◄ back to all posts

More Questionable Employment Data

Written by Michael Lebowitz | Mar 19, 2019

On the heels of the February employment report, we wrote an RIA Pro article entitled Quick Take: Unemployment Anomaly, which highlighted inconsistencies in BLS employment data. In particular, our focus was on the fact that despite jobless claims and the unemployment level reaching nearly 50-year lows, the time it takes for unemployed workers to find a new job is currently 10-15 weeks longer than has been typical at prior levels of unemployment. This study was constructed with over 70 years of monthly data. Based solely on the historically strong statistical correlation of the unemployment rate to the duration of unemployment, the current duration of unemployment lines up with what one would expect for an unemployment rate close to 8%, and not the current 3.8%. Not surprisingly the BLS U6 unemployment rate, a more encompassing measure of inflation, currently sits at 7.3%.

In this article, we present additional data that further questions the health of the labor market. Specifically, we focus on the relationship of the Federal Government’s tax revenue and employment. Once again, something doesn’t add up.

Taxes and GDP

From 1948 until 2010, the annual change in federal tax receipts was 74% correlated with GDP, meaning that 74% of the changes in tax receipts is attributable to the change in economic growth. This makes perfect sense given that during stronger economic periods, wages rise as the supply of labor falls and vice versa. The graph below highlights the relationship.

Of particular interest to us is the shaded area representing the last four years. The graph below magnifies a divergence starting in 2016.

As shown, the deviation from a historical perspective is quite odd. Since 2016, the correlation between the unemployment rate and federal tax receipts has been -58%, meaning that economic growth has a statistically moderate measure of negative correlation. Based on the relationship we should expect tax receipts to decline .58% for every 1% increase in economic growth.

The scatter plot below, dating back to 1990, is another way to highlight how the current relationship compares to recent history. Worth considering is the opposite directions of the slopes of the dotted trend lines of the two periods.

Our Take

The recent change in Federal tax law is certainly responsible for some of this irregular behavior over the last year. However, it is important to note that despite relatively strong economic growth and record low unemployment in recent quarters, weaker tax receipt growth has persisted since the financial crisis.

The graph below shows the steady decline of personal tax receipt growth, despite the significant decline in the unemployment rate. Personal tax receipts were not largely affected by the tax legislation and provide firmer evidence that something is amiss in the data.

The graphs shown in this article, along with the preceding Quick Take on the duration of unemployment, suggest the labor market is not as vibrant as we are led to believe. We also know this from numerous indicators of tepid wage growth despite jobless claims sitting at record low levels.

Our recommendation is to continue to follow the economic data as it can have a large effect on the short-term direction of stock prices. Importantly, however, it may be wise to ignore the narratives surrounding such data. Investors would be well-served to consider the entire scope of incoming data to assess its validity.

Although temporarily suspended by the influences of fiscal and monetary policy, the health of the economy over the long-term drives corporate sales and earnings and ultimately stock prices.

2019/03/19
◄ back to all posts
Holistic