HomeVideos

The Problem with Equal Weight Index Funds

Now Playing

The Problem with Equal Weight Index Funds

Transcript

444 segments

0:00

One of the most common questions in the

0:02

comments on this channel is what about

0:04

equal weight index funds or some version

0:06

of that question. I promise you equal

0:09

weight index funds are not that

0:10

exciting. But people keep asking me the

0:12

question. So here we are. Market

0:15

capitalization waiting, which is what

0:17

typical index funds do, assign index

0:19

weights based on each company's size. An

0:22

equal weighted index fund, as the name

0:24

implies, equally weights all of the

0:27

stocks in a market or market segment.

0:29

Equal waiting seems like the solution to

0:31

many of the perceived problems that

0:32

market capitalization weighted indexes

0:34

have today, like high valuations and

0:36

high levels of concentration in the top

0:38

stocks. Equal weighting has actually

0:40

outperformed cap weighting going back

0:42

decades as well. But it introduces its

0:45

own unique risks, costs, and

0:46

inefficiencies. I'm Ben Felix, chief

0:49

investment officer at PWL Capital, and

0:51

I'm going to tell you why I do not use

0:52

equal weighted index funds.

0:56

Equal

0:59

weight index funds do look really

1:01

compelling right now. They don't have

1:03

the same level of concentration as a

1:05

market cap weighted index fund and they

1:07

have a long track record of strong

1:08

performance even in live fund data

1:11

reflective of real fees and costs. The

1:14

Invesco S&P 500 equal weight ETF has

1:17

outperformed the S&P 500 for most of its

1:20

life since 2003 and for the full period

1:22

since its inception in 2003. Just by a

1:25

hair though. But before getting too

1:27

excited, it's important to understand

1:28

the trade-offs and to give some context

1:31

to the past performance. I'm also going

1:33

to show you why I think the benefits of

1:34

equal weighting can be achieved more

1:36

efficiently without its associated

1:38

drawbacks. Let's start with some quick

1:39

background on equal weighted versus

1:41

market capitalization weighted indexes.

1:43

When you create a stock index, one of

1:45

the most fundamental questions that you

1:47

have to answer is how you're going to

1:48

weight the stocks in your index. Market

1:51

capitalization waiting means holding

1:53

stocks at their market capitalization

1:55

weights. Market capitalization is just

1:57

what the entire company is worth on the

1:59

stock market, like how much it would

2:00

cost to buy every single share of Apple

2:03

or Shopify. A larger company will make

2:05

up a larger portion of a market

2:07

capitalization weighted index than a

2:09

smaller company. Makes sense. Equal

2:12

weighting, on the other hand, assigns

2:13

equal weights to stocks regardless of

2:15

their market capitalization. These

2:17

approaches can result in significantly

2:19

different portfolios given the same set

2:21

of stocks. The logic of market

2:23

capitalization or market cap waiting is

2:26

pretty simple. The market has determined

2:28

how much weight each stock should have.

2:30

This is why market cap index investing

2:32

is often referred to as passive. You're

2:34

passively allowing the market to

2:36

determine the weights of the stocks that

2:37

you own. It's like letting water find

2:40

its own level instead of constantly

2:41

trying to adjust it yourself. The nice

2:44

thing about this is that it does not

2:45

require you to make decisions about how

2:47

to weight the stocks in your index. The

2:49

weights just are. Another benefit is

2:52

that because the index represents the

2:53

market, it doesn't need to be rebalanced

2:55

frequently. As stock prices evolve, the

2:58

market capitalization weighted index

3:00

fund evolves in exactly the same way as

3:02

the market without needing you to make

3:04

trades. One of the perceived problems

3:05

with market cap waiting is that it can

3:07

result in high weights in a relatively

3:09

small number of stocks. Something we are

3:11

seeing an extreme version of in the US

3:13

market, at least relative to its own

3:15

history right now. This index

3:17

concentration kind of es and flows

3:19

naturally over time as some businesses

3:21

become increasingly successful and earn

3:23

a larger share of the market. Market

3:25

concentration can seem concerning, but

3:28

the historical relationship between

3:29

market concentration and future market

3:32

returns has not been strong historically

3:35

in the US or in other countries around

3:37

the world. If we sort 10-year future US

3:40

stock market returns on their starting

3:42

level of market concentration, there's

3:44

really not much of a relationship there.

3:46

Many other countries have been and are

3:48

today more concentrated in the top

3:50

stocks in the index than the US market

3:53

is even today when it's more

3:55

concentrated than it has been in the

3:56

past. In any case, index concentration

3:58

is a current concern for many investors.

4:00

An equal weighted index by construction

4:02

by design won't ever be concentrated in

4:05

a small number of stocks in the same way

4:07

that a market capitalization weighted

4:09

index can be. But it will result in

4:11

significant under and overweights of

4:14

stocks relative to their market

4:16

capitalization weights. Now this is an

4:18

interesting question whether having a

4:19

high weight in large companies in a

4:21

market cap weighted portfolio is riskier

4:24

than having large overweights to smaller

4:26

companies and underweights to larger

4:28

companies in an equal weighted

4:30

portfolio. The answer is not black and

4:32

white but I think there's a legitimate

4:33

argument that the extreme over and

4:36

underweing in an equal weighted

4:37

portfolio is actually riskier than

4:39

accepting concentration in larger

4:41

companies in a cap weighted portfolio.

4:43

This risk can show up in a couple of

4:45

different ways. One is in volatility.

4:47

Looking at an ETF tracking the S&P 500

4:49

and one tracking the S&P 500 equal

4:52

weight, the 15-year standard deviation,

4:54

a measure of how much variability there

4:56

is in index returns, is quite a bit

4:58

higher for the equal weight index than

5:00

for the market cap weighted index.

5:02

Volatility is not the only way to think

5:03

about risk. Another way to measure risk

5:05

is by exposure to certain types of

5:07

stocks that tend to move together. Some

5:09

stocks like small cap and value stocks

5:11

are theoretically priced as being

5:13

riskier by the market. And an equal

5:15

weight index has more exposure to these

5:17

types of stocks than a market cap

5:19

weighted one simply because smaller and

5:21

cheaper stocks make up a smaller portion

5:24

of the market. Equal weighting all

5:26

stocks results in a natural tilt towards

5:28

smaller and lowerpriced stocks. This

5:31

speaks to one of the other perceived

5:32

benefits of equal weighting in the

5:34

current market environment. Stock

5:35

valuations in the US market are high

5:37

relative to their own history. High

5:39

current valuations do have a

5:41

relationship with lower future returns.

5:43

They're not perfectly predictive, but

5:45

there's something there. Due to its

5:47

natural tilt away from the largest and

5:49

highest price stocks, an equal weighted

5:51

index will have a lower aggregate

5:53

valuation than a market cap weighted

5:55

index. We can see this now with the

5:57

price to earnings and price to book

5:58

ratios for an equal weighted S&P 500

6:01

fund being much lower than those for a

6:03

market cap weighted fund. Those

6:04

exposures to smaller and cheaper stocks

6:06

are not a bad thing. I would actually

6:08

argue that they're a good thing when

6:10

they're done right. The problem is that

6:12

if you want to have exposure to these

6:14

types of stocks, they're probably more

6:16

efficient ways to get it. I'll explain

6:18

that in a minute. Equal weighted indexes

6:20

will also tend to have materially

6:22

different sector exposures compared to

6:24

market capitalization weighted indexes.

6:26

Sector over or underweights introduce a

6:29

whole other type of risk that most

6:30

investors probably want to avoid. Again,

6:33

looking at the pair of S&P 500 ETFs as

6:35

an example, we can see the sector

6:37

compositions are materially different

6:40

with a large underweight to technology

6:42

and overweight to industrials for the

6:44

equal weight index. If you want those

6:46

over and underweights because you have a

6:47

view on how those sectors are going to

6:49

perform in the future, I I guess that's

6:51

fine. But sector bets tend to be really

6:54

hard to get right. A lesser known issue

6:57

with equal weights is the implications

6:59

for rebalancing. One issue is more

7:01

frequent rebalancing. When a smaller

7:03

company gets larger, a market cap

7:05

weighted index does not need to do

7:07

anything. But an equal weight index

7:08

likely needs to sell down the position

7:10

back to equalize weights at its next

7:13

rebalancing. The same thing happens when

7:15

a large company gets smaller. For

7:16

example, this S&P 500 equal weight index

7:19

ETF has had average annual turnover more

7:21

than 10 times higher than a market cap

7:24

weighted S&P 500 ETF. Turnover comes

7:26

with implicit and explicit trading costs

7:28

that are ultimately borne by the fund

7:30

investors. The other rebalancing issue

7:32

is that equal waiting is naturally a

7:34

negative momentum strategy. This is a

7:36

big deal. I I'll explain what it means.

7:38

Momentum is the extremely well doumented

7:40

phenomenon first identified by Jagadish

7:43

and Titman in their landmark 1993 study

7:46

that stocks that have recently done well

7:48

tend to continue to do well for a time

7:51

and those that have recently done poorly

7:53

tend to continue doing poorly for a

7:54

time. Equal weight funds to maintain

7:57

their equal weights are always going to

7:59

be selling what has done well recently

8:00

and buying what has done poorly. If a

8:03

stock does well and gets bigger, its

8:05

weight will exceed its equal weight

8:06

target. And a stock that does poorly

8:08

will fall below its target. Maintaining

8:10

equal weights then means systematically

8:12

trading against momentum. We can show

8:15

this with a multiffactor regression.

8:17

That's a statistical test that shows us

8:18

what's driving a fund's returns. It's

8:21

kind of like a like a blood test for a

8:23

fund. It tells you what's in your

8:24

system. Looking again at the equal

8:26

weight S&P 500, we can see that it loads

8:28

positively on the size factor and the

8:31

value factor and loads negatively on the

8:33

momentum factor. These factor loadings

8:35

are all statistically significant,

8:37

meaning they're real patterns, not just

8:39

noise in the data. This means that this

8:41

fund has more exposure to small cap and

8:43

value stocks than the market cap

8:45

weighted total market index and that

8:47

it's trading in the opposite direction

8:49

of momentum. This is exactly what you

8:51

would expect based on the equal weighted

8:53

index's structure. It's naturally

8:55

overweight, smaller and cheaper stocks

8:57

and underweight larger and more

8:59

expensive stocks. and it trades against

9:00

momentum systematically by the way that

9:03

it's designed. Now, here's the big

9:05

question. If you want exposure to

9:08

smaller and lower priced stocks, is

9:10

equal weighting the most efficient way

9:12

to get there? There are other investment

9:14

strategies that look a lot like index

9:16

funds, but don't strictly follow market

9:19

cap weights. For example, the

9:20

dimensional US core equity one fund has

9:24

roughly similar exposures to the small

9:26

cap and value factors as the equal

9:28

weight S&P 500 ETF. But rather than

9:30

naively getting to these exposures

9:32

through equal weighting, Dimensional is

9:34

being super intentional about targeting

9:36

certain types of stocks. This

9:38

Dimensional Fund has a net expense ratio

9:41

of 15 basis points, 0.15%, compared to

9:44

the equal weight S&P 500 ETFs, 0.2%.

9:48

It's basically a wash on fees with a

9:50

slight edge for dimensional. I'm going

9:52

to talk about the long-term data for

9:54

this mutual fund which is the the one

9:57

I'm talking about is a mutual fund. That

9:59

fund is only available through financial

10:01

advisors. That's just how Dimensional

10:02

distributes their products. But just to

10:04

make sure that this is relevant for

10:06

people watching, the same strategy is

10:08

available now through a US listed ETF

10:10

that anybody can buy. There's no no

10:12

restrictions. Uh so Dimensional's

10:15

intentional approach shows up in a

10:17

couple of different ways. One is that

10:19

they limit the sector divergence from

10:22

the market by placing a cap on sector

10:24

weight differences to avoid taking on

10:26

large sector bets. Another intentional

10:28

difference is in their treatment of

10:29

momentum. So remember equal waiting is

10:32

short momentum. It's taking it's got

10:33

negative momentum exposure. Dimensional

10:36

applies a rule when they trade to avoid

10:38

selling recent winners and avoid buying

10:41

recent losers all else equal.

10:43

Dimensional funds are not index funds.

10:44

So they have the ability to do that type

10:46

of flexible trading. They can decide to

10:49

buy this stock instead of that stock.

10:52

Though they'll buy the one that does not

10:54

have the the negative momentum exposure.

10:58

Another important way that their

11:00

intentional implementation shows up is

11:02

in the way that the portfolio turns

11:04

over. How how frequently the holdings

11:05

change. So, it's got similar tilts

11:08

towards smaller and lower price stocks

11:09

to the equal weighted S&P 500 fund, but

11:12

the Dimensional Fund has a really modest

11:14

turnover. Now, a big part of the reason

11:16

is that while Dimensional is tilting

11:18

towards smaller and lower price stocks

11:20

very intentionally, they use market cap

11:23

weights as a starting point. So they

11:25

start with take the market portfolio

11:27

take like a market cap weighted index

11:29

fund and then they're applying modest

11:31

tilts to get the exposures that they

11:32

want to smaller lowerpriced more

11:35

profitable companies. So this starting

11:37

with market cap weights minimizes

11:39

unnecessary deviations from market cap

11:42

weights which then translates to less

11:44

aggressive rebalancing requirements. You

11:46

don't have those extreme movements like

11:49

you have in a in an equal weighted fund.

11:51

on the concentration concern by tilting

11:54

away from larger and higher price

11:56

stocks. The dimensional fund is reducing

11:58

concentration but not to the same extent

12:00

as equal waiting. If your only concern

12:02

is market concentration, equal waiting

12:04

is a is a really good solution, but

12:06

that's probably not the right problem to

12:08

to solve. Looking again at factor

12:10

regressions, we see that the dimensional

12:12

fund and the equal weight S&P 500 fund

12:15

have had roughly similar factor

12:16

exposures. But as expected, the

12:18

Dimensional Fund is not taking a

12:20

negative position on momentum because of

12:22

the way that they trade. Going back to

12:24

2005 when the Dimensional Core Equity

12:26

One Fund first launched, both the equal

12:29

weight S&P 500 ETF and the Dimensional

12:31

Fund trail an S&P 500 ETF, but the

12:35

Dimensional Fund does outperform the

12:36

equal weight fund. This was a period

12:38

where large stocks with high prices

12:40

performed really well. So the result of

12:42

the S&P 500 coming out ahead over this

12:45

period is not surprising. Going back to

12:47

1971,

12:49

which is when back test data start for

12:51

the S&P 500 equal weight index, equal

12:54

weighting looks really, really good

12:56

compared to the market cap weighted S&P

12:57

500. This was a period where small cap

13:00

and value stocks performed well. Again,

13:02

we can see using multiffactor regression

13:04

that the performance of the equal

13:05

weighted fund is largely explained by

13:08

higher exposure to small cap and value

13:10

stocks with a negative contribution from

13:12

momentum. Comparing this equal weight

13:14

result to the backtested dimensional US

13:16

core equity 1 index. So to be clear,

13:19

this is not a fund anymore. This is a

13:21

back tested index is is also a really

13:24

interesting comparison. It had roughly

13:26

similar exposure to small cap and value

13:28

stocks, a lighter negative exposure to

13:30

momentum, and it performs nearly

13:32

identically to the equal weight S&P 500

13:35

index over the full period with less

13:38

volatility. Keep in mind that for both

13:40

of these back tested examples that this

13:42

is back test data, which can be pretty

13:44

unreliable. There's a saying in finance

13:45

that you'll never see a back test you

13:47

don't like. They're built with the

13:49

benefit of hindsight and they don't

13:50

reflect real world implementation costs.

13:52

But it's still interesting to see a

13:54

similar trend to what we saw in live

13:56

funds. Exposure to smaller and cheaper

13:58

stocks and negative exposure to momentum

14:01

in the case of the equal weight fund

14:03

explains most of the results. Equal

14:05

weight indexing seems appealing right

14:06

now because high concentration in the

14:08

top stocks and high market valuations in

14:10

the US market are making investors

14:12

nervous. That in addition to some pretty

14:15

attractive long-term back tests seem to

14:17

make a strong case for equal waiting.

14:20

But it comes with some baggage. Equal

14:22

waiting results in large over and

14:24

underweights at the individual stock and

14:26

sector levels, high portfolio turnover,

14:29

and a systematic bet against momentum.

14:31

The strong past returns of equal

14:33

weighting are largely explained by tilts

14:35

towards smaller and lowerpriced stocks

14:38

rather than some magic from equal

14:40

weights. The implication for investors

14:42

is that if you want to tilt towards

14:44

smaller and lowerpriced stocks, which is

14:46

a reasonable thing to want to do, you

14:48

can do it without introducing all of the

14:50

downsides of equal waiting. Dimensional

14:52

fund advisors creates products that do

14:54

something like that. They tilt towards

14:56

smaller and lower price stocks

14:57

systematically in addition to a couple

15:00

of other factors while controlling for

15:02

sector exposure, limiting portfolio

15:04

turnover, and intentionally avoiding

15:05

trading against momentum among a whole

15:08

bunch of other interesting stuff that

15:09

I've talked about in past videos. To be

15:11

clear about one thing, Dimensional did

15:13

not pay or otherwise incentivize me to

15:15

talk about their funds in this video. My

15:16

firm PWL Capital does use their Canadian

15:19

products and our clients portfolios. I

15:21

did a video a while ago explaining

15:22

Dimensional's history, which is closely

15:24

tied to the history of index funds, and

15:26

comparing their long-term performance to

15:28

comparable Vanguard index funds.

Interactive Summary

The video discusses equal-weight index funds versus market-capitalization-weighted index funds, addressing common questions and perceived problems with the latter, such as high valuations and concentration in top stocks. While equal-weighting has historically outperformed, it introduces its own risks, costs, and inefficiencies. Market-cap weighting assigns weights based on company size, while equal-weighting assigns equal weights to all stocks. The video explains that market-cap weighting is passive, requires no rebalancing, and its concentration, while concerning, has not historically shown a strong negative correlation with future returns. Equal weighting, by design, avoids concentration but leads to significant over- and underweights relative to market capitalization, potentially increasing volatility and exposure to riskier stock types like small-cap and value stocks. It also results in higher portfolio turnover and a systematic bet against momentum, as it requires selling recent winners and buying recent losers to maintain equal weights. The video suggests that if investors want exposure to smaller and cheaper stocks, there are more efficient ways to achieve this, such as through Dimensional Fund Advisors' products, which intentionally tilt towards these factors while controlling for sector exposure, limiting turnover, and avoiding trading against momentum. These products offer similar factor exposures with fewer drawbacks compared to equal-weighted funds.

Suggested questions

5 ready-made prompts