Thrifty 30 progress report
Posted on July 1, 2009 by Richard Beddard
Filed Under Investing, Markets |
In practice:
Good companies at cheap prices
Occasionally on Twitter investors, as opposed to people hawking porn sites, show interest in my tweets. So, when henrio83 asked ‘What’s happened to your thrifty thirty?’ I thought I should explain.
First, a quick recap.
Thrifty 30 is the name I gave a method of discovering good companies at cheap prices. It’s based on the ideas of Benjamin Graham, the long-dead father of value investing, who recognised the main risks facing investors are paying too much for shares and investing in financially weak companies and declining businesses.
Towards the end of his life he proposed a system that any investor could follow, which involved buying shares costing less than seven to ten times their latest profits, measured by earnings per share, in companies that own more than twice what they owe.
By avoiding the riskiest companies Graham figured he’d do well.
The correct PE ratio depended on bond yields, and investors could calculate a company’s financial strength by comparing its shareholders’ equity to its total assets. If equity was at least half of the company’s assets (what it owns) then its liabilities (what it owes) must be less than half.
As now, newspapers reported PE ratios and investors could easily find a company’s equity and assets in its annual report.
The system is thrifty because the companies are cheap, and not overly indebted and thirty because, for safety’s sake, Benjamin Graham advocated holding twenty, or preferably thirty, companies in a portfolio.
If that sounds like gobbledegook you can read my fuller explanation.
Last March I constructed two Thrifty 30 portfolios, one for Interactive Investor and one for Money Observer (PDF). I haven’t tracked their performance, although considering the stockmarket recovery I’d probably look like a genius if I did. That conclusion would be misleading for all sorts of reasons, but even it were true three months is not enough to judge a system promising 15% returns over five years or more.
I’ve been running a slightly longer but less rigorous test in the Share Sleuth column, also published in Money Observer. All but one of the companies I’ve featured in its first year have done OK, and three, EDI, Carluccio’s and ITE met or exceeded Graham’s 50% profit target very quickly. The only failure, so far, is Optare which, coincidentally is the only one that didn’t meet my Thrifty Thirty criteria. However, it’s also too early to judge Share Sleuth and the title of the latest column describing the first year of Share Sleuth ‘Triumph over adversity’ sets me up nicely for a slice of humble pie next year.
Since March I’ve been working out how to improve on Graham. That’s sacrilege of course. Graham is the pre-eminent value investor who invented the discipline, wrote the textbook, and inspired greats like Warren Buffett.
These days, with more data available to the private investor, it’s possible to substitute the long-term PE ratio, an invention of Graham’s by the way, for the one-year historical PE ratio and Piotroski’s F_Score for Graham’s measure of financial strength. Both measures have more potential, I think, judging from the academic research and are a good basis for reducing the market to a manageable number of Thrifty 30 candidates.
Most of the companies I’ve reviewed on the blog since March have met these criteria, and many of them, as long as they are still cheap enough, will feature in a new Thrifty 30 portfolio that I will compile and maintain when I get back from holiday in August.
And that, really, is the mission of this blog:
- To understand the stockmarket and how to profit from it, as I do in the posts categorised ‘markets’ and ‘investing’.
- To identify good companies at cheap prices, as I do in the ‘companies’ category.
All that’s missing is to prove it works, by:
- Running a model portfolio that earns market-beating positive real returns over every five year period of this experiment
Which is coming soon. Meanwhile, here are some companies that have financial year-ends in the last six months and currently meet my Thrifty 30 criteria:
Ggearing is Grahams original measure of financial strength, which I still keep an eye on, and EPS count is the number of years of earnings data used in the long-term PE calculation.
I derive the figures from data from Sharelockholmes.com and Sharescope. Readers interested in recreating Graham’s original screen can do it using the Sharelockholmes website (it charges a small monthly fee).
Here’s the long-term PE ratio for the whole UK market:
As you can see it’s stuck on 11, which looks cheap, but Jim, a reader from the United States sent me an email yesterday warning of complacency:
The following post links to an Excel spreadsheet that has a 10-year PE for the S&P 500 going back over a century… The data there shows that the average 10 year PE for the market over more than a century is just over 16. But it also shows that there are very long periods of time when the 10 year PE was quite depressed. For 10 years from 1915 to 1924 the average 10-year PE was just below 8 or less than half the long-term average. That’s quite a long time for stocks to be so dramatically cheap. For another 10 years from 1975 to 1984 the 10-year PE averaged less than 9.5. And of course there is the aftermath of the Great Depression: for 20 years from 1931 to 1950 the average 10-year PE was under 12.8. That’s forty years overall where the long-term PE was at *least* 20% below the century average, and sometimes over 50% below it. The low PE during the 70s and 80s is really not that long ago, and it was quite far below the average. To know that stocks may trade below 10 times their 10-year average earnings for a decade makes one hesitant to buy the market as a whole based on PE alone.
The Thrifty 30 isn’t an investment in the whole market, it’s an investment in the undervalued part of it. But you have been warned!
In theory:
The economy according to Clarkson
Add Jeremy Clarkson to my list of economists who know what they were talking about.
Nick Train, is the UK’s Warren Buffett, says the Investors Chronicle, but it doesn’t mention whether he measures up, in terms of performance.
George Soros: “When I see a bubble, the first thing I do is buy”
Comments
4 Responses to “Thrifty 30 progress report”
Leave a Reply
Latest iBall episodes- An error has occurred; the feed is probably down. Try again later.
Hi Richard,
Great work/blog which I have been following in the past couple of months.
I am performing the F_Score on VOLEX the first PLC company in your thrifty thirty list. I could not get an F_Score of 8, using the company’s 2009 annual report.
I got the following results
ROA - 1
Cash Flow ROA - 1
Change in ROA - 1
Quality of Earnings - 1
Change in Gearing - 0
Change in working capital - 0
Change in shares in issue - 0
Change in gross margin - 1
Change in asset turnover - 1
FSCORE = 6
Hi Aris, thanks for your comment. The F_Scores come from Sharelockholmes.com and sometimes I find discrepancies (usually just of one) too. If I get a ‘mo I’ll check the Volex calculation and see if I agree with yours, however I would need to use the 2009 and 2008 annual reports to get the beginning and end return on assets numbers for this year and last.
I *think* Sharelock Holmes uses the preliminary results for its calculations and I’m not sure if how it calculates the F_Score so that may explain the discrepancy. When I profile a company, I always check the F_Scores myself, not just because they may be inaccurate but also because once you have the number it’s open to interpretation. A small increase in share capital to fund, say, share options rather than an acquisition, probably isn’t a good reason to mark a company down for example.
When I profile a company, the F_Scores I quote are ones that I’ve calculated. However, I find Sharelockholmes very useful for finding promising companies to investigate. It sounds like you’re doing the right thing by re-calculating the F_Score too.
Hi Aris. I’ve taken a look at Volex’s F_Score. I make it 7. You scored the number of shares in issue 0, but according to the 2009 annual report there were the same number of shares in issue on the balance sheet date in 2009 as in 2008, which would score it 1. See note 26.
Mind you I have doubts about Volex’s PE. If you take out the massive earnings it made in 2000 and 2001 during the telecoms boom, it’s 8 year PE is negative and the company more often than not has made a loss. As a supplier of wire cable to telecoms companies it’s clearly been through a massive boom and bust and I’m not sure how useful its past earnings record is in divining the likely earning power of the company.
Danielle had this to say…
Risky - but could be worth it. Not sure if “cheapest” necessarily correlates with “best,” though….
…