Bill Nygren’s market commentary for the second quarter ended June 30, 2016.
Shortly after I moved to Chicago, a good friend introduced me to Earl the ticket broker. Having grown up in Minnesota, I didn’t even know what a ticket broker was. If you wanted to attend an event, you bought tickets from the box office far in advance. And you paid face value. But that wasn’t very effective for an analyst in the investment management industry, given the long and unpredictable hours. It was pointless to buy tickets months in advance and then miss the event because I had to work late. I often found out if an evening would be free only a day in advance. Of course by then, all the popular events were sold out.
That’s where Earl came in. He knew who had tickets and who was willing to sell them. Earl could get tickets to anything. And I gladly paid his premium price so that the few events I attended could be the ones I was most interested in. Earl had proprietary knowledge and was able to earn a good living from it.
Long after StubHub came on the scene, out of loyalty, I still used Earl. But eventually, StubHub became much cheaper and easier to use. Like most of the old school ticket brokers, Earl closed up shop. Before StubHub, the cost to track down ticket owners who were sellers was prohibitive. With StubHub, that information was one free search away.
A concept that for years has guided our investment thinking is that the Internet makes access to information easier and cheaper. And free-flowing information is bad news for intermediaries. The list of casualties goes far beyond ticket brokers. A travel agent who visited resorts has been replaced by reviews on TripAdvisor. Until recently, a limo service collected half the fare for matching a driver with a rider; now Uber provides that match at a fraction of the price. Retailers could enjoy high margins because comparison shopping was time consuming; now sites like Amazon find the lowest priced seller. Salesforces spun data to make their products look superior; now websites offer unbiased price/performance comparisons. Consumer wins; intermediary loses.
One of our new purchases this past quarter, LinkedIn, has had a similar effect on the employment industry. In the past a good corporate recruiter maintained a proprietary database of employees who were considering changing employers. Within a few weeks of retaining this recruiter, a company could see a flow of candidates that matched its job description. As compensation for that service, the recruiter would receive a percentage of the new hire’s salary. With LinkedIn, a company can now pay a much less expensive subscription fee and see a superior talent pool dominated by candidates not currently looking to change jobs, fully searchable on data the employer chooses. And all of this is available within a few clicks. Lower price, faster service, no adverse selection. If you are a middleman who can be replicated by a computer algorithm, you’re in trouble.
Last month, I sat on a panel discussing the future of investment management amid the growth of passive strategies. As we discussed the evolution of index funds and ETFs, it occurred to me that the asset management business had many similarities to other middlemen being bypassed. When index funds started 40 years ago, much of the asset management industry was charging one percent of assets each year to buy and hold a broadly diversified portfolio of high-quality, large-cap stocks. That allowed the investor to access the higher returns equities achieved compared to bonds, and it was preferable to a stock broker who lost that advantage through frequent trading. But then, index funds came along, which mimicked that same portfolio at almost no cost.
To differentiate themselves from index funds, managers began to define their philosophies as value or growth. Then the indexes followed, establishing sub-indexes to track the “value” and “growth” components separately. Today at very low cost, one can buy an ETF or passive fund designed to track almost any investment style or industry. If you think high-yielding stocks are attractive today (we don’t), you can buy a low-cost ETF that tracks them.
Bill Nygren – Active Management
On the panel, I went through my usual arguments about active management: that value can be added either by providing higher return or by taking less risk, but if you define risk as performing differently than the market, you won’t recognize value added through risk reduction. In years when the S&P 500 has declined, the Oakmark Fund has never declined by as much. Yet some conclude the Fund is riskier than the market because our results don’t closely track the market. I think of risk in terms of losing money; most consultants don’t.
I also talked about how professional sports are the same “zero sum games” as active management, but in sports we all agree that there is some predictability of how different teams will perform. Shouldn’t concepts like talent differential, payroll, winning philosophy, team culture, team depth and investment in player development apply just as much to investment management as to sports?
Eventually I talked about a few stocks and areas of the market that we view as unusually attractive. The concepts won’t surprise you as they are heavily represented in our portfolios. We think banks are cheap because investors are overweighting the memory of a once-a-generation financial collapse; commodity prices need to rise to encourage producers to meet increasing demand but the stocks are priced based on current commodity prices; emerging market growth will exceed developed market growth yet companies with strong exposure to those markets aren’t priced at premiums.
Then another panelist made a statement that to me sounded like fingernails on a chalkboard: “I agree with Bill that ‘value stocks’ are cheap, and that’s why I own our value ETF.” Is there ever a time when “value stocks” aren’t cheap? If so, what does the term “value” even mean? One of my biggest pet peeves is when consultants label companies with the most desirable characteristics “growth stocks” and deem companies with the least desirable characteristics “value stocks,” and their distinction is independent of stock price. But this panelist’s statement helped crystalize for me what we at Oakmark do that a computer doesn’t.
It’s easy for a computer to create a basket of high P/E stocks and another of low P/E—or, for that matter, high versus low on a price-to-anything metric. And it is easy to see when the premium for the “high” stocks gets large relative to history, as my fellow panelist had observed. But very few of our investments at Oakmark rest solely on the idea that a currently low P/E will normalize. Value, to us, is much more complex than whether the trailing P/E ratio is above or below average.
A stock like Alphabet (formerly Google) isn’t likely owned in a value ETF due to its growth rate and P/E ratio both being higher than average. But the P/E ratio is giving no credit to non-earning assets such as its large cash balance, the value of its venture cap portfolio (like autonomous cars) or YouTube—which, based on hours of viewing, would be worth hundreds of dollars per share if valued like cable