The Latest Attack on Ron DeSantis: Florida Vaccinates Residents of Other States

Yet another attack on Florida’s vaccination rate, and on the governor, fails to tell the whole story.

T he Javert-like pursuit of Ron DeSantis’s record on the Covid pandemic never lets up. In the latest installment, a pair of reports by Chris Persaud of the Palm Beach Post contend that Florida has overcounted the number of Floridians vaccinated by at least 622,000, due in large part to Florida vaccinations of vacationing “snowbirds” who are not permanent residents of the state. Persaud also points to Florida’s vaccination of the employees of cruise ships based in Florida.

This is not the damning revelation that DeSantis’s critics seem to think it is. First, nobody is cooking any books. These are actual vaccinations, and they were accurately reported as occurring in Florida. The only question is whether they should be classified with other states’ vaccination rates because of where the people who got the shots in Florida maintain their permanent legal residence.

Second, we’re talking about a modest three percentage point difference in the state’s vaccination rate. Florida has, according to its official figures, vaccinated 15.5 million people compared with a population of 20.8 million people age five or older. (The state’s total population is 21.8 million.) That is a vaccination rate of 74.3 percent compared with the Florida population ages five and up, and 78.9 percent for ages twelve and up. If you take Persaud’s numbers at face value and just back out 622,000 vaccinations, the five-and-up rate drops from 74.3 percent to 71.3. Depending on how you rank different states — more on that below — a three-point adjustment would drop Florida no more than a state or two in the rankings. In either event, it currently stands within one to two points of the national average.

Persaud’s story was picked up by obsessive anti-DeSantis pundits such as Jonathan Chait of New York magazine and Tim Miller of the Bulwark. Chait in particular has gone to comical extremes in hunting for some way to stop DeSantis from denying Donald Trump the 2024 Republican nomination. He is not the only one; the Biden White House seems obsessed with knocking DeSantis on the pandemic even to the point of misrepresenting Florida’s vaccination rates.

It is difficult to overstate how excited Chait is over Persaud’s story. “The Last Defense of DeSantis’s Anti-Vaccine Campaign Goes Poof,” reads the headline of his latest piece, which breathlessly describes Persaud’s reporting as “a devastating revelation about the impact of Ron DeSantis’s war on vaccines.” Persaud specifically notes the three-percentage-point figure; Chait omits it. On Twitter, Chait proclaims that “the conservative defense of DeSantis’s vaccine policy turns out to be based on false statistics” and sneers that he is sure that Charlie Cooke, Karol Markowicz, and I “are working on their corrections now.”

Don’t hold your breath.

Let’s begin with the central accusation here: not that Florida is reporting vaccinations that didn’t happen but that Florida is vaccinating the residents of other states. It is bizarre to argue that this somehow proves that the state’s government is conducting a “war on vaccines.”

In just about any other area of public policy, this line of argument would be laughable. When a state has lots of gun sales to nonresidents, progressives do not say, “See, the governor isn’t as pro-gun as he claims.” Instead, they argue that the state is maliciously exporting an oversupply of guns to other states. When a state is a haven for the legal or tax residence of corporations that are actually headquartered elsewhere, progressives do not accuse it of a war on Big Business; they denounce it for starting a deregulatory “race to the bottom” that undermines the stricter laws of other states — an argument Chait himself once made in a famous piece denouncing the existence of the state of Delaware. I am quite certain that governors inviting out-of-state women to come to their states to have abortions would not be described as conducting a “war on abortion.”

In fact, as Persaud notes, in the first three months that vaccines were available in Florida, DeSantis was criticized by gubernatorial candidate Nikki Fried and other Democrats for requiring identification, because DeSantis did not want “tourists, foreigners” taking up the supply of vaccines intended for Floridians (although he did not object to part-time Florida-resident senior citizens). D.C. had the same problem. DeSantis relaxed those requirements in April 2021, once there was adequate supply — a step that made it easier to get more shots in arms. Is that a bad thing?

*  *  *

Is Persaud correct that Florida’s reported vaccination rate should be lower, compared with other states? That is not so easy to say.

The dirty not-so-secret of Covid data from the very beginning — data on infections, hospitalizations, deaths, and vaccinations — has always been that the data are imperfect. This is partly a problem with data in public policy and politics generally, and with Covid data, it comes in various shapes: population sizes that are a moving target, differing processes and standards for collecting data across jurisdictions, difficulty getting timely data from periodically overwhelmed hospitals, etc. The CDC has had its own foul-ups: It has been repeatedly called out by Florida for misreporting the state’s data, not necessarily out of malice, but simply because the CDC plays catch-up after not reporting data on the weekends. (Florida does, as do some other states, though there is no uniform standard nationally.) I have always tried to offer caveats about doing these sorts of comparisons.

Chait, at least, is compelled to concede in his article that Florida is not faking its data. He admits, regarding the claims of Rebekah Jones and her many admirers of deliberate falsification, “National Review effectively debunked her allegations.” Yet he persists in arguing that Florida’s data are flawed without acknowledging that everybody’s data are flawed. Any meaningful comparison should have some humility about the level of precision that is even possible on this topic. Only the obsessive and unseemly race to make comparisons between political leaders, policies, and voting bases has driven the illusion that we can run global pandemic data as if it were a controlled laboratory experiment.

Back in December, the CDC itself admitted the problems with vaccination data:

CDC officials said the agency may not be able to determine whether people are getting their first, second or booster doses if they got their shots in different states or even from providers in the same city or state. That can lead the CDC to overestimate first doses and underestimate booster doses, CDC spokesperson Scott Pauley said.

In a footnote on its Covid vaccination data tracker webpage, the CDC says: “There are challenges in linking doses when someone is vaccinated in different jurisdictions or at different providers because of the need to remove personally identifiable information (de-identify) data to protect people’s privacy. This means that, even with the high-quality data CDC receives from jurisdictions and federal entities, there are limits to how CDC can analyze those data.”

On its dashboard, the CDC has capped the percentage of the population that has received a vaccine at 99.9 percent. But Pauley said the figures could be off for multiple reasons, such as potential data-reporting errors or the census denominator’s not including everyone who lives in a particular county, like part-time residents.

Still, if we are looking at data and being skeptical about them, we should try to apply our skepticism consistently and with some sense of realism. Persaud uses a rough metric: He looks at zip codes where the vaccination rate exceeds 100 percent, which he sees as “impossible” rates that can only be explained by non-Florida-resident vaccinations. It is questionable, both as a methodology and as a public-health argument, whether he is always correct to complain about places with a lot of workers who aren’t residents in the same zip code as their employer. He focuses, for example, on the fact that the zip code “covering Universal Studios theme park southwest of Orlando, lists 43,956 vaccine recipients but 17,384 residents.” Then there is Port Canaveral:

The Cape Canaveral ZIP code of 32920 counts 19,565 “residents” vaccinated but a population just north of 10,000. That’s because crew members for cruise lines and cargo ships listed the ZIP code as their home address when they got immunized, Port Canaveral spokesman Steve Linden said.

The port coordinated with the state and Parrish Medical Center in Titusville, Florida, to inject the COVID vaccine into about 8,600 people, most of them crew members of cruise or cargo ships, Linden said. Port Canaveral was the first port in Florida to vaccinate crew members following a public health advisory approved by DeSantis, Linden said. Cruise terminals where the vaccine was administered were used as the home addresses of the crew receiving vaccinations while protocols were being developed. “The mission was to get people vaccinated,” Linden said. Since then, the state established protocols so that crew members who are vaccinated at the port — most of whom are from other countries — are listed by their home addresses.

Think about this for a second. This is an infectious disease. It doesn’t care where your legal domicile is. People who work at the Universal theme park are spending a significant amount of time in Florida interacting with big crowds of people (Florida residents and otherwise), regardless of whether they are legally residents of the state. It is good news for public health in Florida, then, for these people to get vaccinated, regardless of whether they actually permanently live there. The same is true of people spending a few months vacationing, or people who have long shore leaves between depositing large numbers of cruise passengers back in Florida.

Moreover, common sense says that there were undoubtedly a fair number of people who work at the theme park and got vaccinated there to start work, but soon relocated somewhere else in Florida. A zip code–based analysis will overlook this. I used to work in the 10048 zip code, which at the time consisted entirely of the World Trade Center; it will not surprise you to learn that the zip code had few permanent residents, yet 55,000 people worked there every day. That is an extreme example, but so are theme parks and ports.

Vaccination rates exceeding 100 percent of known residents is indeed a red flag that your data may not be telling you what they purport to explain, but the problem of drawing conclusions from this metric is well known within the realm of voter fraud. It happens nearly every election cycle: some urban or campus-heavy jurisdiction appears to have above 100 percent voter turnout, casting more ballots than the known number of registered voters in the jurisdiction. Conclusive proof of voter fraud, right? But, frequently, there is a valid reason: The as-of-Election-Day registration numbers are a lagging indicator because they don’t reflect people who registered late, or there are a lot of college students who may be permitted under some states’ laws to vote there. Big mismatches are worthy of investigation, but they do not, by themselves, necessarily prove what they seem to show.

For a variety of reasons, vaccination data have analogous problems. CDC data show greater than 100 percent vaccination rates among people age 65 and up in 29 states:

The CDC responded with a classic “fudge factor” fix: It capped maximum reportable vaccination rates at 95 percent for these populations, a fact now noted in a disclaimer on its website.

Florida, of course, has both a very mobile population — lots of snowbirds, lots of immigrants — and a dramatic rate of people relocating to the state during or because of the pandemic. That can cut both ways, because it also means that it is hard to pin down the population figures used as a baseline, as well as the local residence of Floridians. Persaud hand-waves this factor as inadequate by itself to explain all the discrepancies, but given that the entire controversy is around three percentage points (which he compares with a 1.5 point variation in California and 1.4 points in Arizona, using the same methods), it is worth asking whether population shifts partly explain why Florida has vaccinated people who had not yet established a permanent residence in the state at the time they got their shots.

The population shifts have been significant, and given how people tend to move to Florida in stages, even the official Census Bureau estimates of 243,000 additional residents between April 2020 and July 2021 may be understating things — especially if one continues the trend line since then. As the Palm Beach Post reported in February:

More than 547,000 people exchanged out-of-state driver’s licenses [in 2021] for ones with Sunshine State addresses. That’s a 40% increase from 2020 and nearly 20% greater than the five-year average between 2017 and 2021. The license swaps [were] largely from New York (11%), New Jersey (6%) and foreign countries (14%). . . . Of the 61,728 New Yorkers who in 2021 handed over Empire State driver’s licenses for a Florida license, 19,100 or 31% got IDs with Broward, Miami-Dade or Palm Beach County Zip codes.

Perhaps not coincidentally, Miami-Dade and Palm Beach are two of the counties identified by Persaud as having an impossibly high number of vaccinations compared with official tallies or estimates of their residents. That tracks with the experience of people such as Karol Markowicz, who got vaccinated in Florida while a New York resident — and now lives in Florida.

*  *  *

One thing I have tried to do consistently throughout the pandemic is to use apples-to-apples comparisons: Draw Covid data about a country or state from a source and compare them to the data for other places from the same source. The New York Times, for example, maintains a nicely visualized 50-state vaccination tracker drawn from the CDC’s data. The Times has no obvious incentive to make Florida look better. Its data are no better than the CDC sources but at least allow for 50-state and national comparison.

On that chart, Florida ranks 19th among the 50 states in people receiving a single dose, just ahead of Oregon, Illinois, Nevada, and Minnesota, and well above Wisconsin, Michigan, and most of the southern states. Ranked by fully vaccinated, Florida ranks 22nd. Adjusting the data for Florida only makes methodological sense if you are also adjusting all the other states in order to present a proper ranking.

Persaud does his comparisons selectively, comparing “impossible” zip-code data in Florida only with the same metric in California and Arizona, never looking at other states. Chait, even though his entire argument is that Florida has done poorly on vaccinations compared with other states, does not bother to explain where he thinks Florida should rank if we used better data. In fact, he presents nothing at all as a basis for a better multistate comparison.

There is a reason for that.

Subtracting out 3 percent across the board drops Florida from slightly above to slightly below the national average. But why apply an adjustment only to Florida data, and not anywhere else?

Moreover, if the specific question being asked is how DeSantis’s approach has worked, the proper baseline is Florida’s region of the country. Looking at the CDC’s own maps tells you that Florida has remained consistently ahead of its neighbors by any metric of vaccination:

If we focus on the 13 other red states south of Virginia and Illinois and east of New Mexico — including two states, Louisiana and Kentucky, with Democratic governors — we can see that Florida’s vaccination rates are significantly higher than those of its neighbors, even if Florida and only Florida is adjusted downward by three points. Only Kansas — the most geographically and culturally distant of the peer group — tops Florida in any category:

Persaud, instead, tries to compare Florida’s hospitalizations during the Omicron wave with those in California, Arizona, and Texas. Chait likes Persaud’s chart so much, he reproduces it in his column. Given that we know by now that Covid is seasonally variable at different times in different places, I would question whether the climate of Miami is really more comparable to that of San Francisco than to that of Atlanta or New Orleans. Persaud’s chart somehow leaves out 17 states, including Georgia as well as several more northerly states (New York, New Jersey, Pennsylvania, Maryland, and Ohio) that peaked at higher hospitalization rates than Florida during the Omicron surge.

As with vaccination rates, we can bicker about what is and is not a proper peer comparison for Florida. But that gets us back to an additional apples-and-oranges problem. On the one hand, critics try to adjust Florida’s vaccination figures for factors such as transient residents, theme parks, cruise ships, immigrants, and a vast elderly population. On the other hand, the same critics never seem to count those factors in the unique challenge of handling a pandemic in a state with all of those things happening at once. Florida’s diverse, mobile, and senior-heavy population, surrounded by tourists hitting Disney and spring break, the cruise ships and the Daytona 500, present a nightmarish scenario for a governor trying to keep a mobile virus at bay. Given these challenges, it is surprising that 15 other states have higher per capita death rates during the pandemic. Persaud quotes one Miami-Dade County official discussing “international people who give a local address just to get the vaccine. . . .We have vaccinated almost all of Colombia. That’s one of the jokes.” Fair enough, but half of Colombia isn’t visiting Vermont or Oregon.

*  *  *

This, from Persaud, is also a stretch in suggesting that DeSantis has somehow unreasonably talked down the vaccines based on the Florida data:

During a Jan. 3 news conference in the county, DeSantis referenced [Miami-Dade’s] on-paper immunization rate when he bad-mouthed the shots. “With omicron, you know, the vaccinations are not preventing infection,” he said.

Of course, it should hardly be a controversial point by now that vaccinations, while effective at reducing the severity of the Omicron variant, were notably ineffective at preventing Omicron infections. That was the whole point of waves of media hype at the time about “breakthrough infections” and about the persistence or revival of mask mandates and school closures. It is why the CDC is even now pushing to extend mask mandates in air and train travel. It is why two of the five people in my own household, all vaccinated, got infected during the Omicron wave. So have Nancy Pelosi and numerous other prominent Democratic politicians, all of whom say they are vaccinated. Criticizing DeSantis for pointing out the obvious is not much of a critique.

*  *  *

Broadly speaking, as I have discussed elsewhere, DeSantis proceeded in three stages on Covid vaccines. In the first phase in early 2021, when there was far more demand for vaccines than the available supply, he fought off all manner of silly Democratic attacks to focus on providing access for senior citizens in Florida — such as offering vaccination through the grocery chain Publix. That was consistent with his view from early on that the main public-health priority in the pandemic should be protecting people over 65, rather than on gagging children and shutting down businesses and schools.

In the second phase, running roughly from the spring to the midsummer of 2021, DeSantis was a vocal and ardent advocate of getting vaccinated. This was the crucial period for public persuasion, and DeSantis was a vigorous public presence bluntly urging people to get vaccinated. Florida vaccinated 8.3 million people as of mid April 2021. By early August, that number cracked 13 million.

In the third phase, DeSantis shifted focus and began more actively standing up for the remaining unvaccinated population against mandates. Undoubtedly, this was partly a matter of political calculation: Joe Biden was leaning hard on compulsion to get more people vaccinated, and standing up for the unvaccinated had a constituency (in the Republican Party, but also including independents and Democrats that Biden disdained) that was going unserved by both the president and Donald Trump. Yet DeSantis’s shift in emphasis was also grounded in a rational assessment that we had reached the point of diminishing returns in trying to talk people into the vaccines and that were going to end up inflicting more harm than good with heavy-handed government mandates aimed at the most recalcitrant holdouts. Chait’s relentless focus on the third phase not only ignores what DeSantis accomplished in the first two phases; it ignores even the possibility that the point of diminishing returns on both persuasion and coercion had been passed. Chait regularly assumes, without demonstrating, that people who ignored DeSantis telling them to get vaccinated for seven months would suddenly be receptive to further appeals in the eighth month.

DeSantis pledged in an early August press conference — just before the start of Florida’s school year — both to oppose a “biomedical security state” and to fight to keep Florida schools open in-person. DeSantis has occasionally gone overboard; I criticized him for opposing vaccine mandates on cruise ships. But any public-health campaign, like any effort to stamp out any sort of public immorality by force, eventually reaches the point at which its costs exceed the meager gains to be made at the margins by pushing harder. Florida has vaccinated 2 million more people since Labor Day 2021, but the pace has slowed. It has slowed everywhere, and not only in the United States. The hard reality is that some people will be vaccinated only by escalating government coercion that costs livelihoods and degrades critical public services, and some will resist even beyond that point. Persuasion that doesn’t persuade is simply for show. That DeSantis may have judged the point of diminishing returns earlier than some others did does not make him an anti-vaxxer.

*  *  *

Along the way, Chait also — without any citation or explanation — slags DeSantis for “punishing a state official merely for encouraging his staff to get vaccinated.” “Merely for encouraging” is doing a lot of work in that sentence. The case he references is that of Orange County (Fla.) health director Dr. Raul Pino, who was placed on leave by the Florida department of health in January and reinstated in March. Pino is a close ally of Orange County mayor Jerry Demings, whose wife Val is running for the Senate against Marco Rubio — so skepticism of political motives should run both ways here. The Florida department of health stated in January that it was reviewing Pino’s email to his staff because “the decision to get vaccinated is a personal medical choice that should be made free from coercion and mandates from employers.”

What raised alarms at the Florida department of health was not just that Pino leaned on his staff to get vaccinated but that his email suggested that he had accessed their private medical records. Florida-based NBC reporter Marc Caputo got a copy of the email Pino sent to his subordinates, in which he states that “I ask [sic] our analyst to run vaccination data for our employee [sic].” This raised entirely understandable concerns about potential privacy violations:

If your boss sent this email, wouldn’t you be at least a little concerned that he was prying a little too closely into your medical file?

*  *  *

Ron DeSantis has faced particularly daunting challenges in getting Florida through two years of a global pandemic. As with any governor or head of state during that period, he has handled an unenviable task by making lots of decisions and public statements, some of them admirable, some of them dubious. DeSantis is proud of his record and is running on it for reelection. He may soon be running on it for president. It is fair to scrutinize that record closely. But that task is not served by applying a different standard to Florida than to anywhere else, or by overhyping comparatively small, marginal differences of a few percentage points.