Have a question regarding using the S&P 500 and Nasdaq as benchmarks. I’m sure most people & institutions benchmark to these indices using widely available opening and closing index prices, but isn’t this flawed since it doesn’t take into account dividends and won’t represent the true growth of a fixed investment? For example, the May 2010 return for the S&P 500 is calculated from Yahoo finance based on closing date as (1089.41-1186.69) / 1186.69 = -8.20%, but getting the total return from S&P.com shows May as -7.985%. I’ve been struggling to find a consensus online- which is more widely used and which is a more valid benchmark? It appears to me that the calculation return is more widely used, but the published total return is more valid… any input would be appreciated, thanks.
Difference is total return – reinvestment of dividends… May not be in the prices you are using. My BBG shows the following for May 2010: Returns Holding Per Annual Eq. Simple Price Appreciation -8.1976 % -67.2073 % Gross divs reinvested into index -7.9848 % -66.2024 %
Sorry if I didn’t make the point very clear - I realize the dividend reinvestment factor causes the discrepancy between opening/closing prices and the published total return (thus the description ‘total’ return). My question is with regards to industry practice and correct usage of benchmarks. I don’t want to be using misleading performance data by using the former method. On a related note, the total return is always ~20 bps higher than the opening/closing price return, wouldn’t reinvesting dividends in a down month lead to a comparatively lower total return?
I’ve always seen people / firms use the total return (TR) versions of indexes that include dividends reinvested. Using price return (PR) only can be misleading, making the index easier to beat by a fund manager. This is especially true for indexes that contain companies that pay a lot of dividends, such as the S&P 500 or Russell 1000. If I saw an advertising piece using a PR benchmark, I would find that somewhat suspicious UNLESS there was a really, really good reason for it. The only legitimate reason I could think of would be if TR data wasn’t available (e.g. for distant historical time periods or emerging markets). These days, most data service providers track both TR and PR data series for all widely followed indexes.
sundevl21 Wrote: ------------------------------------------------------- > Sorry if I didn’t make the point very clear - I realize the dividend reinvestment factor causes the discrepancy between opening/closing prices and the published total return (thus the description ‘total’ return). My question is with regards to industry practice and correct usage of benchmarks. I don’t want to be using misleading performance data by using the former method. No problemo. I think you can do it either way - but I believe the convention is to provide the S&P 500 total return to account for CGs and Income. ---------------- > On a related note, the total return is always ~20 bps higher than the opening/closing price return, wouldn’t reinvesting dividends in a down month lead to a comparatively lower total return? No. Divvys are like “free money” reinvested in the index on your behalf. I dont know how they calculate the index total return – modified dietz or twrr or whatever – but im thinking that if they are reinvested @ beginning of the period, you shouldnt do any worse than the index itself. Ive flushed those formulae from my brain, but something tells me that im in the right zip code.
sundevl21 Wrote: ------------------------------------------------------- > Have a question regarding using the S&P 500 and > Nasdaq as benchmarks. I’m sure most people & > institutions benchmark to these indices using > widely available opening and closing index prices, > but isn’t this flawed since it doesn’t take into > account dividends and won’t represent the true > growth of a fixed investment? For example, the > May 2010 return for the S&P 500 is calculated from > Yahoo finance based on closing date as > (1089.41-1186.69) / 1186.69 = -8.20%, but getting > the total return from S&P.com shows May as > -7.985%. I’ve been struggling to find a consensus > online- which is more widely used and which is a > more valid benchmark? It appears to me that the > calculation return is more widely used, but the > published total return is more valid… any input > would be appreciated, thanks. You always only use the TR return, the PR return is not a true return benchmark. it’s very simple, the return of a stock holding a dividend paying stock should be DIV and Stock split adjusted etc. you don’t just calculate the difference between two prices at 2 different dates and call it a return.
as others have stated, the total return is what is widely used by institutions, money managers, etc. I suppose there must be a reason that the Price only return is calculated, but I’ve never seen it used.
I figured that was correct but had seen a lot of conflicting evidence and finding monthly total returns is hardly prevalent on the web. S&P nicely publishes this info on their site, but I can’t find it on the Nasdaq site and have thus far only been able to extract it from a tedious process requiring Bloomberg and an excel add-in- I find it hard to believe everyone who benchmarks against the Nasdaq goes this route. Thanks for the input - charlotte appreciate the dividend reinvestment explanation attempt, can anyone expand on that though? Logically, I’m still stuck on how the reinvestment of dividends always outperforms simple price appreciation. For example, take a $100 index that has a -10% total return including 3% dividend yield, ending index price is $90. If you reinvest these dividends in a down-trending market rather than taking them as distributions, won’t you will have a comparably greater percentage loss?
nevermind… thought i found it
Found several instances, but prime example here, and from a solid source in Bloomberg. Obviously it’s just one article by one author, but they compare Warren Buffett’s performance to the S&P 500 using the simple price appreciation return. http://www.bloomberg.com/apps/news?pid=20601103&sid=aoxammlga6dw Basically, I agree that TR provides a more valid comparison, but given the difficulty in acquiring this data (especially so for the Nasdaq), I’m trying to ascertain if it is widely accepted practice to use the readily available data as a benchmark. It would be a beast of a task to re-do 10 years of monthly, quarterly, and annual performance data (and it has been reported this way for quite some time and I’m not arrogant enough to assume the employees before me have all been incorrect) but if I can’t ethically turn a blind eye. Thanks again for any input.
sundevl21 Wrote: ------------------------------------------------------- > Thanks for the input - charlotte appreciate the > dividend reinvestment explanation attempt, can > anyone expand on that though? Logically, I’m > still stuck on how the reinvestment of dividends > always outperforms simple price appreciation. For > example, take a $100 index that has a -10% total > return including 3% dividend yield, ending index > price is $90. If you reinvest these dividends in > a down-trending market rather than taking them as > distributions, won’t you will have a comparably > greater percentage loss? In your example, let say the dividend distribution happens in the beginning of the period. The TR would still be 2.7% outperforming the PR.
Use SPXT Index on bloomberg for the S&P 500 total return index
I disagree. In that example, the capital gain return is -7%. If you take the 3% dividend distribution at the beginning of the period you’ll have $3 plus $97*(1-.07) = $93.21. If you reinvest this $3 dividend you’d have $100* (1-.07) = $93. I’m pretty darn positive on the math, is there a logical error in my reasoning? Charlotte- thanks a lot, I feel stupid now - XCMP does the trick for the Nasdaq. Still doesn’t clear up why so many people benchmark against the PR indices though, either they’re all misinformed or it is acceptable to use…
sundevl - I think the mistake in your logic is that you are looking more at the investor return, rather than a simple change in price. Lets say you have a share worth $10, it pays a $2 dividend, causing the share price to fall to $8. The change in price is negative 20% on price alone. Yeah, that investor received $2, but that is not part of price only return of the index calculation, I believe. As for the NASDAQ, is it really that widely used? In my experience, it is not widely used at all, but my view of the world has a very strong institutional bias. We tend to use the Russell indices and those are shown as total returns on Russell’s website which is free.
Chi- thanks for the explanation, sometimes the simple things get pushed to the back of the brain in favor of learning the ever-so-practical test material such as, say, IS vs VWAP or bequest vs gift taxes. Makes sense about the Nasdaq as well, took a different perspective to point out the overall infrequent use of it as a benchmark - I work for a tech focused hedge fund, thus the Nasdaq emphasis. I was getting caught up in the “reinvested” aspect of the reinvested dividends. Taking your example, I was envisioning the dividend as $2 vs the FV of $2, where the FV of $2 would be lower if reinvested instead of set aside in a down market. If they simply said the PR index doesn’t adjust for dividends while the TR does it would have made more sense.
sundevl21 Wrote: ------------------------------------------------------- > Chi- thanks for the explanation, sometimes the > simple things get pushed to the back of the brain > in favor of learning the ever-so-practical test > material such as, say, IS vs VWAP or bequest vs > gift taxes. Makes sense about the Nasdaq as well, > took a different perspective to point out the > overall infrequent use of it as a benchmark - I > work for a tech focused hedge fund, thus the > Nasdaq emphasis. > > I was getting caught up in the “reinvested” aspect > of the reinvested dividends. Taking your example, > I was envisioning the dividend as $2 vs the FV of > $2, where the FV of $2 would be lower if > reinvested instead of set aside in a down market. > If they simply said the PR index doesn’t adjust > for dividends while the TR does it would have made > more sense. ofcourse the TR returns are harder to find! you have to pay for them!. I work at a money management company, and we have to pay enough money to S&P, Russell, TSX etc to get any type of information. also, TR will always be higher than PR. it doesn’t matter if you reinvest and lose money, bottom line is you are getting more cash. so simply stated, TR and PR will only equal if the reinvested dividend becomes 0! there’s really no reinvestment, it’s really just getting more money cause dividends are paid out and one indice you recieve divident and they other you do not.
whystudy Wrote: ------------------------------------------------------- > > > ofcourse the TR returns are harder to find! you > have to pay for them!. I work at a money > management company, and we have to pay enough > money to S&P, Russell, TSX etc to get any type of > information. > > also, TR will always be higher than PR. it > doesn’t matter if you reinvest and lose money, > bottom line is you are getting more cash. so > simply stated, TR and PR will only equal if the > reinvested dividend becomes 0! there’s really no > reinvestment, it’s really just getting more money > cause dividends are paid out and one indice you > recieve divident and they other you do not. Calm down buddy, just was asking a few questions, and while I ask some dumb ones from time to time I don’t think these are. For one, I don’t think it’s obvious by any means that TR returns cost money, especially since it’s not true. I welcome any Joe Schmo to go to the S&P home page, choose S&P 500, Index Announcements, and then get your historical TR returns completely free. Cost-cutting 101. Also, did you read what I wrote before you responded? The explanation of why I thought that seems pretty clear. It was the terminology that was throwing me off, check Bloomberg: SPXT DES. You say there’s no reinvestment, but the description of the index says “Calculated intraday by S&P based on the price changes and reinvested dividends of SPX…” However, this reinvestment rate is the 3-mo T-bill rate, while I incorrectly thought it was like a DRIP and was reinvested back into the S&P. If that’s “obvious” I’ll quit the CFA program today…
easy now… Where are you getting that 3-mo T-Bill reinvestment rate? That sounds wrong. Like both metaphorically and factually. If they did that – its not a Total Return index – its a basket of TBills & S&P 500 price index… Anyway - my BB says " Calculated intraday by S&P based on the price changes and reinvested dividends of SPX with a starting date of Jan 4, 1988." No reference to T-Bills, and the S&P index methodology invests the dividends in the price index at the ex ante date. Whatchutalkingaboutwillis?
It must be a proxy for something very similar, because you can change the value of the dividend reinvestment rate, but Bloomberg uses the 3-month T-bill rate and it provides a value pretty close to the actual value. Try Bloomberg’s total return analysis function, SPX TRA. I don’t have Bloomberg open now so won’t be verbatim on this, but it lists 3 returns, the simple price appreciation (regular SPX), total return with gross dividend reinvestment (SPXT) and another dividend reinvestment return that lets you input your own reinvestment rate but defaults to the beginning of period 3-month T-bill return. The first is the PR, second two are TR (latter two not exactly the same but close…I think for May 2010 they were -8.20%, -7.98x%, 7.98x%, respectively)