Low P/B Investing, Net-Nets, Negative Enterprise Value and Cash

[rukawa]

Traditional value investing consists essentially of investing in low price to book companies. In traditional studies of market anomalies there appears to be out performance that comes from low P/B stocks which are identified with "Value" and also an additional alpha that comes from small cap stocks.

 

Normally Net-Nets are considered as a variant of low P/B investing. But one thing observed with Net-Nets is that they get considerably higher returns than low P/B and additionally a lot of the out performance cannot be explained either by value or the size effect. Typically the best performance you get from both value and small cap effect together is less than 20% a year. But net-nets returns are usually substantially in excess of 20% a year. For instance the following study finds returns of 2.55% monthly which translates to 30% a year

http://csinvesting.org/wp-content/uploads/2015/01/Benjamin-Graham-s-Net-Nets-Seventy-Five-Years-Old-and-Outperforming1.pdf

 

Even more surprising is that you get even higher returns when you invest in negative enterprise value stocks. This study finds 50% a year

https://blogs.cfainstitute.org/investor/2013/07/10/returns-on-negative-enterprise-value-stocks-money-for-nothing/

 

Basically net-nets are like low price to book except that instead of including the fixed assets you zero them out completely. In the Net-Net paper by Tobias Carlisle that I cited NCAV (net current asset value) stocks are defined as: Liabilities including preferred stock - current assets. You buy when the stock trades at 2/3 of NCAV. This values all current assets including inventory and accounts receivable at 100%.

 

Negative Enterprise value is a different calculation in that it relies on market values but in some sense the Negative enterprise value calculation can be considered nearly the same as the Net-Nets if you assume that debt trades at par. The only difference is that you completely zero out accounts receivable and inventory. Basically I would summarize this as follows:

 

Negative Enterprise Value Buying Rule

Buy when

Cash - Debt > Market Cap

Return: 50% a year

 

Net-Net Buying Rule

Buy when (2/3)*(Cash + Inventory + Accounts Receivables - Debt) > Market Cap

Return 30% a year

 

The fact that enterprise value > Net-Nets > Low P/B seems to indicate very strongly that in any asset based value investing strategy it hugely pays to buy stocks that are cash rich. Cash really is King.

[frommi]

Na, this is too simple to be working, it has to be a compounder with ROIC > 50%! And look at all the crap that you have to buy!!  ;)

 

Seriously, i think its better to buy a stock where cash, receivables and inventory are evenly split and it trades at a 50% discount to NCAV than to buy a shell company where there is only cash on the asset side of the balance sheet and the discount to NCAV is small. Therefore i would argue that a large discount to NCAV (<50%) and current assets/liabilities>2 is one of the best strategies out there and also delivers roughly 50%. At least there are some studies that have found that to be the case.

I still wonder if its a good idea to exclude companies that have issued a lot of stock in the past because i have not found any studies researching that. Most studies probably include all these stocks and still return 30-50%, thats pretty damn hard to beat. Currently most netnet stocks i found are in Japan, Singapur and Hongkong.

[sculpin]

Thanks for starting this thread. Read O'Shaunnesy's book back about 20 years ago and from then on I have always looked for low price to sales anomalies across different industries. Worth a look or to find an updated study on these types of stocks. As well it is important in combining a deep value strategy with a momentum indicator - I guess it helps to keep one from getting stuck in value traps...

 

However, it may come as a surprise that, according to the 50 years of mechanical investing research performed by James O'Shaughnessy and detailed in his book What Works on Wall Street, the single best statistic for finding undervalued stocks that will outperform the market is the largely ignored price-to-sales ratio. O'Shaughnessy set up a portfolio that bought the lowest price-to-sales ratio stocks and renewed them every year, then compared against other portfolios that used the more popular (and some would argue more meaningful) price-to-earnings and price-to-book multiples. Here were the annualized 50 year results of each strategy:

 

Stocks with Lowest: Annual Return

Price-to-Sales 15.4%

Price-to-Book 14.4%

Price-to-Earnings 11.2%

Price-to-sales outperformed the others by a significant margin. Interestingly, price-to-earnings as the only criteria actually underperformed the market!

 

The phenomenon of low price-to-sales ratio stocks as a good mechanical screening strategy is not completely unknown. Ken Fisher, son of investing legend Philip Fisher and a member of the Forbes 400 richest people in the U.S., wrote the book Super Stocks, where the entire focus of the book is on choosing low price-to-sales issues.

 

[scorpioncapital]

"We assume all stocks are purchased on December 31st, and held for one year.

We assume that the hypothetical investor completely liquidates each portfolio before forming

a new portfolio"

 

Just a hypothesis but I suspect the above methodology could have led to a large part of the positive result. Why? I'm thinking that net-nets are usually one-pop wonders (with a few cases of longer term turnarounds). Like a cigar butt, they rise to IV and then wallow in low returns going forward. But if you sell everything in a year or two when this criteria is met and start again, it might make sense for this result to be achieved. If you select a portfolio of leading stocks for the long term, you wouldn't liquidate and depend on your return from the compounding of the good business over many years.

[rukawa]

Na, this is too simple to be working, it has to be a compounder with ROIC > 50%! And look at all the crap that you have to buy!!  ;)

 

Seriously, i think its better to buy a stock where cash, receivables and inventory are evenly split and it trades at a 50% discount to NCAV than to buy a shell company where there is only cash on the asset side of the balance sheet and the discount to NCAV is small. Therefore i would argue that a large discount to NCAV (<50%) and current assets/liabilities>2 is one of the best strategies out there and also delivers roughly 50%. At least there are some studies that have found that to be the case.

I still wonder if its a good idea to exclude companies that have issued a lot of stock in the past because i have not found any studies researching that. Most studies probably include all these stocks and still return 30-50%, thats pretty damn hard to beat. Currently most netnet stocks i found are in Japan, Singapur and Hongkong.

 

I guess you are arguing that it should be a real business instead of just a cash box. I would agree. This might also be the reason they found that its bad to exclude companies with negative earnings...they have actual businesses that might turn around. I would additionally argue that you should exclude Chinese reverse mergers.

[rukawa]

"We assume all stocks are purchased on December 31st, and held for one year.

We assume that the hypothetical investor completely liquidates each portfolio before forming

a new portfolio"

 

Just a hypothesis but I suspect the above methodology could have led to a large part of the positive result. Why? I'm thinking that net-nets are usually one-pop wonders (with a few cases of longer term turnarounds). Like a cigar butt, they rise to IV and then wallow in low returns going forward. But if you sell everything in a year or two when this criteria is met and start again, it might make sense for this result to be achieved. If you select a portfolio of leading stocks for the long term, you wouldn't liquidate and depend on your return from the compounding of the good business over many years.

 

Two points:

You might want to check out the January effect which relates to your comment and is talked about in the net-net article basically January alone accounts for about 10% of the out-performance of net-nets.

 

Your comment reminds me of a deeper question I am asking myself. Is value investing even about intrinsic value? To me it appears that the reason low P/B does well is because of random hiccups in the market where the business appears to do well for some period of time and the stock price goes way up. I feel like the real value effect comes from the fact that market expectations are so horrible that positive surprises get massive bumps. In this sense low P/B is more an indicator of low expectations than anything else. Net-nets might have the advantage because they simply can outlast companies without cash and they indicate very low expectations. This might also be why negative earning net-nets out perform positive earning net-nets...they have more room to surprise.

 

I think when you think the Buffet way, you expect that eventually a company through the power of its intrinsic value alone will force the market to recognize its value. Coke is a very good example of this. Coke's earnings and dividends increased so much over time that it was impossible for its price not to go up without resulting in absurdities like 100% dividend yields or 0.1 P/E's. Coke's increase in intrinsic value is the catalyst Buffett relied on.

 

In that sense cigar-butt investing seems ridiculous because the "intrinsic value" of a lot of these companies in the long run is zero. You really are making money off short term changes in market perception and then selling. I think this psychological hurdle is why people have a real hard time with these methods. They really do rely on the markets random changes in expectations and perception in a way that Buffett's quality companies method doesn't.

 

But if it works....it works.

[scorpioncapital]

You can further divide the Net-net beast into sub-categories. I.e. you can even specialize within net-nets. For example, some net-nets appear to be similar to SPACs or pools of capital at a discount. Others are biotech stocks. Others are lagging industrials. I can imagine a specialist can probably find some connections as to how to specialize here. I wonder in that study if some of the overall gains came from a few biotech net-nets, or conversely how much return was detracted by that sub-sector.

 

 

 

[Packer16]

I think the key here is to find a cheap decent operating business then figure out what is causing the "cheapness" & determine if that reason can go away.  In some cases like a declining industry cannot be fixed but in the case of a poor management team or neglect the problem can be fixed.  The question then is when & at what cost. 

 

Japan or Korea are the only places today I see businesses like these but they may become more common in the US if more brokers no longer trade non SEC filing firms.

 

Packer

[InspireByReason]

"We assume all stocks are purchased on December 31st, and held for one year.

We assume that the hypothetical investor completely liquidates each portfolio before forming

a new portfolio"

 

Just a hypothesis but I suspect the above methodology could have led to a large part of the positive result. Why? I'm thinking that net-nets are usually one-pop wonders (with a few cases of longer term turnarounds). Like a cigar butt, they rise to IV and then wallow in low returns going forward. But if you sell everything in a year or two when this criteria is met and start again, it might make sense for this result to be achieved. If you select a portfolio of leading stocks for the long term, you wouldn't liquidate and depend on your return from the compounding of the good business over many years.

 

Two points:

You might want to check out the January effect which relates to your comment and is talked about in the net-net article basically January alone accounts for about 10% of the out-performance of net-nets.

 

Your comment reminds me of a deeper question I am asking myself. Is value investing even about intrinsic value? To me it appears that the reason low P/B does well is because of random hiccups in the market where the business appears to do well for some period of time and the stock price goes way up. I feel like the real value effect comes from the fact that market expectations are so horrible that positive surprises get massive bumps. In this sense low P/B is more an indicator of low expectations than anything else. Net-nets might have the advantage because they simply can outlast companies without cash and they indicate very low expectations. This might also be why negative earning net-nets out perform positive earning net-nets...they have more room to surprise.

 

I think when you think the Buffet way, you expect that eventually a company through the power of its intrinsic value alone will force the market to recognize its value. Coke is a very good example of this. Coke's earnings and dividends increased so much over time that it was impossible for its price not to go up without resulting in absurdities like 100% dividend yields or 0.1 P/E's. Coke's increase in intrinsic value is the catalyst Buffett relied on.

 

In that sense cigar-butt investing seems ridiculous because the "intrinsic value" of a lot of these companies in the long run is zero. You really are making money off short term changes in market perception and then selling. I think this psychological hurdle is why people have a real hard time with these methods. They really do rely on the markets random changes in expectations and perception in a way that Buffett's quality companies method doesn't.

 

But if it works....it works.

 

This doesn't make sense. The "intrinsic value" of these companies are worth what they are worth. Net net is just short hand for liquidation value so of course in the long run these companies are not going to grow in intrinsic value but this doesn't mean that they aren't worth something right now.

 

This is the only method of investing I feel comfortable doing. I don't know why so many people shy away from dirt cheap cyclicals, turnarounds and liquidations. It's the most simple and arguably the most predictably successful.

[scorpioncapital]

The major risk I've observed to net-nets is that the discount catches up to the shrinking business/balance sheet as opposed to the discount closing. Example: Say a biotech is losing 20 million per year and is 20 million below cash in the bank. By the end of the year, the discount not only fails to close but the stock may drop another 20 million to maintain the discount. Shrinking is a moving target and the stock sort of follows along. If you can find a net-net that has no shrinkage, or temporary shrinkage, or cold hard assets worth something, that's a better situation. However, you are still assessing whether some management you may not know anything about is capable of expanding or stabilizing a business.