A Review of the Florida Keys Carrying Capacity Study (2002)

The Socioeconomic Module provides population projection inputs to the Fiscal and Human Infrastructure Modules and residential and commercial development data to the Infrastructure Module. These inputs were often derived in a simple, straightforward and reasonable manner using readily available data, which makes it easy for the user to understand the module and should help the user modify assumptions and integrate new information. The choice of planning units also seems reasonable, as it matches both local governmental planning units and census tracks.

The Socioeconomic Module, as it stands in the Draft CCAM report, uses 1990 U.S. Census data to estimate coefficients, which are assumed to be constant. In fact, these coefficients are not constant (see discussion below) but instead evolve over time. Using 2000 U.S. Census data rather than 1990 data will substantially reduce the error in the starting values for these coefficients. It is worth emphasizing that errors in the base population figures will propagate throughout the CCAM because population is used in the Fiscal and Human Infrastructure Modules to calculate various intermediate and final output measures. Although the 2000 U.S. Census data may not have been initially available, most of it has now been made available. Using that data, rather than the 1990 U.S. Census data would significantly strengthen the CCAM.

The decennial U.S. Census is the most “official” population count of jurisdictions, as well as smaller area tracts and blocks in existence. Not only does the Draft CCAM Report indicate a 19 percent lower population count for the year 2000 than the census, but its “Smart Growth” scenario for the year 2020 shows fewer people than the census established for 2000. Since final 2000 Census figures are now available, re-estimation is imperative, particularly if projections are to be legally defensible.

Two methods were used to obtain population estimates in the Socioeconomic Module: a) independent population projections, and b) population derived from scenario-based land use. The Draft CCAM consistently employed the second method. Given growth restrictions in place in the Florida Keys and the limited supply of land available for development, it makes logical sense to start from the amount of land for each type of development as defined by the scenario. Combining the scenario-based land use map along with the assumption that development will occur on suitable land and a coefficient for the number of people per unit for each development type yields an estimate of the increase in population. The first method, which begins with population and projects land use changes based on population changes, is used only in the socioeconomic section and is justified “as a frame of reference” or internal consistency check. Population growth in the Florida Keys is certainly affected by public policies such as the Rate of Growth Ordinance (ROGO). In addition, future growth is likely to be determined in large part by public policy. It is advantageous to make the consequences of land use regulation on population growth explicit as is done in the second method.

Regardless, one consistent method for developing population projections should be used throughout the Draft CCAM. Employing two different methods in the Socioeconomic Module is both unnecessary and confusing. It would be best if all references to the independent population projections were deleted from the Draft CCAM, including references in both the main body of the text and in Appendix C of the Draft CCAM and the socioeconomic section rewritten accordingly.

The Socioeconomic Module makes use of numerous coefficients that are assumed to be constant throughout the analysis. The use of constant coefficients makes the analysis easy to follow and may be justified in some cases on the grounds that certain relationships are relatively constant, at least over the range of analysis considered in the scenarios. There is a danger in assuming constant coefficients, however, especially for models such as the CCAM that are being used as impact assessment tools. Coefficients may in fact be functions of scenario

assumptions. As described in Appendix C (Draft CCAM Section 1.0), residential densities in each planning unit are assumed to remain constant at their current value. As suitable land for development becomes scarce, these densities are likely to increase due either to market forces or regulatory decisions. Increased scarcity of suitable land is also likely to cause an increase in housing prices, which may cause changes in community composition that may, in turn, change everything from the demographic structure of the community to the average number of persons per dwelling. It is incorrect to assume, as is done in Section 3.2.2 of the Draft CCAM, that densities “will remain essentially constant” because of limited population growth.

The analysis could be improved in several ways. At the very least, a sensitivity analysis should be run that allows various coefficients to assume different values. It should also be possible to analyze historical trends in these coefficients to see how they have changed in Monroe Country over time. Comparing coefficient values in Monroe County to coefficient values in other Florida counties may also yield important information. A more sophisticated approach would attempt to estimate these coefficients on the basis of underlying conditions.

Affordable housing is an important socioeconomic indicator to track. In Monroe County, average housing prices are very high relative to state and national averages. The Draft CCAM calculates the Affordable Housing Index (AHI) using existing data on housing prices and income. The study does not attempt to predict what would happen to housing prices under various scenarios, which means that the AHI is a constant, not a function of the scenario. From simple reasoning about supply and demand, it would be expected that allowing less land to be available for development would lead to higher housing prices.

Some care should be exercised in interpreting the AHI. A rise in property values has different implications for different sectors of the economy. For property owners, a rise is good news because it increases the value of their assets. It is also good news for local governments, that increase their tax base with a rise in property values. For those looking to buy property, however, an increase in property values is bad news. In addition, a significant portion of the demand for property in the Florida Keys comes from people living outside of the Florida Keys area. For this segment of demand it is their income, not Keys income, that determines the affordability of the property.

It should be noted that housing affordability is a serious concern for Monroe County and that the County is attempting to find means to provide it, presumably for rental as well as for sale. The County has issued a Request for Qualifications to consultants that specialize in housing to identify possible experts to assist a five-member Affordable Housing Oversight Committee.

As it currently stands, the AHI cannot be used as part of an impact assessment tool to assess outcomes of various scenarios. The measure is static and does not depend on the conditions assumed in the scenario. In order to make the AHI a useful impact assessment variable (IAV), a model of housing demand for the Florida Keys would need to be developed. Though there is a large existing literature on housing demand, existing housing demand models would need to be customized for application to the Florida Keys. At a minimum, such an approach would allow changes in developable land to affect housing prices. A more sophisticated approach might also try to account for the effect on property values given changes in environmental amenities, the major reason why the Florida Keys are in such demand.

Two sets of questions are generated by the description of the Competitive Commerce Index (CCI). First, on a practical level, the process used to compute the CCI is unclear. The description of the CCI in Section 3.3.4 of the Draft CCAM does not agree with the description in Appendix C of the same report which describes a “Retail Concentration Index” but not the CCI. The average disposable income (ADI) used to calculate the CCI is not mentioned in Appendix C of the Draft CCAM. There is no description of the data source for the ADI, whether it is given by planning unit or for the entire county, or whether it attempts to include spending by tourists.

Second, it is unclear as to how to interpret the CCI or what its significance is; it is unclear if the CCI is an important measure on its own and if it impacts traffic flow patterns or other impact assessment variables. There are no threshold values for the CCI defined in Appendix C of the Draft CCAM as there are for the AHI. The CCI should be clearly defined and explained or it should be left out.

The AHI and CCI are the only two impact assessment variables generated by the Socioeconomic Module in Section 3.3.4 of the Draft CCAM report. These measures clearly fail to provide a comprehensive or accurate representation for the quality of life or socioeconomic well-being of people in the Florida Keys. It is unclear why the Draft CCAM did not include measures of income, fishing, recreation, and various social indicators. If the AHI and the CCI are the only two measures assessed, the write-up of the Socioeconomic Module should indicate that the CCAM is not attempting to incorporate or track quality of life or socioeconomic well-being in any systematic manner.

The Quality of Life portion of the Draft CCAM consists of a two-page writeup of the contractor’s Public Information and Involvement Plan (PIIP) (Draft CCAM, pp. 69–70; United State Army Corps of Engineers, Planning Division, 2000; The Market Share Company. 2001.). The PIIP was developed to allow public input on the CCAM development (United State Army Corps of Engineers, Planning Division. 2000; The Market Share Company. 2001). During the PIIP, participants were asked to rank several quality of life issues. Section 3.8 of the Draft CCAM report states that “many of the important Impact Assessment Variables (IAV) that are CCAM outputs are related to quality of life issues as put forth by the community.” The IAVs, however, do not match closely with results from the PIIP as listed in Table 3.21 of the Draft CCAM report. Maintaining current community character (ranked 3rd), views about government regulation (ranked 5th, 8th, 9th, 11th, 15th and 16th), improved safety on U.S. Route 1 (ranked 7th), tourism (ranked 10th, 14th and 17th), and recreational opportunities (ranked 12th and 13th) lack closely related IAVs. Of the Socioeconomic Module IAVs, affordable housing is ranked 6th. Nothing listed in Table 3.21 of the Draft CCAM compares to the CCI.

Encouraging community involvement in order to generate issues against which endpoints can be measured is an important process for the URS Corp. Attendance at public meetings allows input from certain sectors of the community, but is unlikely to gather feedback from the entire community. The contractors should have exerted greater effort to include community input at an early stage. In addition, more insight can be gleaned from the information in hand than has been obtained to date.

A large part of the economy of the Florida Keys depends upon tourism, both those visitors on an extended stay and day-trippers. The number of extended stay visitors are tied to the number of hotel rooms and rental properties available. The number of day-trip visitors, however, are tied more to recreational opportunities, fishing in particular, rather than any measure explicitly related to land use. A number of governmental units collect information on recreation demand. The National Oceanic and Atmospheric Administration (NOAA) has several documents on the subject readily available on the web (National Oceanic and Atmospheric Administration, 1996). This information should have been investigated and incorporated into the Draft CCAM. In failing to incorporate this information, a large part of the economy of the Florida Keys has not been adequately represented in the Draft CCAM.

Few references are given for the Socioeconomic Module. Even when a reference is provided, it includes little detail on where the relevant information can be found within the reference. The report’s description of the Socioeconomic Module in Appendix C contained insufficient information to gain an understanding of particular calculations. Specific comments on these points are given in Appendix D of this report.