Selecting Stocks – Take Two

State of Play: under Motley Fool basic rules there are no companies worth investing in listed on the ASX, so I’ve dug out another book, Tim Hewat’s The Intelligent Investor’s Guide to Share Buying to see what I can glean from it.

In Chapter 8, Solid Signposts for Success, Mr Hewat sets out six guidelines for selecting shares:

1. Diversify by investing in 12 shares. Interestingly the same figure I settled on – clearly great minds think alike.

2. Tend towards Big Caps first and then, if there is no value there, second-line shares. Unfortunately, the book doesn’t define what a second line share is, so I’m not sure how low to go with market capitalization.

3. Give top priority to PE ratios of 10 or below. Mr Hewat has a whole chapter on the Price-Earnings Ratio (current share price divided by earnings per share). He describes a study done by David Dreman (an investment manager) and Professor Michael Berry of the University of Virginia. They made a list of the top 60% of shares by market value listed on the New York Stock Exchange, ranked them by PE ratio and split them into quintiles (ie, shares with the top 20% of PE ratios in one group, next 20% in the next group, etc). They revised the groups each year for 25 years using the same methodology. (I’m guessing this means that at the end of each year they “sold” shares no longer in one group and then “bought” the shares that where now ranked within the group). Over the 25 years, the shares in the lowest quintile returned 17.3% pa, whereas the average for the whole group was 12.3% pa. (There is no comment on whether they took into account taxes and transaction costs).

Another fellow called Qualls did a similar thing with 100 randomly selected industrial companies listed on the (then) Sydney Stock Exchange and also found the lowest group outperformed the rest.

The chapter gives a few more examples, all with the same results. Mr Hewat settled on an absolute benchmark of 10 (note, his book was published in 1994 in the midst of a market boom). This approach saves the time taken to rank all the shares, however, a look over the tables from the weekend edition of the Australian Finance Review shows a fair proportion of them have PE ratios below 10. Also, given we now have the internet and MS Excel, the process of ranking shares by PE ratio and splitting them into quintiles should take all of five minutes, so I think I’ll use that method.

4. Price to Assets Ratio (PAR) should be 1.0 or lower. This is explained along the line of, for example, if the PAR is 0.9, it means you can buy $1 worth of the company or 90c. From this I’m assuming that the ratio is calculated by dividing the market capitalization of the company by the net assets reported on the balance sheet. My concern here is that it assumes the assets in the balance sheet are valued correctly. If I was looking at these figures today (26 January 2009) for a property company, and they were based on 30 June 2008 annual reports, I’d want to discount the properties by, say 10%. But I suppose, as with all of these things, the devil’s in the detail.

5. Satisfy yourself that Dividend Yield meets the demand for an adequate return. I have serious issues with this guideline. Aside from the fact that Mr Hewat provides no real way of determining what an “adequate return” is, the last thing I need is for companies to give me back money in the form of dividends, so I have to pay tax on it and then have to find a place to reinvest it and (unless they have a dividend reinvestment program) pay a fee to a broker to do so. Unfortunately, I think its quite fashionable for Australian companies to pay dividends. I use to the term fashionable quite deliberately, because there is no real reason to pay dividends except that investors in Australia seem to like them so you are an Australian company you are unlikely to attract investors if you don’t pay dividends. We will see when the time comes whether I can find any non-dividend paying shares for profit making companies.

6. Dividends paid without interruption for at least five years. This I see as a lazy way of checking whether the underlying performance of the company has been OK. I would prefer to look at whether the company has actually made a profit for the past five years, because, if a company is making a loss and funding dividend payments through increased borrowings, then it’s a company from which you should run a mile.

Having reviewed all that, I’m going to set a new set of criteria based on a mixture of the Motley Fool rules and the above. So, my new criteria are:

1. Daily Dollar Volume: From $100,000 to $25m;
2. PE Ratio within the lowest 20% of PE ratios for all shares;
3. Price-Assets Ratio of less than or equal to 1.0;
4. Net Profit a positive number and greater than last year;
5. Revenue of $500m or less and greater than last year;
6. Cashflow from operations: a positive number; and
7. Preferably all earnings reinvested (no dividends paid).

So lets see how we go.

1. Daily Dollar Volume: From $100,000 to $25m and
2. Price-Assets Ratio of less than or equal to 1.0

I ran these two tests together, starting with the full list of all securities on the ASX having deleted all the options and other non-share securities. First I ranked the shares by PE ratio and deleted all the shares with a PE ratio of zero or less (ie all companies that had made a loss). I then calculated the daily dollar volume for each share (closing price multiplied by volume traded) and deleted those shares with a DDV below $100,000 and above $25m. There were 201 shares remaining, which I ranked them by PE ratio and kept the bottom 40.

3. Price-Assets Ratio of less than or equal to 1.0

The AFR tables have the Net Tangible Assets (NTA) for each share, which is “the total assets of a company less total liabilities and not including intangible items such as goodwill.” (Refer http://afr.com/home/tables_1.aspx#indus_lnk).

There were nine shares with negative or no NTA so I deleted all of those. Of the 31 shares remaining a further nine had a PA ratio (last sale price divided by NTA) greater than 1.0, so I’m left with 22 potential shares to comprise my portfolio.

4. Net Profit a positive number and greater than last year

I already knocked out the unprofitable companies in the first test, so its just a matter of checking whether the 22 remaining made a greater profit this year than last year. Eight had negative earnings growth, so we are down to fourteen shares.

5. Revenue of $500m or less and greater than last year

Four companies had negative sales growth, so ten remain. Of those, only five had annual revenue less than $500m:

• Macquarie Office Trust
• Mount Gibson Iron
• Nomad Building Solutions
• Macquarie Country Wide and
• Charter Hall Group (one of the final contenders under the Motley Fool criteria)

6. Cashflow from operations: a positive number

The five companies all have positive cashflow from operations.

7. Preferably all earnings reinvested (no dividends paid)

I knew this would be a bridge too far. Only Mount Gibson Iron does not pay dividends, so I will keep the five.

Does this mean I get onto CommSec tomorrow and start splashing my cash around? No. It means I have to start reading annual reports, so I’m going back to “The Motley Fool Investment Guide” to read Chapter 14 “Making Sense of Income Statements” (next post, that is).