Thread by @annastansbury

📢 New WP! Monopsony and Outside Options 📢
(@gregorschub, me, & @Bledi_Taska)

How much do workers' outside job options matter for wages?

This is important to understand the degree of imperfect competition & role of employer concentration in labor markets [1/N]

Paper available at:

SSRN: papers.ssrn.com/sol3/papers.cf…

Direct download: scholar.harvard.edu/files/stansbur…

[2/N]

Takeaway 1: Employer concentration reduces workers' job options. For workers in highly concentrated labor mkts, this matters a lot - esp. for occupations w/ low outward mobility. Within a given occupation, concentration can explain ~21% of wage variation across metro areas [3/N]

Takeaway 2: Having good job options *outside* your occupation also matters. Within a given occupation, differential availability of outside-occupation job options can explain ~13% of the (interquartile) wage variation across US metro areas. [4/N]

So what do we do? We focus on a typical matching model, where the wage is partly determined by the value of worker’s outside option – which we define as the expected value of the job a worker would get if she left her current firm. [5/N]

We then segment these options into job options within workers’ occupation, and job options outside.

We show that employer concentration is a good proxy for job options *within* workers’ occupation: the fewer effective employers, the fewer outside options you have on avg. [6/N]

To identify job options *outside* workers’ occupation, we use occupational mobility patterns to construct “probabilistic” or "revealed" labor markets, using very cool new resume data from @Burning_Glass to measure occupational mobility.

(more on this in a separate thread!) [7/N]

Why use occupational mobility to identify job options, rather than – say – task/skill similarity? We see occupations as a high-dimensional bundle of characteristics: task, skill, qualification requirements, amenities… Transitions are a non-parametric way of capturing this [8/N]

Both these types of outside option should matter: higher employer concentration should reduce wages, and better outside-occupation options increase them.

But also, heterogeneity: for workers in outwardly mobile occupations, employer concentration should matter a lot less [9/N]

We can test this in data covering almost the entire U.S. labor market: 100,000 occupation-city pairs for each year over 2013-2016… but there’s an endogeneity problem. Before we can regress wages on concentration or outside-occupation options, we need good instruments. [10/N]

We instrument for changes in HHI with the Granular IV approach of @xgabaix & @rkoijen: leveraging differential local exposure to the growth of national firms.

(When Starbucks grows faster than Dunkin, the barista HHI in Portland will rise relative to the HHI in Hartford) [11/N]

We instrument for shocks to outside-occupation options by leveraging differential local exposure to national wage trends (a Bartik/shift-share approach).

If wages in an outside option occ grow fast nationwide, you do better if you're in a city where that occ is large. [12/N]

We find significant effects of both employer concentration and outside-occupation options on wages. Concentration matters 3x as much for workers with low outward mobility vs high outward mobility: for low-mobility workers, going from HHI of 0.05 to 0.25 -> ~10% lower wages [13/N]

Finally, interpreting these coefficients through the lens of our Nash bargaining model would suggest that a $1 increase in the value of outside options results in $0.24-0.37 higher wages. [14/N]

Implications?
1. Employer concentration matters a lot for immobile workers in concentrated mkts
2. Labor markets can be modeled as “probabilistic” clusters of related occupations
3. Outside job options matter– but perhaps less than the Nash bargaining model would predict [15/N]

This is still work in progress so we’d love feedback. Thanks very much to the people who’ve given us thoughtful comments already! [16/N]

And thanks so much to @Bledi_Taska & @Burning_Glass for taking the plunge to work w/ two PhD students excited about labor markets and occupational mobility - and for making so much data accessible to researchers more broadly!

[/End]