I received a couple questions/comments from readers lately regarding Buffett and some previous comments I made on strategy before and after the Berkshire annual meeting. I always enjoy interacting with readers. The vast majority of my days are spent researching new investment ideas, so I can’t always respond to everything, but please feel free to email me questions/comments to [email protected]. I read all my emails even if I don’t always have time to respond. When possible, I’ll respond individually, or when I receive numerous emails on the same topic, I’ll do a post.
As for the questions that I wanted to address on Buffett, a few have to do with his evolution from deep value to quality (an evolution that I think is very misunderstood, which I’ll comment on in the next post).
For this post, the question was on how I think about quantitative vs. qualitative analysis and specifically, what do I think about net-nets?
Here is a discussion that I’ll repost from a comment I made a number of months ago:
I’m very much interested in the numbers… but I also work hard to try and understand the qualitative aspects of a business. But I generally agree with Buffett when he said the most “sure” money is made on the quantitative side, while the real big money is made when both qual and quant line up…
But to answer your question–very generally speaking–my ideal investment is a great operating business that produces consistent free cash flow and high returns on capital that for some reason trades at a low price relative to normal earnings. But those investments are relatively rare. So I’m always on the lookout for undervalued situations such as hidden or cheap assets or other special situations resulting from some corporate event like restructuring, or spinoffs, etc…
But very simply, I’m just trying to find large, low-risk gaps between price and value. I try to make everything very simple. I want obvious value…
Net-Nets: Almost Always Poor Long Term Investments
The headline above might aggravate some of my friends. So let me briefly explain: investing in net-nets can be a viable strategy, but net-nets are dubious long term investments on an individual basis. The topic of net-nets reminds me of Winston Churchill’s famous words about Moscow in 1939 that I paraphrased in the title of this post… individually, they are risky and unpredictable.
Ben Graham would agree, most net-nets are inferior businesses that by their very nature lead to inferior long term investment returns. However, the overall strategy of owning a basket of them can work. The key is owning a basket of them and selling them “on the blips”–it’s a strategy that I don’t feel comfortable with, for reasons that I’ll elaborate on below.
However, the general strategy seems to continue to work according to all of the tests, and I know investors who have done very well just mechanically buying these things (net-nets, negative EV stocks, etc…). Although I look at net-nets occasionally, I’m not very excited about them. I would prefer to own businesses that are growing intrinsic value (my experience is that most net-nets today have shrinking intrinsic values). This means that over time, your margin of safety will erode, meaning you need the investment to work out sooner rather than later. As soon as you get in, you’re looking for a way out.
So despite the empirical evidence that the strategy can work and my respect for the investors I know who implement such a strategy, I typically tend to pass on most investments in this area. I have a large amount of discomfort when I know that the fate of my investments depends not only on the price that some other investor is willing to pay me for my assets, but also the timeframe in which he or she must accommodate me. In other words, these backtests tend to work because most of the holding periods are quite short (usually just 1 year).
I have done a lot of research on these tests, and just anecdotally, I’ve noticed that many of the holdings in these mock portfolios do in fact jump up in price from time to time, allowing the investor to exit at a profit. But if you look at some of the old portfolios from say 3-5 years back, you’ll find that most of the holdings (even after some occasional significant rises) ended up either in bankruptcy or continue to wallow along at low prices.
So as long term investments, they often tend to be quite mediocre. This means you have to have more of a trader’s mentality, or a liquidator’s mentality with these types of really cheap cigar butts. You need to get out of them as quickly as you can once you’re involved. That’s just not a relaxing situation to be in, and it just doesn’t suit my personality.
Net-Nets: A Different Ballgame in Graham’s Day
I think Graham and Schloss had a field day with net-nets in the 30?s-50?s because of the opportunities available. The significance of the discounts that were widely available in the shadows of the Depression cannot be overstated. Schloss often found companies that weren’t necessarily great (some might even be losing money temporarily), but they had “$35 worth of liquid assets with a stock price of $15 and a 6% dividend”. It’s hard to lose money with a basket filled of situations like that. Most of the net-nets today are burning cash at a rate that will erase the margin of safety you initially get in just a year or two. Some of those will work, but they have much higher risk than I think Schloss would have wanted to take on.
A&P: The Largest Retailer in America
Another example that Graham referenced: In the late 1930?s, a company called A&P was the largest retailer in America, if not the world, and in 1938 it traded so cheaply that it was a net-net. (Imagine if Walmart had a bad year or two and traded below liquidation value—that’s the kind of environment Graham was working in)…
In the late 20?s A&P traded close to $500 per share. By 1938, it had sunk as low as $36, thanks to a bad year and general pessimism. To put this price in perspective, A&P had net current assets $38 per share, including $24 of cash. So basically, the market was pricing A&P at liquidation levels–assigning no value to their business. But the average earnings per share of the previous 5 years (1933-1937) were about $6 per share. So A&P had a P/E based on avg 5 year earnings of about 6, and it traded below