The US Census Bureau thread

[galleyfox]

Originally Posted by iheartthed

But the census form is tied to the residence, not the person.

That’s the error. Because a person can own multiple residences and fill out multiple forms or respond to more than one census worker.

For example, my parents were divorced and had joint custody of my siblings and I. One year, both households included us in the census form sent to the residences. Whoops, three extra children who don’t actually exist.

The census tries to eliminate duplicates like that, but it gets tricky when people have residences in different states.

[ChiSoxRox]

The 2021 Census city estimates drop at midnight Eastern Time tomorrow night.

They're going to be the same "brunt of work from home" as the county numbers, plus the usual grain of salt for estimates, but still new numbers to scratch the analysis itch.

[C.]

Originally Posted by ChiSoxRox

The 2021 Census city estimates drop at midnight Eastern Time tomorrow night.

They're going to be the same "brunt of work from home" as the county numbers, plus the usual grain of salt for estimates, but still new numbers to scratch the analysis itch.

Hmmm…

[R1070]

Will the census' admitted undercount and overcount start to reflect corrected numbers in these estimates?

[ChiSoxRox]

Originally Posted by R1070

Will the census' admitted undercount and overcount start to reflect corrected numbers in these estimates?

Doesn't look like it. From the Census Bureau announcement:

The PES [corrections] results are necessarily limited due to the limited size of the samples. Results cannot reliably be broken down by demographic characteristics or geographic areas within states. Note that for the 2010 PES, the Census Bureau also was unable to release state estimates of coverage by demographic groups within states. Given the sample size for the 2020 PES and the assumptions required to make unbiased sub-state estimates, the Census Bureau was unable to include county or place estimates in the 2020 PES reports, as well.

For the numbers coming tonight, the stand alone cities over 250k we already have from the county estimates:

City......2020 Census.....2021 estimate
New York, NY.....8,804,190.....8,467,513
Philadelphia, PA....1,603,797....1,576,251
San Francisco, CA.....873,965....815,201
Denver, CO......715,522....711,463
Washington, DC.....689,545....670,050
Baltimore, MD....585,708....576,498
Virginia Beach, VA....459,470....457,672
New Orleans, LA....383,997....376,971
St. Louis, MO.....301,578....293,310
Anchorage, AK....291,247....288,121
Chesapeake, VA.....249,422....251,269 (the new #2 in Virginia)

Suffolk County, MA......797,936....771,245 - Boston plus a bit

The numbers will be ugly from being the peak of WFH (July 1, 2021). However, a key clue on urban recovery will be the state estimates for 2022 this December, as the change for NYS is largely NYC proper.

[Yuri]

More random stuff from the 2020 Census:

-------------------- 2020 -------- 2010 -------- 2000 -------- 1990 -------- Km² -- 2020/2010 - 2010/2000 - 2000/1990

TEXAS --------- 29.145.505 -- 25.145.561 -- 20.851.820 -- 16.986.510 -- 676.653 - 15,9% - 20,6% - 22,8%

Dallas* ------------ 7.834.528 --- 6.538.601 --- 5.307.912 --- 4.108.439 --- 25.975 - 19,8% - 23,2% - 29,2%

Houston ----------- 7.122.240 --- 5.920.416 --- 4.693.161 --- 3.750.883 --- 21.416 - 20,3% - 26,1% - 25,1%

San Antonio ------- 2.558.143 --- 2.142.508 --- 1.711.703 --- 1.407.745 --- 18.941 - 19,4% - 25,2% - 21,6%

Austin ------------- 2.283.371 --- 1.716.309 --- 1.249.763 ----- 846.227 --- 10.928 - 33,0% - 37,3% - 47,7%

Killeen --------------- 475.367 ----- 405.300 ----- 330.714 ----- 268.822 ---- 7.329 - 17,3% - 22,6% - 23,0%

Rest of Texas ----- 8.871.856 --- 8.422.427 --- 7.558.567 --- 6.604.394 -- 592.064 -- 5,3% - 11,4% - 14,4%

*Sherman and Granbury areas included

About 90% of Texas 2010-2020 growth happened in those MSAs while the "rest of Texas" is very close to post negative growth, specially as other former fast-growing areas (McAllen) had slowed down a lot.

[R1070]

Originally Posted by Yuri

More random stuff from the 2020 Census:

-------------------- 2020 -------- 2010 -------- 2000 -------- 1990 -------- Km² -- 2020/2010 - 2010/2000 - 2000/1990

TEXAS --------- 29.145.505 -- 25.145.561 -- 20.851.820 -- 16.986.510 -- 676.653 - 15,9% - 20,6% - 22,8%

Dallas* ------------ 7.834.528 --- 6.538.601 --- 5.307.912 --- 4.108.439 --- 25.975 - 19,8% - 23,2% - 29,2%

Houston ----------- 7.122.240 --- 5.920.416 --- 4.693.161 --- 3.750.883 --- 21.416 - 20,3% - 26,1% - 25,1%

San Antonio ------- 2.558.143 --- 2.142.508 --- 1.711.703 --- 1.407.745 --- 18.941 - 19,4% - 25,2% - 21,6%

Austin ------------- 2.283.371 --- 1.716.309 --- 1.249.763 ----- 846.227 --- 10.928 - 33,0% - 37,3% - 47,7%

Killeen --------------- 475.367 ----- 405.300 ----- 330.714 ----- 268.822 ---- 7.329 - 17,3% - 22,6% - 23,0%

Rest of Texas ----- 8.871.856 --- 8.422.427 --- 7.558.567 --- 6.604.394 -- 592.064 -- 5,3% - 11,4% - 14,4%

*Sherman and Granbury areas included

About 90% of Texas 2010-2020 growth happened in those MSAs while the "rest of Texas" is very close to post negative growth, specially as other former fast-growing areas (McAllen) had slowed down a lot.

Are these updated numbers from a particular source? These don't look like the 2020 numbers I saw before.

[ChiSoxRox]

The 15 Most Populous Cities on July 1, 2021
Rank....Area Name, State Name....2021 Total Population

1 New York city, New York....8,467,513
2 Los Angeles city, California....3,849,297
3 Chicago city, Illinois....2,696,555
4 Houston city, Texas....2,288,250
5 Phoenix city, Arizona.....1,624,569
6 Philadelphia city, Pennsylvania...1,576,251
7 San Antonio city, Texas....1,451,853
8 San Diego city, California....1,381,611
9 Dallas city, Texas...1,288,457
10 San Jose city, California....983,489
11 Austin city, Texas....964,177
12 Jacksonville city, Florida....954,614
13 Fort Worth city, Texas....935,508
14 Columbus city, Ohio....906,528
15 Indianapolis city (balance), Indiana....882,039

The top 15 largest cities remained the same as in 2020 although more than half experienced decreases in their population between 2020 and 2021: New York, New York (-305,465); Los Angeles, California (-40,537); Chicago, Illinois (-45,175); Houston, Texas (-11,777); Philadelphia, Pennsylvania (-24,754); San Diego, California (-3,783); Dallas, Texas (-14,777); San Jose, California (-27,419); and Indianapolis, Indiana (-5,343).

The other six most populous cities experienced some moderate to small growth in population. San Antonio, Texas, had the highest numeric gain of 13,626 people between 2020 and 2021. Followed by Phoenix, Arizona (13,224), and Fort Worth, Texas (12,916).

The remaining three large cities experienced relatively small population growth. Those cities were Jacksonville, Florida (4,151); Austin, Texas (1,056); and Columbus Ohio (adding only 668 people).

Census release

Most surprising to me is that Houston drop.

[Yuri]

Originally Posted by R1070

Are these updated numbers from a particular source? These don't look like the 2020 numbers I saw before.

Those are exactly the 2020 Census. I only added to Dallas MSA, the Sherman MSA and the Granbury micro. Aren’t you mistaken 2020 Census by the 2021 Estimates?

[summersm343]

I wouldn’t put any ounce of validity into these counts.

[iheartthed]

Originally Posted by summersm343

I wouldn’t put any ounce of validity into these counts.

Agreed.

[Steely Dan]

Originally Posted by summersm343

I wouldn’t put any ounce of validity into these counts.

Double agreed.

[Tuckerman]

Originally Posted by Steely Dan

Double agreed.

Triple agreed--this was a very dicey period for estimated change - effects of pandemic will take time to assess

[ChiSoxRox]

Fair enough, I feel similar but I did want to pass th estimates along.

As mentioned above, December is the release of the July 2022 state populations and I am eager to see what direction New York and DC go in, as those will be the post-WFH bellweathers.

Did WFH hit Houston enough to drop its population, or is that just a false signal? Even Harris County dropped!

[Steely Dan]

Originally Posted by ChiSoxRox

Fair enough, I feel similar but I did want to pass th estimates along.

Oh yeah, I wasn't trying to say that I'm not at all interested in seeing the 2021 city estimates, just that with everything that was going on last year, combined with the CB's atrociously awful track record of estimating big messy legacy cities like Chicago, I'm not putting one single iota of stock in the numbers.

An interesting data point, yes.

But ultimately, a meaningless one in my eyes.

[DCReid]

Originally Posted by ChiSoxRox

Fair enough, I feel similar but I did want to pass th estimates along.

As mentioned above, December is the release of the July 2022 state populations and I am eager to see what direction New York and DC go in, as those will be the post-WFH bellweathers.

Did WFH hit Houston enough to drop its population, or is that just a false signal? Even Harris County dropped!

It's hard to believe now, but oil prices briefly went negative in 2020, which I am sure impacted the Houston economy more than other metros. Oil companies had nowhere to store oil supplies, oil storage tanks were full, and oil companies were willing to pay any company to take their excess oil away. The census also shows that the smaller Texas oil centers in Midland and Odessa also dropped quite a bit. Also, it seems than Houston and Harris County may be experiencing outmigration to far flung suburbs and counties, similar to what Dallas city and county seems to have been experiencing for years. There still seems to be a lot of development in the city itself, but most of the growth now appears to be in far flung suburbs.

[Steely Dan]

* FOR ENTERTAINMENT PURPOSES ONLY *


here are the 100 largest municipalities in the nation reversed-ranked by growth %

(i reversed it because most of the cities SSPers care about were negative).

remember to take this all with a GIANT grain of salt considering the monumental shitshow that was 2021.

[ChiSoxRox]

*If* those numbers are close to accurate, not a bad showing by Detroit. Yeah, another dip but a moderate one, and holding population better than LA or Dallas.

But the same Bureau had Detroit holding pretty steady at 910,920 in the 2009 estimates before the 2010 Census dropped its steepest ever decadal decline.

[Crawford]

Originally Posted by ChiSoxRox

*If* those numbers are close to accurate, not a bad showing by Detroit. Yeah, another dip but a moderate one, and holding population better than LA or Dallas.

I mean, you really think a lot of Detroiters were hanging out on Florida beaches, or in the Tetons, during the pandemic?

This imputation was taken at the height of the pandemic. SF had the worst loss bc it's wealthy and mobile. Affluent households with mobile jobs and second homes. If Manhattan were a city, I bet you it would have been much worse than SF. The Upper East Side was basically a ghost town during peak pandemic.

It would be really weird if Detroit (city proper) had extensive temporary Covid migration.

[Tuckerman]

[QUOTE=Steely Dan;9634638]* FOR ENTERTAINMENT PURPOSES ONLY *


A big grain of salt indeed. Any good statistician knows that there is a lot of "static" in estimating trends over time and that accurately predicting a one year interval change is mostly nonsense. Time series type data need measurements over many equal interval data points. Even then these type analysis are noisy and full of errors.