1 August

Blue Chips as Perceived by Sell-Side Analysts: Safe Bet or Boring Stuff?

Brokers matter – Saying bad things about sell-side financial analysts is an easy game. Their independence, competence and cost are continuously challenged and that’s unlikely to ebb away with the new European directive coming into play next year. Yet analysts’ perception of equities still proves to be one of the most effective factors in the back-testing of investment strategies. Try out distinctive investment factors like risk or value, growth, quality, you name it, and check the past performance of a portfolio that would have been invested accordingly and regularly rebalanced to follow your strategy. This will hardly beat a tail-end brokers’ consensus strategy, whereby you would have bought stocks in the top decile of brokers’ recommendation on a large scale index (and sold those that fell off to the lower decile).

45% Buy reco on average – Having expressed such warm feelings, let’s have a look at the playground, i.e. the 600 largest European companies in the Stoxx 600 index. These are covered by 20 analysts on average, ranging from a minimum of 2 to a maximum of 43. At mid-July 2017, each stock was rated with a “buy” recommendation on average by 45% of the analysts covering it (with “hold” at 39% and “sell” at 16%).

Stoxx 600 – Sell-side consensus target price premium (X axis) and percentage of buy recommendations

Classic Gaussian, except Stoxx 50

The above Sismo screen shows how all Stoxx 600 constituents at 15 July 2017 were distributed:

  • On the x axis from left to right, according to analyst’s consensus target price premium, i.e. the short/mid-term upside/downside from current stock price as expected by analysts in their latest note or update
  • Per color, with a gradient from red to green, according to the percentage of analysts’ “buy” recommendation

This looks like a classic symmetrical Gaussian distribution, with both indicators (space and color) highly correlated. The average target price premium stands at 6.4% and the standard deviation is 16%.

On this screen, we have highlighted with white borders the Stoxx Europe 50 constituents, i.e. the largest blue chips out of the 600 constituents. Those 50 stocks mainly populate the right side of the distribution (only 13 are below average), reflecting higher upside overall for blue chip companies (a median of 8.4% vs. 6.2% for Stoxx 600 companies).

However, only 1 stock out of the 50 constituents belongs to the 10th decile for analysts percentage “buy” recommendation and only 3 out of the 50 belong to the 10th decile for target price premium. This is quite intuitive, positioning blue chips as well-ranked “safe” buys rather than best buys.

Non-linear returns. But there is a huge gap in the returns offered by the top decile stocks in analysts’ recommendation in recent years compared to those in the 6th to 8th deciles. A portfolio invested in the 7th decile of brokers consensus target price premium over the last 5 years and monthly rebalanced with equal weight would have outperformed the Stoxx 600 by 7% annually on average, but that comes with hefty brokerage fees as 90% of the portfolio would have changed each month, leaving investor with no outperformance whatsoever.

Conversely a portfolio invested in the same conditions in the 10th decile of brokers consensus’ target price premium over the last 5 years and monthly rebalanced with equal weight would have outperformed the Stoxx 600 by 18% annually on average with much lower turnover (40% monthly vs 90%). This performance (before brokerage fees) is reflected by the orange line below and compared with that of the Stoxx Europe 50 in dotted line.

Stoxx 600 5-Y back-test based on consensus TP premium in 10th decile – Monthly equal weight rebalancing

So blue chips may well be safer investments, but staying out of the top recommended stocks is likely to eliminate most of the upside that you could get by taking financial analysts at their word.

Another way of saying this: if you stick to blue chips, don’t bother reading brokers notes…