The Mechanics of Value Investing

By Jack Forehand (@practicalquant) —

We all have a tendency to want to put investing strategies into groups. That can sometimes be misleading, though. There are more investment products than I can count that use the word value in their name. But beneath the surface, many of those products couldn’t be any more different. Those differences can lead to substantially different outcomes in both the long- and short-term, so it is very important to understand the investment process and parameters used to create any value portfolio before you invest in it. Since I often discuss value investing in my articles, I thought it might be helpful to take a step back and take a look at some of the important details to consider when building or analyzing any value strategy.

Defining Value

But first, I think it is important to define value. When I look at strategies that are referred to as value, I break them down into three distinct categories.

[1] Ben Graham Value – This group of strategies focuses on buying the cheapest stocks. They can use very different metrics to get there, but the end result is a group of companies that are cheap and out of favor. When we talk about the academic research into value investing, this is the group we are talking about. Most academic work will focus on buying the bottom decile (or cheapest 10%) of stocks.

[2] Warren Buffett Value – This group combines value with some sort of measure of quality. The goal is not to buy the cheapest stocks, but rather to buy good companies that are trading at a discount to some measure of their intrinsic value with the hope that the profits these firms make will compound over time. These types of approaches generally end up with more expensive stocks (vs the Graham style above) and stray further away from the academic definition of value.

[3] Bill Miller Value – This style was popularized by Miller in the 90s when he argued that many tech companies were value investments despite the fact that they were very expensive by any traditional value metric. The argument proponents of this type of approach will make is that if you look at companies as the present value of expected future cash flows, then companies with strong projected future growth can be cheap today, even if they trade at high multiples based on things like sales and earnings. These types of strategies stray the furthest from the academic definition of value (they actually have almost nothing in common with it at all).

When I use the term value, I am always referring to the academic definition (#1 above) and mostly discussing it in the context of systematic value models, but you will see the term used in varying ways so I think it is important to establish a clear definition first.

Once you define value, you then need to make a series of important decisions to take that concept and express it as an actual investment portfolio. Here are a few of the most important things that go into the recipe of building a value strategy and what to keep your eye on for each of them.

The Process of Building a Value Portfolio

What is the Universe?

This is a very important question that many people miss. The data shows that the value premium is magnified in smaller stocks, so including them should boost long-term returns. But including smaller stocks also limits the ability of a fund to scale and increases tracking error relative to the S&P 500, which makes it harder for investors to stick with the strategy.

Which Metric(s) to Use

There are a wide range of options here. Most of the academic research into value relies of the Price/Book ratio and many large value firms like DFA use it as their primary value metric, but recent research from O’Shaughnessy Asset Management has shown that in a world where the vast majority of assets are intangible and not included in Price/Book, this ratio may not be very applicable. The most popular metrics outside of Price/Book are Price/Earnings, Price/Sales, Price/Cash Flow and EV/EBITDA. All of these metrics have their strengths and their weaknesses. For example, the PE ratio is subject to earnings manipulation and the Price/Sales treats profitable firms the same as unprofitable firms. Even EV/EBITDA, which is currently one of the favorites of many practitioners, can favor capital intensive firms over firms that have less capital requirements. The fact that each metric has its own inherent flaws leads some to conclude that using a composite of all of them is probably the best option. But either way, if you invest in a value strategy, it is important to understand what you are getting because different metrics can produce very different outcomes (just ask anyone who has followed Price/Book for the past decade).

How Many Stocks to Hold

Like many things in investing, there is no hard and fast rule here, but in general, the more focused a portfolio is (within reason), the more exposure to the value factor it has, and the better the long-term returns will likely be. But with more focus comes more tracking error, which means that focused portfolios will often deviate widely from the market and will go through extended periods of underperformance that make them very hard for investors to stick with. The past decade is a great example of this. Focused portfolios also can’t handle as many assets, and fewer assets equal fewer fees for an investment manager. This is why many value managers will turn their portfolios into closet indexes with hundreds of stocks. That isn’t necessarily a bad thing as long as the fees are similar to the fees of investing in the index, but you would be surprised how many funds are still out there with 1% fees and a portfolio that looks just like the index. These are the portfolios investors should avoid.

How Often to Rebalance

With factor-based portfolios, how often you rebalance can have a significant effect on returns. Momentum portfolios typically require more frequent rebalancing to produce their best performance, but value can do well with quarterly or annual rebalancing. In fact, many of the value models we follow like our Value Investor model based on Ben Graham perform best when rebalanced annually.

How to Handle Industry Concentration

Some value managers feel that concentrating in sectors and industries that are cheap enhances returns. Others feel that value works best when it is used to pick individual stocks and not to make industry bets. I don’t think there is one correct answer, but strategies that will concentrate in specific sectors do tend to have higher tracking error, which means they can be tough to stick with.

How to Weight Individual Positions

Many quantitative managers tend to equal weight their portfolios, but some will weight their highest conviction positions more heavily. Closet index funds typically market cap weight so they don’t stray too far from the benchmark. Others will use things like inverse volatility weighting, which overweights the least volatile stocks in an effort to have each position contribute an equal risk to the portfolio.

This certainly isn’t an all-inclusive list, and there are many other decisions that go into building a value portfolio. In the end, some of the smartest investors in the world differ on the answers to these questions so my point here is not to suggest there is one correct way to do it. The point is that like any investing strategy, it is important to understand what you are getting with a value strategy before you invest. Doing the work up front can prevent surprises and help you find a fund, manager or strategy that best fits your investing goals.