Overview

This chapter tests whether the higher average pay of non-compete-covered jobs gives low-education workers a better career path. In the 2026 Chicago Fed working paper’s retained transition sample, covered jobs pay about 15% more on average, but the wage change at signing is statistically indistinguishable from zero. Four years later, workers entering jobs with a non-compete show significantly slower wage growth than comparable transitioners. The pattern is consistent with weaker outside options and bargaining power, not proof of that mechanism or a claim that every NCA harms every worker. It describes a measured four-year endpoint, not each annual raise or lifetime welfare.

Learning outcomes

For low-education workers in the studied transitions, the roughly 15% average wage gap is descriptive, not a proven NCA premium.
The estimated wage change at signing was statistically indistinguishable from zero, so a clear immediate premium was not found.
Four years after signing, wage growth was significantly slower; this is consistent with weaker outside options, not direct proof of the mechanism.

Keywords

Non-competeLow-educationWage growthSortingPremiumMobility

The Higher-Pay Trap: What Happens to the Low-Education Ladder

The low-education results in the Chicago Fed study answer a deceptively simple question: if non-compete-covered jobs pay more on average, do workers who sign an NCA actually receive a better wage path? This chapter focuses on the difference between a wage level and a wage trajectory. It shows why the descriptive premium is not a compensating differential, what the study finds at the moment of signing and four years later, and how far the evidence supports an outside-option explanation.

The result is a useful warning against reading a contract from its first visible number. A worker can enter a higher-paying job and still face slower subsequent wage growth. That is not a contradiction once sorting and timing are separated. The relevant comparison is not between two unrelated workers at one date; it is between comparable job changers whose NCA status changes in different ways.

A higher starting wage is not necessarily a better bargain

In the retained sample, jobs covered by NCAs pay about 15% more on average for low-education workers than jobs without NCAs. NCA-covered jobs are also a substantial, though minority, share of this group’s employment: roughly 12–13%. At first glance, those facts could support a straightforward story. Perhaps workers surrender some future mobility and are paid for it.

But the average difference is not measured at the moment a worker accepts the restriction, and it is not a comparison of the same worker before and after signing. It combines workers, employers, occupations, industries, and job matches that may differ in ways the NCA does not cause. A higher wage may reflect ability, a more productive employer, a different occupation, or a role with more valuable responsibilities. The NCA may simply be one feature of that job bundle.

This is what economists mean by sorting in this context. People are not assigned randomly to NCA and non-NCA jobs. Workers with particular abilities or preferences may be more likely to take one kind of job, while employers that use NCAs may offer different forms of work. If those pre-existing differences raise pay, the raw wage gap gives the agreement credit for an advantage that belonged to the worker-job match already.

The paper reports that imposing worker fixed effects and controls for occupation or industry makes the wage gap shrink substantially and become insignificant. That does not mean every remaining difference is proven to be sorting. It means the initial cross-sectional gap is a poor causal answer and that accounting for obvious, persistent differences changes the picture. The study therefore turns to within-worker changes around job transitions.

The distinction is easiest to see with two comparisons:

Level comparison: workers in NCA jobs earn more than workers in non-NCA jobs.
Transition comparison: after a worker changes jobs, does the wage path differ when the new job changes NCA status relative to a comparable transition where status remains unchanged?

The first can describe the labor market. The second is designed to say more about what happens when the restriction changes.

Published by DeepMox

FAQ

What does “statistically indistinguishable from zero” mean for the signing effect?

It means the study cannot reliably separate the immediate wage change from zero; it does not prove the true effect is exactly zero.

What happened to low-education wage growth four years after entering an NCA job?

It was significantly slower than for comparable transitioners whose new jobs did not include an NCA. This is a cumulative difference at the stated horizon.

What does the four-year estimate not tell us about a worker’s raises?

It does not show that every annual raise was smaller; it measures a cumulative log-hourly-wage difference at the stated horizon, not a personal forecast.

Previous9 minNon-Compete Transitions: Sorting or Causation?

Next6 minNon-Competes: Two Wage Paths, Uncertain Tenure

Non-Competes: The Low-Education Wage Ladder

The Higher-Pay Trap: What Happens to the Low-Education Ladder

A higher starting wage is not necessarily a better bargain

FAQ