I’ve said before it’s worth noting that while loss of employment is the most common downside brought up about increases in the minimum wage, it’s far from the only downside. Employers can cut costs among multiple margins – perhaps avoiding cutting jobs by cutting hours and benefits or putting less effort into ensuring a pleasant work environment. But a newly published paper takes another look at a deleterious consequence of minimum wage increases. 

The author, economist Seth Hill, starts by making the same observation – that while “employment might be the consequence of minimum wage laws most often evaluated…economists have also looked at alternative effects” such as:

An increased minimum wage might drive declines in nonwage compensation such as fringe benefits, job flexibility, or incidental experience at the workplace. An increased minimum wage might influence prices or customer service, might cause substitution of higher-skill for lower-skill labor, and might lead to increased loitering or petty crime from young males losing hours or employment.

Hill also points out that these negative effects are not evenly distributed – they most heavily impact the people who are struggling the most:

If there are negative distributional consequences of minimum wages, they most likely fall on the lowest-skilled workers whose marginal revenue product falls below the wage floor. The causes of low-marginal revenue product – low skills, drug addictions, mental illness – likely also cause other hardships meaning that negative consequences of minimum wages might more often fall on those already struggling at the economic margins. There could be different and compounding negative consequences of minimum wage increases for Americans at the margins. Low-skill workers could lose employment; could retain employment but see hours or benefits reduced; could have others in their support network experience employment disruptions; or, could face higher prices in the goods and services – for example housing – consumed by low-wage workers. Employment or support network disruptions might be particularly challenging for those with mental illness or drug addiction, two factors that harm marginal revenue product and relate to housing insecurity.

And housing insecurity is the consequence this new paper seeks to examine. Many activists cite the cost of living and the expense of housing as urgent reasons to drive up the minimum wage. But might this backfire? Could it be the case that higher minimum wages actually increase rather than reduce the incidence of homelessness? Hill examine this question using three different statistical methods. The results? According to the first method,

Municipalities that increased minimum wages by up to $2.50 per hour from 2013 to 2018 saw an average increase of 14 percent in homeless counts in the years 2014 to 2019 relative to municipalities with no nominal change in the minimum wage (real decline) or with changes pegged to inflation (real no change). Municipalities that increased minimum wages by more than $2.50 per hour from 2013 to 2018 saw an average increase of 23 percent in homeless counts in the years 2014 to 2019 relative to municipalities with no change.

The second method of analysis uses a stacked regression estimator method which “creates event-specific data sets to eliminate the staggered treatment problem.” According to this test,

Increases of $0.75 or more in local minimums increased relative homeless counts by about 25 percent in the years following the increase. All results hold with controls for changes in local income and local population.

The third method of analysis uses “a local projection difference-in-differences model” in order to “estimate the effect of continuous changes in the minimum wage rather than breaking changes into categorical treatments as required by the event study and stacked regression approaches.” The result?

This estimator suggests that when cities raise their minimum wage by 10%, relative homeless counts increase by three to four percent.

The overall conclusion:

Overall, these findings imply that minimum wages have negative distributional consequences not limited to disemployment. If a higher minimum wage causes economic harm more often for individuals with characteristics that cause hardship – low skills, lower education, mental or physical disabilities, criminal records, drug addiction – minimum wages could push some at the bottom of the economic ladder into housing, health, or physical insecurity.

It’s also worth noting that the minimum wage is not the only bad policy to blame here. The paper also contains the following observation:

While homelessness declined in most American cities from the end of the Great Recession up to the Covid-19 pandemic, in a small number of cities such as New York, Seattle, Los Angeles, and San Francisco, homelessness surged. O’Flaherty (2019, section 6) calls this one of the “mysteries” of scholarly understanding of homelessness. Minimum wages increased in these cities by 110, 98, 71, and 63 percent from 2006 to 2019. My evidence suggests these increases could have been an important factor driving increases in homelessness.

Another important factor is that these are cities where NIMBY policies are firmly entrenched. It seems very likely that the negative effects of NIMBYism and the negative effects of mandating wages above their market-clearing level are additive, and they can very well create results worse than the sum of their parts. And this is why its important to remember that job losses are just one of the ways minimum wages can inflict harm on the least well-off. Even if a minimum wage results in no net job losses, it can still cause lower skilled workers to be substituted for higher skilled workers, obscuring the full consequences of the policy:

The evidence presented here implies that focus on net employment could mislead evaluation of minimum wage policy. Aggregate employment might move very little in response to an increase in the minimum wage yet individual consequences for some of the lowest-skilled workers could be large, particularly if such workers are already on the economic margin or if loss of wages leads to loss of income-based subsidies.

Bastiat warned us not to look only at what is seen, but also what is unseen, when considering the consequences of a policy. This data helps shine a light on how the people who are the worst off can be made to suffer by minimum wage policies, in a way that would be rendered invisible by remaining focused solely on net employment changes.