Improving the Morningstar Rating for Stocks

At Morningstar, we're always thinking about how we can improve the quality of our ratings and analyses. To that end, we've implemented a change in our star-rating calculation that we think will result in better recommendations for our readers and clients.

What's Changing, What's Not
What's not changing is how we assign 5 stars. We will still require a larger margin of safety for lower-quality (more uncertain) stocks to reach a 4- or 5-star rating, relative to higher-quality (less uncertain) stocks. So, a company with future cash flows that are hard to forecast accurately--such as a fashion retailer--would need to trade at a substantial discount to our fair value estimate to receive a high rating; a company with future cash flows that are relatively more certain--say, a diversified consumer-products company--would need to trade at a smaller discount to our fair value estimate to receive a high rating.

We're now applying that same thinking to our 1-star ratings as well. Stocks in our higher-risk categories formerly triggered a 1-star rating at a relatively low premium to our fair value estimate, while lower-risk stocks had to trade substantially above our fair value estimate to receive a 1-star rating. In other words, our star rating bands were not symmetrical.

But uncertainty works both ways, and we think our process would be improved if the premium needed for a 1-star rating were largely the same as the discount to fair value needed for a 5-star rating. We want our 5-star and 1-star ratings to reflect a more symmetrical "confidence interval" around a company's fair value. After all, we should have as much confidence in a 1-star call as a 5-star call. If a company's future is uncertain enough that we'd want a substantially large discount to recommend its shares for purchase (a 5-star rating), then it follows that we'd also want a substantial premium to fair value to recommend that investors consider selling the shares (a 1-star rating).

So, we're raising the 1-star cutoff (our "consider selling" price) for the riskier stocks that we cover, with the result that we'll have fewer 1- and 2-star stocks after the change, mainly in the "above average" and "speculative" risk groups. We're also widening our 3-star band for these same stocks, so you'll see more 3-star ratings among the riskier companies we cover. The chart below shows exactly where our new rating bands fall.

You should see very little rating changes to our less-risky companies, and very few changes among our 5-star stocks, because our "consider buying" prices are remaining essentially the same.

In practice, the number of 1-star stocks drops from 156 to 89, and the number of 2- and 3-star stocks increases accordingly, from 779 to 986 (as of mid-January). There's virtually no change in the number of 5-star stocks.

Balancing the Confidence Intervals
Think of our rating calculation change this way: The future of any company can follow a number of different paths. In estimating a fair value, our job as analysts is to assess which of those paths have some reasonable likelihood of occurring, assign reasonable probabilities to the various scenarios, and reach an expected fair value for the company's shares.

However, as physicist Niels Bohr famously said, "Making predictions is hard, especially about the future." It's entirely possible that our estimates will be too high--hence the need for a margin of safety. Of course, it's also entirely possible that our estimates will be too low, which is a big reason why we're changing our star rating bands to be more symmetrical, so that our confidence interval is equally wide (or narrow) on both the upside and the downside.

Click here to read about our star rating bands in more detail.