Census population estimates 2011-2019

[cinyc]

You probably already knew this, but sometime this month. Last year, it was May 23.

It looks like the 24th

https://www.calendarwiz.com/calendars/calendar.php?crd=cens1sample&PHPSESSID=1376188eb446e0e6b8ec05b90f763b0f&jsenabled=1&winH=424

It's almost always the Thursday before Memorial Day weekend. That usually gives me the weekend to make my population change map. Now that I'm a little more adept with QGIS, I might have a quicker turnaround.

[cinyc]

May 24 is the date that the masses get to see the locality population estimates.

12:01AM, to be exact. Last year, I think we had to wait until the morning, around 10AM for the actual data. We still might have to this year.

I probably should start updating my shapefiles so I can update the pop estimate maps. The 2016 series is still available on Carto here, for now:
https://cinyc.carto.com/viz/2179fb11-c942-4958-b0e8-147324cb395a/public_map (2016 estimate vs the 2010 Estimates Base).

The 2016 series won't be around after Thursday. I'm going to have to nuke it to make space in my Carto account for the 2017 maps.

[cinyc]

A little under 5 hours until we start getting reports about the fastest-growing incorporated places. Any last guesses as to what was the fastest-growing city with over 10,000 people last year? Will Williston, ND make the list of the biggest percentage losers?

[cinyc]

The list of the fastest-growing cities is out.

The top 10 % gainers (50K+ pop):
1. Frisco, Texas 8.2%
2. New Braunfels, Texas 8.0%
3. Pflugerville, TX 6.5%
4. Ankeny, IA 6.4%
5. Buckeye, AZ 5.9%
6. Georgetown, Texas 5.4%
7. Castle Rock, Colorado 5.1%
8. Franklin, Tennessee 4.9%
9. McKinney, Texas 4.8%
10. Meridian, Idaho 4.7%

The top numerical gainers were San Antonio, Phoenix, Dallas, Fort Worth and Los Angeles.

Fort Worth passed Indianapolis to become the 15th largest city.

More here: https://www.census.gov/newsroom/press-releases/2018/estimates-cities.html

As expected, full data won't be available until morning, supposedly before 10AM.

[cinyc]

Here's the first set of this year's interactive maps:

2016-2017 Percentage Change in Incorporated Places & County Remainders
2010 Estimates Base - 2017 Percentage Change in Incorporated Places & County Remainders (takes into account annexations and the like)
2010 Census - 2017 Percentage Change in Incorporated Places & County Remainders

[cinyc]

Charlotte, NC will probably pass Indianapolis next year:

Charlotte: 731,424 (census)/735,688 (est. base 2010) --> 842,051 (2016) --> 859,035 (2017)
Indianapolis: 820,445 (census)/820,443 (est. base 2010) --> 855,164 (2016) -->863,002 (2017)

Charlotte gained 16,984 (+2.02%) last year vs. Indianapolis's gain of just 7,838 (+0.92%).

Although we can't predict what will happen exactly, if both cities maintain their 16-17 growth rate:

Charlotte 2018: 876,662
Indianapolis 2018: 870,911

So unless something crazy happens, Indianapolis will most likely drop to #17 in for 2018.

On a different note, my hometown of Elkhart, IN (52,558 in 2017) started growing again after getting screwed by a housing shortage.

Also, Austin, TX's growth slowed dramatically, wow.

Other random facts that occurred last year:

Chicago gained about 12k
The unthinkable happened: Detroit grew
Salt Lake city Joined the 200k club
Denver hit 700k
Cincinnati is back over 300k
Akron grew

That's enough for now.

I think Detroit grew because Census back-upgraded its prior year estimates, as it's been known to do. Detroit's population actually fell from this year's 2017 to 2016 estimates.

Indianapolis fell behind Fort Worth this year to end up out of the top 15.

[cinyc]

Doesn't NY have a redistricting commission for 2022?

The commission is toothless, and can be ignored about as easily as the one in Ohio.

The commission is not as toothless as you think, because of the supermajority requirement to reject the commission’s map when there is single party government rule.

As for what’s losing population in New York, yes, it’s largely Upstate, but IIRC, some NYC boroughs lost population in the last estimate, too. I don’t think the math works to get rid of 2 Upstate seats if NY loses 2 seats. NYC will lose one, too.

[cinyc]

The drop in population growth nationally really seems to have hit a lot of the larger counties. Of the 10 most populous counties, all 10 had positive growth between 2010-2017 and 3 of them lost population between 2017-18. Only Maricopa is still seeing strong growth, even Harris county which grew by 1.9% p.a. in 2010-2017 grew by only 0.7% in 2017-18. Miami-Dade which grew by 1.3% p.a. in 2010-2017 grew by only 0.6%. Dallas county after growing by 1.5% p.a. in 2010-2017 slowed down to 0.6%.

As fertility continues to fall in America and deaths rise as the population ages I expect growth to slow down even further.

One year trends are always subject to revision.

The exurbs look like they're growing again. Even the Los Angeles Metro lost population 2017-18 - but San Bernardino/Riverside grew.

[cinyc]

Brooklyn, Queens, and Bronx all peaked in 2016 after many years of growth and have had small but significant declines since then. Immigration crackdown? I'm really surprised to see Brooklyn losing population.

It is due to less reported foreign immigration. Whether that is due to a different way of asking the foreign migration question in the ACS (instead of in what year did you last live outside of Brooklyn, did you live in Brooklyn last year, or something like that), illegal immigrants being less likely to fill out the ACS in the age of Trump, or an actual halt in immigration is unclear. That probably won't be answered until after the 2020 Census.

[cinyc]

Any press release yet on what date this month the 2018 city and town (or "county subdivision") estimates will be released (with likely two days of embargoed media access - they'll list both dates in the press release)?

They usually come out the Thursday before Memorial Day.

Edited to add: Census’ most recent tip sheet confirms this. They are scheduled for release on Thursday, May 23 with the media getting them on May 21.

[cinyc]

Here’s the new interactive map with the 18 vintage data. Default is percentage change of 2018 vs the 2010 Estimates base. To change maps, click on 18 Est % and select a year. 2010C-18 is vs. actual census numbers, which doesn’t take into account annexations and the like.

2018 Est # show numerical changes. The prior vintage estimates are in 17/16 %# – though they’re largely irrelevant now except to track changes in how Census estimated.

There’s also an isolate button on the main menu if you want to isolate a particular type of growth, like cities that increased by >20%.

https://cinycmaps.com/index.php/pop-change

Please let me know if you encounter any errors.

[cinyc]

This is a work in progress in the beta testing stage, but here's an interactive map of the percentage/numerical change in total population by census tract. There are submaps for 2010-"15", 2000-"15" and 2000-10 (click on Tot Est % or Tot Est #, then the year).

The "2015" data is from the 2013-17 American Community Survey (2015 is the midpoint of that data); the 2000 estimates were based on percentage land area from the 2000-10 tract relationship tables.

https://cinycmaps.com/index.php/population-change/tract-population-change

Red is an increase; blue is a decrease.

Let me know if you encounter any errors or something that looks just plain wrong. Like I said, this is in beta.

Jimrtex - would using percentage population from the relationship tables yield a more accurate result? I'm ultimately going to estimate the changes in racial data, too.

[cinyc]

What minimum population threshold would you suggest? 10? 25? 50? 100?

My other question is this: The 2000-10 Census Relationship file compares the 2000 census tracts to fit into the 2010 a number of ways. One way is by comparing percentage of land area in a given tract, AREALANDPCT00PT. Another is by comparing population, POPPCT00. Which should I be using when guestimating 2000 in the 2010 tracts?

So far in the beta, I’ve been using land area. Does Pop make more sense? Eventually, I’m going to guestimate change in Hispanic and Non-Hispanic White, etc. populations, too. Neither will be perfect there.

[cinyc]

POPPCT00 would show the percentage of the 2010 population that was in a particular 2000 Census Tract???

It shows the PCT Of 2000 population in the new 2010 census tract. I've changed the maps to be based on that instead of land area.

The interactive maps are now all there at the above link - though I might change one thing:

Maps under the “Tot Pop” tab show change over time in Total Pop, “NH Wh” change over time in the Non-Hispanic White population, etc. Within those tabs, the default maps set a minimum tract population size to get colored – it’s currently 100 for Total Pop and 50 of the type for races/Hispanic origin. I will probably lower these to 70/35 to solve a problem in Loving County, TX over the next few days.

Within each of those tabs, the maps are sorted by “PCT” Change or “NUM”erical change, then by year. “Min” maps have the minimum threshold; “All” maps color all tracts without the 100/50 min threshold.

The 10th tab, Pop PCT, contains maps of the overall percentage share of each race. The “Top” maps show the top race in each tract & PCT for the 8 racial/Hispanic groups. The “Race PCT” maps map the percentage of the selected race in each tract. The “Rel CHG” maps map the change from one cycle to another. Unfortunately, I was only able to map NH White, Hispanic, NH Black and NH Asian in the Rel CHG maps due to the size of the data file.

Let me know if you encounter any errors. If some maps aren’t appearing, you might have to clear your browser cache.

[cinyc]

The 2018 1-year ACS came out last Thursday. The top 10 CDs by population as of 7/1/18 are:

MT-AL 1,062,305
DE-AL 967,171
TX-22 935,386
ID-01 912,950
FL-09 902,812
TX-03 899,784
TX-10 896,798
TX-26 894,192
TX-31 883,347
SD-AL 882,235

RI-01 and 02 are the smallest, followed by WV-03 and WY-AL. RI and WV will likely lose a seat after 2020. MT was estimated to be larger than RI as of 7/1/18 - but we knew that last December.

I'm working on an interactive CD mapping tool. It currently shows estimated population change, racial and ancestry characteristics of each CD. I will eventually add election maps. They're on my website, here.

[cinyc]

Great job on the website cinyc!

I was looking at the population change map for Norfolk, Va.   I saw what I thought must have been a mistake for Census tract 38, a 87% decrease in a nice area that's been pretty stable. Looked around a bit and found this:
https://data.census.gov/cedsci/table?q=virginia%20population&hidePreview=true&table=P1&tid=DECENNIALSF12010.P1&g=0400000US51_1400000US51710003800&vintage=2018&layer=censustract&cid=DP05_0001E&lastDisplayedRow=15
has a note with the revised count and mentions the 2010 Census Count Question Resolution - which is where cities and towns can question what the count was, and the census makes a revision.   I don't know if there is data anywhere that incorporates all the revised counts from this.

Yeah - there will be errors in the 2010 Census data that have since been corrected. I downloaded the data from the actual 2010 Census, so the CCQR program has not been taken into account (it probably has in some of my incorporated place pop change maps of using Census' population estimates program maps where I used the 10 Estimates Base, instead of 10 Census).

There's also a known error in some Ancestry tracts that I need to fix because I used the 10 Census CBS 500k tract map, and Census changed the GEOIDs after 2010 to relect mistakes. The 2013-17 ACS uses the corrected GEOIDs. I made all the necessary changes for the Pop Change/Racial CD maps, but not the Ancestry CD maps - yet.

The other minor issue is that the legend/popup says Votes instead of Population/NH Whites/Hispanics, etc. in some places. That's a legacy coding issue (the code was written for election maps) that's on my to-do list to fix- along with allowing you to choose what minimum pop/votes you want to grey out. The latter is going to be a bit more difficult to implement, since it involves math.

The final issue is that I didn't break down tracts that are split between CDs, so the total pop will appear in both. It's not clear to me how to fix that without spending way more time than I'd want to.

[cinyc]

I don't know if there is data anywhere that incorporates all the revised counts from this.

I found Census' 2010 errata notes.The reason for the errors in Norfolk (and San Diego, Groton, CT, Portsmouth, NH, Pascagoula, MS & Everett, WA) was... a systematic mistake in determining the placement of Navy vessels.

I'll put a fix on my to-do list.

[cinyc]

The mid-2019 estimates will be released on December 19.

Link

When is the 2014-18 5-year ACS coming out? The following Thursday?

[cinyc]

The mid-2019 estimates will be released on December 19.

Link

The link now says December 30. The 2014-18 5-year ACS will be released on the 19th.

[cinyc]

The 7/1/19 state estimates weren't released at midnight, best I can tell. The link still says December 30, so I suspect they'll drop in the morning.

[cinyc]

The Northeast region as a whole lost population last year.    10 states lost population too,  one higher than last year. 

New York lost almost 77k people, does that mean it'll lose two congressional districts?

Its looking extremely likely at this point.

And one could make the argument that both seats should come at the expense of Upstate New York.

How does that math work when NYC + LI + Westchester have been growing at half the rate of the country, if even that much? I would think that they can no longer support more than 15 districts.

It doesn’t.

[cinyc]

The Northeast region as a whole lost population last year.    10 states lost population too,  one higher than last year. 

New York lost almost 77k people, does that mean it'll lose two congressional districts?

Its looking extremely likely at this point.

And one could make the argument that both seats should come at the expense of Upstate New York.

The math doesn’t add up for two upstate losses. The more likely result is that one seat upstate is lost along with one Long Island/East NYC seat.

The math might not add up to 2 lost seats at all. NY-26 is literally on the bubble in many of the analyses I've seen.

I'd bet that it will lose 2, though.

[cinyc]

DKE came out with its new reappointment prediction:
https://mobile.twitter.com/DKElections/status/1211721512106901504
As Cinyc points out, NY’s 26th seat is on the cusp of being lost, though this is better than the seat’s previous status as being lost.

New York's problem is that its rate of growth keeps on dropping like a rock. It's almost in a linear nosedive at this point:

https://twitter.com/cinyc9/status/1211734191253139457

[cinyc]

Maps of the relative state change in population vs the nationwide average (a.k.a. the bear). Red = you beat the bear and may gain a seat or two in redistricting. Blue = you lose to the bear and the bear may eat you and one or more of your seats.

All you really need to do to avoid the bear is outrun your peers, though, not necessarily the bear.

https://twitter.com/cinyc9/status/1211719646837137408

[cinyc]

VA is interesting.

They used to grow faster than the US, but not any longer.

Why is that ?

Are they going the way of CA with skyrocketing property/rent prices that are keeping Americans out of the state (immigrants are still moving there) ?

DC's growth rate has been trending downwards, too (though it's still net positive compared to the US). It's probably due to the size of the government workforce in the age of Trump.