- Used to determine the strength of a company’s financial position, does the Piotroski Score work?
- Like a fine wine the score seems to have gotten better with age.
- I show you a very simple strategy based on Piotroski that would have returned 21% annually since 2000.
The Piotroski F Score is a number between 0-9 that uses nine criteria to determine the strength or weakness of a company’s financial position. Named after its inventor, Joseph Piotroski, the score tries to differentiate quality stocks that are likely to show price appreciation when compared to their lesser quality peers.
Piotroski’s 2002 paper “Value Investing: The Use of Historical Financial Statement Information to Separate Winners from Losers” showed returns of 23% annually between 1976 and 1996 using an investment strategy that created a universe of high book to market stocks and bought the expected winners and shorted the expected losers. The research that I have performed indicates that a simpler long-only strategy would also have been profitable in the last 18 years.
How The Score Works
The Piotroski F Score is very simple to calculate yet slightly time-consuming. The 9 ratios used can be broken down into 3 categories:
1. Return on assets
Score 1 if net income before extraordinary items for the year divided by total assets at the beginning of the year is positive.
2. Cash flow return on assets
Score 1 if operating cash flow divided by total assets at the beginning of the year is positive.
3. Change in return on assets
Score 1 if this year’s return on assets (above) is higher than last year’s return.
4. Quality of earnings
Score 1 if the company’s cash flow return on assets (2) is higher than its return on assets (1).
5. Change in gearing or leverage
Score 1 if this year’s gearing (long-term debt divided by average total assets) is lower than last year’s.
6. Change in working capital (liquidity)
Score 1 if this year’s current ratio (current assets divided by current liabilities) is higher than last year’s.
7. Change in shares in issue
Score 1 if this year’s shares in issue are equal to or less than last year’s.
8. Change in gross margin
Score 1 if this year’s gross margin (gross profit divided by sales) is higher than last year’s.
9. Change in asset turnover
Score 1 if this year’s asset turnover (total sales divided by total assets at the start of the year) is higher than last year’s.
Piotroski’s original research specified that you select your stocks from a universe of the top 20% of companies when you rank them by price to book value. My own research has seen that book value’s predictive power when it comes to price increases has dramatically reduced during this century, and I haven’t been able to improve this specific strategy using it.
How The Score Performs On Its Own
I have simulated returns using a very simple strategy – investing $10,000 and once a month buying the top 10 US shares and depository receipts with a market cap of at least $50m and a Piotroski score of 8 or 9, ranked by the Piotroski F Score itself:
Click image to view the strategy
Profits were fairly evenly distributed over the years, with the strategy only losing money in 2008 and 2015:
Finally, the companies the strategy invests in tend to be valued between $2-100bn which tells me that the strategy will scale well and is less prone to liquidity issues if you needed to get out of the market in a hurry.
Why The Strategy Works
As you can see from the inner workings of the factor the Piotroski F Score is a measure of the quality of a company, and quality is king. Stipulating that we only consider stocks with a score of 8 or 9 reduces our list of candidate stocks to those that are more profitable than last year, that utilize their assets more efficiently and that do all this without increasing their debt loads or share count.
When I analyze strategies I tend to split them into two parts by asking two questions – 1) which factors am I using that make the investment case for why I would want to hold this company?, and 2) why is now a good time to buy the stock? What is surprising about the success of this strategy is that there is no inherent timing element – the vast majority of high-scoring companies are good ones to hold so to an extent it makes little difference when or which ones you buy.
While there are risks involved in trading any strategy, the Piotroski F Score goes a long way toward helping you sleep well at night. Typically we analyze a number of different risks to understand if a strategy will meet our requirements: Two risks stand out as the obvious ones to gauge:
Continuity – will the strategy stop working anytime soon?
Piotroski’s original research covered from 1976-96 and our own from 2000-18, so the strategy has worked for 42 years and is still going strong. Again, this is where the factor surprises us as you typically see a strategy such as this stop working as soon as it is published (such as the Magic Formula strategy which I reviewed here which hasn’t performed at all well since its publication in 2008).
Perhaps in this case the amount of legwork you have to undertake to calculate the score across all US stocks has improved its longevity.
Liquidity – can we get out of the stock when we really need to?
In a downturn you may need to liquidate your positions quickly, which is easier if you invest in stocks with high trading volumes. The backtesting system I use (InvestorsEdge.net) has a built-in liquidity filter to ensure that simulated trades could actually have been made in real life, and as we have already seen we would have been investing in stocks which are large enough to be tradeable in most market conditions.
Similarly, some strategies work well for a few years with smaller amounts invested and then suffer from diminishing returns as the invested pot grows. This strategy performs almost as well with $1m invested as $10k.
If you want a simple method that consistently identifies high-quality stocks you could do a lot worse than use the Piotroski F Score, either on its own or in combination with other factors.
Before I started my research into the Piotroski F Score I fully expected to see diminishing returns over the last 18 years – I tend to use the score as an addition to my strategy factors or as part of my due diligence. Other research I’ve read on the score over the years suggests that it tended to select smaller companies, so I was also pleasantly surprised to see a simulated cash investment of $1m returning 18% a year over the test period.
I’m going to be investing my own cash in this strategy and will report on its progress on a monthly basis from now on.