NCAV NNWC Screen Strategy Backtest

February 16, 2010 | Comments (26)

Stock Screen Strategy and Backtest Series

Over the past year, I’ve focused mainly on cheap, ugly, beaten down stocks. My 2009 performance exceeded any expectation and confirmed that stocks trading at near floor levels with the downside well protected outperformed everything in a rising market.

Over the next few weeks, I’ll be going through other strategies and backtesting them to see how it would have worked. You could have the best tools in your belt, but in the end, picking the right stocks is what it’s all about. Also check the updated test of the NCAV NNWC strategy

Graham NCAV & NNWC Screen Strategies

Here, I’ve backtested the basic strategy of screening for NCAV and NNWC and the results are quite surprising. Financial stocks have been filtered out and the Graham NNWC screener will now reflect the results you are about to see.

In this test, I’ve compiled two sets of data. A portfolio that consists of 15 stocks and another more concentrated portfolio of 10 stocks. Both strategies are based on rebalancing the portfolio every 4 weeks.

Screen Descriptions

1. Net Current Asset Value (NCAV)

NCAV is greater than market cap.

NCAV = Current Assets – Total Liabilities

View the best NCAV screener.

2. Net Net Working Capital (NNWC)

NNWC is greater than market cap. (Don’t forget to download the free NNWC valuation spreadsheet)

NNWC = Cash + (0.75 x Accounts receivables) + (0.5 x  Inventory) – Total Liabilities

View the best NNWC screener.

3. NNWC Increasing

NNWC is positive and the latest NNWC has increased compared to the previous quarter. In this screen, NNWC doesn’t have to be less than current market price. Since the requirement is that NNWC is greater than 0, most large caps automatically fail to make the cut due to the large quantity of intangibles, goodwill and total debt.

View the increasing NNWC screen.

4. S&P500

S&P500 market index e.g. S&P500 ETF

Graham NCAV NNWC Screen Performance – 15 Stocks

Assuming you rebalanced your portfolio with the top 15 stocks from the screen every 4 weeks for the past 3 years, $100 would now be worth the following based on each of the strategies.

  • NCAV: $105
  • NNWC: $328
  • NNWC Incr: $283
  • S&P500: $74

I was quite surprised to see how the stark contrast between the NCAV and NNWC strategy.

Graham NCAV NNWC 15 Stocks from 2008-2009

The purpose of looking from 2008-2009 was to see how the strategies performed during a boom, bust and then boom again. 2008 started riding the tails of a market high in 2007 followed by the crash of 08 and uproar we had from March 09.

Turns out NCAV is having a miserable time. Investing $100 on Jan 2008 would have become about $25 by the end of 2009. The graph states that NCAV stocks are of low quality.

On the other hand, a surprising result is that stocks, where the NNWC has increased compared to the prior quarter, not only outperformed incredibly, but also held up very well while regular NNWC stocks were explosive, but at the same time, stomach twisting.

Graham NCAV NNWC Screen Performance – 10 Stocks

Here you can look at the performance if the portfolio was concentrated to the top 10 stocks.

An original $100 would have become

  • NCAV: $103
  • NNWC: $544
  • NNWC Incr: $503
  • S&P500: $74

That’s a gain of over 400% for NNWC stocks!

Graham NCAV NNWC 10 Stocks from 2008-2009

Top 10 Stocks with Increasing NNWC

[table id=11 /]

Top 10 Graham NNWC Stocks

[table id=12 /]

Disclosure

None

About Jae Jun


Jae Jun is the founder of Old School Value. He is on a mission to provide practical and actionable value investing tools, tutorials and educational material to help empower the individual investor. Keep in touch with Jae via any of the methods linked below.

  • tedk81

    i agree with you strongly that NCAV and NNWC are terrific approaches to investing, and what you did here is cool but it doesn’t mean much, the sample size is too small (and we all knew that net-nets had an amazing year last year, i would bet it was one of the best net-net years in the last 30.)

    i would also ask how you took into account commissions and slippage? rebalancing every 4 weeks with stuff like this would be a very bad idea i think because these things are often extremely illiquid and you’re having to sit on the bid all week to get as much as you want or pay a nasty spread. 15 round trips a month aren’t cheap either (especially if you’re using IB and it’s a penny stock. also the tax implications aren’t that hot when you’re paying short-term taxes on everything.

    but there is other evidence for this method of investing, i was talking about this with an investing buddy of mine the other day, i’ll copy/paste from the e-mail

    - the oppenheimer paper 1970-1983, 18% outperformance
    - james montier numbers (1985-2007, worldwide, 18% outperformance)
    - walter schloss’s 50 years of outperformance and the fact that schloss studied with buffett in the 50s and directly under graham (he did more than net-nets obv)
    - grahams claim that over 30 years he averaged 20% a year with the NCAV approach (no official stats but we know graham-newman did well and the people he taught all put up huge numbers for the next 50 years)
    - performance of buffett partnership, which is buffett from like 1958-1972 or something, he did a lot of bottom-feeding with NCAV stocks (berkshire hathaway was a NCAV stock when he bought it!)
    - http://papers.ssrn.com/sol3/papers.cfm?abstract_id=966188 which covers 1980-2005 net-nets in the london stock market finding around 20% outperformance

    it’s unfortunate that none of these studies go back before 1960, but we have ample evidence that it worked before then and it just makes sense. i do something along these lines and rebalance yearly. oh, and finally, i own INOC which came up here… check it out it’s just a part of my basket, not a major holding but cheap any way you measure it.

  • http://www.oldschoolvalue.com Jae Jun

    Thanks Ted. I went as far back as 5 years, could have gone 7, but 2009 obscured the results. Slippage I took at around 0.5% which is probably too low. Your comment about the liquidity is right.

    Regarding the sample, OTC and financial stocks were ignored.

    But what I wanted to point out was that according to the recent results, NCAV doesn’t work very well. NNWC also offered some more stability for the most part which was surprising.

    Thanks for the INOC tip.

  • tedk81

    One other thing too, I’m interested in your definition of NCAV and NNWC? I think I’m missing something here bc you’re saying NCAV includes goodwill and PP&E and things of that nature- how’d you do that? When Graham talked about NCAV he wasn’t talking about any of that stuff, just current assets. I don’t think you can include PP&E in a NCAV screen.

    Instead of NCAV did you maybe mean book value?

  • tedk81

    Also, yeah, slippage is much nastier than you think with stuff like this, I own a basket of stuff like this and that’s by far the most difficult thing about doing this. You’re often looking at 5 or 10% spreads and next to no volume. Rolling it over monthly would kill your results.

    I’d be interested in what the results would be for companies with reasonable avg daily volumes, say > $30k or > $75k, I don’t know where I’d draw the line but upon looking at a few it’d be pretty obvious. Did you do this by hand? If that’s something you’re interested in, e-mail me…

  • http://www.oldschoolvalue.com Jae Jun

    Oops. I was confusing myself with book value rather than NCAV. The formula I used in the screen is correct. My post isn’t.

    I’m going to test out NCAV NNWC with volume greater than 20k. Should be interesting.

  • Jim Allen

    Although much testing is done with random stocks meeting the screen criteria, I don’t think you would want to just buy a random bunch of issues, say, every third one that fell out of the screen. In real life you would be much more selective. Of course, like Chris Browne once said, “To thrive as a value investor, you have to risk being called a dummy from time to time.”

    I found one NCAV stock that showed up on every value screen imaginable, but when I pulled the statements, it had never made a penny, in something like 8 or 9 years! It was cheap, and likely to remain so until it shows it can make a profit, or goes broke, which ever comes first.

    The other comment I have is that there is very likely nothing, no technique, no strategy, that will protect you in a real no-holds-barred bear market like 2007-2009, 1972-74, or like the 1929-32, etc. If you get caught standing when the music stops, you just have to hunker down and hope you aren’t using borrowed money.

  • http://chromainvesting.com Chroma

    Jae,

    I wonder if the difference in results of NCAV and NNWC is that the latter is a form of NCAV with a margin of safety built in. Graham suggested only buying NCAV stocks that had a margin of safety of at least 33% (discounted from their NCAV value). Perhaps the results would be different if the comparison was one of NCAV with a margin of Safety vs. NNWC.

    I love your blog, thanks for keeping the convesation going!
    .-= Chroma´s last blog ..Good Decisions, Bad Outcomes in Investing =-.

  • peekay

    Nice post Jae!
    Two things comes out of the above discussions -

    1. One needs to understand the Business prospects .. something offered cheap than its True value .. can have different implications.

    2. Needs to have a decent trading vol atleast .. so that one can get in & out without problems, if & when the situation arises.

    Would be interesting to see the screens with the vol part added.

    Thanks.

  • http://www.greenbackd.com Greenbackd

    This is superb.

  • http://www.greenbackd.com Greenbackd

    Where does the net net working capital concept come from?
    .-= Greenbackd´s last blog ..Walking the talk: Applying back-tested investment strategies in practice =-.

  • http://www.oldschoolvalue.com Jae Jun

    @ Jim Allen,
    Im sure if you were selective with the net nets, the returns would be phenomenal.

    @ Chroma,
    Good point. I ran some tests based on the closing price being less than 66% of NCAV. I’ll update the article and include the new additional graphs.

    @ peekay,
    Very true. After my latest stint, I’m no longer trading net nets that have 0 volume. Just so hard to sell even if it hits intrinsic value and worse if you made a mistake and want to get out.

    @ Greenbackd,
    NNWC is from Graham himself. Klarman also mentions it often in Margin of Safety. It’s the purest form of NCAV you can get.

  • http://www.greenbackd.com Greenbackd

    That’s interesting. I’ve never heard of it before. I’ve covered your post for my site. It’ll come out tonight.
    .-= Greenbackd´s last blog ..Walking the talk: Applying back-tested investment strategies in practice =-.

  • zehua

    excellent post, Jae! I am wondering what is your selection criteria to reduce 15 stocks into 10?
    From the results, it seems to imply that companies who are cheap and also have good performance will do better than companies that are merely cheap.

  • tedk81

    Ya, the NNWC thing is the graham thing, isn’t it?

    I have to disagree with Jim Allen as well. There are two ways to do something like this, you can use it as a screen or you can just buy a basket of a bunch of them (30 stocks minimum), rebalance every year/2 years/whatever. Some of the stocks in your basket will go to zero. This is ok and doesn’t bother me, it’s the nature of the game. Some will double.

    I find the basket approach extremely appealing, personally. I had a very very good year last year, but what small-cap value guy didn’t crush it? Keeping a decent chunk of my portfolio in a couple “value baskets” is a nice way to play it safe and guard against rash decisions- it’s all automatic.

  • Adam

    Did you ever rerun the test with 20k volume? If so, what were the results? Could you post those? Very cool stuff.

  • http://www.oldschoolvalue.com Jae Jun

    I put up the results in the following post.

  • Mike

    Do we have stock screener spreadsheets we can download and use? Don’t they come with the $65 srock valuation spreadsheet?

    cheers

    Mike

  • http://www.oldschoolvalue.com Jae Jun

    The screener spreadsheet isn’t available for purchase or download I’m afraid.

  • Jim

    I have a question regarding the backtest results. In the chart for NCAV/NNWC from 2/15/2007 to 12/15/2009 it appears as if the NNWC goes up around 400% wereas in the next chart focusing just on 2008/2009, the same NNWC strategy only stays about level, maybe a 10% gain. I know the charts measure different type periods, but if you look at the 2007-2009 chart, at around beginning of 2008, the NNWC and SP500 are about even, so most of the 400% comes after the 1/1/2008. Any ideas why there is this discrepancy?

  • http://www.oldschoolvalue.com Jae Jun

    I’ve never seen a single strategy that will beat the market every single year. Over the long term, yes, but not every year.
    The NNWC is the same. Depending on market conditions, it will outperform or underperform.
    When the market is extremely cheap, I have experienced NNWC to be the best performers. When market multiples become high, fewer NNWC exists, and those that remain usually are horrible companies.

x
Receive Email Tips and Strategies on Achieving Higher Stock Returns

Bonus: FREE Set of 9 Investing Spreadsheets