Leading, coincident and lagging indicators

There is a version of macro that is entertainment. You read the headline, you have a view about the headline, and the view never becomes a decision. It feels like work and it changes nothing.

The version that is useful starts with a filing question: when this number is released, is it telling me about the future, the present, or the past?

That single question sorts almost every economic release you will ever see, and it is the difference between macro that narrows your universe and macro that just gives you opinions.

The three kinds of series

Leading indicators move before the economy does. New orders, building permits, the yield curve, jobless claims, consumer expectations, credit conditions. They are noisy and they are wrong often enough to be humbling, but they are the only ones that give you time.

Coincident indicators move with the economy. Industrial production, non-farm payrolls, retail sales, personal income. They tell you where you are standing. They are far more reliable than leading indicators and far less useful, because by the time they confirm something, the market has usually moved.

Lagging indicators move after. Unemployment rate, corporate profits, inventory-to-sales ratios, the average duration of unemployment. They are the most accurate and the least tradeable. They exist to confirm, not to anticipate.

The trap is treating all three as news. A lagging indicator that surprises is not new information about the future. It is new information about the past, and the market has had months to work it out.

Why the distinction decides whether macro is worth doing

If you only follow coincident and lagging data, you will always be describing a regime that is already priced. You will have well-informed opinions and no edge, and you will conclude that macro does not work.

The value in leading indicators is not that they are accurate. Individually they are not. It is that they are early enough to be actionable, and that a handful of them moving together is a different signal from one of them moving alone.

That is the practical skill: not predicting the print, but noticing when several early series start pointing the same way, and knowing what that implies for which sectors should lead.

The one that gets misread most often

Jobless claims are a leading indicator. The unemployment rate is a lagging one. They are both about employment and they belong in completely different columns.

Claims are a weekly, high-frequency read on companies deciding to let people go. The unemployment rate is a monthly survey that reflects decisions made months earlier and is affected by whether people are still looking for work at all.

If you treat them as two versions of the same thing, you will be surprised when they disagree, and they disagree at exactly the turning points that matter.

What to actually do with this

Build the list once, in three columns, for the handful of series you will genuinely follow. Five leading, three coincident, two lagging is plenty. A longer list gets abandoned.

Then, when a number lands, do not start with whether it was good or bad. Start with which column it is in. A weak lagging print in a strengthening leading picture is a very different situation from a weak leading print in a strong coincident one, and only one of those is a reason to change what you hold.

The output of macro should be a decision about which sectors deserve your research time over the next few months. If a release does not move that, it was interesting rather than useful, and interesting is the thing that quietly eats the most hours.

Where this sits

This is the first of the four pillars, and it comes first for a reason: it narrows the universe before you start looking at individual companies. Without it you are picking from thousands of listed names with no filter and calling the result a watchlist.

What the macro pillar covers inside 4 Pillar Trading is set out on the four pillars page, including how it hands off to fundamentals.