Lately there has been a lot of talk about ETF investing and how flawed some of the strategies can be, so I really wanted to take a little bit of a deep dive to see if this was true or not. The biggest issue I hear is that so many S&P 500 ETFs are weighted too heavily vs an equal weight S&P 500 ETF, but my question is this – is that a bad thing?
First, let’s explain the differences between the two ETFs. You’ve likely heard of some of the most common S&P 500 ETFs like my personal favorite, SPY. That ETF contains all 500 companies in the S&P 500 but it is weighted by market cap.
What that means is that the largest companies are going to have more of an impact on the share price of that ETF, which might be good but also can be bad.
One of my favorite charts to show this variance is from Market Watch and I’ve included it below for your viewing pleasure:
As you can see, there is basically 5 major companies (AAPL, AMZN, GOOG, MSFT & FB) that make up essentially half of the S&P 500 ETF. Now, I’m not a mathematician but 5/500 is 1% while this chart is showing about 50%, so these 5 companies combine to make up 50x what they would if it was equally weighted.
Which brings me to the next option – an equally weighted S&P 500 ETF. An equal weight S&P 500 ETF is just that – it’s perfectly equal. Every company, regardless of market cap, has the exact same weighting.
As I mentioned earlier, the question is really if this is a bad thing or not. In essence, it really boils down to two different opinions:
1 – You think that it is a bad thing because your investment isn’t very diversified as 50% is made up by 5 companies
2 – You think that it is a good thing because theoretically, the larger that the company is, the better they are
Honestly, I flip back and forth on my opinion of these two viewpoints and can legitimately see it both ways.
If you’re not in favor of the ETF because you’re not diversified, I can’t disagree at all with that because you’re really not. A large majority of the companies that you’re investing in are making up a much, much smaller amount than those Top 5 companies are.
Another issue with this is that if you look at those five companies that make up the 50% – what do they all have in common?
They’re all tech companies! Personally, I love tech companies and think that they’re the way of the future, but I understand why you wouldn’t want HALF of your investment to be in tech if this was where you were parking all of your money for the long-term.
If you are in favor, I think that makes sense too – big companies have grown because they have shown long periods of sustained success. If you’re investing in an S&P 500, that’s what you’re looking for – somewhere that you can dump your money and not have to worry about it again until you decide to take it out for retirement.
Personally, I would want a majority of my money to be invested in companies that are continuing to crush it rather than those that are company 499 in the S&P 500.
All of these are merely just qualitative factors though. If you know anything about you me, I am a man that lives and dies by the numbers, so let’s dive deep!
For this comparison, I am going to compare two different ETFs:
- SPY is an S&P 500 ETF that is weighted by market cap
- RSP is an equal weight S&P 500 ETF
Before you go any further, I challenge you to take a step back and think about what you think the outcome is going to be. I think that taking some time to think about the answer beforehand, and then wiping your memory completely clean and going into the analysis with an open mind is key.
Something that I have an issue with is confirmation bias, where I’m essentially looking for data to support the conclusion that I want to find. I have found that taking the time to think about any preconceived notions and then making the conscious effort to ignore them will make you a better investor.
Personally, I think SPY is going to crush RSP. I would want my money to be invested in the companies that have grown the most because hopefully, that means they’re poised to continue their dominance.
Ok – mind is swept clean. Time to look at the data!
I went back to 1/1/2004 as RSP started mid-year 2003 to keep the data clean. Below is an annual breakdown of the ETF returns for RSP and SPY, as well as a comparison between the two ETFs:
The chart shows a lot of data, but the thing that I immediately get sucked into is the Total Return. The total for RSP is 366% while SPY is nearly 351%, so you’re talking just about a 15% improvement over SPY over the long-term.
Now, another thing that immediately sticks out to me is that a very large majority of that occurred in 2009 when SPY was nearly 20% worse than RSP. Based on the next largest variance being in 2019 when RSP was 6.27% worse than SPY, I would venture to guess that might be an outlier.
Now you can never throw out data because it is real data that actually occurred, but it’s important to think about if it’s repeatable or not. Below I breakdown the average and median annual returns for both ETFs – one dataset includes all years and the other excludes 2009:
Clearly a pretty big shakeup between RSP and SPY when you exclude 2009. The average outperformance for RSP goes from .84% to be a .35% underperformer.
Personally, if I was making an investing decision, I would likely exclude 2009 because that 20% variance doesn’t really seem repeatable since the next closest was 1/3 of the variance at 6.27%, but that’s just me.
So, my opinion would be that so far, these ETFs are actually pretty equal. That means that you can invest in either one, right?!
Well, maybe – but you’re forgetting something…DIVIDENDS!
Dividends play a huge factor when you’re investing so it’s imperative that you take that into account. I have talked before about the importance of looking at the Total Yield, which is the annual return + dividend yield because if not, you’re comparing apples to oranges from the get-go.
Below shows the dividend paid each year for the respective ETF as well as the dividend growth:
The RSP dividend growth absolutely puts the SPY dividend growth to shame! I am pretty surprised by this because typically when you think of older, well-established companies, which typically would make up more of SPY because they have a higher market cap, I would think that they would pay larger dividends.
The more that I thought about it, though, the more it made sense. FB, AMZN and GOOG all do not pay dividends currently, so about 30% of SPY isn’t paying a dividend while those three companies would all be equally weighted with RSP, therefore not dragging down the total dividend nearly as much.
However, AAPL does pay an incredible dividend so hopefully they can try to make up a little bit for the other three laggards!
Similar to how I threw out an outlier in the share price return section, I want to see if there are any major outliers in this dividend growth section as well.
I instantly see that in 2017 there was a 43% outperformance of RSP vs SPY, and that seems a little bit high! Looking at some of the other years, there were some pretty major discrepancies such as when RSP underperformed by 28% in 2016 and outperformed by 19% in 2020, but neither of those are close to that 43% level.
And also, it’s a good point to bring up that since I only had 2 months of dividend data for 2020, I made an assumption that the dividend would be the same as it was the first two quarters of 2020 so that I could make a fair comparison.
Ok, back to the outlier conversation – you can see that the average RSP growth rate is nearly 6% more than SPY and the median is nearly 9%! Even if I take out the massive outperformance in 2017, RSP is still crushing SPY:
Now, it’s always important to remember that everything that I have shown so far is simply just the past and not an indication of how these ETFs are going to perform in the future, but that’s really all that we have to go on when we’re evaluating ETFs.
One of my favorite things to do as a “last check” when I am investing is to go back and look at the history of how my real, hard-earned money would’ve actually performed if I had invested it in this way. To do this, I go to a DRIP Return Calculator from the Dividend Channel (https://www.dividendchannel.com/drip-returns-calculator/)
It’s really cool because you can enter the pertinent info and pick a specific date and then see how actual returns would’ve looked for you. I entered in the information for RSP and then started it on 1/1/2004 and compared it to SPY, as you can see below:
How do you think the results turned out? Probably a slight advantage for RSP right? That’s what I am thinking!
The data shows that you would’ve turned $10K into $41,556 with RSP and $37,690 with SPY. Might not seem like a crazy amount being about $4K, but it actually is a huge difference.
Once you take out your $10K investment, that means you made $31,556 with RSP. That amount is nearly 13% more than what you would’ve made with RSP. Sure, that’s only $4K in this case, but if you had invested $100K then it’s $40K. $500K investment would mean $200K more. You can see how it can really add up.
And these returns are if you DID NOT reinvest those dividends back into the ETF, which I always recommend that you reinvest them as long as you still like the stock. Wonder why I advise that? I’ll show you:
Simply by putting those dividends to work you now have amounts that are about $6K more regardless of the ETF that you had chosen to invest in. Putting your money to work as soon as you can is mathematically proven to work better than holding onto the cash and trying to time the market, and that’s why I also recommend lump sum investing instead of dollar cost averaging.
One thing that I find really interesting is that the outperformance for RSP is only $2K when the dividends are reinvested vs. $4K when they weren’t reinvested. What this tells me is that the performance of SPY must have grown faster later on, and when we look at the data, we do see that SPY actually outperformed RSP on 6 of the last 7 years.
As it turns out, investing in an equal weight S&P 500 ETF mathematically has shown to be more prosperous than investing in a market cap weighted S&P 500 ETF like SPY. As I mentioned, this is not at all what I was expecting the data to show me, and this is why you should always go into analysis with an open mind.
Personally, this information isn’t enough to make me run out to my Fidelity account and change all of my investments, but I will be investing in RSP in the future.
I don’t have a bunch of money in ETFs as I mostly use ETFs for exposure to different sectors that I’m not super knowledgeable on like gold or cloud stocks, but I will occasionally use them as a more ratable return for an account like my 401k or HSA – something that I’m not necessarily looking for a homerun on.
For my other accounts, however, well, you all know me – I’m a diehard stock picker!