Originally published on Nov. 16, 2014
I figured now was the perfect time to play catch-up and finally publish the once promised and often delayed posting on how to spot my least favorite category of fund managers; closet indexers.
There is literally nothing that upsets me more than people who take your money for a promised service and then fail to deliver. It's happened to me dozens of times in my personal and professional lives and it's always an infuriating experience. Investing in mutual funds is in an easier and more liquid way to invest with some of the greatest money managers in the world despite what you may have heard from any number of investment "professionals" you see on the nightly news. Finding myself increasingly as one of the few who's still willing to defend mutual funds and active management, I decided it was time to man up and try to arm you with the tools you need to spot managers who whisper into your ear about their great process only to hand you lackluster performance. Of course, anytime you have a system that involves trillions of dollars, vague benchmarks and misaligned rewards systems, you have the potential for a handful of managers to abuse the public's faith in the system.
How does it go? Well first, you've got cash and you need to invest it and being of sound mind, you search for a great mutual fund, do your research (or often your adviser claims to have done it for you) and you invest in a fund that promises NOT to outperform a particular benchmark but simply that it has a system it believes adds long-term value. Looking at some very glossy and sexy handouts, you make your investment (hopefully without paying a load of some kinds) and in your naivety feel it's okay to pay an "average" fee for the potential to outperform. What happens then? Well you most likely forget about the fund until it's time for your annual review or if your fund family is hopeful, you forget you have the account until it's time to retire. That's when you discover this top-rated fund managed to deliver hopefully "blah" performance in line with the median fund in the category and soaked up about 1% to 1.5% of your annual performance doing it.
Now I haven't come to bury active mutual fund managers but rather to sing their praises and I'll be the first to admit, managing money is not easy. Even when you have a well-defined benchmark with a specific subset of stocks to choose from, people are always demanding more and if you fail to meet the benchmark for one to two years, you'll probably find yourself out of a job and hence the origin of the expression "career risk" and why managers in certain categories tend to cluster, all holding the same securities in similar amounts lest they get left behind. When it comes to managing a mutual fund, there's no sin in being the average performer, but if you find yourself in the lower quartile once too often, you'll be managing money for the "Widows of Retired Circus Freaks Retirement System" and guess what that pays?
So how do we go about spotting a closet indexer? First we have to take a deeper dive into the terrifying math behind the concept of "Tracking Error!"
Tracking error is actually a fairly simple concept that even a liberal arts major like myself can understand. Before we get into the basic formula, remember that every mutual fund has a benchmark, which for the lay people in the audience is simply an index that the manager of the fund has decided best represents his universe of available investments and has accepted as a valid metric to compare his performance to. For the rest of this paper, we'll be looking at the universe of managers that Morningstar has pegged as "large cap blend" managers where the typical benchmark is either the S&P 500 Total Return index or the Russell 1000 Total Return Index. All the managers we will be reviewing have the S&P 500 TR listed as their primary benchmark and came from a Morningstar Direct list of 309 mutual funds where the only search parameter was their category and oldest share class. For the sake of simplicity, all share classes other than "A", "Investor" or the lowest possibly minimum investment for that fund have been eliminated.
Tracking error is simply defined as the difference between the return of the portfolio in a given period minus the return of the benchmark. Using monthly returns, this would be represented as (Rp-Rb) but the number generated is typically useless by itself, so we look at the tracking error over a longer period of time. For example, if I had sixty monthly returns for the XYZ Fund and the S&P 500, I would simply calculate the tracking error for each month, find the standard deviation and then multiply it by the square root of 12. This gives me a more practical number that can be compared to funds across the same segment of funds to determine how closely they track their index.
It's easier to understand tracking error by looking at a basic example and since the Yinzer Analyst is about as basic as it gets, this should work for us:
Sticking with the Large Blend theme, what if you could actually have invested $100 in the S&P 500, reinvested those dividends and resisted the urge to up and bail in the summer of 2012 like everyone else? Well as you can see below, 2013 was a very good year for the large cap stocks and you would have ended the year up 32.41%. Not too shabby, except for the fact that you can't actually invest in an index but have no fear, Vanguard is here. No, I don't work for Vanguard (although I do invest in their funds), I just like stupid rhymes.
So if you had purchased $100 worth of the Vanguard 500 Index Fund (MUTF:VFINX) you would have earned the return of the S&P 500 minus about 17 bps for expenses. But those 17bps, while on the low end of the fee scale, introduced a certain amount of tracking error into the process. For the sake of simplicity, let's assume the .17% or 17bps fee is deducted monthly in equal installments throughout the year instead of quarterly or annually.
Now if you looked at our table, you'd see that each month a few cents are deducted from your account as the fees are liquidated and paid to Vanguard for its trading and record keeping. By the end of the year, you'd see both a slight difference in performance and minor tracking error from the fee liquidation. So tracking error isn't a negative, it's an inevitable part of the investing process. Anytime you look at the difference between your investment returns and a particular benchmark you have referenced your investments against.
Now what if you decided to come to The Yinzer Funds (Ticker: YINZX -trademark pending) to invest in our new Large Cap Core Super A Awesome Fund where we charge an industry average fee of 1% or 100 bps annually and then just take your money and index it to the S&P 500? Well, to reuse the same tired chart, you'd see the following:
Yes, literally all of the tracking error in these last two examples comes from fees, not from having holdings that aren't part of the S&P 500 or over/underweighting sectors/individual securities. Literally, just whacking money out of your account (often to repay your adviser for investing with me in the first place) gets you a tracking error of 1%.
Now this was a very simple exercise to show you how to calculate tracking error, in practice it's much more difficult because you first have to establish what the appropriate benchmark for the fund is. You would think that fund companies have a vested interest in presenting you with the most relevant and pertinent information but truthfully, they like having the process be opaque. As I mentioned before, there are two indexes that are commonly used for benchmarking large blend funds, but how do you determine which to use?
You can find everything you'll need to do a quick-and-dirty check for closet indexers on the free side of Morningstar.com. First type VFINX into the quote field and then look at "Ratings and Risk" and scroll down to MPT Statistics. You can see here that the Best Fit and Standard Index both use the S&P 500 and the R2 (how much changes in the index affect changes in the mutual fund) is 1, indicating that all of the change in the fund is due to changes in the benchmark. The beta is also 1 and the alpha is slightly negative, exactly what we would expect from an indexed mutual fund.
Now I understand that many of you reading this have no interest/desire to actually download returns and calculate tracking error manually, so is there a quick and dirty way to get it from a website devoted to mutual fund information like Morningstar.com? Nope and there's a great reason why. Tracking error is half of what's needed to derive the information ratio, which is simply the active return of the fund (Rp-Rb) divided by the tracking error. It's one of the most powerful tools used to analyze how effective a manager is at adding value against a benchmark, so logically the last thing they would want is for it to be something you could easily find on any website. These tools are available either by downloading the returns of the mutual funds from Yahoo Finance, sometimes from the fund families themselves and to those who subscribe to the more advanced Morningstar systems, Factset, Bloomberg, etc.
Now that we know what tracking error is as well as some of the more common tools for analyzing mutual funds, let's start tracking down our closet indexers.
So how can we use tracking error to sort out the true active managers from the closet indexers out to collect your fees for below average performance? First I re-screened my list of 309 mutual funds in the large blend category looking at their five year tracking error versus the S&P 500 TR index and pulled together a list of true passive index funds as an example of how they would appear.
You can see a few common traits that are easy to find on Morningstar.com:
- They actually have index in the name. Very obvious but always start by eliminating the obvious!
- Holdings: They typically have the same number of holdings as the index
- Average Market Cap: Their average market capitalization is close to the index
- Fees: Typically very small; index replication is pretty easy and a very competitive market.
- Risk Statistics: If you go to Morningstar.com, type in any of these tickers and then go to Risk, you should look at their Alpha generation. It should be slightly negative as their simply replicating the index minus their small management fee. Beta should also be close to 1.
Two obvious questions that stand out are why is PIIAX's tracking error/fee so large relative to the other 4 and why does PREIX have a large number of holdings? Large banks typically offer their own index funds as a convenience to clients and often will charge a higher fee as the client prefers the convenience of having all of their assets at one institution to opening a separate account to save on fees (and potentially losing the savings in ticket charges). PREIX might be a case of semi-active management that we'll be discussing later on.
And now that you can spot the index funds, here are a few examples of true active managers that I culled by picking 4 managers with a relatively high R2 to the S&P 500 and tracking error below 10, 1 manager with a lower R2 and a high tracking error that indicates a misfit benchmark.
What stands out? Typically they have smaller and more concentrated positions. Second, their performance is all over the map. Remember, just because they have higher tracking error doesn't mean they actually performed any better; just that these fund managers actually try to outperform and you'll have the opportunity to stand out more from the pack. Doesn't mean they're any good at it. Once you have a short list of funds, you have to go beyond the quantitative to the qualitative. USA Mutuals Barrier Fund (MUTF:VICEX) is more commonly remembered by its former name, the VICE Fund. Supposedly focused on companies in very specific industries, it really screened fund based on a strong name brand and prohibitive barriers to entry or other economic moat.
So now that we know what to search for to spot index funds (again besides looking at their names), how can we spot closet indexers? First we need to delineate closet cases from what are probably best called "semi-active" managers. Semi-active managers ("semi") represent a perfectly valid strategy where they tend to hug the benchmark but will have small, active mis-weights against the index. For example, they might benchmark 90% of their assets to the index but overweight healthcare stocks while underweighting energy, which would have helped them be in the top tier in 2014. In fact, Hartford Disciplined Equity has followed almost that exactly playbook and is nearly in the top decile for the trailing 1, 3, and 5 year periods. Their goal isn't to drastically outperform the benchmark but to generate a small amount of alpha over time that should compensate investors for their fees plus a small return.
- The semis typically have a slightly lower R2, slightly less negative or even positive Alpha and slightly higher tracking error. The first three managers all meet this criteria so I classified them as SA for semi-active.
- ClearBridge Appreciation might seem to be a borderline case of closet indexing, but if you look closely at its beta and average market cap, you'll see that it's really a giant core fund being shoved into the Large Blend category by Morningstar. For the amount of systematic (market) risk the fund has taken on, they've nearly generated enough of a return to compensate investors (less negative alpha).
- The next four funds all meet my standard for closet indexers. High R2 to the index, High Beta, negative alpha and greater than -1 and most importantly, their tracking error less net expense ratio is typically around .5 to .6. These funds…are closet indexers. PNC Large Cap recently had a change in management in October of 2014 and hopefully their new team will be taking the fund in a more active direction.
The worst of the managers still in the closet is the Spirit of America Large Cap Value Fund, a fund so awful it's in the lowest quartile on a 3 and 5 year annualized basis. Of course, with all the difficulties active managers have had in the last year, it's actually only trailing the S&P 500 TR by 127 bps in the last year. This fund represents almost everything I hate about closet indexers but if you're wondering why you've never heard of the fund before, it's because SOAVX is only available on one platform, the David Lerner Associates; a Long Island based broker dealer with around $9 billion in assets (according to their website). And why would they use this fund? Spirit of America Funds was created entirely to provide products to David Lerner Associates which is probably the reason behind the high expense ratio and above average 12b-1 fee of 30 bps.
So to recap, why do I hate this fund? It charges high fees for essentially indexed performance and was designed and marketed solely to retain assets of clients who probably don't know any better and never check their portfolio statements to see how much they're missing out on. Over the last five years, VFINX has returned 15.53% annually versus 12.03% for SOAVX and that may not sound like a big difference; think about what that 3.5% difference would look today. If you have invested $10,000 in VFINX you'd have $20,581 through the market close yesterday compared to $17,647 with SOAVX. That's over $2900 you would have missed out in just fees!