Some key highlights:
- 55% of homes in the park are owner-occupied
- The rest are second homes or rentals
- Adirondacks’ population density is the lowest in the East
- Adirondack Park population is down 5% since 2010
By Tim Rowland
New 2020 census numbers point out the difficulties in finding housing for year-round residents and a decline of nearly 7,000 residents in the Adirondack Park.
The U.S. Census Bureau data show nearly half the houses in the park are now vacation homes, short- or long-term rentals or otherwise unavailable to people seeking residences.
In some nooks and crannies of the park there are census tracts where four out of five houses are “unoccupied,” which is to say they are dedicated to vacationers or part-time residents.
The Adirondack Explorer thanks its advertising partners. Become one of them.
Adirondack Park Agency Commissioners got a peek at the data points during a presentation by mapping technologists John Barge and Molly Jordan at the agency’s February meeting. Some of the 2020 data are available for perusal by board members and the general public in the APA’s online map room.
Statistics related to population and housing can be found by visiting apa.ny.gov and selecting Adirondack Maps and GIS under the Popular Topics heading.
The trends found are not particularly new, nor would they surprise local governments and nonprofit advocacy groups that have been working to solve some of the problems reflected, including affordable housing, child care and an aging population, with fewer young families that support schools and add to community vibrancy.
Overall, the park population was down 5% since 2010 from 130,137 to 123,316. Barge said that for the first time planners have an exact count of park residents, due to the elimination of Census districts that straddled the Blue Line. Prior, planners had to guess at the number of people in these border districts who lived in the park. “To us it’s really exciting that the population no longer has to be an estimate,” he said.
The Adirondack Explorer thanks its advertising partners. Become one of them.
Barge also cautioned about making sweeping assumptions based on percentages in jurisdictions where the population is so small — the park’s smallest town, Morehouse, has 90 residents — and the simple gain or loss of one or two families could look severe in terms of percentages.
There are other census vagaries as well, including in North Elba, where an apparent 16.5% plunge in population has been tracked to 1,000 federal prisoners at FCI Ray Brook, who for unknown reasons, seemingly vanished from the 2020 count.
Housing issues
The 2020 census count also preceded the dramatic demographic changes caused by the pandemic, which touched off a wave of property transactions, as wealthier newcomers bought second homes or moved to the park to escape tight city confines. Essex County, for example, Dave Wainwright, director of real property, told supervisors last month that the pandemic had driven property transfers to record levels, and that the market was still going strong.
At the same time, it became more profitable for landlords to rent to short-term vacationers than to working class residents, according to supervisors across the North Country, who are considering the prospects of establishing land banks to get affordable homes into the hands of local residents.
The Adirondack Explorer thanks its advertising partners. Become one of them.
All this has driven up the price of housing beyond the means of local residents with local jobs, many of which are in the relatively low-paying service sector. Dan Kelleher, APA special assistant for economic affairs, said surveys indicate that new housing is beyond the budgets of 80% of Adirondack residents.
Even those who have the means may have trouble finding available housing, due to the lack of buildable land. More than half of the park population lives in its 100 hamlets because of the large percentage of state-owned land and tighter regulations outside population centers. APA Commissioner Jerry Delaney said growth is further restricted by easements on private land, which don’t show up on maps of state-owned tracts.
And of the existing housing stock, just 55% is occupied by Adirondack residents, the rest being rentals or second homes. Barge said that’s not necessarily unusual for tourist destinations, citing high numbers of “vacant” homes in places such as Ocean City, N.J., and Breckenridge, Colo.
But someone looking for a home in Hamilton County — along with Essex, one of two counties entirely in the park — would be up against statistics showing that only 32 percent of the houses are locally occupied.
The Adirondack Explorer thanks its advertising partners. Become one of them.
Even in Essex County, with its working class hamlets of Ticonderoga, Port Henry and Au Sable Forks, more than a third of the homes are effectively off limits to local buyers.
Lowest population density in the East
The numbers also show a general dearth of children in the Adirondacks. While nationally, a quarter of the population is under the age of 18, the number of children in the Adirondacks is 17%. In parts of Hamilton County, nearly 90% of the population is adult.
Hamilton was virtually the only part of the park where population increased (by 5.6%) from a decade ago. Put together, these numbers appear to indicate that the population increase is skewed more toward retirees.
But again, these percentages have to be couched by the region’s low population; expressed in numbers, the county gained 271 people over the past decade.
The Adirondacks’ population density is the lowest in the East, and Barge said when hunting for similar sparsity he didn’t find it until he got to West Texas.
Sad statistics on affordable housing. Trends will likely continue until there are sufficient incentives for developers to build affordable housing instead of or addition to McMansions for the wealthy. Perhaps counties requiring 3 affordable units built for every luxury residence built before an application is granted.
Housing and people are sparse…and that’s the way we like it. We know our way to Syracuse but most of us don’t go any more often than we have to, so don’t try to draw the masses here. We neither want nor need them. You know what else is sparse here? Roads and cars. Many towns have no traffic lights. I graduated from the local k-12 central school in the largest graduating class ever. 42 kids graduated that year. I knew all of them, many of their parents and some of their grandparents. Life is good.
Well said.
Yes we need “affordable housing”, definitely section eight, it will bring in the best and brightest people to contribute.
How about all the people who believe in “affordable housing” donate theirs
so these people have a cheap place to live.
“The numbers also show a general dearth of children in the Adirondacks. While nationally, a quarter of the population is under the age of 18, the number of children in the Adirondacks is 17%. In parts of Hamilton County, nearly 90% of the population is adult.”
Yea if you include places like the Harlem or the Bronx you would have a higher number of children.
This is a filler article that does not provide any real data.
Plenty of data there. Even more if you bother to click on the links to APA data!
I watched the meeting on Webex specifically to watch the population count presentation. Barge’s methodology for estimating Park population is very innovative: many census blocks that are adjacent to the Blue Line –the lowest level of granularity in the public census dataset–are divided into halves by it. To deal with this, my understanding is that Barge has been using NYS Parcel Centroid Data since at least the 2010 Census, calculating the percentage of occupied units inside of the Blue Line versus outside of it based on these centroids and then dividing up the population tabulations for those blocks using the resulting proportions.
Although I don’t know the extent to which the total population count has been affected, I have serious questions as to the veracity of this methodology with the latest census data, which utilizes a new method of obfuscation to protect privacy known as differential privacy. Experts have raised serious concerns that this new technique will cause important calculations that use the data to be inaccurate (https://apnews.com/article/business-census-2020-technology-e701e313e841674be6396321343b7e49); and this in fact has happened, as populations in certain blocks have been demonstrably and grossly misrepresented in the data (https://apnews.com/article/religion-wisconsin-new-york-tampa-florida-68c96e7eb701da74ae7c8df3c3476705). Even Dr. Ron Jarmin, the acting director of the Census Bureau, himself warns against relying upon the granularity of census block data, such as in the way outlined above: “Noise in the block-level data will require a shift in how some data users typically approach using these census data. Instead of looking for precision in an individual block, we strongly encourage data users to aggregate, or group, blocks together.” (https://www.census.gov/newsroom/blogs/director/2021/07/redistricting-data.html)
We at least need to look towards this latest population data with skepticism, even when only considering it as a snapshot of population on the census reference data of April 1st, 2020. Better yet, an in depth statistical analysis of population changes in these contested blocks should be performed to look for red flags (I have a few ideas…). And ideally–if APA and NYS are going to utilize this data in shaping policy–the Census Bureau, since they are unlikely to change all blocks to exactly follow the Blue Line, should probably be providing APA with a specially tailored dataset for internal use by APA.
All of this highlights the huge challenges faced by the APA GIS staff and their hard work to constantly improve the tools needed by the APA to carry out their mission. It is possible that these concerns may be overblown on my part, and that Barge’s experience with the data may give him good reason to be more confident. But a few remarks that he made has led me to believe that he himself noticed some discrepancies vis a vis the peripheral blocks. Lastly, Jerry Delaney’s comment on overlooked private easement lands I found particularly insightful; the PROP_CLASS (9**) of the NYS Parcel Centroid Data may provide for an easy way of gaining more insight into this.
Kierin, Thanks for your thoughtful comments. Perhaps I didn’t make myself clear in my presentation. I used my methodology applying the ratio of residential parcel centroids for blocks straddling the park boundary in 2000 and 2010.
Due to advocacy of the APA, the Governor’s office, and others in NYS demographic and economic work, the US Census Bureau DID AGREE to break all census blocks at the park boundary for the 2020 Decennial Census and for future census enumeration. This is significant for park analysis!
I just added all NYS census blocks to my Park Census application referenced in the article. Previously it only had the in-park blocks. Now, if you zoom in to the neighborhood level along the edge of the park, you can see how the blocks no longer straddle the park boundary. Also, click on any of these blocks and the pop-up will show aggregate population, housing, and race information with the adjacent blocks.
– Regards, John Barge
John Barge, I’m glad my comment made its way to you. Thanks for the additional information! Not needing to perform interpolation on peripheral census blocks, along with a baked-in ability for the public to easily analyse aggregate populations of adjacent blocks, is huge! (That was definitely part of the confusion.) However, I did not explain my full concern very well in my initial comment. My real question relates to the extent of the noisiness introduced by differential privacy–more the effects of splitting towns rather than those of “splitting” blocks. Since total demographics for towns split by the Blue Line are essentially tabulations for ad hoc aggregations of blocks–not officially recognized and targeted by the Bureau’s algorithms as “noise-proof” (as far as I know)–what is the probability that splitting towns produces inaccuracies? …I.e., for towns straddling the Blue Line, could noise from blocks on the outside of the Blue Line affect blocks, even distant ones, on the inside?
…At first glance, it doesn’t look like any red flags are jumping out. …But, in the absence of an official specification on differential privacy (I have never heard about one), some geospatial analysis to test these effects may be worthwhile. …Even if it may be smaller issue in light of the whole picture.
The exchange between Mr Bell and Barge got a bit technical.
Could one or both of you explain the implications of your comments to ordinary folks living ordinary lives in the Adirondacks ?
The article says that”nearly half the houses in the park are . . .unavailable to people seeking residences.” This is based on the statistic that 55 %% of the housing units are owner-occupied. Surely those who rent are Adirondackers! In Essex County about a quarter of occupied units are rentals, about 15% of the total housing stock. Units occupied by people whose usual residence is elsewhere are considered “vacant” by the Census. There is a very high vacancy rate in the Park, which includes units available for rent or sale, under construction or repair, abandoned (but not clearly destined for demolition), or held off the market for legal reasons, etc.
Another insidious product of the Census Bureau which adversely impacts residents of the Adirondack Park and other small marginalized communities involves the Median Household Income (MHI) data sets illustrated through the American Communities Survey(ACS). The ACS data sets are compiled by the census as a statistically valid indication of community household wealth utilized by state and federal agencies in awarding grant and loan packages for water and sewer infrastructure. The data sets are broken down by Municipality so every town, village, city has an MHI based not on a complete data base of all the residents but by statistical samples from a very small population (>10%) in the relevant community. Small Towns in the AP where seasonal residents and waterfront properties are located project a much higher MHI than what exists in the hamlets where water and sewer infrastructure exists. Meanwhile the entire towns’ MHI is used to determine grant and loan packages based on the ACS data not the real income of those within the improvement district. Federal agencies such as the Dept. Of Agriculture’s’ Farmers Home Administration and State Agencies of the Department of Health and NYS DEC by deferment to the Environmental Facilities Corporation (EFC) annually administer Clean Water and Drinking Water Revolving Funds supporting infrastructure through the annually modified Intended Use Plan (IUP).
The population dynamics in the AP are working to marginalize communities in the Catskill and Adirondack communities that must compete with other communities with much larger populations, more resilient economies of scale and without similar geologic constraints. The EFC’s threshold for small communities incorporates populations below 300,000. The vast majority of AP water and sewer districts have under 300 users and many have under 150, some are under 50. Construction of water and sewer infrastructure requires blasting of bedrock making for more costly installations. Many hamlets were founded along water courses the historically significant highways of pre-industrialization and the hamlets are concentrated in those areas with small lot sizes high groundwater and sparse populations making development of infrastructure less affordable.
This subject is more complex than noted here and requires systemic review and improvement. One such avenue would provide additional priority on the IUP for those communities inside the respective Catskill and Adirondack Park as an investment by the state in Clean Water infrastructure and the communities which struggle to be caretakers and stewards of the garden.
can u add the population uv the eastern adirondacks?