Commerce Department Asks For Appeal To End Census Counting Before Oct. 31
STEVE INSKEEP, HOST:
The man overseeing the census is on the line. The Constitution tells the United States to enumerate the whole number of persons in this country. And Commerce Secretary Wilbur Ross wants to finish right away. His department appealed a court ruling to keep counting until October 31. They expect a ruling today. The Trump administration wants numbers in time to remove unauthorized immigrants from the count that is used to apportion congressional seats. Many other things are at stake here, so we'll talk it over with Secretary Ross. Why is it so important to rush to finish?
WILBUR ROSS: It isn't a rush to finish. There is a statutory deadline of December 31, which Congress enacted quite a few years ago, in which, as far as I know, just about every census has completed in time for. The reason for that deadline is the apportionment material, which the president gives to the state, is then used to allocate congressional seats in state legislature. And for the next 10 years as well, it allocates federal funds.
INSKEEP: I want to grant, Mr. Secretary, there is a statutory deadline coming up at the end of the year. You're right about that. But you have a court that says you can keep counting. Everyone agrees an undercount is bad. In fact, people of color are often undercounted more. And you yourself said earlier this year you needed more time because of the pandemic. Why not just take it?
ROSS: Well, what we did when the papers came out and the coronavirus hit and then subsequently we were hit by some bad weather situations, particularly down in the Southeast, what we did was we thought through with the census career management, what could they do to improve things so that we would have a complete count and yet could meet the deadline? And what we opted to do was several basic things. No. 1, we hired some 30,000 more enumerators than had been planned. Second, we put in a bonus system so that they would work more hours per day than they had. In 2010, the average census taker worked 19 hours a week. We got them up to about 21 hours a week. So with more people working more hours per week and, frankly, because of better training, doing it with greater efficiency, (unintelligible) we have them computer scheduled in advance so that they have much more efficient route as they go house to house...
INSKEEP: Are you telling me, Mr. Secretary - forgive me. I'm...
ROSS: I'm sorry. I'm not quite finished. Given all the...
INSKEEP: Time is short.
ROSS: ...Factors, we didn't need as many calendar days to complete the census. And that's why we are already at - 99.7% of all the household have already been enumerated, and that's a tenth of a percent better than in 2010.
INSKEEP: That is an impressive number and maybe that's a good moment to bring in Hansi Lo Wang of NPR News. He's been covering the 2020 census and has a question for you. Hansi, go.
HANSI LO WANG, BYLINE: Good morning, Secretary Ross. Does that number, 99.7%, does that really tell us anything about how accurate the count is at this point? Doesn't that rate include homes where a census worker knocked on the door and nobody answered?
ROSS: No, we've - those are the cases that have been settled. All censuses end up doing after the door to door is complete what the code imputations. In 2010, every single state had some degree of imputation. Imputation is where a court-approved - in fact, approved by the Supreme Court - mathematical process for dealing with those who just don't answer. So the two are not logically inconsistent. We have gone up to six times to houses that didn't answer, and that's an awful lot of door knocks in addition to all the mailings that were done and all the other things. In some cases where we had phone numbers, we also made phone calls. So this is a true year-to-year, census-to-census comparison. And I'm very proud that we had already exceeded the prior census number. And we will actually go even slightly beyond that because that was as of the 5 of October. And now because of the court order, we'll be at least two days beyond that. Now, you don't get much toward the end because the law of diminishing returns sets in. So the next couple of days will be fairly microscopic improvement, but nonetheless will be even better than the 99.7% that we have already achieved.
INSKEEP: Let me ask about what happens once you turn over these numbers, whenever a court allows you to do so, maybe today, maybe October 31, we will see. The numbers go to President Trump. He then has an opportunity to remove unauthorized immigrants from that count. Is there a way you found to estimate how many unauthorized immigrants are in your count?
ROSS: Well, a couple of things to that. First of all, the president has the ability to do all kinds of modifications to the material that we send in. All we do is provide data. That's the job of the enumeration. What gets done with the data and indeed how they use it for allocations, that's beyond the ken of the census. Now, his paper, his executive order, describing the idea of removing certain parties from the rosters, that is subject to litigation. And there's litigation underway right now saying that he shouldn't be doing that. So we don't know how that will turn out.
INSKEEP: Meaning that there is no way to do that. Will you tell the public at the same time you tell the president what the count is so that we have a way to check how he might change it?
ROSS: Well, the census data will certainly be made public. There's no question about that. And indeed, it will ultimately be turned over by the president to the Congress. Now, what - as I say, what he will do with the raw data that we give him, that is something we have no control over.
INSKEEP: Secretary Wilbur Ross of the Commerce Department. We can continue this discussion, I hope.
NPR transcripts are created on a rush deadline by an NPR contractor. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.