In any given pay period, hundreds of thousands of Americans have an extra deduction on their paystubs.
Wage garnishment allows creditors to collect on defaulted debts by taking money directly from workers’ paychecks. But historically, we haven’t known how many people are impacted by this practice or how much their creditors take.
That’s because wage garnishment is largely hidden. Employers are aware of it because they implement garnishment orders, but there is little reason for anyone else to know, and the stigma surrounding garnishment means that many people keep their experiences private. Not only is the topic absent from casual conversation, there has also been a dearth of data for researchers to analyze.
But a new study from Kellogg’s Anthony DeFusco is shedding light on the practice. DeFusco, an associate professor of finance, worked with Brandon Enriquez and Maggie Yellen of MIT to analyze data from Automatic Data Processing (ADP). The company provides administrative payroll services to other businesses and distributes paychecks to about a fifth of U.S. private-sector workers. ADP recently began collaborating with academics to gain a richer understanding of the data it generates.
“The institution of wage garnishment is part of people’s lives, but we really don’t know anything about it. Data just has not been available,” DeFusco says. “We wanted to know how big of a deal it is for people, and how it could be interacting with the choices they make in the labor market.”
DeFusco’s analysis of ADP’s data showed that wage garnishment is widespread and onerous. He looked at the pay data of private-sector workers between 2014 and 2019 and found that about 1 percent of workers were experiencing wage garnishment at any given time. And creditors took a very significant chunk out of those workers’ checks. The average garnishment is around 10 percent of gross earnings—roughly the same amount that people spend on groceries.
“We were somewhat surprised to find how common it is. One in a hundred people is a lot,” says DeFusco.
The prevalence and severity of garnishment were not the only pieces of fundamental information that were missing. Because garnishment is hidden behind a veil of secrecy, researchers and policymakers have not known which groups of people are most likely to experience garnishment, how long garnishments tend to last, and whether garnishment affects workers’ employment decisions. The ADP data yielded preliminary answers to each of these questions—and raised several more.
Garnishment is brief but painful
When people default on their debt, creditors can go to state courts to seek wage-garnishment orders. You might think, then, that data about the practice would be relatively straightforward to access via court records.
But historically, this has not been the case. For one, the data is inconsistent because jurisdictions differ in how much information they make public. Plus, states vary in how much protection against garnishment from private creditors they offer. (Four have banned the practice altogether.) Moreover, different rules may apply to different types of debt: for instance, student-loan debts versus medical debts. This regulatory patchwork makes it difficult to draw widespread conclusions using data from just one location.
But ADP’s dataset provided detailed information from many different places for the first time. This allowed DeFusco and his colleagues to get a clearer sense of the big picture.
The team examined a 1 percent random sample of anonymized data from workers aged 16–64 who live in the 46 states that allow wage garnishment. This information spanned a five-year period, and included each worker’s total hours worked, earnings, taxes paid, and, most importantly, wage garnishments. It also included basic demographic information.
In addition to determining how many workers’ wages are garnished and how much is taken from their checks, DeFusco and his coauthors also learned that garnishment is generally brief. Student-loan garnishments spanned an average of 7.6 months, while other creditor debt lasted just 4.8 months.
These durations could indicate that some people are filing for bankruptcy after several months to end the financial pain, although it’s clear from the sheer number of garnishments that not everyone is taking advantage of this exit strategy.
“In some ways, it is easy to get out of wage garnishment. That’s what bankruptcy is designed to do,” DeFusco says. “So, it was surprising that so many people are stuck in this kind of purgatory, with creditors garnishing their wages.”
DeFusco and his colleagues also found that workers who experienced garnishment were less likely to be working for the same employer one year after garnishment began. While 65 percent of non-garnished workers remained with the same employer, only 61 percent of garnished workers did.
It is possible that some workers could be switching jobs to escape garnishment. Garnishment orders are made to employers, and when a worker finds a new position, the garnishment doesn’t automatically follow them. Creditors need to figure out who a debtor’s new employer is and repeat the process. This takes time and money, and creditors don’t always bother.
“This evidence is suggestive, but it is still an open question whether garnishing people’s paychecks is actually what’s causing them to transition jobs at higher rates,” says DeFusco. “If it is, it could matter for labor markets, and this is a possible avenue for future research.”
Disparities could point to systemic problems with garnishment—or reflect societal inequality
Finally, the researchers wanted to understand who bears the brunt of wage garnishment and how adversely it affects them. Given inequalities in other measures of economic well-being, they hypothesized that garnishment would not be equally distributed, and the data bore out that suspicion.
“We found it was concentrated in particular segments of the population, among people with lower levels of education, people from minority backgrounds, and middle-aged households,” says DeFusco. “We expected to see that, but we think it is important to document these disparities.”
While detailed demographic information was not available, zip code–level data showed that garnishments were much more common in neighborhoods with higher proportions of Black residents. Neighborhoods with the lowest proportion of Black residents had a garnishment rate of 0.7 percent, while neighborhoods where more than three-quarters of residents were Black had a garnishment rate of 1.8 percent.
The precise contributors to this inequality aren’t entirely clear. “Is wage garnishment unequally distributed because debt is unequally distributed? Or is it because there is something systemic about the way this institution works?” says DeFusco.
In other words, researchers can’t determine whether workers in majority-Black neighborhoods are subject to higher levels of garnishment because they are more likely to be in debt in the first place or because they are targeted for garnishment more aggressively. If it’s the latter, DeFusco says, “there could be something specific to the garnishment regime that regulators should be taking into account.”
Additional research is needed to deepen understanding
Wage garnishment is common, but knowledge of the practice is still in its infancy.
While DeFusco and his coauthors established valuable information about how garnishment, race, education level, and neighborhood interact, more granular data is still lacking. For example, nobody knows how factors such as immigration status affect workers’ likelihood of experiencing garnishment. DeFusco would like to see research into wage garnishment continue. “Regulators will benefit most immediately from additional information. The more they have, the better targeted regulations will be,” he says.
Eventually, workers, employers, and even creditors all stand to gain from a richer understanding of wage garnishment if it in turn leads to better policies. For instance, workers could receive greater protection against predatory behavior, while creditors could see a reduction in red tape.
But the journey from data to policy is just beginning. “At this stage,” DeFusco says, “we’re still documenting basic facts.”