How do we measure "recovery"?

Levels don't set the bar high enough; we need to look at trends

Update since Tuesday: more discussion in the papers about the shifting housing market. An uncertain sentiment continues to fill the market. If anything, these stories show how difficult it is to infer a lot from the data.

My subscriber base is quickly growing these days – thanks to everyone who has recently signed up for the newsletter. Please continue to spread the word: sending this post to a work colleague or friend in the area helps a ton. Thank you in advance for your support!

Also, let me know if there is ever anything I can do for you. If you’re looking for something more specific for you or your company, I’d be happy to help. I write Python scripts to collect and process most of the data I use in these posts, which makes it easy for me to quickly find the answer to a question or document a trend based on publicly available sources.

At any rate, for today’s special topics post (that’s what Friday's are for - Tuesdays are your normal economic update), I wanted to revisit the concept of “recovery”. One of the initial reasons for starting this newsletter was to track the economy’s recovery from the pandemic. At the time, a lot of analysts were looking at how many jobs were lost in an area since the pandemic. For them, the sign of a recovery was whether we had gained all those jobs back.

There are a few problems with that perspective:

·         The labor force continued to grow, even during the pandemic: just as in normal times, high school and college graduates entered the workforce, contributing to the pool of job seekers

·         While some left the work force permanently and retired early, others left or scaled down hours temporarily to take care of kids, and many were ready to go back to work once school closures ended

·         High wage growth during the pandemic in certain sectors meant that it was not an economically sound decision to stay out of the labor force for very long

All these observations suggest that what mattered was not whether we returned to pre-pandemic levels of employment, but whether we returned to pre-pandemic trends in employment.

Basically, I’m working off the assumption a large part of the effects of the pandemic are temporary. I realize this is not the safest assumption to make, and things have changed permanently for many workers. But so many things have happened over the past 2 years that it’s hard to keep track of what has permanently changed and what hasn’t, and so I think the focus on pre-trends is still useful. And, at the end of the day, people still need to work to support themselves and their families.

Here’s how this looks for a few SoCal metro areas. In the charts above and below, I constructed a pre-pandemic trend based on employment growth 2010-February 2020. Then I extracted that trend (dashed line) through to compare to actual employment (green line).

On this metric, the IE has done best – we’ve recovered all jobs lost and are just about 1.2% off pre-pandemic trend. San Diego is not far off – it has fully recovered (though by a small margin) and is 2.8% off trend. Orange (not shown) is a little worse than these two. And at the bottom are Los Angeles (hasn’t recovered and 4.2% off trend) and Bakersfield (not shown; not recovered and 3.6% off trend).  

While this is not a perfectly summative assessment of each region’s labor market, I do think it puts together a story based on many of the factors I identify above. Feel free to leave your thoughts in the comments, and thank you for reading!