Understanding Your Bias - Naive Diversification

Part 1: Why you aren't as 'diversified' as you think

You are no slouch. You are in the top percent of knowledge when it comes to investments and finance (I can say that definitively if you are reading this substack since we are teaching REAL Chad-level stuff).

Diversify Yo Bonds…The GOATs know you need to diversify

So you know you need to diversify and there are benefits of diversification. Maybe you took a class and can even drop terms like ‘efficient frontier’, ‘capital allocation lines’, and ‘Markowitz’ in conversation.

Did you build a risk-specific Tangency Portfolio you big brain Anon? (source Wikipedia)

So you choose a few different funds in your 401k: an S&P500 or domestic large-cap ETF, a NASDAQ or tech-focused ETF, maybe some smaller allocation to international ETFs, small- & mid-cap ETFs, and a bond fund (not 40% bonds, you know better than that - we showed it in our Retirement with the ‘Tism series). Then you buy a few blue-chip stocks in your individual account.

Suffice to say, you got this diversification thing in a bag, right?

….right?….

What if, despite all your separate choices, you aren’t as diversified as you think?

Let’s take a little deep dive into diversification and see.

Naïve Diversification

Naïve diversification is a ‘choice’ heuristic that describes how people will overly spread out their bets across all the options given to them. It is also known as ‘diversification heuristic’ and is a known fallacy when giving users lots of choices.

The original study on the concept was done with snacks. The researchers gave 6 choices to individuals. If the individual had to plan multiple weeks out, they would usually choose roughly equal amounts of each of the 6 snacks. They expected to want the variety and diversified across the choices. However, if individuals were allowed to choose a snack everyday, they would usually choose the same 1 or few options over the study period. Their choices were concentrated on the ones they actually really liked.

The study has been replicated with children and candy (simulating Halloween). If you gave a big bowl of candy and asked kids to choose a bunch, they would diversify across many different types. Maybe they would grab a bunch of good candy like Reese’s, but also would get some less good candy (looking at you ‘paper dots’). However, if you allowed kids to go ‘house-to-house’ they would choose their favorite candy at each house. The resulting bags would be very very different.

Sorry dots, turns out no one wants chunks of paper with their unflavored sugar & food dye after all

Cool. What’s this have to do with investing?

Turns out, despite all your models and education, adults aren’t much different than children picking candy. When given 3 options in their 401k, many adults will put 1/3 into each. When you expand it to 10 options, they tend to put 1/10th into each. This is done regardless of what the options were.

The more options an individual is given, the more they diversify. Even if the choice wasn’t 1/10th in each, most people put at least some allocation into every investment option given. Even if they are given 5 different S&P ETFs, they will put money in each.

Let’s say the 3 options are 1) S&P500, 2) International, and 3) Bond ETF. And the 10 options are 8 different types of large-cap domestic funds, 1 international fund, and 1 domestic bond funds.

That same person would be 33% domestic large cap in the first example vs 80% in the second (with 8 / 10 funds) domestic large cap based solely on the options they were given.

[F’er Note - If you don’t think this is true, bless your heart. We can 100% say that at least 1 person anecdotally does this. Mama F’er has us allocate her 401k, and the first time we went in, she was contributing exactly 1/n to each option. 10ish or so were target date funds with different dates.

Target date funds just have different allocation to the same underlying funds and shift to a more bond focused allocation as you get close to retirement. Hopefully you see how silly it is to be invested in 10 of them.]

Naïve Diversification - Even Deeper

You aren’t some financial illiterate though. You read the above and had a chuckle…I mean, how could anyone do that and think they are diversified?

Let’s hypothetically assume you follow the standard advice of buying low-cost Vanguard ETFs. You buy some S&P500 and Nasdaq ETFs. You also want to own some of the nice ‘safer’ blue-chip individual names and follow allocation rules like 5% max to any individual names.

What are some blue-chip names you think of?

Apple. Microsoft. Berkshire Hathaway. Amazon. JP Morgan Chase. etc.

At the surface you would look pretty diversified. You have a few small allocations to relatively safe names you like and most of your money is in a nice diversified low-fee ETF.

But you need to look at the holdings of the ETFs if you want to know how diversified you really are. So what are the top holdings in the Vanguard S&P500 ETF?

Vanguard S&P500 ETF Holdings

What jumps out? The big blue-chip stocks you think of are a huge portion of the index. If you wanted to limit your total exposure to 5% allocation to Apple, having it make up 7% of the index means you are stealth allocating a large amount of money to AAPL already.

But I am sure the other indices are better…

Vanguard Growth & Tech ETF Holdings

And, those top 2 companies look familiar. But Apple now makes up 24% of the Vanguard Tech ETF. If you put 25% of your portfolio into this NASDAQ tracking ETF, you would have over 6% of your total portfolio already exposed to Apple. As in that would already put you over your 5% exposure limit.

Do international ETFs help?

Vanguard World Index ETF

Even on a world index, American companies make up a hugely outsized percent of the index.

Suddenly you don’t look as diversified. Apple makes up 7%, 24%, and 6% respectively across the 3 Vanguard indices. You think you only have 5% exposure to Apple with the individual stock you bought, but you have a lot more than that due to how much of the ETFs are made up of AAPL.

And if you think that this is a Vanguard specific issue, here is a mix of common non-Vanguard ETFs. The SPDR, Invesco and MSCI indices are other low-cost ETFs that you can purchase in your account. What are the top holdings of these names?

To really beat the proverbial dead horse, we will walk through an example. You allocate your account as:

• Nasdaq (QQQ) 35%

• SP500 (SPY) 35%

• MSCI World 10%

• Apple 5%

• Microsoft 5%

• Amazon 5%

• Misc Individual names 5%

You would look & feel pretty diversified on the surface. No individual name is more than 5% and you are spread across what looks like 3 pretty unrelated indices. But as we showed above your actual portfolio exposure is:

• Nasdaq (QQQ) 25% (Net AAPL, MSFT, AMZN)

• SP500 (SPY) 30% (Net AAPL, MSFT, AMZN)

• MSCI World 9% (Net AAPL, MSFT, AMZN)

• Apple 12%

• Microsoft 11%

• Amazon 9%

• Misc Individual names 5%

Sure you could argue what you own is still diversified, however it is nowhere near as diversified as you thought. Apple, Microsoft, and Amazon all make up a larger part of your exposure than the actual international exposure you have in your portfolio.

Additionally, all 3 make up well over your 5% max allocation to a single name. You are twice as exposed to them as you wanted. You have a lot more risk tied to the iPhone release than you probably wanted.

This isn’t naive diversification in the traditional sense (of 1/n allocation across all options), but we would argue it is the naive diversification trap that ‘more sophisticated’ investors fall into. You are selecting options without looking at the holistic ending point.

Why Do These Indices Have High Allocations To The Same Names?

Stock market indices can be market-weighted, equal-weighted, or fundamental-weighted. Market capitalization weighted indices have historically been the most popular type of index with equal weighted indices being the second most popular choice. Fundamental-weighted indices are relatively new and were largely introduced in 2005.

Market Capitalization Weighted Index

A market capitalization (market cap or market weighted) index is what most people think about when they think of an index and ETF. They hold the names of each component of the index at roughly the same size as the company makes up of that index.

Therefore, mega-cap companies like Apple that make up a large percent of the S&P500 index will also make up a similarly large allocation of the market-weighted ETF.

Apple is ~$2.5 Trillion market capitalization and the sum of all 500ish companies in the S&P500 is ~$36 Trillion which means Apple is around 7% of the entire S&P500 market cap (2.5/36 = ~7%). As such, Apple will be ~7% of a market cap-weighted ETF.

[Note - market capitalization is the price of the stock x the number of outstanding shares.]

There are other ETF weighting methods out there but they are less commonly used, especially rarely offered in things like 401ks.

Equal Weighted Index

Equal weighted indices will put the same weight in each underlying component of the index. So the S&P500 equal weighted index will put 1/500 weight on each of the 500 stocks (0.2% to each name).

An example of an equal weighted index is Invesco S&P 500 Equal Weight ETF (RSP) if you want to look one example up.

One thing to note, to keep an equal weighting there is a higher rebalancing cost than in a market weighted index. If one stock goes up and another goes down, an equal weighted index needs to sell and buy to get them back to parity. However, a market weighted index wouldn’t need to make any trades since the change in price changes the market cap which changes the market weighting naturally.

Fundamental Weighted Index

Fundamental weighted index is a relatively new concept. They are essentially made around certain factors and adjust the weighting as those factors change. Factors can be things like:

• Valuation ratios (ie P/E or back-value)

• Financial metrics (ie weight by revenue or cashflows)

• Dividend yield

Additionally, the index doesn’t need to use just one factor as many fundamental weighted indices use a multi-factor approach. There is almost no limit to the combination of factors and complexity that can be done.

Should You Choose A Different Weighting Methodology On Your ETF?

Which methodology is better and which should you choose? From an investment perspective, this is fully up to you to decide. There are some general rules of thumb for each type of ETF performs well in:

• Market Cap Weighted ETF

  • Since these are highly weighted toward the larger companies, when outperformance is concentrated in the already large companies, that is a good guy

  • Similarly, these will perform well when ‘momentum’ outperforms (also called trend following). As a company gets larger, it makes up a bigger allocation in the index, and if it continues to outperform you now have a larger weight on a name as it continues to see its price go up

    • For example, if you think of the tech bubble. As tech stocks got larger, tech made up a bigger % of the S&P500. As the bubble kept inflating, you had outsized returns due to the higher allocation in tech

  • This gets into the downside of market-weighting indices. They don’t rebalance & take gains on names that have run up in value. Therefore, any reversion to the mean or ‘corrections’ in the market also get an outsized loss in the portfolio

• Equal-Weighted ETFs

  • Equal weighted ETFs will outperform when the smaller names outperform since these names get outsized allocations to the index.

    • For example, the smallest stock in the S&P right now is Embecta Corp (EMBC) with a 0.0005% allocation on market cap. Giving it a 0.2% allocation would be ‘overweighting’ it by like 400x. If EMBC doubles, it would have essentially no impact on the market-weight but a meaningful impact on the equal weight index

  • Similarly, when there is a reversion to mean or correction, you will outperform in the equal weighted index vs market cap weighting.

  • Consequently, in momentum markets or when large caps are outperforming, the equal weighted index will underperform - which has been the recent historical markets

• Fundamental-Weighted ETFs

  • Since fundamental-weighted ETFs can vary greatly based on which factor they are using, it is difficult to make a generalized rule. However, the general sales pitch is that fundamentals should lead to lower volatility and moderate positive performance due to looking at fundamental factors.

Summary: Naïve Diversification

Naïve diversification takes on many forms. The most basic of which is someone just haphazardly allocating the same percent to all the options they are given. This results in a portfolio allocation driven solely on what is offered and without any strategic planning.

However, even more sophisticated investors can fall into the trap of mental accounting fallacy and thinking they are properly diversified & adhering to their exposure rules when mixing ETFs.

To really ensure you are diversifying at the level you want, you need to look at the holdings of your various funds. But lets be honest, that can be a lot of work.

What do we do?

We have a 401k that is roughly the same size as our IRA accounts. Since our 401k is largely invested in large cap funds, we keep our IRA largely invested in smaller cap names that are not in the S&P500. This is an easy way to avoid double-allocating to names.

Another easy solution people use, is to avoid buying individual names and just buying ETFs. If you spread your ETFs into mutually exclusive indices, you avoid overlap. For example, buying ETFs that are large cap, mid cap, small cap, and international excluding US stocks you have 4 funds that would have essentially no potential for overlap.

Whatever you decide, you now have the knowledge to be aware of the potential issue with your diversification and can make a more informed choice when choosing your allocations.

And don’t think this is limited to investing. Just like the studies using snacks, diversification heuristics impact every aspect of life. The more you are aware, the better decisions you can make.

Good luck Anon