Model portfolios enter real life

At least as far as Intelligent Investor is concerned, life as we know it began in 1998. Back then, as remains the case today, our aim was simple: to help you get great investment results. Along the way we've also tried to educate and enlighten, and perhaps occasionally smile, but only in the context of helping you make more money.

Overall (and not without a few hiccups) we've achieved that objective. Our audited performance report shows average annual returns of our Buy and Speculative Buy recommendations of 14.0% a year since 2001, compared to 10.8% a year for the All Ords (adjusted for franking).

Our model Growth and Income portfolios have also done well, returning 10.9% a year and 13.9% a year since inception in 2001, each comfortably ahead of the All Ords return of 8.8% a year.

Our research and advice attracts a broad church. Many of our members are highly sophisticated investors, using our research as a source of ideas on which they build. Others blend their own research and ours, and some – often those just starting out or time-poor investors – follow our recommendations to the letter, effectively using our service as a high-performing managed fund without the high fees.

For this latter group especially, this presents a few problems. If you followed all of our recommendations you'd soon end up with dozens of holdings. This is one of the reasons why we introduced our model portfolios: to help people apply our advice to real-life situations, putting higher weightings into better ideas, balancing risks, taking profits and cutting losses – all the usual hurly-burly of portfolio management.

There are, however, a number of problems with them. First, our model portfolios don't take account of trading costs. Instead, they assume you can buy and sell at zero cost when of course you can't. As benchmark indices make the same assumption the comparison is still valid, but both cases are far removed from real life.

Secondly, because trades are conducted when the article announcing them is published, they must therefore take place when the market is open. This means we occasionally have to delay important updates.

Thirdly, the pricing of trades lacks transparency. Because no actual trade takes place, we have no way of being certain that the recorded price (that is, the last price available before the publication of the article announcing the trade) would reflect the actual price had the trade been real.

Fourthly, tax implications may have a bearing on your buy and sell decisions. In our model portfolios, they are irrelevant.

And finally, when we sell a stock from the portfolios to make way for something else, without having first downgraded it to Sell, it can be frustrating to those members that own it. Although it's a normal part of portfolio management to substitute good ideas for better ones, trying to interpret these signals can be confusing. In trying not to give the wrong signals to members our model portfolios have expanded to almost 30 holdings. In our view this is too many.

Of the five reasons why our model portfolios don't accurately mirror real life, this last reason is the most important, especially for those members trying to accurately mimic them. Not only will share prices move between when the portfolios announce trades and when the member is able to complete them, having almost 30 stocks makes this a complicated, time-consuming task.

Real money

So, it's time for a little house cleaning of our own. First up, we're going to start running each model portfolio by investing $20,000 of our own company money in each one. This will introduce transparency to their operation (not to mention trading costs) and means that the trading can be separated from the publishing of articles.

In future, when we publish research on a stock we won't conduct trades in its shares until the following business day. This will avoid any suggestion that we might benefit from the price action of members buying the stock.

We consider this to be a very minor handicap, since the portfolios will continue to focus on relatively large, liquid stocks and will carry on with the same long-term, value-based approach that we've always followed. We won't, however, flag the trades in advance, to avoid anyone trying to take advantage of us. When a trade has been completed, we'll post details on our website for members to view.

Since the portfolios will now be investing real money, we'll also start taking a more ruthless approach to portfolio management in an effort to maximise returns (within acceptable levels of risk). This means we'll hold fewer stocks (typically between 15 and 25) and aim to hold higher weightings of our best ideas (perhaps as high as 15% for less risky stocks). The tables within this article explain the details.

Following along

We're also going to make the portfolios available for people to invest in alongside us, through 'separately-managed accounts' (SMAs). Our sister company, InvestSMART, already offers a number of SMAs so it's an easy step to take. If you wish to follow the portfolios closely this will make life a lot easier for you.

With these SMAs, buy and sell decisions for the portfolios will automatically be replicated across all relevant accounts but, unlike a managed fund, your investment won't be pooled with money from other investors. You'll remain the direct, beneficial owner of the assets, providing greater control and transparency, and improved tax-effectiveness.

Last but not least, the platform we're using allows for extremely low costs. Trades within the portfolio will typically be conducted for a fee of 33 cents, plus 0.05%, which works out at just 83 cents for a $1,000 trade. The platform's efficiency also means we can offer the SMAs with a management fee of just 0.60% a year, plus an admin fee of 0.37% (plus GST).

Both portfolios will soon be available for investment and details of how to set up an account will follow in due course.

To get the portfolios into shape, we're conducting a number of transactions to reduce the number of holdings and increase the weightings towards our best ideas. We're also going to sell a couple of low yielders (notably ResMed) from the Income Portfolio to increase its income credentials. Tables 1 and 2 list the trades for the Growth and Income portfolios respectively, while Tables 3 and 4 show the composition of the portfolios as they stand after these trades have been completed.

As a result of these changes, people who want to follow along with our portfolios will be able to do so much more quickly and easily, and at low cost. If, on the other hand, you're happy to run things yourself you'll be able to use our model portfolios as a more useful base because they'll be more transparent and more closely mirror your own trading experiences.

We think these changes are a big improvement but we'd love your feedback, too. Please leave your comments, thoughts and suggestions below.