Value investing specialist Bill Nygren is a partner at Harris Associates and a portfolio manager of three funds: the Oakmark Fund, Oakmark Select Fund, and Oakmark Global Select Fund. Oakmark Funds was named Lipper's best equity large fund group in both 2015 and 2016, and Harris Associates (Oakmark's parent company) won the same award in 2014, an indication that the firm has a winning investing team, philosophy, process, and culture.

In this interview, he discusses the qualities he sees in great analysts and portfolio managers and shares his thoughts on the big banks, Alphabet, and more.

John Rotonti: Which qualities or skills should great analysts have? What about great portfolio managers? Are they different?

Bill Nygren. Image source: Oakmark Funds.

Bill Nygren: The answer depends on how each firm uses their analysts and portfolio managers. At our company, analysts are generalists and therefore migrate to areas of the market with the greatest perceived opportunity. That is very different than a firm that assigns an analyst to cover one industry and basically just rank order by attractiveness the companies in the industry he or she covers. Our analysts are, effectively, doing what portfolio managers do at other firms: attempting to identify the most attractive parts of the market at any given time.

Warren Buffett has said that investors don't need math skills beyond eighth-grade algebra, and though true, I think that understates the importance of math skills. Good analysts need to have great intuitive math skills. To a good analyst, it should sound like fingernails on a chalkboard when numbers don't make sense. Good analysts also need intellectual curiosity, they need the confidence to question conventional thinking, and unless they are only investing their own money, they need great communication skills to give decision-makers the conviction needed to invest in their recommendations. Last, they need a good dose of humility, which eventually comes with experience working in a profession where half of your decisions will prove to be wrong.

Our portfolio managers are making fairly modest changes to the portfolio of ideas the analysts have researched that have survived the process of making our approved list of investments. The portfolio manager's job is to tailor that list to meet any client-specific needs, client communication, tax management and so on.

We have many analysts who also work as portfolio managers on various Oakmark portfolios. In that hybrid role, I think the biggest hurdle for a good analyst in becoming a good portfolio manager is learning to not favor the ideas they have done the research on. We have a rule in our meetings for Oakmark and Oakmark Select that nobody can lobby for adding or increasing the weighting in any stock they personally cover. My door is always open, and like any analyst they can promote their own stocks most any time, but not during meetings when they are performing the role of a portfolio manager.

Do you use an investing checklist?

Our checklist is brief -- just three items -- good business, good price, and good management. It is printed at the top of my commentary to shareholders in each quarterly report:

"At Oakmark, we are long-term investors. We attempt to identify growing businesses that are managed to benefit their shareholders. We will purchase stock in those businesses only when priced substantially below our estimate of intrinsic value. After purchase, we patiently wait for the gap between stock price and intrinsic value to close."

When should a company pay a dividend or repurchase stock?

A company should return capital to its owners when that adds more to per-share value than would reinvestment in the business. With the exception of those few companies that have supernormal growth opportunities, we believe companies should be consistently returning capital to their owners. Our preference for how that capital is returned is repurchase of undervalued shares since that adds more value than a taxable dividend would. If the shares are fully valued or overvalued, then a dividend is preferable to repurchase.

How should value investors try to avoid value traps?

I think the path of least resistance when you are wrong on a stock is to say that the declining stock price has more than reflected the fundamental shortfall. When stocks fall on bad news, the typical analyst response is to say that the stock is a better value now than it was before it fell. We've gone back and analyzed stocks we've recommended, and we've found that companies that start to fall short of our analysts' fundamental expectations typically continue to fall short both on fundamentals and stock price. Those that meet our expectations are likely to continue doing so. Statistically speaking, the best course of action is to admit the mistake early, sell the stock, and revisit it at a later date. That is much easier said than done.

How do you evaluate a company's balance sheet? Do you look for particular coverage or debt ratios?

We don't have targets for debt ratios. As long-term investors, we want to see a balance sheet strong enough to make it highly likely the company can survive for the long term. That can be especially important for cyclical companies that tend to lose money when the economy is weak. During recessions, we typically have confidence the economy will get better but are not confident of when that will happen. So the ability to survive an extended downturn becomes critical.

When we value companies, we value the business separate from the balance sheet. Once we've estimated the value of the business, we will add excess cash and other non-earning assets and then deduct claims ahead of the equity investor, such as debt or preferred stock.

We will reduce our normal position sizes to offset a levered capital structure as opposed to making levered businesses off limits.

What do you think of the big banks?

Our thesis on banks (and AIG (AIG 1.85%)) is that investors have properly penalized them for increased capital requirements but have improperly ignored the offsetting positive of lower risk. If a company keeps its business the same but doubles its capital, the business becomes half as risky. Many investors fear that an overzealous regulatory environment will end up making big banks and insurers become like utilities. Well, utilities sell at about 1.5 times book value and most banks are selling well under book.

How do you calculate free cash flow at the big banks? Is GAAP earnings a good proxy?

We generally believe GAAP income is a reasonable estimate of cash flow. Where we've been wrong the past two years is that capital requirements kept moving higher, so less cash was available for dividends or repurchase. We don't believe capital requirements will keep moving higher and we don't believe interest rates will stay near zero. In a rate environment we think of as normal (interest rates slightly higher than inflation), we believe these companies can earn 10% on equity and if they don't have organic growth opportunities, can return all of it to shareholders. In a market selling at an upper-teens P/E multiple, it seems logical that these large financials could sell at mid-teens multiples.

What do you think about Google parent company Alphabet?

A lot of investors take a superficial look at Alphabet (GOOG -1.24%) (GOOGL -1.36%) and say the stock sells for $780 and it only earned $23 last year, so the P/E of 34 is about twice the market. Search is a great business, but is it really worth more than 2 times the market multiple? Using the same sum-of-the-parts analysis we've used on so many companies, we come to a very different conclusion.

First, Alphabet is sitting on a huge mound of cash, about $120 per share by the end of this year, and like most cash, it is earning next to nothing. So valuing Alphabet with just a P/E multiple is effectively valuing that cash at zero.

Next, Alphabet is investing heavily in venture cap type projects such as autonomous vehicles. This year alone they invested about $4.5 billion in such projects. Had they made those investments through a venture capital firm, that $4.5 billion would show as an asset on the balance sheet, and there would be no effect on the income statement. However, since they made those investments themselves, the cost flows through the income statement, and reduces reported income by about $4 per share. The investor looking only at P/E is effectively valuing these businesses at negative $136 per share ($4 of losses multiplied by the 34x P/E ratio).

Finally, at YouTube, Alphabet management has made the decision to not fully monetize viewing hours in exchange for more rapid growth of those hours. By making most content free to view and not embedding as many ads as on traditional TV, YouTube has been able to enjoy 40%-plus annual growth in hours viewed while traditional TV viewing is in decline. Nobody knows what this business will look like at maturity, but if current hours of YouTube viewing were valued comparably to how the stock market values hours of viewing for the cable networks, YouTube would be worth several hundred dollars per share.

So we add the cash, a guess at the value of the cumulative investments in other bets, and a guess at the value of YouTube, and subtract that from the market price of Alphabet. We don't think it is a stretch to say you get half the current price of Alphabet in assets other than the Google search business. If you look at the implied market price of the Google division and divide that by the earnings of just the Google division, you get a Google P/E that is much less than the market on expected earnings. I don't pretend to have a precise estimate of the appropriate P/E for Google, but given the strong tailwind of advertising moving online from traditional media, I think it is very difficult to argue that Google doesn't deserve an above-market P/E multiple. Add in the value of Alphabet's other assets, and we believe it is bargain-priced.

As an aside, I don't think growth investors are as accustomed to using sum-of-the-parts models as value investors are, so that lack of familiarity creates an opportunity for us.

I think it's admirable that you have more than $1 million invested in each of the funds you manage. Is it common for Harris and Oakmark employees to invest significantly in the funds they work for?

I've always been surprised by how few mutual fund managers are anxious to invest their personal assets side by side with their shareholders. Only after regulations required disclosure of personal investment were some managers shamed into finally investing in their own funds.

At Harris Associates, one of the reasons we started the Oakmark Fund back in 1991 was that the younger investment professionals wanted the opportunity to invest their own capital in the same manner we were investing for our clients. We launched each subsequent fund only after investment professionals had expressed their desire to invest personal assets. People often ask why we don't offer a bond fund, given how aggressive investors are today to get yield. The answer has nothing to do with whether or not that fund would be saleable. We don't offer it because none of us think we'd get returns high enough to justify investing our own capital.

Each year we voluntarily disclose the total investment of our employees, their families and Oakmark trustees in our funds. As of December that number totaled $440 million. Across all of our funds, the portfolio managers have the bulk of their investable capital in the funds they manage. When we invest in companies, we expect management to be heavily invested in their own stock. You should expect nothing less of us or your other fund managers. For myself, I invest in all seven of our funds, but my investments in the three I manage (Oakmark, Oakmark Select, Oakmark Global Select) constitute the overwhelming majority of my investable assets.

How do you narrow down your investable universe?

For starters, on the U.S. side of our business, we limit our universe to companies either based in the U.S. or conducting a substantial part of their business inside the U.S. We don't do that out of xenophobia, but rather out of respect for our peers on the international side of our firm. David Herro manages an outstanding international team for Oakmark that primarily invests in companies headquartered outside of the U.S. For non-U.S. investing, I'd urge you to strongly consider Oakmark's offerings of international and global funds.

Further reductions in the investable universe depend on goals of each specific fund, but generally revolve around liquidity and suitability. For Oakmark Select, as an example, we want typical position sizes of about 4% of assets because we are targeting a 20-stock portfolio. We also want reasonable liquidity in each position. If we want 4% of our assets in a stock without owning more than 4% of the outstanding shares, that means we can't buy companies with a smaller market cap than our total NAV. Today we have about $5 billion in Select, so we only look at companies larger than that.

In the Oakmark Fund, we are focused more on risk than we are in Select because many of our investors use it as their sole domestic fund. To keep the risk lower, we only invest in big businesses. Many studies of risk have, incorrectly in our opinion, concluded that large-cap stocks, usually defined as the 250 largest market caps, are the least risky. While often true, at important turning points such as 2000 when overvalued small technology companies achieved large-cap status, investors were hit by the double whammy of risky small businesses combined with excessive valuations. Large-cap investors in 2000 were in a very risky spot.

We believe that the risk reduction associated with "large" is due to business characteristics, not valuations. So we limit Oakmark investments to large businesses as measured by the 250 largest in sales, earnings or book value. These two sets of businesses are largely overlapping; they don't sum to 750 companies, but more like 300 to 400. Eliminating the ones with below average per-share growth prospects, poor management, or excessive valuations then gets us down to about 100 big businesses on our approved list. About half of those names make it into the portfolio.

How do you measure the performance of the analysts who work for Oakmark?

Our goal is to compensate analysts based on the returns their investment recommendations achieve for our clients. The difficulty comes in distinguishing between good luck and a job well done. Serendipity can create short-term results that are not reflective of long-term ability. To fight that, we compensate the newest analysts primarily based on our qualitative assessment of their output: How sound is the analysis? How well is it communicated? How does their quantity of output compare to other analysts? At the other end of the spectrum, our longest-tenured analysts are compensated primarily on the results of their stock recommendations over a multi-year period.

Harris Associates and Oakmark Funds recently announced that on January 1, you will be named chief investment officer (CIO) for U.S. equities. Congratulations! You also previously served as Director of Research (DoR) at Harris Associates for eight years. How do the roles and responsibilities of the DoR and CIO differ at your firm?

Thank you. Our company has a very horizontal structure with blurry lines of responsibility between our investment leaders. In his role as Director of Research, Win Murray is responsible for identifying and hiring talented investors, evaluating their contributions, motivating them, allocating the workload, and instituting policies to maximize consistency across the department. Our U.S. Chief Investment Officer, Bob Levy -- who is retiring at year-end -- chairs our weekly Stock Selection Group meetings and is the external face to Harris Associates clients. Additionally, it is his responsibility to make sure the investment team is getting proper support from the other departments at our company. Our investment leaders like Clyde McGregor and myself also work with Win and Bob to make sure we are always looking for ways to tweak our systems to improve them. Thanks to the great job Bob has done, I inherit no pent-up need for change and instead have the luxury of starting next year knowing that business as usual is a good outcome.

We've chatted on various occasions. No one has been as generous with his time and knowledge as you have. Is there a question about Oakmark I've missed?

Thank you -- I enjoy sharing our thinking with anyone who is interested in hearing it. But that isn't just being altruistic. One of the biggest difficulties in managing money today is that investors have developed very short time frames -- much shorter than the five-to-seven-year time horizon we use to evaluate the companies we invest in. I think the main reason investors have become so impatient is they are bombarded with short-term performance data and are encouraged to almost constantly reevaluate their investments. They tend to shoot themselves in the feet buying after a stock or fund has gone up, then selling after it has gone down. Rapid in-and-out trading also increases the cost of managing a fund, and those costs are borne by the fund shareholders.

By sharing with investors and potential investors how we think, rather than just how we've performed in the past quarter, we attempt to get investors to understand and believe in our process. An investor who appreciates the thinking that goes into our decisions is more likely to be a long-term investor than a performance chaser. So when you see Oakmark getting awards for industry-leading shareholder communication, know that we believe better informed shareholders are less likely to suffer from self-inflicted wounds. And what is good for our shareholders is good for our business.