Can someone please explain why “some” misfit return is beneficial? From my understanding misfit return and true active return are independent so how can misfit risk “cause” active return (which is presumably why it is beneficial)?
I’ve done a search on this but not found anything that explains it well…
Thanks,
Misfit return is the active return you get from taking a certain style.
So it’s style benchmark - market benchmark.
Active return is the portfolio return - style benchmark.
Adding both would give you the total active return.
Thanks for the response, but it doesn’t really address the question? In the formulae you posted it would be best to have misfit return = 0 i.e. the sponsor uses the manager’s “normal benchmark” so that there is no distortion through misfit “return”
My question is why would some misfit return (and misfit risk) be positive for the portfolio if it distorts performance evaluation?
No, remember, active return = true active return + misfit return
So, the manager being better than the “correct style” benchmark is the “true active return”, but if you select a correct style as a benchmark, you will generate positive alpha or positive active return in the overall by having a positive misfit return.
Regards
Jorge
I still don’t understand your point (am I just being really stupid?)
If the correct benchmark is used by the sponsor then:
True active return = Portfolio return - “normal” benchmark return = X
Misfit return = “Normal” benchmark return - “normal” benchmark return = 0 (zero due to use of correct benchmark)
Total active return = true active return + misfit return = X
Misfit risk above would also be zero
So my question still stands, why is having some misfit return (and misfit risk) better than having none? From my point of view all it seems to do is distort the performance evaluation.
Misfit is when a manager will create a benchmark more suited to HIS style, which makes it DIFFERENT from the market benchmark. So if the market benchmark is the S&P 500 he will create his own, say, growth benchmark. He then trades to attempt to outperform his own growth benchmark.
Therefore, part of his active return is simply because he created a different benchmark than the market -i.e the “MISFIT” and the other part is when he outperforms his own benchmark “true active”.
This doesn’t distort portfolio evaluation because this helps one distinguish if the manager outperformed the market only because that sector outperformed or if he has added value beyond it.
Hope this helps
Having misfit return > 0 indicates that the benchmark I selected generated additional return from the general market. Specifically, this means the style of MY benchmark chosen outperformed the global/market benchmark. This is misfit return.
I’ve selected a ‘style’ benchmark that is different from the normal/global market benchmark that had a return that was above 0. This is a good thing.
I would assume it’s because if
misfit rtn = mgr’s benchamrk - investor’s benchmark
then a positive value would imply that the manager is trying to outperform a higher hurdle than the one held by the investor.
this is the key
Misfit return = “Normal” benchmark return - “normal” benchmark return = 0 (zero due to use of correct benchmark)
It doesnt has to be 0,
You have MR= “Style” benchmark return - “Normal” benchmark return. (its not the same benchmark)
An appropriate style provides alpha
Jorge
Alisha - not necessarily true. Instead of using an inappropriate benchmark e.g. S&P 500, which carries misfit risk why not just use the custom benchmark ONLY and eliminate the misfit return?
Aydub - you are assuming that misfit return will always be positive. Additionally, even if it is positive the problem is that the sponsor is acknowledging this as something the manager did
cokemicho - unless I’m mistaken returns based on style is NOT alpha
Gersonides - this cannot be true because there is no guarantee that misfit return is positive (if it is negative it will lower the return “target”)
im trying to explain the positive side of the misfit return, not alpha, but you get positive return vs the client benchmark, call it whatever you want.
I think you are confused somewhere.
The market benchmark does NOT carry any misfit risk. The misfit is when you make changes to the market portfolio. If a manager choses to invest only in the Financial Services sector, thats the misfit, hes deviating from the market i.e ALL sectors in the market. Therefore if the manager then trades around only the Financial sector and earns excess return beyond what the market is earning.
Perhaps numbers will help?
S&P 500 annual return - 6%
Financial services - 10%
Poftfolio manager who invested in financial services sector - 11%
When his perfomance is evaluated the investors will compare his performance to the MARKET, not his own benchmark (as this would allow the manager to misuse it and show excessive return when there is none).
So when the return is calculated can the entire 5% (11-6) be attributed to his own skills? No, because the sector itself outperformed the market by 4%. So if he had chosen to just buy all he shares in the financial services sector and done nothing about it, it would have earned 4% and his only active decision making was to put the money in that sector. However he earned 5% beyond the general market out of which 4% is because he chose a specific style and 1% because he used his own skills and analyses to earn beyond the sector performance.
This divides his perfomance between what he earned by chosing a style and what he earned by actively using his skills.
ayousaf, try looking at this from the point of the investor, not the manager. The investor establishes his benchmark to beat. This is appropriately named the " Investor’s Benchmark". This is the base which “Total Active Return” is calculated.
But Mr. Manager doesn’t use the Investor’s Benchmark, he uses one more inline with his style. He uses the “Manager’s Benchmark”. His outperformance is the managers’ true outperformance.
Now don’t forget that we’re looking at everything from the point of view of the investor. There isn’t a more appropriate BM to use because it’s investor’s choice. The alpha from the investors POV is going to be the Total Active return versus the Invstor’s Benchmark.
Therefore the Total Active Return = True Active Return (Manager’s AR above Manager’s BM) + Misfit Active Return (The difference between the Manager’s BM vs. the Investor’s)
True Active and Misfit Active Return can either be positive or negative.
I honestly hope I’m not being stupid here BUT:
greyound86 - OK, let’s assume that the investor chooses the S&P 500 as their benchmark and the manager selects the S&P 500 GROWTH as his benchmark. The investor invest $1m with this manager.
Portfolio return: 20.00%
Return on S&P 500: 5.00%
Return on S&P 500 GROWTH: 10.00%
Return on S&P 500 VALUE: -5.00%
In this case, the active return = [20.00%-10.00%]+[10.00%-5.00%] = 15.00%
10.00% is “true” active return and 5.00% is misfit return
HOWEVER, the sponsor could have used a completeness fund to eliminate misfit risk by investing an equal amount with an indexed value manager.
SO, if we have $1m in S&P 500 VALUE this gives a return of -5.00%. The total return to the investor is -50,000+150,000 = 10,000 which is purely the active management component - MISFIT RETURN IS ELIMINATED.
The book however states that a non zero misfit risk is optimal. My question is, why is eliminating misfit risk a bad thing? Another way of saying the same thing is, why is misfit risk beneficial?
ayousaf,
I think you got it. Eliminating Misfit active return is reducing the total active return of the portfolio. Essentially the goal of using a Completeness Fund is to reduce the active risk in the a portfolio to hopefully improve the overall Information Ratio of the total portfolio. However, the downside to this is that you lose some active return, as well. The portifon of active return you lose is the misfit active return.
Yeah i think you just answered your own question there… In your example eliminating the misfit risk lowered your return.
Errrrr… no?
The true active management component is still the same at 10.00%.
The book states (book 4 pg. 220) " Although it may seem that no “misfit” risk is desired, a non-zero amount is optimal, because a high level of “true” active return may more than compensate a given level of misfit risk"
But, increasing misfit risk has no bearing on how much “true” active return is earned (although it does increase total active return) - this is what I’m not understanding…
I found the section of the book you’re referencing. It’s saying that one would might think that you wouldn’t want any misfit risk (not return), but actually having some misfit risk is acceptable.
While it doesn’t explain it further I believe that it’s implying that it’s OK to use a manager that doesn’t fit the investor benchmark perfectly because even though you will have some additional misfit risk from not matching the benchmark, the “true” active return you receive from the manager (you would get true and misfit, but all you really want is true) will make up for the additional risk you’re taking.
Bascially you’re increasing both the numerator and the denomiator of the information ratio, but the increase can be acceptable because the return will go up more than the risk.
Did that do it?
guys very nice high-level discussion here…
In a seperate note, why would an investor’s benchmark be different than the manager’s?
This is actually an important point. There’s several reasons. First of all, remember that a benchmark is specified in advance. Therefore when you’re analysing performance, you’re analysing it against the set benchmark. Scenario:
Bob invests in the market on your behalf. Your default benchmark for Bob is Total US Market. He does analysis, and decides smallcaps are the way to go this year, and buys a bunch of stocks, all smallcaps. Returns come out like this:
US Market: 10%
US smallCap: 15%
Bob: 16%
What’s the result here? Well, Bob outperformed your benchmark (Total US Market) by 6%.
“Wait, he only bought small-caps, shouldn’t he be measured against the smallcap index? In that case, didn’t he only beat it by 1%?” No. Remember, benchmark is specified in advance. His analysis showed smallcaps will outperform, so he bought them. He should be rewarded for that. Bob’s benchmark may be smallcaps, but the investor never put any such restriction. Investor just said, “go beat the US Market for me”. Bob did.