Jobless Claims vs Jobs: Charting The Relationship

Tyler Durden's picture

Tomorrow the BLS will announce that last week's initial claims number was revised to over 400K, the first time this important level has been breached, this time in an adverse fashion, in the past 2 months. But why is 400K important, and why do economists and pundits put impact on this particular number? Here is Bank of America with the explanation in the form of a historical matrix, correlating the historical relationship between these time series, highlighting the notable patterns observed in the past several decade, and what it all means for the big picture.

Here is Bank of America "Chart of the day"

A frayed relationship: Since January 1990 there has been a tight relationship – as claims decline, payrolls increase – between jobless claims and private payroll growth. However, the nearby chart, shows that this relationship has shifted recently.

Amd the explanation:

Several months ago we published a report called A Simple Relationship Souring where we discussed a breakdown in the relationship between jobless claims and payroll growth. In particular we noted that the breakeven level on jobless claims – the level of claims consistent with positive payroll growth – has likely risen. New research from the Federal Reserve Bank of St. Louis1 comes to a similar conclusion.

 

The Federal Reserve Bank of St. Louis comes to three key conclusions:

 

1. The researches found that the current threshold – the particular level of claims associated with an improving labor market (faster job growth andfalling unemployment) or a faltering labor market (declining job growth and rising unemployment) – for jobless claims are 400,000. That is consistent with the current rule of thumb estimate many analysts quote.

 

2. The report find that the claims threshold varies significantly over time. The researches note that the varying threshold is a function of changing labor market dynamics driven by changes in potential output, population growth, and labor force participation rates. For example, from 1957 to 1975, the threshold averaged about 318,000. That was similar to the average from 1995 to 2006. However, during the period 1976 to 1994 the threshold level for initial jobless claims was 474,000.

 

3. Lastly, the report found that there is a high negative correlation between monthly employment changes and the difference between the actual level of initial claims and the threshold estimate. Or in other words, when claims rise above the threshold level, employment growth slows or falls and when claims fall below the threshold level employment growth accelerates or turns positive.

 

As we noted in A Simple Relationship Souring the recent threshold level of initial jobless claims has probably risen since the end of the recent recession. See the chart of the day. The bottom line is that while the breakeven level on jobless claims has probably risen. Our advice to investors today is to not fixate over the level of jobless claims, but the general direction. Over, the last 12 months jobless claims have been moving sideways. That suggests that employment growth will not accelerate meaningfully in the near term. Looking farther ahead, our forecast assumes a slowing in economic growth in the second half of this year and that is why we’ve penciled in a slowing in payroll growth.