Photo: Paul The Writer / Wikimedia Commons, CC0

vizmaya desk · 2 November 2026

Can a higher minimum wage make workers study less?

Take two workers with the same job: one earns $11 an hour, the other $22. A 2026 economics paper argues the second has less reason to sign up for night classes.

Is that true? And how big could the effect be?

Example from the paper, p. 344. Journal: Review of Economic Analysis, 18(3).

For 30 years, the fight was about jobs.

Raise the minimum wage and pay goes up for some workers. Do others lose their jobs? That argument has run since at least 1994.

Far fewer studies ask a different question: what happens to learning?

A paycheck is half the story. What is the other half?

Source: paper, pp. 343-344.

Black and white 1917 photo of a shop floor with a turnstile entrance, wicker baskets and a check-out counter — Piggly Wiggly, Tennessee, 1917, the first self-service grocery in the US. Photo: Library of Congress via Wikimedia Commons, public domain.

Studying costs you now. It pays you back later.

A worker can stay in the same job. Or take classes for a few years and move to a better one.

Classes mean less pay and fees today. The better job pays more after.

So what does a raise change?

Schematic of the paper's model, Eq. 6 and 7, pp. 346-347.

Every class hour is an hour off the clock.

The higher your hourly pay, the more each hour in class costs you.

So in the model, a raise makes studying more expensive.

That is one force. A second one hits the reward.

Source: paper, p. 346 (less work while studying) and Eq. 16, p. 353.

A blue workplace time clock with a round dial and a slot for punch cards — A workplace time clock. Photo: Jonathan Mauer / Wikimedia Commons, CC BY-SA 4.0.

A raise also shrinks the prize.

In the paper's example, a skilled job pays 38 an hour. When the minimum wage is 5, that is 7.6 times more.

When the minimum wage is 30, it is only 1.27 times more.

Cost up, prize down. So raises can only hurt learning? Not so fast.

Computed: 38 divided by the minimum wage. Example setting, p. 352; idea, p. 353. No currency is given.

Being broke makes tomorrow feel far away.

When money is tight, planning ahead is hard. In the model, a raise eases that, and the future starts to matter.

Patience is scored from 0 to 1. It rises from 0.24 at a minimum wage of 5 to 0.43 at 12.

Two forces push against studying. One pushes for it. Who wins?

Computed from Eq. 10 at the paper's example settings, pp. 348 and 352.

Up first. Then down.

Put the forces together and the model draws a hill. As the minimum wage rises from 5 to about 15.7, more workers choose to study.

Past that, fewer do.

That sounds like a cliff. Is it?

Three points read off Figure 1, p. 352, give or take 0.005. The vertical axis starts at 0.34, as in the paper.

Same three numbers. Zoomed out.

The climb is 0.128. The slide after the peak is 0.014.

9to 1

The paper calls the drop a "poverty cliff" (p. 350). In its own example, it is a gentle slope.

A tidy model. What does real data say?

Computed: 0.473 minus 0.345 = 0.128; 0.473 minus 0.459 = 0.014. Read off Figure 1, p. 352.

Does real data back the hill?

Two 2024 studies looked at community college enrollment after minimum wage raises. They found opposite things. Contested

Canada

More students sign up for community college. Fewer sign up for university.

Alessandrini and Milla, 2024

United States

Fewer students sign up for community college.

Schanzenbach, Turner and Turner, 2024

Nobody has tested the full up-then-down shape. So is the hill real?

Sources: paper, pp. 344 and 352; NBER w31540; CRDCN summary.

This is a what-if machine. Nobody has measured it yet.

Engineers test a small model before they build the plane. This paper tests an idea the same way, with equations.

No new data was tested (p. 355).
The model has no job losses in it (Eq. 3, p. 345).
Change the settings and the peak moves (p. 352).

So how would we find out?

How would we find out?

1Find places with different minimum wages.
2Count who enrolls, who drops out, who moves up.
3Look for the hill.

Until then, what should you take from this?

Tests proposed in the paper, pp. 355-356.

A raise has two stories.

Pay now.Climb later.

Most of us ask one question: did the raise cost jobs? This paper adds a second: did it change who gets to climb?

It does not say to cut the minimum wage. It says to pair raises with help for training (p. 357).

Three questions to carry into any headline.

Source: paper, pp. 343 and 357.

Three questions for any "trap" headline.

Was it measured, or modeled?Here: modeled. No new data (p. 355).
Where does the chart's axis start?Here: 0.34. From zero, the hill is flat-ish.
How big is the drop?Here: 0.014, after a climb of 0.128.

Sources and credits.

Sources and method

The paperAlkhowaiter, M. (2026). The Minimum Wage Poverty Trap: A Theoretical Analysis. Review of Economic Analysis, 18(3), 343-361. doi.org/10.15353/rea.v18i3.6815. Free copy: LSE Research Online. CC BY-NC 4.0.

Studies it citesSchanzenbach, Turner and Turner (2024): NBER w31540. Alessandrini and Milla (2024): CRDCN summary.

ChartsPatience and the skill ratio are computed from Eq. 10 and the paper's example settings (p. 352). The hill uses three points read off Figure 1, give or take 0.005. Figure 1 itself is not reproduced.

Not foundA citation count for the paper. The formula behind Figure 1's curve. Sizes of the two studies' effects were not checked, so none are shown.

PhotosLecture theatre: Paul The Writer, CC0. Piggly Wiggly, 1917: Library of Congress, public domain. Time clock: Jonathan Mauer, CC BY-SA 4.0. Wind tunnel model: Jonathan Cardy, CC BY-SA 3.0. Solano Community College: Kiddo27, CC BY-SA 4.0. All via Wikimedia Commons.

Byvizmaya desk, 2 November 2026. More stories at vizmaya.fyi/stories.