Invest Like Buffett: No Looking Back

Includes: BRK.A, IBM, JNJ, KO, PG
by: Eli Inkrot

Following in the investing footsteps of Warren Buffett is kind of like riding in the back seat of a car: It's relatively simple, you're buckled up for the long haul and you usually get to where you want to go. Problem is, if one were to invest like Buffett, they would need significant insight into his upcoming equity selections. Without this formidable foresight, we are left with Buffett and Graham proxies such as the understanding that "time is the friend of a wonderful company," but only within the appropriate margin of safety.

A forefront example came up recently with Buffett's purchase of IBM (NYSE:IBM). He had been buying IBM shares for 8 months, but he didn't have to disclose the information to the public until he was good and ready. Buffett's cost was about $167 a share, while investor's today are looking at a near all-time high for IBM around $194; this certainly would not be investing like Buffett. More than the timing issue that results from following Buffett's disclosures is the fact that Warren is able to conjure up massive deals that you or I simply could not get a piece of. The special circumstances with Bank of America (NYSE:BAC) and Goldman Sachs (NYSE:GS) come to mind.

However, there is a reason that so many look to the Oracle of Omaha for an applicable investing thesis. Surely most flock to Omaha each year not for the Omaha Steaks (they ship those, you know) but instead for the investing prowess of a storied track record. Without Buffett's magnificent mark on the investing world, I would imagine most would fall out of tune to Warren's song. But that doesn't mean that his thesis would be any less attractive. In today's CNBC age of "5 stocks you have to buy by 11 AM and sell by lunch," Buffett's "in it for the longest of terms" almost seems contrarian. Obviously being contrarian just to be different is foolish, but it can be refreshing to slow things done a bit and focus on the eventual. Logic is on your side.

Buffett has so many simple yet eloquent quotes to choose from; however, I believe this one captures the essence of long-term investing best: "Only buy something that you'd be perfectly happy to hold if the market shut down for 10 years."

This quote can be taken quite literally with Berkshire Hathaway's (NYSE:BRK.A) buy-and-hold stalwarts such as Coca-Cola (NYSE:KO) and Wells Fargo (NYSE:WFC). But fundamentally the idea is to purely consider your investing options for the long-term. If you think you might regret your investment in a week or a month or even in a couple years, then you should probably think twice. On the other hand, if you feel comfortable owning an investment for the next couple of decades, initiating at a given price for a given reason, then you might just have something.

I was recently browsing the histories of a few of my holdings within the Berkshire Hathaway (BRK.A) annual letters. We have in common the typical companies dividend growth investors love, firms like Coca-Cola (KO), Procter & Gamble (NYSE:PG) and Johnson & Johnson (NYSE:JNJ). The first two companies fit the Buffett mantra quite well: Coca-Cola has long been known as "Buffett's favorite holding," yet he hasn't initiated new capital in about 15 years. Similarly, Buffett backed into Procter & Gamble (PG) through his Gillette holdings and has held course since. However, Johnson & Johnson (JNJ) stood out as having a relatively interesting pattern.

It isn't necessarily interesting why Buffett owns Johnson & Johnson. After all, it has a wide and tangible moat that was been able to not only pay dividends but also increase them for the last 49 years. More than that, if a kid skins his knee, you can bet Moms will continue to rely on Band-Aids and Neosporin. If you have a headache, the go-to is Tylenol.

The interesting part is that over the 2006 to 2010 period, Buffett initiated, bought more, sold, sold again and rebought shares of JNJ. Sure doesn't sound like the market shut down for 10 years to me. Here's a consolidated look at Berkshire Hathaway's JNJ holdings over that particular period.

Berkshire Hathaway's JNJ Holdings
Year Shares Cost Avg. Cost per Share
2006 21,334,900 $1,250,000,000 $58.59
2007 64,271,948 $3,943,000,000 $61.35
2008 30,009,591 $1,847,000,000 $61.55
2009 28,530,467 $1,724,000,000 $60.43
2010 45,022,563 $2,749,000,000 $61.06

At first glance the $61.06 average JNJ share cost doesn't appear to be that far from today's price at just over $65 a share. A lot of people make this mistake, as is seen regularly in BRK-A's ConocoPhillips (NYSE:COP) $69.67 average cost basis. Some might even be clamoring that you might be able to buy it for what Buffett did! But it is paramount to consider that one cannot forget about dividends received. In this example, on at least the 21 million initial shares, Berkshire received a little over $11 a share in payouts from JNJ. Without doing the time value of money calculations, this brings the average basis to around $50 a share. True, this only equates to about a 4 and a half % average gain over the last 6 years, but it tells a more compelling tale than simply looking at a cost basis around $60 and a price today around $65.

Simply aggregating the Berkshire letters can make comparisons difficult to discern. Thus I wanted to break the net movements out by year:

Year Action Net Shares Sold/Bought Cost / Gain Price
2006 Initiate 21,334,900 $1,250,000,000 $58.59
2007 Add 42,937,048 $2,693,000,000 $62.72
2008 Sell 34,262,357 $63.88
2009 Sell 1,479,124 $56.21
2010 Add 16,492,096 $1,025,000,000 $62.15

It should be noted that the sell prices are difficult to quantify and that the prices in place for 2008 and 2009 are merely an estimate for the whole year. To this regard I simply averaged the high and low monthly value. It is more than probable that these do not accurately reflect the price Buffett actually sold for. In this way, only buy prices should carry weight. We see that Buffett paid $58.59 for JNJ in 2006, $62.72 in 2007 and bought shares of JNJ for $62.15 in 2010.

However, as described previously, we cannot use these numbers to make a comparison to today. In doing so one would disregard both dividends and the time value of money. A more applicable way of approaching the comparison would be to find comparable evaluation metrics. I used the Price to Earnings Ratio and Dividend Yield to make an assessment.

Year Action Price Earnings Dividend / Share P/E Dividend Yield
2006 Bought $58.59 $3.73 $1.46 15.70763 2.48%
2007 Bought $62.72 $3.63 $1.62 17.27816 2.58%
2008 Sold $63.88 $4.57 $1.80 13.97812 2.81%
2009 Sold $56.21 $4.40 $1.93 12.775 3.43%
2010 Bought $62.15 $4.78 $2.11 13.0023 3.39%


Surely many would debate what metrics to use, but I felt that the PE ratio and dividend yield were both widely used and easily quantifiable. I would again like to stress that sold prices are included, but only as place holdings. They represent averages and not known values. It is quite possible that Buffett sold shares of JNJ at a much higher value. However, the buy prices can be used to make a comparison, as they are based purely on average cost basis.

Here's what we know. In 2006 Buffett bought shares of JNJ with an actual PE of 15.7 and a Dividend Yield of 2.48%. In 2007 JNJ was bought with an actual PE of 17.27 and a dividend yield of 2.58%. In 2010 shares of JNJ were bought with an actual PE of 13 and a dividend yield of 3.39%.

If you compare that to today's numbers, you find a PE around 18.6 and a dividend yield around 3.45%. Put together, it seems like the PE is historically high compared to Buffett's purchase points and the dividend yield is better. However, I'm going to go out on limb and suggest that buying JNJ today is just as good of a decision as when Buffett made his purchases. I acknowledge that dividends paid are already lost, but as any present value calculation will allude, you cannot include sunk costs. The best we can do is make the most appropriate decisions moving forward.

I liken JNJ's high PE to that of KO's low PE. If you looked up KO, you would find a quoted price to earnings ratio of about 12.5 for the world's largest beverage maker. However, this is due to a one time jump in accounting profit; which can be explained in more detail here. Likewise, JNJ had a variety of issues that caused the company to report earnings of just $0.08 a share, as compared to $0.70 a year earlier. If you look at full year earnings per share excluding special items you find an EPS of $5 for 2011. This results in a PE around 13. Couple this with the 3.45% dividend yield and you have a PE that is lower or at least comparable to the average Buffett purchase and a dividend that's higher.

Of course simply buying a stock because Warren Buffett did does not pass as due diligence. Although as many investors who watch insider transactions will tell you, "there are a lot of reasons to sell, but only one reason to buy." To this point, it should be noted that while the price at which Buffett sold JNJ is difficult to quantify, he did in fact sell JNJ shares. During this time earnings grew in the mid-single digits and dividends grew by nearly 10% a year. Yet, Buffett sold more than half of his position just 2 years after initiating. Sure doesn't seem like he was "staying with a wonderful business as long as it stays wonderful."

To this point I would make two observations. First since 2006 Buffett has held at least 21 million shares of JNJ, representing over a $1.2 Billion stake of confidence. Second, much like his decision not to reinvest KO dividends, it is possible that he found better opportunities along the way. The purchases of Burlington Northern Santa Fe and Lubrizol, along with say a more recent $10 Billion stake in IBM (IBM) come to mind.

It should be noted that there are a variety of problems associated with looking at "Buffett prices." However, "Buffett valuations" might come a bit closer to investing like Buffett. While it is interesting to note that Buffett sold millions of shares of JNJ just a couple of years after purchasing, it isn't necessarily telling. It remains that Buffett would not be holding a single share of JNJ if he did not think that it was a wonderful company. Certainly one's own evaluation should be completed. But if you stopped me on the street and said "here's a stock that Buffett owns, and you can buy the same amount of earnings and larger payout for less than he did" I think I might be interested in digging in a bit deeper. Look for JNJ to increase its dividend for the 50th straight year this April, but more importantly look for JNJ to increase shareholder wealth over the next half century.

Disclosure: I am long JNJ, KO, PG, IBM.