WaMu's Toll on Oakmark

The government seizure of  Washington Mutual (WM) this week marks the largest bank failure in United States history and the end of a saga for embattled investor Bill Nygren. Nygren held (and was adding to) WaMu through June 30, 2008, according to reports for all of the funds he manages:  Oakmark I (OAKMX),  Oakmark Select (OAKLX), and  Oakmark Global Select (OAKWX).

The June 30 reports indicate that WaMu constituted 1.14% of Oakmark's portfolio, 4.85% of Oakmark Select, and 2.78% of Oakmark Global Select. The stock had absorbed a whopping 16% of Select's assets and 3.5% of Oakmark's assets in the middle of 2006 and occupied the top slot in both funds. A message on the Oakmark Web site now informs investors that Nygren, in fact, sold WaMu earlier this month and that he'll have more to report about it in his next shareholder letter.

Nygren's Thesis
Nygren devoted considerable space to WaMu in his 2004 year-end shareholder letter. He remarked that WaMu's mortgage-banking division had a disappointing year, causing earnings to fall short of expectations. Noting that he received more shareholder mail questioning his judgment on WaMu than on any other topic, Nygren went on to explain that WaMu's stock had a strong year, returning 10%, and that this reflected the success of its other business, retail banking.

Despite some poor quarters, Nygren noted that WaMu's shares were trading at around five times earnings in 1998, when he first purchased it in Oakmark, and paying out a $0.47-per-share dividend on $1.43 per share of earnings. Moreover, the dividend had increased more than threefold, and earnings had grown at a double-digit rate (using 2005 consensus numbers), as had retail banking deposits. Nygren concluded that although the stock had outperformed the market from 1998 through 2004, the stock later struggled because Wall Street analysts paid too much attention to a few quarters of disappointing earnings.

At the end of 2005, Nygren mentioned that WaMu could become an acquisition target, but he was satisfied to hold it based on his estimate of its value as a going concern. At various points in 2006, Nygren reported that WaMu was doing well. Indeed, the stock outperformed other banks, returning 9.6% for the year, though it trailed the index.

Early in 2007, Nygren again addressed concerns about WaMu, this time about its subprime exposure. He acknowledged that the subprime market negatively affected WaMu, but argued that WaMu had built a "very valuable retail bank" and that the stock "appeared inexpensive based on its low P/E, high yield and above average growth in retail banking income." He also noted that it was priced inexpensively compared with deposit premiums paid for other recently acquired banks. Then, Nygren made a worst-case scenario analysis, arguing that if 10% of the bank's loans experienced foreclosure and proceeds from the sales of the foreclosed homes were 20% lower than the mortgage amounts, the aftertax loss would still only be 10% of WaMu's earnings. Additionally, WaMu sold its riskier loans and kept its safer ones. Nygren simply couldn't envision the bank cracking under an even worse case than the one he laid out.

What Nygren Missed
In conversations with Morningstar, Nygren has acknowledged some mistakes regarding his assessment of financials. He understood that the housing market was slowing, but he didn't anticipate the amount of bad loans written. He also didn't anticipate house prices declining as much as they have. With defaults high and home prices low, Nygren also may have underestimated the extent to which institutions soured on holding mortgage-backed bonds. Even though mortgage-backed bonds are believed to be cheap in many quarters and their discounted cash flow values higher than their current market value, mathematical models are suggesting that institutions should not hold additional mortgage-backed bonds. Nygren may have underestimated how long the market for those securities would remain frozen and acknowledges that has hurt financials now. Additionally, Nygren likely didn't forecast the amount of credit card losses WaMu experienced recently. Finally, Nygren probably misjudged how a "run" on WaMu, or depositors removing their assets, would crush the bank.

What Can We Learn from the Financial Meltdown?
We don't mean to pick on Nygren, whose 10-year returns on Oakmark Select are stellar. He remains one of our favorite managers; Oakmark and Oakmark Select remain Morningstar Fund Analyst Picks in the large-blend category (although Select has tried our patience), and Oakmark Global Select looks promising to us, despite an inauspicious start in its young life. Nygren's arguments about WaMu and other financials have been repeated by other value managers heavily invested in banks and financial firms. Such managers include Richard Pzena at  JHancock Classic Value (PZFVX), Bill Miller at  Legg Mason Value Trust (LMVTX), and David Dreman of  DWS Dreman High Return (KDHAX). We also appreciate how candid Nygren has been in his shareholder letters, laying out his case over the years for WaMu and other investments. Most fund managers are not as open and honest with their shareholders.

However, as mutual fund analysts, we've learned that value managers, despite perfunctory genuflection to Benjamin Graham, sometimes aren't generous enough with their margin of safety when it comes to financials. Financials often trade at low P/Es and P/Bs precisely because it's impossible to judge the quality of the loans (and other assets) on their balance sheets. They look like cheap stocks at first blush, but that's because it's awfully hard to know what book value (the "B" in P/B) really is and how quickly the earnings (the "E" in P/E) can deteriorate. Consequently, value funds don't always provide the downside protection they're famous for.

Also, we've learned something about concentration. We have long disliked index-hugging funds--those that charge fees well above index funds but are constructed so closely to index funds that they have no hope of ever outperforming them. We've often thought that investors who are paying for active management should get it in the form of relatively concentrated funds of talented managers' best ideas. However, having 16% of Oakmark Select in WaMu turned out to be a bit much.