Women cannery workers, a Works Progress Administration depression-era project. (WPA/Associated Press)

This Labor Day, it looks like workers’ rights advocates finally have some things to celebrate. Last week the National Labor Relations Board issued a major ruling in favor of workers; the Fight for $15 movement won wage increases at several major corporations and across New York state; President Obama’s Labor Department will soon dramatically expand overtime eligibility; and in the past two years alone, minimum-wage increases were enacted in 14 states and several major metropolitan areas.

These and other changes have altered the trajectory of low-wage work in the United States and given many workers hope for still better days.

But will a higher minimum wage actually deliver higher earnings for workers?

Many workers don’t get paid what they earned.

Maybe not. As it turns out, minimum-wage laws don’t guarantee that workers will receive what they’ve earned.

In many low-wage industries, wage theft — when employers don’t pay their employees the full amount they’ve earned and to which they are legally entitled — is pervasive and endemic.

In a recent study, I used Current Population Survey (CPS) data to generate estimates of minimum-wage violations nationwide (methodology here). I found that between 2005 and 2013, about 16 percent of low-wage workers had some of their wages stolen each year.

These were not just paycheck rounding errors or a few lost tips. On average, these workers didn’t receive about 26 percent of what they were owed.

If they had earned the minimum wage in 2013, for example, these workers would have made $12,441 while working on average 33 hours a week. Instead, they were paid only $9,095, well below the poverty line for an individual.

The CPS data doesn’t tell us how employers withheld the wages. But we know that employers often commit wage theft by mandating off-the-clock work, paying their employees a flat rate irrespective of hours worked, making illegal deductions, withholding tips, misclassifying their employees as exempt, or simply refusing to pay for work performed.

Who is most likely to have wages stolen?

Some minimum-wage workers were at higher risk than others. The odds were significantly higher if you were a woman, nonwhite, younger than 30, not a U.S. citizen, had not completed high school, were not a union member, or lived in the South.

Wage theft was most prevalent in certain industries, including private households (25 percent of low-wage workers were underpaid), personal and laundry services, such as nail salons (22 percent), social assistance, such as home-care workers (20 percent), and food services and drinking establishments (19 percent).

Some states protect their workers; others leave them vulnerable.

States also vary widely in their employment laws and enforcement abilities.

Systematically coding the wage and hour laws in all 50 states and the District as of 2013, I found that in states with the toughest penalties and strongest state agencies, workers had significantly lower probabilities of facing wage theft than comparable workers in states with weaker regulatory regimes, holding constant demographic, economic and other factors.

For example, the probability that a low-wage worker would suffer a minimum-wage violation in Washington state (which has one of the strongest regulatory regimes in the country) was about 12 percent, while a comparable worker in Virginia (with some of the weakest employment laws) had an 18 percent probability of experiencing wage theft, with all else being equal.

Some states have tried to strengthen their regulatory regimes. Between 2006 and 2012, 10 states (California, Illinois, Iowa, Maryland, Massachusetts, New Mexico, New York, Ohio, Texas and Washington) passed substantial wage theft laws.

In all but one of those states, Democrats controlled both houses of the legislature and the governorship. Equally important for the bills’ success were the protests and persistent lobbying by broad coalitions that included informal worker groups (“alt-labor”), traditional labor unions, legal advocates and even some business groups.

Not all solutions are created equal. Which ones actually work?

But even when there is unified Democratic state government and a determined workers’ rights coalition pushing for reform, not all policy reforms are equally successful in deterring wage theft.

Among the various laws passed to combat wage theft in the past decade, I found that only one type of reform was associated with significant declines in the probability of wage theft: mandatory treble damages, which eliminate judicial or agency discretion in awarding the worker three times the back wages owed. These laws were enacted in New Mexico, Massachusetts and Ohio.

The laws introduced in the seven other states — more modest penalty increases (Iowa, New York and Texas), new small-claims administrative processes (Illinois and Maryland), and failure-to-pay-up penalties (Washington and California) — did not reduce the incidence of wage theft.

In other words, to effectively combat the problem, three elements were needed: Democrats in charge of state government; advocates tirelessly pressing for change; and strong policy solutions.

A Labor Day agenda: Work on state and local policies

With private-sector unions continuing their steep decline and little possibility of federal action, public policy at the state and local levels has become the main prize in the battle for workers’ rights.

That’s why we’re seeing ramped-up political activism from workers’ advocates. They’re building broader coalitions, organizing larger protests, developing more creative policy proposals, and working more closely with politicians so that when windows of opportunity open, they can get new laws passed that better protect workers and defend their right to get paid.

Daniel J. Galvin is an associate professor of political science and a faculty fellow at the Institute for Policy Research at Northwestern University. He is the author of “Presidential Party Building: Dwight D. Eisenhower to George W. Bush” (Princeton University Press).