Buffett Clarifies Retained Earnings Policy

Published on March 3, 2010

Warren Buffett includes an “Owner’s Manual” for Berkshire Hathaway shareholders in each annual report which is also available separately on the company’s web site.  The Owner’s Manual does not change very often which is appropriate since it is supposed to communicate basic business principles that are not likely to change each year.  For this reason, it was easy to miss a change that has significant implications for Berkshire Hathaway’s earnings retention policy going forward.

Original Retained Earnings Test

The following statement has been documented as business principle #9 ever since Mr. Buffett published the Owner’s Manual in 1983:

We feel noble intentions should be checked periodically against results. We test the wisdom of retaining earnings by assessing whether retention, over time, delivers shareholders at least $1 of market value for each $1 retained. To date, this test has been met. We will continue to apply it on a five-year rolling basis. As our net worth grows, it is more difficult to use retained earnings wisely.

The basic meaning of this business principle is that earnings retention must, in the long run, deliver at least $1 in market value to shareholders for each $1 that management retains.  It had the virtue of simplicity and was also very easy to measure.  Anyone can calculate Berkshire’s retained earnings for a five year rolling period and then examine whether the retained earnings resulted in a corresponding rise in market value.

Limitations With Original Principle

While the principle is simple and measurable, there are clearly problems with the way it is formulated.  It is obvious that over the past decade, valuation extremes were common for companies both on the upside and downside.  As Mr. Buffett noted in his latest shareholder letter, Microsoft CEO Steve Ballmer and General Electric CEO Jeff Immelt both had the misfortune of taking over as CEO near the peak of a bubble in their company’s stock valuation.  Just as it is difficult to evaluate the performance of these CEOs over the past decade based on share price performance alone, it is difficult to evaluate the wisdom of earnings retention using the same standard.

Berkshire’s Modified Earnings Retention Test

The updated version of Berkshire Hathaway’s earnings test reads as follows:

I should have written the “five-year rolling basis” sentence differently, an error I didn’t realize until I received a question about this subject at the 2009 annual meeting.  When the stock market has declined sharply over a five-year stretch, our market-price premium to book value has sometimes shrunk. And when that happens, we fail the test as I improperly formulated it. In fact, we fell far short as early as 1971-75, well before I wrote this principle in 1983.

The five-year test should be: (1) during the period did our book-value gain exceed the performance of the S&P; and (2) did our stock consistently sell at a premium to book, meaning that every $1 of retained earnings was always worth more than $1? If these tests are met, retaining earnings has made sense.

It must be noted that any modification of a long standing test that recently failed should be treated with a healthy dose of skepticism.  Is management changing the test due to a legitimate problem in the formulation of the original wording or is the goal line simply being moved?  The fact that the CEO is Warren Buffett does not mean that this question should not be asked.

Mr. Buffett’s argument is that the original test was improperly formulated because markets can remain extreme for a long period of time, which is certainly true.  During such times, Berkshire’s price to book value often falls.  This is certainly the case as we noted in The Rational Walk’s Berkshire Hathaway Briefing Book. As Mr. Buffett notes, this also happened during the early 1970s far before he formulated the Owner’s Manual principles.

Allowing Mr. Market to dictate earnings retention policy even over a five year period can cause unintended consequences.  For example, Berkshire Hathaway failed to meet the test at the market lows in 2009.  A strict interpretation of the original rule would have forced a dividend in February or March 2009 and would have limited the capital available to Mr. Buffett to take advantage of opportunities caused by the market crash.  This would not have served shareholder interests.

Does the New Rule Make Sense?

The new retention principle says that the litmus test should be whether Berkshire’s book value gain exceeded the performance of the S&P 500 and whether the stock consistently sells at a premium to book – meaning a price to book ratio of at least 1.  Based on this formulation, Berkshire would have passed the earnings retention test even at the 2009 lows.

One obvious problem with the new rule is that book value is only a rough proxy of changes in Berkshire’s intrinsic value, as Mr. Buffett himself tells us in his shareholder letters.  In addition, Mr. Buffett has told us that intrinsic value far exceeds book value.  From a directional standpoint, changes in book value are likely to signal changes in intrinsic value, but a price to book ratio of 1.0 or 1.1 is almost sure to signal an undervaluation of Berkshire shares.

Under the new rule, future managements at Berkshire could argue that earnings should be retained under the new test even if the price to book value is only slightly above 1.0 provided that the change in book value over a five year period at least exceeds the S&P 500 change.

One other objection is that looking at Berkshire’s overall price to book value ratio does not measure the wisdom of retention of incremental capital.  It is perfectly possible to have value destroying earnings retention coincide with maintenance of a price to book value ratio well in excess of 1.0 because of the cumulative effect of decades of good decisions that have created the bulk of the intrinsic value.  At the margin, earnings retention could still destroy value while the price to book ratio remains above 1.0, although below what it otherwise might have been without earnings retention.

No Substitute for Management Judgment

The bottom line is that few shareholders would have wanted Mr. Buffett to declare a dividend in March 2009.  Shareholders trust his judgment based on his cumulative history at Berkshire and are willing to grant a huge amount of latitude based simply on the track record.

The problem with attempting to define this type of rule is that some element of management judgment is always going to be required when deciding on earnings retention policy.  Only after a period of time passes will shareholders be able to evaluate whether the retained earnings created value or not.  Future CEOs at Berkshire Hathaway will find it impossible to alter any of the business principles in any way whatsoever because they will be accused of trying to modify the company culture.  Therefore, Berkshire shareholders must be comfortable with these principles as they apply to the next CEO, not just Mr. Buffett.

The fact that Berkshire Hathaway has an Owner’s Manual with clear principles is a great example for other companies to follow but the recent revision to the earnings retention test demonstrates the inherent limitations associated with static principles that meet the irrational behavior of Mr. Market.

The author owns shares of Berkshire Hathaway and is the author of The Rational Walk’s Berkshire Hathaway 2010 Briefing Book which provides a detailed analysis of the company along with estimates of intrinsic value.

Buffett Clarifies Retained Earnings Policy