Investing With Style: How Size and Value Actually Work

Advanced, Theory

Sometimes the simplest investing concepts we take for granted are actually a lot more complicated than we think. For example, reasonable people might rationally assume that two small cap value index funds should have identical returns since they theoretically follow the same asset. But it doesn’t really work that way because definitions matter. What does “small cap” mean? Who defines “value”? Read the fine print and two similar funds may be a lot more unique than you realize.

I’ve been thinking about that subtle complexity recently, as in the process of finalizing the most recent annual Portfolio Charts data update I refined the definition of “small cap” to better match common index fund methodologies. The tweak was simple enough, but succinctly explaining what it means and why it matters got more and more difficult the longer I thought about it. The entire process made me realize it might be a good time to have a longer discussion on how stock index fund definitions work.

So if you’ve ever looked at an assortment of large, mid, small, blend, value, and growth stocks and wondered what all of that actually means, this article is for you. It’s going to get a little technical, but if you stick with it you’ll learn something not only about how indices are constructed but also how to use that info to interpret the data you find both here and elsewhere.

How large, mid, and small caps are defined

To understand how size is defined, we first need to talk about how stock data is sorted. A common method that organizations like the New York Stock Exchange (NYSE) use to track companies by size is to group them into ten different bins called “deciles”. Each decile is roughly equally populated by company count, so decile 1 contains the top 10% largest companies and decile 10 contains the bottom 10% smallest companies. For example, here are the NYSE companies contained with the Vanguard total market index fund VTSAX sorted by decile.

(Special thanks to Siamond at the Bogleheads forum for help with this data. If you aren’t already using Simba’s Backtesting Spreadsheet or reading his blog, it’s worth your time to check them both out!)

That’s simple enough. So given that distribution of companies, what deciles would you expect to qualify as large cap or small cap? Perhaps you would simply divide it in half, with the largest 5 deciles called “large” and the smallest 5 deciles called “small”. While it makes perfect sense on the surface, it’s not at all how index funds do it. And to understand why, it’s important to put the size of the companies in context.

Here is the same decile sort scaled not by company count but by the percentage of the total market capitalization contained within each decile.

Now the problem is pretty obvious. The largest 10% of companies are WAY larger than everything else, which majorly skews any size classifications based on decile. For example, look at decile 3. In terms of size, are companies in that grouping more similar to decile 1 or to decile 10?

Visualizing the same data another way, let’s stack the columns to see how each additional decile adds up to 100% of the market.

As you can see, the 10% of companies contained in decile 1 make up nearly 70% of the total market. The next ~15% or so is covered by decile 2, and deciles 3-10 make up the rest. See that thin purple sliver at the bottom? That’s 7 through 10 combined.

Various indices thus have an interesting challenge in how to divvy that uneven market stack into consistent definitions of large, mid, and small cap stocks. And they all have different approaches. Companies like FTSE Russell and S&P prefer round-number company count per index, while MSCI and CRSP generally use free-floating market percentage definitions. It may sound confusing and disorganized, but by mapping several different index providers to the same chart style we can see how they all stack up.

Image courtesy of CRSP

Even with different technical size definitions and a few verbiage nuances (like CRSP “Mega” being the same as MSCI “Large”), the end results are actually not that far apart in terms of market coverage. So how did four different companies with competing methodologies all arrive at similar ranges of large, mid, and small?

Even though none of these indices simply track NYSE deciles in their size definitions, I intentionally used the same colors in my last chart to highlight an important fact about how the market breakpoints (the lines between size categories) are influenced by industry standards.

See the resemblance? That’s no accident. Stock indices may define size different ways, but the people who build the indices are creatures of habit. Large caps are pretty close to decile 1, mid caps are close to decile 2, and small caps cover everything else save for the smallest micro caps. Even as more and more companies are added to any given index over time (usually skewed heavily towards new micro caps), these same general NYSE market breakpoints are often referenced to maintain historical standards in the data.

In summary, every index defines size a little differently. But because they share a common history in decile data, each of the index interpretations can be roughly mapped to standard market capitalization breakpoints. Think of it like understanding a new word you’ve never heard before by studying the Latin root words. The familiarity is by design.

How blend, value, and growth stocks are defined

Company valuation definitions take a slightly different approach. They’re generally not based on deciles at all, and instead focus on individual objective ratings per company.

The simplest model is to use a metric like the book-to-market ratio, with low B/M representing growth stocks and high B/M representing value stocks. But index funds generally take it much further with proprietary ratings formulas based on a wide variety of corporate factors. Simply put, it’s all about finding companies that can be considered either undervalued or overvalued based on current market prices.

The idea of studying annual reports to identify good deals has a long history in investing and appeals to value investors everywhere. Industry titans like Benjamin Graham and Warren Buffett have made entire careers on the topic, and scores of like-minded investors are naturally attracted to the concept of buying low and selling high. But in this case the storied history of value investing actually works against people trying to interpret index funds sorted by valuation, as indices don’t work the way you might expect. The difference between the classic value investing concept and the modern index version comes down to the difference between relative and absolute value.

I’ve found that many data-driven investors have a habit of thinking in absolute terms. They look at the B/M or P/E ratio and use it to make an objective decision on whether a stock is properly priced. If it’s a true value they may add it to their portfolio, and if it’s grossly overpriced they may sell it and prefer to wait for a good deal. After all, that sort of discipline is exactly how Warren Buffett became the most famous value investor of all time. When you’re willing to sit on $128 billion in cash because no investment looks appealing right now, you know you’re committed to the process.

But index funds don’t work like that. To minimize capital gains taxes, reduce fund turnover, and keep shareholders properly invested in the markets at all times, most index funds define value and growth in relative terms. And it’s usually a lot simpler than you may think. Here’s our previous market size stack with an additional split between value and growth.

Rather than messing with complicated decile definitions, index funds simply divide the total market capitalization in two*. The cheap half of the market is “value”, the expensive half of the market is “growth”, and “blend” is the entire market including both value and growth companies.

(*) It’s actually a little more complicated than that with extra features like trading bands to help prevent whiplash trades in volatile markets, which is one reason why your favorite index fund composition may not be as pure as the definitions imply. But that’s getting too into the weeds and the basic idea holds for how they initially allocate companies to an index.

Let the idea of relative valuation sink in, and the distinction between value indexing and deep value stock picking becomes clear. If the entire market is overvalued or undervalued by any objective definition, it doesn’t move the halfway line at all! Value index funds don’t chase fleeting academic value. They take what they can get in the most practical and cost-effective way possible and invest in the most undervalued half of the market at all times.

Portfolio Charts definitions

The goal of Portfolio Charts is to provide good data that is a reasonable approximation of real-world index fund definitions without getting too fund-specific. Technically speaking, the calculations are designed to match industry-standard CRSP and MSCI free-float cap-weighted methodologies and are bound by neither company count nor hard decile categories. You can read this if you want to know the details of how that’s done, but to keep it simple let’s stick to pictures. For example, here’s how I break things up by size:

In terms of value and growth, the Portfolio Charts numbers sort everything by book-to-market and allocate the cheap half of the market to value and the expensive half to growth. Following common index terminology, “blend” is simply the entire market in the given size category including both growth and value.

Combine it all together and it looks like this:

Hopefully if you’ve read this far that chart should make sense, but if you take only one thing away let it be this:

The numbers follow standard real-world fund methodologies and should model the vast majority of index funds within a reasonable margin of error.

Long story short, I’m a practical guy with a short attention span for academic arguments based on theoretical data with no foundation in the real world. And because of that, I do a lot of work behind the scenes researching index methodologies to keep the analysis grounded and useful. Good decisions require good data, and that’s what Portfolio Charts is all about.

Interesting implications

So we’ve talked about how index funds define size and value. And we’ve also covered how Portfolio Charts seeks to track the same practical methodologies. To build on that knowledge base, I think it would be helpful to discuss a few examples of how that information can be useful in everyday financial research.

Fama-French Data

While the numbers here are built in part using the popular factor research published by Pr. Ken French, it’s important to realize that the returns from my own calculations probably won’t match what you see in a paper on the topic. That’s because the Fama-French definitions of both size and value are very different from what you’ll find in any index fund. I have no problem with that at all, as their goal is to study contributing factors in market returns rather than accurately model an index. But it does make applying the numbers to your own investments more complicated than you might think.

I love the Fama-French research and reference it all the time, but I also understand what the authors are modeling and how to work with the data. Never assume that the default assumptions describe the behavior of a real-world index fund, and take any research that uses the data in that way with an appropriate grain of salt. You now know way more than the average investor about index definitions, so dig into the details and be smart about it!

Morningstar Style Box

One of the most widely used reference points for size and value distribution within a fund is the Morningstar style box. By illustrating the percentages of a fund in a 9-square grid, it allows you to quickly see the distribution of companies across size and value factors. For example, here’s the style box for VTI:

The tricky thing about using the style box, however, is to understand that the definitions may not match those of your favorite fund.

For example, the Morningstar valuation definition splits the market into three parts. Value represents the third of the market most undervalued, growth represents the third of the market most overvalued, and blend represents the middle part of the market classified as neither growth nor value. That’s not how most index funds define the same terms, so be careful about using that information to choose your funds. Contrary to the Morningstar definition, buying a “blend” fund in addition to value and growth usually adds no additional company diversification at all!

It’s also somewhat common for investors to complain that their favorite small cap fund became too weighted towards mid caps over time. And to be fair, if you look at the style box data the shift is indeed pretty noticeable. In reality, though, the only thing that usually changed was the Morningstar definition of “small”. They do periodically change methodologies, and that shift combined with definitions that differ from the funds they’re studying can easily cause confusion. For example, a change shrinking small caps to the bottom 10% of the market bumped many companies into mid-cap territory without them growing in size at all.

There also seems to be lots of conflicting information online about the style box size breakpoints. It’s probably just a result of articles written at different times, but there also could be variations in how different tools sort the same data. So when using it to study your own portfolio be sure to take the time to also compare definitions.

Large Cap Definitions

Speaking of definitions, the delineation between large and mid caps is not always so clean-cut. So let’s look again at the size compositions:

The first thing I’ll point out is that CRSP “Mega Caps” are actually normal large caps compared to most alternatives. And somewhat confusingly, they define “large” as inclusive of all mid caps. That definition is not reserved exclusively to CRSP, by the way, as MSCI refers to the combination of large and mid as a “standard” index that some fund providers may still refer to as “large cap”. And the S&P500 is normally classified as large cap but also has a few more mid caps than you might think. So if you own a large cap fund and are considering purchasing mid caps, just know that you may already be covered! When in doubt, read the prospectus for the true market coverage.

Detail-oriented investors using the Portfolio Charts tools might also consider adding a dash of mid caps to best match the proportion held in their large cap CRSP fund. It probably won’t make a huge difference in the numbers and I’ll wager you go back the simpler version, but the PC tools are designed for that type of flexibility for people who need it.

Selecting an Index Fund

One way to look at all of this information is to conclude that the specific definitions of size and value are very important, and to work really hard to find just the right index fund. But another way to look at it is that a bunch of extremely smart index managers all have different highly-educated ideas about definitions. Are you sure you know who is right?

Take that second perspective to heart, and a healthy mindest might be to simply acknowledge that there is always some amount of uncertainty in index returns. Most are on the same general page, and none are necessarily more “right” than others. There’s no point in getting caught up in the idea that recent returns are any indication of one index outperforming or under-performing another. Sometimes it’s just normal market noise.

I suggest learning to embrace that noise. Focus on the more important issues like where that index fits into your larger portfolio plan and how you’ll handle a little inevitable error in your financial projections, and you’ll be much better off in the long run than someone spending all of their energy on academic index perfection.

Portfolio Charts Small Caps

This may be the longest post ever to explain a simple update, but with the full explanation under our belts let’s finally circle back to the recent change in the definition of “small cap”. You can see the reasoning here.

The old small cap numbers covered the market space between 85% and 100%, but I recently cropped them at 98% to be consistent with the exclusion of the smallest micro caps (the purple band) from other small cap indices. While the end result should not be particularly noticeable in most portfolios, it’s all about doing my best to accurately replicate index methodologies as closely as possible.

Next Steps

As I’m sure you’ve gathered by now, index fund construction is a surprisingly complicated topic and is admittedly a lot of information to soak in at once. No worries! If you’re like me and most effectively learn by approaching problems from different angles, I have several resources that can help.

  • For deep information on how the Portfolio Charts stock calculations work, check out the Stock Index Calculator.
  • To learn more about index funds, including background information and a working list of ticker options for each asset on the site, visit the Index Funds page.
  • To experiment with different percentages of size and value funds to see how differing definitions can affect portfolio returns, play around with the My Portfolio tool.
  • And if you’re interested in a more detailed academic explanation of the size and value factors including why they matter and what contributes to their performance, I’m personally a fan of Your Complete Guide to Factor Based Investing by Andrew Berkin and Larry Swedroe. You can probably find it at your local library, and it’s a very educational read.

No matter whether you’re truly interested in the little details about how indices work or simply need reassurance that the data here at Portfolio Charts is well-researched and as realistic as possible, hopefully you learned what you needed! If you have any questions, feel free to contact me as I’m always happy to talk about good data. My ultimate goal isn’t to overwhelm anyone with little details that distract from the big picture, but to provide people with the knowledge they need to make educated decisions.

Be it big or small, value or growth, or just a single total market fund, there are lots of great portfolio options for all types of investors. Learn a little bit about how each of them work, and it will be that much easier to find your personal style and face the markets with confidence.

Did you learn something new about how index funds work?