A Revolutionary Take on Morningstar’s Style Box

“If we could first know where we are, and whither we are tending, we could better judge what to do, and how to do it.”

Abraham Lincoln

It must seem mighty strange for a blog about investment products to be ripping off quotes from the great emancipator, especially one that began his famous “House Divided” speech as the nation tore itself apart over the issue of slavery and states’ rights.  Then again, anyone working in the mutual fund side of the asset management business probably feels that it’s a pretty apt comparison to what’s going on in the business today.  Seems every armchair portfolio manager can only talk about underperformance by active managers and why passive index replicators charging low fees are the only way to go, a topic even the DOL enshrined in its recent Conflict of Interest rules.   Most are probably wondering if the situation could possibly get any worse as Morningstar says over $300 billion in AUM fled actively run funds for the year ending 8.31.

Now I’m sure you’re reaching for the world’s smallest violin thinking they have no one to blame but themselves after so many in the business got rich by running what amounted to closet index funds charging high fees which in turn helped spark the ETF revolution and brought the Feds down on them.  Picture the powdered wigs and finery of the French revolution but instead of aristocrats saying “let them eat cake” the masters of the mutual fund universe were saying “fees less than 1% and no 12b-1?  I never!”

powdered-wig_5(Fee’s below 1%?  The dogs!)

Ultimately all revolutions turn on themselves and the shift from active to passive mutual funds and ETF’s is no exception.  That same fund flows report from Morningstar showed passive funds of all types brought in over $400 billion in new assets over the last year with taxable bond funds taking in almost as much as U.S. equity and ratcheting up tensions over whether ETF’s exacerbate volatility.  Parsimonious providers of passive products (say that five times fast) were supposed to be the big winners of the new DOL rules but with the SEC announcing a “sweeping review” of the ETF products and their market influence, it must feel like there’s no way to win in this business anymore.

Fortunately, and I can’t believe I have to say this, Morningstar might have found a way to help actively run funds level the playing feel with a proposed revamp of their style box for alternative funds.  It’s a good place to start a conversation on some long overdue changes but if you want a revolution, they’ll need to tear the place down and start all over again.

Where We Are:

I haven’t been broadcasting a lot of love letters to Morningstar lately, but believe it or not, I still think they offer one of the most comprehensive data solutions out there for anyone who works with mutual funds.  Morningstar Direct was my constant companion when I was a fund analyst; I never bothered much with the graphic packages or asset allocation modules, but Direct could offer up nearly any data point I could possibly imagine.  My productivity skyrocketed by a factor of four when we finally got a second license that gave me unfettered access.

My chief complaint has been with their analyst commentary, most recently John Rekenthaler’s weekly commentary from August 30th (here) where he offers some very sage advice on tactical funds as if it was still 2011.  He takes tactical managers to ask for underperforming the classic 60/40 portfolio, here using Vanguard Balanced Index (VBIAX), not underperforming their respective benchmarks or Morningstar’s various target risk benchmarks.  Seems reasonable, especially after I pointed out that to even be in the tactical category you have to have allocations to bonds, international equities, cash, etc, along with frequent turnover.

He also didn’t take them to task for underperforming on a risk-adjusted basis, which is hard to do with a heterogeneous category like tactical allocation which is a dumping ground for misfit funds.  There are funds in the space, mostly income oriented, that have delivered risk-adjusted returns in line with VBIAX, but instead most commentators focus strictly for not delivering the same level of absolute return as a 60/40 fund, largely on trailing three and five year returns, while managing to avoid saying anything about the category’s 2016’s performance and the number of funds outperforming VBIAX this year.

Most active managers are underperforming their benchmarks but since no one will pay Morningstar thousands for a Direct license to tell them which S&P 500 index fund is best, you don’t want to bite the hand that feeds you by criticizing all active managers so better to go after low-hanging fruit I guess.  I can’t defend the performance record of tactical managers, but given the amount of assets in the space, picking on them based strictly on performance is like throwing rocks at a blind kid.  Active management in general and tactical funds in particular were an easy target after years of underperformance, but with volatility rising and equities relatively flat after two years, maybe it’s time to start thinking more outside the box.

Wither We are Tending:

Fortunately, Morningstar still has its share of forward thinkers and Exhibit A is a recent column, “A New Framework for Analyzing Alternative Mutual Funds” by their alternative fund analyst, Jason Kephart.  Maybe it’s because alt funds are his bread and butter, but he at least seems willing to consider them for what they are, a tool for portfolio diversification, then for what they aren’t, which is an S&P 500 index fund or a generic 60/40 portfolio.  His solution is simple and elegant, taking Morningstar’s ubiquitous style box and reorient it towards their Global Equity Index.  The Y-Axis would measure the degree of correlation between a fund to the index while the X-Axis represents the relative volatility between the two.  Check out the chart below and for a more detailed description, you can read the article here.

4633

Senor Kephart’s example focuses on the more well-known alt categories like long/short, market neutral, managed futures, etc although it could easily be adapted for other purposes.  After all, it taps into one of the most important foundations of modern portfolio theory, the use of assets with low correlations to each other to reduce the overall risk of a portfolio.  Recognizing that the bulk of the volatility of any portfolio is likely to be driven by the equity allocation (unless you’re messing around with 3x levered long bonds and if you are, God help you), Kephart decided that most investors will consider the degree of correlation between a fund and a common and comprehensive equity benchmark, in this case one that includes both emerging and developed stocks with roughly 97% of the world’s market cap according to Morningstar’s literature, as a useful way to approximate the diversification benefit of a fund.

Given that he spends a hefty chunk of the article defending his use of relative volatility versus beta, I’m assuming he’s already gotten some pushback from within Morningstar although he makes a sound argument on why he didn’t use the more common metric.  What he could have done a slightly better job at is explaining is how beta could be irrelevant between two funds depending on your correlation to the index.

At my former shop, we ran a series of tactical programs that were insanely difficult to benchmark since their investment universe was essentially the ENTIRE INVESTMENT UNIVERSE.  Theoretically, healthcare stocks might be the biggest return contributor one year and gold miners the next.  We could have large bond allocations or cash positions so there was no consistency which gave us a very low correlation to the broader market and if you’re familiar with the math behind ordinary least square regressions, that low correlation translates into a very low R2 to the market meaning shifts in the S&P 500 did a very poor job of explaining the variation in our returns.  Our colorful illustrations still showed beta and subsequently alpha, but the fact our correlation to the market was so low made those stats unreliable as an indicator for predicting future performance.

If you want a more graphic example, consider these five funds in Morningstar’s tactical allocation category that have two things in common.  First, they all have the word “income” in their name, likely implying a bond oriented process more focused on current yield rather than capital gains.  Second, they were all outperforming VBIAX in 2016 when I ran the screen in Morningstar.  While my focus was on diversifying a 60/40 portfolio, the relative benchmark for this first graph is MSCI ACWI NR (USD) to test the philosophy behind Kephart’s work with correlation on the Y axis, relative volatility on the X.

tb1

First impressions would make you think it looked very like the efficient frontier you learned (or should’ve learned) about in school although like most things in life, appearances are deceiving.  On one end is relatively low risk SEI Multi-Asset Income A (SIOAX) with relatively high risk Cornerstone Advisors Income (CAIOX) on the other and since this is using trailing 36-month data, I’ll let you guess who outperformed who on this one.  Or you can look at the table.  The other quick takeaway is that the chart is not that useful; CAIOX may have more volatility but that could good or bad depending on the asset mix of your existing portfolio.  You still need a table of returns, details on the existing portfolio and CAIOX along a viewpoint on global asset class returns.

What it is useful for is that it’s tell you Hancock Horizon Diversified Income Fund (HHIAX) and Direxion Hilton Tactical Income (HCYIX) have very similar correlations to the market but significantly different relative volatility which would indicate very different management strategies.  HHIAX is a newer fund with less history to review, but investors would probably be questioning the relative merits of the fund versus other’s in the space for the amount of risk they’re taking on and whether their strategy will outperform going forward.

Now anyone without a strong opinion on global bonds or time to do research on the funds would probably pick SIOAX and move on, theorizing that lower volatility and correlation mean instant diversification.  Well yes and no, the management team at SIOAX has done a great job with a trailing three year Sharpe WAY above that of VBIAX or MSCI ACWI, but that doesn’t tell you squat about what happens next.  Time to dive into holding reports and management updates to get a better feel for their strategy.  Or better yet, just breakout Direct and go to the town.  Sell the sizzle, not the steak.

A correlation of .61 translates into an R2 of .41 and meets the test of statistical significance, which maybe be a pretty big number for financial researchers, but doesn’t provide that much reassurance for retail investors or advisors.  You know that SIOAX has weak relationship to the ACWI, but not much more than that while HYCIX has a higher beta but also a much higher correlation and thus R2 to the index.  If you’re 100% invested in an ACWI replication ETF, you might prefer having that higher correlation and potential higher predictability versus rolling the dice on the relatively unknown qualities of SIOAX.

You Say You Want a Revolution:

By this point you’re starting to wonder what I was so excited about, I build up the idea of this exciting new way to screen alt funds and then point out all the major flaws in the concept.  Remember that most revolutions don’t go from minor disputes over tax policies to public beheadings and mass executions in an hour or less, there’s usually a long ramp up period of public arguments until some yahoo on their soap box (yo!) starts the name-calling and then mass executions can begin.  And since the mutual fund industry isn’t well known for bloodletting (unless you’re involved in the PIMCO-Gross dispute), think of Mr. Kepharts modest proposal like Luther nailing his 95 theses’ to the cathedral door; Luther was looking to start an academic debate, not a bloody revolution.

I think the real value of this proposal isn’t in evaluating funds within the same space, but using it to compare similar strategies across a wide variety of Morningstar’s different categories.  One should wonder if Kephart was working towards the same goal all-along; if you look back at the article, you’ll see that his example includes a jumble of alternative funds across multiple categories, not a handful within the same space.  I know Morningstar has a variety of tools you can use to show the correlation between two funds, but the issue with that is you must know how to search for the relevant funds or have previously identified them as relevant.  A preset map showing all the alt funds is a simpler and more powerful tool if you’re more interested in diversification and volatility rather than a specific investment style.

Still more evolutionary than revolutionary, for Morningstar to unlock the power of this simple tool, they’d need to consider a total revamping of their existing fund category system.  Or rather, they’d need to settle on just ONE system since the last time I checked, there was the category, institutional category and then their “Premium Scanner” tool in the retail subscription has a “role in the portfolio” screening tool.  But instead of overwhelming you with hypotheticals, let’s go back to the example of tactical allocation funds.

Lately I’ve been looking for alternative funds that could be useful for diversifying the standard 60/40 portfolio (here being VBIAX) which means a lot of time spent combing through Morningstar looking at the same funds that Kephart studies but I also subscribe that just about fund can be a diversification tool as long as the correlation is less than 1 so in addition to the alts, I’ve been looking at the entirety of the allocation universe along with target date and retirement income funds.  Screening those categories and looking only at unique strategies (oldest share class) along being open to new money and outperforming VBIAX on a YTD basis gets me to around 360 plus strategies.

Now if you’re like me and you want a manager with a good long-term record, you’re focusing on three and five year returns along with a strong Sharpe ratio which is where things get interesting.  The only funds that have a Sharpe ratio in line with VBIAX did it through having lower volatility which typically indicates higher bond allocations, which should be making you nervous right now.  But another thing I noticed is that there’s an ample supply of fund’s using the word “income” in their name across all these different categories which got me wondering.  Anyone who says you can judge a book by its cover never worked in the ETF industry where funds tend to have extremely descriptive names describing their underlying investment strategy.  Is the same true of income funds?

tb2Well I pulled 15 funds more or less at random with 5 coming from the tactical side, 5 from world allocation and 5 from different parts of the allocation space and you can check the table for more details but outside of the tactical funds you can see a distinct amount of clustering among the other 10.  Compare AAIAX and EKSAX, over the last three years they have very similar correlations to the ACWI not to mention almost the same relative volatility but the key difference is their return with EKSAX leaving AAIAX in the dust over the last three years.  Investors looking for a diversification tool with an income focus would obviously prefer EKSAX (assuming the future looks like the immediate past) but if they only looked at “tactical allocation funds” how would they ever know this?  Why not have income funds with a similar correlation and subsequent R2 to the benchmark in the same place?

Or what about MSMAX and HBLAX?  Two allocations funds with very similar correlations and relative volatilities to ACWI but on different parts of the allocation fund curve.  MSMAX is in the “50%-70% Equity” portion indicating it should have a higher equity allocation and be more volatile than HBLAX in the “30%-50% Equity” subcategory but instead they have almost identical standard deviations over the last three years.  Both funds are actively managed with HBLAX currently having 50% more equity than its category average while MSMAX has less equity and more bond exposure than its category average fund.  Even so, Manning & Napier’s offering (which is available in a higher fee but less expensive S share) is still a highly ranked fund in its space, but a true “apples-to-apples” comparison should be made to HBLAX to find the best manager.

Investors and portfolio builders who are more rigid followers of Morningstar’s category system might not even consider the possibility of using allocation funds as diversification tools, which is why in addition to using Herr Kemphart’s system, Morningstar should consider whether it’s time to completely rethink how they place funds within categories in the first place.

More than a few aristocrats learned the hard way that being a forward thinker and staying relevant by offering a valuable service is a much better way to avoid being first against the wall when the revolution comes.

(For those who are really interested in nitpicking details, here’s the table with all the beta’s, correlations and r2 for the above funds relative to MSCI ACWI)

cor

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s