Secrets Of Trading Earnings: Estimates Matter

Includes: AAPL, AMZN, CRM
by: Paulo Santos

There are a few well known effects that most traders don’t know about, yet are very powerful and can easily add to their stock selection and trading edges. These have to do with the behavior of estimates, and how that behavior influences the stock price. Like mostly every other edge, these don’t work every time, but they work more often than not.

Positive and negative surprises are not discounted right away

This applies both to unforeseen events and to regular earnings surprises. Basically, due to conservatism bias, investors (and people in general) under react to unforeseen events. This means that even though you might see a huge move on some surprise, odds are better than it will continue in the same direction, that that it will reverse.

Seen that news on (NYSE:CRM) not meeting the analyst’s deferred revenue projections? It fell a lot less in reaction to those news, than it ended up falling – of course, the market helped, it could have reversed, etc, but it going further in the same way is more likely than it reversing.

Seen the same behavior on Apple (NASDAQ:AAPL) after missing earnings? Again, it went deeper afterwards. But one should also stress that although odds are favorable, the opposite will also happen – just less frequently.

The earnings consensus has inertia

This is rather mechanical – the consensus is derived from the views of a group of analysts, and not all of them will update their estimates at once when there are fundamental news (this effect will be more pronounced with unscheduled news). We can derive two consequences:

  • If an analyst raises or lowers his estimates on some fundamental ground, it is likely that others will follow – and since every change in estimates or opinion can have an impact, there’s an edge in going the same way as the most recent revisions are going;
  • Since one analyst revision is just one among many analysts, the consensus will partially reflect stale information, so a company with a rising earnings estimate consensus has a higher chance of beating it than a company whose earnings estimates are going down (here, bear in mind that there is a powerful bias in the market towards companies beating whatever the estimates are). Obviously, beating estimates more often than not has a positive effect (indeed, that’s the reason why over time the bias towards a large percentage of companies beating came to be).

This inertia was on display when (NASDAQ:AMZN), whose estimates had coming down hard, ended up missing those estimates, though again we must say that here, too, it’s all about the odds.

The sector benefits from the same effects

It’s never too late to repeat it: the sector in which a company trades is very important. The effects spoken about in this article are just one more reason to believe so. Very often, a good fundamental development for a company in a given sector is also positive for several others in the same sector, so earnings estimates rising in one competitor more often than not increase the odds of earnings estimates rising in another, and the other way around, as well.

This means that there are better odds in buying companies in sectors exhibiting earnings estimates momentum, and selling them in sectors where earnings estimates are going down.

Obviously, one should also consider other factors, this just moves the odds in a trader’s favor, but other factors, such as valuation or cyclicality also move them. Indeed, cyclicality is probably the most dangerous of the effects for whoever uses this kind of earnings/estimates momentum in deciding what and how to trade – this happens because in cyclical industries, the stocks will top out before the earnings momentum does (and to make it worse, they’ll top out at LOW apparent valuations).

In short

It is desirable to buy stocks that show increasing estimates, had a recent positive surprise, had an analyst increase estimates meaningfully for a reason others might follow or had a competitor seeing increased earnings estimates. This might seem obvious, but many times people neglect it.

It is a good idea, also, if you are going to short something (which is particularly dangerous), for that stock to show decreasing earnings estimates, have some kind of recent negative surprise, had some analyst decrease estimates meaningfully for a reason others might follow or had a competitor seeing decreased earnings estimates.

This is just common sense, and it keeps a trader from, for instance, shorting momentum names that are incredibly overvalued when they still haven’t shown any signs that turn the odds in the trader’s favor.

Disclosure: I have no positions in any stocks mentioned, but may initiate a short position in CRM, AMZN over the next 72 hours.