Last August I downloaded population and earnings data from the U.S. Bureau of Economic Analysis, from its Local Area Personal Income series, to use in my compilation of state and local government employment per 100,000 people. The data was for 1997, 2007, and 2017. As always I divided the state into four regions. New York City, whose population I got by adding up the five boroughs. The Downstate Suburbs, which I got by adding up Nassau, Putman, Rockland, Suffolk and Westchester Counties. The Upstate Urban Counties, the sum of Albany, Broome, Dutchess, Erie, Monroe, Niagara, Oneida, Onondaga, Orange, Rensselaer, Saratoga and Schenectady Counties. And the rural and small Rest of New York State, which I got by subtracting the other three areas from the state total. The data showed a big population drop for this part of the state from 2007 to 2017 – and a thus huge increase in local government employment per 100,000 people.
Local Area Personal Income data has been updated to 2018 recently.
And I started downloading it for possible use in another analysis. New York State’s 2017 population was exactly the same as the estimate released a year earlier. But New York City’s 2017 population was slashed by 184,427 (2.1%), with smaller decreases for the Downstate Suburbs and Upstate Urban Counties. Which means that since the Rest of New York State was obtained by subtraction, its population 2017 had soared by 247,319, a full 10.3% increase! Despite the fact that the 2017 population estimate for virtually every individual county in the Rest of New York State has gone down! It isn’t a surprise that the numbers are different. Numbers are revised all the time based on new information. But changes of this magnitude, despite NO change in the state total?
The best case scenario is a screw up. Which is pretty much what I believe about next year’s 2020 Census of Population.
A spreadsheet with the new and old population figures for 2017 is here.
Obviously, I’m not going to go back and redo all the work I did to produce the analysis of state and local government employment I just finished, which took a huge amount of my spare time over a couple of months. For what it’s worth, however, I did produce a revised version of the spreadsheet for all areas for March 2017.
And a revised spreadsheet showing the 1997, 2007 and 2017 figures for the U.S., different areas of New York State, New Jersey and Fairfield County, Connecticut.
The difference in local government employment per 100,000 area residents is very small for New York City, the Upstate Urban Counties, and the Downstate Suburbs. They all have slightly more employment per 100,000 residents than I reported.
But for the Rest of New York State as a whole, the revised population data puts 2017 employment per 100,000 residents lower by a significant 10.0%.
And yet for any individual county in the Rest of New York State, the employment figures per 100,000 residents are slightly higher, because the 2017 population estimate for all these areas was cut, with the exception of tiny increases in Cortland, Delaware, Greene, Lewis and Tioga Counties.
Which figures are right? What if neither of them are? A quick addition shows that the 2017 figures for individual New York counties downloaded in August added up to more than the state total, whereas the figures downloaded in November add up to the state total.
The 2007 figures did not change. The U.S. figures did not change. All the 2017 figures for individual counties of New York State changed, but the state total did not.
Moreover, for New York City the new plan seemed to be to just cut the 2017 estimate by 2.0% to 2.6% in every borough save Staten Island. Or the old plan was to inflate the population by that amount.
And the updated figures showing only a small population decline in rural Upstate NY are now in conflict with administrative data showing a massive decrease in the number of public school children, and data on employment decreases.
This doesn’t raise my slowly declining confidence in government statistics. Between defunding, chaotic management at the top of the federal government. And a growing bipartisan trend toward deceit as things get worse for most people in later born generations while those older, richer and better connected continue to cash in. The possibility of errors – or something worse – grows.
Under Baby Boomer leadership this country’s future is Argentina’s present.
This is a generation for which what matters is getting whatever they want right now whatever way they can, regardless of the effect on other people. And they want anything the calls the fairness of this into question to go away, so they don’t have to feel bad about it. Something has to change, and fast.
What’s next? Revised data “proving” that New York City’s unionized public employees owe the rest of us nothing because they are underpaid, and revised data showing the economy grew 4.0% per year under President Trump, double the previous estimates? Data showing the Millennials are paid 25.0% more than Baby Boomers had been at the same age, rather than 25.0% less, and rising life expectancy for those under age 60?
I could see the 2020 census of population being rejected as false, and not used for reapportionment and redistricting, something that previously happened exactly a century ago.