&w=1440″ class=”ff-og-image-inserted” />

But even when the administration succumbed, some warned that damage might already have been done — that certain immigrants might still shy away from responding because of fears engendered by the lengthy battle. And there was data to back that up.

So did it happen? It’s not quite clear that it did in significant measure, but there are some indications it might have — though perhaps to the detriment of Trump’s red-state allies rather than Democrats.

On Monday, we learned which states will gain and lose seats thanks to the new census population numbers. Perhaps the three biggest surprises to come from the reapportionment announcement were in Arizona, Florida and Texas — all of which failed to gain an extra seat despite projections that they would. What else do those states have in common? They all have among the highest Hispanic populations in the country, each ranking in the top six nationwide, at around one-quarter Hispanic or more.

A report from redistricting expert Kimball Brace’s Election Data Services spotlighted those three states, noting the disconnect with 2019 population estimates from the Census Bureau late last year. In it, Brace “speculated that it’s possible the southern state changes, with their large Hispanic populations, have been caused by the Trump administration’s efforts to keep noncitizens from being counted in the census.”

Michael Li, a redistricting expert at New York University’s Brennan Center for Justice, pointed specifically to Texas.

“We’ll have to wait for more granular data, but it certainly looks like the Texas Legislature’s decision not to budget $ to encourage census participation combined with the Trump administration efforts to add a citizenship question cost Texas a congressional district,” Li said in a tweet.

At this point, this is a somewhat speculative exercise, given we don’t have detailed population data from the census — including on race. There is also a history of certain groups such as Hispanics being undercounted relative to other groups.

“We’ll have a much better sense starting in August when the block-level data becomes available, but even then it won’t be a complete picture,” Li told me. “But at least we’ll have more data to compare.”

He added that “right now, people mostly have educated guesses.”

In addition, Li noted that some states with large undocumented populations did relatively well and beat expectations. New York, for instance, came extremely close — within 89 people! — of not losing any seats at all, despite projections that it was a shoo-in to do so. Two other states with large Hispanic and immigrant populations — Illinois and, especially, New Jersey — also did relatively well.

New York was the closest to miss out on a seat, according to EDS, followed by Ohio, which also lost one seat. After that, the next closest were the three mentioned at the top: Texas, Florida and then Arizona. Texas came within 190,000 people of gaining a third seat, in a state with about 5 million immigrants, according to 2019 census estimates. Florida came within 172,000 of winning a second seat, in a state with more than 4 million immigrants. And Arizona came within 80,000 of gaining one seat, in a state with around 1 million immigrants.

And despite the exceptions, most of the states with large Hispanic populations did relatively poorly, failing to beat 2019 estimates by as much as most states. Beyond Arizona, Florida and Texas, they include California, Colorado and Nevada. (Only California lost a seat, which was expected, but the others were also outpaced by most other states.)

Arizona, California, Colorado and Florida all ranked in the bottom 10 relative to 2019 population estimates, according to numbers crunched by The Fix. The only others in the bottom 10 were rural Southern states and Wyoming.

And here comes the but: Some of those states have fared poorly in the census before, with Arizona, Florida and Texas also ranking in the bottom 10 when it came to census omission rates after the 2010 Census, according to a study from demographer William P. O’Hare. There is indeed some question as to whether this is a normal occurrence — even one involving immigrant non-responses — or something potentially exacerbated by the census citizenship effort (to whatever degree) or even the coronavirus pandemic.

But these states also lost out on seats by relatively fine margins — not quite 89 people, but by less than most any other state. So even if relatively few people were discouraged from participating for certain reasons, including a fear among undocumented immigrants that they would be giving the government information that could later be used against them, that could matter.

Arizona fared the worst relative to both expectations and its 2010 omission rate, coming in more than 1 percent shy of its 2019 population estimate. Its governor Doug Ducey (R) said in 2019 that he favored the census citizenship question, despite cautions that it could cost the state federal aid by depressing responses and perhaps even a congressional seat if immigrants didn’t respond.

We’ll have to await more data, but it’s a valid question as to whether the whole botched effort did just that. And if it did, it might have actually worked against Trump’s party by depriving red states of more seats — as multiple people, in fact, suggested it might.