Return to the Brecht Associates Site Assisted Living: Do We Know How Much Is Too Much? by Susan B. Brecht As the assisted living industry continues its aggressive growth, the question of how much is enough, and more disturbingly, how much is too much is heard with increasing frequency. Efforts to gauge the size of the growing market for assisted living are hampered, to a certain extent, by both the infancy of this burgeoning segment of the senior housing industry and the issue of whether it will, in fact, occupy not only its own new position in the continuum, but also infringe upon the market for both congregate and nursing home care. Efforts have begun to try and quantify market potential on a national basis. The National Investment Conference (NIC/Price Waterhouse), in its 1996 publication, "The Case for Senior Living and Long Term Care Properties", established a methodology for forecasting the number of assisted living units that would be needed by the year 2005. In its baseline scenario for 1996, this publication estimated a need for 492,000 beds. An alternative scenarios indicated a need for as many as 522,000 beds. While this effort to establish the potential size of the national market is of some value, it has little bearing on the ability to judge the need for assisted living in any specific local market. In most markets, the development of state-of-the-art, residential assisted living communities has only recently reached the stage in which those communities are striving to reach their initial fill-up, thereby providing little insight into the question of how much product a given market can support over time. However, we have examined an area of suburban Philadelphia, in which the race to develop assisted living significantly preceded the national starting gun that is resulting the aggressive development activities that are currently being witnessed elsewhere. The performance of the assisted living communities in Bucks County, Pennsylvania, provide an instructive and sobering picture of what happens when overbuilding occurs. Bucks County is located to the north and west of Philadelphia and has a diverse character. Those areas located immediately outside the City of Philadelphia were the first to experience extensive development, fueled in part by the post-war housing boom created by the GI bill allowing many city dwellers to realize their dream of a home in the suburbs in the vast development known as Levittown. The suburbanization of Bucks County continued throughout the following decades, experiencing overall population growth of nearly 62,000 to approximately 541,000 between 1980 and 1990. The senior population grew by 54% during this period, increasing from 38, 300 to 58,900. Of equal importance to the story of Bucks County's assisted living boom, was the size of the adult child market. In 1980, a total of 100,500 households comprised this segment of the market. By 1990, the number had grown to 109,000. Much of Bucks County's growth continued to be concentrated in the areas known as lower and central Bucks, the area farthest from the city. Upper Bucks County, despite some growth, continues to be significantly less densely developed and more rural in character. The period examined in this analysis begins in 1987, when there were approximately 530 assisted living beds serving the private pay segment of the population. These were contained in 10 communities, of which four, comprising 30 percent of the beds, were continuing care retirement communities in which, essentially, a captive market was being served. Four others shared their campuses with nursing homes, offering a more abbreviated continuum for the frail elderly. The largest assisted living community at that time contained 144 beds and was a free-standing. The ten existing assisted living communities were operating at full occupancy levels. Between 1988 and 1990, five new assisted living communities opened their doors adding 647 new beds and more than doubling the existing supply. Unlike the majority of existing providers, four of the five new communities were quite sizeable, ranging from 117 to 203 beds. All but one were free-standing, with no ties to existing long term care or senior housing communities that might have acted as a feeder in filling their beds. Most were developed and owned by for-profit organizations with limited experience in the assisted living industry, and few of these operators understood the market that they had set out to serve. Expecting that their residents would be relatively independent, these providers were surprised to learn that the market for their product was a frail senior, in need of substantial support in activities of daily living. Further development occured between 1993 and 1996, when three additional assisted living communities were built increasing the number of beds in the market by 142. Two of these three, comparatively small facilities were designed and marketed exclusively to serve those with Alzheimer's Disease or related disorders. (See Table 1 for Development Chronology) The development pattern and the rate of growth in assisted living beds must be examined in the context of the demographic statistics that characterized the specific geographic area served by these communities. In 1996, just prior to the initial burst of development, the total population of individuals age 75+ was 14,783. This segment of the population grew by to 23,400 in 1996, reflecting the strong and growing market of seniors in the area. However, the pace at which assisted living beds were added to the area, significantly exceeded population growth within the target market. In 1986, there was one bed for every 28 people age 75 and above. By 1990, this ratio had increased to one bed for every 15 seniors in this age group. A comparison with national beds per population statistics is even more revealing. It is estimated that in 1991, there were approximately 11.4 beds per thousand seniors age 65+ in the United States, and 114.8 per thousand seniors age 85+. By 1990, in Bucks County, there were 25 beds per 1,000 65+ and 262 beds per 1,000 at age 85+. While these ratios diminished somewhat as development slowed down during the 1990's and population growth continued, the assisted living ratios in Bucks County (23 beds per 1,000 65+/224 per 1,000 85+) still significantly exceeded U.S. statistics by 1996 (estimated at 14 beds per 1,000 65+ and 114 beds per 1,000 85+ based on base case statistics from the previously referenced NIC/Price Waterhouse report). The development activities of the late 1980's, in particular, glutted the Bucks County market. During the period between 1989 and 1991, the market was able to fill over 300 units per year including turnover at existing communities and absorption of new beds. This represented an absorption rate that was more than 75 percent greater than the market had been exposed to in previous years and was not sufficient to absorb new product in a timely fashion. Overbuilding, compounded by the relative unfamiliarity of the market with the assisted living concept (particularly in its free-standing format), and the lack of industry experience on the part of most of the new providers contributed to the fact that the new communities took four to six years to reach full occupancy. The lengthy absorption period resulted in at least one of the new communities being taken back by the bank, and required the infusion of a significant amount of additional capital by all providers in order to remain in operation and offset operating losses. The slow and difficult absorption pace caused several of the organizations to diversify their services adding special dementia care programs, for example, which occurred at three of the five communities that opened between 1987 and 1990. This latter dimension created competition for the new AZ/RD facilities that opened in the mid 1990's. All of the new providers pointed to the level of competition as one of the major factors in the difficulty in reaching full occupancy levels in a reasonable time frame. There were other factors that contributed to this troubled period in assisted living development in Bucks County. The large size of most of the new communities had a significant impact on their ability to fill their beds. Turnover, which began well before full occupancy was reached, created an impediment to reaching stabilized occupancy levels. The experience in Bucks County underscored the notion that market saturation is not, however, simply a function of too much product. It is clear that too much product was introduced in too short a period of time. The existing supply of 530 beds that were serving this market when the overbuilding began had been developed over a period of many years and the relatively slow growth of supply was supported by the comparatively rapid growth in both the senior and adult child market. By contrast, during the period between 1988-1989, 476 new beds were opened, nearly equaling (90%) the number of existing beds. Another 167 were built in 1990-1991, not allowing enough time to absorb the initial burst of development of the previous two years. Happily, and somewhat miraculously, the overbuilding in Bucks County did not result in any of the new or existing communities going under. All communities finally reached occupancy levels of 93 percent or better by 1996-1997 and, although they work hard to remain full, they have been able to sustain high occupancy levels during the last year, demonstrating the the saturation that occurred in this market was a temporary condition that was corrected because of an ultimate slowing in the pace of new development and the continued strong growth in the target market. The lessons of Bucks County are instructive when evaluating the current assisted living development boom. A state which has captured the attention of much of the industry is New Jersey, where Certificate of Need (CN) approvals have been secured for over 20,000 new assisted living beds in 218 proposed new communities. Many have expressed concern about the potential for overbuilding in this densely populated northeastern State and so an examination of this possibility seemed in order, particularly in light of what has been learned about early overbuilding. When New Jersey passed its legislation, establishing the assisted living program, it set forth the requirement that projects apply for and gain a CN before they could be developed. At that time, the Department of Health divided the State into six geographic regions referred to as LABs. The LABs comprised the 21 counties in the State. (Recently, these LABs have been redefined). During the four year period since the CN was instituted, 35 new assisted living communities (containing 2,537 beds) conforming to the regulations set forth by this particular program have been licensed and an additional 33 are under construction which will add 3,611 beds to the existing supply. Prior to the initiation of New Jersey's assisted living program, there were approximately 307 existing communities in the State, containing nearly 10,000 beds licensed under other programs that permitted services similar to those offered in most assisted living communities. Based on extensive experience in evaluating various New Jersey submarkets, it is becoming clear that many of the early entrants into assisted living in this populous state are filling rapidly and doing well. Many have reported during interviews that we have conducted, that large proportions of their residents are not coming from the immediately surrounding communities, but are moving to their facilities because of their proximity to adult children. In addition, it is becoming clear, now that four years have passed since the inception of New Jersey's assisted living program, that many CNs have been secured by organizations that will not actually build their assisted living communities and so it isn't possible to be certain of the actual development activity that will result from the pipeline of approved projects. An examination of the six regions comprised by the LABs reveals that the beds that are currently under construction will result in increases in the total supply in each region ranging from 11 percent to 60 percent between 1997 and 1998. Arguably, there was a substantial under supply of assisted living product throughout much of New Jersey prior to the initiation of the assisted living program in late 1993. As reflected in the statistics in Table 2, by 1997, beds per 1,000 population age 85+ were below the estimated national ratio of 114 in four of the six LABs. However, if all, or even a large proportion of the beds currently in the assisted living planning pipeline were to be built, the ratio of beds to 85+ population would significantly exceed this ratio. Clearly, the potential for overbuilding exists in many areas of New Jersey. If all beds currently approved and in the review process were to be built, the supply, as of today, could increase in the regions by approximately 100 to nearly 250 percent over a five year period and cause saturation conditions reminiscent of those in Bucks County in the early 1990's, when supply was increased by 90 percent during a three to four year period. There is much to be learned from recent history, such as that in Bucks County as well as from the emergence of the assisted living industry in New Jersey. These lessons may help to frame the way in which absorption and saturation are defined and perceived in the future. Return to the Brecht Associates Site