Berkshire’s Repurchase Policy: Too Little, Too Late?

Published on November 4, 2018

Berkshire Hathaway acquired $928 million of its own stock during the third quarter of 2018 through a series of purchases of Class A and Class B shares from August 7 to 24.  A repurchase of this size, relative to the company’s cash flow and market capitalization, would normally attract relatively little attention.  However, Berkshire has maintained an unusual policy with respect to return of shareholder capital in general and stock repurchases in particular.

Rational capital allocation is a problem that many companies never adequately resolve because the skill set of management is often more attuned to solving operational problems rather than allocating free cash flow.  However, Berkshire shareholders have benefitted from Warren Buffett’s superior operational and capital allocation skills for decades.  For almost all of Berkshire’s history, there was never any question regarding whether capital should be retained within the company.  Few shareholders would have been better off receiving dividends or selling their shares back to the company and reinvesting the after-tax proceeds elsewhere.  Mr. Buffett’s investment prowess fully justified full retention of all free cash flow generated by the business.

In 2015, Berkshire marked its 50th anniversary of Mr. Buffett taking full control of the company.  This was a natural time to look back on the history of Berkshire and, more importantly, to take a look at where the company might be a decade in the future:

The implications of Berkshire continuing to retain all earnings over the next decade while growing book value per share at a compound rate of approximately 10 percent are staggering.  If we take Berkshire’s 2015 net earnings of $24 billion as a baseline, reinvestment of all earnings would need to result in enough incremental earnings power to generate approximately $62 billion of net income for Berkshire by 2025.  We would expect retained earnings to increase by about $420 billion over the next decade.  Berkshire’s shareholders’ equity would approximate $675 billion by the end of 2025 based on these assumptions.  With this kind of track record, the market would most likely value Berkshire in excess of $1 trillion.

The idea of a trillion dollar company is no longer as novel today as it was a few years ago, but the numbers are still staggering.  We also pointed out in the article that Berkshire’s repurchase policy may well need to be changed in order to allocate a portion of the massive free cash flow likely to accrue over the coming decade and noted that the novel repurchase policy adopted by Berkshire could make it difficult to deploy cash via repurchases.  Berkshire had, for several years, limited repurchases to times when shares could be repurchased under a fixed price-to-book ratio which had been set at 120% since late 2012 after being established at 110% in 2011.

For critics of Berkshire’s repurchase policy, Berkshire’s updated repurchase policy announced on July 17, 2018 was welcome news.  A fixed price-to-book ratio limit was eliminated in favor of allowing repurchases anytime both Warren Buffett and Charlie Munger believe that the stock is trading “below Berkshire’s intrinsic value, conservatively determined” and when doing so would not reduce Berkshire’s holdings of cash and equivalents below $20 billion.  The news was very surprising to the market with shares closing up over 5 percent on July 18.  One caveat was that repurchases under the new policy would not be permitted until after Berkshire released second quarter earnings.  As a result, the first day that repurchases were permitted under the new policy was August 6.

Let’s take a look at Berkshire’s repurchase activity under the new policy.  Here is an excerpt from the recently released 10-Q report showing repurchase activity during the quarter:

It helps to put these repurchases into proper context by looking at Berkshire’s trading activity during that time.  The charts below, presented separately for Class A and Class B shares, were derived using data from Yahoo Finance:

We can see from the charts that there were two significant bumps in Berkshire’s trading price following the revision to the repurchase policy on July 17.  On July 18, the stock price rose significantly due to market expectations that repurchases were more likely to occur and, on August 6, the stock rose again in response to the release of second quarter earnings.  These price increases did not deter Mr. Buffett from repurchasing close to 9 percent of the company’s trading volume, in aggregate, over the period from August 7 to 24.  This is not to say that he regularly purchased some fixed percentage of volume, although perhaps he did, and repurchases could have been concentrated on specific days.  However, the average price paid of $312,807/A and $207.09/B is pretty typical for that range of dates.

After August 24, the stock price rose to a level where Mr. Buffett decided to halt repurchase activity.  We know that no further shares were repurchased between August 25 and September 30.  However, we can infer that additional shares were repurchased between October 1 and October 25.  We know this because Berkshire is required to release a recent share count along with its 10-Q report.  On the first page of the recently released 10-Q, we see the following:

Given the conversion ratio of 1,500 Class B shares per Class A share, the Class A equivalent share count on October 25 was 1,641,681 shares whereas the Class A equivalent share count was 1,642,269 shares on September 30.  Berkshire typically issues a very small number of shares every year so we cannot simply subtract the October 25 share count from the September 30 share count.  However, we can infer that at least the difference of 588 A equivalents were repurchased between October 1 and October 25.  This is a relatively small deployment of capital – probably around $185 million.  Obviously shares were available during this timeframe well under the levels paid during the August repurchases.

What can we infer from these repurchases regarding what Mr. Buffett and Mr. Munger think of Berkshire’s intrinsic value, “conservatively calculated”?  Since Berkshire’s repurchase policy was based on book value for several years, it is logical to start by considering what price-to-book ratio was paid for the August repurchases.  Mr. Buffett made a point to wait until after second quarter results were released prior to initiating any repurchases.  June 30, 2018 book value per share was $217,677/A and $145.12/B.  The average price paid for the August repurchases was 1.437x book value for the A shares and 1.427x book value for the B shares.  This is substantially higher than the prior 1.2x book value limit and quite surprising.

We do not know the price or timing of the repurchases that took place from October 1 to 25, and there are sometimes “quiet periods” that limit companies from freely repurchasing stock prior to releasing earnings.  However, we know that Berkshire traded below the average level of the August repurchases from October 11 to 25.  Additionally, we know that Berkshire’s book value rose to $228,712/A and $152.47/B as of September 30.  Berkshire’s A class stock price on Friday, November 2 closed at $308,411 which is slightly under 1.35x book value.  With the caveat that the recent stock market correction has put pressure on Berkshire’s book value due to its large holding of marketable securities, it seems like we can still infer that Mr. Buffett and Mr. Munger view Berkshire’s stock to be trading below conservatively calculated intrinsic value.

Although we are still awaiting the release of Berkshire’s 13-F report which will reveal the company’s U.S. equity portfolio as of September 30, astute observers have already inferred from the 10-Q that Berkshire was a heavy buyer of common stocks during the second quarter.  Although an analysis of what Berkshire may have been doing in its equity portfolio during the quarter is beyond the immediate scope of this article, it is worth pointing out that the scale of Berkshire’s repurchases are quite small compared to its overall free cash flow and its allocation of capital into the stock of other companies.  This could be due to a variety of factors.  Mr. Buffett and Mr. Munger could believe that Berkshire is undervalued but not as undervalued as other opportunities in public markets.  It could also be due to the relative lack of trading volume in Berkshire’s stock which could put a cap on how much Berkshire can repurchase without pushing up the price dramatically.

Shareholders who were hoping for very large repurchases relative to Berkshire’s market capitalization or free cash flow will probably be disappointed with the magnitude of actual repurchase activity in the third quarter.  It could be that the peculiarities of Berkshire’s shareholder base and trading volume will make it difficult to execute large repurchases and that activity will be more modest.  However, over long periods of time, a significant reduction of the share count is still a possibility.

One cannot help but wonder how much lower Berkshire’s share count might be today had the present repurchase policy been adopted in 2011 rather than the policy that imposed the 110 percent of book value limit.  We now know that Mr. Buffett and Mr. Munger view a price below 140 percent of book value as a definite bargain.  Although we must note that the ratio between intrinsic value and book value can change over time, we can see that the price-to-book has been below 140 percent for substantial periods since 2011:

Would Berkshire shareholders have been better off with aggressive repurchase activity at various times since 2011 given that the stock price has traded below management’s view of intrinsic value for much of this time?  This is a complicated question to answer because obviously Berkshire has utilized its resources for other purposes during the past several years.  However, the presence of a large amount of undeployed cash on the balance sheet for a long period of time has signaled a dearth of opportunities and, for much of this time, management was unable to capitalize on an undervalued bargain in the form of the company’s own stock due to the constraints imposed by the 2011 and 2012 repurchase authorizations.

The updated repurchase policy is well overdue and hopefully will not be “too little, too late” when it comes to Berkshire’s overall capital allocation over the next decade.  We should, however, note that this new repurchase policy places no more of a “floor” on the stock price than the old policy.  Markets can and will do crazy things and Berkshire’s stock will certainly trade at depressed levels at times in the future.  Now, management has full flexibility to take advantage of such times.  Despite occasional criticism by the uniformed and/or politically motivated, repurchases can be a very intelligent way to deploy capital under the right circumstances.  The key issue is to avoid overpaying and there seems to be little risk of that happening at Berkshire.

It is also worth noting that for many longtime shareholders with a low cost basis, Berkshire represents a particularly tax advantaged way of compounding wealth.  Allocating capital to repurchases rather than dividends, as long as repurchases are below intrinsic value, will have the effect of increasing unrealized capital gains and continue tax efficient compounding for continuing shareholders.  The alternative of receiving dividends would result in immediate tax consequences and interrupt the snowballing tax deferred compounding machine known as Berkshire Hathaway.

Disclosure:  Individuals associated with The Rational Walk LLC own shares of Berkshire Hathaway.

Berkshire’s Repurchase Policy: Too Little, Too Late?