How do you define outperformance by stock funds?
Portfolio managers want to outperform their benchmarks. There’s no question in my mind about this. But how much of an 
advantage do you need before you can claim outperformance?
Outperformance for stocks
To keep things simple, let’s focus on portfolios investing in stocks.
Is it okay to claim outperformance if your return exceeds the benchmark’s by more than 1 basis point (0.01%), 25 bps, 50 bps, or 100 bps?
Or should the margin be calculated relative to the benchmark’s return? After all, exceeding the benchmark’s return by 26 basis points (0.26%) looks better when the benchmark returns 0.01% than when it returns 45%.
Please answer the poll in the right-hand column of this blog. I’ll report on the results in my February e-newsletter.
Diverse opinions on “outperform”
I’m literal-minded. To me, a fund “outperforms” when it beats its index by the tiniest margin, though I doubt that I’d crow about that. However, asset management companies often report such returns as “in line with” or “closely tracking” the benchmark. The concerns of their legal or compliance departments probably influence this decision.
Here’s one example:
…the Wasatch Heritage Fund posted a return of 6.22% for the quarter. These results closely tracked those of the Fund’s benchmark, the Russell 1000 Value Index, which returned 6.78% over the same period.
Meanwhile, some managers–including the manager of the Wasatch Global Science & Technology Fund–question whether their returns should be compared to benchmarks.
Typically, the first paragraph of our quarterly letter to shareholders includes a statement regarding the Fund’s performance relative to its benchmark. We intend to move away from this approach beginning with this letter, as
we think the industry norm of tracking performance versus a broad index on a quarter-by-quarter basis distracts from the Fund’s long-term investment strategy. Our new mantra, forged by the pressure of the 2008–2009 credit crunch, is that we must invest “away from the market” as we attempt to deliver exceptional long-term returns.
I’m looking forward to learning what YOU think.
Dec. 27. Oops. I made a miscalculation in discussing the Heritage example, so I’m deleting the offending sentence thanks to David Lufkin.
Outperforming for a year is a silly concept to me. If I were reviewing for a client, in order for a manager to say they had outperformed for a year, I’d like to see them outperform by at least 1 standard deviation (1.5 would be better for 1 year only though), putting them in the top 30% (theoretically). To really be outperforming, I’d like to see them outperform by at least 1 standard deviation 3 years out of 5 and cumulatively over the 5-year period.
John,
Using standard deviation provides a different perspective. It takes the scale of outperformance into account.
Thank you for your comment!
Technically, a mere 1 bp excess return should arguably count for “outperformance.” I you take into consideration the fact that the portfolio is subject to transaction costs while the index isn’t, one might even say if you were equal to the index you outperformed, but this isn’t likely. We tend to focus on specific time periods (2010, 4th quarter, December) and each is independently examined for “outperformance.” Typically, we find managers outperforming in certain periods and not in others, but in each case we only require that their return exceeds the index’s.
Thank you, David! It’s good to get the performance professional’s perspective on this issue.
Susan good post. I agree with the comments that fund performance needs to be measured in the context of risk and over a period longer than one year.
Thanks, Roger.
It looks as if the readers’ perspective has a big impact on how they define outperformance.
In the Heritage example, the fund UNDERperformed the benchmark by 56 bp? I’ve found that fund managers like the index benchmark when they aren’t highly competitive against peers, and they need to beat at least half their peers to want to mention it.
We constantly battle with marketers who want to cherry-pick performance and at the end of the day we need a positive spin; poor performance will soon improve, good performance will continue.
Oh my goodness, thank you for catching my mistake, David!
You’re right about managers liking the index benchmark when their peer group comparisons don’t look as good.
Leaving aside the degree of outperformance, two baseline criteria are also required: 1) choosing the proper benchmark (i.e. “best fit” index) and 2) having a very high (mid-90%) R squared.
I believe that a stock, fund, or variable annuity that outperforms a benchmark by any margin, no matter how small, it can claim outperformance. But in my opinion, to say that a fund “closely tracked” a benchmark when it underperformed by 57 basis points is potentially misleading. The misleading nature of the claim could have been mitigated by including an expression such as “net of fees,” “without sales charges,” or “before fees and sales charges.” If a portfolio manager tells a client that his fund performed “closely in line” with the benchmark when his return is 8.39469% lower than the benchmark, the client might have reason question the credibility of other claims regarding the fund’s performance.
Steve and Jeff,
Thank you for your comments.
I think you’re both saying that it’s tough to say whether a fund “outperformed” if you don’t know the context.
You can find more lively conversation on this topic on the Financial Writing/Marketing Communications group on LinkedIn (membership required).
Well, how do you define speeding? Easy, you measure the speed of the car against the posted speed limit. The real question is “how do you go about figuring out an appropriate speed limit?” Once you determine that, speeding is measure relative to that. If you go 55mph in a 35 that is too fast, but go 55 in a 65 and you may not get to where you are going on time. Investors can get by with going 70 in a 65, but much faster than that and you risk a ticket or worse, an accident.
Portfolios should always be the output of a complete and comprehensive financial plan. The financial plan will dictate the return required to meet goals. That’s the speed limit.
With that said, I define outperformance as this – when an investor’s portfolio does better then the return required by the financial plan to meet the investors goals – that’s outperformance.
Moderately outperforming the return required in a financial plan is probably ok – most investors can get away with that safely from time to time. It’s when your outperformance is like going 95 in a 65 that investors have a problem. How many investors experienced a ticket or a wreck in their portfolios in late 2008? Or better yet – how many advisors sat in the back seat of the car and let their clients drive 95 mph…drunk!
@DavidBArmstrong on Twitter
Dave,
I love the speeding analogy! Thank you for your thoughtful comment.
I think the yearly portfolio [ here I mean all asset classes- not just equity]review for each client has to be judged against whether we [as wealth managers/financial planners] are on track to achieve his goals[short, medium and long term] and is it in sync with his risk-return profile that was arrived upon at the start of the year.
Partha,
Yes, for individual investors, progress toward their goals is critical.
I don’t think you can claim “practical” outperformance between a fund and its benchmark until it is analyzed on an after-fee basis. The fee analysis should extend to the benchmark as well because investment in a benchmark isn’t costless.
For example, if you have a fund with a 50BPs expense ratio being compared with the SP500 (which investors can access via ETF with an 8BPS expense ratio), you should subtract the fees from both numbers to get an accurate view of relative performance by the manager.
In this case, the fund would need to outperform the SP500 by 42BPS to claim “outperformance” over the benchmark. I’m surprised that mutual funds and portfolio managers (especially in high benchmark cost asset classes like small cap and emerging merkets) haven’t latched onto this. It would make their funds’ performance look better.
Frazer,
Thank you for adding your voice to the discussion! Your suggestion makes sense from the viewpoint of the investor.
I imagine that SEC reporting requirements get in the way of your suggestion to compare funds with an ETF rather than the benchmark. Also, the data collection might be a nightmare.
“How many bps are required to say you outperformed?” The answer is: “It depends.” If you’re running an equity portfolio, then 100 bps is probably a minimum hurdle rate, but if you’re running a money market fund then 25 bps seems like quite a lot (if you are truly staying within your mandate and not taking inappropriate types or levels of risk.) So, the expectations for outperformance will be proportional to both the risk and the market return of your mandate.
That said, outperformance is really not about return; it’s about having more money than you need to meet your tangible financial goals. We invest money in order to have more money. We don’t invest to beat benchmarks; that’s just an abstraction. As fiduciaries, we need to start thinking in terms of our loyalty standard, and start thinking about meeting he client’s financial goals – and these are money goals. So, we need to “show them the money” and when we talk about return we need to show them an internal rate of return over a long period. We need to show them the return that incorporates their beginning wealth, the money they were able to pull out of the portfolio for their goals, and their ending wealth. Then (and only then) will we be acting in the best interests of the client.
Sorry for the noob question. Can you tell me what this blog skin is? I really like it. Or is it customized one, perhaps? I believe it will be a decent option for adsense too. I’d like it if you are able\will tell me about it. Thanks.
Ted, this is a custom design by Cold Spring Design, Inc.
I believe it would differ from person to person regarding what would categorize as outperforming the benchmark. Interesting post btw… minus the errata of course 🙂