History, Nature, Art & Good Food In The Hudson River Valley

In the second week of July my wife and I spent 5 days on vacation in the Hudson River Valley.   It is a place several friends and family members have visited and recommended and it is reachable from our home in Baltimore in a 4 – 5 hour drive.   Our primary interest was in visiting Franklin Roosevelt’s home, museum and library in Hyde Park, NY near Poughkeepsie but there are a broad range of attractions and accommodations on both sides of the River between Westchester County north of New York City and Albany.

We found very attractive accommodations on the west side of the river in Milton, NY at the Buttermilk Falls Inn and Spa.  The Inn is located on a 75 acre site overlooking the Hudson bisected by a stream with a small waterfall and several ponds with ducks, geese and swans.    Accommodations include a main house dating to 1764 with 10 rooms and a number of houses and cottages.   Breakfast is included and there is a very nice farm-to-table restaurant on (Henry’s), as well as event space, including a great outdoor wedding venue overlooking the Hudson.  A farm and animals provide food for the restaurant and another diversion for guests.   There is an exercise room, indoor pool and spa.   We stayed in the Sage Right room in the main house, which comes with a queen bed, gas fireplace, refrigerator, patio with views of the Hudson and bath with combination whirlpool tube and shower.    The room was attractively furnished with antiques but a bit cluttered with limited closet and drawer space.    There was no way to control the air-conditioning temperature in the room and we ended up having to run the gas fireplace to maintain the room temperature as a reasonable level – nothing environmentally conscious in that.

Buttermilk Falls Inn From Our Patio

We ate at Henry’s, the on-site farm to table restaurant, our first night and liked it so well that we ended up having light suppers two additional nights during our stay.   Both the food and the wait staff at the restaurant were excellent and the menu offers lots of appetizer/small plate options as well as substantial entrees and different white, red and rose sangria nightly.  The owners of Buttermilk Falls also own a bakery and cafe, called Frieda’s, a few miles from the Inn on Milton’s main street.   It provides the baked goods for the Inn and Henry’s and also offers good breakfast, lunch and take away/picnic options.

We maintained an active but measured pace during our trip, blending visits to historic, natural, and art attractions and the Culinary Institute of America with time at the Inn for afternoon tea, reading and relaxation.   We spent more than half a day on our first full day visiting Franklin Roosevelt’s home, Springwood https://www.nps.gov/hofr/index.htm, and the adjoining Presidential Library and Museum https://fdrlibrary.org.    The house is large but surprisingly modest and comfortable compared to other Gilded Age mansions.   The library and museum, the first Presidential library, were designed by Roosevelt himself and have excellent exhibits chronicling Roosevelt’s life as well as housing his personal study and being a repository for Presidential archives.   The museum exhibits are very well designed and many are interactive.

Springwood

Day two we parked on the western approach to Walkway Over The Hudson and crossed the world’s longest elevated pedestrian bridge spanning 1.28 miles over the Hudson River http://walkway.org/visit/.    The bridge is a converted rail span built in the nineteenth century. The walkway is free of charge and provides great views up and down river with the east end landing in Poughkeepsie, which offers some restaurant options.    You can enter and exit the walkway at-grade on both sides of the  River and there are also elevator and stair options on the Poughkeepsie side but the elevator to the Poughkeepsie waterfront wasn’t working the day we visited.   Information panels along the walkway acquaint visitors with the River and the history of the bridge and the area.

Mid Hudson Bridge From Walkway Over The Hudson

Day three we drove south to Storm King Art Center, a 500 acre sculpture park located in Cornwall, NY https://stormking.org/about/.   Storm King offers a vast array of monumental and smaller sculpture on an attractive rolling site.   We very much enjoyed and were impressed by the art but believe Storm King should offer more tram service options to help visitors get around.   A tram circulates through the site but only about once an hour. We walked more than two miles and by no means saw all of the sculpture.   More frequent tram circulation and shuttles between parking, dining, and shop/museum locations so you can concentrate your walking to see the art would make Storm King much more accessible to visitors.    There is a bike rental option that you may want to try but we did not discover it until we were on our way back to our car.   If you visit, be prepared to walk and bring water and sun protections with you.

Zhang Huan – Three Legged Budda
Ronald Bladen – Untitled
Alexander Lieberman – Adonai
Alexander Calder – The Arch

Day four we returned to Hyde Park to tour Val-Kill https://www.nps.gov/elro/index.htm, Eleanor Roosevelt’s cottage home and we also visited Top Cottage, Franklin’s personal retreat https://www.nps.gov/hofr/planyourvisit/top-cottage.htm.  The Roosevelt home, Springwood, Val-Kill and Top Cottage are all administered by the National Park Service.   A visitor’s center and the Roosevelt Library and Museum adjoin Springwood but Val-Kill and Top Cottage are located on separate nearby sites.   You can drive yourself or take a shuttle bus to Val-KIll from the visitors center but Top Cottage is only reachable by a strenuous 1.5 mile hike from Val-KIll or by shuttle.  Val-KIll was acquired by the National Park Service at the time of the bicentennial and is dedicated to Eleanor Roosevelt personal accomplishments, not her role as First Lady.   Val-KIll offers attractive grounds, a small gift shop and welcome center, an orientation film about Eleanor’s life and a tour of several rooms in Val-Kill and the adjoining Stone Cottage, which also houses some exhibits.   We very much enjoyed our visit to Val-Kill but it’s offerings are much more modest than those of Springwood and the Roosevelt Library and Museum.

Val-Kill Stream and Pond

Top Cottage was designed by Franklin Roosevelt to be his retreat after completion of his second term and only saw limited use as he went on to serve a third and a portion of a fourth term as President.    It has almost no original furnishings and the volunteer docent who we toured with had only limited information to offer on the property.    Top Cottage is only open limited hours and should not be a high priority for a visit.    We got there by hiking a somewhat steep and rocky trail from Val-Kill but arrived in time to catch a tour and were able to return to Val-KIll on the shuttle.

Top Cottage – Rear Porch

Two nights during our visit to the Hudson River Valley we dined at restaurants operated on campus by the Culinary Institute  of America (CIA).   Advanced reservation, best made exactly 30 days in advance, are a must and can be done on OpenTable.com.   The CIA operates five restaurants, four of which are open for dinner – American Bounty with a focus on the seasons and products of the Hudson Valley, Bocuse a French restaurant named for the most famous chef in France, Paul Bocuse, Ristorante Caterina de’ Medici and Al Forno Trattoria offering authentic regional Italian cuisine and Post Road Brew House http://www.ciarestaurantgroup.com/new-york-restaurants/.    We tried both American Bounty and Bocuse but preferred Bocuse, which is a bit more upscale and where we had a table next to the glass enclosed kitchen.   A signature item at Bocuse is lavender ice cream made fresh at your table using liquid nitrogen to deliver hand-churned ice cream in only about five minutes.

Desserts Accompanying Ice Cream At Bocuse
Making Lavender Ice Cream at Bocuse
Pike At American Bounty
Duck at American Bounty
Dessert At American Bounty

There is a lot more to see and do in the Hudson River Valley including wineries, local farms, cute small towns, cruising the river and West Point but we intentionally did not try and squeeze too much in so we had time to relax and enjoy the picturesque setting as well as tour some sites.

 

Slow 80+ Pop Growth, Elevated Construction Spark Concern For Seniors Housing

Cited In Senior Care Investor Podcast

I am republishing this blog post, originally issued in December, 2015, because it was recently cited in a podcast by Steve Monroe, Editor of Senior Care Investor, a well regarded industry newsletter, as being prescient about predicting issues being faced by senior housing and care facilities today.

Background

I have been pessimistic about maintaining occupancy and pricing and the risk of overbuilding in private-pay seniors housing.   I shared these concerns, along with lessons learned from the last industry downturn, with the NIC Future Leaders Council at the annual conference of the National Investment Center for Seniors Housing and Care (NIC) in early October and will express similar views when I speak at the Senior Living 100 Conference in March.

Since I am more pessimistic about the risk of overbuilding than NIC MAP® Data Service and many industry professionals, I recently reviewed my assessment by examining the most recent census population projections to estimate demand and updated 3Q15 NIC-MAP information on supply. This blog summarizes the results of that review.

NIC MAP Assessment

NIC MAP data indicates a total supply of U.S. institutional quality private-pay seniors housing units (independent living, assisted living and memory care) as 1,404,000 units as of 4Q14.   It shows construction as a share of inventory for the top 99 markets as of the 3Q15 of 3.3% of existing majority IL supply and 7.9% of majority AL including memory care. If I apply these same shares to the inventory of seniors housing for the nation (1.404 million units), then I estimate that there may be 72,838 units under construction as of 3Q15.

NIC staff estimates that these 72,838 units will be delivered over a two-year period for average annual construction of approximately 36,400 units. This compares to peak construction levels of approximately 45,000 units in the late 1990s when the last significant overbuilding occurred.

NIC MAP’s statistics on demand and supply focus on two key items, % growth in the supply of private-pay seniors housing and the percent of the 75+ household population, or penetration rate, required to fill anticipated construction.  Comparing NIC MAP 4Q14 supply in the top 99 markets to the most recent U.S. Census 2015 population forecast for the entire U.S. 75 + population, NIC-MAP data shows a penetration rate for occupied private-pay seniors housing of 6.25% of the 75+ population in 2015 at a 90.05% occupancy level.  The 75+ population is projected to grow at a compound annual rate of 2.9% between 2016 and 2020 while the seniors housing supply is projected to increase by about 2.6% in 2016 if we assume that half of the units NIC MAP estimates are under construction as of 3Q15 are completed in each of the next two years.

The absolute growth in the entire U.S. 75+ population at a 2.9% annual rate is expected be nearly 626,000 annually.   At a 6.25% occupied penetration rate, this equates to demand for 39,125 new seniors housing units annually between 2016 and 2020 compared to annual unit growth from new construction according to NIC staff of 36,419 (72,838/2).   The market’s ability to absorb projected levels of new construction would appear even better on a net basis if obsolete units being removed from the market were to be deducted from the estimated growth in supply based solely on units under construction.   Using NIC MAP estimated supply growth rate (without any assumed demolition) the 75+ occupied penetration rate could actually decrease to 6.1% in 2020 while still keeping private-pay senior housing occupancy at the 90.1% level as of 3Q15 and filling projected development at its current rate to this same level of occupancy.

The key takeaways from this analysis of Census and NIC MAP data are:

  • Private-pay seniors housing construction levels in the US are elevated compared to recent years but below late 1990s peaks.
  • Demand is sufficient to accommodate current levels of construction because the 75+ population is growing at 2.9% annually between 2016 and 2020 vs. supply growth of about 2.6%.
  • Growth in the 75+ population between 2016 and 2020 will produce sufficient absolute growth in demand at a 6.25% penetration rate (39,125 units annually) to absorb projected seniors housing supply growth (36,125 units annually).
  • With the exception of some select markets, NIC MAP data indicates occupancy can be maintained without an increase, and even with a small decrease, in the 75+ occupied penetration rate of private-pay seniors housing.
  • Some older obsolete units will be removed from the market, further brightening the prospects for private pay seniors housing compared with estimates of supply growth based solely on units under construction.
  • Many industry leaders report little evidence of overbuilding in their markets.

Why I Am Concerned

I don’t dispute the NIC-MAP data factually or the view of many industry leaders but I believe they overlook three key items: (1) the increasing age of entry of new residents into private-pay seniors housing, (2) near-term growth in the senior population is concentrated in the “younger” 75 – 79 age group and (3) high turnover means newly constructed seniors housing is very competitive with the existing supply. These are the items that make me pessimistic about the near-term performance of private-pay seniors housing.

Increasing Age Of Entry – Different studies report different numbers for average age of senior housing residents and average entry age, but it is fair to say that in 2008/09 studies the average age of residents ranged from 82 in majority IL properties to 84 in majority AL properties and has moved higher.   Estimated entry ages for IL and AL are now closer to the mid-80s according to many operators.   This is important because much of the growth in the supply of private-pay seniors housing is in AL and Memory Care units that appeal to seniors over age 85, while much of the growth in the 75+ population will occur in the younger end of this age cohort.

Growth75+

Near-Term Growth Concentrated in Seniors Less Than 80 – The chart above shows projected population growth from the most recent projections of the US Census Bureau for the 75-79, 80-84 and 85+ age groups for the periods 2016–2020, 2021–2025 and 2026-2030.   Focusing on the 2016–2020 period you can see that growth is highest for the 75-79 age group, while much lower for seniors 80 and above.   As a result, when NIC MAP and others use a 75+ penetration rate it may overstate demand for private-pay seniors housing because residents are not moving in on average until 82 – 84 and perhaps 85 or higher for AL.

The chart below further refines population growth for seniors between age 80 and 87 to illustrate how dramatically growth is skewed toward seniors less that 85 between 2016 and 2020.

Growth 80 - 87

Near-Term Outlook Looks Worse On 80+ Penetration Rate – If we look at private-pay seniors housing penetration rates for the 80+ rather than 75+, the 4Q14 penetration rate for occupied units is 10.1% at the national level.   Annual projected demand between 2016 and 2020 for the entire 80+ population at this penetration rate is only 23,123 units, compared to current construction levels of 36,400 units per year and the 80+ penetration rate would have to rise to 10.8% in order to maintain senior housing occupancy and accommodate unit growth at current levels to 2020. (This analysis assumes that the rate of construction as a share of inventory exhibited currently for the 99 markets is the same for the non-99 markets as well.)   Slow growth in the 80 – 87 age group most likely to move into private-pay seniors housing (particularly in the 85+ age group) and the need for a significant increase in the 80+ penetration rate in order to maintain current occupancy levels raise concern about the industry’s ability to maintain private-pay seniors housing occupancy and rate and accommodate new unit growth near term, even if we assume some reduction in the supply as obsolete units are removed from the market.

Turnover – Data for YE2014 as reported in ASHA’s The State of Seniors Housing 2015 shows turnover rates of 26.2% for majority IL properties and 51.6% for majority AL properties for a weighted average of 36.5%.   With a total private-pay seniors housing supply of 1.404 million units and a 90.05% occupancy level, this means that 462,000 units need to be filled annually just to maintain current occupancy.    These relatively high rates of turnover, particularly for AL properties, mean that the existing stock of private-pay seniors housing is constantly competing with any newly constructed units and any degree of overbuilding is likely to quickly put pressure on occupancy and pricing in the existing stock, in my view.

When Will Supply Demand Improve – In order to assess when demand/supply conditions for private-pay seniors housing will improve, in the chart below I project growth in the supply of seniors housing into the future assuming the same rate of annual growth in supply seen in 3Q15.   This rate of growth (5.24% weighted average) is applied to the supply at the beginning of each five year period and held constant over each five-year period.  Once we pass 2020, as the chart indicates, the future of private-pay seniors housing is increasingly bright, with higher demand driven by increased longevity and, after 2026, the long-touted and final arrival of the baby boomers to an age when they might actually consider seniors housing.

Supply Demand

However, when you look closely at the above numbers, you see that 80+ demand begins to exceed the growth in supply only slightly in the 2021-2025 and really strong demand from 80+ seniors relative to the level of supply growth does not begin to appear until after 2026, when the Baby Boomers (1946 to 1964) begin turning 80.

Reasons For Near-Term Pessimism – While not every seniors housing market will get overbuilt and many high-barrier-to-entry markets may avoid the adverse impact of additional private-pay seniors housing development, I believe the data above supports my pessimistic view on private-pay seniors housing occupancy, rate and the risk of overbuilding over the next 3 – 5 years. If we keep building at current levels as a percent of supply and we focus on demand from 80+ seniors, it appears that the seniors housing industry will substantially overbuild the market over the next five years. In the period from 2021-2025, the amount by which projected 80+ demand will exceed projected supply growth should be sufficient to help absorb some of the excess supply created in 2016-2020 but may not be high enough to support a significant increase in occupancy or rate or a true senior housing boom.   The golden age in terms of demand is really a post-2026 event assuming supply growth continues at today’s rate, the health of baby boomers at 80 is about the same as today’s 80 year olds and boomers will find seniors housing as it is currently being designed and built attractive.

Keys To Success In A More Competitive Environment And Future Arrival Of The Boomers

In order to outperform in the more competitive environment for private-pay seniors housing that I see over the next 3-5 years, I believe operators should:

  • Limit new development near term
  • Focus on high barrier to entry markets
  • Try to reduce turnover
  • Design/Redesign/Market properties to attract under-80 or early 80s seniors by focusing on IL rather than AL and rethinking locations and amenities to appeal to “younger” seniors
  • Increase their equity cushion and line up capital in order to be able to bid for more attractively priced acquisitions if occupancy and rates fall and some new product cannot be filled as anticipated

In future blogs, I will discuss some of the cutting edge product that I believe will appeal to Under-80 seniors and look at the housing alternatives to private-pay seniors housing for this age group such as staying in their homes or choosing mixed age condos and apartments, using support services where necessary.

Technical Notes

I want to acknowledge the help of my friends at NIC in preparing this blog, even though it takes a more negative view on the near-term outlook for the industry.   I particularly want to acknowledge the help of Robert Kramer, Beth Burnham Mace and Chris McGraw.   Dave Schless at ASHA also reviewed an early draft and gave me his feedback.

I do not intend to malign NIC MAP data in this blog post. The advent and growth of NIC MAP data is a great tool for the industry and one that should help us avoid the rampant overbuilding seen in private-pay seniors housing in the late 1990s.   NIC MAP makes no statement about the appropriateness of the 75 plus penetration rate and demand, per se. NIC MAP adopted the 75 plus household cohort a number of years ago because it has been traditionally been used in the sector by feasibility analysts and others.

I also want to acknowledge two industry reports that cover some of the same material noted here but reach somewhat different conclusions. These are: Beth Mace’s Demographic Update Commentary, circulated by NIC in July, 2015 and Phil Downey’s and Larry Rouvelas’ A Projection of Demand for Market Rate U.S. Seniors Housing 2010 – 2030 published by American Seniors Housing Association Winter 2013.

NIC defines institutional quality private-pay seniors housing as properties with 20 or more units. NIC normally calculates penetration based by comparing the total supply of private-pay seniors housing in the top 99 markets to the total U.S. 75+ household cohort (not the entire household and institutional 75+ population).

In this analysis, I compare the total number of estimated occupied private-pay senior housing units in the U.S. to the total U.S. population of 75 and over and 80 and over seniors.   I believe use of a penetration rate based on actual occupancy rather than including vacant units is more accurate but use of either an occupied or total supply penetration rate would produce essentially the same result as indicated above.

While households are the standard unit of demand for housing of all types, NIC, other researchers and I also use population to measure future demand because the Census Bureau does detailed population projections by age but not projections of households.   Various commercial data services do project households by age.   One other cautionary note when thinking about demand projections for seniors housing is that male longevity has been improving, meaning more very old two person households and potentially less unit demand for private-pay seniors housing than population projections alone may indicate.

There are also some limitations in how I project supply growth.   I use NIC MAP construction estimates as of 3Q15 for the top 99 markets, make the assumption that these units will be delivered evenly over two years and that this same rate of growth is occurring in the rest of the country outside the top 99 markets and will continue in the future.

While I believe the assumptions used in this assessment are reasonable and have reviewed them with NIC staff, I believe it would be very helpful for the industry for NIC, ASHA or an independent academic researcher to undertake a demand / supply analysis using household projections by age, seniors housing supply and construction data for just the top 99 markets and include in this an updated survey of the actual entry ages for seniors housing today.   Such a study would allow us to better select an aged-based penetration rate at 75, 80 or 85 and would eliminate some of the uncertainty created by mixing population and supply data for the entire U.S. with occupancy and construction stats for the top 99 markets.

I welcome your comments on this blog post.