Detroit’s population is 620,376, according to the U.S. Census Bureau’s 2023 estimate. That’s an incredibly precise number. But the bureau’s methods for determining it are actually rather fuzzy.
The City of Detroit also thinks those methods are deeply flawed. It filed a lawsuit in March challenging the census population count, arguing in its filing that the method it employs to count cities is “irrational and discriminatory.”
If Detroit wins the lawsuit, the effects could be profound. Billions of dollars in federal appropriations are determined by population. Detroit and other cities could see millions more dollars each year.
It could also be a narrative-shifting event. Mayor Mike Duggan said in 2014 after his first election that his tenure should be defined by “whether the population of the city is going up or going down.”
What’s the city’s case about, and how much does Detroit and potentially other cities stand to gain? Here’s what the city and other experts are saying.
What is Detroit’s argument for the population undercount?
The City of Detroit says in its lawsuit that the way the Census Bureau determines population fundamentally disadvantages “older urban cities” like Detroit.
Every 10 years, the bureau attempts to count every person living in the United States. This decennial census determines how many U.S. congressional seats each state is awarded. It was last done in 2020.
The decennial census also sets a baseline number used for annual population estimates. City populations are counted by multiplying the number of housing units by the average number of people living in each unit. In Detroit’s case, that number is just over two.
According to the bureau’s methodology from 2022, the census tracks the number of housing units in two simple ways: demolition and new construction. If a unit is demolished, regardless of whether it’s occupied, the census subtracts about two residents. For every new unit constructed, it adds two residents.
Unfortunately for Detroit, the city has been on a demolition spree since Duggan took office. It demolished more than 4,000 homes in 2021 and 2022 alone, costing over 8,000 residents in the census count.
“To somehow say that homes that have fallen into disrepair for decades and have to be demolished is evidence of population loss is just incorrect,” said Conrad Mallett, the city’s corporation counsel.
That’s not the only way the city says it’s being punished.
The census only counts units in new buildings as adding to the housing stock. Detroit doesn’t think this accounts for all the vacant homes that have been renovated and occupied.
In a letter to the Census Bureau, Duggan wrote that more than 6,000 previously abandoned homes were renovated then inhabited in 2021 and 2022, but the census “does not recognize the existence of a single one” of them.
The city estimates the bureau undercounted the number of housing units during those two years by 11,449.
The census also has a so-called county cap rule, requiring the county’s population to match the total from all towns and cities within it. The numbers rarely align because county population is calculated using a different method — mostly births, deaths and migration. When the numbers don’t match, city counts are adjusted up or down using a single multiplier applied to all of them.
The city says in the lawsuit this cap has cost Detroit more than 20,000 residents over the last two years. Overall, the lawsuit puts the undercount at more than 38,000.
Is Detroit right that the Census Bureau got it wrong?
Census experts agree with the city.
“I believe the city has been screwed by the Census Bureau’s methodology,” said Kurt Metzger, founder of Data Driven Detroit, who spent 15 years with the Census Bureau’s Detroit office. “The census’s tool is very crude. Population doesn’t even enter into it — it’s all housing.”
Almost all the demolished housing had likely been vacant for years, he said, but the census still docked Detroit residents.
Metzger thinks the county cap also penalized Detroit.
“To use totally different methodologies and then at the end have everything add up makes very little sense,” he said. “If Wayne County is losing population, the census is going to act as if Detroit’s losing population, too.”
The Census Bureau declined to answer questions for this story, saying it doesn’t comment on pending litigation.
Detroit likely suffered from an undercount during the last decennial census in 2020 as well.
For the first time ever, the bureau had residents respond online. It did not go well for Detroit, often called one of the least connected major cities in the U.S. The vast majority of Detroit’s census tracts were in the bottom 20% nationally in response rates.
The census estimated Detroit’s population going from around 670,000 in 2019 to only 639,000 the next year — a drop of nearly 5%.
Experts didn’t buy it.
“A single-year decline of 31,000 residents is anomalous and implausible,” noted the University of Michigan’s Poverty Solutions initiative in its report on the 2020 census in Detroit. It added that from 2014-2019, the city lost an average of 2,000 residents per year.
If a resident doesn’t respond to the census, a surveyor is supposed to follow up in person. Ren Farley, one of the report’s authors, said there weren’t enough of these surveyors in Detroit, and they often looked at the wrong things.
“If a home is really dilapidated, it could be put down as vacant,” Farley said. “But there are some pretty derelict homes in Detroit that are occupied.”
What is Detroit asking for in its lawsuit?
The city wants the bureau to take more variables into account for its population counts than just demolition and new construction. U.S. Postal Service address registration and new utility accounts could provide a more accurate number.
The 2023 appropriations bill actually directed the bureau to consider new methodologies to improve accuracy in its counts.
Detroit also wants the bureau to incorporate other data the city has collected on Detroit Land Bank Authority certificates of compliance, permitted residential construction and verified nonpermitted rehabs.
“What we don’t understand,” Mallett said, “is why it’s so hard in this age of technology, particularly for the federal government, to use all the tools at its disposal to count everyone. Older urban cities need different kinds of considerations.”
How much money is at stake?
It’s impossible to know and incredibly difficult to estimate how federal funding would change if the census estimate for Detroit was revised. The city doesn’t even want to estimate its actual population.
“I don’t want to throw out a number in a cavalier manner before we’ve proven before a judge that the methodology is flawed,” Mallett said.
Metzger conservatively estimates the city’s population at around 650,000 — a difference of about 30,000 residents.
How many federal dollars that would equate to is hard to know. There are more than 300 programs that rely on census data to distribute funds, and 22 of them fluctuate quite a bit depending on population. The largest of these is Medicaid, and the percentage subsidized by the federal government varies wildly by state. States also have some flexibility in how they distribute funds to cities.
“It’s really difficult to come up with total dollars,” Metzger said. “We just know it’s a lot.”
The city says in its lawsuit that a higher population recount in Detroit could make a difference of “hundreds of millions of federal and state dollars.” Metzger and Farley estimate it’s closer to $4 million-$10 million a year that would come directly from federal appropriations.
The upstream effects of the court’s decision could be significant beyond Detroit. Other cities might challenge the census’s count on similar grounds. Perhaps entire county or state counts would then have to be adjusted.
“Black and brown and poor people across the country benefit. We want that to be the case,” Mallett said.
Beyond the federal appropriations, the city also wants to provide people with tangible proof that Detroit is no longer losing population.
“Detroit is thriving and that’s an important message to get across,” Mallett said.
This story was originally published by Aaron Mondry of Outlier Media.
