2012: Now That Wasn’t So Bad, Was It?
Daily headlines provided investors with an abundance of gloom to feed their doubts last year. Take your pick. A weak U.S. recovery. The debt crisis in Europe. The U.S. elections. The fiscal cliff. Investors who acted on those fears missed an opportunity to participate in strong returns across the global financial markets. Major market indexes delivered double-digit total returns. As a group, the non-U.S. developed (international) and emerging markets outperformed the U.S. equity market.
Yet many investors were in cash, earning nothing, scared out of the markets by anticipation of bad events. Sadly, this isn’t the first time investors have been hiding while investments provided handsome returns. Some data from Dalbar Inc.1 tells the story about fear and the typical individual investor.
In 2008 the S&P 500 index lost 37.72 percent, but the average investor lost 41.63 percent. Over a much longer period — from January 1989 through December 2008 (20 years) — the average equity investor earned an annual return of only 1.87 percent. This means the average investor underperformed the S&P 500 index by 6.48 percent and trailed inflation by 1.02 percent
There are some big messages here. One is that investors tend to buy high and sell low. Another is that returns are more dependent on investor behavior than actual fund performance. Still another is that investors who don’t act on news or anticipations — buy-and-hold investors — typically earn higher returns over time than those who time the market.
We believe, along with the behavioral economists, that four major factors drive the sorry decisions that separate investors from the returns they ought to earn.
- Markets are like spaghetti sauce. When it comes to news, “It’s in there.” News can be worrisome, but markets are forward-looking and absorb information very quickly. By the time we read about it in the newspaper, the markets have usually gone on to worrying about something else.
- The economy and the market are different things. News is important to stock prices only if it is different from the information the market has already priced in.
- Forecasted events and markets are different, too. If you invest via forecasts, you need to realize that markets may react differently to those events than you expect.
- Markets are different in other places. There is a variation in the market performance of different countries. That’s not surprising given the differences in each market in sector composition, economic influences and market dynamics. That variation is the rationale for diversification — spreading your risk to smooth the performance of your portfolio.
Don’t get us wrong. We think it’s fine to take an interest in what is happening in the world. But care needs to be taken in extrapolating headlines into your investment choices. It’s far better to let the market do the worrying for you and diversify around risks you are willing to take. We think that’s the job we do here at AssetBuilder. That’s why our new tag line is “Simple investing for the smarter investor.”
We know — and you know — that there is a psychology that affects investing decisions. There is a clash between well-reasoned investment decisions and decisions driven by fear after markets have fallen, or greed after they have been rising. We believe our clients are smarter investors because they know how human behavior can hurt their wealth. They don’t extrapolate news headlines into their investment strategy. They let the market do the worrying. They diversify around the risks they are willing to take.
2012 Performance – Size, Value
So, just how did low-cost rational investing do last year?
AssetBuilder portfolios have a “tilt” toward value stocks and small-cap stocks for a single reason: Thoroughly examined historical data shows that there is a return premium for both small-capitalization and for relatively low-priced stocks. Mr. Market doesn’t always provide that return premium, but last year these factors were rewarded in four of six market segments.
The U.S. Market
- Positive value premium: Value outperformed growth by 2.24 percent.
- Negative size premium: Small-cap stocks underperformed large-cap stocks by 0.07 percent.
Non-U.S. Developed (International)
- Positive value premium: Value outperformed growth by 2.11 percent.
- Positive size premium: Small-cap stocks outperformed large-cap stocks by 1.11 percent.
- Negative value premium: Value underperformed growth marketwise by 4.10 percent.
- Positive size premium: Small-cap stocks outperformed large-cap stocks by 4.43 percent.
The Roulette Wheel of Asset Class Performance
Yes, we would have loved it if the score had been six out of six. But that isn’t real life. In real life, things are probable, but not inevitable. Skeptics should consider the randomness of asset class returns data that we shared last year. We present it again, below, in our “Roulette Wheel of Asset Class Performance,” which demonstrates that telling the future is more than difficult. It is impossible. (See Figure 1.)
Figure 1. The Roulette Wheel of Asset Class Performance
- International Small-Cap Value
- Emerging Markets
- U.S. Small-Cap Value
- U.S. Large-Cap Value
- International Small-Cap
- U.S. Real Estate
- International Large-Cap Value
- U.S. Large-Cap
- U.S. Small-Cap
- Five-Year Global Fixed
- Five-Year U.S. Gov’t Fixed
- One-Year U.S. Fixed
Based strictly on 2012 results with regard to the additional return we get by tilting toward small and value, we remain confident that these factors will continue to add value. But, as you will soon see, we intend to update our models for another positive factor as well as expand the menu of portfolios we offer.
Are AssetBuilder Portfolios “Delivering the Goods”?
We ask this question out of existential as well as intellectual interest. Our original premise was that we wanted to add value by building risk-efficient portfolios with low-cost index funds. By definition, we could not deliver more than market returns for any asset class used in our portfolios. What we could do was try to combine asset classes in a way that clients would get “more bang for the buck” — in this case, more return for any given amount of risk.
Recently, we examined the performance of several of our portfolios against their nearest risk-comparable Morningstar category. We categorize our Portfolio 10 as the most aggressive of the three AssetBuilder portfolios in the “balanced” category. It had a standard deviation over the last five years of 17.97 percent. Over the five years ending Dec. 31, 2012, it had an annualized return of 3.05 percent. That’s a return and risk that thoroughly reflects the terror and disappointment of the last five years.
In the Morningstar mutual fund database, our Portfolio 10 would be classified as an "Aggressive Allocation" fund. This means such funds have more equities than a "Moderate Allocation," or traditional balanced, fund. Over the last five years, the average annualized return of the 270 funds with five-year histories in this group was 0.95 percent, with an average standard deviation of 17.42 percent. In other words, Portfolio 10 provided 2.1 percentage points more return for about the same amount of risk. If we rank Portfolio 10 by its performance against this fund category, it shows up as No. 18 of 270 — better than 93 percent of its group.
Using the same method, we found that our Portfolio 9, with a five-year standard deviation of 15.40 percent, fit comfortably with Morningstar category “World Allocation.” That category had 157 funds with five-year histories and a standard deviation of 14.87 percent. Portfolio 9, however, provided an annualized return of 3.34 percent, while the average return of the World Allocation category was only 1.84 percent. Portfolio 9 would have ranked 36th of such portfolios, better than 80 percent of its competition.
Our Portfolio 8, with a standard deviation of 12.79 percent and an annualized return over the last five years of 3.67 percent, fit the Morningstar category “Moderate Allocation,” which has an average standard deviation of 13.87 percent and an average return of 2.41 percent. That’s 1.26 percentage points less than Portfolio 8. Ranked in the moderate allocation category, Portfolio 8 did better than 84 percent of its competition.
Our Portfolio 7, with a standard deviation of 10.41 percent and a five-year return of 3.87 percent, did only slightly better than funds Morningstar categorizes as “Conservative Allocation.” This group has an average standard deviation of 9.86 percent and an average five-year return of 3.60 percent. Here, Portfolio 7 was a disappointment. It did better than only 53 percent of the funds in this group. We believe the reason for this is that fixed-income holdings in all of our portfolios are skewed toward lower risk, short maturity holdings. As a consequence, our portfolios benefit less from interest rate declines than more typical portfolios. (The figures are summarized in Table 1 below.)
AssetBuilder Portfolios vs. Mutual Fund Comparison Categories
This table examines AssetBuilder portfolio returns against the four major hybrid mutual fund categories in the Morningstar fund universe.
||5-Year Annualized Return
|AB Portfolio 10
|M* “Aggressive Allocation”
|AB Portfolio 09
|M* “World Allocation”
|AB Portfolio 08
|M* “Moderate Allocation”
|AB Portfolio 07
|M* “Conservative Allocation”
|Sources: AssetBuilder calculations, Morningstar Principia 12/31/12
The data strongly suggests that AssetBuilder portfolios are meeting the goal we set for them. They are providing more return for a given amount of risk.
In most time periods a higher risk portfolio will deliver a higher return than a portfolio with lower risk. As you know, this wasn't the case for the last five years. If you examine our posted trailing results you will see that annualized returns declined as risk increased in AssetBuilder portfolios invested for the last five years.
This was not unique to AssetBuilder: Everyone experienced the same thing. When markets panic, stocks fall and bonds rise, so you get a high return for relatively low volatility and a low or negative return for high volatility. The graph below, from our “Growth of Wealth” tool on our website, shows the inversion. AssetBuilder Portfolio 6 declined less in the financial crisis, and the more aggressive portfolios, such as Portfolio 14, have yet to catch up. (See Figure 2.)
Figure 2: Low-Risk Portfolios Win in a Bear Market
Change the starting point for measurement to the last three years, missing the financial crisis, and the results literally flip. Higher risk portfolios provide the higher returns. (See Figure 3.)
Figure 3: In Recovery, High Risk Wins, As Expected
Rather than trying to time the markets, our goal has been to help clients select a degree of risk they can tolerate. While all of us enjoy the periods of rising markets, tolerating the sinking market periods is very difficult. Today we know that the financial crisis of 2008 was a life-changing experience for everyone. Most people entered 2008 with a higher tolerance for risk than they have today. Today, anxiety about the future is at the highest level we’ve experienced in our adult lives.
That’s why we’ve taken a fresh look at our portfolios and decided to make some changes. These changes aren’t dramatic. We’re still focused on risk efficiency, but we’re using new research to find ways to reduce volatility. We’re also offering a new portfolio with still lower risk than Portfolio 6.
Needless to say, we’re not alone in looking for ways to reduce risk while focusing on returns. Indeed, Wall Street continues to respond to every situation as a product opportunity, so we’d like to introduce you to some of the approaches that are likely to make money for Wall Street while losing it for investors.
Danger, Will Robinson! – Wall Street and the Search for Higher Yield
One regularly born-again idea is the high-yield closed-end fund (CEF). It always seems attractive until you realize that the vast majority of the CEFs with yields over 5 percent achieve that yield by leveraging the portfolio. The result is great for fund manager fees, but it means every market cough is pneumonia for fund shareholders.
A less dramatic approach is to build a small side portfolio of high-yield stocks that will boost overall portfolio income — somewhat. But the operative word is “somewhat.” If you committed 10 percent of a portfolio yielding 3 percent to a group of stocks yielding 6 percent, the net result would be an increase in average yield to only 3.3 percent.
Wall Street achieved record junk bond issuance of $350 billion in 2012, with many of the issues carrying fewer protections for bondholders. Investors also flocked to high-yielding alternative investments, such as energy partnerships and venture capital funds.
One Wall Street product that screams danger in 2013 is a complex investment known as an “equity-linked structured product.” Structured anything in the investment world should emit warnings. Equity-linked products link a bond to a well-known stock like Apple (APPL). If all goes perfectly, these structured products pay high monthly interest (10 percent, annualized) for a year and can’t be liquidated before the term of the bond.
In 2012, an amazing 450 new structured products were linked to Apple stock, according to Securities Litigation and Consulting Group (SLCG). Firms such as JP Morgan Chase, Morgan Stanley, UBS and Barclays alone sold more than $722 million. Easy sell … free lunch! Yield based on one of the most popular companies in recent times —Apple!
Consider one specific structured product issued by UBS on Sept. 26. The face value of each note was set at $700.71, almost precisely Apple’s closing price three days earlier. The one-year notes pay monthly interest at an 8.03 percent annual rate, or about eight times what the average short-term bond fund offered at the time. Despite the 2 percent sales fee, investors bought $1 million of the notes. The notes are structured to repay the full $700.71 on Sept. 26, 2013. However, if Apple closes below $595.60 on Sept. 23, 2013, then investors get one share of Apple in lieu of their original principal. For investors to get all their money back, Apple’s shares must climb 35 percent. As of now, investors would lose at least 30 percent, even after counting the income they will earn. The drastic drop in Apple’s stock price to its current mid-$400 level means, according to SLCG: “The vast majority of them are now underwater.”
We believe the record inflows into these Wall Street products should be telling us all: “Danger, Will Robinson!” The combination of unchecked risk appetites, low interest rates and high bond prices presents danger for investors who are pursuing yield.
So what can we do?
Not much. Basically, we’re stuck here as long as the current Fed policy persists. While it does, we can (1) keep our exposure to rising interest rates low and (2) diversify our portfolios as broadly as possible to include asset classes that generate somewhat more income than domestic fixed-income securities or equities. As a practical matter, this is exactly what all of our AssetBuilder portfolios do — we have very limited maturities on the fixed-income side, and our international diversification raises dividend yield a bit.
Does that answer the problem of needing a 4, 5 or 6 percent annual spending distribution? Sadly, no. But, as noted, we have found that most of the offered “solutions” are burdened with a great deal of risk — more risk than most investors would take if they understood how large it actually was. There are, however, some things we can do and we’re doing them.
Model Portfolio Changes
We’re dropping commodities from our investment models.
We had used the exchange-traded fund DBC as our investment proxy for the commodity asset class. Our reasoning at the time we included it was based on three factors: (1) diversification, (2) a hedge against future inflation, and (3) the influence oil has as the global currency. We also liked the liquid approach to gaining commodity exposure without inconveniences of direct ownership such as storage, shrinkage and insurance.
We chose DBC because it tracked the Deutsche Bank Liquid Commodity Index (DBLCI). The main difference between DBLCI and other commodity indexes was the number of commodity components — seven for DBLCI and 19 to 24 for other commodity indexes. Three years ago, DBLCI doubled its commodity components to 14.
We are dropping our exposure to commodities as an alternative asset class based on three factors: (1) It has become much more equity-like in its behavior, reducing its ability to hedge equity risks. (2) Costs and liquidity make it less desirable—we never liked its 85 basis-point operating expense. Nor did we like the three-day hold before cash distribution on sell orders. And (3) the expected return profile is significantly below equities because the return potential is tied more to the inflationary environment than equity market fundamentals.
We’re introducing a new portfolio: AB Portfolio 05.
This portfolio has 20 percent in equities and 80 percent in fixed income. It extends our range of capital preservation portfolios. Portfolio 05 keeps even more of your monies in fixed income — 80 percent — where we take little to no risk. We try and place all of our risk in equities where we have the potential for a higher return.
We’re introducing a new tool for reducing volatility and capturing return for all of our portfolios
New research by Robert Novy-Marx (2012) has identified a way to capture a dimension of expected returns related to profitability. Novy-Marx found that profitability, as measured by gross profits-to-assets, has roughly the same power as book-to-market in predicting the cross section of average returns. While this changes a very basic premise in the Fama/French value model, controlling for profitability dramatically increases the performance of value strategies, especially among the largest, most liquid stocks. And controlling for book-to-market ratios improves the performance of profitability strategies.
Another way of explaining this is that strategies based on profitability are growth strategies that also provide an excellent hedge for value strategies. Adding profitability on top of a value strategy reduces the strategies’ overall volatility. According to Novy-Marx, both profitability and value strategies generally performed well over the sample time periods, but both had significant periods in which they lost money. However, profitability generally performed well in the periods when value performed poorly, and vice versa. As a result, the mixed profitability-value strategy never experienced a losing five-year period (data set July 1963 – December 2010).
Dimensional Funds has aided in this new research and has identified a way to capture a dimension of expected returns related to profitability. Dimensional has made major advances with research in this area and has found a robust proxy that can be applied to portfolio management.
We have stayed in tune with this research and the recent announcements from Dimensional Funds. We’re excited to bring this new key ingredient to our clients and will be incorporating the profitability strategy into our AssetBuilder portfolios.
We will be introducing model portfolios in distribution.
For most of our lives we accumulate assets. But when we retire, our portfolios go into distribution, distributing interest, dividends and assets as needed. Research has shown that we can make our money (and income) last longer by annuitizing a portion of our assets
Annuitized income doesn’t increase the yield of a portfolio. It defines a guaranteed income stream and works to reduce the demand on the remaining portfolio for distributions in the first years of retirement. That, in turn, increases the odds of portfolio survival. If you need a relatively high annual withdrawal and don’t have a pension, this solution may work for you. We believe this is a better solution for the income problem than reaching for higher yields and taking higher risks.
From Individual Portfolios to 401(k) Plans
We built AssetBuilder (AB) on the premise that costs and transparency matter to the individual investor. We also felt that all but the largest investors were ignored or disadvantaged when dealing with traditional financial firms. Soon, we will be offering our low cost and transparency to a market that sorely needs it: 401(k) plans for professionals and small businesses.
The legacy retirement plan providers have been overcharging employees (and their employers) for decades — creating a huge drag on returns. But the government has introduced new rules and regulations requiring transparency and fee disclosure. We believe this has created an opportunity for AssetBuilder.
Defined-contribution plans fall under the provisions of the Employee Retirement Income Securities Act of 1974 (ERISA), regulated by the Department of Labor. Since Nov. 14, 2012, plan sponsors had to personalize all fees for the employee/participants. Some may learn that they paid as much in fees as they got in return.
The most excessive 401(k) plan fees exist in the small business sector, an area dominated by insurance companies with armies of brokers. Those brokers work on commission. They have an incentive to push mutual funds with higher fees — fatter margins from which to pay commissions — instead of the best or lowest-cost funds for employees.
The AssetBuilder Cost Advantage
One way to demonstrate the opportunity we have in an era of full cost disclosure is to compare costs with a major legacy vendor. Boston-based John Hancock is the No. 1 full-service provider to 401(k) plans.
Let’s take a look at the cost to the employees of a medical partnership group with $3.5 million in plan assets. Plan expenses can be divided into three distinct categories:
- Portfolio manager and wrapper;
- Fund expenses; and
- Record keeping.
The Hancock plan has manager and platform expenses of 60 basis points. This is more than double the 25 basis points AssetBuilder charges for an account of the same size. So we have an immediate cost advantage of 35 basis points.
Even more is lost when fund expenses are compared because the majority of funds offered in traditional plans are managed funds that are burdened with substantially higher expenses than index funds. The average moderate allocation fund in the Morningstar database, for instance, has an expense ratio of 0.98 percent, net of 12b-1 expenses.By strategically partnering with Dimensional, we can offer a complete asset allocation solution (AB Model Portfolio 08) for 0.33 percent a year, or 33 basis points. The 65 basis-point difference in costs brings the total cost saving to 1.0 percent a year, or 100 basis points. (See Table 2.)
Sadly, the smallest plans pay the highest costs. According to a recent study, smaller 401(k) plans get hammered on fees. Plans with $1.25 million in assets paid an average of 1.42 percent in investment fees, somewhat less than the 1.58 percent of the Hancock plan being discussed.
At 0.58 percent, the AssetBuilder total investment costs will be a fraction of the comparable cost for much larger plans. The average investment cost for $50 million plans with 5,000 participants, for instance, is 1.20 percent.
Table 2: AssetBuilder vs. The Traditional Firms
This table compares costs for the two largest costs in the operation of small company 401(k) plans.
|Portfolio manager & Wrapper
|Sources: Morningstar, Plan documents
The third element of costs, record keeping, also works against small plans. According to “The 401(k) Averages Book” cited above, plans with $1 million to $5 million in assets can figure on costs between 40 basis points and 22 basis points for plans with 200 participants.
Dimensional Funds will provide record keeping services for 30 basis points. We believe our “all-in” cost for smaller plans will be about 88 basis points. That’s nearly half the cost most small plans face. It’s also far less than most plans that are 10 times larger.
And, Finally, a Solution for the Fiduciary Problem.
Many brokers, consultants and advisers create confusion in the market by positioning themselves as “investment representatives” to a plan. A careful reading of their contract documents reveals that they disclaim any fiduciary responsibility for the plan. As a consequence, the entire fiduciary liability falls on the business owners who sponsor the plan.
ERISA Section 3(38) is an “investment manager” and is a fiduciary because they take discretion, authority and control of the 401(k) plan’s assets. ERISA provides that a plan sponsor can delegate the significant responsibility (and significant liability) of selecting, monitoring and replacing investments to the 3(38) investment manager fiduciary.
AssetBuilder is offering our services as a Section 3(38) investment manager, and we do this without additional charge.
The AssetBuilder Premise
Our nine AB Model Portfolios represent the AB investment strategy. We match client investment goals for risk, expected return and investment horizon with these portfolios. We are not market timers. We do not believe anyone can see the future. We believe deeply in our low-cost index investing strategy.
In their seminal work, Brinson, Hood and Beebower (1986) estimated that over time 90 percent of the variance in returns of a typical portfolio is explained by the variance of the portfolio’s asset allocation. Ibbotson and Kaplan (2000), among others, confirmed this important finding supporting the notion that strategic asset allocation is the most important decision in the investment process.
AssetBuilder (AB) uses Harry Markowitz’s Nobel Prize-winning mean-variance optimization as a quantitative tool to construct our asset allocation portfolio construction strategies — our AB Model Portfolios. Mean-variance optimization is one of the cornerstones of modern portfolio theory. Over the last half century it has become the dominant asset allocation model. The procedure maximizes expected return for a given level of risk, or equivalently, minimizes risk for a given return.
Unfortunately, without “taming the optimizer,” the optimization process tends to produce highly concentrated asset allocations, especially with the most recent best-performing asset classes. AssetBuilder employs two additional techniques to tame the optimization process that result in well-diversified asset allocations: the Black-Litterman model and resampling mean-variance optimization. The first approach focuses on building capital market expectations that behave better within the optimizer. The second is an attempt to build a better optimizer.
To date we have gone a step beyond market capitalization-based indexes by taking advantage of higher returns to small-cap and value-priced equities. The book-to-market (a value measure) effect was popularized after several studies have shown the rationality of using it — well described in the Fama and French research (1993).
In 2008, we employed a great deal of historical data and the methods described above to build a forward-looking investment strategy. Since that time our model portfolios have not changed. It is not for lack of intense analysis … we have revisited our portfolios many, many times. But the final question in each visit quells most change: “Is this a long-term investment strategy based on statistical proof, or is it merely noise?” With the introduction of the Novy-Marx gross profit to assets factor, we believe we’ve found another way to add value and reduce risk to our client portfolios.
Does all this effort produce the perfect portfolio? Sorry, no. Both the back-testing and the actual operating data, however, show that it is possible to deliver greater risk-efficiency.
Dalbar Inc.,“Quantitative Analysis of Investor Behavior 2009”
If you’d like to review the ideas behind our portfolios, we’ve added a quick summary at the end of this document. It begins on page 19.
You can do this on our website at: http://assetbuilder.com/our_portfolios/assetbuilder_results
You can see how our portfolios have done from a start date of your choice at this link: http://assetbuilder.com/growthofwealth/growth_of_wealth.aspx
It could be done with a package of securities focused on telephone stocks, utility stocks, MLPs and REITs. Even so, we believe the average yield would likely be less than 6 percent for the package. Verizon, for instance, currently yields 4.5 percent; Equity Residential (the largest apartment REIT) currently yields 5.5 percent (taxable at regular income rates); and Kinder Morgan Energy partners yields 6.0 percent. So putting that 6 percent together is far from a slam-dunk. Worse, it’s likely to make your tax return pretty complicated; witness the K-1 filings you’d have to do for Kinder Morgan.
Scott Burns, “Annuity Income May Increase Portfolio Survival,” 02/26/2002, http://assetbuilder.com/scott_burns/annuity_income_may_increase_portfolio_survival
In traditional distribution system plans, 12b-1 charges are typically credited against the broad charge for the service.
Scott Burns, “OK, How Much Is Your 401(k) Plan Costing You?” 10/12/2012. http://assetbuilder.com/scott_burns/ok_how_much_is_your_401(k)_plan_costing_you
Record keeping costs decline rapidly as the number of participants declines so we have to consider this figure less indicative than the other two categories.