How do you validate if your idea works?
How do you know if not for sure, then at least historically, that what you do works?
The answer is simple.
You see what happens if you performed this strategy over the past ten years.
You test it. And you test it using tools the professionals have access to and consider their golden goose.
We used a Bloomberg terminal to test and determine if our chosen strategy would have performed well over the past ten years.
What does ‘perform well’ mean?
I would call outperforming the major indexes around the world by a large margin “performing well”.
There are two reasons why.
First: Most investors don’t beat the index. That includes hedge funds and professional money managers.
Second: investing in the index is typically low cost, automated, and fuss-free.
So if a) you can’t beat the index, and b) you need to spend more time only to lose to it – why bother trying?
If this were me, I’d probably buy the index and just spend my time doing something I liked.
Here is how the various indexes did over the past ten years, compared to the strategy employed.
| Indexes | Returns |
| Hong Kong, Hang Seng Index: | 118.1% (dividends not included) |
| USA, S&P Index: | 159.995%(dividends included) |
| USA, Dow Jones Index | 152.132%(dividends included) |
| Our Growth Strategy | 1566.10% (dividends if any, included) |
As is obvious, the strategy outperformed all major indexes by approximately 1300% or more.
That’s what we’re after.
Let’s talk about the “how“.
Methods
All stocks we selected must pass the following criteria:
- Must also have a price to book ratio of less than one, latest financial year (undervalued)
- Must also have an Altman Z Score of 3 or greater (strong financials)
- Must also have a Return on Common Equity of 10% or Greater, latest filing (profitable by a decent margin)
- Must be listed in Singapore Exchange since we enjoy no taxes on stock market gains
- Not based in China to reduce the risks of fraud
Stocks which passed the 5 criteria were bought each year.
Stocks which failed the criteria were sold each year.
The portfolio was arranged such that each stock had equal weight. (If I had $10,000 and 10 stocks, each stock would have at maximum, $1000. This is to diversify against risks.)
Stock Selection Process Explained
1) Price-to-Book (P/B Ratio) less than One
Price is dictated by the market. Book value is determined by whatever the company owns.
If the company is worth $10, but people are only willing to pay $5 for it, then the company has a price to book ratio of 0.5 Where Price = $5, and Book = $10.
This then allows us to find undervalued stock more easily since we can pay $5 for $10 worth of a company’s value.
As investors, we want to buy $1 for $0.50. This is a common theme among all investors. As much as possible, we buy cheap, and we never, ever, overpay for something.
In fact, the cheaper, the better.
That way we get to own what the company owns cheaply, and we get to own the business it has for free.
The aim should always be to buy good businesses at great prices.
Talking about good businesses…
2) Altman Z Score Above Three (Is the business financially sound and unlikely to go bankrupt?)
When we buy companies, we always want to be sure to cover our asses.
We don’t want to buy into a company that seems to be temporarily making a good profit only to have it die on us out of the blue thanks to a lack of cash, or a lack of long term sustainable profit margins.
That’s what the Altman Z was designed to do: check if a company was at risk of going bankrupt.
The other half of the reason we decided to use the Altman Z Score is that its academically and statistically supported.
What do I mean?
“In its initial test, the Altman Z-Score was found to be 72% accurate in predicting bankruptcy two years before the event, with a Type II error (false negatives) of 6% (Altman, 1968).
In a series of subsequent tests covering three periods over the next 31 years (up until 1999), the model was found to be approximately 80%–90% accurate in predicting bankruptcy one year before the event, with a Type II error (classifying the firm as bankrupt when it does not go bankrupt) of approximately 15%–20% (Altman, 2000).”
Ignore the score’s predictive power at your own peril.
Z Score Summarised
| Altman Z Score | How to Interpret |
| 3 or Above (can check further to invest) | Unlikely to go bankrupt / Financially Stable |
| 1.8 or Below (don’t bother checking) | Likely to go bankrupt |
Formula: Z-Score = 1.2A + 1.4B + 3.3C + 0.6D + 1.0E
A = working capital / total assets
B = retained earnings / total assets
C = earnings before interest and tax / total assets
D = market value of equity / total liabilities
E = sales / total assets
3) Return on Common Equity/ ROCE (Is the business actually profitable/good?)
How do we tell if a company is actually making money, and in a way that matters?
That’s where we turn to Return on Common Equity.
The Return on Common Equity (ROCE) ratio refers to the return that common equity investors receive on their investment.
ROCE is different from Return on Equity (ROE) in that it isolates the return that the company sees on its common equity, rather than measure the total returns that the company generated on all of its equity.
Capital received from investors as preferred equity is excluded from this calculation, thus making the ratio more representative of common equity investor returns.
This provides a cleaner measure of profitability that is meaningful to a shareholder.
So why an ROCE of 10% and above?
I simply wanted to only be looking at companies with ROCE of 10% and above to ensure we look at good companies only.
Some additional points to note:
- Return on Capital Employed can still be inflated on the back of prevailing losses, large debts, or even perpetual securities disguised as equity.
- Despite these drawbacks, the strategy still returned substantially higher returns versus the index. We should test if removing these false inflations of ROCE can improve investment performance.
The full list of stocks that currently pass the criteria are listed here.
Stocks That Passed The Criteria As Of Time of Testing
| Ticker | Short Name |
| AZEUS | AZEUS SYSTEMS |
| FRKN | FRENCKEN GROUP L |
| BLT | BAN LEONG TECH |
| IPC | IPC CORP LTD |
| FUJI | FUJI OFFSET PLAT |
| HG | HOUR GLASS LTD |
| AVARGA | AVARGA LTD |
| SERL | SERIAL SYSTEM |
| INNOT | INNOTEK LTD |
| SPE | SPINDEX IND |
Note that this is not an incitement to invest all the stocks at once. If you notice any glaring issues with the stocks listed, feel free to tell me so.
Also note that the strategy has not been tested in a bear market, which we seem to be heading into. My colleague Khin Wai will be updating all of you shortly on that as soon as he can, and we’ll get to
Criticisms and Closing Thoughts
Point One: Will it always work?
Anyone who tells you “yes” is a liar and you probably should stay quite far away from them.
As a whole, this strategy has not yet been stretched out over 30 years or more. I have one hour with the terminal a day at maximum so I haven’t yet had the chance to test everything. But I look forward to the day we get one in the office and I can go about tinkering to my heart’s content.
So no, there’s no guarantee it will work forever. I will simply say that the old adage of buying cheap, quality businesses can never go too wrong.
I also have not been able to test if this 3 specific criteria (pb ratio below 1, ROCE of 10% and above, and an Altman-Z score of 3 and above) only applies to Singapore, or if it also applies in other parts of the world.
If the strategy both predicts returns far above the norm internationally and across all time-frames, we will have found a new golden rule to implement and follow. As of now, I can only state that it has worked for the past 10 years.
No one can say it will work forever. This goes double for value investing, arguably the most popular approach to investing in the world over – and yet also one of the oldest.
The truth is that all investors must be prepared to look back, see what has worked, understand why it has worked, and hope it continues to work, while always staying on guard.
That’s just life.
Point Two: Years of Slight Underperformance
Out of ten years, the portfolio actually underperformed the S&P index in 4 years and it outperformed the index by a very large margin in the other 6 years of testing.
My concern is what happened in those 4 years, and what drove the returns in those 6. When the portfolio outperformed, it did so in leaps and bounds, but it underperformed in very small margins.
This is good thing no doubt. And there are a variety of reasons as to why.
Perhaps the investible set of stocks was too stringent and very few remained for the picking. Perhaps there was some level of unsystematic risk.
Without taking a detailed look at the model (something I can only do with unlimited time), I can only take very large, macro snapshots of how the strategy has worked.
This is not all bad.
I look forward to smoothing out the return on equity from latest financial filings, to 5-year spans to eliminate temporarily, or arbitrarily inflated stock prices.
Once I figure out how to code the Benenish M score into the Bloomberg terminal to further remove aggressive accounting/fraudulent companies, I can rest a bit easier.
I’ll also be looking to add in low debt criteria and experiment with percentiles rather than absolutes. That means I might input “buy only companies in the bottom 20% for price” instead of “price to book ratio of less than 1” to see if that actually improves performance.
Most of my further tinkering will be to remove “impurities” under ROCE, namely to try and get a clearer sense of profits, owner’s earnings, and the price paid to acquire the asset.
I believe doing so will provide us with greater returns over the years as opposed to having years of underperformance, though of course, the underperformance might just be due to outside factors.
Food for thought.
Point Three: Strategy Updates and Applying Foreign Strategies
On Applying Foreign Strategies to Singapore Markets
All too often, new investors or even seasoned investors are drawn to fanciful new strategies that have never tested well.
We must ensure we do not fall into that trap of simply accepting that whatever has worked overseas, can work just as well here.
If a strategy does not bear fruit, we should not simply execute it without a strong basis of belief in its historical performance.
On Updates
I look forward to updating the strategies as such:
- Extend the period of testing to 30 years
- Extend the area of testing to HK, MY, China, US
- Check if individual ratios predict performance above the baseline (baseline being the major indexes around the world) and if so, by how much.
- Check if ratios combined exponentially increase ratio predictive performance, and if so, how much.
- Check if Altman Z model actually predicts bankruptcy well in Singapore for the development of possible short strategies, or minimally, a list of companies to avoid for Singapore Investors
- Test passive income strategies and returns
- Test Buffett and Lynch investment methodologies if possible (since their methods are not entirely quantitative) for local application
Investment strategies cannot simply be trusted to work in the local context. We must always be willing to ensure that a strategy which works overseas works locally before implementing it.
Stocks that are published in my articles will also be generated via the terminal to ensure reliability and transparency as far as possible. I will also provide additional analysis when appropriate.
So make sure to stay tuned for tested strategies and up-to-date stocks to look into.
I look forward to sharing more Bloomberg tested results with all of you in the future.
Stay tuned. Stay safe. And be on the defensive.
PS: The factors and criteria chosen today were loosely based off of the original Profitability and Value factors under Intelligent Investing Immersive with a long track record of performance. Its roots came from the mentor of Buffett himself: Benjamin Graham.
ROCE was meant as a proxy for the percentile of profitability and price to book ratio was meant to be a proxy for conservative net asset valuation. Altman Z score was added in to ensure financial stability and was researched by myself. If you’re curious as to how we actually implement and use the factors to achieve supernormal returns, you can click here to take an introductory class. It’s free of charge.
Is this strategy better CNAV strategy? CNAV focuses on discount in asset prices and margin of safety.