Sorting terminal stocks from turnarounds
Posted on September 17, 2008 by Richard Beddard
Filed Under Editor's choice, Investing |
How can you tell the difference between a company getting into trouble, and one getting out of it? It’s an important question for investors who try to buy shares on the cheap, when the price is low relative to the value of the company’s assets or earning power.
Typically, these companies are financially distressed. Profitability may be falling or negative and they may have to raise money just to keep going. They might not sound like good investments, but turnarounds can earn investors extreme returns when previously investors had written them off.
Buy while business is still deteriorating, though, and at best you’ll have to wait before you make any money. At worst the company goes bust, and you loose everything.
Why not reduce the waiting, and the risk? I’ve been thinking about this question more actively since I lost money on SCS Upholstery, a turnaround that didn’t. Checking a company’s level of debt didn’t save me from SCS’ fall. It had none. Perhaps a more complete description of a company’s financial health would be a better test.
The F_Score, invented by an academic, Joseph Piotroski*1, distinguishes between financially weak and financially strong companies. By selecting only the strongest companies from a pool of cheap stocks as measured by the price to book ratio, Piotroski increased average annual returns by 7.5%. Companies with low F_Scores are also five times more likely to go bust than companies with high scores.
Intuitively, I like the F_Score because it combines common sense measures of financial health; profitability, debt, and productivity. The information needed to calculate it is all in a company’s financial statements, and the calculation involves basic addition.
Most of all, I like it because it promises to resolve one of the psychological hurdles facing value investors and pundits alike. Although value investing produces market-beating returns, Piotroski showed that they depend on the strong performance of a minority of stocks. In other words, typical value investors are wrong more often than they are right, but when they are right the returns more than compensate them for their many wrongs. By picking companies with high F_Scores, value investors can be right about half the time!
As an investor it’s tough being wrong most of the time, as a pundit it’s even harder, which, explains Piotroski, is one reason analysts prefer to recommend expensive momentum stocks to value stocks.
It also explains why I’m checking the F_Scores of the stocks with the lowest Naked PEs*2. I’m hoping to identify companies that are already turning themselves around, and thereby improve my ratio of winners to losers.
Calculating the F-Score
Each variable scores one or zero to give a total score out of nine. Stocks that score two or less are the weakest, and best avoided. Stocks that score eight are nine are financially strong and consequently they’re:
- More likely to recover, and…
- Go on to make the highest returns.
Which is another way of saying they’re low risk, and high reward.
In the descriptions of the variables that follow, I’ve attempted to simplify and change some of the terminology Piotroski used to British equivalents and included his terms in brackets afterwards. Piotroski chose these variables because they’re stalwarts of financial analysis, not because they’re an optimal blend. Potentially the F_Score could be improved by tinkering - at your own peril!
Profitability
A recovering company should be profitable, profitability should be increasing, and cash profit should cover accounting profit.
(1) Return on assets
Shareholders’ profit (net income) before extraordinary items divided by total assets at the beginning of the year.
Score 1 if positive, 0 if negative
(2) Cash flow return on assets
Net cash flow from operating activities (operating cash flow) divided by total assets at the beginning of the year.
Score 1 if positive, 0 if negative
(3) Change in return on assets
Compare this year’s return on assets (1) to last year’s return on assets.
Score 1 if it’s higher, 0 if it’s lower
(4) Quality of earnings (accrual)
Compare Cash flow return on assets (2) to return on assets (1)
Score 1 if CFROA>ROA, 0 if CFROA<ROA
Funding
A recovering company should be able to fund itself without borrowing more, or selling more shares.
(5) Change in gearing (leverage)
Compare this year’s gearing (long-term debt divided by average total assets) to last year’s gearing.
Score 1 if gearing is lower, 0 if it’s higher.
(6) Change in working capital (liquidity)
Compare this year’s current ratio (current assets divided by current liabilities) to last year’s current ratio.
Score 1 if this year’s current ratio is higher, 0 if it’s lower
(7) Change in shares in issue
Compare the number of shares in issue this year, to the number in issue last year.
Score 1 if there are the same number of shares in issue this year, or fewer. Score 0 if there are more shares in issue.
Efficiency
A recovering company should be reducing costs, increasing prices or both, and selling more (relative to the assets it uses).
(8) Change in gross margin
Compare this year’s gross margin (gross profit divided by sales) to last year’s.
Score 1 if this year’s gross margin is higher, 0 if it’s lower
(9) Change in asset turnover
Compare this year’s asset turnover (total sales divided by total assets at the beginning of the year) to last year’s asset turnover ratio.
Score 1 if this year’s asset turnover ratio is higher, 0 if it’s lower
Does the F_Score still work?
Piotroski tested the F_Score on US stocks between 1976 and 1996. In its 2007 newsletter (PDF), Validea – a commercial stock-screening company, reported annual returns of 39.9%, 9%, 17.9% and -4.9% for its implementation of the F_Score, demonstrating once again that last year was a bad one for value investors.
Why does it work?
Piotroski found that the smaller the company the better the returns. He also found that the majority of companies not covered by analysts in the previous year did better than the minority that were.
Because there is so little information about small, thinly traded companies, investors aren’t expecting them to recover. That may explain why share prices fall so low, and bounce back so strongly when the companies do.
Footnotes:
- See: Value Investing: The Use of Historical Financial Statement Information to Separate Winners from Losers
- See Identifying Wagon’s smoking carburettor and The six cheapest stocks in August.
Comments
28 Responses to “Sorting terminal stocks from turnarounds”
Leave a Reply
Hi Richard,
This is a fascinating article with loads of useful background on financial terminology. Many thanks.
Once you publish the F-scores for the companies with the lowest naked PE ratio’s then perhaps I will take a punt on the best ones (assuming they reach 8 or 9).
This would be my first foray into direct stock investment since my disastrous Northern Rock effort
Hi Robin, thanks for your encouragement.
I suspect there might not be many eights or nines - particularly in the current market. By trawling the bottom 20% of the US market (by price:book) Piotroski had a much bigger pool to draw from.
I’m working on the assumption that at some point in the future most of these low PE companies will recover and then they should show positive fundamentals. So it’s a bit of a waiting game.
I’m not sure it would be necessary to stick rigidly to companies that score 8 or 9. For example, if the company had issued a small percentage of shares (say to satisfy options exercised by staff) it would score zero for that criteria. That’s a bit harsh. Similarly if the current ratio fell, but was still well above one it doesn’t necessarily indicate the company is in trouble (in fact Piotroski mentions this).
One of the things I like about the F_Score is it is a good entry point for investigation of the company’s financial statements. So some interpretation is possible.
Hi Richard,
Thanks for the post:
Is there a Piotroski Score Generator tool out there anywhere that allows to to simply enter a ticker symbol and see what the score is?
Thanks—Tony
Hi Tony. I don’t think there is. If you Google Piotroski you’ll find some US sites have Piotroski screens. I don’t think they faithfully use the same criteria Piotorski did, though, but get as close as possible using the data they have available.
The Investors Chronicle has a stock screening newsletter and which, I believe, has a Piotroski screen. It’s a subscription service though, and I haven’t seen it.
[...] by cutting costs and embracing multimedia but judging by the figures, it could be a ‘buy’. Its F_Score, a measure of financial strength is a magnificent eight (out of nine), and its long-term [...]
[...] by cutting costs and embracing multimedia but judging by the figures, it could be a ‘buy’. Its F_Score, a measure of financial strength is a magnificent eight (out of nine), and its long-term [...]
[...] I may have part of the answer. [...]
Very interested in what you have to say about the F score. At the risk of being pedantic, how do you define ‘operating cash flow’? Is it ‘ebitda’ or something more comprehensive like ‘ebitda’ plus or minus the change in working capital less capital expenditure less tax. Your clarification would be much appreciated.
best wishes, John.
Hi John, thanks for your question. I take the figure from the cash flow statement (usually Net cash flow from operating activities i.e. after interest, tax, depreciation, and changes in working capital but before capital expenditure).
[...] Piotroski taught me the importance of looking at annual reports when evaluating smaller, unfashionable companies. It’s easy to shuffle glibly through doorstoppers from FTSE companies, indeed if you committed yourself to reading them thoroughly you’d probably retire before you’d read enough to invest all your money. [...]
[...] (an earnings yield of 20%) it’s certainly cheap, and although profits were down last year, it’s F_Score of six suggests financial [...]
[...] has an F_Score of three, which puts it right in the danger-zone. Tags: EZJ, ITL, RNSM Category: [...]
[...] or projections, it means checking it’s healthy. Piotroski measured nine variables in his F_Score, but I tend to rule out companies with large amounts of debt and companies with apparently doomed [...]
[...] Is that good enough to crown it pick of the week? Dunno. I have four more annual reports to look at tomorrow. It would appeal to Piotroski, though. The shares are cheaper than the value of EDP’s assets, net of liabilities and it’s a financial ox, with an F_Score of 7. [...]
[...] Energy scores eight out of nine for financial strength according to Piotroski’s F_score, and it fails to score 9 by a whisker. It could be the ultimate safety-first investment offering a [...]
[...] its latest annual report1 reveals declining profitability and cash flows, and rising debt. It’s F_Score is four out of nine, which means, in terms of financial strength it’s close, but not in the death [...]
[...] use Piotroski’s F_Score, which I think captures elements of business risk and financial risk. Tags: AGS, BA., SGC [...]
[...] If it weren’t for a long list of exceptional costs excluded from the headline figure in 2008, TIG would have made a loss this year, as it did in 2001 and although it had more cash than debt at the end of the year, taking into account its profitability ratios and working capital it’s financially quite weak. TIG scores four out of nine on Piotroski’s F_Score. [...]
[...] favour the long-term price earnings ratio as a measure of value, and the F_Score as a measure of financial strength, but the problem with these measures is the data is not freely [...]
[...] value stocks and bad value stocks. That was Piotroski’s project, and it culminated in the F_Score. Category: [...]
[...] a low valuation are prerequisites for investment. Going into recession Holders looks strong. Its F_Score, a test of financial strength, is seven out of nine. In November, shareholders owned 73% of the [...]
[...] Whichever way you look at it, Molins’ (MLIN) price earnings ratio is very low. The shares cost just one times last year’s earnings, or nearly 1.5 times the average of its last ten years of earnings. The share price is just one fifth of its book value, the accounting value of its assets minus its liabilities, and it’s almost a fabled Benjamin Graham net-net. [...]
[...] (DWSN) oppose a major shareholder’s attempt to replace its chairman, Mike Hartley. With its F_Score of eight, the cashmere business is a potential turnaround situation. Category: Companies, [...]
[...] Financial strength. Preferably they own more than twice what they owe, and they’re showing signs of getting stronger. [...]
[...] financial strength report. He uses a proprietary service called the H-Score. Ditto, I calculate the F_Score, and also check for large pension funds, a nasty not picked up by the F_Score. Other investors use [...]
[...] of Morgan Stanley’s measures will be familiar to readers of this blog. It’s Piotroski’s F_Score. The other is Altman’s Z-score. These calculations determine financial health. Generally speaking [...]
[...] balance sheet date of the last annual report, the company was in good financial health. It had an F_Score of eight out of nine, and its ratio of equity to assets was 66%, meaning the value of the [...]
[...] good news is judging by Titon’s F_Score of six out of nine, the business appears to be stabilising. The bad news is of course, that it’s [...]