So the New York City region has the biggest collection of nighttime lights in the world, and that’s where I would go to find people. How many people will live in New York City itself in 2081? That is, after two devastating wars in the mid-21st century (fictional, of course)… I projected numbers from some experts and applied some adjustments from history to arrive at a very interesting, but somewhat problematic number.
Disclaimer: this is a back-of-the-serviette (slightly larger than a napkin) cave-girl exercise to flesh out the world of my novel (“SFT” is my very imaginative working title). What follows is not academically rigorous and should not be used to make investment decisions.
We begin with a bit of background. According to the 2010 United States Census (the next one’s due in 2020), the population of New York City is:
The 2013 estimate* is:
It’s the most populous place in the United States. And this doesn’t count tourists. Here’s a shot of a nice interactive map by the US Census Bureau:
New York City is made up of five counties (also called boroughs, essentially): New York (Manhattan), Kings (Brooklyn), Queens, Bronx and Richmond (Staten Island) counties. In the map above, I highlight New York County, population 1.6 million, because as was obvious from “Chapter 1”, that’s one of the places in which I’m most interested – Manhattan. More on why below.
Plain Vanilla Projection
First, I used existing studies to get expert views as far into the future as I could. I picked this September 2011 Cornell study because it has population projections by county (borough) all the way out to 2040. The New York Department of Labor uses this study, too. In 2040, the Cornell team expects New York City to have a population of:
Using compound annual growth rate formulas (the ones in this land use planning software manual were handy), I got to a growth rate of 0.44 percent for the total of the five New York City boroughs. It’s the annual growth rate backed out from the study’s data from 1990 to 2040. Then using this rate, I projected the population of New York from 2040 out to 2047:
I used the average percentage distribution by borough of the Cornell data from 1990 to 2040 for my total population in 2047 and got this:
Census 2010 puts New York County (Manhattan) at 1.6 million people. By my reckoning, if all goes well, Manhattan will have 1.7 million inhabitants in 2047. So far, so good.
But why 2047? Because big things start happening in 2048 (pure coincidence***), at least in the world of SFT.
Assumptions from History
There is a massive global war in 2048 and then another one in 2050 that lasts for several years. To adjust my forecasts for devastation and destruction, I look to the past: World War II. To cities where bombings happened, because in the world of SFT, the Wars were everywhere – even in New York City.
One night of bombing in May 1945 led Nordhausen and Pforzheim, Germany to lose 20% and 22% of their respective populations. I put a 21% loss factor for the 2048 war.
In Tokyo, more than 100,000 people died after one year of relentless bombing from 1944 to 1945. The city lost almost 60% of its population in only one year (death and migration):
The 35 wards of Tokyo had a population of 6.78 million in October 1940, and the figure fell only slightly in the midst of the war, to 6.56 million in early 1944. But by November 1945, two and a half months after the surrender, only 2.77 million people remained in the city; almost 60 percent of its prewar population had been bombed out or fled (Tokyo-to 1979, 6:69)” (Bestor 72. Boldface type emphasis mine.)
This checks out with official population figures. I give a 58% loss factor for the 2050-52 war.
I would have looked more closely at refugee migration patterns, but during the SFT world wars, there were no definitively safe places. Zero. Other cities and countries faced the same problems. And transportation systems were crippled.
Survivors who could afford it or who were employed stayed relatively local and couldn’t even dream of going to another state. They would get killed by Creatures. So whoever got through the Wars went to nearby places with the most forcefield-protected areas (more on these later).
The best such protected area in New York City was Manhattan. Why Manhattan? It:
- Is the second-richest of the five boroughs after Richmond by median income (US Census 2010. $67,000 versus Richmond’s $72,000. Nicely visualized in this JusticeMap.org interactive map)
- Has 15 of 19 of the richest neighborhoods in the US located in New York City (a very cool list by Professor Higley called the Higley 1000) – this is a win over Richmond (Staten Island). Sorry, Staten Island.
- Is home to 45% of firms located in New York City (Kings has 23%, Queens 20%), half of NYC’s biggest firms (more than 500 employees), and pays almost 80% of the NYC’s total payrolls (Calculations from US Census 2011 County Business Patterns).
A note on a neighbor: What about New Jersey? Maybe some of the 58% of the lost population didn’t die and went to New Jersey instead. We will never know.
So immediately after the SFT Wars, the population of New York was:
And it was distributed in 2052 as shown in the table.
Yup. Approximately a third of New York City’s current population, mostly concentrated in Manhattan. Immediately after some sad, scary and super destructive wars (and imagine the same or similar happening all over the world).
The Population of New York City in 2081
But the human race is mighty scrappy and, of course, figures out how to get on with things. So I can take these figures and project them all the way to 2081 – when SF takes place. Again, I look to history as a guide.
In Tokyo, the annual population growth rate in the 5 years after World War II was almost double that of the five year period before the war (see below). People came back, people had children. Annual growth rates declined by about 60% and 25% in the two 5-year periods afterwards, then later tapered off to some declining steady state.
I use the same pattern of population growth rates for SFT post 2050 war and arrive at my forecasts for New York City in 2081:
Population distribution within the city is as follows:
That blasts SFT New York City population back by almost 3 centuries to somewhere around 1905: in 900, the population was 3.3 million and in 1910, it was 4.7 million. Department of City Planning, City of New York, “NYC Total and Foreign-born Population 1790 – 2000“.
And so happily starts the novel in 2081 with 3.80 million folks living in New York City. (Technically, “Chapter 2” happily starts: “Chapter 1” takes place a few years before 2081.)
The Problem Part
But what was that wee problematic thing to which I referred above?
Population density. 3.45 million people living on Manhattan Island is more than twice as many people as there were in 2010 (1.6 million). (We can forget about tourism in SFT 2081.) This poses a challenge as to where they all live, work and play – especially since much of Manhattan will look like this in SFT’s 2081:
See image source here.
Compressed housing, office and leisure space for 3.45 million people would require, for example, 7,600 London Shards standing 73 stories tall (assuming 4 people for each of the 10 apartments and 2 people in each of the 202 hotel rooms). What about a taller tower? That many people would need 768 Burj Khalifas at more than 160 stories tall to house (assuming that the 900 apartments house five people each and that the 160 hotel rooms house 4).
There is a lot cooking today in the supertall building space, so not to worry. Supertall building technology will take care of our grand- and great-grandchildren (at least the more fortunate ones) in the SFT future.
I leave below a line-and-numbers overview of this post with colours and speech bubbles. Would love to hear any thoughts you might have on the method and forecast.
Some Graphs: Overview of SFT Population Projections to 2081
Y-axis scales are the same for both graphs to facilitate comparison. Click to enlarge.
Notes and Sources
* On US census figures and estimates. The United States Census Bureau conducts a detailed population census every 10 years – they interview and take information directly from people living in the United States. In the years in between censuses, population figures are estimates – they use models to arrive at population figures.
** What’s the difference between a projection and a forecast?
For a projection, trends from the past continue into the future. You decide on mechanical things, such as time horizon, whether to use mean or geometric mean, whether to use linear growth rates or exponential growth rates. For a forecast, you make more involved decisions – you change or make some big assumptions based on what you expect from events in the future.
In the SFT case, population grew by 0.44% per year to 2047 (projection). I wiped out almost 60% of New York City in 2050-52 because there was a war (forecast).
I like this distinction for cognitive neatness, but some may not agree with it. There is a nice step-by-step explanation of this difference and how land use planners forecast populations in that land use planning software manual.
*** A very stimulating and addictive game. Play it. Screenshot courtesy of Samsung Galaxy Note 3.
A Real Book I Cited (but Found on Google Books, So Maybe It Doesn’t Count So Much)
Bestor, Theodore C. Neighborhood Tokyo. Stanford, California: Stanford University Press, 1989.