Someone who reads my articles asked me this question:
Hi Geoff,
I had a few questions about your article: "Are Most Net-Nets Uninvestable?"
Concerning investing as a group instead of individually.
You said: if (you were) buying 30 net-nets or more you would use a set of standard metrics like the F-Score. You would rank the whole universe of net-nets according to numbers (you) thought indicated quality when bought as a group and buy a group of the top ones of that group.
What other criteria besides the F-Score would you use and what numbers would be good to help rank them by? I see that another strong criteria is the highest insider ownership to institutional ownership ratio. What other things would you use to come up with and rate this list?
Thanks,
Ron
That’s a great question. But I’m going to take it in a slightly different direction than I think you intended. You probably wanted a list of metrics that can add to your returns in net-nets. I’m going to focus on one metric that can give you some common sense advice on whether a net-net is solid or shaky.
I don’t want to make this a complicated discussion of exactly what metrics to use to rank net-nets so as to provide the best annual returns possible. I don’t think that’s a discussion worth having for a two reasons:
1. A strategy of randomly choosing net-nets provides an adequate annual return – and beats the market – when practiced over 10 years or more.
2. Net-net portfolios don’t fail. Net-net investors fail. They don’t stick to the strategy.
Goals of Net-Net Investing
Your goals in net-net investing should be:
1. Don’t lose money.
2. Achieve an acceptable absolute return over 10 years.
3. Achieve an acceptable relative return over 10 years.
Don’t lose money is self-explanatory.
I’d define an acceptable absolute return as being a portfolio that compounds at 10% a year for at least 10 years.
And I’d define an acceptable relative return as being a portfolio that beats the stock market over at least 10 years.
Taxes, Fees, and Turnover
All of these milestones need to be achieved on an after-fees and after-tax equivalent basis. So, if you are paying 35% a year on your net-net portfolio’s gains in taxes and would only be paying 15% a year on your gains in a general investment in the stock market (index fund, buy and hold blue chips, etc.) then you would need to earn 13% a year in net-nets to match your 10% a year in blue chips. The same is true if you are paying high commissions on net-nets.
There are easy ways to avoid doing both. And, in general, the drag on net-net performance from taxes, fees, etc. is actually quite manageable. There’s no reason to pay more in taxes on your net-nets than you would in any other portfolio. Because holding net-nets beyond one year works out fine.
The major issue is illiquidity. But that’s a cost you can control. If you leave bids out for net-nets at the last trade price before you made your buy decision – you will usually have your order filled within a couple weeks to a couple months.
I’ve not experienced trading costs in net-nets that made more than a 2% annual return difference from what my same costs are in stocks $1 billion in market cap and over. But it’s worth noting that I generally don’t bid more than a penny over the last trade price. (To be clear: I don’t change my bid when the price moves. I leave the bid in place. The price moves. In a few days, weeks, or months the price will usually fall back below my bid and I will get my shares then).
Trading costs are determined by turnover more than anything else. And I don’t turn my net-nets over faster than my other investments.
This is not true for many people. And, so, you’ll hear a lot of people talk about the high costs of net-net investing. There are definitely inconveniences in net-net investing. For example, it can take weeks or months to get your order filled. But I’m not convinced there are high costs in a low turnover net-net portfolio.
Principles vs. Percentages
I'm going to show you some backtest results in this article. Don't focus on the annual return percentages. Don't go away thinking this proves you can make a certain return in net-nets. It doesn't. It proves some basic principles: net-net investing works, depending on what kind of net-nets you pick you can get very different results, and the volatility of net-net portfolios comes from the shakiness of the balance sheets of the net-nets they own.
I'm using backtests to illustrate those principles. Not to sell you on the idea of net-net investing. Whether or not you invest in net-nets should be based on whether or not you think you can stomach the experience – and stick to the strategy year after year.
Don't just go out and follow the strategy with the best backtest. Follow the strategy you can stick to.
Any System Works…
It’s actually hard to design a net-net portfolio that will underperform the market over 10 years. Net-nets are so cheap that even doing things like picking the 10 net-nets with the biggest sales declines this year – a truly frightening list of stocks – ends up giving you a basket of stocks that has tended to beat the market. In fact, doing that every year – buying the 10 net-nets with the biggest sales declines, holding them for exactly one year, selling them all, and repeating the process – returned 14% a year over the last 10 years.
To give you an idea of how bad the one-year sales drop in net-nets can be, a list of the 10 net-nets with the biggest percentage year over year sales declines currently features revenue declines ranging from 45% to 100%. Four of the 10 companies have year over year sales declines of 97% or more – meaning, yes, 40% of that list essentially has zero revenues right now.
And yet this group tends not to underperform the market. It’s a terrible strategy. I don’t recommend picking net-nets on this basis – but if somehow you stuck to this insane plan to buy the net-nets with the biggest sales declines, you’d do fine over 10 years.
Again, the thing to focus on is not why such a bizarre strategy would work at all – it’s to focus on how you can screw up net-net investing. You can fail to stick with it.
…If You Stick to It
Of course, you wouldn’t stick to that system. Half the stocks you’d buy each year would probably have almost no sales. No one is going to stick to a plan that involves buying companies that have just seen their sales anywhere from cut in half to totally obliterated.
So, the investor’s goal in net-net investing should not be to find the system that works best. It should be to find the system that is workable. Workable means you can stick to it.
Today, I’ll recommend one super simple system for net-net investing that I think is quite workable.
It’s not the elaborate ranking system you asked for. In fact, it’s insanely simple. But before we get to the system, let’s start with the results.
What if I told you net-net investing could give you results like these:
Well, it can.
And what if I told you net-net investing could give you results like these:
It can do that too.
Which set of results you see depends on which kind of net-nets you buy.
And if you preferred the first set of returns – you’re not alone. But the second set of annual returns actually leads to higher returns: 22% a year versus 15% a year for the first group.
So, I guess we should rank net-nets so we get the second set of annual returns, right?
I wouldn’t mind earning 22% a year. But I’d rather own the stocks that provided the first set of annual returns.
Why?
Because I can stick to that system. I really could earn 15% a year investing in those net-nets. I would never stick with the second system. I would never earn that 22%. The issue is what results a screen would earn. The issue is what results a human would earn.
Two Systems: One Criterion
Those are both net-net backtests. And they both used the same criterion:
Current Assets/Total Liabilities
There couldn’t be a simpler number. It’s basically a measure of the “leverage” in your net-net purchase. A top heavy fraction means the company has to lose a lot of its current assets to reach insolvency and cause you a permanent loss.
But a ratio very, very close to 1 gives you a huge amount of upside. Think about it. If a net-net has $100 in current assets and $90 in total liabilities it has a current assets to total liabilities ratio of 1.11. If you pay 60% of NCAV for that stock – which is $6 a share – and the company grows current assets by just 10%, you could suddenly have the potential for another 100% return on your investment. That’s because 10% of $100 is $6 and you only paid $6 for the stock. You can easily see how your $6 investment just became a $20 stock.
Okay, but what about the downside?
Oh, it’s big. And likely. A 10% decline in current assets eliminates all of the company’s net current asset value. If the company is losing money and doesn’t have access to long-term financing it could go to zero. So, a stock that cost you anywhere from $6 to $10 could very quickly be worth either $20 a share or $0.
That’s if the current assets to total liabilities ratio is low.
What if it’s high?
Let’s look at the another stock. Assume it’s also trading for $6 a share. And it also has $10 in NCAV. But this time current assets are $11 a share and total liabilities are $1 a share. Again, let’s imagine that current assets increase by 10% to $12.10 a share. That makes NCAV $11.10 a share versus your $6 purchase price. Your upside is now 85% versus the mind blowing 230% it was in our leveraged example. And what about the downside? A 10% decline in current assets would still leave the company with $8 in net current assets. That means a 10% shrinkage in current assets actually leaves more than 30% upside for you.
In other words, bad things happened to the company – and you still made money.
That’s a Ben Graham bargain.
Enough with the hypotheticals. Let’s see what these lists looked like in action.
Why do I think that picking net-nets with the lowest ratio of Current Assets to Total Liabilities is a system nobody would follow – even if they knew it offered the possibility of 20% annual returns?
Two reasons.
Let’s start with the biggest loser among the 10 net-nets with the lowest ratio of current assets to total liabilities for each of the last 10 years.
Remember, if you constructed a portfolio of the 10 net-nets with the lowest Current Assets/Total Liabilities ratio you’d have 10% of your portfolio in one of these awful, awful stocks every year:
Also, this is what the years 2007 through 2011 would have felt like:
Yeah. No one is going to lose 80% of their portfolio only to stick with the same approach for two more years. Of course, if you did, you’d get an acceptable return – 13% from 2007 through 2010 would be a fine return. The emotions you’d be feeling in late 2008 and early 2009 – would not be fine. And you’d bail on net-net investing entirely.
Compare this to the experience of a portfolio made up of the 10 net-nets with the highest Current Assets/Total Liabilities ratio:
You would end up with less money at the end of 2010 than you started with in 2007. But, you’d have been down less than 25% from your starting point in the middle of a worldwide financial panic. And you’d basically be flat over some very tumultuous times. This is a strategy you could stick to.
Two Types of Net-Nets
I’m in favor of a net-net portfolio built from net-nets with a high Current Assets/Total Liabilities ratio. And I’m against including net-nets with a low Current Assets/Total Liabilities ratio. Of course, we all break rules. And I happened to pick a net-net with a low Current Assets/Total Liabilities ratio in this month’s issue of the Ben Graham: Net-Net Newsletter. I hope the stock works out. But I know it’s one of the riskiest picks in the portfolio. In fact, I think it’s the single riskiest stock in the Ben Graham: Net-Net Newsletter’s model portfolio right now. But it’s one of 13 stocks in the portfolio. And I like the odds on the whole group a lot.
It’s important to use common sense when picking net-nets. There’s this idea out there that a net-net portfolio’s performance always has to look like the second column in this table:
Phil Fisher had this idea that you should never invest in the marginal companies in an industry. Sure, they had more upside when the economy was booming – because their margins would expand more. But they had too much downside in bad times.
It was easy for Phil Fisher to tell a marginal company from a leading company.
And it’s easy for you to tell a solid net-net from a shaky net-net.
The solid net-nets may not outperform the shaky net-nets – just like leading companies may not outperform marginal companies – but it’s your choice which group you want to invest in.
I can’t tell you not to invest in shaky net-nets – after all, they did provide the better 10 year returns. But I can tell you that I prefer investing in the solid net-nets.
I always try to pick solid net-nets for the Ben Graham: Net-Net Newsletter.
You can do the same by looking for net-nets with the highest Current Assets/Total Liabilities ratio.
And I know we didn’t talk enough about specific stocks in this article. So, I’ll end with two columns of stocks. On the left, you have some of the most solid net-nets available today. And on the right, you have some of the most shaky net-nets available today.
Peruse whichever column you think is most interesting.
But I’ll stick to the first column when picking stocks for the Ben Graham: Net-Net Newsletter.
Remember, those lists are just based on one number: Current Assets/Total Liabilities. There are other risks – like fraud. I would eliminate a couple stocks on the left for that reason.
If you’ve read my past net-net articles, you can probably guess which ones.
Ask Geoff a Question about Net-Nets
Check out the Ben Graham: Net-Net Newsletter
Hi Geoff,
I had a few questions about your article: "Are Most Net-Nets Uninvestable?"
Concerning investing as a group instead of individually.
You said: if (you were) buying 30 net-nets or more you would use a set of standard metrics like the F-Score. You would rank the whole universe of net-nets according to numbers (you) thought indicated quality when bought as a group and buy a group of the top ones of that group.
What other criteria besides the F-Score would you use and what numbers would be good to help rank them by? I see that another strong criteria is the highest insider ownership to institutional ownership ratio. What other things would you use to come up with and rate this list?
Thanks,
Ron
That’s a great question. But I’m going to take it in a slightly different direction than I think you intended. You probably wanted a list of metrics that can add to your returns in net-nets. I’m going to focus on one metric that can give you some common sense advice on whether a net-net is solid or shaky.
I don’t want to make this a complicated discussion of exactly what metrics to use to rank net-nets so as to provide the best annual returns possible. I don’t think that’s a discussion worth having for a two reasons:
1. A strategy of randomly choosing net-nets provides an adequate annual return – and beats the market – when practiced over 10 years or more.
2. Net-net portfolios don’t fail. Net-net investors fail. They don’t stick to the strategy.
Goals of Net-Net Investing
Your goals in net-net investing should be:
1. Don’t lose money.
2. Achieve an acceptable absolute return over 10 years.
3. Achieve an acceptable relative return over 10 years.
Don’t lose money is self-explanatory.
I’d define an acceptable absolute return as being a portfolio that compounds at 10% a year for at least 10 years.
And I’d define an acceptable relative return as being a portfolio that beats the stock market over at least 10 years.
Taxes, Fees, and Turnover
All of these milestones need to be achieved on an after-fees and after-tax equivalent basis. So, if you are paying 35% a year on your net-net portfolio’s gains in taxes and would only be paying 15% a year on your gains in a general investment in the stock market (index fund, buy and hold blue chips, etc.) then you would need to earn 13% a year in net-nets to match your 10% a year in blue chips. The same is true if you are paying high commissions on net-nets.
There are easy ways to avoid doing both. And, in general, the drag on net-net performance from taxes, fees, etc. is actually quite manageable. There’s no reason to pay more in taxes on your net-nets than you would in any other portfolio. Because holding net-nets beyond one year works out fine.
The major issue is illiquidity. But that’s a cost you can control. If you leave bids out for net-nets at the last trade price before you made your buy decision – you will usually have your order filled within a couple weeks to a couple months.
I’ve not experienced trading costs in net-nets that made more than a 2% annual return difference from what my same costs are in stocks $1 billion in market cap and over. But it’s worth noting that I generally don’t bid more than a penny over the last trade price. (To be clear: I don’t change my bid when the price moves. I leave the bid in place. The price moves. In a few days, weeks, or months the price will usually fall back below my bid and I will get my shares then).
Trading costs are determined by turnover more than anything else. And I don’t turn my net-nets over faster than my other investments.
This is not true for many people. And, so, you’ll hear a lot of people talk about the high costs of net-net investing. There are definitely inconveniences in net-net investing. For example, it can take weeks or months to get your order filled. But I’m not convinced there are high costs in a low turnover net-net portfolio.
Principles vs. Percentages
I'm going to show you some backtest results in this article. Don't focus on the annual return percentages. Don't go away thinking this proves you can make a certain return in net-nets. It doesn't. It proves some basic principles: net-net investing works, depending on what kind of net-nets you pick you can get very different results, and the volatility of net-net portfolios comes from the shakiness of the balance sheets of the net-nets they own.
I'm using backtests to illustrate those principles. Not to sell you on the idea of net-net investing. Whether or not you invest in net-nets should be based on whether or not you think you can stomach the experience – and stick to the strategy year after year.
Don't just go out and follow the strategy with the best backtest. Follow the strategy you can stick to.
Any System Works…
It’s actually hard to design a net-net portfolio that will underperform the market over 10 years. Net-nets are so cheap that even doing things like picking the 10 net-nets with the biggest sales declines this year – a truly frightening list of stocks – ends up giving you a basket of stocks that has tended to beat the market. In fact, doing that every year – buying the 10 net-nets with the biggest sales declines, holding them for exactly one year, selling them all, and repeating the process – returned 14% a year over the last 10 years.
To give you an idea of how bad the one-year sales drop in net-nets can be, a list of the 10 net-nets with the biggest percentage year over year sales declines currently features revenue declines ranging from 45% to 100%. Four of the 10 companies have year over year sales declines of 97% or more – meaning, yes, 40% of that list essentially has zero revenues right now.
And yet this group tends not to underperform the market. It’s a terrible strategy. I don’t recommend picking net-nets on this basis – but if somehow you stuck to this insane plan to buy the net-nets with the biggest sales declines, you’d do fine over 10 years.
Again, the thing to focus on is not why such a bizarre strategy would work at all – it’s to focus on how you can screw up net-net investing. You can fail to stick with it.
…If You Stick to It
Of course, you wouldn’t stick to that system. Half the stocks you’d buy each year would probably have almost no sales. No one is going to stick to a plan that involves buying companies that have just seen their sales anywhere from cut in half to totally obliterated.
So, the investor’s goal in net-net investing should not be to find the system that works best. It should be to find the system that is workable. Workable means you can stick to it.
Today, I’ll recommend one super simple system for net-net investing that I think is quite workable.
It’s not the elaborate ranking system you asked for. In fact, it’s insanely simple. But before we get to the system, let’s start with the results.
What if I told you net-net investing could give you results like these:
| 2002 | (-12%) |
| 2003 | 48% |
| 2004 | 44% |
| 2005 | 49% |
| 2006 | 41% |
| 2007 | 20% |
| 2008 | (-37%) |
| 2009 | 33% |
| 2010 | (-3%) |
| 2011 | 6% |
Well, it can.
And what if I told you net-net investing could give you results like these:
| 2002 | (-14%) |
| 2003 | 124% |
| 2004 | 7% |
| 2005 | 49% |
| 2006 | 18% |
| 2007 | (-44%) |
| 2008 | (-61%) |
| 2009 | 279% |
| 2010 | 96% |
| 2011 | 21% |
It can do that too.
Which set of results you see depends on which kind of net-nets you buy.
And if you preferred the first set of returns – you’re not alone. But the second set of annual returns actually leads to higher returns: 22% a year versus 15% a year for the first group.
So, I guess we should rank net-nets so we get the second set of annual returns, right?
I wouldn’t mind earning 22% a year. But I’d rather own the stocks that provided the first set of annual returns.
Why?
Because I can stick to that system. I really could earn 15% a year investing in those net-nets. I would never stick with the second system. I would never earn that 22%. The issue is what results a screen would earn. The issue is what results a human would earn.
Two Systems: One Criterion
Those are both net-net backtests. And they both used the same criterion:
Current Assets/Total Liabilities
There couldn’t be a simpler number. It’s basically a measure of the “leverage” in your net-net purchase. A top heavy fraction means the company has to lose a lot of its current assets to reach insolvency and cause you a permanent loss.
But a ratio very, very close to 1 gives you a huge amount of upside. Think about it. If a net-net has $100 in current assets and $90 in total liabilities it has a current assets to total liabilities ratio of 1.11. If you pay 60% of NCAV for that stock – which is $6 a share – and the company grows current assets by just 10%, you could suddenly have the potential for another 100% return on your investment. That’s because 10% of $100 is $6 and you only paid $6 for the stock. You can easily see how your $6 investment just became a $20 stock.
Okay, but what about the downside?
Oh, it’s big. And likely. A 10% decline in current assets eliminates all of the company’s net current asset value. If the company is losing money and doesn’t have access to long-term financing it could go to zero. So, a stock that cost you anywhere from $6 to $10 could very quickly be worth either $20 a share or $0.
That’s if the current assets to total liabilities ratio is low.
What if it’s high?
Let’s look at the another stock. Assume it’s also trading for $6 a share. And it also has $10 in NCAV. But this time current assets are $11 a share and total liabilities are $1 a share. Again, let’s imagine that current assets increase by 10% to $12.10 a share. That makes NCAV $11.10 a share versus your $6 purchase price. Your upside is now 85% versus the mind blowing 230% it was in our leveraged example. And what about the downside? A 10% decline in current assets would still leave the company with $8 in net current assets. That means a 10% shrinkage in current assets actually leaves more than 30% upside for you.
In other words, bad things happened to the company – and you still made money.
That’s a Ben Graham bargain.
Enough with the hypotheticals. Let’s see what these lists looked like in action.
Why do I think that picking net-nets with the lowest ratio of Current Assets to Total Liabilities is a system nobody would follow – even if they knew it offered the possibility of 20% annual returns?
Two reasons.
Let’s start with the biggest loser among the 10 net-nets with the lowest ratio of current assets to total liabilities for each of the last 10 years.
Remember, if you constructed a portfolio of the 10 net-nets with the lowest Current Assets/Total Liabilities ratio you’d have 10% of your portfolio in one of these awful, awful stocks every year:
| Worst Performing Stock in Portfolio | |
| 2002 | (-87%) |
| 2003 | (-28%) |
| 2004 | (-67%) |
| 2005 | (-99%) |
| 2006 | (-90%) |
| 2007 | (-100%) |
| 2008 | (-100%) |
| 2009 | (-100%) |
| 2010 | (-97%) |
| 2011 | (-66%) |
Also, this is what the years 2007 through 2011 would have felt like:
| Start 2007 | $10,000 |
| End 2007 | $5,600 |
| End 2008 | $2,184 |
| End 2009 | $8,277 |
| End 2010 | $16,224 |
Yeah. No one is going to lose 80% of their portfolio only to stick with the same approach for two more years. Of course, if you did, you’d get an acceptable return – 13% from 2007 through 2010 would be a fine return. The emotions you’d be feeling in late 2008 and early 2009 – would not be fine. And you’d bail on net-net investing entirely.
Compare this to the experience of a portfolio made up of the 10 net-nets with the highest Current Assets/Total Liabilities ratio:
| Start 2007 | $10,000 |
| End 2007 | $11,989 |
| End 2008 | $7,726 |
| End 2009 | $10,283 |
| End 2010 | $9,965 |
You would end up with less money at the end of 2010 than you started with in 2007. But, you’d have been down less than 25% from your starting point in the middle of a worldwide financial panic. And you’d basically be flat over some very tumultuous times. This is a strategy you could stick to.
Two Types of Net-Nets
I’m in favor of a net-net portfolio built from net-nets with a high Current Assets/Total Liabilities ratio. And I’m against including net-nets with a low Current Assets/Total Liabilities ratio. Of course, we all break rules. And I happened to pick a net-net with a low Current Assets/Total Liabilities ratio in this month’s issue of the Ben Graham: Net-Net Newsletter. I hope the stock works out. But I know it’s one of the riskiest picks in the portfolio. In fact, I think it’s the single riskiest stock in the Ben Graham: Net-Net Newsletter’s model portfolio right now. But it’s one of 13 stocks in the portfolio. And I like the odds on the whole group a lot.
It’s important to use common sense when picking net-nets. There’s this idea out there that a net-net portfolio’s performance always has to look like the second column in this table:
| 10 Most Solid Net-Nets | 10 Most Shaky Net-Nets | |
| 2002 | (-12%) | (-14%) |
| 2003 | 48% | 124% |
| 2004 | 44% | 7% |
| 2005 | 49% | 49% |
| 2006 | 41% | 18% |
| 2007 | 20% | (-44%) |
| 2008 | (-37%) | (-61%) |
| 2009 | 33% | 279% |
| 2010 | (-3%) | 96% |
| 2011 | 6% | 21% |
Phil Fisher had this idea that you should never invest in the marginal companies in an industry. Sure, they had more upside when the economy was booming – because their margins would expand more. But they had too much downside in bad times.
It was easy for Phil Fisher to tell a marginal company from a leading company.
And it’s easy for you to tell a solid net-net from a shaky net-net.
The solid net-nets may not outperform the shaky net-nets – just like leading companies may not outperform marginal companies – but it’s your choice which group you want to invest in.
I can’t tell you not to invest in shaky net-nets – after all, they did provide the better 10 year returns. But I can tell you that I prefer investing in the solid net-nets.
I always try to pick solid net-nets for the Ben Graham: Net-Net Newsletter.
You can do the same by looking for net-nets with the highest Current Assets/Total Liabilities ratio.
And I know we didn’t talk enough about specific stocks in this article. So, I’ll end with two columns of stocks. On the left, you have some of the most solid net-nets available today. And on the right, you have some of the most shaky net-nets available today.
Peruse whichever column you think is most interesting.
But I’ll stick to the first column when picking stocks for the Ben Graham: Net-Net Newsletter.
| Solid Net-Nets | Shaky Net-Nets |
| Cadus (KDUS) | Empire Resources (ERSO) |
| Maxygen (MAXY) | SED International (SED) |
| ARC Wireless (ARCW) | Hastings (HAST) |
| Steel Excel (SXCL) | TreeCon (TCOR) |
| Capstone Therapeutics (CAPS) | iParty (IPT) |
| OPT-Sciences (OPST) | QKL Stores (QKLS) |
| Tegal (TGAL) | Lazare Kaplan (LKII) |
| Sen Yu International (CSWG) | Hampshire Group (HAMP) |
| Pioneer Oil & Gas (POGS) | Integrated Electrical Services (IESC) |
| Heelys (HLYS) | Duckwall-ALCO (DUCK) |
Remember, those lists are just based on one number: Current Assets/Total Liabilities. There are other risks – like fraud. I would eliminate a couple stocks on the left for that reason.
If you’ve read my past net-net articles, you can probably guess which ones.
Ask Geoff a Question about Net-Nets
Check out the Ben Graham: Net-Net Newsletter
