Someone who reads my blog sent me this email:
“Hi Geoff,
Thanks for the good work in digging out these Japanese net-nets. After purchasing your report, I hesitated in placing orders. Would be interested to hear your response to any of the following:
1. Most of these 15 stocks have had little price movements in the last 5-10 years. Given that these are typically sub-par business, at least by American standards, with either low returns or no growth, it could take a long time, say five years, for these stocks to move up significantly. Staring at their price history seems to indicate that looking for a 50+% pickup within 1-2 years could be overly optimistic. While time is the friend of a good business, it could become very testy to wait for the market to recognize the value of a mediocre company.
2. Japanese companies seem to generally have low returns on capital, and are not necessarily run with the best interests of shareholders. That makes investing in Japan more difficult. Warren Buffett often cited low ROE as a reason not to go there. While net-nets offer opportunity of a different kind, have you looked at how Japanese net-nets performed historically vs. U.S. net-nets? I somehow have a superstitious feeling that, since Japanese corporate culture is not focusing on shareholders and profits, market players might have less reason to believe values in Japanese net-nets to be realized, but wonder if that's completely unfounded.”
Okay, let’s start with…
1) A valid point. Japanese net-nets could go nowhere for a long time. But I don't think that’s much of a criticism.
In the United States, I don't see the S&P 500 priced to return double digits from here. You may disagree. But, we're basically talking about single digit returns. The stocks listed in the report are profitable. Over time, you should expect that "normal" operating income times 0.55 (in other words, EBIT minus the 45% Japanese corporate tax) will be added to the company's cash pile.
So, you are increasing the value of the stock over time. Obviously, it may be much less of an increase than the time value you put on the money you have to tie up in these stocks, but that's a separate issue.
So, for instance, if a company is adding 5% a year in terms of earnings relative to your purchase price and the business should trade at an NCAV of 1 and now trades at say 70% of NCAV, we're really talking about two separate streams of returns. There's the earnings adding up over time and then there's the one-time valuation gap.
Assume it take five years for a one-time valuation gap of 0.70 times NCAV to close to one times NCAV. That's a 7.4% compound annual growth rate over a five-year time period needed to correct the asset value gap between price and value. Then you also have the build up of earnings — these companies are historically profitable net-nets — which could add any number to value depending on the P/E you are buying at. If it's say a "normal" price of even 20 times earnings — a lot of these Japanese net-nets are trading at much less than 20 times normal earnings — we're still talking about adding 5% of your purchase price to the company’s cash pile each year.
We don't want to get too deep into the timing of the returns and math-y stuff like that, but obviously what I've said can support returns in the 8%-13% range even if it takes five long years for the stocks to trade up just to NCAV. You can see this by just using a paper and pen to add 5% of your purchase price to the stock’s NCAV each year for five years and then calculate what kind of stock price rise you’d need to get the stock from today’s below NCAV price to a price equal to the future NCAV five years from today. Your annual returns are going to be above 8%. Usually, quite a bit closer to 15% for many of these stocks. And that’s if you have to wait patiently for five years – and if the stocks only reach one times NCAV.
Historically — around the world — we've seen that stocks don't stop at NCAV, they blow right past it. And it rarely takes five years to correct an NCAV undervaluation. Now, Japan's stock market is at levels — or was recently as levels — similar to where it was almost a quarter century ago. So, obviously, Japan is a special case in terms of having stocks take a long time to get to appropriate valuation levels.
But, overall, we're talking about a basket of stocks with a downside that is probably not far from the expected upside of a basket of U.S. stocks. In other words, if it takes five years instead of one year or three years for these stocks to reach NCAV (which is actually still considered cheap in most of the world), these net-nets will still give you returns that should match the S&P 500 over those five years.
Again, you may think the S&P 500 is going to do better than 8% – 12% over the next five years. I don't.
As for little or no growth …
The companies on the list are a mixed bag. When compared to net-nets in the United States in terms of real growth, I'd say they actually are more growth-y than U.S. net-nets.
Which isn’t saying much.
I don't see many U.S. net-nets with growth rates over the last 10 years that exceed inflation. I see many that have negative growth rates even in nominal terms.
This brings us to the edge of the currency question.
If growth comes from inflation — does it count if you have to get that growth in a currency that loses its value through the same inflation that helps sales?
Likewise, if no growth comes from no inflation — how badly does that stagnation hurt you if you can hold it in a currency that is not falling in value relative to the goods/assets inside the country?
But then, this depends on what your home currency is too. Most investors want to increase their future purchasing power at home by investing in stocks today. So, again, we see that the risk is in the currency. And that’s the big risk you’d need to be comfortable with if you want to buy a basket of Japanese net-nets.
Eventually, a basket of Japanese net-nets will work in yen. But will it work in dollars or euros orb…
That’s the question each investor has to answer for himself based on his home country’s currency.
Now, back to things which — unlike exchange rates — we have some hope of actually understanding.
Basically, you’re right, these Japanese net-net aren't good businesses, nor are they fast growing businesses. But they are at least as good and fast growing as American net-nets. In fact, I'd say some of them are much, much better. The question you're asking — which is a legitimate one — is whether net-nets are so cheap their cheapness makes up for their lack of quality. Historically, the answer is yes. Net-nets — despite being of abysmal business quality — outperform the highest quality companies in the same country because net-nets are assigned the wrong odds by Mr. Market.
But, yes, lack of quality/growth – the risk of stuck money – is the big risk with individual net-nets in all countries at all times. As a basket, this risk disappears. It’s simply too hard for the market to be right in its extreme pessimism about a basket of expected losers if it prices them this low. You can lose money in individual net-nets. But, it’s hard to lose money in baskets of net-nets because the level of pessimism baked into the group would require extraordinary prescience on the part of Mr. Market.
No. I haven't looked at how Japanese net-nets performed historically. It's hard enough getting good data on U.S. net-nets. I'd love to have data on Japanese net-nets, but I just don't have that kind of info.
I know that the Japanese net-nets listed in my report on 15 Japanese net-nets have the traits that lead to outperformance with lower risk when seen in U.S. net-net portfolios.
But I don't know if those traits work in Japan as well.
2) Yes. Japanese companies have low ROEs.
To be entirely honest, I actually think people have this backwards when talking about Japanese net-nets versus blue chip Japanese stocks. If anything, growth in Japan is a lot less interesting than value because Japanese companies do a terrible job of making growth profitable.
The risk to net-nets in Japan is not from low ROE (many U.S. net-nets have low-to-negative ROE over the last 10 years; some have zero retained earnings or have lost money cumulatively over their entire business history, etc.). The risk for Japanese net-nets is poor capital allocation.
Although exactly what poor capital allocation means in Japan is hard to say. These companies aren’t actually investing additional capital in their business at rates lower than the yield on long-term Japanese bonds. Remember, 30-year government bonds in Japan yield 2.3%. If you go over the last 10 years for these net-nets, you actually don’t see much evidence that they invested additional capital in the business at rates of 3% or whatever would be clearly value destroying in Japan.
You see them holding more and more cash.
But if long-term bond yields are at 3% or below, that’s what you should do — hold cash. There’s little reason to pay out cash if the people you pay it to — your shareholders — are only going to earn 3% or 4% on their money. There’s not much of a margin of safety there.
Now, it’s true that a Japanese company that hoards cash isn’t going to earn 3% or 4% on its cash, but it is going to preserve the option of investing that money in the future, and you’re going to eliminate all risks of insolvency.
So, it’s actually hard for me to condemn public Japanese companies for holding cash.
Not buying back stock is a different story. But Japan is very different from the U.S. in that way. Obviously, Japanese companies and Japanese investors should be buying stock.
But, again, it’s hard to criticize a company for holding additional cash. That’s actually preferable — from my perspective — to earning 3% or 4% on money they invest in their business. I’d rather they have dry powder earning zero percent than get locked into long-term rates of return of 4%. And that’s a real possibility in Japan.
If you look at many of the net-nets in the report, they’re hoarding cash rather than investing heavily in their business.
What about dividends?
How do Japanese net-nets compare to American net-nets?
Now, this is a mixed bag. Japanese net-nets seem to pay at least as much — in fact, to me it looks like a lot more — in dividends than U.S. net-nets. This is not due to high payout ratios. It is due to the higher earnings power of Japanese net-nets. It’s also due to having many fewer Japanese net-nets who pay no dividends, unlike in the U.S., probably because Japanese net-nets often lack a serious, devastating recent sales/earnings downturns. Japanese net-nets are less in peril than American net-nets. American net-nets are stocks going through very serious — often existential — threats due to scandal, loss of customer, technological change, legal issues, etc.
Japanese net-nets are often net-nets because they have seen no price improvement in the stock for a decade — in fact, maybe a price decline — despite accumulating more and more cash during that period.
Some U.S. net-nets are like that. Those are actually the ones I buy for myself.
Yes. As strange as it sounds, I prefer the stocks that have gone nowhere for a decade. I like it when you have a decent business that has been chugging along — building a cash pile — for the last decade while the stock price has gone nowhere or slowly fallen. That’s not to say that huge one-time drops in a stock’s price due to some devastating event don’t provide buying opportunities. They might.
But I prefer to look for net-nets with an absence of change.
So, when I look at Japanese net-nets, I’m just looking for the same kinds of situations I like in the United States.
Basically, I like boring businesses with lots of cash and a history of profitability over the last 10–15 years.
That's what I looked for in Japan.
Now obviously, yes, there are bad businesses in Japan and some of them are net-nets. It's a risk there. I think it's a much, much bigger risk here right now.
If I showed you a list of U.S. net-nets — there are about 60 American net-nets right now — it would turn your stomach. They're very, very bad businesses. A lot are young, have retained deficits, have pretty questionable businesses in the sense that they have never shown they can make money in that industry. Lots are super tech-oriented. They depend on specific customers, standards, products, etc. that could change or have changed and rendered them totally incapable of making money in their historical business.
It's not a good looking group. Net-nets always have warts. The current American crop of net-nets is the worst looking bunch I've ever seen.
The quality of Japanese net-nets is much, much higher than the quality of American net-nets. No doubt about that.
Yes. Japanese net-nets are extremely unlikely to be made profitable investments through activism. Forget about shareholder activism. However, net-nets work in countries and times where there is little or no shareholder activism.
A net-net with an activist shareholder would be a wonderful thing. It doesn't happen very often. Anywhere. And it’s never going to happen in Japan.
But it's a myth that net-nets realize value through liquidations. It's really about Microeconomics 101. If you've got capital invested in a lousy business, over time, people either stop putting capital into the business, or they remove capital, or they earn higher returns on capital in some other way.
Basically, net-nets go up in value because the market reappraises them and because their earnings normalize, not just in the sense of normal sales or margins, but in the sense of earning normal returns on capital.
Think about why buying stocks at a low price-to-book ratio works …
… Now think about net-nets.
Price-to-book is measuring very special capital, very stuck capital. It's looking at machines and stuff that would be worthless in liquidation because maybe only five companies in the world actually want to buy a machine that makes that particular product and if there’s a worldwide downturn in that industry then probably nobody needs that machine. Price-to-book is sometimes measuring stuff that is literally worthless. No one will pay what it takes to move the thing. Competitors are praying they burn down the plant rather than hoping to buy any of it. Equipment like that is included in price-to-book. Factories like that are included in price-to-book.
Net-nets have more general, more liquid capital. Even invested capital — which is only part of Japanese net-nets NCAV — is in stuff like inventory and receivables that will be liquidated as sales decline. Investors underestimate this self-liquidation process. But you'll actually see some net-nets report accrual losses and positive free cash flow simultaneously because they are reducing working capital year after year. That’s what overcapitalized, dying businesses do. They shrink.
My point is just that the whole activism thing is a myth. Ben Graham didn't buy net-nets because he actually expected them to liquidate. Ben Graham bought net-nets because if an asset can be bought for less than its scrap value and it’s actually producing income as long as it’s in service, you know you're getting a bargain. You don't need to actually scrap the asset. You just need to pay less than scrap value when you buy it.
So, yes, there's no chance of any shareholder activism in Japan. You are completely at the mercy of the managers of these companies. You will only get what capital they choose to return. No shareholder value will ever be unlocked.
That's just something you've got to live it. If you can't, I wouldn't buy any of the stocks on that list of 15 Japanese net-nets.
Follow Geoff at Gannon On Investing
“Hi Geoff,
Thanks for the good work in digging out these Japanese net-nets. After purchasing your report, I hesitated in placing orders. Would be interested to hear your response to any of the following:
1. Most of these 15 stocks have had little price movements in the last 5-10 years. Given that these are typically sub-par business, at least by American standards, with either low returns or no growth, it could take a long time, say five years, for these stocks to move up significantly. Staring at their price history seems to indicate that looking for a 50+% pickup within 1-2 years could be overly optimistic. While time is the friend of a good business, it could become very testy to wait for the market to recognize the value of a mediocre company.
2. Japanese companies seem to generally have low returns on capital, and are not necessarily run with the best interests of shareholders. That makes investing in Japan more difficult. Warren Buffett often cited low ROE as a reason not to go there. While net-nets offer opportunity of a different kind, have you looked at how Japanese net-nets performed historically vs. U.S. net-nets? I somehow have a superstitious feeling that, since Japanese corporate culture is not focusing on shareholders and profits, market players might have less reason to believe values in Japanese net-nets to be realized, but wonder if that's completely unfounded.”
Okay, let’s start with…
1) A valid point. Japanese net-nets could go nowhere for a long time. But I don't think that’s much of a criticism.
In the United States, I don't see the S&P 500 priced to return double digits from here. You may disagree. But, we're basically talking about single digit returns. The stocks listed in the report are profitable. Over time, you should expect that "normal" operating income times 0.55 (in other words, EBIT minus the 45% Japanese corporate tax) will be added to the company's cash pile.
So, you are increasing the value of the stock over time. Obviously, it may be much less of an increase than the time value you put on the money you have to tie up in these stocks, but that's a separate issue.
So, for instance, if a company is adding 5% a year in terms of earnings relative to your purchase price and the business should trade at an NCAV of 1 and now trades at say 70% of NCAV, we're really talking about two separate streams of returns. There's the earnings adding up over time and then there's the one-time valuation gap.
Assume it take five years for a one-time valuation gap of 0.70 times NCAV to close to one times NCAV. That's a 7.4% compound annual growth rate over a five-year time period needed to correct the asset value gap between price and value. Then you also have the build up of earnings — these companies are historically profitable net-nets — which could add any number to value depending on the P/E you are buying at. If it's say a "normal" price of even 20 times earnings — a lot of these Japanese net-nets are trading at much less than 20 times normal earnings — we're still talking about adding 5% of your purchase price to the company’s cash pile each year.
We don't want to get too deep into the timing of the returns and math-y stuff like that, but obviously what I've said can support returns in the 8%-13% range even if it takes five long years for the stocks to trade up just to NCAV. You can see this by just using a paper and pen to add 5% of your purchase price to the stock’s NCAV each year for five years and then calculate what kind of stock price rise you’d need to get the stock from today’s below NCAV price to a price equal to the future NCAV five years from today. Your annual returns are going to be above 8%. Usually, quite a bit closer to 15% for many of these stocks. And that’s if you have to wait patiently for five years – and if the stocks only reach one times NCAV.
Historically — around the world — we've seen that stocks don't stop at NCAV, they blow right past it. And it rarely takes five years to correct an NCAV undervaluation. Now, Japan's stock market is at levels — or was recently as levels — similar to where it was almost a quarter century ago. So, obviously, Japan is a special case in terms of having stocks take a long time to get to appropriate valuation levels.
But, overall, we're talking about a basket of stocks with a downside that is probably not far from the expected upside of a basket of U.S. stocks. In other words, if it takes five years instead of one year or three years for these stocks to reach NCAV (which is actually still considered cheap in most of the world), these net-nets will still give you returns that should match the S&P 500 over those five years.
Again, you may think the S&P 500 is going to do better than 8% – 12% over the next five years. I don't.
As for little or no growth …
The companies on the list are a mixed bag. When compared to net-nets in the United States in terms of real growth, I'd say they actually are more growth-y than U.S. net-nets.
Which isn’t saying much.
I don't see many U.S. net-nets with growth rates over the last 10 years that exceed inflation. I see many that have negative growth rates even in nominal terms.
This brings us to the edge of the currency question.
If growth comes from inflation — does it count if you have to get that growth in a currency that loses its value through the same inflation that helps sales?
Likewise, if no growth comes from no inflation — how badly does that stagnation hurt you if you can hold it in a currency that is not falling in value relative to the goods/assets inside the country?
But then, this depends on what your home currency is too. Most investors want to increase their future purchasing power at home by investing in stocks today. So, again, we see that the risk is in the currency. And that’s the big risk you’d need to be comfortable with if you want to buy a basket of Japanese net-nets.
Eventually, a basket of Japanese net-nets will work in yen. But will it work in dollars or euros orb…
That’s the question each investor has to answer for himself based on his home country’s currency.
Now, back to things which — unlike exchange rates — we have some hope of actually understanding.
Basically, you’re right, these Japanese net-net aren't good businesses, nor are they fast growing businesses. But they are at least as good and fast growing as American net-nets. In fact, I'd say some of them are much, much better. The question you're asking — which is a legitimate one — is whether net-nets are so cheap their cheapness makes up for their lack of quality. Historically, the answer is yes. Net-nets — despite being of abysmal business quality — outperform the highest quality companies in the same country because net-nets are assigned the wrong odds by Mr. Market.
But, yes, lack of quality/growth – the risk of stuck money – is the big risk with individual net-nets in all countries at all times. As a basket, this risk disappears. It’s simply too hard for the market to be right in its extreme pessimism about a basket of expected losers if it prices them this low. You can lose money in individual net-nets. But, it’s hard to lose money in baskets of net-nets because the level of pessimism baked into the group would require extraordinary prescience on the part of Mr. Market.
No. I haven't looked at how Japanese net-nets performed historically. It's hard enough getting good data on U.S. net-nets. I'd love to have data on Japanese net-nets, but I just don't have that kind of info.
I know that the Japanese net-nets listed in my report on 15 Japanese net-nets have the traits that lead to outperformance with lower risk when seen in U.S. net-net portfolios.
But I don't know if those traits work in Japan as well.
2) Yes. Japanese companies have low ROEs.
To be entirely honest, I actually think people have this backwards when talking about Japanese net-nets versus blue chip Japanese stocks. If anything, growth in Japan is a lot less interesting than value because Japanese companies do a terrible job of making growth profitable.
The risk to net-nets in Japan is not from low ROE (many U.S. net-nets have low-to-negative ROE over the last 10 years; some have zero retained earnings or have lost money cumulatively over their entire business history, etc.). The risk for Japanese net-nets is poor capital allocation.
Although exactly what poor capital allocation means in Japan is hard to say. These companies aren’t actually investing additional capital in their business at rates lower than the yield on long-term Japanese bonds. Remember, 30-year government bonds in Japan yield 2.3%. If you go over the last 10 years for these net-nets, you actually don’t see much evidence that they invested additional capital in the business at rates of 3% or whatever would be clearly value destroying in Japan.
You see them holding more and more cash.
But if long-term bond yields are at 3% or below, that’s what you should do — hold cash. There’s little reason to pay out cash if the people you pay it to — your shareholders — are only going to earn 3% or 4% on their money. There’s not much of a margin of safety there.
Now, it’s true that a Japanese company that hoards cash isn’t going to earn 3% or 4% on its cash, but it is going to preserve the option of investing that money in the future, and you’re going to eliminate all risks of insolvency.
So, it’s actually hard for me to condemn public Japanese companies for holding cash.
Not buying back stock is a different story. But Japan is very different from the U.S. in that way. Obviously, Japanese companies and Japanese investors should be buying stock.
But, again, it’s hard to criticize a company for holding additional cash. That’s actually preferable — from my perspective — to earning 3% or 4% on money they invest in their business. I’d rather they have dry powder earning zero percent than get locked into long-term rates of return of 4%. And that’s a real possibility in Japan.
If you look at many of the net-nets in the report, they’re hoarding cash rather than investing heavily in their business.
What about dividends?
How do Japanese net-nets compare to American net-nets?
Now, this is a mixed bag. Japanese net-nets seem to pay at least as much — in fact, to me it looks like a lot more — in dividends than U.S. net-nets. This is not due to high payout ratios. It is due to the higher earnings power of Japanese net-nets. It’s also due to having many fewer Japanese net-nets who pay no dividends, unlike in the U.S., probably because Japanese net-nets often lack a serious, devastating recent sales/earnings downturns. Japanese net-nets are less in peril than American net-nets. American net-nets are stocks going through very serious — often existential — threats due to scandal, loss of customer, technological change, legal issues, etc.
Japanese net-nets are often net-nets because they have seen no price improvement in the stock for a decade — in fact, maybe a price decline — despite accumulating more and more cash during that period.
Some U.S. net-nets are like that. Those are actually the ones I buy for myself.
Yes. As strange as it sounds, I prefer the stocks that have gone nowhere for a decade. I like it when you have a decent business that has been chugging along — building a cash pile — for the last decade while the stock price has gone nowhere or slowly fallen. That’s not to say that huge one-time drops in a stock’s price due to some devastating event don’t provide buying opportunities. They might.
But I prefer to look for net-nets with an absence of change.
So, when I look at Japanese net-nets, I’m just looking for the same kinds of situations I like in the United States.
Basically, I like boring businesses with lots of cash and a history of profitability over the last 10–15 years.
That's what I looked for in Japan.
Now obviously, yes, there are bad businesses in Japan and some of them are net-nets. It's a risk there. I think it's a much, much bigger risk here right now.
If I showed you a list of U.S. net-nets — there are about 60 American net-nets right now — it would turn your stomach. They're very, very bad businesses. A lot are young, have retained deficits, have pretty questionable businesses in the sense that they have never shown they can make money in that industry. Lots are super tech-oriented. They depend on specific customers, standards, products, etc. that could change or have changed and rendered them totally incapable of making money in their historical business.
It's not a good looking group. Net-nets always have warts. The current American crop of net-nets is the worst looking bunch I've ever seen.
The quality of Japanese net-nets is much, much higher than the quality of American net-nets. No doubt about that.
Yes. Japanese net-nets are extremely unlikely to be made profitable investments through activism. Forget about shareholder activism. However, net-nets work in countries and times where there is little or no shareholder activism.
A net-net with an activist shareholder would be a wonderful thing. It doesn't happen very often. Anywhere. And it’s never going to happen in Japan.
But it's a myth that net-nets realize value through liquidations. It's really about Microeconomics 101. If you've got capital invested in a lousy business, over time, people either stop putting capital into the business, or they remove capital, or they earn higher returns on capital in some other way.
Basically, net-nets go up in value because the market reappraises them and because their earnings normalize, not just in the sense of normal sales or margins, but in the sense of earning normal returns on capital.
Think about why buying stocks at a low price-to-book ratio works …
… Now think about net-nets.
Price-to-book is measuring very special capital, very stuck capital. It's looking at machines and stuff that would be worthless in liquidation because maybe only five companies in the world actually want to buy a machine that makes that particular product and if there’s a worldwide downturn in that industry then probably nobody needs that machine. Price-to-book is sometimes measuring stuff that is literally worthless. No one will pay what it takes to move the thing. Competitors are praying they burn down the plant rather than hoping to buy any of it. Equipment like that is included in price-to-book. Factories like that are included in price-to-book.
Net-nets have more general, more liquid capital. Even invested capital — which is only part of Japanese net-nets NCAV — is in stuff like inventory and receivables that will be liquidated as sales decline. Investors underestimate this self-liquidation process. But you'll actually see some net-nets report accrual losses and positive free cash flow simultaneously because they are reducing working capital year after year. That’s what overcapitalized, dying businesses do. They shrink.
My point is just that the whole activism thing is a myth. Ben Graham didn't buy net-nets because he actually expected them to liquidate. Ben Graham bought net-nets because if an asset can be bought for less than its scrap value and it’s actually producing income as long as it’s in service, you know you're getting a bargain. You don't need to actually scrap the asset. You just need to pay less than scrap value when you buy it.
So, yes, there's no chance of any shareholder activism in Japan. You are completely at the mercy of the managers of these companies. You will only get what capital they choose to return. No shareholder value will ever be unlocked.
That's just something you've got to live it. If you can't, I wouldn't buy any of the stocks on that list of 15 Japanese net-nets.
Follow Geoff at Gannon On Investing
