The Haugen Model

A Proven Model for Predicting Expected Returns

Based on the work of Dr. Bob Haugen, the Haugen Model has an extremely strong track record of success. Today at Haugen Equity Signals, Bob Haugen’s work is being continued at the same high level by Brian Boyer, PhD, who works as a professor of finance at Brigham Young University, who routinely develops new ideas for improving the model. These ideas are back-tested and vetted for data mining and other biases before being incorporated into the model that we use.


Tracking four different groups of stocks

The Haugen model assesses how each stock in a given universe of stocks is affected by over 60 different factors at a given point in time. These factors include measurements of:

  • Risk
  • Liquidity
  • Stock price relative to several measures of corporate income and cash flow
  • Profitability
  • Stock return history
  • Analysts' estimates
  • Macroeconomic influences
  • Sector influences

Haugen Equity Signals’ clients receive weekly expected return data for four different groups of stocks:

  • US Model (ranks expected returns for more than 4,100 US stocks)
  • European Model (ranks expected returns for more than 3,100 European stocks)
  • Japanese Model (ranks expected returns for more than 3,000 Japanese stocks)
  • Emerging Markets Model (ranks expected returns for more than 3,300 stocks in 24 emerging countries)

In addition, customized/proprietary versions of the Haugen Model can be built; please inquire about details and pricing.

How The Model Works

  • Each year Haugen Equity Signals creates a universe of stocks in each of the four markets that we track (i.e. US, European, Japanese and emerging markets) and determines the top stocks in each group based on market capitalization and liquidity.
  • At the end of each month the stocks are put through advanced statistical modeling algorithms, producing a projected “payoff” or “score” for over 60 different factors for each stock. For example, if in a given month the payoff for the book-to-price factor is negative, this means that in that month the model is favoring companies with low book-to-price values over stocks with high book-to-price values. Payoffs for all factors are estimated simultaneously.
  • To calculate the expected returns, we get the latest data in order to look at each stock’s current exposure level to each factor, multiply by that month’s projected payoffs for each factor, and then add these 60 numbers up. This is done using the following formula:

Projected payoff for factor #1 for all stocks for the coming month

Each stock's current exposure to factor #1

#1's contribution to expected return for each stock for the coming month

  • The model computes this formula for all factors and then sums all factor components for each stock.
  • Finally we rank each stock based on its expected return for the coming month.

Benefits of the Model

Unlike statistical models with a fixed set of rules for picking stocks, the Haugen model with its huge array of more than 60 factors is able to replicate the kind of “feel for the markets” that experienced money managers demonstrate.

What makes it so powerful? It does what no human can – it analyzes the impact of 60+ factors on the expected returns of more than 4,000 stocks simultaneously. This enables it to forecast future performance on the basis of constantly changing conditions.


Flexibility is therefore one of the key benefits of our expected return factor model. It can:

  • Be used to improve portfolio performance by accurately predicting near term winners and losers within a group of value or growth stocks
  • Enhance the payoffs to long, short and market neutral strategies
  • Respond effectively to changing market conditions

The model accurately projects not only size changes in the payoff to a given factor, but also sign changes. For example, payoffs to stocks paying relatively high dividends might be positive one month, but negative the next. As a result, you can predict with a high degree of accuracy a forthcoming change in the payoff to a given stock. This enables you to more accurately time portfolio adjustments, i.e., to buy at bargain prices and sell at the top of a slide.

Our model has also been shown repeatedly to respond to current market conditions. In the long run, the model favors portfolios of stocks that are in the aggregate relatively cheap, profitable, liquid, and minimally volatile. However, the Haugen model will, over shorter periods, shift from favoring value, to growth, to growth at a reasonable price.

Our model is even flexible enough to predict a rise in the price of a particular energy stock, while "judging" that the sector as a whole will go down. In the hands of a savvy investment professional, the Haugen model is clearly a winning tool.

High Performance Profile

Bob Haugen spent his career studying what pays off in the market and why. One of his studies, titled Case Closed which evaluated U.S. Stock performance between 1963 and 2007, validated his contention that the market is inefficient.

The study showed that throughout the entire period, the characteristics of stocks in the top performing 10% (Decile 10) are the mirror image of the lowest expected return stocks in Decile 1.


The highest expected return stocks consistently demonstrate this surprising profile:

  • Low risk
  • Highly profitable companies
  • Improving profitability
  • Large capitalization
  • High liquidity
  • Inexpensive stock price relative to earnings, dividends and cash flow

Evaluating the Model

If you’re considering becoming a client of Haugen Equity Signals, here are a few ways to evaluate our performance:

Against real world outcomes Investars.com has audited our monthly results for more than 10 years against about 20 other research organizations. Over the long term, Haugen has consistently ranked in the top three. We are often ranked #1.

Through hypothetical portfolios The chart below examines hypothetical portfolios constructed from our real-time signals. Keep in mind that we don’t predict whether the market is going up or down. Instead, we predict the relative returns of roughly the largest 4,100 US stocks by market cap.

Each month the model ranks these from lowest to highest expected return. Then we divide the 4,100 stocks into deciles, which amounts to 10 portfolios of 410 stocks each. At the end of each month we compare our predictions to the realized returns.

Consistently, decile 10 outperforms decile 1 and over a ten year period, as you can see from the chart below, each decile outperforms the one below it. This is powerful evidence that the model is doing exactly what it is designed to do.


Our Performance: Decile vs S&P 500 Returns
January 1996 Through December 2021
Cumulative Values of $1 Invested

Lowest
$1.74
Second
$6.83
Third
$9.44
Fourth
$8.19
Fifth
$16.19
Sixth
$26.52
Seventh
$25.38
Eighth
$39.46
Ninth
$48.07
Highest
$148.17
S&P 500
$11.05

We can also tell you that once you’re a client, you’re likely to remain one. We still have clients who have been with us since the inception of the company more than a decade ago. That’s because the model has proven its ability to enhance almost any fund manager’s performance, regardless of investment strategy.

Please see the Performance section of this site for more statistics and evaluation tools. Please feel free to contact us!

Pricing

For one monthly fee you will have access to the data from all four markets that we track: our US Model, European Model, Japanese Model and Emerging Markets model.

4 Pricing Options

Haugen Equity Signals offers a choice of four different pricing structures, all of which include end of month data along with periodic updates for all of the stocks that we track. Choose the pricing option that works best for you...a set subscription fee, a flat fee based on your AUM, a fee based on the difference that our data makes in your portfolio’s performance, or a fee you pay only if you make money:

  •  Standardized Pricing – A set fee for your subscription OR
  •  Flat fee – Based on your assets under management OR
  •  Flat fee + performance-based – Based on your assets under management (lower than the first option) plus your portfolio’s performance OR
  •  Performance-based – A portion of the Alpha produced for your portfolio by our model.

In addition, customized/proprietary versions of the Haugen Model can be built. You can also have an additional exclusive weekly or monthly run of the data on the trading day of your choice, giving you data that no one else gets. Please inquire about details and pricing for these options.