Unemployment & Inflation: Has One Been Fired?

To: Cardiff Garcia – Co-host: The Indicator, Planet Money - NPR Stacey Vanek Smith – Co-Host: The Indicator, Planet Money – NPR

During your Dec 13th podcast you raised the issue of whether low Unemployment has become disconnected from Inflation. You indicated that that is what Fed Chairman Powell has speculated on. Now that is a possibility but there are other possibilities, so it’s an open question about what ‘causal’ factor(s) are there between Employment & Inflation or other factors besides employment.

What if the linkage between unemployment and inflation are still connected but the underlying data that connects them is a different version of the data then assumed/thought? The BLS’s Report on Unemployment presents a percentage conditioned on number of people who were ‘currently’ seeking employment. BLS also provides data on the ‘Participation Rate’ of the ‘working’ population. There’s a relationship between Participation and Unemployment then. So, what if you ‘normalized’ the data?

There’s also a base upon which each months’ rate calculated which changes with the population. Every month the US population grows. This means that the composition of the workforce is constantly changing from month to month, year to year, and decade to decade, and so on. The workers enter the workforce, age through it and then exit it at some point.

Given just these two factors, what would the calculated ‘normalized’ unemployment rates look like? Now the simplest thing to do would be to just show you; but then there’s no challenge, no opportunity to test. The chart below shows a ‘normalized’ set of unemployment rates for 10 different years (Jan. data). Note: the years selected included 3 for G.W. Bush, 3 for B. Obama, and 4 for D. Trump; and 2019’s last Jobs Report was used as an estimate for a 2020 approx. of Jan. 2020.

The Test? Assign each year to a President. Now this should be trivial right. And therein lies the question asked earlier.

If Inflation is triggered by Unemployment, which basis of Unemployment are we to assess and use? How do we know which it is more closely related to? Because in order for the relationship to be ‘disconnected’ you may have to know this. If we are going to be plotting our national economy policies based on current data, it may just be important to have a firm understanding of and grip upon the data that we use. Add to this that the relationship may have more than one factor. This is probably hard to believe, but out economy may be slightly more complex than a single variable.

Well, at least the jobs that make up our economy are stable, unvarying types of jobs across the years even if we only look at a 12-year period; and they are basically the same occupations as say 50 years ago. Perhaps, the type of work that employs our workforce has some connection to how the level of unemployment impacts inflation. How many other factors could be important to inflation is difficult to determine, if there’s too little attention paid to studying them. Does the equation that connects Inflation and Unemployment contain a factor for the level of the National Debt, Consumer debt, Corporate debt? What about other conditions representing our economy? There’s a rumor that manufacturing jobs have declined in the US and that service-sector jobs have increased. Whether there are any implications to inflation related to this shift might be something worth looking into. I wonder what Adam Smith might have had to say regarding this aspect of an economy.

As a final point to ponder, the Consumer is now responsible for the economy. Consumers dominate the US’s economy by controlling 70% or more of it. Does this consumer-centric economy impose a suppressing force on prices, hence inflation?

Powell may be right that there’s a disconnect in the old model; but not because it’s broken. Rather, there are likely other influences that can change the economic reality. And, if you are relying upon the wrong method for calculating the data points then you have a problem that was best described as computers gained supremacy in analysis and prediction: “Garbage In, Garbage Out.”

Not to leave you hanging. The following chart shows both the current BLS data and the normalized data for Unemployment rates. Which years belong to each President should be easy to determine now. However, the meaning of the different interpretations that might have to be made from the differences depictions of what has occurred could be more difficult. Does one view predict something about Inflation that the other doesn’t?