This is John Rotonti's third interview with Bill Nygren, the portfolio manager of three Oakmark Funds and the chief investment officer for U.S. equities at Harris Associates (Oakmark's parent company). By now, then, it's no secret that John considers Bill one of the all-time value investing greats. (You can read the first two interviews here and here.)
Oakmark Funds was named Lipper's best equity large fund group in both 2015 and 2016, and Harris Associates won the same award in 2014. The Oakmark Global Select Fund, one of three that Nygren co-manages, won the 2018 Lipper Fund Award in the Global Large-Cap Value Fund category for the three-year and five-year periods.
In this interview, John and Bill discuss how the Oakmark investing team only models out seven years (and therefore does not use the phrase "long-term compounding machine"). They also discuss Apple (NASDAQ:AAPL), Alphabet (NASDAQ:GOOG)(NASDAQ:GOOGL), Mastercard (NYSE:MA), Visa (NYSE:V), Moody's (NYSE:MCO), bitcoin, and more.
John Rotonti: The Oakmark Fund (5-star rating by Morningstar) returned 18% in 2016 and 21.1% in 2017. Since inception in August 1991, the Oakmark Fund has generated 13.3% annualized returns, compared to 10% for the S&P 500. What is driving these impressive results both in the last couple of years and since inception?
Bill Nygren: Investment discipline and people.
Every quarter, my investment commentary piece starts with the same statement: "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."
I tell people we bring a private equity perspective to public equity investing. What private equity is typically good at is identifying value that is beyond the one- or two-year time horizon of most investors. That's the same thing we do -- forecast how the business is likely to change over the next five to seven years, and identify value that other investors have missed.
As for our people, we have several important advantages:
- Value is the only type of investing our firm does. I can't imagine how different the environment is for an analyst who serves a value portfolio manager one day and a momentum manager the next.
- We have tremendous depth. I always tell people that to really understand Oakmark, you'll learn much more about why we've done so well by talking to our 10th best analyst than by talking to David or me. [Note: Nygren is referring to David Herro, the chief investment officer for international equities at Harris Associates. Herro was also named Morningstar International's stock fund manager of the year in 2006 and 2016, and Morningstar International's stock fund manager of the decade for 2000-2009.]
- Continuity of our investment team. Throughout our history, including one-and-a-half generational transitions, at every point in time our top investment team has always worked together for more than a decade. Only by knowing each other so well are we able to get the maximum value from our research output.
JR: What is your definition of a business compounder? Which long-term secular tailwinds do you find most compelling?
BN: "Compounder" is a word we don't use. For all of our investments and potential investments, we create two years of specific estimates for both the income statement and balance sheet. Then, we apply a five-year EPS growth rate that incorporates both growth in earnings and reinvestment of excess cash flow (or, in some unusual cases, additional financing needs). So, for all companies, we have a compound growth estimate for EPS going out seven years.
When we assume significantly above average compound growth, it tends to be for businesses that have high ROIC, that could benefit from secular change and that have significant advantages over their future competitors. Some examples from our current investments include advertising moving online away from traditional media (Alphabet), the move for payment with plastic instead of cash (Mastercard, Visa), video consumption via streaming instead of linear TV (Netflix (NASDAQ:NFLX)), making travel plans online instead of with a local travel agent (Booking Holdings (NASDAQ:BKNG)) and outsourcing of building maintenance (CBRE Group (NYSE: CBRE)).
JR: Mastercard shares are up about 55% in the past year, and they trade at a next-12-month P/E of about 30 (according to S&P Global Market Intelligence). What is your investment thesis on Mastercard? Is the share price still reasonably valued?
BN: With a company growing as fast as Mastercard, one needs to really pay attention to which P/E they use. Mastercard trades at about 39 times last year's earnings and at about 30 times this year's consensus EPS, and that falls further to about 25 times consensus 2019 earnings. For the five years after that, we expect Mastercard to grow EPS at a 13% rate. Using our P/E model, that suggests Mastercard's fair P/E next year would be over 20% above its current P/E on 2019 earnings. The basic thesis is that world consumption will grow, plastic will take share from cash and excess cash generation will enable share repurchases that will help EPS growth exceed net income growth.
JR: Are you more excited by Visa's acquisition of Visa Europe or Mastercard's acquisition of VocaLink?
BN: They are obviously very different acquisitions, and we like them both. Given how much bigger Visa Europe is than VocaLink, I'd have to say it added more to our sell target. But both Mastercard and Visa have made nice tuck-in acquisitions of new electronic transfer technology, which increases our confidence that both companies will be the leaders in this business for many years.
JR: Square currently accepts Visa and Mastercard. Can you describe the competitive dynamic between Square and Visa and Mastercard, and does Square present a long-term threat to the two largest payment processors?
BN: Most of the innovation in payment transfer has been at the consumer or the merchant level, rather than on the pipes that connect them. Square, Apple Pay, and PayPal (NASDAQ:PYPL) to name the larger ones, each present a way to pay with your Mastercard or Visa that is potentially more convenient for the customer. I think what all these companies have found is that the fee to use Mastercard or Visa is well worth it, especially considering their industry-leading fraud technology both employ.
JR: What is a fair price-to-earnings (P/E) multiple to pay for shares of the highest-quality growth business in the world?
BN: I would simply say that we don't find current P/E to be a very useful indicator of value for rapidly growing businesses. Allow me to explain by an example. One of our holdings, Alphabet, owns YouTube. So to make an educated guess of Alphabet's value, we have to take a stab at valuing YouTube. We believe that creating and selling video content is a great business. The number of hours people spend watching YouTube videos has about doubled every two years, whereas publicly traded competitors, like cable networks, have been slowly declining. That would suggest that YouTube deserves a much higher P/E than the networks do. But because YouTube is currently heavily reinvesting to maximize its growth, estimates are that it makes little or no current income. The closer those current earnings are to zero, the closer the fair current P/E is to infinity.
Instead of working from current income, we project out over our seven-year horizon: what will hours viewed likely be in seven years, how much can an average hour be monetized, and what does that imply for earnings (assuming the super normal growth spending has ceased)? For almost all businesses, our crystal ball goes dark after seven years, so we assume all businesses trade at similar P/Es after seven years. With an estimate of fair value seven years in the future, we can discount that back at an appropriate risk-adjusted discount rate to estimate today's value. Whether that results in a near infinite or a negative P/E on current earnings is not of concern to us.
JR: Do you incorporate the P/E-to-growth (PEG) ratio into your valuation analysis? If so, what PEG do you look to pay for high-quality growing businesses?
BN: We don't use PEG ratios, though we agree with the concept that, for most businesses, higher growth leads to a higher fair-value P/E. We establish a target EBITA multiple for the business based on its growth, margins and capital requirements. What that model shows is that the fair multiple is typically more sensitive to changes in ROIC than it is to changes in sales growth.
JR: In the Oakmark Select Fund, you have a 7% position in CBRE Group. What is your investment thesis?
BN: CBRE is the market leader for providing services to owners of commercial real estate. The stock trades at a discount P/E to the S&P 500. Most analysts think that CBRE is a transactions-based company due to its history as the leading commercial real estate broker. Brokerage businesses tend to be quite cyclical, especially if the underlying market that is served (commercial real estate) is cyclical. Were that CBRE's only business, we would agree that the stock probably doesn't deserve a premium multiple. What that misses is the rapidly growing outsourcing business that CBRE has developed. Most large companies would prefer to not spend time taking care of the real estate they occupy. For a recurring fee, as opposed to transaction-based fee, CBRE will manage the properties, including services like building maintenance. Because of its scale, CBRE can usually offer these services not only as a convenience but as a cost save relative to do-it-yourself. With the growth of this business, we expect CBRE to be priced at a P/E similar to industrial services companies, and that tends to be a premium to the market P/E.
Last, I think many value investors have avoided CBRE because it doesn't allocate much capital to share repurchase. Instead, capital is spent acquiring smaller real estate service providers. We believe that after cost synergies these acquisitions are of comparable valuation to share repurchase. They also grow CBRE's scale and, therefore, its competitive moat.
JR: Not too long ago, Apple was dramatically undervalued at 10 times earnings. But Apple still only trades at a forward P/E of about 15, which is a steep discount to the S&P 500. Does Apple deserve to trade at a higher P/E multiple? If so, why?
BN: We own Apple, so I have to believe it deserves to sell at a higher multiple than it currently sells. I'll turn the question around: Adjusted for cash, Apple sells at about 80% of the S&P 500 P/E multiple. How much worse do you think Apple is than the average business?
JR: Apple has about $160 billion in net cash, and Alphabet has about $100 billion. What should they do with all that cash sitting on their balance sheets?
BN: We want any company to invest cash in opportunities that are the most NPV [net present value] positive using a realistic cost of capital. One input we like to see in the cost of capital calculation is the return on share repurchases. If we are correct and the stock is selling well below intrinsic value, then the company's cost of capital will be higher because an easy high return alternative is to grow per-share value by shrinking the denominator. When investment opportunities with returns above the cost of capital don't exist, the best thing to do is to return cash to the owners. We would always be pleased to see our companies, including Apple and Alphabet, use cash for share repurchase because, by definition, we believe the stocks we own are undervalued. If management doesn't believe the stock is undervalued, they should return the cash as dividends.
Unfortunately, companies seem to get attacked in the press when they return capital. The alternative is that mature businesses make sub-par investments, which is worse for the economy than returning capital to investors who can then invest in other companies that are better positioned to capitalize on growth opportunities. Share repurchases fuel a natural cycle: A business like IBM matures and begins to distribute most of its cash flow to its shareholders, and those shareholders then use that capital to fund start-ups like Facebook (NASDAQ:FB) or Netflix. Where would we be today if IBM had kept all of its capital instead of giving it back to investors who in turn funded new technologies and companies?
JR: When should a company pay a special dividend?
BN: When a company generates excess cash flow, beyond what can be invested above its cost of capital, the value maximizing move is to return that capital to shareholders. If, in addition, the stock is selling at a discount to its fair value, we would prefer that cash be returned via share repurchase rather than dividends. When the cash to be returned via dividends is a predictable annual amount, a regular dividend is fine. When it is a cyclically inflated amount, or when it comes from a non-recurring event (such as sale of a business) we would prefer a special dividend.
Cyclical businesses are perfect candidates for specials. The stocks tend to be cheap when business conditions are poor and there is no excess cash flow, and are often overvalued when conditions are good and earnings are high. Rather than repurchasing overvalued stock with the excess cash, we would much prefer a special dividend.
JR: I see that you recently bought shares of Priceline, now called Booking Holdings. Priceline's returns on invested capital (ROIC) have been falling, according to S&P Global Market Intelligence. I think this is because Priceline is getting lower return on investment on its advertising spend and because newer business lines and acquisitions are simply not as profitable as booking.com (which it bought for a song and which provides Priceline with a massive competitive advantage in Europe). Are there any other reasons that explain its falling ROICs?
BN: In 1963, a player for MLB's then Houston Colt .45s, John Paciorek, became the first player to retire with a career 1.000 batting average with more than two at bats. After his five trips to the plate, he had three hits and two walks. Due to injuries, with his 1.000 average intact, he never again stepped into the batter's box. Does that make him the most successful hitter in baseball history? I think most reasonable analysts would say no. The definition of success is not based solely on career average, but on the total number of hits above what an average player would have accomplished in the same number of at bats.
Applying the baseball analogy to a business, a declining return on invested capital isn't cause for alarm so long as the newly deployed capital is also earning an above average return. Booking Holdings has grown its scale by a factor of 10x over the past decade. Yes, return on shareholders' equity has declined from a peak of over 40% to a low 30% number, but total equity has grown by 5x. So, despite diluting returns somewhat, the new capital has been invested at a much higher return than its cost of capital. Two additional specifics for Booking are that cash is now a large component of shareholders' equity (with very low returns), and its investment in shares of Ctrip (NASDAQ:TCOM) also earn lower returns than the base business. We don't see the lower returns as any indication of problems at Booking.
I'll take Ted Williams over John Paciorek any day!
JR: Some (perhaps many) traditional value investors avoided investing in technology stocks because rapid technological change could lead to risk of obsolescence (or disruption) and because those investors aimed to pay a discount to the current earnings power of the business without paying anything for the potential growth. But you and your team stand out as value investors that have done exceptionally well buying stock in technology companies. You made about 100% in both Amazon (NASDAQ:AMZN) and LinkedIn, and you've got substantial unrealized gains in Alphabet. Why do you think Oakmark has done so well investing in technology companies while also sticking to your value investing principles?
BN: I think the view that two high school kids in a garage could disrupt leading technology companies was prevalent in the '80s and has been largely dead since, except in the value investor community. A generation ago, Warren Buffett said that technology stocks were outside his circle of competence. Ever since, it has become somewhat of a badge of honor among value investors to say, "I don't buy tech."
It's ironic that the three examples you cite as Oakmark's successes in technology are companies we didn't think of as tech businesses. We bought Amazon because on enterprise value-to-sales it was cheaper than other retailers, LinkedIn because as an employment placement firm it was cheaper on EV/Sales than almost any other business services firm and Alphabet because on a sum-of-the-parts basis it was priced as if Google was cheaper than other advertising-supported media companies.
For years at Oakmark, we have used sum-of-the-parts valuation models for multi-divisional industrial or consumer companies. That isn't really unique at all today. What is somewhat different is using that same valuation approach for technology companies. Our Alphabet model, for example, separately values Search, Cash, Other Bets and YouTube. Anyone just looking at the nearly 30x P/E on Alphabet misses all those other assets.
JR: What is the biggest risk facing big tech?
BN: We have about 25% of the Oakmark Fund portfolio in technology, so this is a question I get asked frequently. My answer is to compare our "tech risk" to our financials risk. We have nearly 30% in financials, and most of that is invested in levered lenders. Credit quality is a big risk for Ally Financial Citigroup (NYSE:ALLY), Bank of America (NYSE:BAC), JPMorgan Chase (NYSE:JPM), Capital One Financial: (NYSE:COF), and Wells Fargo (NYSE:WFC). That 15% of our portfolio tends to move together--higher on good credit news and lower on bad news.
Our tech exposure isn't nearly as monolithic. About 5% is MasterCard and Visa, which I think of as tied to consumer spending, Alphabet is shy of 4% and is effectively an advertising company, Apple at just under 3% is a consumer product company, Automatic Data Processing (NASDAQ:ADP) (2%) is a payroll processor that moves up or down with employment and TE Connectivity (NYSE:TEL) is over 2% and moves more like an auto supplier than a tech company. So nearly two-thirds of our industry exposure isn't really what I would call technology, though they all certainly use a lot of technology.
I guess Intel (NASDAQ:INTC) and Oracle (NYSE:ORCL) have some overlapping risk with faster conversion to cloud being good for both of them and vice versa, but they are the exception among our tech investments. I just don't see a common industry risk running throughout our investments in this sector.
JR: What are Moody's competitive advantages and what would have to happen for it to become a larger position in your portfolio?
BN: Moody's makes most of its income from rating bonds. I guess anyone could theoretically produce an algorithm that would assess riskiness of default comparable to what Moody's does. But "Bill's Bond Ratings," even if it was as statistically accurate as Moody's, would not be recognized by bond buyers. The bond buying community looks for ratings by S&P (NYSE:SPGI) and Moody's (and to a lesser extent Fitch) and has very little interest in how other services rate bonds. That is the moat. Additionally, in its analytics division, which is much smaller than credit ratings, Moody's has become the go-to for regulatory filings, especially for banks and other financial firms. The rapid pace of regulatory change makes this a good "economies of scale" business--Moody's does the work once to be up to date on all changes, instead of each bank duplicating the same work.
We would increase our weighting in Moody's for the same reasons we would increase any holding: either we get more confident that our view of fundamentals is playing out, or we get a price that is more attractive.
JR: What are your thoughts on current stock market valuations?
BN: At Oakmark, we aren't market timers, so we don't spend much time looking at stock market valuations. We start with the belief that, over time, there is more money to be made by owning businesses than by lending to them. We set sell targets for each of our investments and potential investments that are based on providing an appropriate risk-adjusted return relative to corporate bonds, assuming our forecasts are exactly right. We have never found a time in our history, or today, when we haven't been able to fully populate a portfolio with stocks that were selling below our sell targets.
Today the S&P multiple is a little higher than its historic average, but both bond and cash yields are also less attractive than their historic averages. Paraphrasing Warren Buffett, would you rather own a business at a P/E of 18 whose earnings will likely grow, or a bond at a P/E of 35 with earnings that will absolutely not grow?
JR: Any thoughts on bitcoin?
BN: I don't own Bitcoin, and Oakmark doesn't own Bitcoin. If I were smarter, I'd probably stop right there and say we don't comment on investments we don't own. But I can't help myself.
I think the fascination with Bitcoin is simply due to the remarkable increase in price it has enjoyed. Big price changes are newsworthy, and people get entertainment value from following volatile trading. Bitcoin has no relevance to the vast majority of people. It isn't an investment, in the sense that you can't create a DCF to estimate its value. It is designed to be a store of value (like gold) that is more reliable than government-issued currency. The tremendous price volatility that Bitcoin has experienced suggests to me that it is failing at its main purpose.