1. How to use GuruFocus - Tutorials
  2. What Is in the GuruFocus Premium Membership?
  3. A DIY Guide on How to Invest Using Guru Strategies
Geoff Gannon
Geoff Gannon
Articles (304) 

Risk in Net-Nets

December 16, 2011 | About:

In my last net-net article I said I’d discuss the issue of technology risk at net-nets. I scrapped that idea. I have a good reason. I decided I wasn’t qualified to talk about the subject. Instead, I’m going to talk about risk generally. How risky are net-nets? How should we price a net-net? Should we use probabilities? How can we invest in businesses that are so inherently uncertain? That sort of thing.

A net-net is a stock selling for less than the value of its current assets – cash, receivables, inventory, and prepaid expenses – minus all liabilities. Basically, it’s a stock selling for less than its liquidation value.

In normal times, very few stocks trade at such a low level. Today, there are 132 net-nets. Compare this to the number of stocks that could be net-nets. Generally, a net-net will not be in financial services or real estate. Technically, this isn’t impossible. But the kind of net-nets we are looking for – actual operating businesses selling for less than their current assets per share – are almost never found in these two sectors. So, if we narrow the universe of possible net-nets down to U.S. stocks that aren’t in financial services or real estate we end up with roughly 4,800 candidates.

Of course, some of those 4,800 stocks will never be net-nets – no matter how much investors hate them – simply because their current assets are less than their total liabilities. A net-net – by definition – must have higher current assets than total liabilities. Otherwise, there would be zero chance that the stock’s price could be lower than the excess of its current assets over its total liabilities. If current assets are less than total liabilities, a stock will never become a net-net – even if it trades at $0 a share.

So, how many stocks could actually become net-nets?

That number is lower. Right now, it’s about 2,100 stocks. This is the number of U.S. stocks outside of financial services and real estate that have current assets greater than total liabilities. Any of these 2,100 stocks could become net-nets today if only investors traded their shares at low enough prices. For these 2,100 stocks, the question of whether or not they are net-nets is answered by investor sentiment.

How much do investors love or hate the stock? That’s the question that decides whether one of these 2,100 stocks will qualify as a net-net. Right now, only 132 of these more than 2,100 potential net-nets are actual net-nets. That means investors choose to price 15 of every 16 of these stocks above their net current assets per share.

So when we talk about net-nets we are really talking about stocks with three traits:

1. They are not financial services or real estate companies.

2. Their balance sheets are in the top 45% of least leveraged, most liquid balance sheets in the U.S.

3. And their stock prices are in the bottom 6% of most-loved stocks.

There are two ways a net-net stands out. They must have better than the average balance sheets. And they must have lowers than average stock prices.

All 132 of today’s net-nets share these two traits. Which raises an obvious question. Isn’t there something wrong with them?

Yes. There is always something wrong with a net-net. Technically, there doesn’t actually have to be something wrong with a net-net. There just has to be a perception of wrongness. Not actual wrongness. After all, investors could be wrong. But let’s assume investor perceptions are decently accurate in this respect – at least when dealing with large groups of stocks (like 132 of them).

Something is wrong with all of these stocks. They are – at a minimum – a little less leveraged and more cash, receivable and inventory-rich than most public companies. Cash is used to pay bills as they come due. And banks lend against receivables and inventory. So, the ability of these companies to meet their obligations should be high. They shouldn’t be likely bankruptcy candidates.

That would be true independent of their stock price. But we happen to know that these net-nets aren’t just in the top half of public companies in terms of balance sheet strength. They are also in the bottom one-sixteenth of stocks in terms of investor optimism. And stock prices do help indicate bankruptcy risk. Stocks that trade at very low levels relative to current assets may trade that low because investors think bankruptcy is much more likely than the company’s balance sheet suggests.

Why would a company with a good balance sheet be at risk of bankruptcy?

The company could:

1. Be a fraud.

2. Face dangerous and disruptive changes in its industry. Or

3. Suffer from perpetual losses.

Those are really the only three reasons why a company with a lowly levered and highly liquid balance sheet would be at risk of bankruptcy. There is either something terribly wrong with the company’s management, its future or its business model.

The third source of risk is the easiest to detect. A company with current assets greater than total liabilities should have a low risk of bankruptcy as long as it reports an operating profit. While it is technically possible that such a company could run into cash flow problems even while posting a profit, this is very unlikely considering the excess of current assets over total liabilities and the ease with which current assets can be used to borrow cash. In other words, a profitable company with low obligations and a high ability to increase borrowing is a low bankruptcy risk.

So, yes, some net-nets have a poor record of converting reported earnings into cash. That’s a real flaw. And investors should be reluctant to pay a normal price relative to such a poor cash flow-generating company’s reported profits. But that doesn’t mean investors necessarily need to pay a low price relative to current assets. Net-nets are asset value bargains – not earning power bargains.

At the top end of companies in terms of stock prices versus asset values, the big concern is return on assets and equity. Companies that trade at 3, 4 or 9 times tangible book value need to earn very high returns on their tangible equity to justify these high price-to-asset value ratios.

At the bottom end of companies in terms of stock prices versus asset values, the big concern is safety. The reason for this is simple. Say you get it wrong when it comes to a company trading at a high P/B value like Microsoft (NASDAQ:MSFT). You believed the company could continue to earn high returns on tangible book value. It turns out you were wrong. What happens? The company’s return on its assets drifts towards the central tendency of returns on assets at U.S. companies. Returns on assets mean revert. And the stock price falls.

If the same thing were to happen with a low price to asset value stock – you misjudged the company’s future in terms of returns on assets – the most likely outcome is a drift toward this same central tendency. If the company does things you don’t expect with its assets it’s likely to end up earning returns on those assets close to the average of American business. But in the case of net-nets, such mean reversion would cause the stock price to rise.

By definition, every net-net is trading at a discount to its book value. This is because a stock’s book value is always higher than a stock’s net current asset value. Most public companies trade at a premium to their book value. Therefore, any mean reverting tendency a company’s return on assets has will cause high P/B stocks to fall in price and low P/B stocks to rise in price.

This is an important point. It’s totally different than when we are talking about a stock with a low price to earnings ratio. Microsoft has a high P/B ratio. But it has a low P/E ratio. Investors are optimistic about Microsoft’s ability to earn a good return on its assets. However, they are pessimistic about Microsoft’s ability to earn as much in the future as it has in the past.

If you are considering buying shares of Microsoft, the question you need to ask is whether investors are right or wrong about that second part. The question at Microsoft is future earnings relative to past earnings. That’s because Microsoft is only cheap relative to its past earnings. The company is not cheap relative to its present book value.

And that’s totally different from a net-net. A net-net may be cheap or expensive relative to its past earnings. I prefer that the stocks I pick for GuruFocus’s Net-Net Newsletter are cheap relative to their past earnings. But that’s not always the case. Usually, they are – at a minimum – cheap on an enterprise value to EBIT basis. But not always.

One of the stocks that’s now in the newsletter’s portfolio has a very spotty record of operating profits. A lot of past profits came from sources that can no longer be relied on. That’s a problem. But it’s tangential to the big question. Because in that case, the company’s stock was selling for less than its net cash per share. The company’s enterprise value to EBIT wasn’t even a meaningful figure, because enterprise value was negative.

Past earnings would be our main concern if the company was selling at a low price relative to earnings – like Microsoft – but it wasn’t. It was a net-net. So our top concern was the company’s assets and the reliability of those assets. In that extreme case, the company actually had an investment portfolio worth more than its stock price. So, the two questions were how reliable is the value in that investment portfolio and then what was the chance the operating business would destroy value over time instead of just breaking even. The question of the company’s operating business adding value was treated as nothing but a kicker. It might add value. So the stock came with what I hoped was a business that would likely break even and might possible have some positive value. That’s all you need in a net cash bargain.

We’ll see whether or not that stock pick works out. The net-net newsletter holds stocks for 13 months and then sells them. After we sell the stock, I’ll talk about it here in this weekly net-net series. Later in 2012, I’ll start writing about the net-net the newsletter just sold each month. And you can judge whether the decisions I made were smart or stupid.

That’s not the point of this article. The point of this article is that you don’t analyze net-nets from the same perspective you analyze Microsoft. You don’t start with earning power. You start with asset values.

That makes safety paramount. At Microsoft, future earnings relative to past earnings will be critical in deciding whether or not you make money in that stock. This is not true of a net-net. Past earnings may be lousy. And future earnings may turn out to be wonderful. It’s very hard to predict this sort of thing. I don’t try.

All I try to do is look at whether the stock is probably worth more than what we are paying. Usually, that means the biggest concern is safety. What is the risk of bankruptcy? What is the risk of a total failure to produce any profits in the future? Those are the kinds of questions you start with when looking at a net-net.

Now, it’s true that some net-nets have poor returns on assets built into their business model. I’ve mentioned Duckwall-ALCO (DUCK) before. I don’t mean to beat up on the stock. But I think it’s an easy to understand example. ALCO is in the business of general retailing in small town America. Store size is modest at 21,000 square feet. And the operating margin is below 3% in a good year. Even if we were to take only the positive years in the company’s last 10 years – in other words, throw out the two operating losses as one-time events – we would find the company’s return on equity isn’t much better than 4%. Such a low return on equity suggests an appropriate price-to-book value for the company – in its current state of profitability – would be about 0.5. In other words, the company deserves to trade for less than its book value.

Now, I’m actually not saying that ALCO wouldn’t be a good investment if bought below book value. It might be. But you’d be betting on a turnaround. A reversion to the mean.

Is that possible?

Actually, yes. ALCOs gross margins are consistent with a company earning two to three times as much on equity. ALCO’s issue is sales relative to selling, general, and administrative expenses. If ALCO could turn its inventory over faster, it could turn out to be a very good net-net.

But that’s an example of a net-net where you would want to look into how the business works. Is it possible ALCO could always earn an abysmal return on assets simply because of the business it wants to be. Is running what are basically general stores ever going to be as profitable as the average business in America?

Maybe not. But you can see how the past earnings record at ALCO is poor enough to – almost but not quite – justify the very low price-to-book value. In fact, you could say ALCO deserves to be a net-net.

I’m not sure I would go that far. The problem here is one of odds. I talked about mean reversion. Well in the case of Microsoft any drift to normal returns on assets for American business generally would be very bad for that stock. But in the case of ALCO any drift toward normal returns on assets for American business generally would be miraculous for that stock. At a minimum, it would result in the stock doubling. And more likely, it would result in something closer to a quadrupling of the stock price.

You see the problem. Imagine there is a one in four chance that ALCO will improve its long-term return on equity to say the 8-12% range? And there is a one in four chance ALCO will see its results deteriorate further – to the point of bankruptcy. The other 50% chance is basically muddling through around a 4% to 5% return on equity for as far as the eye can see.

This is the problem investors face when they pick stocks. They have to handicap the stocks. They can’t just say one business is better than another. They have to price them to level out that difference.

I just made those odds up. They are for the purposes of illustration only. But they give some idea of what net-net investing is really like. A lot of companies fall into this sort of category. Under the most likely scenario – muddling through – the stock’s price will rise 30% to 50%. Under the best case scenario, it will rise 200%. And under the worst case scenario, it will fall to zero.

Would you buy it?

This is a good representation of what investors really face. Some people like thinking in these terms. I don’t. I have serious doubts about the ability of anyone – certainly I have doubts about my own ability – to handicap this kind of situation. The return estimates are fairly accurate. Muddling through really will result in a gain of 30% to 50% for the stock. Turning things around really will result in a home run return in the neighborhood of 200%. And, of course, bankruptcy really would mean the stock falls to zero.

But what about the probabilities? How likely is each scenario? That’s the part where I think human beings are pretty close to hopeless. In 9 out of 10 cases, it’s simply going to be too hard for me to know the probabilities well enough to disagree with Mr. Market. Often, I can’t tell the difference between a 1-in-3 chance and 1-in-9 chance. And that’s kind of critical to this approach.

In net-net land, the numbers of cases where I’m willing to disagree with Mr. Market is a lot higher. That’s because the odds are better. There are more extreme examples of mispricing. ALCO is actually an example of an extremely fuzzy – for a net-net – handicapping situation. The past is inadequate. But how likely is a future where nothing changes? And if you were going to bet on change, the odds of change helping a company as cheap as this are pretty high. Uncertainty may be your best friend in a stock as cheap as ALCO.

So again I come back to safety. If net-nets generally are mispriced bets – then a safe group of net-nets should work out better than the market. This focus on safety is not because I think a safety first approach actually achieves higher results. I don’t. I think a probability based approach like the one I showed you with ALCO is the approach that would work best. It is however merely a theoretical approach with no support in actual practice.

I just don’t think it’s a strategy human beings can adopt. I don’t know about you. But I can’t recognize small gradations of expected values. What I can do is recognize huge gaps in safety and reliability.

To give you some idea, the net-net newsletter has stocks in its portfolio that were bought at around their net cash per share. These companies didn’t have any operating losses in their recent history. So, the idea was simple. When you buy the stock, you are paying for the cash and getting the business for free. A profitable business is worth more than nothing. And we are paying nothing for a profitable business. Therefore, the stock is worth more than we paid for it.

So the newsletter’s approach is to look more for compartments of defense than for probabilities. A good example would be buying a net-net with an above average history of returns on its invested assets at a price close to its net cash. In this case, we would have 3 defenses:

1. Protection provided by cash per share

2. Protection provided by above average business

3. Protection provided by average business

In other words, if the operating business turns out to be worth nothing, the surplus cash per share should still cover our purchase price. If the cash per share somehow gets squandered, we still have protection from an above average business. And if this historically above average business mean reverts to the kind of return on assets most businesses achieve – well then we still paid a price that was lower than what an average business would be worth. So we can suffer a couple failures and still survive.

That example of 3 defenses against loss is the ideal. Most stocks in the net-net newsletter’s portfolio get more protection from one of these areas than the others. In one case, cash alone provided full protection. In other cases, it was more of a combination of all three defenses.

This isn’t to say that relying on only one of these defenses is a bad strategy. In fact, I don’t think it is a bad strategy. If all you did was buy mediocre businesses with no surplus cash at less than two-thirds of their net current asset value, I think you’d outperform the market. Some stocks would fail miserably. But others would more than make up for these failures.

Portfolios of randomly picked net-nets embody this strikeout and homerun approach. And they tend to outperform the market. So the approach works.

But it’s not what I do in GuruFocus’s net-net newsletter. The approach I use in the newsletter is to pick the net-nets that have the highest chance of delivering an acceptable performance for the time we hold them.

Which, by the way, is just a year for the newsletter. This is to keep new picks coming month after month. It is not an endorsement of such a short-term approach to your own net-net investing.

When buying net-nets yourself, you should always hold them for at least a year. And I’d suggest an average holding period of 2-5 years. That’s the kind of net-net holding period Ben Graham, Walter Schloss, etc. actually practiced.

They certainly didn’t turn their portfolio over 100% a year.

You shouldn’t either.

Check out GuruFocus’s Net-Net Newsletter

Follow Geoff at Gannon On Investing

About the author:

Geoff Gannon

Rating: 3.5/5 (26 votes)


Tkervin - 6 years ago    Report SPAM
Clear and informative as always.

Well done.
Dgenchev - 6 years ago    Report SPAM
Don't you think that suggesting an average holding period of 2-5 years for net-nets is excessive?

I don't know about Schloss, but Graham has, at different points, suggested holding net-nets up to a maximum of one to three years or 50% appreciation if that comes first, and selling at market afterwards. He said something along the lines of: sell after the stock has appreciated 50% or no later than the end of the second calendar year after purchase. This would be 2 years max.

Overall, one should be careful hanging on to such issues for too long, because they won't be growing one's capital in the meantime. In the context of compounding and opportunity cost, one would be losing money with every passing year by not being invested in a growing business.
Augustabound - 6 years ago    Report SPAM

I don't know about Schloss, but Graham has, at different points, suggested holding net-nets up to a maximum of one to three years or 50% appreciation if that comes first, and selling at market afterwards. He said something along the lines of: sell after the stock has appreciated 50% or no later than the end of the second calendar year after purchase. This would be 2 years max.

I thought Graham's 2 year or 50% "rule" was from his 1976 interview after his view on investment had changed somewhat?
Dgenchev - 6 years ago    Report SPAM
Overall, you buy a net-net at, say, 1/3-1/2 discount to intrinsic value. This is 50-100% return on the way up. Over 3 years, that's roughly 15-25% annually. Cut this in half to account for the ones in the bunch that don't work out and you've got decent 7-12% annual returns - not quite Graham's 20% or Buffett's 25-30%, far from the hypothetical 50% Buffett would be making now with less money, but still good.

Over more years, things start getting stretched.

When picking up cigar butts, you do need quite some turnover. You don't have to turn them every year, but hanging on for more than 3 years one could be punished for negligence or obstinance.

I don't have any useful information about Buffett's partnership portfolio turnover and I would be glad if someone who has shares it, but I am under the impression that his favorite holding period was not forever back then. As far as I know, he stuck only with the companies that didn't go anywhere and tried to fix them via activism.

Please leave your comment:

Performances of the stocks mentioned by Geoff Gannon

User Generated Screeners

cyebrahimiGrowth Stocks
DBrizanALL 22Mar2018 1133p
DBrizanALL 22Mar2018 1015p
HOLKLSUTop 20 Group Utilities
efredbacYahoo Watchlist
jasonw.mosleyRule # 1
pbarker46FCF & dividend
Get WordPress Plugins for easy affiliate links on Stock Tickers and Guru Names | Earn affiliate commissions by embedding GuruFocus Charts
GuruFocus Affiliate Program: Earn up to $400 per referral. ( Learn More)

GF Chat