rukawa Posted August 17, 2014 Posted August 17, 2014 I was wondering what peoples favorite stocks screens are? Right now I am running one on PE using average 10 year earnings.
west Posted August 18, 2014 Posted August 18, 2014 If you're really interested about this, I would recommend looking at the book Quantitative Value and also looking at what Damodaran has written about/lectured about on different price and ev ratios and what the fundamental drivers/companion variables are for those ratios. This being said, at least in the US you're probably going to have better luck cloning ideas from others versus trying to screen for things...
LC Posted August 18, 2014 Posted August 18, 2014 I really only look at screens for business metrics (margins, sales growth, etc.) and not valuation metrics. So I wake up one day and say, today I feel like reading about a company with really nice gross margins, let me go find one and see what it looks like.
rukawa Posted August 18, 2014 Author Posted August 18, 2014 I don't get it. Is there a reason why people don't use screens more? I have observed this before that value investors tend to avoid using screens but I don't understand the reason. And why wouldn't this work in the US? Is the idea that the market is too efficient for screens to work?
topofeaturellc Posted August 18, 2014 Posted August 18, 2014 I'm a huge believer in screens. I think anyone really interested in this topic should read the book West reccs: "Quantitative Value" Personally I mostly use a modified version of their approach that mean reverts returns a bit. I disagree that screens don't work in the US generally, its just that valuation spreads are tight right now in the US, so a quantitative value approach is probably not primed to perform particularly well until the next crisis blows spreads back out. ETA: The Quantitative Value guys use an EV/EBIT screen with some other factors to try to avoid value traps. Personally I think if you screen to source ideas rather than construct a PF from the process, you can manually find and reject the value traps.
west Posted August 18, 2014 Posted August 18, 2014 Yes. I'd say the US stock market is either too "efficient" or tends to be on the overvalued side versus undervalued side for classic screening techniques to work. I think as far a simple screens go, you can still get some cheap stocks by looking for P/B bargains, and then looking at their earnings history. You can also use the price/10 year earnings to some effect as you suggested. (An aside: These techniques still ignore the effect on balance sheet or off balance sheet debt has, unlike say EV/EBIT or EV/10 year EBIT. But good luck finding a free screener that does that.) However, even with the above mentioned screening techniques, *good* opportunities (not just ugly and justifiably priced low stocks) in the US don't seem to come along that often. If you look outside the US (or in the overlooked parts of the US like the OTC markets) it's much easier to screen for things. I think Nate of Oddball Stocks (www.oddballstocks.com) has a couple of posts on this topic. It's probably worth digging through his website to find them.
topofeaturellc Posted August 18, 2014 Posted August 18, 2014 I was wondering what peoples favorite stocks screens are? Right now I am running one on PE using average 10 year earnings. You are probably aware of this, but that's a Graham P/E. Its generally a decent tool, but I find that the 10 year look back is too long. You end up with many permanently impaired names that screen very cheaply. Personally I use 5 year ROE to drive my normal earnings and screen on that. I also have this kind of funky thing I do that forecasts normal earnings based purely on mean reversion in margins, capital required, leverage and coverage rations, and sector metrics. It is of more use for businesses that don't have a lot fixed assets.
Ross812 Posted August 18, 2014 Posted August 18, 2014 I don't get it. Is there a reason why people don't use screens more? I have observed this before that value investors tend to avoid using screens but I don't understand the reason. And why wouldn't this work in the US? Is the idea that the market is too efficient for screens to work? It is far better to learn about one industry in depth and build your circle of competence. There is no magic screen that returned companies that would allow you to compound at 15% a year. Value investing today requires that you understand something about a business that the market is missing. The market doesn't miss data that shows in a few clicks of a mouse. The reason there are so many piggy back value investors or value investors that troll for insider buying is its much easier to focus on companies where management is confident the company's value is increasing, or where an institutional investor with a small army of analysts has found something. Go look at the new buys on dataroma. There are about 50 new buys to keep you busy. If you choose to focus on one industry within your circle of competence, there may only be a name or two to check out.
west Posted August 18, 2014 Posted August 18, 2014 To clarify, as topofeaturellc suggests, the market may be a little overvalued *now* for screens to work well, but they have worked well in the past five or so years. However, you have to believe that some day stocks will be as cheap as they have been 2009-2012 and, perhaps, that the market is overvalued now and will revert to being fairly valued. And when the tide goes out, there will be some easy-to-screen-for bargains out there. I personally believe that if you wait for that you may be waiting for a while. And why wait when there's cheap stuff abroad or through not-picked-up-by-classical-screeners methods?
topofeaturellc Posted August 18, 2014 Posted August 18, 2014 Yes. I'd say the US stock market is either too "efficient" or tends to be on the overvalued side versus undervalued side for classic screening techniques to work. I think as far a simple screens go, you can still get some cheap stocks by looking for P/B bargains, and then looking at their earnings history. You can also use the price/10 year earnings to some effect as you suggested. (An aside: These techniques still ignore the effect on balance sheet or off balance sheet debt has, unlike say EV/EBIT or EV/10 year EBIT. But good luck finding a free screener that does that.) However, even with the above mentioned screening techniques, *good* opportunities (not just ugly and justifiably priced low stocks) in the US don't seem to come along that often. If you look outside the US (or in the overlooked parts of the US like the OTC markets) it's much easier to screen for things. I think Nate of Oddball Stocks (www.oddballstocks.com) has a couple of posts on this topic. It's probably worth digging through his website to find them. While it is absolutely correct that P/B does not take leverage into account, the difference in performance between P/B and EV/EBIT is really pretty small. Its not really worth worrying about at the screening stage, and instead is something to focus on at the name level. Also somewhat surprisingly to me the levered names in the cheapest quintile of P/B don't reliably underperform the unlevered names. I think because the levered names that survive have some massive returns. An I def disagree about screening not working in the US. Donald Smith has basically crushed the benchmarks for decades doing nothing but buying cheap Price/Tang Book.
topofeaturellc Posted August 18, 2014 Posted August 18, 2014 I don't get it. Is there a reason why people don't use screens more? I have observed this before that value investors tend to avoid using screens but I don't understand the reason. And why wouldn't this work in the US? Is the idea that the market is too efficient for screens to work? It is far better to learn about one industry in depth and build your circle of competence. There is no magic screen that returned companies that would allow you to compound at 15% a year. Value investing today requires that you understand something about a business that the market is missing. The market doesn't miss data that shows in a few clicks of a mouse. The reason there are so many piggy back value investors or value investors that troll for insider buying is its much easier to focus on companies where management is confident the company's value is increasing, or where an institutional investor with a small army of analysts has found something. Go look at the new buys on dataroma. There are about 50 new buys to keep you busy. If you choose to focus on one industry within your circle of competence, there may only be a name or two to check out. spending time learning an industry in-depth is actually detrimental to returns. You become over-confident about your ability to predict the future and essentially force yourself to buy things because of all the time you've invested. In reality there are only two kinds of businesses - Return on Capital Businesses, and Margin businesses. Figure out how to look at those and you can look at anything. At the end of the day research itself is a commodity. No one consistently gets paid for knowing businesses better then someone else.
west Posted August 18, 2014 Posted August 18, 2014 While it is absolutely correct that P/B does not take leverage into account, the difference in performance between P/B and EV/EBIT is really pretty small. Its not really worth worrying about at the screening stage, and instead is something to focus on at the name level. Also somewhat surprisingly to me the levered names in the cheapest quintile of P/B don't reliably underperform the unlevered names. I think because the levered names that survive have some massive returns. An I def disagree about screening not working in the US. Donald Smith has basically crushed the benchmarks for decades doing nothing but buying cheap Price/Tang Book. topofeaturellc, I should clarify, I think it's hard to find individual cheap stocks in the US using classic screening techniques, at least outside of market pullbacks. Too many people are already combing the market using them. If you do a basket based approach where you don't know anything about the stocks beyond the numbers (I'm assuming this is what Donald Smith is doing?) you will most likely do well. This isn't the route I would take though because I like to understand what's going on with my stocks. I'm not sure I wouldn't panic in a market downturn if I didn't know everything that was going on with the stocks I was holding.
oddballstocks Posted August 18, 2014 Posted August 18, 2014 I'm a fan of screening, I don't know why they're so overlooked. Interesting notes from topofeaturellc on Donald Smith. I know Tweedy Browne crushed the benchmarks for years just buying ALL net-nets at 2/3 of NCAV and selling at NCAV. They then moved to BV and have had to change their strategy as they've become more successful. To the poster who said you need to become an expert on an industry before investing I disagree. There are a number of investors like Schloss who did 15% and were never experts. I'd argue that 99% of value investors who fancy themselves as 'experts' are only fooling themselves. If you want to be an expert in an industry you're bound to become a consultant in some industry. When you can charge a few grand an hour for advice why would you be investing? Build a business instead. For us mere mortals who aren't fooling ourselves the name of the game is to make a profit. I don't need to impress anyone, and I've found that buying cheap things and selling them when the market fairly values them has resulted in earnings for myself. I'll stick with this, so far I haven't really had to learn how much of what I own works, I don't think I ever will either. I'm making money on market psychology, not because I can identify the smartest widget manager. On screening I find it generates a great list to look through further. I had a list of companies I ran through last night, all earning 15% ROE's or higher and below 80% of BV. None were worth further research, each had what I'd consider a fatal flaw eliminating them from consideration. Two had averse auditor opinions. The point of a screen for me is to limit the universe, not find a perfect investment. I tend to conduct broader screens. I'll do screens like P/B < .8 and ROE > .01% (profitable and below BV). Or do 5 year growth and below BV or something. My hope is to get 25-30 results that I can pick through on my own. I've screened, and I've also gone A-Z on lists of stocks. For my money screening is worth it. Going A-Z I noticed that I was manually screening on factors that a computer could have done. There were a few companies I've found that a screener wouldn't turn up, but I'm not sure of the time tradeoff.
oddballstocks Posted August 18, 2014 Posted August 18, 2014 West, I remember Yahoo! had a free screener that could do EV/EBIT. It's a very clunky Java screener that you need to dig around for. I don't know if it's still updated, I'm not even sure if it's screening over Yahoo's data (which is CapIQ).
west Posted August 18, 2014 Posted August 18, 2014 West, I remember Yahoo! had a free screener that could do EV/EBIT. It's a very clunky Java screener that you need to dig around for. I don't know if it's still updated, I'm not even sure if it's screening over Yahoo's data (which is CapIQ). Oddball, thanks for the recommendation. Are you fearful that beginning value investors using these one year earnings screens might put too much faith in them? I'm always a little fearful of that, which is why it's hard to recommend one year of earnings screening techniques without knowing how sophisticated the person is I'm talking too... My hunch is that it's best for most beginner value investors (or seasoned value investors who like to keep things simple) to avoid one year P/E or EV/EBIT or any other earnings based data and focus on P/B and build up historic earnings, EBIT and/or EBITDA by hand themselves. I could be wrong on this though...
yseric Posted August 18, 2014 Posted August 18, 2014 While we're on the topic of stock screening, any good screening tools you guys can suggest? I used to use Yahoo Screener, but I think it's not longer being updated... Any help would be greatly appreciated! :)
LC Posted August 18, 2014 Posted August 18, 2014 Also screening is just one tool in the toolbox. Sometimes it doesn't turn up obvious bargains like Apple when it was 400/share (pre-split). Cash adjusted p/e was really low but a screen wouldn't show that. That's why I use business metrics, make straightforward adjustments (excess cash, accrual vs cash flow, etc.) to find a rough multiple for the business, and then try to decide whether any "cheapness" is due to temporary or permanent issues.
topofeaturellc Posted August 18, 2014 Posted August 18, 2014 I agree in theory re: one year numbers, but the data show that its not a terrible option. Certainly better than none at all. OBS - I think its a mistake to throw out businesses that are losing money today. If you don't want to throw those out by hand maybe pick a minimum ten year ROE or ROTC number. I usually use ROTC over time for my minimum quality threshold for Capital businesses. Margins biz can be a little more complex - tho broken margin bizzes tend to not fix themselves anyway. AFAIK Donald Smith is not basket guy. My impression is that they generate a universe of 200 names (10% of the R2K) and then use fundamental research to narrow things down. Right now he's got 81 holdings but half his portfolio is in his top ten names. Here is a decent interview with him in the G&D letter. He's pretty low profile as he's been closed for a long time and his only MF AFAIK is a portion of a subadvised vanguard fund. He's got a great business for himself though. Like 4 bil in Small Cap AUM (so prob like 1% fees) and 7 employees. IIRC they basically refuse to speak to other investors. http://www8.gsb.columbia.edu/sites/valueinvesting/files/files/Newsletter%20Issue%2010_Fall2010_v4.pdf
topofeaturellc Posted August 18, 2014 Posted August 18, 2014 Also screening is just one tool in the toolbox. Sometimes it doesn't turn up obvious bargains like Apple when it was 400/share (pre-split). Cash adjusted p/e was really low but a screen wouldn't show that. That's why I use business metrics, make straightforward adjustments (excess cash, accrual vs cash flow, etc.) to find a rough multiple for the business, and then try to decide whether any "cheapness" is due to temporary or permanent issues. Any process - any process - is going to miss ideas that turn out to be good ones. What you are basically saying is that the process saves you from more supposedly good ideas that turn out to be crap than it keeps you from buying good ideas that actually are good ideas. I think that's a pretty important idea in investing.
Ross812 Posted August 18, 2014 Posted August 18, 2014 I don't get it. Is there a reason why people don't use screens more? I have observed this before that value investors tend to avoid using screens but I don't understand the reason. And why wouldn't this work in the US? Is the idea that the market is too efficient for screens to work? It is far better to learn about one industry in depth and build your circle of competence. There is no magic screen that returned companies that would allow you to compound at 15% a year. Value investing today requires that you understand something about a business that the market is missing. The market doesn't miss data that shows in a few clicks of a mouse. The reason there are so many piggy back value investors or value investors that troll for insider buying is its much easier to focus on companies where management is confident the company's value is increasing, or where an institutional investor with a small army of analysts has found something. Go look at the new buys on dataroma. There are about 50 new buys to keep you busy. If you choose to focus on one industry within your circle of competence, there may only be a name or two to check out. spending time learning an industry in-depth is actually detrimental to returns. You become over-confident about your ability to predict the future and essentially force yourself to buy things because of all the time you've invested. In reality there are only two kinds of businesses - Return on Capital Businesses, and Margin businesses. Figure out how to look at those and you can look at anything. At the end of the day research itself is a commodity. No one consistently gets paid for knowing businesses better then someone else. If you are a basket type investor who is going to buy 50 names that you turn up with a screen I would agree with you. If you choose a more concentrated approach you need to understand the relevant metrics within a company's given industry. Maybe I implied too much by saying "in depth knowledge" as I was not talking about anything remotely close to becoming a consultant of some sort. I am referring to having the knowledge of an industry to at least understand what is important. You may screen for two energy companies with low P/B ratios. Which one is better? Does one have more implied reserves than the other? At what stage are they allowed to count oil in the ground into their BV? How much does an efficient producer pay to extract oil per barrel? This is not particularly in depth knowledge. You could teach yourself most of it in a day or two of reading.
oddballstocks Posted August 18, 2014 Posted August 18, 2014 I agree in theory re: one year numbers, but the data show that its not a terrible option. Certainly better than none at all. OBS - I think its a mistake to throw out businesses that are losing money today. If you don't want to throw those out by hand maybe pick a minimum ten year ROE or ROTC number. I usually use ROTC over time for my minimum quality threshold for Capital businesses. Margins biz can be a little more complex - tho broken margin bizzes tend to not fix themselves anyway. AFAIK Donald Smith is not basket guy. My impression is that they generate a universe of 200 names (10% of the R2K) and then use fundamental research to narrow things down. Right now he's got 81 holdings but half his portfolio is in his top ten names. Here is a decent interview with him in the G&D letter. He's pretty low profile as he's been closed for a long time and his only MF AFAIK is a portion of a subadvised vanguard fund. He's got a great business for himself though. Like 4 bil in Small Cap AUM (so prob like 1% fees) and 7 employees. IIRC they basically refuse to speak to other investors. http://www8.gsb.columbia.edu/sites/valueinvesting/files/files/Newsletter%20Issue%2010_Fall2010_v4.pdf Good points on money losing companies. I find the one year earnings screens tough as well. In terms of tools I know Screener.co is decent, I've also used FT.com, Bloomberg and OTCMarkets. Screener.co seems like the best tool for someone who isn't going to pony up for a Bloomberg Terminal. They pull from CapIQ I believe, so it's a way to pay for just the CapIQ screening and not their data feed. The Donald Smith interview is fascinating. Here's the secret of value investing, there are a lot of guys like him running simple investment strategies and generating great returns. The problem is it isn't 'sexy'. Readers want to know about someone who spent 500 hours reading old annual reports and uncovered some secret franchise value of some company. They don't want to know that buying low P/B companies and low EV/EBIT companies and selling them when they rise works, it's boring and has no curb appeal.
topofeaturellc Posted August 18, 2014 Posted August 18, 2014 I don't get it. Is there a reason why people don't use screens more? I have observed this before that value investors tend to avoid using screens but I don't understand the reason. And why wouldn't this work in the US? Is the idea that the market is too efficient for screens to work? It is far better to learn about one industry in depth and build your circle of competence. There is no magic screen that returned companies that would allow you to compound at 15% a year. Value investing today requires that you understand something about a business that the market is missing. The market doesn't miss data that shows in a few clicks of a mouse. The reason there are so many piggy back value investors or value investors that troll for insider buying is its much easier to focus on companies where management is confident the company's value is increasing, or where an institutional investor with a small army of analysts has found something. Go look at the new buys on dataroma. There are about 50 new buys to keep you busy. If you choose to focus on one industry within your circle of competence, there may only be a name or two to check out. spending time learning an industry in-depth is actually detrimental to returns. You become over-confident about your ability to predict the future and essentially force yourself to buy things because of all the time you've invested. In reality there are only two kinds of businesses - Return on Capital Businesses, and Margin businesses. Figure out how to look at those and you can look at anything. At the end of the day research itself is a commodity. No one consistently gets paid for knowing businesses better then someone else. If you are a basket type investor who is going to buy 50 names that you turn up with a screen I would agree with you. If you choose a more concentrated approach you need to understand the relevant metrics within a company's given industry. Maybe I implied too much by saying "in depth knowledge" as I was not talking about anything remotely close to becoming a consultant of some sort. I am referring to having the knowledge of an industry to at least understand what is important. You may screen for two energy companies with low P/B ratios. Which one is better? Does one have more implied reserves than the other? At what stage are they allowed to count oil in the ground into their BV? How much does an efficient producer pay to extract oil per barrel? This is not particularly in depth knowledge. You could teach yourself most of it in a day or two of reading. Don't disagree with that at all. Indeed I actually think most good basket investors do that much DD.
topofeaturellc Posted August 18, 2014 Posted August 18, 2014 The Donald Smith interview is fascinating. Here's the secret of value investing, there are a lot of guys like him running simple investment strategies and generating great returns. The problem is it isn't 'sexy'. Readers want to know about someone who spent 500 hours reading old annual reports and uncovered some secret franchise value of some company. They don't want to know that buying low P/B companies and low EV/EBIT companies and selling them when they rise works, it's boring and has no curb appeal. But that's what is so wonderful about value investing. Its precisely because of this that value investing always works. I bet Donald Smith has a company file on him on average once a year. Could you imagine telling a potential investor that in a meeting? Just as behavioral biases create the opportunity for us as investors, they create huge barriers to building a business. I've sometimes thought analysts are more valuable for marketing than investing.
Packer16 Posted August 18, 2014 Posted August 18, 2014 One question I have always had about basket investors is how they tradeoff cheapness and weighting? I have a tendency to weight cheaper stocks heavier than more expensive stocks and use the Kelly Formula to provide a ceiling for my security weightings. For an enterprising investor, interested in capital growth, I have always been reluctant to reduce the weighted average cheapness of my portfolio for diversification. How do you handle this tradeoff? Packer
writser Posted August 18, 2014 Posted August 18, 2014 Thanks for the DS interview, never heard of the guy before. Good read.
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now