After earning a degree in finance from UC Berkeley, I was a market maker in Treasury options at the Chicago Board of Trade, and then began a PhD program in finance at the University of Chicago.  There was an opportunity to do research at Holt Value Associates, which was what I really wanted to do.  I started to work for Holt while going to school and by the end of the year was their Director of Research. In 1985 Daniel Obrycki and I started our own research company, with the goal of building a better mousetrap, which ultimately led to the development of the Economic Margin Framework, which today sits at the core of our analytical offerings.  In 1985 we had no clients and no understanding of how to market our services to potential clients.  We quickly realized we needed to effectively communicate our story, and it took a while to put together the right marketing plan to deliver our message to the investment community.  Today we have about 180 investment houses that subscribe to our services, including relationships with various international consulting houses that use our services to advise corporate executives on strategic management issues that affect security valuation.  AFG today has 18 people with principal offices in Chicago and Fresno, California, and satellite offices in Seattle and St. Louis.

The AFG 50 has outperfomed the S&P 500 since its creation, in June of 2004, by 16.94% (as of July 12, 2007).  A primary input to the security selection process is quantitative modeling done by the AFG.  Can you describe that modeling, and how it differs from other approaches that are used, for example, by mutual funds?

The Economic Margin Framework catches distortions in reported accounting information.  For example, Microsoft expenses over $6 billion on R&D each year.  Contrast this to a traditional bricks and mortar business that builds a factory and is able to amortize and those costs over the lifetime of the assets, sometimes over periods as long as 30 years.  While it is hard to argue that R&D costs are not generally long term investments, similar to a new factory, intellectual property companies have fundamentally different accounting approaches from traditional bricks and mortar companies, and analysts need to properly account for those differences in their models. There is no guarantee that R&D will pay off, but our model corrects for the underlying distortions by treating R&D as a long term investment.  In addition to adjusting for R&D, we build out a set of metrics that adjust for the age of the assets, the inflation-adjusted value of the assets, and the riskiness of the cash flows. 

After building a research business around our modeling, many of our clients came back to us and said ‘great,’ but can you put together a portfolio that reflects this research.  We said ‘ok’ and launched such a product in 2004 called the AFG 50.  We do quantitative research on approximately 4,500 securities, and combine that research with our analyst team to create this product. The AFG 50 is for clients that need a large cap product that beats the S&P 500, which we have done in each year of its 3 year history.  We choose only from stocks in the S&P 500 – we actively seek to avoid style drift.  We remain sector neutral to the S&P 500 (i.e., the AFG 50 is sector-weighted proportionally to the sector weightings in the S&P 500).  Our approach is to pick the best stocks in each sector.

The last piece of the puzzle is to have low turnover.  There are lots of trading strategies out there, but for the traditional advisor with HNW/UHNW clients, we need an approach that conveys peace of mind, even if the advisor is fee-based.  Since inception, turnover on the AFG 50 is running at approximately 20% per year.  Our fundamental goal is to offer a tax efficient, systematic process that is consistently replicable through time.  So far we have managed to achieve our goal of beating the S&P 500 by 200 to 300 basis points annually.

The Economic Margin Framework is based on discounted cash flow analysis, and the output is the intrinsic value of a company.  We provide many ancillary statistics, including the Economic Margin, which describes whether and the extent to which a company is creating or destroying value.

In addition to the quantitative modeling, you also apply qualitative measures to the security selection process.  Can you describe those measures, and how those differ from the approaches used elsewhere by market analysts?

Every day we run screens on the S&P 500 to identify buy candidates, which we then compare to the securities we already own (and which passed our screen originally).  Our analysts look at each security, assessing the economic risks faced by the company, any executive changes that might be significant, the characteristics of any acquisitions, and what has been happening with respect to quarterly earnings guidance.  With respect to acquisitions, which we generally do not like, we prefer that companies make ‘bolt-on’ acquisitions for existing business lines, rather than whole-scale paradigm shifts, as we see bolt-ons are more manageable and have superior track records.  With respect to earnings guidance, we look beyond the numbers themselves to see whether management is dealing effectively and realistically with any issues they face.  Lastly, our team looks closely at the competitive landscape.  For example, Apple just entered the phone business, so we looked closely at the effect this would have on Motorola, and concluded that Motorola would likely face significant structural operational issues.  At the same time, Cisco began showing up on our screens as being very attractively valued after a recent sell-off.  We moved to sell Motorola and add Cisco to the portfolio, illustrating how we combine qualitative and quantitative factors to make decisions. We find that when the qualitative and quantitative factors are aligned for a company, over the long run these stocks tend to outperform.  We are happy with this structure, as it is an approach that has worked well for us.

If I compare ourselves to a typical mutual fund, I believe we place much less emphasis on traditional valuation metrics such as P/E ratios, earnings growth rates, and price-to-book ratios.  These traditional metrics are flawed through the distortions I have mentioned in a company’s reported accounting statements.  For example, in 1995 Cisco was trading with a growth P/E in the high 20s or 30s.  But, when modeled in our framework, we saw that they were expensing an enormous amount of R&D, and the market did not realize Cisco’s latent capacity to generate cash flow.  It was truly a deep value stock.  Cisco went on to outperform the S&P 500 by 60% per year.  In July of 2000, our models told us that Cisco needed to continue to grow sales in excess of 60% per year just to maintain its current price, which was unrealistic in that environment, and we issued a sell order.

What sets us apart is our ability to deconstruct the as-reported accounting numbers to understand the true economics of a business.  It allows our clients to have confidence in the stocks we select, and to withstand short term volatility.  By design it is highly unlikely that we will be the best performing product in a given year, but rather we strive to  hit 'singles and doubles,' which might be boring, but will pay off handsomely over the long haul.

What is your personal role in the management of the AFG 50?

My partner and I act as advisors and quasi-portfolio managers, not as direct analysts.  The people that know the most about companies are our analysts that do the primary research.  Each analyst covers a sector, and presents his or her analysis and recommendations to the team on a weekly basis.  We lead the team, and make the final decisions when a security is bought or sold in the AFG 50.

How do advisors typically utilize the AFG 50 for their clients?  Do they use rebalancing software, with a model portfolio, to allocate holdings across many accounts?  Or do they create an internal “mutual fund” and allocate shares?

We create a research partnership between AFG and advisors.  Advisors are resource constrained; they don’t have analysts tracking each sector.  Our advisor clients have access to our analysts as an outsourced research team.  Our published reports go to clients, and clients come back to us with questions.  We want our clients to be in a position where, if one of their clients walks in, they can explain which securities they bought and why, and the rationale behind their investment approach.  Our greatest success with the AFG 50 product is working with advisors managing less than $1.5 Billion and our typical client is a retail RIA with about $500 million under management.

Because our clients typically serve HNW and UHNW clients, they often have to work around concentrated positions in a particular security.  We offer a backup list of securities that can be substituted for any security that is in the AFG 50.  These are stocks that are ready to be added to the AFG 50 but, for whatever reason, are not currently in it. 

Our clients have found a lot of different ways to incorporate the AFG 50 into their process.  Some use rebalancing software, which works well as they get new money that they want to efficiently allocate across clients and accounts.   Others do it the old fashioned way, with Excel spreadsheets.  Relative to a typical fund, we have low turnover.  Since 2004 we have made only 32 trades, which is about nine per year.  With less than 20% annual turnover, we expect to hold companies five to six years, and this minimizes the rebalancing issues.  It also takes out many of the concerns around short term volatility.  We sell something only when we see there is a structural difference vis-à-vis when we bought the stock; we don’t react to short term price movements.

Basically, the AFG 50 is a large cap core product. About 40% of our universe is growth, and 60% is value. But we don’t feel constrained to own any proportion of growth or value stocks, and freely shift between growth and value depending on where we see the greatest opportunities. 

Advisors are often concerned about the learning curve associated with the AFG 50.  While our approach is very rigorous and comprehensive, it is not rocket science.  We build on concepts most advisors learned in school or through company training programs.  Furthermore, to make the transition to our products painless, we offer a series of on-line training programs, and quarterly seminars, enabling an advisor to be up-to-speed in a couple of days.

What is the current AUM of assets that are driven by the AFG 50, and will you limit the amount of assets that you will support?

It is difficult to say, but I believe there is between $500 and $600 million directly managed through the AFG 50, and this is growing about 30% per year.  We have no limit in mind, and will play that by ear.  Our universe, the S&P 500, is deep, so this question is premature, and we are just scratching the surface.  It would be very different if the product were in a small or micro cap segment.

Why not create a hedge fund or a mutual fund of your own, based on the AFG 50?

AFG is a research firm and I place great value on that charter, and our record of client service and retention.  I don’t see fund management as fitting directly within this research model.  However, we have a majority interest in a fund management company, Toreador Research and Trading, which offers the Toreador Large Cap Fund (TORLX), with the mandate to beat the Russell 1000.  While TORLX has a broader mandate than the AFG 50, its selection process s based on the same Economic Margin Framework.  The fund has an overlap of about 30% to 40% with the AFG 50.  This has appealed to our smaller clients, who do not have a need for the AFG 50.   It is managed with the same principles and low turnover, and as clients grow they often move to the AFG 50.

Lastly, can you give us your current thoughts on the relative valuations in the market today and what sectors look attractive?

We are still in a market of extraordinarily low risk premiums.  We have entered a period where the 10 year Treasury crossed 5% and the price of oil is going up.  The same scenario existed a year ago.  Who would have thought, a year ago, that we would see a 20% rise in the markets?  But that said, there are many factors that lead us to believe US stocks are still an attractive asset class.  Strength in international markets a default positive for US stocks.  Junk credit spreads continue to be near all time lows.  Lastly, stocks are still priced to return 4% to 5% above Treasury Bonds, which we feel is more than fair in today’s environment.  I would not put money into high yield debt, and believe that large cap stocks look attractive relative to small caps.  There may be a short term sell off, but five years from now I believe that the US markets, in particular large cap stocks, will have done well.  I would not be rattled by short term volatility.

Note: The AFG50™ is not a security; it is a proprietary composite index of 50 securities selected by AFG based on AFG’s proprietary research process. 

Display article as PDF for printing.

Would you like to send this article to a friend?


Remember, if you have a question or comment, send it to .