Berkshire Hathaway and "sprawl"

Warren Buffett may be the most well know man in business and many people have used his ideas in classrooms and boardrooms alike. In the past 50 years he has used the idea that sprawl is good. It is the idea that he used to build his massive conglomerate, Berkshire Hathaway. The company has been outperforming the stock market for decades but over the past several years has fallen short of beating the average gains.

Buffett has recently stated that while sprawl is good it may not be enough to continue as the company has been before. To me, a novice business person, this makes sense and has a lot to do with diminishing returns. His company is so large that it is hard for all the companies inside to continue growing above average. Basically the company is an example of a mutual fund, and I hope this is not news to anyone but some funds perform better than others.

One thing that I believe will help pull Berkshire Hathaway back out of the "average" stage is it's great reputation. This gives it a competitive advantage over others who run other conglomerates. They will continue to have the advantage in merging and acquiring fast up and coming businesses. Though the last couple of years have been below average I would not put it past Buffett to come up with the strategy to change things for the better.


Dave Tufte said...

Pedro: 94/100 (you mean "known" not "know").

What Buffett calls "sprawl" is just another word for the conglomeration that the takeover boom of the mid 80's showed was a bad idea.

Why would anything have changed now? I'll give three reasons.

First, we need to always remember that it's possible that Buffett is just the lucky one. Over any sample size, someone must come out with the best return. Efficient markets theory suggests that the luckiest one should have no better returns going forward. That doesn't seem far off from the story Pedro is telling. I can tell you that 20 years ago, no one paid any attention to Berkshire at all ... because it wasn't at the top of the heap yet.

Second, it's possible that the mergers and acquisitions people that are supposed to keep inefficient conglomerates from thriving have been suppressed to much by regulators over the last 25 years. Certainly the regulators would have us believe exactly that. To the extent they're right, we should see conglomeration be a successful business strategy again.

Third, maybe Buffett really is better. What I know about Berkshire is that they have very light hand over the management of the individual components. Perhaps it's their discipline to not micromanage that is the key to their success.

Sebastian said...

Adding onto Dr. Tufte's first point, I tend to think that this may finally be the law of averages rearing its head. Buffett has done very well for himself over the years. Part of it may be due to a system or an amazing understanding of business, but at some point he has to get down on his luck and lose a little bit. He has been able to benefit from the saying "the rich get richer", but all success stories come with a few bad years.

Dave Tufte said...

Sebastian: 50/50

I'm not sure I agree with your interpretation of the "law of averages".

What we know about efficient markets is that the past return performance (above or below average) of someone like Buffett has little or no predictive power for whether future return performance is above or below average.

Now, I'm making a distinction between returns and value, so make sure you follow me.

That behavior of returns implies that the value of something like Berkshire Hathaway will be a random walk with drift. The drift may be in one direction, but it isn't better or worse than any other investor's drift.

There's a tendency (and I'm not sure that Sebastian has fallen for this, but I worry about a chunk of a class this size falling for it) to think that the law of averages means that there's some trend line for value that any investment returns to. We call this trend reversion. In this case, it would mean that because Berkshire Hathaway's value has been above that trend, that it's due for some bad times that bring it back closer to the trend. Again, our evidence on efficient markets is that investments do not behave this way.

The result that value follows a random walk with drift does not imply reversion to some trend from the past with a fixed y-intercept, but rather no tendency to be above or below a trend of the same slope that goes through the value we're at right now (that probably has a different y-intercept).

ty said...
This comment has been removed by the author.
ty said...

I like to think that the strategy adopted by conglomerates like Berkshire Hathaway and GE have caused their success. All companies have to decide their level of diversity. As noted, most large conglomerates fail. Could it be that managing a conglomerate takes a different management style then what made successful companies large enough to see diversification as a feasible option? If that were not the case why wouldn’t someone like Warren Buffet simply spin off several segments of the conglomerate and remain as a major shareholder in each? It appears that just like in everything else some managers do a better job at managing conglomerates.

The fact that so many companies have tried to succeed as large conglomerates might also be a sign of greed. If a company has already become a major player in their own industry, should their shareholders pressure them to continue growth? Differentiation obviously adds a new level of risk to a company. Attention should be paid to that risk unless there is a compelling strategic reason sprawl should not be a goal for growth within a company.

Dave Tufte said...

Ty: 47/50 (you mean "than" not "them")

Maybe, Ty. I can't disagree with you that there may be better conglomerate managers. But the point in the finance literature about conglomeration is that it's the managers reducing their risk with the owner's money that makes it an issue.

It's only been for 30 years or so that financiers have been pushing the conglomeration as an ROI hit idea to M&A investors. So I think your second paragraph is OK, as long as we limit the evidence on firms to the last 30 years or so. Before that no one appreciated the problem enough to change their behavior.