Analyst ratings, decoded

A sell-side analyst works for a bank or brokerage, covers perhaps fifteen to thirty companies in one industry, and publishes research on them: a rating, a price target, and an earnings model. That research reaches the firm's clients and, in summary form, everyone else.

The vocabulary is not standardised

Every firm uses its own labels, which is why the same opinion appears under half a dozen names. They generally map onto three tiers:

• Positive: Buy, Outperform, Overweight, Add, Accumulate, Sector Outperform.
• Neutral: Hold, Neutral, Market Perform, Equal Weight, Sector Perform.
• Negative: Sell, Underperform, Underweight, Reduce.

"Overweight" and "Underweight" are portfolio language rather than instructions: they mean hold more, or less, of this stock than its weight in the relevant index.

The distribution is lopsided

Sell ratings are rare. Across the market, positive ratings vastly outnumber negative ones, and outright Sells are a small fraction of coverage — a pattern that has persisted for decades and through several rounds of reform.

The reasons are structural rather than sinister. An analyst who publishes a Sell can find management stops returning calls and stops including the firm in meetings, which makes the job harder. Banks also have relationships with the companies their colleagues cover. Regulations adopted in the early 2000s separated research from investment banking and required firms to publish their rating distributions, but the tilt did not disappear.

Price targets are a twelve-month opinion

A price target is where the analyst expects the shares to trade over roughly the next year. It comes out of a valuation model — a multiple applied to forecast earnings, or a discounted cash flow — and it moves whenever the inputs move.

Targets are worth reading for their reasoning rather than their number. The useful question is what assumption changed: did the analyst raise the multiple, or raise the earnings forecast? The second is a genuine change of view about the business. The first is often a response to the market having already moved.

The change matters more than the level

A stock carrying a Buy rating for three years contains no new information. An upgrade from Hold to Buy does, because it says an informed observer changed their mind and clients are hearing about it today.

This is why ratings changes move prices and static ratings generally don't, and why upgrades and downgrades — not rating levels — are what alert services report. The effect is usually largest when the analyst is well regarded in that sector, when the change skips a tier, or when it runs against the consensus on that name.

How to use them without over-trusting them

• Treat a single rating as one informed opinion, not a verdict. Look at where the consensus sits and how wide the spread of targets is.
• Read the reasoning if you can get it. The argument is more valuable than the label.
• Note that analysts revise in herds; several downgrades in a week may reflect one piece of news rather than several independent conclusions.
• Remember the incentives. Coverage is a business, and the rating scale is compressed at the bottom.

This is general educational information, not investment advice, and not a recommendation about any security. Do your own research or speak with a licensed professional before investing.