Housing Issues or “Where Do We Live?”

One question that confronts regional economists is “Where do We Live?”. This question provides two possible tracks upon which we can initially travel. First, it is important to look at the expansion of the housing base. This serves the economic consideration of home building as an economic engine. It also serves as a consideration of sprawl and regional growth. The second track to explore is housing valuation. This serves the question of wealth creation and affordability.

Sprawl or Population Growth: A Chicken and Egg Question Within the first question of home building exists a nagging question—who will buy these new homes? This question is regularly debated and addressed. It is an important discussion as it relates to issues of urban sprawl and population growth. An initial consideration in evaluating this issue is to realize is that Wisconsin, and this region, are experiencing increases in population.

Through projections provided through the Wisconsin Department of Administration by the US Census Bureau, it is estimated that the population of the State of Wisconsin will grow by 3.73% between 2000 and 2005 (see Table #4—Projected Population Change). It is estimated to grow an additional 3.37% between 2005 and 2010. As a result of this projected growth, both through migration and natural growth, it is appropriate that these individuals will expect housing that serves their needs. It must be noted that a 3% increase in population does not command an equal increase in housing units. Since the average density of a housing unit in Wisconsin is slightly over 2 persons per unit there is a need for less than one unit per person.

This section will review the over- and under- supply expectations throughout a variety of communities in the region. This process begins by calculating the individual density numbers for the various communities. It then looks at the population projections. Through these two numbers, it is estimated how many more units will be needed in the short term (2000-2005) to accommodate these individuals. These estimates are then placed next to the actual number of housing permits (single family, duplex, and multi-family) pulled in 2001-August 2004 (see Table #3—Single Family Housing Unit Permits). This number is compared to the estimated number of units necessary to house the expected population change.

The simplest example provided is the City of Brodhead in Green County. Note that no communities in Lafayette County were included due to a lack of consistent building permit data. The US Census Bureau estimates that the population of Brodhead will decline by 20 persons between 2000 and 2005. This decrease of 0.63% (see Table #4—Projected Population Change) is then compared to the density calculated in the 2000 census of 2.35 persons per unit in Brodhead. As a result, it is estimated that Brodhead should eliminate 9 units over this period or witness an oversupply of housing. This assumes that density will not change over the period. It also assumes (not without the recognition that housing does wear out over time) that abandonment and teardowns do not occur. When the fact that 23 units were issued building permits in this period, it can be argued that there is an oversupply of 32 units in Brodhead at this time (see Table #5—Housing Unit Growth). Since the permit data only goes through September 2004, any additional permits will compound this theoretic issue.

For clarity, several real world issues need to be considered. First, since Brodhead is an extreme example, it turns out that if the city merely witnesses population growth of 74 persons during this period, the oversupply will be depleted. Second, as stated earlier, while housing enters the market through new construction it also exits the market through destruction (tear-downs, natural disasters, etc.). As a result, the Brodhead example can show how, on the margin housing expansion may lead to oversupply. It can also show how minor population changes can accommodate such supply problems.

Within this example, two communities stand out: Lake Geneva and Elkhorn. These two communities, both in Walworth county, anticipate future population growth. The census predicts Lake Geneva’s growth will be 1.57% over the five-year period, while Elkhorn’s will be 9.24%. As a result of their housing density calculation (1.94 persons in Lake Geneva and 2.37 persons in Elkhorn) it appears that Lake Geneva has overbuilt while Elkhorn has under built.

According to the census, Lake Geneva will need 58 units. However, it permitted 519 units. In order to accommodate that supply, Lake Geneva will need to grow 12 times faster than the census bureau projects. At the same time, Elkhorn required 258 units while it only permitted 197. As a result, it can only accommodate a 6.4% growth rate over this period.

While these numbers appear to foretell an oversupply of housing, several issues should be considered. First, the population is fluid. Barbed wire and barking dogs do not surround these communities. As a result, population growth estimated for Elkhorn may fluidly relocate to Lake Geneva, causing demand to accelerate there. Second, due to the vacation nature of Lake Geneva, population estimates (conducted on year round residents) may not apply. The same is valid in Whitewater where the University of Wisconsin-Whitewater creates a demand for housing without commensurate population changes. Finally, on the margin, these numbers are not considerable. These permits allow for increases in the housing stock that amount to an increase in housing of about 5.6%. These estimates are based on the census bureaus projection of a 3.26% increase in population.

As a result, any combination of increased population (an additional 5500 persons), second home owners, college students, and the retirement of existing units will accommodate the development of this supply of new housing. Within these selected communities there were 63,901 housing units in 2000. This variation in estimates accounts for only 2.4% of supply.

Valuation vs. Appreciation A second interesting issue that faces the housing market is the cost of single-family houses. This data focuses on countywide data (with the exception of Lafayette County where a full data set was not available). It is important to note that within the counties are unique markets, such as lake communities, such as Delavan and Lake Geneva. These communities offer unique externalities that may drive up prices. In order to make more generalized comments on the markets, county data is incorporated. Future issues of the newsletter will focus on individual communities.

In addition, it is important to acknowledge the difference between valuation and appreciation. As the data shows, there has been dramatic appreciation in real estate in Green and Rock counties (see Figure #3—Single Family Home Sales Appreciation 1997-2003) over the past few years. However, the low basis upon which this appreciation occurred creates a mild statistical distortion that needs to be noted.

As the data shows, there has been consistent appreciation of existing single-family houses between 1997 and 2003. Outside of one year in Green County, every year has witnessed higher housing prices than the previous (see Table #6—Average Single Family Home Price 1997-2003). In fact, while Green County did see a reduction in the average sales price in 1999, the rebound in 2000 more than compensated for that poor performance.

In terms of valuation, Walworth and Jefferson possess more expensive existing single-family homes. This is not a statement of quality; this data does not provide any insight into the types of homes sold. It may reflect supply and demand considerations. It may also reflect different externalities possessed in the market. As noted earlier, lake properties command higher prices due to the positive externalities they offer. On the other hand rural properties (consider Green County) may be more remote from employment opportunities.

Home values in both Walworth and Jefferson Counties benefit from their median location between employment centers. Jefferson County sits between Dane County and Waukesha County. As a result, Jefferson County (which is an exporter of workers according to the Wisconsin Department of Administration and the US Census) provides various opportunities for employment in a regional format. For example, there is a net export of 5% of Jefferson County’s labor force to Dane County and 10% to Waukesha County. The three largest importers of labor in the State of Wisconsin (Milwaukee, Waukesha and Dane Counties) are connected to Jefferson County by I-94.

Walworth County has a similar advantage, although it abuts both Racine and Kenosha Counties, both exporters of workers. However, this is a net export of 11% of Walworth County workers to Illinois. In addition, Waukesha County is a net importer of workers from Walworth County. 3.7% of Walworth County workers commute to Waukesha County for employment.

The remaining counties have less robust export markets. For example, Rock County two largest job export markets include Illinois, which is a net importer of 3.4% of jobs. Dane County creates a net import of 4% of jobs. Green and Lafayette County follow similar patterns of job development, thus limiting the external demand for single-family homes by commuting workers. While this has kept valuations low, appreciation in all markets is high.

Single Home Values—Wisconsin Realtors Association. http://www.wra.org Housing Unit Data—United States Census Bureau. http://www.census.gov Population Estimates—Department of Administration. http://www.doa.state.wi.us Commute to Work Data— United States Census Bureau. http://www.census.gov

Table 3: Click on image for larger view

Table 4: Click on image for larger view

Table 5: Click on image for larger view