How jobless claims tick up before a recession hits

I am going to start commenting on the weekly jobless claims figure more actively because I like it as a real-time indicator. For me, it is the best real-time data point we have on how the employment picture intersects with consumption demand and GDP because it is released every week.

This past week’s figure was 241,000, down from an unrevised 243,000 the week prior. That brings the 4-week average up to 237,250, just a hair above last week’s 4 decade low.

Why it matters: Most people post the following chart — or some variation of it — to give you a sense of what’s going on.

But this isn’t the chart that matters because economic growth is a first derivative statistic. It’s about the change in the economic situation. And what we want to see is how fast that change is occurring.

Here’s what that picture looks like going back to the 1980-82 double dip recession.

The big takeaway here is that – except in the special circumstances of the double dip in 1982 – claims signalled ahead of time that labor market distress was feeding through into spending and growth. In fact, every recession since claims record keeping began in 1967 was preceded by a rise in jobless claims. We are not seeing that now. Let’s watch these numbers, because if they rise, it is a sign of economic weakness – though not necessarily recession. But that means it is also a bullish marker for long-dated government bonds.

Source: St. Louis Fed