6% worst annual drawdown, and 7.3% worst 3-year drawdown.

FOR THOSE CURRENTLY RETIRED

Figure 2 shows the allocation from each of the fund families for those currently in retirement.  It also averages their allocations for all 7 and for the top three (Fidelity, Vanguard, T. Rowe Price).

(click image to enlarge)

2015_Funds

You will note substantial ranges for allocations from fund family to fund family.  For example, MFS using about 19% US stocks while Fidelity uses about 38%; and MFS uses about 64% bonds and Fidelity uses about 36%.

The average bond allocation for the 7 families is about 54%, but the top three by assets average about 42%; and their average cash allocation is about 9% versus the top 3 average of 5%.

Figure 3 shows how a portfolio using Vanguard index funds would have performed over the past 10 years with monthly rebalancing if it was based on the average of the top 3 families.

We recalculated the allocations to exclude “Other” which is undefined, but which is relatively minor in size in each fund.

We also note that the Vanguard index funds have a small cash component, so that the effective cash allocation is higher than the model.

FIGURE 3 – Backtest Performance:
(retired now: average of top 3 families)

20151004_2015Av3

Observations: 

  • Yield is somewhat higher (2.28% versus 2.10%).
  • Worst 3 months were somewhat better (-18.4% versus – 19.7%)
  • Worst 1 year was somewhat better (-26.2% versus -27.6%)
  • Worst 3 year drawdown was better (-5.7% versus -7.3%)
  • Underperformed benchmark over 10, 5, 3 and 1 year and 3 months (10 years underperformed by annualized 1.05%).

Reasons For Underperformance:

Inclusion of non-US equities may be the biggest contributor to underperformance versus the balanced fund with 100% US securities.

Another part of the underperformance is maintenance of a cash reserve position that is over and above any cash position within the benchmark balanced fund.  Part is also  due to a higher bond allocation.

Those factors probably account most of the performance difference.  We did not try to determine the exact contributions of each attribute to performance differences.

The historical underperformance due to non-US stocks could possibly turn out to be a long-term reason for future outperformance.

FIGURE 4 – Backtest Performance:
(retired now: average of top 7 families)

20151004_2015Av7Observations: 

  • Yield is somewhat higher (2.19% versus 2.10%).
  • Worst 3 months were significantly better (-12.6% versus – 19.7%)
  • Worst 1 year was somewhat better (-17.7% versus -27.6%)
  • Worst 3 year drawdown was a lot better (-2.3% versus -7.3%)
  • Underperformed benchmark over 10, 5, 3 and 1 year & outperformed over the last 3 months (10 years underperformed by annualized 1.32%).
  • Incurred less drawdown in exchange for lower cumulative return.

FOR THOSE EXPECTING TO RETIRE WITHIN 5 YEARS

FIGURE 5 – Allocation:
(expected retirement within 5 years)

(click image to enlarge)

2020_Funds

Again, we see substantial variation between fund families, and also between the averages for the top 3 by assets and for all 7 of the Gold or Silver rated target date families.

The average bond allocation for the 7 families is about 44%, but for the top 3 it is only about 34%.  For the 7 families the average non-US stocks are about 15%, but for the top 3 families it is about 21%.

FIGURE 6 – Backtest Performance:
(up to 5 years to retirement: average of top 3 families)

20151004_2020Av3Observations: 

  • Yield is higher (2.30% versus 2.10%).
  • Worst 3 months were somewhat worse (-21.1% versus – 19.7%)
  • Worst 1 year was somewhat worse (-30.1% versus -27.6%)
  • Worst 3 year drawdown was the same (-7.3% versus -7.3%)
  • Underperformed benchmark over 10, 5, 3 and 1 year and 3 months (10 years underperformed by annualized 0.95%).

FIGURE 7 – BacktestPerformance:
(up to 5 years to retirement: average of top 7 families)

20151004_2020Av7Observations: 

  • Yield is somewhat higher (2.19% versus 2.10%).
  • Worst 3 months were better (-16.2% versus – 19.7%)
  • Worst 1 year was better (-22.9% versus -27.6%)
  • Worst 3 year drawdown was better (-4.4% versus -7.3%)
  • Underperformed benchmark over 10, 5, 3 and 1 year & slightly outperformed over the last 3 months (10 years underperformed by annualized 1.19%).

PERFORMANCE OF INDIVIDUAL PROXY FUNDS

Figure 8 presents the current yield and rolling returns of the five individual proxy funds used in this review.

FIGURE 8:

(click image to enlarge)

20151004_proxyfdreturns
PERFORMANCE OF THE TARGET DATE FUNDS

FIGURE 9:

(click image to enlarge)

20151004_TgtDateFunds

Symbols for funds mentioned in this article are: VMMXX, VTSAX, VGTSX, VBTLX, VBIAX, TRRGX, AABTX, JSFSX, LFTDX, VTXVX, FLIFX, BAPBX, TRRUX, AACFX, JTTAX, MFLAX, VTWNX, FPIFX, BAPCX