|
Meeting #16 October 26, 2000
Agenda October 26, 2000
(The Assessment Report is very large and will require a long loading time.)
Assessment Report, October 19, 2000
Appendix 1- Alternatives Assessment Transportation & Circulation
APENDIX 2
TECHNICAL MEMORANDUM
| Date: |
October 24, 2000 |
| From: |
Sally Nielsen |
| Subject: |
Feasibility Analysis of Central Larkspur Specific Plan Alternatives |
This memorandum summarizes the results of Hausrath Economics Group's (HEG) feasibility analysis of the Central Larkspur Specific Plan alternatives. Our analysis is based on the Alternative Plans Development Program Descriptions prepared by Thomas Cooke Associates (TCA) dated 9/26/00 and on subsequent clarifying communications. TCA also provided the construction costs and site preparation costs. HEG assembled the development revenue and income assumptions based on review of comparable Marin County development, conversations with Thompson Residential Builders, and our prior experience preparing pro-forma feasibility analyses.
Structure of the Analysis
We conducted a land residual feasibility analysis for the four key properties affected by the alternatives: the city parking lot, Nazari property, Niven West, and Niven East. This type of analysis identifies any development income left over, after accounting for all development costs except land. That amount is the "residual" that determines the land value for a given development program.
Our analysis is structured as follows:
Add all development costs:
- Construction costs by development type (hard costs);
- Site preparation costs (hard costs);
- Soft costs (design, engineering, financing, fees, insurance, and other incidentals) as a percentage of total hard costs;
- Land costs for replacement parking, where applicable; plus
- Contingency, as a percentage of all hard costs, soft costs, and applicable land costs.
Add all development income:
- Capitalized rental income net of vacancy and operating expenses; plus
- Income from sale of residential units, net of selling expenses and builder's profit.
The difference between development cost and development income is the net development income. This amount is the income residual that represents the land value for the alternative development programs.
Assumptions are generally conservative
We assume no real increase in rental income or sales prices over time. This is a conservative assumption. Similarly, no real increases in development costs are assumed. Other components of the feasibility analysis are also relatively conservative:
- a 25 percent factor for soft costs (architecture and engineering, fees, financing, insurance, and other incidentals),
- a 10 percent contingency on all development costs,
- a 15 percent rate of return on investment (a 15 percent capitalization rate applied to the income property),
- a 7.5 percent factor for the sales expense on for-sale residential development, and
- a 10 percent builder's profit on the for-sale residential development.
Table 1 (at the end of this memorandum) summarizes the development cost assumptions. Table 2 displays all of the development income assumptions used in the analysis, including sales prices and rental rates.
Answers to Specific Topics and Issues for the Four Development Sites
Residual land value of city lot assuming mixed use development
Development of the city lot does not generate a positive residual land value under any of the alternatives evaluated here (see Table 3). Excluding the library alternatives, the net values associated the other alternatives are all negative, ranging from a negative $29 per square foot (Alternative X2) to a negative $70 per square foot (Alternative Y-B/Z1-B). For the city lot, the alternatives with revenue-generating development result in lower residual land values than does Alternative X2 (the plaza alternative).
The proposed revenue-generating development for the city lot consists of a mixed use building with ground floor retail and residential rental apartments above. The amount of space is not large, and residential rental development does not generate very much income. The net income from new mixed use development barely covers the hard construction costs of the mixed use development, even before accounting for the costs of structured parking, replacement parking (relatively small), site preparation (also relatively small), and soft costs and contingency. A development program incorporating for-sale residential development (townhouses above street-level retail) and no structured parking would generate a positive residual land value under the development cost and revenue assumptions used for the feasibility analysis of the alternatives.
Comparison of the economic viability of Nazari property development under different development assumptions
Mixed use development of the Nazari property generates a positive net land value under Alternatives Y and Z1, in the range of $10 per square foot. The residual land values are negative under Alternative X1, X2, and Z2, ranging from negative $5.60 per square foot under X2 to negative $19.38 per square foot under Z2. (Note: The costs of relocating the existing railroad buildings under Alternative Y are not included in this feasibility analysis.)
Under Alternatives X1 and X2 the costs of the plaza and structured parking are high relative to the amount and type of revenue-generating development. As seen in the city lot analysis, mixed use development with residential rental units does not generate enough income to offset development costs, even without structured parking. Replacing residential rental units with hotel development (X2) boosts the income from new development but the added costs of structure parking and additional plaza under that alternative drag down the net value calculation.
The positive results for Y and Z1 are very similar, reflecting the similar development programs. Both have a larger hotel than does X2 and both also have for-sale residential development (townhouse or live-work). This feasibility analysis assumes that there is no sales price discount for the live-work building type, just as there is no construction price discount. In other words, the live-work units function in the market essentially as townhouse units. Alternative Z2 with no hotel and with high structured parking costs results in the most negative residual land value for the Nazari property.
We conducted a sensitivity analysis to evaluate the impact on feasibility of the higher proportion of land area and improvement costs devoted to publicly accessible plaza area under those alternatives. The amount of additional public plaza area under X1 is small (about 2,700 square feet) and removing that cost from the feasibility calculation only changes the net value by two percent. Under X2, the higher amount of public plaza area is larger (about 10,000 square feet); subtracting those plaza improvement costs improves the net residual land value calculation by 20 percent. Table 2: Central Larkspur Specific Plan Area Alternatives - Public Development Costs in the companion fiscal analysis memorandum (Appendix 3) lists these public plaza improvement costs, and associated land value compensation to the Nazari developer, as potential additional public costs.
Residual land value of the Niven West site assuming 4,000 square feet of retail development and either five cottage units or 10 live/work units
The residual land values for the Niven West site are the highest per square foot among the four sites analyzed (see Table 5). For this analysis, the site area is reduced by the amount of land area that would be devoted to surface public parking (50 downtown-serving spaces and varying amounts of surface spaces replacing spaces lost with development of the city lot). Site preparation costs are also reduced by the cost of the public parking. Table 2: Central Larkspur Specific Plan Area Alternatives-Public Development Costs in the companion fiscal analysis memorandum (Appendix 3) accounts for the public parking costs. The costs of the city lot replacement spaces are also incorporated in the feasibility analysis of city lot development (see Table 3).
The residual land values range from about $18 per square foot for the option of 4,000 square feet of retail space and five cottage units to about $46per square foot for the option of 4,000 square feet of retail space and 10 live-work units. The live-work option provides a substantially better return because more units are provided, and those units are sold for higher per-square-foot prices than are the cottage units. The per-square-foot construction costs are also higher for the live-work units than they are for the cottage units, implying a substantial degree of finish work, comparable to what would be found in a typical residential townhouse. The differential between the per-square-foot construction cost factors is greater than the differential between the two per-square-foot sales price factors.
It is also important to note that we have assumed the live-work units do not count as residential development for the purpose of the City of Larkspur's General Plan affordable housing policy. If they did count as residential development, then one (or two) of the units would have to be provided at a discounted price. This affordable housing discount is not incorporated in the current feasibility analysis.
The residual land value of the Niven West site is overstated to the extent some site preparation costs are absorbed by the Niven East development. Table 7 shows the residual land value results for the combined Niven West and Niven East sites. To estimate land cost for replacement parking and for other public parking provided on the Niven site, we used a residual land value reflecting the combined results for Niven West and Niven East.
Residual land value of three residential alternatives for the Niven East site
The residual land values for the three residential development alternatives for Niven East are about the same in all cases-in the range of $15 to $16 per square foot of land (see Table 6). The alternatives reflect different development programs and different means of meeting the affordable housing requirement.
Alternatives X and Y accommodate the affordable housing requirement within the private development program by making the requisite number of cottage units (four units in Alternative X and 7 units in Alternative Y) available at prices affordable to low and moderate income households. Those less-than-market-rate prices are incorporated in the feasibility analysis for Alternatives X and Y and in the land value results.
Alternative Z takes another approach to the affordable housing requirement, reserving about two acres for land for development of affordable housing by a non-profit housing developer. (A larger amount of affordable housing could be developed in this way-36 apartment units could be accommodated on this two-acre site.) The balance of the Niven East site is assumed to be developed with entirely market-rate residential development (and associated parks and open space). The feasibility analysis of Alternative Z accounts only for that development, not the non-profit component. Under the assumption that these would be the responsibility of the non-profit developer, the development costs, including site preparation costs, and development revenues associated with the affordable housing development are not included in the Alternative Z calculations for Niven East. The land value results for Alternative Z represent residual land values for a development program that does not incorporate any subsidy for affordable housing. It also does not reflect any value from transfer of the two-acre site to a non-profit developer.1
In general, the residual land values for the three Niven East alternatives are positive but are lower than the land values for the Niven West site. This is because Niven East has higher costs to provide park, open space, and transportation improvements. Niven East also bears the cost of site preparation that is required for Niven West. (Table 7 shows the residual land value results for the two sites combined.) The affordable housing requirement also lowers the residual land values for Niven East compared to Niven West.
We conducted a sensitivity analysis of the Niven East alternatives to evaluate potentially higher costs for roads within the development area. If the road cost component of site preparation costs were 20 percent higher than assumed in the basic analysis, then residual land values would be about four percent lower than indicated in Table 6.
For the basic feasibility analysis, all Niven East alternatives incorporate as private development costs the habitat and park land improvements required by the city. We conducted a sensitivity analysis to evaluate the impact of those requirements on residual land value. The habitat improvement costs under X apply to 9,200 sf of land. Eliminating the costs of habitat improvements to that land area increases the net value per square foot for the balance of the site by just over two percent. The change is greater under Z where more land area is designated for habitat improvement. Subtracting those costs increases the net value per square foot by about seven percent. Also under Alternative Z, more improved park land is assumed than would be required by the City of Larkspur Park Dedication ordinance. Under the Alternative Z residential development program (including affordable housing units), the park dedication requirement would be 44,274 square feet of land area;2 64,000 square feet of park land is included under that alternative, resulting in an excess of 19,726 square feet beyond the dedication requirement of the ordinance. Subtracting this amount of land area and associated park improvement costs from the private development cost calculation for Alternative Z increases the residual land value per square foot by about five percent. Subtracting both the excess park and the habitat improvement requirements from Alternative Z increases the residual land value per square foot by about 12
percent. Table 2: Central Larkspur Specific Plan Alternatives-Public Development Costs in the companion fiscal analysis memorandum (Appendix 3) shows the magnitude of the land and improvement costs associated with these city requirements/policies.
Answers to related questions and issues
What is the trade off between a standard single family lot and other types of housing proposed, i.e., how many cottages are required to equal the return on one single family unit?
Sensitivity analysis indicates that two cottage units provide about the same return as one standard single family unit. We used Alternative Z as a baseline for this sensitivity analysis, since it evaluates a purely market-rate development. In the base case for Alternative Z, the residual land value is $15.54 per square foot. Subtracting one single family unit from the mix (14 units instead of 15 in the base case) lowers the residual land value. Subsequently increasing the number of cottage units by two (from 39 to 41) and maintaining 14 single family units results in a residual land value of $15.61 per square foot.
This sensitivity analysis is based only on the differentials in hard construction costs, sales prices, and unit sizes. It does not assume any differences in site preparation costs that might be associated with a change in the configuration of unit types.
What would be the income to the city if they sold the parking lot for mixed use development?
Under the alternative development programs analyzed here there would be no incentive to sell the city parking lot for mixed use development. None of the development proposals generates a positive land value. A scheme that incorporated for-sale townhouse or live-work development would show a positive land value. One such scenario (townhouses built over retail and "tuck under" parking, with no structured parking) shows a residual land value of about $10 per square foot. Income to the city from sale of the lot would be about $196,000 at this land value.
_____________________
| 1 |
An appropriate land value assumption for the transfer of that two-acre site would incorporate the consideration that affordable cottages would otherwise be provided as part of the Alternative Z development scenario. That factor reduces the residual land value under Alternative Z by about 13.5 percent, from $15.54 per square foot to $13.43 per square foot. |
| 2 |
The park dedication ordinance provides for waiving of requirements and fees for affordable housing development (Section 17.13.150 Exemptions.) This sensitivity analysis assumes that the requirement would not be waived. |
APENDIX 3
TECHNICAL MEMORANDUM
| Date: |
October 25, 2000 |
| From: |
Sally Nielsen |
| Subject: |
Fiscal Analysis of Central Larkspur Specific Plan Alternatives |
This memorandum summarizes the results of Hausrath Economics Group's (HEG) fiscal analysis of the Central Larkspur Specific Plan alternatives. The assessment is based on HEG's analysis of the City of Larkspur's 1998-99 Actual Budget, on the Alternative Plans Development Program Descriptions prepared by Thomas Cooke Associates (TCA) dated 9/26/00 and subsequent clarifying communications, as well as on the results of HEG's feasibility analysis of the Alternatives, presented in a companion memorandum (Appendix 2).
The fiscal impact analysis consists of an assessment of the implications of each alternative for the City of Larkspur's General Fund operating budget. We also provide a summary of the potential public costs associated with each alternative, e.g., costs for a new library, costs for public parking spaces to serve the downtown area, costs for replacement parking, costs for publicly accessible plaza area on private property, and costs for habitat improvements and park land in excess of city requirements. The memorandum concludes with an evaluation of the development impact fee revenue generated by each alternative for the Larkspur School District.
Alternatives Analyzed
The feasibility analysis indicated that some development programs would generate higher net values than others and would thus be the preferred development options. We have simplified the range of alternatives analyzed for the fiscal impact analysis to reflect those conclusions. We analyze Alternatives X1, X2, Y, Z1, and Z2. In all cases, the live-work development option for the Niven West site is analyzed, because the returns generated by live-work residential development substantially exceed those generated by the cottage option. Similarly, the fiscal impact analysis uses the mixed use development options for the city lot, not the library options. The non-profit affordable housing development is incorporated in the analysis of Alternatives Z1 and Z2.
Fiscal Impact on City of Larkspur General Fund
Table 1 summarizes the fiscal impact analysis of the Central Larkspur Specific Plan Alternatives on costs and revenues for the City's General Fund operating budget. The results shown are annual revenues and costs and net revenue in Year Five, after initial absorption, when all new development is assumed to be built and occupied. All of the alternatives would generate a positive net revenue to the City's General Fund. The alternatives that include hotel development (X2, Y, and Z1) generate the most net revenue because of the transient occupancy tax-accounting for about one quarter of total general fund revenue under those alternatives. The annual net revenue ranges from a low of about $151,000 per year under X1 to a high of about $303,000 per year under Y.
Comparing the results of Alternatives Y and Z1 illustrates other components of the fiscal impact analysis. Property tax revenues in Z1 are lower than they are in Y because of the different mix of development assumed for Niven East. The average market value of the market-rate residential units is lower under Z1 compared to Y, and Z1 includes the non-profit affordable housing development (assumed to be rental housing affordable to low and very low income households). On the other hand, sales tax revenue (and other per-capita revenue) is higher under Z1 because of the larger residential population associated with that alternative (including the 36 affordable units). Offsetting these higher revenues associated with a larger residential population are the higher costs that would also come with that larger population.
After Year Five, under the assumptions of this analysis (no real increases over and above inflation in development revenues, property values, per capita retail sales, or city operating costs), annual net revenue declines by a small percentage each year because of the impact of Proposition 13 limits on annual increases in assessed value (on which property tax revenue is based). Those limits, which apply unless property is sold, mean that assessed value increases at less than the assumed rate of inflation, so property tax revenues decline over time in real terms. Property tax revenues are an important revenue source, but they are not the only revenue source (as indicated by the table), so the annual real decline in net revenue would be relatively small.
The fiscal impact analysis evaluates several components of the operating budget:
- property tax revenue based on assessed value of new development and changes in that assessed value over time, including re-assessment at market value for periodic re-sales,
- property transfer tax revenue that accounts for the initial sale and the periodic turn-over (re-sale) of real property,
- sales tax analysis that accounts for the taxable spending in Larkspur of people living in the Specific Plan Area and of people working in the Specific Plan Area, as well as the additional sales tax revenue that would be generated by taxable sales in new retail development in the Specific Plan Area,
- transient occupancy tax revenue associated with new hotel development,
- other revenue sources that make up the operating budget and the public service costs funded by the operating budget (both calculated on a per-capita basis-per resident and per employee).
Public Development Costs Associated with the Alternatives
Table 2 summarizes the potential one-time public development costs associated with the Central Larkspur Specific Plan Alternatives. The biggest one-time cost item is the public library at $3.2 million. Without the public library, the potential one-time public costs range from around $350,000 - $400,000 under Alternative Y to about $1,700,000 under Alternative X2 (with the large public plaza on the city lot and the full cost of replacement parking spaces). The costs under X1 are lower because, aside from foregone general fund revenue, there are no public development costs associated with the existing city parking lot. The habitat costs associated with Niven East contribute to potentially higher public costs under X1 and X2. Similarly, under Z1 and Z2, it is the potential park land and habitat costs associated with Niven East development that could increase public development costs.
The downtown parking lot (50 surface spaces provided at the Niven West site) is common to all alternatives. The cost is in the range of $500,000 ($10,000 per space). The cost difference among alternatives reflects different land cost assumptions, based on the residual land values for the combined Niven property generated by the feasibility analysis.
The variations in use of the existing city parking lot introduce a number of public cost factors. One potential public cost is the cost of replacement parking assuming the city lot is developed more intensively. Again, the major component of this cost is the land cost assumption. In a parallel to this cost, the table indicates the net general fund revenue foregone (assuming no mixed use development on the existing city parking lot site) as a potential public cost under X1 and X2. Those alternatives would either retain the city parking lot (X1) or develop the site as a public plaza (X2). The improvement cost for the public plaza on the city lot under X2 is also shown on the table ($256,000). The table also indicates offsetting revenue to the City from sale of the city lot for more intensive mixed use development under Alternatives Y, Z1, and Z2.
The final potential public costs indicated on Table 2-Nazari public plaza, Niven East park land, and Niven East habitat represent the costs of improvements and associated land value for these public amenities. Again, the land cost is the most significant factor.
Development Impact Fee Revenue to the Larkspur School District
Table 3 presents estimates of the development impact fee revenue that the Larkspur School District would collect from new residential and non-residential development in the Central Larkspur Specific Plan Area. (The Tamalpais Union High School District does not assess a development impact fee.) These fees are established to partially offset the cost of new capital facilities required to accommodate the students associated with new development in the district. The current fees are $1.93 per square foot of residential development and $0.31 per square foot of non-residential development.
The table shows the one-time fee revenue that would be generated by development on each of the development sites under each alternative. The residential development on Niven East would clearly generate the most fee revenue, ranging from about $304,000 under Alternative Y to $340,000 under Alternative X. Because the fee is assessed per square foot of residential development, the relative size of the unit makes a significant difference. According to the school district, there is no exemption for affordable housing, so the fees associated with that development are calculated in the table.
For this analysis, it is assumed that live-work development is not counted as a residential use. Therefore, the fees associated with live-work development are calculated based on the lower non-residential rate. Also, as noted in the notes to the table, the fee does not apply to garage space, so the actual amounts would be somewhat less than shown here since garage space is included in the total square footage of residential development.
|
|