The URL can be used to link to this page

Home My WebLink AboutCOM 0896.106 2022-2024 P/Courlci( Pail( M COMM.8% From: Chuck Flaherty Sent: Sunday,July 7, 2024 8:17 PM. To: Council Testimony Subject: Bill 169, Draft 2, Kona Vistas FEMA floodway example C C=:B Aloha Madam Chair and members of the county council, Here is a video narrated by Sherry Bracken for Big Island Video News of the Federal Emergence ' Management Agency's designated floodways running through the Kona Vistas property. https://youtu.be/baf0pdtc5xE == As rainfall events reach ever-more historic levels, Kona's floodways will begin to overflow more often. Because Kona's FEMA floodways are immature and subject to channel overflows and course changes, the impacts to downhill properties will be catastrophic. Look at the General Plan, Flooding, North Kona section and see that this is an area that, while in the urban expansion area, should not be developed. This guidance is further described in the Kona CDP and Chapter 27 of the county code. Clearly, this area should be preserved as open space and for natural and cultural resource preservation. It is a high flood risk area not suitable for residential or commercial development. Mahalo, Chuck Flaherty I n'n Comm. :•. ► Ir J. 6UW Ref. To: ( !IItI 1 Ref. Dote .41 IL 1 0 2024 July 10, 2024 Re: Opposition to Bill 169, Draft 2 Aloha Madam Chair and members of the county council, There is no economic or political justification for this development for the following reasons: Housing This development will not provide additional affordable housing. I repeat. This development will not provide additional affordable housing. It will only provide the number of affordable housing units required for a residential development. I have attached a study completed by Keyser-Marston Associates for the City and County of Honolulu. As you can see from the table below, 20% affordable housing unit will only supply the number of affordable units needed to accommodate the 80% market rate homes. It does not provide additional housing units for our county. Cumulative Inclusionary Percentage to Mitigate Increased Affordable Housing Need Single Family Low-Rise Mid-Rise Condo High-Rise Condo Rental Townhomes (PUC) Apartment Extr. Low (up 3.8% 3.5% 3.1% 4.1% 3.1% to 30%AMI) Very Low (up 10.5% 9.5% 8.8% 11.1% 8.7% to 50%AMI) Low (up to 16.5% 15.1% 14.0% 17.4% 13.8% 80%AMI) Moderate (up 19.8% 18.2% 16.9% 20.9% 16.7% to 120%AMI) 140% Tier 20.5% 18.9% 17.6% 21.7% 17.3% (up to 140% AMI) In addition, the Governor's housing emergency proclamation is being implemented to accelerate and prioritize 201 H affordable housing developments. These are developments that construct 50% + 1 affordable housing units. Bottomline, there is no economic or political justification for passing Bill 160, Draft 2. Flooding The county council should consider and apply Kona Community Development Plan Chapter 4.3 "Environmental Resources", Section 2. "Overall Strategy", paragraph b. "Turning Stormwater Management into an Asset. If the mauka lands are able to mitigate some of the surface stormwater runoff, there is still a challenge in managing stormwater runoff because of Kona's steep topography and undefined drainageways. The objective is to identify the drainageways where the major stormflows would be directed, provide a buffer to these drainageways to account for our imperfect knowledge, and to design these drainageways to function as recreational or open space amenities (e.g., linear parks). The following KCDP Policies and Actions must be considered and applied when considering Bill 169, Draft 2. Policy ENV-1.7: Flood Corridors. The County's Central Environmental Resources Inventory (see Policy ENV-1.1) should include the FederalEmergency Management Agency (FEMA) Flood Insurance Rate Map (FIRM) 100-year floodplains, as well as planned natural flow ways identified by the Stormwater Management Program (see Policy PUB —4.7), where predevelopment offsite runoff from existing and future development will be directed. Collectively, the FIRM floodplains and the planned flow ways may serve as open space amenities, such as linear parks and/or greenbelts between urbanized areas. Action ENV-1.7a: Develop priorities and financing strategies to improve accuracy and comprehensiveness of flood mapping (DPW, 2-3). Action ENV-1.7b: Budget and hire contractor to study feasibility of regional stormwater management systems, such as flow ways (DPW, PD, 2-3). Action ENV-1.7c: Identify corridors to be recommended for public open space pursuant to Policy ENV-2.2 Open Space Network Program (PD, DPW, 3-5) Policy ENV-1.9: Improvements Proposed Adjacent to Drainage Facilities. New construction, improvements to repetitive loss structures, and substantial improvements proposed adjacent to drainage facilities outside of the special flood hazard areas identified on the Flood Insurance Rate Maps shall be subject to review and approval of the director of public works in compliance with HCC 27-25. Action ENV-1.9a: Supports Chapter 27 pursuant to Policy ENV-1.9 (on-going) Cultural Resources The county council must consider Policy CR-1.1, "Cultural Resources Commission (CRC): "Cultural Resources Commission (CRC). The Action Committee should work closely with the County of Hawaii Cultural Resources Commission to implement the Kona CDP's goals, policies, and actions for Kona, along with the General Plan..." I hope the county council will respect the recommendation of the Cultural Resources Commission and the KCDP Action Committee. Please respect the constitutionally- protected rights and resources of kanaka maoli. For these reasons, as well as many more involving infrastructure, please vote against Bill 169, Draft 2. Mahalo, Chuck Flaherty • ••.....„..„,„:„.,..1., 3 Np= E .1. i „ ., ‘ ,...,„ ,,,,,, ,,,,,:, .. .,. .3 )0.. , ,•,,,,, , , ,, , ,,....::,,,,t,..t.v, .nip+. )0.1i '12:J 1 i .. ° 1. ... :'wKE L R 'y R T,-, W=::b , .sA p v. e. : 4;p='\ .. " , . y N ASSOCy,,fie•s.3:\ •''':i. `'yq,.;�'Jz:,::';,,,,;,. ..,•' ,.a,.- ,A;•,• s...1. ,':i'v,,.u'fs`S?',i'\iiY j.^.,"'-�•^Y, _ r- • ry ;si':." S'3/ :h • i a is:: �u;:, , s r�';� • RESIDENTIAL NEXUS ANALYSIS Honolulu, Hawaii Prepared for City and County of Honolulu Prepared by: Keyser Marston Associates, Inc. September 2015 ..::a .: .� :t'x-:.i5 .7{ .,y°-:-=1,. �. —�,ii.k?,'„:E'.w?1= ii= eg TABLE OF CONTENTS Page I. EXECUTIVE SUMMARY 1 II. INTRODUCTION AND OVERVIEW 10 III. NEXUS ANALYSIS 14 A. MARKET RATE UNITS AND HOUSEHOLD INCOME 14 B. THE IMPLAN MODEL 29 C. THE KMA JOBS HOUSING NEXUS MODEL 32 D. MITIGATION COSTS 48 ADDENDUM: ADDITIONAL BACKGROUND AND NOTES ON SPECIFIC ASSUMPTIONS 58 APPENDIX 1: MARKET SURVEY 60 APPENDIX 2: SUPPORTING TECHNICAL ANALYSIS TABLES 70 I. EXECUTIVE SUMMARY Keyser Marston Associates (KMA) prepared this residential nexus analysis for the City and County of Honolulu pursuant to a contractual agreement. This Executive Summary contains a concise overview of the residential nexus analysis; full documentation of the analysis is contained in the body of the Report and its Appendices. A. Residential Nexus Analysis A residential nexus analysis demonstrates and quantifies the impact of new market rate housing development on the demand for affordable housing. The underlying nexus concept is that the newly constructed market rate units represent net new households in Honolulu. These households represent new income in Honolulu that will consume goods and services, either through purchases of goods and services or`consumption' of government services. New consumption translates to jobs; a portion of the jobs are at lower compensation levels; low compensation jobs relate to lower income households that cannot afford market rate units in Honolulu and therefore need affordable housing. The City and County of Honolulu has requested this Residential Nexus Analysis in conjunction with the consideration of potential inclusionary requirements applicable to new residential development in Oahu as one component of the proposed Housing Oahu: Islandwide Housing Strategy. The purpose of this Residential Nexus Analysis is to provide information about the impact that new residential development has on the need for affordable housing and to determine inclusionary housing percentage and in-lieu fee requirements that are proportionate to these impacts and sufficient to fully mitigate them. 1. Impact Methodology and Models Used The analysis is performed using two models. The IMPLAN model is an industry accepted, commercially available model developed over 30 years ago to quantify the impacts of changes in a local economy, including the employment impacts of changes in personal income. The input into the IMPLAN model is net new personal income in Honolulu available for expenditures; the IMPLAN model then estimates a distribution of expenditures and ultimately produces a quantification of jobs generated by industry. IMPLAN is based on a similar methodology to the Hawaii's State Input Output Study developed by the Department of Business Economic Development and Tourism. The analysis uses the IMPLAN data set for Honolulu. The KMA Jobs Housing Nexus model, which was initially developed over 25 years ago to analyze the income structure of job growth, is used to determine the household income of new employee households and identify how many are in five housing affordability tiers ranging from Extremely Low-Income up through 140% of Area Median Income (AMI). Keyser Marston Associates, Inc. Page 1 \\Sf-fs 2\wp\14\14100\002\001-002.d o cx Nexus Analysis Concept 40 4 • newly constructed units . k101 • new households , 11111 • new expenditures on goods and services 1110 • new jobs, a share of which are low paying 110 s • new lower income households 4 • new demand for affordable units . To illustrate the linkages by looking at a simplified example, we can take an average household that buys a house at a certain price. From that price, we estimate the gross income of the household (from mortgage rates and lending practices) and the portion of income available for expenditures. Households will "purchase" or consume a range of goods and services, such as purchases at the supermarket or services at the bank. Purchases in the local economy in turn generate employment. The jobs generated are at different compensation levels. Some of the jobs are low paying and as a result, even when there is more than one worker in the household, there are some lower and middle-income households who cannot afford market rate housing in Honolulu. An underlying assumption of the analysis is that households that purchase or rent new units represent net new households in Honolulu. The nexus does not make the argument that construction of new units is solely responsible for population and household growth. Household growth in Honolulu occurs through a combination of natural increases in population and relocations from off-island. Construction of new residential units is a major contributing cause to population and household growth because without new housing supply, population and household growth would not continue to occur over a sustained period. In the short-term, population growth may occur without additions to the housing supply through accommodating additional people within the existing housing stock. However, over the long-term, households would not continue to relocate to Honolulu from off-island if they could not find adequate housing available. Without construction of new housing, out-migration could also become more of a factor offsetting natural increases in population as households seek places where housing is more available. Families may respond to a lack of adequate housing by delaying childbearing or having fewer children. Recent college graduates born in Honolulu may decide not to return based on challenges in finding adequate housing. Keyser Marston Associates, Inc. Page 2 \\Sf-fs2\wp\14\14100\002\001-002.d ocx 2. Market Survey and Residential Prototypes The first step of the nexus analysis is to identify residential prototypes that are representative of what is generally being built by the private marketplace in Honolulu. KMA developed programmatic assumptions in consultation with the City and County of Honolulu for five residential prototypes—four ownership prototypes and one rental prototype. KMA then undertook a market survey of projects covering these prototypes to estimate sales prices and rent levels for the prototype units. The prototypes are designed to be representative of averages for residential development activity occurring island-wide as described in the Appendix 1 market survey. The prototypes are summarized in the following table. Prototypical Residential Units Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo(PUC) Apartment Avg. Unit Size 1,700 SF 1,200 SF 1,000 SF 1,000 SF 900 SF Avg. Sales Price/Rent $700,000 $575,000 $525,000 $700,000 $2,500/mo. From the sales prices and rent levels, household income is determined using assumptions with respect to a share of income spent on housing and housing purchase terms. For ownership units, 37% of income is spent on housing (including mortgage payments, property taxes, home owner association dues, and insurance) based on the current average for new purchase home loans being underwritten in Honolulu. Renters are assumed to spend 30% of their income on rent, a relationship commonly used in housing policy to establish affordable rent levels relative to income. Gross household income is adjusted to a net amount available for expenditures after deducting the portion of income dedicated to income taxes, contributions to Social Security and Medicare, savings, and repayment of household debt. Housing costs are not deducted as part of this adjustment step because they are addressed separately as expenditures within the IMPLAN model. In addition, an adjustment is made to account for rental vacancy and a share of ownership units likely to be used as second homes and occupied only part of the year. The adjusted household income available for expenditures becomes the input into the IMPLAN model. As a result, household income and expenditures associated with each of the prototypes is as follows: Keyser Marston Associates, Inc. Page 3 \\Sf-fs2\wp\14\14100\002\001-002.docx Household Income and Expenditures Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo (PUC) Apartment Gross Household Income $115,000 $101,000 $95,000 $127,000 $100,000 Percent Income available for 67% 71% 72% 67% 65% Expenditures Spending adjustment for vacancy/ 99% 96% 96% 96% 95% 2nd homes occupied part of year Household Income Available for Expenditures $76,300 $68,800 $65,700 $81,700 $61,800 [Input to IMPLAN model] The nexus analysis is conducted on 100-unit project modules (i.e., 100 new households) for ease of presentation and to avoid awkward fractions. 3. IMPLAN Model Results The IMPLAN model was applied to link household income to job growth occurring in Honolulu. IMPLAN data sets are available for each county in the United States and are tailored to reflect the economic base in each area. The analysis uses the IMPLAN data set for Honolulu. The IMPLAN model distributes spending among various types of goods and services based on data from the Consumer Expenditure Survey and the Bureau of Economic Analysis Benchmark input-output study, to estimate employment generated. Job creation, driven by increased demand for products and services, is projected for each of the industries that will serve the new households. The employment generated by this new household spending is summarized in the following table. Jobs Generated Per 100 Units Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo(PUC) Apartment Annual Household $7,630,000 $6,880,000 $6,570,000 $8,170,000 $6,180,000 Expenditures (100 Units) Total Jobs Generated per 67.1 60.5 55.7 71.8 54.3 IMPLAN, 100 Units Net New Jobs after 17% reduction for declining 55.7 50.2 46.2 59.6 45.1 industries The IMPLAN model quantifies jobs generated at establishments that serve new residents directly (i.e. supermarkets, banks or schools), jobs generated by increased demand at firms which service or supply these establishments (wholesalers, janitorial contractors, accounting Keyser Marston Associates, Inc. Page 4 \\Sf-fs2\wp\14\14100\002\001-002.docx firms, or any jobs down the service/supply chain from direct jobs), and jobs generated when the new employees spend their wages in the local economy and generate additional jobs. Retail, restaurants, and health care represent the largest share of jobs generated by household expenditures. Employment estimates represent net new jobs after making a 17% downward adjustment to the IMPLAN employment estimates based on the expectation that a portion of jobs will be filled by existing workers who already have housing. The 17% adjustment is based upon job losses in declining sectors of the local economy over a historic period. "Downsized" workers from declining sectors are assumed to fill a portion of the new jobs in sectors that serve residents. 4. Compensation Levels of Jobs and Household Income The output of the IMPLAN model —the numbers of jobs by industry— is then entered into the Keyser Marston Associates jobs housing nexus analysis model to quantify the compensation levels of new jobs and the income of the new worker households. The KMA model sorts the jobs by industry into jobs by occupation, based on national data, and then attaches local wage distribution data to the occupations, using recent data for Honolulu from the Bureau of Labor Statistics Occupational Employment Survey. Further description is provided in Section III. C. The KMA model makes a conversation from number of employees to the number of employee households, recognizing that there is, on average, more than one worker per household, and thus the number of housing units in demand for new workers is reduced. The calculation is shown in the table below. For purposes of the adjustment from jobs to housing units, the average of 1.92 workers per working household in Honolulu is used, which is a higher number of workers per household than in other jurisdictions KMA has performed similar analyses. Application of the 1.92 factor effectively assumes the existing pattern of high numbers of workers per housing unit will continue and result in a reduced need for affordable units. Adjustment from No. of Workers to No. of Households Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo(PUC) Apartment Net New Jobs 55.7 50.2 46.2 59.6 45.1 Divide by No. of Workers per 1.92 1.92 1.92 1.92 1.92 Worker Household in Honolulu Net new worker households 29.0 26.2 24.1 31.1 23.5 The output of the model is the number of new worker households by income level (expressed in relation to the Area Median Income, or AMI) attributable to the new residential units and new households in Honolulu. Five categories are addressed: Extremely Low(under 30% of AMI), Very Low (30% to 50% of AMI), Low (50% to 80% of AMI), Moderate (80% to 120% of AMI), and a "140% AMI Tier" representing household incomes from 120% to 140% of AMI. Keyser Marston Associates, Inc. Page 5 \\Sf-fs2\wp\14\14100\002\001-002.docx Following are the numbers of worker households by income level associated with the Honolulu prototype units. New Worker Households by Income Level per 100 Market Rate Units Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo(PUC) Apartment Extr. Low(0% -30%AMI) 4.0 3.6 3.2 4.3 3.2 Very Low(30% - 50%AMI) 7.7 7.0 6.4 8.3 6.2 Low(50%-80%AMI) 8.0 7.2 6.6 8.6 6.5 Moderate(80% - 120%AMI) 5.0 4.5 4.2 5.3 4.0 Subtotal through 120% AMI 24.7 22.2 20.4 26.4 20.0 140% Tier(120% -140% AMI) 1.2 1.1 1.0 1.3 1.0 Subtotal through 140% AMI 25.8 23.3 21.4 27.7 20.9 Greater than 140%AMI 3.2 2.9 2.7 3.4 2.6 Total, New Households 29.0 26.2 24.1 31.1 23.5 The above findings represent the number of new affordable units required to offset the new affordable housing demand associated with services to each 100 new market rate residential units. 5. Inclusionary Percentages Supported Nexus findings regarding the number of affordable units needed per 100 market rate units can be converted to a percentage of units provided on-site within a project that would fully mitigate the affordable housing impacts. The percentages are calculated including both market rate and affordable units (for example, 25 affordable units per 100 market rate units translates to a project of 125 units; 25 affordable units out of 125 units equals 20%). Each tier is cumulative, or inclusive of the tiers above. The purpose of showing the figures on a cumulative basis is so they can be readily compared to potential inclusionary requirements that may be considered. As an example, for new single family projects, the analysis indicates that an inclusionary requirement of 19.8% with affordable units available to households earning up to 120% of AMI would be sufficient to mitigate the affordable housing needs of service worker households earning up through 120% of AMI. The percentages represent the inclusionary requirement that would be sufficient to fully offset the increased affordable housing need from the services and service workers that support the new residential development. Keyser Marston Associates, Inc. Page 6 \\Sf-fs2\wp\14\14100\002\001-002.docx Cumulative Inclusionary Percentage to Mitigate Increased Affordable Housing Need Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo (PUC) Apartment Extr. Low(up to 30%AMI) 3.8% 3.5% 3.1% 4.1% 3.1% Very Low(up to 50%AMI) 10.5% 9.5% 8.8% 11.1% 8.7% Low(up to 80%AMI) 16.5% 15.1% 14.0% 17.4% 13.8% Moderate (up to 120%AMI) 19.8% 18.2% 16.9% 20.9% 16.7% 140%Tier(up to 140%AMI) 20.5% 18.9% 17.6% 21.7% 17.3% 6. Impact Fee Levels Supported by the Nexus Analysis The last step in the analysis puts a dollar amount on the cost of mitigating the affordable housing impacts. The conclusions of the nexus analysis, expressed as the number of worker households by income affordability category, are linked to the cost of delivering housing to the households in need. Each income or affordability tier is associated with a subsidy needed to produce and deliver a unit at the specified affordability level; this subsidy is referred to as the `affordability gap.' Affordability gaps are calculated for each of the five affordable tiers. The analysis assumes households earning less than 80% of Area Median Income will be assisted in rental units, while households earning between 80% and 140% of Area Median Income will be assisted in ownership units. The resulting affordability gaps are as follows: • $367,300 for households in the under 30% AMI category; • $288,300 for households in the 30% to 50% AMI category; • $169,300 for households in the 50% to 80% AMI category; • $69,850 for households in the 80% to 120% AMI category; and • $0 (no affordability gap) for households in the 120% to 140% AMI category. No affordability gap is indicated for the 140% AMI Tier based on sales prices affordable to this income level and development costs for affordable townhome units'. When the affordability gap conclusions for each income tier are linked to the number of affordable units required per 100 market rate units and divided by 100 units, the result is a Total Nexus Cost per new market rate residential unit. The results per unit are: ' Development costs are higher for other for-sale unit types such as high-rise.There would be an affordability gap associated with providing 140%AMI affordable units in other more expensive product types. Keyser Marston Associates, Inc. Page 7 \\Sf-fs2\wp\14\14100\002\001-002.docx Nexus Cost Per Market Rate Unit Affordability Single Low-Rise Mid-Rise High-Rise Rental Income Category Gap Family Townhomes Condo Condo(PUC) Apartment Ext. Low(30% - 50%AMI) $367,300 $14,600 $13,200 $11,900 $15,700 $11,800 Very Low(30% - 50%AMI) $288,300 $22,200 $20,100 $18,300 $23,800 $18,000 Low(50%-80%AMI) $169,300 $13,500 $12,200 $11,200 $14,500 $11,000 Moderate (80%-120%AMI) $69,850 $3,500 $3,100 $2,900 $3,700 $2,800 140% Tier(120%-140% AMI) None $0 $0 $0 $0 $0 Total Nexus Costs $53,800 $48,600 $44,300 $57,700 $43,600 The chart below illustrates how the above nexus costs per unit are calculated: Calculation of Nexus Cost Per Market-Rate Unit ffordable. _ Affordability units required - Nexus cost gap per X per 100 - per market- affordable r unit market-rate rate unit units The Total Nexus Costs, or Mitigation Costs, indicated above, may also be expressed on a per square foot level. The results per square foot of building area (net rentable or sellable Sq.Ft.) are as follows: Total Nexus Cost Per Sq.Ft. of Building Area Affordability Single Low-Rise Mid-Rise High-Rise Rental Income Category Gap Family Townhomes Condo Condo(PUC) Apartment Prototype Size 1,700 SF 1,200 SF 1,000 SF 1,000 SF 900 SF Ext. Low(30%- 50%AMI) $367,300 $8.60 $11.00 $11.90 $15.70 $13.10 Very Low(30% -50%AMI) $288,300 $13.10 $16.80 $18.30 $23.80 $20.00 Low(50%-80%AMI) $169,300 $7.90 $10.20 $11.20 $14.50 $12.20 Moderate (80%-120%AMI) $69,850 $2.10 $2.60 $2.90 $3.70 $3.10 140% Tier(120%-140%) none $0.00 $0.00 $0.00 $0.00 $0.00 Total Nexus Costs $31.70 $40.60 $44.30 $57.70 $48.40 These costs express the total linkage or nexus costs for the five prototype developments in Honolulu. These total nexus costs represent the cost of creating new affordable units to offset increased affordable housing needs associated with new market-rate residential development. The totals are not recommended levels for fees; many other policy considerations may be brought to bear in selecting appropriate in-lieu fee requirements. Keyser Marston Associates, Inc. Page 8 \\S f-fs 2\w p\14\14100\00 2\001-0 0 2.d o c x The flow chart below provides a graphical illustration of the nexus analysis. Overview of Nexus Analysis [Figures in Chart Apply to High-Rise Condo Prototype] Startinci Point: i 17 Jobs its Distinguish 100 New MarketSi t rs Serving Housing Need by Rate Units 1 Residents Affordability Tier High-Rise Condo Sales (71.8 Jobs in retail; 26.4 units needed Price:$700,000 restaurants,other services through 120%AMI estimated using IMPLAN) (based on worker compensation I I — levels for Honolulu) 100 New Less: 17% Supported adjustment for net new jobs . Households Inclusionary Avg. Income:$127,000/Yr after losses in declining sectors o g ` Percent:20.91� ($12.7 Million for 100 HHs) 71.$—12.2=59.6 jabs estimated based on sales price € • =26.E air-units/(100 mkt+26A aft.units) Adjust from °^�, No.Workers to Households Multiply by Net Cost New Demand for at 1.92 Workers per Household Goods and To Produce Affordable Units 59.6 jobs/1.92=31.1 HH (Affordability Gap) Services ($8.2 M in new spending after deductions for taxes, Housing Need for 1 NexusFinding; 1 etc.) j Workers in Sectors itioation Cost Serving Residents , r„t 31.1 units at all income levels ` $57.73 per sq. ft. Note:figures in chart apply to High-Rise Condo prototype. Keyser Marston Associates, Inc. Page 9 \\Sf-fs2\wp\14\14100\002\001-002.docx II. INTRODUCTION AND OVERVIEW This report documents and quantifies the linkages between new market-rate residential development in Honolulu and the demand for additional affordable housing. The analysis, which demonstrates support for an affordable housing requirement, has been prepared by Keyser Marston Associates (KMA) for the City and County of Honolulu in accordance with a contractual agreement. Analyses of the impacts of new development are called linkage or nexus analyses. This nexus analysis establishes inclusionary requirements and fee levels that are proportionate to the impact that new market rate residential development has on the need for affordable housing. The City's Draft Housing Oahu: Islandwide Housing Strategy proposes that a new inclusionary housing requirement be considered as one of many strategies to produce affordable housing islandwide. Under the strategy's proposed requirements, residential projects with 10 or more units would be required to set-aside a percentage of units as affordable, construct units offsite, or pay a fee in-lieu of providing units. This analysis is intended to provide information to assist in the design of proposed requirements by identifying a set of affordable housing requirements that are proportionate to the impact new market rate residential development has on the need for affordable housing and which are reflective of the affordable housing needs of workers employed in services to new market rate residential development. The Nexus Concept At its most simplified level, the underlying nexus concept is that the newly constructed units represent net new households in Honolulu. These households represent new income in Honolulu that will consume goods and services, either through purchases of goods and services or "consumption" of governmental services. New consumption translates to jobs; a portion of the jobs are at lower compensation levels; low compensation jobs relate to lower income households that cannot afford market rate units in Honolulu and therefore need affordable housing. Purpose and Use of This Study The City and County of Honolulu requested this Residential Nexus Analysis in conjunction with the consideration of potential inclusionary requirements applicable to new residential development as a component of the proposed Housing Oahu: Islandwide Housing Strategy. The purpose of this Residential Nexus Analysis is to provide information about the impact that new residential development has on the need for affordable housing and determine inclusionary housing percentage and in-lieu fee requirements proportionate to these impacts and which are sufficient to mitigate them. We caution against the use of this study, or any impact study for that matter, for purposes beyond the intended use. All impact studies are limited and imperfect, but can be helpful for understanding the externalities created by new development. The nexus analysis presented in this report is an impact analysis only and the nexus amounts are not recommended requirements. Keyser Marston Associates, Inc. Page 10 \\Sf-fs2\wp\14\14100\002\001-002.docx Methodology and Models Used The methodology or analysis procedure for this nexus analysis starts with the sales price or rental rate of a new market rate residential unit, and moves through a series of linkages to the gross income of the household that purchased or rented the unit, the income available for expenditures on goods and services, the jobs associated with the purchases and delivery of those services, the income of the workers doings those jobs, the household income of the workers and, ultimately, the affordability level of the housing needed by the worker households. The steps of the analysis from household income available for expenditures to jobs generated were performed using the IMPLAN model, a model widely used for the past 35 years to quantify the impacts of changes in a local economy, including employment impacts from changes in personal income. From job generation by industry, KMA used its own jobs housing nexus model to quantify the income of worker households by affordability level. To illustrate the linkages by looking at a simplified example, we can take an average household that buys a house at a certain price. From that price, we estimate the gross income of the household (from mortgage rates and lending practices) and the portion of income available for expenditures. Households will "purchase" or consume a range of goods and services, such as purchases at the supermarket or services at the bank. Purchases in the local economy in turn generate employment. The jobs generated are at different compensation levels. Some of the jobs are low paying and as a result, even when there is more than one worker in the household, there are some lower and middle-income households who cannot afford market rate housing in Honolulu. The IMPLAN model quantifies jobs generated at establishments that serve new residents directly (e.g., supermarkets, banks or schools),jobs generated by increased demand at firms which service or supply these establishments, and jobs generated when the new employees spend their wages in the local economy and generate additional jobs. The IMPLAN model estimates the total impact combined. Net New Underlying Assumption An underlying assumption of the analysis is that households that purchase or rent new units represent net new households in Honolulu. If purchasers or renters have relocated from elsewhere in Honolulu, vacancies have been created that will be filled. An adjustment to new construction of units would be warranted if Honolulu were experiencing demolitions or loss of existing housing inventory. However, the rate of housing unit removal is so low as to not warrant an adjustment or offset. On an individual project basis, if existing units are removed to redevelop a site to higher density, then there could be a need for recognition of the existing households in that all new units might not represent net new households, depending on the program design and number of units removed relative to new units. Keyser Marston Associates, Inc. Page 11 \\Sf-fs2\wp\14\14100\002\001-002.docx The nexus does not make the argument that construction of new units is solely responsible for population and household growth. Household growth in Honolulu occurs through a combination of natural increases in population and relocations from off-island. Construction of new residential units is a major contributing cause to population and household growth because without new housing supply, population and household growth would not continue to occur over a sustained period. In the short-term, population growth may occur without additions to the housing supply through accommodating additional people within the existing housing stock. However, over the long-term, households would not continue to relocate to Honolulu from off- island if they could not find adequate housing available. Without construction of new housing, out-migration could also become more of a factor offsetting natural increases in population as households seek places where housing is more available. Families may also respond to a lack of adequate housing by delaying childbearing or having fewer children. Since the analysis addresses net new households in Honolulu and the impacts generated by their consumption expenditures, it quantifies net new demands for affordable units to accommodate new worker households. As such, the impact results do not address nor in any way include existing deficiencies in the supply of affordable housing. Geographic Area of Impact The analysis quantifies impacts occurring within the City and County of Honolulu. The majority of jobs related to services to new households are anticipated to be located on Oahu. The IMPLAN model computes the jobs generated in Honolulu /the island of Oahu and sorts out any jobs located off-island. The KMA Jobs Housing Nexus Model is then used to analyze the income structure of the jobs and their worker households. In summary, the KMA nexus analysis quantifies all job impacts occurring within Honolulu and related worker households. Market Rate Residential Project Types Five prototypical residential project types were selected for analysis in this nexus study. The prototypes were intended to represent the range of product types currently being built in Honolulu or which are expected in the future including: • Single Family • Low-Rise Townhomes2 • Mid-Rise Condo • High-Rise Condo (PUC) • Rental Apartments 2 The Townhome prototype is typically all wood frame construction and can include conventional townhomes and other similar all wood frame prototypes such as stacked flats. Keyser Marston Associates, Inc. Page 12 \\Sf-fs2\wp\14\14100\002\001-002.docx Affordability Tiers The nexus analysis addresses the following five income or affordability tiers: • Extremely Low Income (under 30% of Area Median Income or AMI) • Very Low Income (30% to 50% AMI) • Low Income (50% to 80% AMI) • Moderate (80% to 120% AMI) • 140% AMI Tier (120% to 140% AMI) The analysis includes the 140% AMI Tier representing households from 120% - 140% of AMI given that Honolulu's existing unilateral agreement rules address housing needs up to 140% of AMI. Report Organization The report is organized into four sections as follows: • Section A. presents information regarding the prototypical new market rate residential units and the estimated household income of purchases or renters of those units. • Section B. describes the IMPLAN model which is used in the nexus analysis to translate household income into the estimated number of jobs in retail, restaurants, healthcare, and other sectors serving new residents. • Section C. presents the linkage between employment growth associated with residential development and the need for new lower income housing units required in each of five income categories. • Section D. quantifies the nexus or mitigation cost based on the cost of delivering affordable units to new worker households in each of the five lower income categories. Keyser Marston Associates, Inc. Page 13 \\Sf-fs2\wp\14\14100\002\001-002.docx III. NEXUS ANALYSIS A. MARKET RATE UNITS AND HOUSEHOLD INCOME This section describes the prototypical market rate residential units and the income of the purchaser and renter households. Market rate prototypes are representative of new residential units currently being built in Honolulu or that are likely to be built in Honolulu over the next several years. Household income is estimated based on the amount necessary for the mortgage or rent payments associated with the prototypical new market rate units and becomes the basis for the input to the IMPLAN model described in Section B of this report. These are the starting points of the chain of linkages that connect new market rate units to incremental demand for affordable residential units. This section provides a summary of the prototypes and household income. More description and supporting tables are provided in Appendix 1. Recent Housing Market Activity and Prototypical Units KMA identified five residential prototypes in consultation with City staff; these prototypes are representative of the types of development that are being built in Honolulu today and expected to be built in the coming years. KMA then undertook a market survey of new construction projects covering these prototypes in fall 2014, as well as obtained data on sales of existing homes in Honolulu, focusing on units built since 1990. Further discussion of the market survey is included in Appendix 1. The results of the market survey and the selection of five prototypes are summarized in the table below. The main objective of the survey was to establish current sales prices or rents per unit and per square foot for the various residential project types recently developed, or expected to be developed in the future, in Honolulu. Table A-1 at the end of this section provides a more detailed summary of the five market rate prototypes. It is important to note that the prototypes analysis is intended to reflect average or typical residential projects in the Honolulu market rather than any specific project. It would be expected that specific projects would vary to some degree from the prototypes. In summary, the prototypes tested in the nexus analysis are as follows: Prototypical Residential Units Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo (PUC) Apartment Avg. Unit Size 1,700 SF 1,200 SF 1,000 SF 1,000 SF 900 SF Avg. Sales Price/Rent $700,000 $575,000 $525,000 $700,000 $2,500/mo. Keyser Marston Associates, Inc. Page 14 \\Sf-fs2\wp\14\14100\002\001-002.docx Income of Housing Unit Purchaser or Renter After the prototypes are established, the next step in the analysis is to determine the income of the purchasing or renting households in the prototypical units. Ownership Units To make the determination for ownership units, terms for the purchase of residential units used in the analysis are slightly less favorable than what can be achieved at the current time since current terms are not likely to endure. The selected terms for the analysis are: 20% down payment, 30 year fixed rate mortgage, 5.5% interest rate. The assumption of a 20% down payment is based on the median for purchase loans in Honolulu3. The interest rate at 5.5% reflects an estimate of the longer term average based on the experience over the past fifteen years.4 Tables A-2 through A-5 at the end of this section provide the details. All ownership product types include an estimate of homeowners' insurance, homeowner association dues, and property taxes which are included along with the mortgage payment as part of housing expenses for purposes of determining mortgage eligibility5. Incomes for households in the prototypical market rate units is estimated based on the amount necessary to afford mortgage payments and other housing costs. Data from Freddie Mac on the average "debt to income ratio"for new purchase mortgage loans originated in Honolulu of 37% is used to make the calculation. The 37% debt to income ratio means that housing costs, along with other debt, represent an average of 37% of household incomes. Apartment Units Household income for renter households is estimated based on the assumption that rent represents, on average, 30% of gross household income. The 30% factor was selected as it is widely used as a standard for relating income to affordable rent levels including by the U.S. Department of Housing and Urban Development. Selection of 30% represents a conservative 3 Median down payment at 20%is based on Freddie Mac data on its portfolio of mortgages within zip codes corresponding to Honolulu and is specific to principal residence purchase loans originated during the 2nd quarter of 2013, the most recent period available at the time the data was accessed. 4 Based on Freddie Mac Primary Mortgage Market Survey weekly average rates for 30 year fixed rate mortgages during the period from September 1999 through September 2014. 5 Housing expenses are combined with other debt payments such as credit cards and auto loans to compute a Debt To Income(DTI) ratio which is a key criteria used for determining mortgage eligibility. 6 New purchase loans in Honolulu have an average debt to income ratio of 37%based on data from Freddie Mac on its portfolio of mortgages within zip codes corresponding to Honolulu and specific to principal residence purchase loans originated during 2nd quarter of 2013,the most recent period available at the time the data was accessed. Debt to income ratio includes other forms of debt such as student loans,credit cards, and auto loans which suggests a ratio including only housing expenses would be less than 37%. Applying a ratio below 37%in the analysis would have produced a higher estimate of gross household income and higher resulting nexus findings;therefore, application of a 37%ratio represents a conservative assumption for purposes of the nexus analysis. Keyser Marston Associates, Inc. Page 15 \\Sf-fs2\wp\14\14100\002\001-002.docx approach for purposes of the nexus analysis because it produces a lower estimate of gross household income and lower resulting nexus conclusions than if the Census average of 16% of income spent on rent for renter households earning $100,000 and above were applied, the income category applicable to households in the prototypical new market rate rental units. While leasing agents and landlords may permit rental payments to represent a slightly higher share of total income, use of the 30% factor, which is representative of the average, is appropriate. Further, many renters will choose to spend less than 30% of their income on rent where possible, since, unlike an ownership situation, the unit is not viewed as an investment with value enhancement potential. The resulting relationship is that annual household income is 3.3 times annual rent. The estimated gross household incomes of the purchasers or renters of the prototype units are calculated in tables A-2 through A-6, and summarized below. Household Income Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo (PUC) Apartment Gross Household Income $115,000 $101,000 $95,000 $127,000 $100,000 Income Available for Expenditures The input into the IMPLAN model used in this analysis is the net income available for expenditures. To arrive at income available for expenditures, gross income must be adjusted for Federal and State income taxes, contributions to Social Security and Medicare, savings, and payments on household debt. Per KMA correspondence with the producers of the IMPLAN model (IMPLAN Group LLC), other taxes including sales tax, gas tax, and property tax are handled internally within the model as part of the analysis of expenditures. Housing costs are addressed separately, as described below, and so are not deducted as part of this adjustment step. Table A-7 at the end of this section shows the calculation of income available for expenditures. Income available for expenditures ranges from 67% to 72% of gross income for the ownership housing prototypes. The estimate is based on a review of data from the Internal Revenue Service and State of Hawaii Department of Taxation tax tables. Residents of the Single Family and High-rise Condo prototypes are estimated to pay an average of 12.4% of gross income in federal taxes based on data from the Internal Revenue Service applicable to households earning between $100,000 and $200,000 per year. Residents of the townhome and mid-rise condo prototypes are estimated to pay 8.8% of their income toward federal taxes, which is the average for households in the $75,000 to $100,000 income range'. State taxes are estimated to average 4% to 6°/0 of gross income based on tax rates per the State of Hawaii Department of 'Average tax rate for the$75,000 to$100,000 income range was applied to the townhome prototype despite estimated income slightly above$100,000 because the$75,000 to$100,000 range is likely more representative than the$100,000 to$200,000 category which covers a much broader range of incomes. Keyser Marston Associates, Inc. Page 16 \\Sf-fs2\wp\14\14100\002\001-002.docx Taxation. The employee share of FICA payroll taxes for Social Security and Medicare is 7.65% of gross income (conservatively assumes all earners in the household are within the $118,500 ceiling on income subject to Social Security taxes). Savings and repayment of household debt represent another necessary adjustment to gross income. Savings includes various IRA and 401 K type programs as well as non-retirement household savings and investments. Debt repayment includes auto loans, credit cards, and all other non-mortgage debt. Savings and repayment of debt are estimated to represent a combined 8% of gross income based on the 20 year average derived from United States Bureau of Economic Analysis data. The percentage of income available for expenditure for input into the IMPLAN model is prior to deducting housing costs. The reason is for consistency with the IMPLAN model which defines housing costs as expenditures. The IMPLAN model addresses the fact that expenditures on housing do not generate employment to the degree other expenditures such as retail or restaurants do, but there is some limited maintenance and property management employment generated. After deducting income taxes, Social Security, Medicare, savings, and repayment of debt, for purchasers of one of the new ownership prototypes, the estimated income available for expenditures ranges from 67% for the Single Family prototype to 72% for the mid-rise condo prototype. This is the factor used to adjust from gross income to the income available for expenditures. As indicated above, other forms of taxation such as property tax are handled internally within the IMPLAN model. Income available for expenditures for the prototypical renter household is based on the same evaluation, but a higher tax rate applies to a renter household. Renter households are assumed to pay higher average income tax rates because many renter households will not have the ability to itemize deductions on income tax returns without mortgage interest and real estate taxes to deduct. The result is that the renter household would have an estimated 65% of income available for expenditures. The rate of savings and debt repayment is assumed to be the same for the renter household as for households in the ownership prototypes. Another adjustment made to spending is to account for the use of a share of units as second homes which may be occupied for only a portion of the year. The adjustment is made using U.S. census data for Honolulu on the percentage of single family and multifamily units that are used as second homes. Household expenditures for single family homes are reduced by 1% on average, to account for the fact that some units may be used as second homes and left vacant part of the year. A 4% reduction to expenditures for the townhome and condo prototypes is made based on Census data for multifamily units in Honolulu which have a higher incidence of second home owners who do not occupy the unit year-round. For the apartment prototype, we apply a 5% adjustment for standard operational vacancy. Keyser Marston Associates, Inc. Page 17 \\Sf-fs2\wp\14\14100\002\001-002.docx Estimates of household income available for expenditures are presented in the following table: Household Income Available for Expenditure Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo (PUC) Apartment Gross Household Income $115,000 $101,000 $95,000 $127,000 $100,000 Percent Income available for 67% 71% 72% 67% 65% Expenditures Spending adjustment for vacancy/ 99% 96% 96% 96% 95% 2nd homes occupied part of year Household Income Available for Expenditures $76,300 $68,800 $65,700 $81,700 $61,800 [Input to IMPLAN model] The nexus analysis is conducted on 100-unit building modules for ease of presentation, and to avoid awkward fractions. Tables A-8 and A-9 summarize the conclusions of this section and calculate the household income for the 100-unit building modules. This is the input into the IMPLAN model. Keyser Marston Associates, Inc. Page 18 \\Sf-fs2\wp\14\14100\002\001-002.docx TABLE A-1 SUMMARY OF MARKET RATE RESIDENTIAL PROTOTYPES RESIDENTIAL NEXUS ANALYSIS CITY AND COUNTY OF HONOLULU For-Sale Prototypes Rental Prototype 1 Prototype 2 Prototype 3 Prototype 4 Prototype 5 Detached Homes Townhomes Mid-Rise Condos (PUC) Apartments Average Unit Size 1,700 sf 1,200 sf 1,000 sf 1,000 sf 900 sf Market Rate Price/Rent $700,000 $575,000 $525,000 $700,000 $2,500 /mo. $/SF $412 /sf $479 /sf $525 /sf $700 /sf $2.78 /sf Source: Pricing estimated by KMA based on market survey Page 19 Prepared by Keyser Marston Associates, Inc. \\Sf-fs2\wp\14\14100\002\Honolulu Prototypes 10 3 14;Prototypes;8/31/2015; 10:25 AM TABLE A-2 PROTOTYPE 1: SINGLE FAMILY SALES PRICE TO INCOME RATIO RESIDENTIAL NEXUS ANALYSIS HONOLULU, HI Prototype 1 Single Family Sales Price $412/SF 1,700 SF ' $700,000 Mortgage Payment Downpayment @ 20% 20% 2 $140,000 Loan Amount $560,000 interest Kate 5.50% 3 Term of Mortgage 30 years Annual Mortgage Payment $38,200 Other Costs Property Taxes 0.31% of sales price° $2,200 HOA Dues/Maintenance $100 per month 5 $1,200 Homeowner Insurance 0.15% sale price 6 $1,100 Total Annual Housing Cost $42,700 % of Income Spent on Hsg 37% ' Annual Household Income Required $115,000 Sales Price to Income Ratio 6.1 Notes (1)Based on Market Survey. (2)Median down payment at 20%is based on Freddie Mac data on its portfolio of mortgages within zip codes corresponding to Honolulu and specific to principal residence purchase loans originated during 2nd quarter of 2013(most recent available). (3)Average interest rate for 30-year fixed rate mortgage over the past 15 years derived from Freddie Mac Primary Mortgage Market Survey(rounded down from average of 5.66%). (4)Effective tax rate assuming home exemption applies. (5)HOA dues for newer single family units are estimated at$100 based on projects covered in market survey which have HOA dues ranging from$0 to$257/mo. (6)Estimated from quote obtained from Liberty Mutual Insurance. (7)Based on Freddie Mac data on mortgages originated in Honolulu which reflect an average debt to income ratio of 37%including both housing expenses and other debt like auto loans and credit cards.Were other debt excluded,the ratio would be lower than 37%. Using a ratio of 37%produces conservative nexus results that are understated compared to results that would be produced using a lower ratio. Prepared by: Keyser Marston Associates, Inc. \\Sf-fs2\wp\14\14100\002\Honolulu Nexus model 8-31-15; 8/31/2015; dd Page 20 TABLE A-3 PROTOTYPE 2: LOW-RISE TOWNHOMES SALES PRICE TO INCOME RATIO RESIDENTIAL NEXUS ANALYSIS HONOLULU, HI Prototype 2 Low-Rise Townhomes Sales Price $479/SF 1,200 SF 1 $575,000 Mortgage Payment Downpayment @ 20% 20°%o 2 $115,000 Loan Amount $460,000 Interest Kate 5.50% 3 Term of Mortgage 30 years Annual Mortgage Payment $31,300 Other Costs Property Taxes 0.30% of sales price° $1,700 HOA Dues/Maintenance $300 per month 5 $3,600 Homeowner Insurance 0.15% sale price 6 $900 Total Annual Housing Cost $37,500 % of Income Spent on Hsg 37% 7 Annual Household Income Required $101,000 Sales Price to Income Ratio 5.7 Notes (1)Based on Market Survey. (2)Median down payment at 20%is based on Freddie Mac data on its portfolio of mortgages within zip codes corresponding to Honolulu and specific to principal residence purchase loans originated during 2nd quarter of 2013(most recent available). (3)Average interest rate for 30-year fixed rate mortgage over the past 15 years derived from Freddie Mac Primary Mortgage Market Survey(rounded down from average of 5.66%). (4)Effective tax rate assuming home exemption applies. (5)Estimated based on review of HOA dues for townhome resale listings and new townhomes identified in market survey. (6)Estimated from quote obtained from Liberty Mutual Insurance. (7)Based on Freddie Mac data on mortgages originated in Honolulu which reflect an average debt to income ratio of 37%including both housing expenses and other debt like auto loans and credit cards.Were other debt excluded,the ratio would be lower than 37%. Using a ratio of 37%produces conservative nexus results that are understated compared to results that would be produced using a lower ratio. Prepared by: Keyser Marston Associates, Inc. Page 21 \\Sf-fs2\wp\14\14100\002\Honolulu Nexus model 8-31-15; 8/31/2015; dd TABLE A-4 PROTOTYPE 3: MID-RISE CONDO SALES PRICE TO INCOME RATIO RESIDENTIAL NEXUS ANALYSIS HONOLULU, HI Prototype 3 Mid-Rise Condo Sales Price $525/SF 1,000 SF ' $525,000 Mortgage Payment Downpayment @ 20% 20% 2 $105,000 Loan Amount $420,000 interest Kate 5.50% 3 Term of Mortgage 30 years Annual Mortgage Payment $28,600 Other Costs Property Taxes 0.30% of sales price° $1,600 HOA Dues/Maintenance $350 per month 5 $4,200 Homeowner Insurance 0.15% sale price 6 $800 Total Annual Housing Cost $35,200 % of Income Spent on Hsg 37% 7 Annual Household Income Required $95,000 Sales Price to Income Ratio 5.5 Notes (1)Based on Market Survey. (2)Median down payment at 20%is based on Freddie Mac data on its portfolio of mortgages within zip codes corresponding to Honolulu and specific to principal residence purchase loans originated during 2nd quarter of 2013(most recent available). (3)Average interest rate for 30-year fixed rate mortgage over the past 15 years derived from Freddie Mac Primary Mortgage Market Survey(rounded down from average of 5.66%). (4)Effective tax rate assuming home exemption applies. (5)Estimated based on review of HOA dues for condo resale listings and condos identified in market survey. (6)Estimated from quote obtained from Liberty Mutual Insurance. (7)Based on Freddie Mac data on mortgages originated in Honolulu which reflect an average debt to income ratio of 37%including both housing expenses and other debt like auto loans and credit cards.Were other debt excluded,the ratio would be lower than 37%. Using a ratio of 37%produces conservative nexus results that are understated compared to results that would be produced using a lower ratio. Prepared by: Keyser Marston Associates, Inc. \\Sf-fs2\wp\14\14100\002\Honolulu Nexus model 8-31-15; 8/31/2015; dd Page 22 TABLE A-5 PROTOTYPE 4: HIGH-RISE CONDO (PUC) SALES PRICE TO INCOME RATIO RESIDENTIAL NEXUS ANALYSIS HONOLULU, HI Prototype 4 High-Rise Condo (PUC) Sales Price $700/SF 1,000 SF 1 $700,000 Mortgage Payment Downpayment @ 20% 20% 2 $140,000 Loan Amount $560,000 Interest Rate 5.50% 3 Term of Mortgage 30 years Annual Mortgage Payment $38,200 Other Costs Property Taxes 0.31% of sales price° $2,200 HOA Dues/Maintenance $450 per month 5 $5,400 Homeowner Insurance 0.15% sale price 6 $1,100 Total Annual Housing Cost $46,900 % of Income Spent on Hsg 37% Annual Income Required $127,000 Sales Price to Income Ratio 5.5 Notes (1)Based on Market Survey. (2)Median down payment at 20%is based on Freddie Mac data on its portfolio of mortgages within zip codes corresponding to Honolulu and specific to principal residence purchase loans originated during 2nd quarter of 2013(most recent available). (3)Average interest rate for 30-year fixed rate mortgage over the past 15 years derived from Freddie Mac Primary Mortgage Market Survey(rounded down from average of 5.66%). (4)Effective tax rate assuming home exemption applies. (5)Based on HOA dues for high-rise condo units in Honolulu from the Market Survey. (6)Estimated from quote obtained from Liberty Mutual Insurance. (7)Based on Freddie Mac data on mortgages originated in Honolulu which reflect an average debt to income ratio of 37%including both housing expenses and other debt like auto loans and credit cards. Were other debt excluded,the ratio would be lower than 37%.Using a ratio of 37%produces conservative nexus results that are understated compared to results that would be produced using a lower ratio. Prepared by: Keyser Marston Associates, Inc. Page 23 \\Sf-fs2\wp\14\14100\002\Honolulu Nexus model 8-31-15; 8/31/2015; dd TABLE A-6 PROTOTYPE 5: RENTAL APARTMENT RENT TO INCOME RATIO RESIDENTIAL NEXUS ANALYSIS HONOLULU, HI Prototype 5 Rental Apartment Market Rent Monthly $2.78/SF 900 SF ' $2,500 Annual $30,000 % of Income Spent on Rent 30% 2 (excludes utilities) Annual Household Income Required $100,000 Annual Rent to Income Ratio 3.3 Notes (1)Based on the results of the market survey. Represents rent levels applicable to new units. (2)Renter households are assumed to spend 30%of income on rent.The 30%figure is selected based on its frequent use in housing policy.While Census data for Honolulu indicates only 16%of households earning$100,000 or above spend 30%or more of their income on rent,30%produces results that are conservative or understated as compared to those that would be produced by selection of a lower figure. Prepared by: Keyser Marston Associates, Inc. \\Sf-fs2\wp\14\14100\002\Honolulu Nexus model 8-31-15; 8/31/2015; dd Page 24 TABLE A-7 INCOME AVAILABLE FOR EXPENDITURES' RESIDENTIAL NEXUS ANALYSIS HONOLULU, HI Prototype 1: Prototype 2: Prototype 3: Prototype 4: Prototype 5: Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo(PUC) Apartment Gross Income 100% 100% 100% 100% 100% Less: Federal Income Taxes 2 12.4% 8.8% 8.8% 12.4% 12.9% State Income Taxes 3 5% 5% 4% 5% 6% FICA Tax Rate 4 7.65% 7.65% 7.65% 7.65% 7.65% Savings&other deductions 5 8% 8% 8% 8% 8% Percent of Income Available 67% 71% 72% 67% 65% for Expenditures 6 (Input to IMPLAN model) Notes: I Gross income after deduction of taxes and savings. Income available for expenditures is the input to the IMPLAN model which is used to estimate the resulting employment impacts. Housing costs are not deducted as part of this adjustment step because they are addressed separately as expenditures within the IMPLAN model. 2 Reflects average tax rates(as opposed to marginal)applicable to estimated household incomes from tables A-2 to A-6 based on U.S. Internal Revenue Services,Tax Statistics,Tables 1.4 and 2.1.Figures are for the 2012 tax year,the most recent for which data is available. Homeowners are assumed to itemize deductions. Renters are assumed to take the standard deduction. 3 Average tax rates applicable to estimated household incomes from tables A-2 to A-6 estimated by KMA based on marginal rates per the State of Hawaii Department of Taxation and ratios of taxable income to gross income estimated based on U.S.Internal Revenue Service data. The higher average tax rates applicable to single or married filing separately tax filers is applied in the analysis so as to produce a conservative(likely understated)estimate. 4 For Social Security and Medicare.Conservatively assumes all income will be subject to Social Security taxes.The current ceiling on applicability of Social Security taxes is$118,500(ceiling applies per earner not per household). 5 Household savings including retirement accounts like 401k/IRA and other deductions such as interest costs on credit cards,auto loans,etc,necessary to determine the amount of income available for expenditures.The 8%rate used in the analysis is based on the average over the past 20 years computed from U.S. Bureau of Economic Analysis data,specifically the National Income and Product Accounts,Table 2.1 "Personal Income and It's Disposition." 6 Deductions from gross income to arrive at the net income available for expenditures are consistent with the way the IMPLAN model and National Income and Product Accounts(NIPA)defines income available for personal consumption expenditures.Deductions include income taxes,contributions to Social Security and Medicare,savings and household debt.The income available for expenditures does not reflect deductions for items accounted for as expenditures within the IMPLAN model iteslf including property taxes,sales taxes and housing expenditures based on guidance from IMPLAN. Prepared by: Keyser Marston Associates, Inc. Page 25 \\Sf-fs2\wp\14\14100\002\Honolulu Nexus model 8-31-15; 8/31/2015; dd TABLE A-8 FOR SALE PROTOTYPES: SALES PRICE TO INCOME SUMMARY RESIDENTIAL NEXUS ANALYSIS HONOLULU, HI 100 Unit Per Unit Per Sq.Ft. Building Module Page 1 of 2 PROTOTYPE 1: SINGLE FAMILY Units 100 Units Building Sq.Ft. (net salable area) 1,700 170,000 Sales Price $700,000 $412 $70,000,000 Sales Price to Income Ratio 6.1 6.1 Gross Household Income $115,000 $11,500,000 Income Available for Expenditure' 67% of gross $77,000 $7,710,000 Adjusted Expenditures/Second Homes2 1% adjustment $7,630,000 PROTOTYPE 2: LOW-RISE TOWNHOMES Units 100 Units Building Sq.Ft. (net salable area) 1,200 120,000 Sales Price $575,000 $479 $57,500,000 Sales Price to Income Ratio 5.7 5.7 Gross Household Income $101,000 $10,100,000 Income Available for Expenditure' 71% of gross $72,000 $7,170,000 Adjusted Expenditures/Second Homes2 4% adjustment $6,880,000 Prepared by: Keyser Marston Associates, Inc. \\Sf-fs2\wp\14\14100\002\Honolulu Nexus model 8-31-15; 8/31/2015;dd Page 26 TABLE A-8 FOR SALE PROTOTYPES: SALES PRICE TO INCOME SUMMARY RESIDENTIAL NEXUS ANALYSIS HONOLULU, HI 100 Unit Per Unit Per Sq.Ft. Building Module Page 2 of 2 PROTOTYPE 3: MID-RISE CONDO Units 100 Units Building Sq.Ft. (net salable area) 1,000 100,000 Sales Price $525,000 $525 $52,500,000 Sales Price to Income Ratio 5.5 5.5 Gross Household Income $95,000 $9,500,000 Income Available for Expenditure' 72% of gross $68,000 $6,840,000 Adjusted Expenditures/Second Homes2 4% adjustment $6,570,000 PROTOTYPE 4: HIGH-RISE CONDO (PUC) Units 100 Units Building Sq.Ft. (net salable area) 1,000 100,000 Sales Price $700,000 $700 $70,000,000 Sales Price to Income Ratio 5.5 5.5 Gross Household Income $127,000 $12,700,000 Income Available for Expenditure' 67% of gross $85,000 $8,510,000 Adjusted Expenditures/Second Homes2 4% adjustment $8,170,000 Notes: (1)Represents net income available for expenditures after income tax,payroll taxes,and savings.See Table A-8 for derivation. (2)Adjustment to expenditures based upon the expectation that a share of units may not be occupied year round because they are second homes.The adjustment is based upon American Community Survey data for Honolulu which identifies the portion of single family and multi- family units used as seasonal or vacation homes.Data is for 2005,the most recent period in which the breakout between single family and multifamily units used as vacation homes is provided.Second homes are assumed to be in use three months of the year. Source:See Tables A-2 to A-5. Prepared by:Keyser Marston Associates,Inc. Page 27 \\Sf-fs2\wp\14\14100\002\Honolulu Nexus model 8-31-15; 8/31/2015; dd TABLE A-9 NEW MARKET RATE RESIDENTIAL HOUSEHOLD SUMMARY RESIDENTIAL NEXUS ANALYSIS HONOLULU, HI 100 Unit Per Unit Per Sq.Ft. Building Module PROTOTYPE 5: RENTAL APARTMENT Units 100 Units Building Sq.Ft. (net rentable area) 900 90,000 Rent Monthly $2,500 $2.78/SF $250,000 Annual $30,000 $33.36/SF $3,000,000 Rent to Income Ratio 3.3 3.3 Gross Household Income $100,000 $10,000,000 Income Available for Expenditure' 65% of gross $65,000 $6,500,000 txpendltures adjusted tor vacancy 5°�° vacancy $62,000 $6,180,000 Notes: (1)Represents net income available for expenditures after income tax,payroll taxes,and savings.See Table A-8 for derivation. (2)Adjustment to expenditures to account for standard operational vacancy.Vacancy rates consistent with current figures for Honolulu based on the 2011-2013 American Community Survey. Source:Table A-6. Prepared by: Keyser Marston Associates, Inc. \\Sf-fs2\wp\14\14100\002\Honolulu Nexus model 8-31-15; 8/31/2015; dd Page 28 B. THE IMPLAN MODEL Consumer spending by residents of new housing units will create jobs, particularly in sectors such as restaurants, health care, and retail, which are closely connected to the expenditures of residents. The widely used economic analysis tool, IMPLAN (IMpact Analysis for PLANning), was used to quantify these new jobs by industry sector. IMPLAN Model Description The IMPLAN model is an economic analysis software package now commercially available through the IMPLAN Group, LLC. IMPLAN was originally developed by the U.S. Forest Service, the Federal Emergency Management Agency, and the U.S. Department of the Interior Bureau of Land Management and has been in use since 1979 and refined over time. It has become a widely used tool for analyzing economic impacts for a broad range of applications from major construction projects to natural resource programs. IMPLAN is based on an input-output accounting of commodity flows within an economy from producers to intermediate and final consumers. The model establishes a matrix of supply chain relationships between industries and also between households and the producers of household goods and services. Assumptions about the portion of inputs or supplies for a given industry likely to be met by local suppliers, and the portion supplied from outside the region or study area are derived internally within the model using data on the industrial structure of the region. The output or result of the model is generated by tracking changes in purchases for final use (final demand) as they filter through the supply chain. Industries that produce goods and services for final demand or consumption must purchase inputs from other producers, which in turn, purchase goods and services. The model tracks these relationships through the economy to the point where leakages from the region stop the cycle. This allows the user to identify how a change in demand for one industry will affect a list of over 400 other industry sectors. The projected response of an economy to a change in final demand can be viewed in terms of economic output, employment, or income. Data sets are available for each county and state, so the model can be tailored to the specific economic conditions of the region being analyzed. This analysis utilizes the 2012 IMPLAN data set for Honolulu which was released in December 2013. As will be discussed, much of the employment impact is in local-serving sectors, such as retail, eating and drinking establishments, and medical services. The vast majority of these jobs are expected to be located in Honolulu as most services to Honolulu residents cannot be readily provided from off- shore. Employment estimates from the IMPLAN model represents employment in Honolulu. Keyser Marston Associates, Inc. Page 29 \\Sf-fs2\wp\14\14100\002\001-002.docx Application of the IMPLAN Model to Estimate Job Growth The IMPLAN model was applied to link income to household expenditures to job growth. Employment generated by the household income of residents is analyzed in modules of 100 residential units to simplify communication of the results and avoid awkward fractions. The IMPLAN model distributes spending among various types of goods and services (industry sectors) based on data from the Consumer Expenditure Survey and the Bureau of Economic Analysis Benchmark input-output study, to estimate employment generated. Job creation, driven by increased demand for products and services, was projected for each of the industries that will serve the new households. The employment generated by this new household spending is summarized below. Jobs Generated Per 100 Units Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo (PUC) Apartment Annual Household $7,630,000 $6,880,000 $6,570,000 $8,170,000 $6,180,000 Expenditures, 100 Units Total Jobs Generated, 67.1 60.5 55.7 71.8 54.3 100 Units Table B-1 provides a detailed summary of employment generated by industry. The table shows industries sorted by projected employment. The Consumer Expenditure Survey published by the Bureau of Labor Statistics tracks expenditure patterns by income level. IMPLAN utilizes this data to reflect the pattern by income bracket. In the case of the Honolulu prototypes, the mid- rise condo is in the $75,000 to $100,000 category while all other prototypes fall into the $100,000 to $150,000 income category. Estimated employment is shown for each IMPLAN industry sector representing 1% or more of total employment. The jobs that are generated are heavily retail jobs,jobs in restaurants and other eating establishments, and in services that are provided locally such as health care. The jobs counted in the IMPLAN model cover all jobs, full and part time, similar to the U.S. Census and all reporting agencies (unless otherwise indicated). Keyser Marston Associates, Inc. Page 30 \\Sf-fs2\wp\14\14100\002\001-002.docx TABLE B-1 IMPLAN MODEL OUTPUT EMPLOYMENT GENERATED RESIDENTIAL NEXUS ANALYSIS HONOLULU,HI Per 100 Market Rate Units Prototype 2: Prototype 3: Prototype 4: Prototype 5: Prototype 1: Low-Rise Mid-Rise High-Rise Rental %of Single Family Townhomes Condo Condo(PUC) Apartment Jobs Household Expenditures(100 Market Rate Units)' $7,630,000 $6,880,000 $6,570,000 $8,170,000 $6,180,000 Jobs Generated by Industry 2 Retail Stores-Food and beverage 2.4 2.2 1.9 2.6 2.0 4% Retail Stores-General merchandise 2.3 2.1 1.7 2.4 1.8 3% Retail Stores-Motor vehicle and parts 1.3 1.2 1.0 1.4 1.1 2% Retail Stores-Clothing and accessories 1.2 1.1 0.9 1.3 1.0 2% Retail Stores-Miscellaneous 1.1 1.0 0.8 1.2 0.9 2% Retail Stores-Health and personal care 1.0 0.9 0.7 1.0 0.8 1% Retail Stores-Sporting goods,book, music 0.6 0.5 0.4 0.6 0.5 1% Retail Stores-Building and garden supply 0.5 0.5 0.4 0.6 0.4 1% Retail Stores-Electronics and appliances 0.4 0.3 0.3 0.4 0.3 1% Retail Stores-Gasoline stations 0.3 0.3 0.2 0.3 0.3 0% Retail Stores-Furniture and home furnishings 0_3 0_2 0_2 0_3 0_2 000% Subtotal Retail 11.3 10.2 8.6 12.1 9.1 17% Offices of physicians and dentists 3.8 3.4 3.3 4.1 3.1 6% Private hospitals 3.1 2.8 2.8 3.3 2.5 5% Home health care services 0.8 0.7 0.7 0.8 0.6 1% Nursing and residential care facilities 1.8 1.6 1.6 1.9 1.5 3% Medical and diagnostic labs and outpatient care 1.1 1_0 0_9 1_1 0_9 2% Subtotal Health Care 10.6 9.5 9.3 11.3 8.5 16% Food services and drinking places 7.9 7.1 6.7 8.5 6.4 12% Real estate including property management 2.3 2.0 2.4 2.4 1.8 3% Private household operations 2.0 1.8 1.7 2.2 1.7 3% Wholesale trade businesses 2.0 1.8 2.2 2.2 1.6 3% Individual and family services 1.5 1.4 1.2 1.6 1.2 2% Employment services 1.3 1.2 1.1 1.4 1.0 2% Civic,social,professional organizations 1.3 1.1 1.0 1.4 1.0 2% Banking and depository credit 1.1 1.0 0.9 1.2 0.9 2% Insurance carriers 1.1 1.0 0.9 1.2 0.9 2% Services to buildings and dwellings 1.1 1.0 0.9 1.1 0.9 2% Securities,investments,and related 1.0 0.9 0.8 1.1 0.8 2% Child day care services 1.0 0.9 0.7 1.1 0.8 2% Elementary and secondary schools 1.0 0.9 0.7 1.0 0.8 1% Personal care services 0.9 0.8 0.8 1.0 0.8 1% Other private educational services 0.9 0.8 0.6 1.0 0.7 1% Colleges,universities,and professional schools 0.8 0.8 0.6 0.9 0.7 1% Amusement parks,arcades,and gambling industries 0.8 0.7 0.6 0.8 0.6 1% Legal services 0.7 0.7 0.6 0.8 0.6 1% Automotive repair and maintenance 0.6 0.6 0.5 0.7 0.5 1% All Other 15.8 14.2 13.0 16.9 12.8 24% Total Number of Jobs Generated 67.1 60.5 55.7 71.8 54.3 100% I Estimated employment generated by expenditures of households within 100 prototypical market rate units.Employment estimates are based on the IMPLAN Group's economic model,IMPLAN,for Honolulu. 2 For Industries representing more than 1%of total employment. Page 31 Prepared by:Keyser Marston Associates,Inc. \\Sf-fs2\wp\14\14100\002\Honolulu Nexus model 8-31-15;8/31/2015;dd C. THE KMA JOBS HOUSING NEXUS MODEL This section presents a summary of the analysis linking the employment growth associated with residential development, or the output of the IMPLAN model (see Section B), to the estimated number of lower income housing units required in each of five income categories, for each of the five residential prototype units. Analysis Approach and Framework The analysis approach is to examine the employment growth for industries related to consumer spending by residents in the 100-unit modules. Then, through a series of linkage steps, the number of employees is converted to households and housing units by affordability level. The findings are expressed in terms of numbers of affordable units per 100 market rate units. The analysis addresses the affordable unit demand associated with single family, low-rise townhomes, mid-rise condos, high-rise condos and rental units in Honolulu. The table below shows the 2014 Honolulu Area Median Income (AMI), as well as the income limits for the five categories that were evaluated from 30% of AMI through 140% AMI. The income definitions used in the analysis are those published by the Department of Planning and Permitting, City and County of Honolulu (DPP). 2014 Income Limits for Honolulu Household Size (Persons) 1 2 3 4 5 6 + 30%AMI $20,150 $23,000 $25,900 $28,750 $31,050 $33,350 50%AMI $33,550 $38,350 $43,150 $47,900 $51,750 $55,600 80%AMI $53,700 $61,350 $69,000 $76,650 $82,800 $88,950 120%AMI $80,560 $92,020 $103,550 $114,980 $124,200 $133,420 140%AMI $93,980 $107,360 $120,810 $134,140 $144,900 $155,660 Median (100% of Median) $67,100 $76,700 $86,300 $95,800 $103,500 $111,200 The analysis is conducted using a model that KMA developed and has applied to similar evaluations in many other jurisdictions. The model inputs are all local data to the extent possible, and are fully documented in the following description. Analysis Steps The tables at the end of this section present a summary of the nexus analysis steps for the prototype units. Following is a description of each step of the analysis. Keyser Marston Associates, Inc. Page 32 \\Sf-fs2\wp\14\14100\002\001-002.docx Step 1 — Estimate of Total New Employees Table C-1 commences with the total number of employees associated with the new market rate units. The employees were estimated based on household expenditures of new residents using the IMPLAN model (see Section B). Step 2—Changing Industries Adjustment and Net New Jobs The local economy, like that of the U.S. as a whole, is constantly evolving. In Honolulu, employment in certain sectors of the economy has been declining including manufacturing, information (includes newspapers), and finance and insurance (includes banks which have become increasingly reliant on ATMs). Jobs lost in these declining sectors were replaced by job growth in other industry sectors. Step 2 makes an adjustment to take ongoing changes in the economy into account recognizing that jobs added are not 100% net new in all cases. A 17% adjustment is utilized based on the long term shifts in employment that have occurred in some sectors of the local economy and the likelihood of continuing changes in the future. Long term declines in employment experienced in certain sectors of the economy mean that some of the new jobs are being filled by workers that have been displaced from another industry and who are presumed to already have housing locally. Existing workers downsized from declining industries are assumed to be available to fill a portion of the new retail, restaurant, health care, and other jobs associated with services to residents. This is a conservative assumption given some displaced workers may exit the workforce entirely by retiring rather than seek a new job in one of the industries serving new residents. The 17% downward adjustment used for purposes of the analysis was derived from data available through the State of Hawaii Research and Statistics Office on employment by industry in Honolulu over the eighteen year period from 2013 to 1995. Employment in 1995 was compared to the most recent data available for 2013 because the unemployment rate was the same. Selection of a year with a similar unemployment rate is to separate short-term or cyclical declines from long-term changes warranting an adjustment in the nexus analysis. Over the period from 1995 through 2013, approximately 9,400 jobs were lost in declining industry sectors. Over the same period, growing and stable industries added a total of 56,800 jobs. The figures are used to establish a ratio between jobs lost in declining industries to jobs gained in growing and stable industries at 17%8. The 17% factor is applied as an adjustment in the analysis, effectively assuming approximately one in every six new jobs is filled by a worker down-sized from a declining industry and who already has housing. 8 The 17%ratio is calculated as 9,400 jobs lost in declining sectors excluding defense divided by 56,800 jobs gained in growing and stable sectors= 16.5%(rounded to 17%). Keyser Marston Associates, Inc. Page 33 \\Sf-fs2\wp\14\14100\002\001-002.docx Step 3—Adjustment from Employees to Employee Households This step (Table C-1) converts the number of employees to the number of employee households, recognizing that there is, on average, more than one worker per household, and thus the number of housing units in demand for new workers is reduced. The workers-per- worker-household ratio eliminates from the equation all non-working households, such as retired persons, students, and those on public assistance. The average for Honolulu of 1.92 workers per worker household (from the U. S. Census Bureau 2011-2013 American Community Survey) is used for this step in the analysis. The number of jobs is divided by 1.92 to determine the number of worker households. This ratio is distinguished from the overall number of workers per household in that the denominator includes only households with at least one worker. If the average number of workers in all households were used, it would have produced a greater demand for housing units. The 1.92 ratio covers all workers, full and part time. Step 4— Occupational Distribution of Employees The occupational breakdown of employees is the first step to arrive at income level. The output from the IMPLAN model provides the number of employees by industry sector, shown in Table B-1. The IMPLAN output is paired with data from the Department of Labor, Bureau of Labor Statistics May 2013 Occupational Employment Survey (OES) to estimate the occupational composition of employees for each industry sector. Step 4a— Translation from IMPLAN Industry Codes to NAICS Industry Codes The output of the IMPLAN model is jobs by industry sector using IMPLAN's own industry classification system which consists of 440 industry sectors. The OES occupation data uses the North American Industry Classification System (NAICS). Estimates of jobs by IMPLAN sector must be translated into estimates by NAICS code for consistency with the OES data. The NAICS system is organized into industry codes ranging from two- to six-digits. Two-digit codes are the broadest industry categories and six-digit codes are the most specific. Within a two-digit NAICS code, there may be several three-digit codes and within each three digit code, several four-digit codes, etc. A chart published by IMPLAN relates each IMPLAN industry sector with one or more NAICS codes, with matching NAICS codes ranging from the two-digit level to the five-digit level. For purposes of the nexus analysis, all employment estimates must be aggregated to the four digit NAICS code level to align with OES data which is organized by four- digit NAICS code. For some industry sectors, an allocation is necessary between more than one four-digit NAICS code. Where required, allocations are made proportionate to total employment at the national level from the OES. The table below illustrates analysis Step 4a in which employment estimates by IMPLAN Code are translated to NAICS codes and then aggregated at the four digit NAICS code level. The Keyser Marston Associates, Inc. Page 34 \\Sf-fs2\wp\14\14100\002\001-002.docx examples used are Child Day Care Centers and Food and Drinking Places. The process is applied to all the industry sectors. Illustration of Model Step 4a. A. IMPLAN Output by B. Link to Corresponding C.Aggregate at 4-Digit NAICS Code Level IMPLAN Industry Sector NAICS Code % Total Jobs IMPLAN Sector Jobs NAICS Code Jobs Employment 4-Digit NAICS 1.0 399 -Child day 1.0 6244 Child day 1.0 100% 6244 Child day care care services care services services 7.9 413 -Food and 7.9 722 Food and 7.1 91% 7225 Restaurants Drinking Places Drinking Places and Other Eating Places 0.5 6% 7223 Special Food Services 0.3 3% 7224 Drinking Places (Alcoholic Beverages) Step 4b—Apply OES Data to Estimate Occupational Distribution Employment estimates by four-digit NAICS code from step 4a are paired with data on occupational composition within each industry from the OES to generate an estimate of employment by detailed occupational category. As shown on Table C-1, new jobs will be distributed across a variety of occupational categories. The three largest occupational categories are office and administrative support (17%), sales (15%), and food preparation and serving (13%). Step 4 of Table C-1 indicates the percentage and number of employee households by occupation associated with 100 market rate units. Step 5—Estimates of Employee Households Meeting the Lower Income Definitions In this step, occupations are translated to employee incomes based on recent Honolulu wage and salary information obtained from the Bureau of Labor Statistics Occupational Employment Survey for Honolulu. The wage and salary information summarized in Appendix 2 Tables 2 and 4 provided the income inputs to the model. For each occupational category shown in Table C-1, the OES data provides a distribution of specific occupations within the category. For example, within the Food Preparation and Serving Category, there are Supervisors, Cooks, Bartenders, Waiters and Waitresses, Dishwashers, etc. In total there are over 100 detailed occupation categories included in the analysis as shown in Appendix 2, Tables 2 and 4. The analysis uses a separate OES data set on compensations to reflect the distribution of wages specific to workers in Honolulu as of 2013 for each of the detailed occupation categories. Keyser Marston Associates, Inc. Page 35 \\Sf-fs2\wp\14\14100\002\001-002.d ocx For each detailed occupational category, the distribution of wages is used to calculate the percent of worker households that would fall into each income category. The calculation is performed for each possible combination of household size and number of workers in the household. For households with more than one worker, individual employee income data was used to calculate the household income by assuming multiple earner households are, on average, formed of individuals with similar incomes. The table below illustrates Step 5 as applied to food preparation and serving workers. Annual compensation for food preparation and serving workers in Honolulu as of 2013 is distributed9 around a mean of$19,700. For households with one worker, 64% of one person households are estimated to qualify as Extremely Low and 100% of households with two or more people are estimated to qualify. For households with two workers, no households of five persons or less are estimated to qualify as Extremely Low and only 17% of six person households are estimated to qualify. No households that have three workers are expected to qualify as Extremely Low. Step 5 Illustration for Food Preparation and Serving Worker Households Percent Qualifying as Extremely Low for Each Possible Household Size/No. of Workers Combination Percent of Worker Households That Would Qualify as Extremely Low For Each Possible Combination of Household Size and No. of Workers Applying 2014 Income Limits for Honolulu HH Size 1 Person 2 Person 3 Person 4 Person 5 Person 6 Person Limit $20,150 $23,000 $25,900 $28,750 $31,050 $33,350 No. Workers in Household 1 64% 100% 100% 100% 100% 100% 2 N/A 0% 0% 0% 0% 17% 3 or more10 N/A N/A 0% 0% 0% 0% The step illustrated above is repeated for each detailed worker occupation category and each of the five affordability tiers. At the end of Step 5, the nexus model has established a matrix indicating the percentages of households that would qualify in the affordable income tiers for every detailed occupational category and every potential combination of household size and number of workers in the household. Step 6—Distribution of Household Size and Number of Workers In this step, we account for the distribution in household sizes and number of workers for Honolulu households using local data obtained from the U.S. Census. Census data is used to develop a set of percentage factors representing the distribution of household sizes and number of workers within working households in Honolulu. The following table presents the percentage factors used in the analysis. s In addition to the mean compensation, BLS reported 25th, 50th, and 75th percentile compensations are utilized. 10 Census data aggregates households with three or more workers;therefore, a corresponding aggregation is necessary for purposes of the analysis. Keyser Marston Associates, Inc. Page 36 \\Sf-fs2\wp\14\14100\002\001-002.docx Step 6: Percentage of Honolulu Households by Size and Number of Workers Household Size (Persons) 1 2 3 4 5 6+ No. Workers in Household 1 16.78% 13.33% 7.69% 4.52% 2.44% 2.97% 2 N/A 14.86% 8.92% 6.06% 3.27% 3.98% 3 or more N/A N/A 3.10% 5.50% 2.96% 3.61% Note:percentages sum to 100% The percentage factors are specific to Honolulu and are derived from the 2011 —2013 American Community Survey. Application of these percentage factors accounts for the following: • Households have a range in size and a range in the number of workers. • Large households generally have more workers than smaller households. The result of Step 6 is a distribution of Honolulu working households by number of workers and household size. Step 7— Estimate of Number of Households that Meet Size and Income Criteria Step 7 is the final step to calculate the number of worker households meeting the size and income criteria for the five affordability tiers. The calculation combines the matrix of results from Step 5 on percentage of worker households that would meet the income criteria at each potential household size/ no. of workers combination, with Step 6, the percentage of worker households having a given household size/ number of workers combination. The result is the percentage of households that fall into each affordability tier. The percentages are then multiplied by the number of households from Step 3 to arrive at number of households in each affordability tier. Tables C-2.1 through C-2.5 shows the result after completing Steps 5, 6, and 7 for each of the five individual affordability tiers. Summary Findings Table C-3 indicates the results of the analysis for each of the residential prototypes. The table presents the number of households generated in each affordability category and the total number over 140% of Area Median Income. The findings in Table C-3 are presented below. The table shows the total demand for affordable housing units associated with 100 market rate units. Keyser Marston Associates, Inc. Page 37 \\Sf-fs2\wp\14\14100\002\001-002.docx New Worker Households by Income Level per 100 Market Rate Units Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo(PUC) Apartment Extr. Low(0% -30%AMI) 4.0 3.6 3.2 4.3 3.2 Very Low(30% -50%AMI) 7.7 7.0 6.4 8.3 6.2 Low(50% -80%AMI) 8.0 7.2 6.6 8.6 6.5 Moderate (80% - 120%AMI) 5.0 4.5 4.2 5.3 4.0 Subtotal through 120% AMI 24.7 22.2 20.4 26.4 20.0 140%Tier(120% -140%AMI) 1.2 1.1 1.0 1.3 1.0 Subtotal through 140% AMI 25.8 23.3 21.4 27.7 20.9 Greater than 140%AMI 3.2 2.9 2.7 3.4 2.6 Total, New Households 29.0 26.2 24.1 31.1 23.5 Housing demand for new worker households earning less than 120% of AMI ranges from approximately 26 units per 100 market rate units for High-Rise Condo units, to 20 units per 100 market rate units for apartments. Including housing needs at the 120% to 140% AMI tier increases these figures by approximately one unit for each 100 market rate units. Housing demand is distributed across the lower income tiers with the greatest number of households in the Low-Income tier from 50% to 80% AMI. The finding that the jobs associated with household spending tend to be low-paying jobs where the workers will require housing affordable at the lower income levels is not surprising. As noted above, household spending results in employment that is concentrated in lower paid occupations including food preparation, administrative, and retail sales. Inclusionary Percentages Supported The analysis findings identify how many lower income households are generated for every 100 market rate units. These findings are adjusted to an inclusionary percentage which represents the percentage of units provided on-site within a project that would fully mitigate the affordable housing impacts as documented in this nexus analysis. The percentages are calculated including both market rate and affordable units (for example, 25 affordable units per 100 market rate units translates to a project of 125 units; 25 affordable units out of 125 units equals 20%). The table below presents the results of the analysis, drawn from Table C-4. Each tier is cumulative, or inclusive of the tiers above. The purpose of showing the figures on a cumulative basis is so they can be readily compared to potential inclusionary requirements that may be considered. As an example, for new single family projects, the analysis indicates that an inclusionary requirement of 19.8% with affordable units available to households earning up to 120% of AMI would be sufficient to mitigate the affordable housing need through 120% of AMI. Keyser Marston Associates, Inc. Page 38 \\Sf-fs2\wp\14\14100\002\001-002.docx The percentages represent the inclusionary requirement that would be sufficient to fully offset the increased affordable housing need from the services and service workers that support the new residential development. Cumulative Inclusionary Percentage to Mitigate Increased Affordable Housing Need Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo(PUC) Apartment Extr. Low(up to 30%AMI) 3.8% 3.5% 3.1% 4.1% 3.1% Very Low(up to 50%AMI) 10.5% 9.5% 8.8% 11.1% 8.7% Low(up to 80%AMI) 16.5% 15.1% 14.0% 17.4% 13.8% Moderate (up to 120%AMI) 19.8% 18.2% 16.9% 20.9% 16.7% 140%Tier(up to 140%AMI) 20.5% 18.9% 17.6% 21.7% 17.3% Keyser Marston Associates, Inc. Page 39 \\Sf-fs2\wp\14\14100\002\001-002.docx TABLE C-1 NET NEW HOUSEHOLDS AND OCCUPATION DISTRIBUTION EMPLOYEE HOUSEHOLDS GENERATED RESIDENTIAL NEXUS ANALYSIS HONOLULU,HI Prototype 1: Prototype 2: Prototype 3: Prototype 4: Prototype 5: Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo(PUC) Apartment Step 1-Employees' 67.1 60.5 55.7 71.8 54.3 Step 2-Adjustment for Changing Industries(17%) 55.7 50.2 46.2 59.6 45.1 Step 3-Adjustment for Number of Households(1.92)2 29.0 26.2 24.1 31.1 23.5 Step 4-Occupation Distribution 3 Management Occupations 4.3% 4.3% 4.4% 4.3% 4.3% Business and Financial Operations 4.7% 4.7% 4.6°% 4.7% 4.7% Computer and Mathematical 1.4% 1.4% 1.4% 1.4% 1.4% Architecture and Engineering 0.3% 0.3% 0.4% 0.3% 0.3% Life,Physical,and Social Science 0.4% 0.4% 0.4% 0.4% 0.4% Community and Social Services 1.8% 1.8% 1.8% 1.8% 1.8% Legal 0.9% 0.9% 0.8% 0.9% 0.9% Education,Training,and Library 3.4% 3.4% 2.9% 3.4% 3.4% Arts,Design,Entertainment,Sports,and Media 1.5% 1.5% 1.5% 1.5% 1.5% Healthcare Practitioners and Technical 7.7% 7.7% 8.0% 7.7% 7.7% Healthcare Support 4.1% 4.1% 4.3% 4.1% 4.1% Protective Service 1.5% 1.5% 1.5% 1.5% 1.5% Food Preparation and Serving Related 13.2% 13.2% 13.3°/0 13.2% 13.2% Building and Grounds Cleaning and Maint. 5.9% 5.9% 5.9% 5.9% 5.9% Personal Care and Service 5.5% 5.5% 5.3% 5.5% 5.5% Sales and Related 14.7% 14.7% 14.3% 14.7% 14.7% Office and Administrative Support 16.8% 16.8% 16.9% 16.8% 16.8% Farming,Fishing,and Forestry 0.1% 0.1% 0.1% 0.1% 0.1% Construction and Extraction 0.8% 0.8% 0.8% 0.8% 0.8% Installation,Maintenance,and Repair 3.6% 3.6% 3.8% 3.6% 3.6% Production 1.9% 1.9% 1.9% 1.9% 1.9% Transportation and Material Moving 5.6% 5.6% 5.6% 5.6% 5.6% Totals 100.0% 100.0% 100.0% 100.0% 100.0°/0 Management Occupations 1.3 1.1 1.1 1.3 1.0 Business and Financial Operations 1.4 1.2 1.1 1.4 1.1 Computer and Mathematical 0.4 0.4 0.3 0.4 0.3 Architecture and Engineering 0.1 0.1 0.1 0.1 0.1 Life,Physical,and Social Science 0.1 0.1 0.1 0.1 0.1 Community and Social Services 0.5 0.5 0.4 0.6 0.4 Legal 0.3 0.2 0.2 0.3 0.2 Education,Training,and Library 1.0 0.9 0.7 1.1 0.8 Arts,Design,Entertainment,Sports,and Media 0.4 0.4 0.4 0.5 0.4 Healthcare Practitioners and Technical 2.2 2.0 1.9 2.4 1.8 Healthcare Support 1.2 1.1 1.0 1.3 1.0 Protective Service 0.4 0.4 0.4 0.5 0.4 Food Preparation and Serving Related 3.8 3.4 3.2 4.1 3.1 Building and Grounds Cleaning and Maint. 1.7 1.5 1.4 1.8 1.4 Personal Care and Service 1.6 1.4 1.3 1.7 1.3 Sales and Related 4.3 3.8 3.4 4.6 3.4 Office and Administrative Support 4.9 4.4 4.1 5.2 4.0 Farming,Fishing,and Forestry 0.0 0.0 0.0 0.0 0.0 Construction and Extraction 0.2 0.2 0.2 0.2 0.2 Installation,Maintenance,and Repair 1.0 0.9 0.9 1.1 0.9 Production 0.5 0.5 0.5 0.6 0.4 Transportation and Material Moving 1.6 1.5 1.3 1.7 1.3 Totals 29.0 26.2 24.1 31.1 23.5 Notes: ' Estimated employment generated by expenditures of households within 100 prototypical market rate units.Employment estimates based on economic model,IMPLAN. `Adjustment from number of workers to households using average of 1.92 workers per worker household derived from the U.S.Census American Community Survey 2011 to 2013. 3 See Appendix 2,Tables 1 through 4 for additional information on Major Occupation Categories. Page 40 Prepared by Keyser Marston Associates,Inc. \\Sf-fs2\wp\14\14100\002\Honolulu Nexus model 8-31-15;8/31/2015;dd TABLE C-2.1 EXTREMELY LOW-INCOME EMPLOYEE HOUSEHOLDS' GENERATED RESIDENTIAL NEXUS ANALYSIS HONOLULU, HI Per 100 Market Rate Units Prototype 1: Prototype 2: Prototype 3: Prototype 4: Prototype 5: Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo(PUC) Apartment Step 5&6-Extremely Low Income Households(under 30%AMI)within Major Occupation Categories 2 Management 0.00 0.00 0.00 0.00 0.00 Business and Financial Operations 0.00 0.00 0.00 0.00 0.00 Computer and Mathematical - - - - - Architecture and Engineering - - - - - Life,Physical and Social Science - - - - - Community and Social Services - - - - - Legal - - - - - Education Training and Library 0.06 0.05 0.04 0.06 0.05 Arts,Design,Entertainment,Sports,&Media - - - - - Healthcare Practitioners and Technical 0.00 0.00 0.00 0.00 0.00 Healthcare Support 0.14 0.13 0.12 0.15 0.12 Protective Service - - - - - Food Preparation and Serving Related 1.12 1.01 0.94 1.20 0.91 Building Grounds and Maintenance 0.29 0.26 0.25 0.31 0.23 Personal Care and Service 0.36 0.32 0.29 0.38 0.29 Sales and Related 0.85 0.77 0.67 0.91 0.69 Office and Admin 0.37 0.33 0.29 0.39 0.30 Farm, Fishing,and Forestry - - - - - Construction and Extraction - - - - - Installation Maintenance and Repair 0.01 0.01 0.01 0.01 0.01 Production - - - - - Transportation and Material Moving 0.22 0.20 0.18 0.24 0.18 ELI Households-Major Occupations 3.4 3.1 2.8 3.7 2.8 ELI Households'-all other occupations 0.6 0.5 0.5 0.6 0.5 Total ELI Households' 4.0 3.6 3.2 4.3 3.2 1 Includes households earning from zero through 30%of Honolulu Area Median Income. 2 See Appendix 2 Tables 1 through 4 for additional information on Major Occupation Categories.Note that the model places individual employees into households.Many households have multiple income sources and therefore household income is higher than the wages shown in Appendix 2 tables 2 and 4.The distribution of the number of workers per worker household and the distribution of household size are based on American Community Survey data. Prepared by:Keyser Marston Associates,Inc. Page 41 \\Sf-fs2\wp\14\14100\002\Honolulu Nexus model 8-31-15;8/31/2015;dd TABLE C-2.2 VERY LOW-INCOME EMPLOYEE HOUSEHOLDS' GENERATED RESIDENTIAL NEXUS ANALYSIS HONOLULU,HI Per 100 Market Rate Units Prototype 1: Prototype 2: Prototype 3: Prototype 4: Prototype 5: Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo(PUC) Apartment Step 5&6-Very Low Income Households(30%-50%AMI)within Major Occupation Categories 2 Management 0.03 0.03 0.03 0.04 0.03 Business and Financial Operations 0.06 0.06 0.05 0.07 0.05 Computer and Mathematical - - - - - Architecture and Engineering - - - - - Life,Physical and Social Science - - - - - Community and Social Services - - - - - Legal - - - - - Education Training and Library 0.23 0.20 0.16 0.24 0.18 Arts,Design,Entertainment,Sports,&Media - - - - - Healthcare Practitioners and Technical 0.05 0.05 0.04 0.06 0.04 Healthcare Support 0.40 0.37 0.35 0.43 0.33 Protective Service - - - - - Food Preparation and Serving Related 1.34 1.21 1.12 1.43 1.08 Building Grounds and Maintenance 0.57 0.52 0.49 0.61 0.46 Personal Care and Service 0.54 0.49 0.43 0.58 0.44 Sales and Related 1.37 1.23 1.10 1.46 1.11 Office and Admin 1.37 1.23 1.12 1.47 1.11 Farm,Fishing,and Forestry - - - - - Construction and Extraction - - - - - Installation Maintenance and Repair 0.15 0.13 0.13 0.16 0.12 Production - - - - - Transportation and Material Moving 0.52 0.47 0.43 0.55 0.42 Very Low Households-Major Occupations 6.6 6.0 5.5 7.1 5.4 Very Low Households'-all other occupations 1.1 1.0 0.9 1.2 0.9 Total Very Low Households' 7.7 7,0 6.4 8.3 6.2 'Includes households earning from 30%through 50%of Honolulu Area Median Income. 2 See Appendix 2 Tables 1 through 4 for additional information on Major Occupation Categories.Note that the model places individual employees into households.Many households have multiple income sources and therefore household income is higher than the wages shown in Appendix 2 tables 2 and 4.The distribution of the number of workers per worker household and the distribution of household size are based on American Community Survey data. Page 42 Prepared by:Keyser Marston Associates, Inc. \\Sf-fs2\wp\14\14100\002\Honolulu Nexus model 8-31-15;8/31/2015;dd TABLE C-2.3 LOW-INCOME EMPLOYEE HOUSEHOLDS'GENERATED RESIDENTIAL NEXUS ANALYSIS HONOLULU, HI Per 100 Market Rate Units Prototype 1: Prototype 2: Prototype 3: Prototype 4: Prototype 5: Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo(PUC) Apartment Step 5&6-Low Income Households(50%-80%AMI)within Major Occupation Categories 2 Management 0.17 0.15 0.14 0.18 0.14 Business and Financial Operations 0.29 0.26 0.24 0.31 0.23 Computer and Mathematical - - - - - Architecture and Engineering - - - - - Life,Physical and Social Science - - - - - Community and Social Services - - - - - Legal - - - - - Education Training and Library 0.31 0.28 0.22 0.33 0.25 Arts,Design,Entertainment,Sports,&Media - - - - - Healthcare Practitioners and Technical 0.20 0.18 0.17 0.21 0.16 Healthcare Support 0.38 0.34 0.33 0.41 0.31 Protective Service - - - - - Food Preparation and Serving Related 0.93 0.84 0.78 1.00 0.75 Building Grounds and Maintenance 0.48 0.43 0.40 0.52 0.39 Personal Care and Service 0.46 0.41 0.37 0.49 0.37 Sales and Related 1.22 1.10 0.99 1.30 0.99 Office and Admin 1.62 1.46 1.35 1.73 1.31 Farm,Fishing,and Forestry - - - - - Construction and Extraction - - - - - Installation Maintenance and Repair 0.31 0.28 0.27 0.33 0.25 Production - - - - - Transportation and Material Moving 0.52 0.47 0.44 0.56 0.42 Low Inc.Households-Major Occupations 6.9 6.2 5.7 7.4 5.6 Low Inc. Households'-all other occupations 1.1 1.0 0.9 1.2 0.9 Total Low Income Households' 8.0 7.2 6.6 8.6 6.5 1 Includes households earning from 50%through 80%of Honolulu Area Median Income. 2 See Appendix 2 Tables 1 through 4 for additional information on Major Occupation Categories.Note that the model places individual employees into households.Many households have multiple income sources and therefore household income is higher than the wages shown in Appendix 2 tables 2 and 4.The distribution of the number of workers per worker household and the distribution of household size are based on American Community Survey data. Prepared by:Keyser Marston Associates, Inc. Page 43 \\Sf-fs2\wp\14\14100\002\Honolulu Nexus model 8-31-15;8/31/2015;dd TABLE C-2.4 MODERATE-INCOME EMPLOYEE HOUSEHOLDS GENERATED RESIDENTIAL NEXUS ANALYSIS HONOLULU,HI Per 100 Market Rate Units Prototype 1: Prototype 2: Prototype 3: Prototype 4: Prototype 5: Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo(PUC) Apartment Step 5&6-Moderate Income Households(80%-120%AMI)within Major Occupation Categories 2 Management 0.26 0.24 0.22 0.28 0.21 Business and Financial Operations 0.40 0.36 0.33 0.43 0.32 Computer and Mathematical - - - - - Architecture and Engineering - - - - - Life, Physical and Social Science - - - - - Community and Social Services - - - - - Legal - - - - - Education Training and Library 0.23 0.20 0.16 0.24 0.18 Arts,Design,Entertainment,Sports,&Media - - - - - Healthcare Practitioners and Technical 0.52 0.47 0.45 0.55 0.42 Healthcare Support 0.19 0.17 0.17 0.20 0.15 Protective Service - - - - - Food Preparation and Serving Related 0.26 0.24 0.22 0.28 0.21 Building Grounds and Maintenance 0.26 0.23 0.21 0.28 0.21 Personal Care and Service 0.15 0.13 0.12 0.16 0.12 Sales and Related 0.45 0.40 0.37 0.48 0.36 Office and Admin 1.02 0.92 0.88 1.10 0.83 Farm,Fishing,and Forestry - - - - - Construction and Extraction - - - - - Installation Maintenance and Repair 0.28 0.26 0.25 0.30 0.23 Production - - - - - Transportation and Material Moving 0.25 0.23 0.21 0.27 0.20 Moderate Inc.Households-Major Occupations 4.3 3.9 3.6 4.6 3.5 Moderate Inc.Households'-all other occupations 0.7 0.6 0.6 0.7 0.6 Total Moderate Income Households.' 5.0 4.5 4.2 5.3 4.0 1 Includes households earning from 80%through 120%of Honolulu Area Median Income. 2 See Appendix 2 Tables 1 through 4 for additional information on Major Occupation Categories.Note that the model places individual employees into households.Many households have multiple income sources and therefore household income is higher than the wages shown in Appendix 2 tables 2 and 4. The distribution of the number of workers per worker household and the distribution of household size are based on American Community Survey data. Prepared by:Keyser Marston Associates, Inc. Page 44 \\Sf-fs2\wp\14\14100\002\Honolulu Nexus model 8-31-15;8/31/2015;dd TABLE C-2.5 EMPLOYEE HOUSEHOLDS FROM 120%TO 140%AMI'GENERATED RESIDENTIAL NEXUS ANALYSIS HONOLULU, HI Per 100 Market Rate Units Prototype 1: Prototype 2: Prototype 3: Prototype 4: Prototype 5: Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo(PUC) Apartment Step 5&6-Households from 120%to 140%AMI within Major Occupation Categories 2 Management 0.13 0.12 0.11 0.14 0.10 Business and Financial Operations 0.14 0.12 0.11 0.15 0.11 Computer and Mathematical - - - - - Architecture and Engineering - - - - - Life,Physical and Social Science - - - - - Community and Social Services - - - - - Legal - - - - - Education Training and Library 0.06 0.05 0.04 0.06 0.05 Arts, Design, Entertainment,Sports,&Media - - - - - Healthcare Practitioners and Technical 0.23 0.21 0.20 0.25 0.19 Healthcare Support 0.01 0.01 0.01 0.02 0.01 Protective Service - - - - - Food Preparation and Serving Related 0.02 0.02 0.02 0.03 0.02 Building Grounds and Maintenance 0.02 0.02 0.01 0.02 0.02 Personal Care and Service 0.01 0.01 0.01 0.01 0.01 Sales and Related 0.07 0.07 0.06 0.08 0.06 Office and Admin 0.18 0.17 0.17 0.20 0.15 Farm,Fishing,and Forestry - - - - - Construction and Extraction - - - - - Installation Maintenance and Repair 0.09 0.08 0.08 0.10 0.07 Production - - - - - Transportation and Material Moving 0.04 0.03 0.03 0.04 0.03 Households: 120%-140%AMI-Major Occupations 1.0 0.9 0.9 1.1 0.8 Households 120%-140%AMI'-other occupations 0.2 0.1 0.1 0.2 0.1 Total Households from 120%-140%AMI 1 1.2 1.1 1.0 1.3 1.0 1 Includes households earning from 120%through 140%of Honolulu Area Median Income(AMI). 2 See Appendix 2 Tables 1 through 4 for additional information on Major Occupation Categories.Note that the model places individual employees into households.Many households have multiple income sources and therefore household income is higher than the wages shown in Appendix 2 tables 2 and 4.The distribution of the number of workers per worker household and the distribution of household size are based on American Community Survey data. Prepared by:Keyser Marston Associates, Inc. Page 45 \\Sf-fs2\wp\14\14100\002\Honolulu Nexus model 8-31-15;8/31/2015;dd TABLE C-3 IMPACT ANALYSIS SUMMARY EMPLOYEE HOUSEHOLDS GENERATED RESIDENTIAL NEXUS ANALYSIS HONOLULU, HI RESIDENTIAL UNIT DEMAND IMPACTS PER 100 MARKET RATE UNITS Prototype 1: Prototype 2: Prototype 3: Prototype 4: Prototype 5: Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo(PUC) Apartment Number of New Households' Under 30%Area Median Income 4.0 3.6 3.2 4.3 3.2 30%to 50%Area Median Income 7.7 7.0 6.4 8.3 6.2 50%to 80%Area Median Income 8.0 7.2 6.6 8.6 6.5 80%to 120%Area Median Income 5.0 4.5 4.2 5.3 4.0 Subtotal through 120%of Median 24.7 22.2 20.4 26.4 20.0 120%to 140%Area Median Income 1.2 1.1 1.0 1.3 1.0 Subtotal through 140%of Median 25.8 23.3 21.4 27.7 20.9 Over 140%Area Median Income 3.2 2.9 2.7 3.4 2.6 Total Employee Households 29.0 26.2 24.1 31.1 23.5 Percent of New Households 1 Under 30°A)Area Median Income 14% 14% 13% 14% 14% 30%to 50%Area Median Income 27% 27% 26% 27% 27% 50%to 80%Area Median Income 28% 28% 28% 28% 28% 80°A°to 120%Area Median Income 17% 17% 17% 17% 17% Subtotal through 120%of Median 85% 85% 85% 85% 85% 120%to 140%Area Median Income 4% 4% 4% 4% 4% Subtotal through 140%of Median 89% 89% 89% 89% 89% Over 140%Area Median Income 11°A° 11% 11% 11% 11% Total Employee Households 100% 100% 100% 100% 100% Notes 1 Households of retail,education,healthcare and other workers that serve residents of new market rate units. Prepared by: Keyser Marston Associates, Inc. Page 46 \\Sf-fs2\wp\14\14100\002\Honolulu Nexus model 8-31-15; 8/31/2015; dd TABLE C-4 INCLUSIONARY REQUIREMENT SUFFICIENT TO MITIGATE INCREASED AFFORDABLE HOUSING NEEDS RESIDENTIAL NEXUS ANALYSIS HONOLULU, HI Prototype 2: Prototype 3: Prototype 4: Prototype 5: Prototype 1: Low-Rise Mid-Rise High-Rise Rental Single Family Townhomes Condo Condo(PUC) Apartment Supported Inclusionary Requirement Affordable Housing Need Per 100 Market Rate Units-Cumulative Through 30%OF MEDIAN INCOME 4.0 Units 3.6 Units 3.2 Units 4.3 Units 3.2 Units 50%OF MEDIAN INCOME 11.7 Units 10.5 Units 9.6 Units 12.5 Units 9.5 Units 80%OF MEDIAN INCOME 19.7 Units 17.8 Units 16.2 Units 21.1 Units 16.0 Units 120%OF MEDIAN INCOME 24.7 Units 22.2 Units 20.4 Units 26.4 Units 20.0 Units 140%OF MEDIAN INCOME 25.8 Units 23.3 Units 21.4 Units 27.7 Units 20.9 Units Inclusionary Percentage Sufficient to Mitigate Impacts-Cumulative Through 1 30%OF MEDIAN INCOME 3.8% 3.5% 3.1% 4.1% 3.1% 50%OF MEDIAN INCOME 10.5% 9.5% 8.8% 11.1% 8.7% 80%OF MEDIAN INCOME 16.5% 15.1% 14.0% 17.4% 13.8% 120%OF MEDIAN INCOME 19.8% 18.2% 16.9% 20.9% 16.7% 140% OF MEDIAN INCOME 20.5% 18.9% 17.6% 21.7% 17.3% Notes: 'Calculated by dividing the number of affordable units needed by the total number of units(affordable units needed+100 market rate units). Prepared by: Keyser Marston Associates, Inc. Page 47 \\Sf-fs2\wp\14\14100\002\Honolulu Nexus model 8-31-15; 8/31/2015; dd D. MITIGATION COSTS This section takes the conclusions of the previous section on the number of households in the lower income categories associated with the market rate units and identifies the total cost of assistance required to make housing affordable. This section puts a cost on the units for each income level to produce the "total nexus cost." This is done for each of the prototype units. The estimate of total nexus cost is needed in order to quantify the amount of the in-lieu fee that would be sufficient to fully mitigate the increased affordable housing impacts of the new market rate units. A key component of the nexus cost analysis is the size of the gap between what households can afford and the cost of producing new housing in Honolulu, known as the 'affordability gap.' Affordability gaps are calculated for each of the five categories of area median income: Extremely Low (under 30% of median), Very Low (30% to 50%), Low (50% to 80%), Moderate (80% to 120%), and the 140% AMI Tier (120% to 140%). The following summarizes the analysis of mitigation cost which is based on the affordability gap or net cost to deliver units that are affordable to worker households in the lower income tiers. Detailed affordability gap calculations are presented in Tables D-1 through D-4 at the end of this section. Affordable Unit Prototypes Subsidized with In-lieu Fees For estimating the affordability gap, there is a need to match a household of each income level with a unit type and size according to governmental regulations and City practices and policies. The analysis assumes that the City will help subsidize affordable ownership units for households in the Moderate and 140% AMI Tier. The prototype affordable unit should reflect a modest unit consistent with what the City is likely to assist and appropriate for housing the average worker household in these income tiers, which in the case of Honolulu is assumed to be a four person household in a three-bedroom townhome unit. The analysis assumes households earning less than 80% of Area Median Income will be housed in rental units. The analysis uses a two bedroom affordable rental prototype. For the purposes of estimating the affordability gaps, we do not assume additional sources of affordable housing financing such as the federal income tax credit program. While some recent affordable housing developments in Honolulu have utilized these additional funding sources, it is not assured that these sources will be available in the future. Accessing these sources is also highly competitive due to the limited supply. Of importance, Honolulu has a sizable existing deficiency of affordable housing units and the limited amount of outside subsidy sources has not been sufficient to fully address the existing needs of the community let alone the future needs created by new market rate residential units. The value of tax credits to the project can fluctuate widely. Determining the affordability gap assuming no outside sources is a sound and legitimate approach. Keyser Marston Associates, Inc. Page 48 \\Sf-fs2\wp\14\14100\002\001-002.docx Development Costs KMA prepared an estimate of total development cost for a typical two bedroom affordable rental unit (inclusive of land, all fees and permits, financing and other indirect costs) based on a review of development costs for affordable multi-family projects in Hawaii. On this basis, KMA concluded that on average, the new affordable rental units would have a total development cost per unit of$350,000. For ownership units, total development costs for a typical three bedroom, 1,200 square foot townhome were estimated by KMA to be $491,000 per unit ($409 per square foot). Development Costs Income Group Tenure/Type Type Development Cost 0%through 80%AMI Rental 2BR Apartment $350,000 80%through 140%AMI Ownership 3BR Townhome $491,000 Unit Values For affordable ownership units, unit values are the affordable purchase prices. Affordable purchase prices for ownership units are calculated based on the purchase price affordable to a household earning 120% of AMI for Moderate-Income households and 140% of AMI for the 140% AMI Tier. For a three bedroom unit, KMA calculated the affordable sales price as $421,000 for a Moderate-Income Unit and $513,000 for the 140% AMI Tier. Details of the calculation are presented in Tables D-3 and D-4. For the Extremely Low, Very Low, and Low-Income units, the unit values were estimated based on the amount of permanent debt that can be supported by the project's net operating income (NOI) plus a small amount for deferred developer fees, which is a common source of funding for affordable apartment projects. Maximum Affordable Sales Prices and Rent Levels Household Maximum Monthly Unit Values/ Income Group Unit Tenure Size Housing Costs' Sales Price Under 30%AMI Rental 3 persons $719/Month ($17,300) 30% to 50%AMI Rental 3 persons $1,199/Month $61,700 50% to 80%AMI Rental 3 persons $1,918/Month $180,700 80%to 120%AMI Ownership 4 persons $3,193/Month $421,150 120%to 140%AMI Ownership 4 persons $3,726/Month $512,650 " For rental units, maximum housing costs are the affordable rents before utility allowance. For ownership unit, maximum monthly housing costs includes all housing expenses such as mortgage, insurance, property taxes, HOA dues, and utilities. Keyser Marston Associates, Inc. Page 49 \\Sf-fs2\wp\14\14100\002\001-002.docx Affordability Gap The affordability gap is the difference between the cost of developing a residential unit and the unit values at the affordable rents or sales prices. The resulting affordability gaps are as follows: Affordability Gap Calculation Unit Value Development Affordability /Sales Price Cost Gap Affordable Rental Units Extremely Low(Under 30%AMI) ($17,300)1 $350,000 $367,300 Very Low(30%to 50%AMI) $61,700 $350,000 $288,300 Low(50%to 80%AMI) $180,700 $350,000 $169,300 Affordable Ownership Units Moderate (80% to 120%AMI) $421,150 $491,000 $69,850 140%AMI Tier(120% to 140%AMI) $512,650 $491,000 none 'Negative unit value represents capitalized operating subsidy required. For the 140% AMI Tier, the affordable sales price exceeds the development cost for the unit, so there is no affordability gap for that income tier when affordable units are provided in a townhome product. Tables D-1 to D-4 present the detailed affordability gap calculations. Caveat on Affordability Gaps The development costs that have been estimated for purposes of the affordability gaps are based on "average" costs for affordable units, not on specific projects or in specific locations. Because they are based on average costs, the affordability gaps quantified herein likely underestimate the costs to construct units in more expensive areas of Honolulu and overestimate the costs to construct in less expensive areas. The primary cost variable from one part of the island to another is land acquisition costs. Total Nexus Costs The last step in the linkage fee analysis marries the findings on the numbers of households in each of the lower income ranges associated with the five prototypes to the affordability gaps, or the costs of delivering housing to them in Honolulu. Table D-5 summarizes the analysis. The Affordability Gaps are drawn from the prior discussion. The "Total Nexus Cost per Market Rate Unit" shows the results of the following calculation: the affordability gap times the number of affordable units demanded per market rate unit. (Demand Keyser Marston Associates, Inc. Page 50 \\Sf-fs2\wp\14\14100\002\001-002.docx for affordable units for each of the income ranges is drawn from Table C-3 in the previous section and is adjusted to a per-unit basis from the 100 unit building module.) The total nexus costs per market rate unit are as follows: Nexus Cost Per Market Rate Unit Affordability Single Low-Rise Mid-Rise High-Rise Rental Income Category Gap Family Townhomes Condo Condo (PUC) Apartment Ext. Low(30% - 50%AMI) $367,300 $14,600 $13,200 $11,900 $15,700 $11,800 Very Low(30% -50%AMI) $288,300 $22,200 $20,100 $18,300 $23,800 $18,000 Low(50%-80%AMI) $169,300 $13,500 $12,200 $11,200 $14,500 $11,000 Moderate (80%-120%AMI) $69,850 $3,500 $3,100 $2,900 $3,700 $2,800 140%Tier(120%-140%AMI) none $0 $0 $0 $0 $0 Total Nexus Costs $53,800 $48,600 $44,300 $57,700 $43,600 The chart below illustrates how the above nexus costs per unit are calculated: Calculation of Nexus Cost Per Market-Rate Unit Affc r lability #11 . 1 1�per _(di a E l - Nexus Cost affordable unit - per market- rate unit _. The Total Nexus Costs, or Mitigation Costs, indicated above, may also be expressed on a per square foot level. The square foot area of the prototype unit used throughout the analysis becomes the basis for the calculation. Again, see Appendix 1 for more discussion of the prototypes. The results per square foot of building area are as follows: Total Nexus Cost Per Sq.Ft. of Building Area Affordability Single Low-Rise Mid-Rise High-Rise Rental Income Category Gap Family Townhomes Condo Condo(PUC) Apartment Prototype Size 1,700 SF 1,200 SF 1,000 SF 1,000 SF 900 SF Ext. Low(30% - 50%AMI) $367,300 $8.60 $11.00 $11.90 $15.70 $13.10 Very Low(30%- 50%AMI) $288,300 $13.10 $16.80 $18.30 $23.80 $20.00 Low(50%-80%AMI) $169,300 $7.90 $10.20 $11.20 $14.50 $12.20 Moderate (80%-120%AMI) $69,850 $2.10 $2.60 $2.90 $3.70 $3.10 140% Tier(120%-140%) none $0.00 $0.00 $0.00 $0.00 $0.00 Total Nexus Costs $31.70 $40.60 $44.30 $57.70 $48.40 Keyser Marston Associates, Inc. Page 51 \\Sf-fs2\wp\14\14100\002\001-002.docx These costs express the total linkage or nexus costs for the five prototype developments in Honolulu. These total nexus costs represent the cost of creating new affordable units to offset the increased affordable housing needs associated with new market-rate residential development. The totals are not recommended levels for fees; many other policy considerations may be brought to bear in selecting appropriate fee levels. Keyser Marston Associates, Inc. Page 52 \\Sf-fs2\wp\14\14100\002\001-002.docx TABLE D-1 AFFORDABILITY GAP FOR FOR-SALE UNITS RESIDENTIAL NEXUS ANALYSIS CITY AND COUNTY OF HONOLULU N120%AIM s r 140%AMI I. Affordable Sale Price Building Type 3-BR Townhome 3-BR Townhome Household Size 4-Person HH 4-Person HH Unit Size 1,200 SF 1,200 SF Affordable Sale Price $421,150 (1) $512,650 (1) II. Development Costs - Per Unit Land Acquisition $140,000 (2) $140,000 (2) Direct Construction $260,000 (3) $260,000 (3) Indirects & Financing $91,000 (4) $91,000 (4) Total Costs $491,000 (5) $491,000 (5) III. Affordability Gap - Per Unit Affordable Sale Price $421,150 $512,650 (Less) Development Costs ($491,000) ($491,000) Affordability Gap ($69,850) $21,650 (1)See Table 4 for affordable home price estimates. (2)Land acquisition cost based on review of residential land values for lower density prototypes from the City and County of Honolulu property information database. (3)Direct construction cost based on RS Means estimate for 2-story, 1,200 sq. ft.townhome unit with 2- car garage. Cost includes major appliances. (4)Indirect& Financing costs based on 35%of direct cost based on data on Hawaii affordable multi-family rental projects(35%to 40%).This figure is considered conservative because multi-family projects typically do not have additional costs of unit sales or carrying costs related to the sales absorption period. (5)As a cross-check on total development costs,the total development profit at the average island-wide sale price of$575,000 would be$84,000($575,000 minus$491,000), or approximately 15%of the market rate sale price.This is an indication that the development cost estimate is not excessive. Page 53 Prepared by: Keyser Marston Associates Filename:\\Sf-fs2\wp\14\14100\002\Honolulu Affordability Gaps 12 18 14 TABLE D-2 AFFORDABILITY GAPS FOR 30% TO 80% AMI HOUSEHOLDS RESIDENTIAL NEXUS ANALYSIS CITY AND COUNTY OF HONOLULU 30%AMI 80%AMI 80% AMC I. Affordable Rent - Per Unit Average Number of Bedrooms 2 Bedrooms 2 Bedrooms 2 Bedrooms Average Household Size 3 Person 3 Person 3 Person Annual Household Income $25,890 $43,150 $69,040 Monthly Household Income $2,158 $3,596 $5,753 Income Allocation to Housing 33% 33% 33% Monthly Housing Cost $719 $1,199 $1,918 (Less) Utility Allowance(') ($387) ($387) ($387) Maximum Monthly Rent $332 $812 $1,531 II. Net Operating Income - Per Unit Annual Rental Income $3,986 $9,739 $18,369 Other Income $250 $250 $250 (Less) Vacancy 5.0% ($212) ($499) ($931) Effective Gross Income (EGI) $4,024 $9,490 $17,688 (Less) Operating Expenses ($5,500) ($5,500) ($5,500) Net Operating Income (NOI) ($1,476) $3,990 $12,188 III. Affordabiliy Gap - Per Unit Sources of Funds Supported Mortgage ($21,000) (2) $58,000 $177,000 Deferred Developer Fee $3,700 $3,700 $3,700 Total Sources ($17,300) $61,700 $180,700 (Less) Total Development Costs(3) ($350,000) ($350,000) ($350,000) Affordability Gap ($367,300) ($288,300) ($169,300) (1) Utility allowances from DPP Home Price and Rental Rates Guideline(effective January 2014) (2) Negative mortgage amount for 30%AMI units represents capitalized operating subsidy required. (3) Development costs based on the average cost per square foot of Hawaii tax credit projects from 2010 to 2014, excluding land acquisition costs, multiplied by an average 900 square foot 2-bedroom apartment unit, plus a land cost estimated at $30,000/unit. Page 54 Prepared by:Keyser Marston Associates Filename:\\Sf-fs2\wp\14\14100\002\Honolulu Affordability Gaps 12 18 14; TABLE D-3 ESTIMATED AFFORDABLE HOME PRICES-120%AMI RESIDENTIAL NEXUS ANALYSIS CITY AND COUNTY OF HONOLULU Unit Size 1-Bedroom Unit 2-Bedroom Unit 3-Bedroom Unit 4-Bedroom Unit Household Size 2-person HH 3-person HH 4-person HH 5-person HH Median Income-Honolulu County 2014 $76,700 $86,300 $95,800 $103,500 Annual Income @ 120% $92,040 $103,560 $114,960 $124,200 %Available for Housing Costs 33% 33% 33% 33% Income Available for Housing Costs $30,680 $34,520 $38,320 $41,400 (Less) Property Taxes ($1,023) ($1,143) ($1,263) ($1,353) (Less) HOA ($3,600) ($3,900) ($4,200) ($4,500) (Less) Utilities ($2,664) ($3,324) ($3,936) ($4,512) (Less) Insurance ($600) ($700) ($800) ($900) (Less) Mortgage Insurance ($4,212) ($4,706) ($5,200) ($5,577) Income Available for Mortgage $18,581 $20,747 $22,921 $24,558 Mortgage Amount $324,300 $362,100 $400,100 $428,700 Down Payment(homebuyer cash) $17,050 $19,050 $21,050 $22,550 Affordable Home Price* $341,350 $381,150 $421,150 $451,250 Key Assumptions -Mortgage Interest Rate 4.00% 4.00% 4.00% 4.00% -Down Payment 5.0% 5.0% 5.0% 5.0% -Property Taxes(%of sales price) 0.30% 0.30% 0.30% 0.30% -HOA(per month) $300 $325 $350 $375 -Utilities(per month) $222 $277 $328 $376 -Mortgage Insurance(% of loan amount) 1.30% 1.30% 1.30% 1.30% *Current DPP Calculated Price $571,493 $673,101 $714,087 $771,348 Source:All assumptions by KMA.Note:current DPP pricing schedules do not account for taxes,insurance,utilities,or HOA.Thus DPP's price schedules indicate significantly higher affordable prices than shown in this table.DPP staff do not currently keep data on actual sale prices of affordable units. Page 55 Prepared by:Keyser Marston Associates Filename:\\Sf-fs2\wp\14\14100\002\Honolulu Affordability Gaps 12 18 14;120%Prices TABLE D-4 ESTIMATED AFFORDABLE HOME PRICES-140%AMI RESIDENTIAL NEXUS ANALYSIS CITY AND COUNTY OF HONOLULU Unit Size 1-Bedroom Unit 2-Bedroom Unit 3-Bedroom Unit 4-Bedroom Unit Household Size 2-person HH 3-person HH 4-person HH 5-person HH Median Income-Honolulu County 2014 $76,700 $86,300 $95,800 $103,500 Annual Income @ 140% $107,380 $120,820 $134,120 $144,900 %Available for Housing Costs 33% 33% 33% 33% Income Available for Housing Costs $35,793 $40,273 $44,707 $48,300 (Less) Property Taxes ($1,242) ($1,392) ($1,539) ($1,650) (Less) HOA ($3,600) ($3,900) ($4,200) ($4,500) (Less) Utilities ($2,664) ($3,324) ($3,936) ($4,512) (Less) Insurance ($600) ($700) ($800) ($900) (Less)Mortgage Insurance ($5,122) ($5,720) ($6,331) ($6,799) Income Available for Mortgage $22,565 $25,237 $27,901 $29,939 Mortgage Amount $393,900 $440,500 $487,000 $522,600 Down Payment(homebuyer cash) $20,700 $23,200 $25,650 $27,500 Affordable Home Price $414,600 $463,700 $512,650 $550,100 Key Assumptions -Mortgage Interest Rate 4.00% 4.00% 4.00% 4.00% -Down Payment 5.0% 5.0% 5.0% 5.0% -Property Taxes(%of sales price) 0.30% 0.30% 0.30% 0.30% -HOA(per month) $300 $325 $350 $375 -Utilities(per month) $222 $277 $328 $376 -Mortgage Insurance(%of loan amount) 1.30% 1.30% 1.30% 1.30% *Current DPP Calculated Price $666,763 $750,295 $833,081 $899,906 Source:All assumptions by KMA.Note:current DPP pricing schedules do not account for taxes,insurance,utilities,or HOA.Thus DPP's price schedules indicate significantly higher affordable prices than shown in this table.DPP staff do not currently keep data on actual sale prices of affordable units. • Page 56 Prepared by:Keyser Marston Associates Filename:\\Sf-fs2\wp\14\14100\002\Honolulu Affordability Gaps 12 18 14;140%Prices TABLE D-5 NEXUS COST RESIDENTIAL NEXUS ANALYSIS HONOLULU,HI TOTAL NEXUS COST PER MARKET RATE UNIT Nexus Cost Per Market Rate Unit' Prototype 1: Prototype 2: Prototype 3: Prototype 4: Prototype 5: Affordability Single Low-Rise Mid-Rise High-Rise Rental Gap' Family Townhomes Condo Condo(PUC) Apartment Household Income Level Under 30%Area Median Income $367,300 ' $14,600 $13,200 $11,900 $15,700 $11,800 30%to 50%Area Median Income $288,300 1 $22,200 $20,100 $18,300 $23,800 $18,000 50%to 80%Area Median Income $169,300 ' $13,500 $12,200 $11,200 $14,500 $11,000 80%to 120%Area Median Income $69,850 2 $3,500 $3,100 $2,900 $3,700 $2,800 120%to 140%Area Median Income no gap 2 $0 $0 $0 $0 $0 Total Nexus Cost Per Unit $53,800 $48,600 $44,300 $57,700 $43,600 TOTAL NEXUS COST PER SQUARE FOOT OF BUILDING AREA4 Nexus Cost Per Square Foot(Net Rentable/Sellable)' Prototype 1: Prototype 2: Prototype 3: Prototype 4: Prototype 5: Single Low-Rise Mid-Rise High-Rise Rental Family Townhomes Condo Condo(PUC) Apartment Unit Size(SF) 1,700 SF 1,200 SF 1,000 SF 1,000 SF 900 SF Household Income Level Under 30%Area Median Income $8.60 $11.00 $11.90 $15.70 $13.10 30%to 50%Area Median Income $13.10 $16.80 $18.30 $23.80 $20.00 50%to 80%Area Median Income $7.90 $10.20 $11.20 $14.50 $12.20 80%to 120%Area Median Income $2.10 $2.60 $2.90 $3.70 $3.10 120%to 140%Area Median Income $0.00 $0.00 $0.00 $0.00 $0.00 Total Nexus Cost $31.70 $40.60 $44.30 $57.70 $48.40 Notes: I Assumes affordable rental units. 2 Affordability gap for moderate income households based on ownership unit(townhome)priced at 110%AMI. No affordability gap in a townhome product at 140%AMI. 3 Nexus cost per unit computed by multiplying affordable unit demand per 100 units from Table C-3 by the affordability gap and dividing by 100 units. `Computed by dividing the nexus cost per unit by the square footage of the unit. Prepared by: Keyser Marston Associates, Inc. Page 57 \\Sf-fs2\wp\14\14100\002\Honolulu Nexus model 8-31-15; 8/31/2015;dd ADDENDUM: ADDITIONAL BACKGROUND AND NOTES ON SPECIFIC ASSUMPTIONS No Excess Supply of Affordable Housing An assumption of this residential nexus analysis is that there is no excess supply of affordable housing available to absorb or offset new demand; therefore, new affordable units are needed to mitigate the new affordable housing demand generated by development of new market rate residential units. Based on a review of the draft Housing Oahu: Islandwide Housing Strategy, conditions in Honolulu are consistent with this underlying assumption. According to the Islandwide Housing Strategy, Oahu needs over 24,000 additional housing units to address pent- up demand and near-term projected housing needs through 2016. An estimated 75% of the demand is for units affordable to households earning less than 80% of AMI. Census data (2011 to 2013 ACS) indicates that the percentage of households spending more than 30% of their income on housing is 43% of all households in Honolulu. Households who are spending more than 30% of their income on housing are heavily those in the lower income categories. Affordability Gap The use of the affordability gap for establishing the nexus costs is grounded in the concept that a jurisdiction will use in-lieu fee revenues to help subsidize new affordable units needed to mitigate impacts. The nexus analysis has established that units will be needed at one or more different affordability levels and the type of unit to be delivered depends on the income/affordability level. In Honolulu, the City and County is anticipated to assist in the development of rental units for household incomes less than 80% of median and for moderate income households, ownership units are assumed to be assisted. The units assisted by the public sector for affordable households are usually small in square foot area (for the number of bedrooms) and modest in finishes and amenities. As a result, in some communities these units are similar in physical configuration to what the market is delivering at market rate; in other communities (particularly very high income communities), they may be smaller and more modest than what the market is delivering. Parking, for example, is usually the minimum permitted by the code. In some communities where there is a wide range in land cost per acre or per unit, it may be assumed that affordable units are built on land parcels in the lower portion of the cost range. KMA tries to develop a total development cost summary that represents the lower half of the average range, but not so low as to be unrealistic. If the affordability gap is the difference between total development cost and the affordable sales price, the question sometimes arises as to how total development cost is defined. KMA defines total development costs as including land costs, construction costs, site improvements, architectural and engineering, financing and all other indirect costs, and an allowance for an industry profit (non-profit developers receive a development fee instead). Keyser Marston Associates, Inc. Page 58 \\Sf-fs2\wp\14\14100\002\001-002.d ocx Excess Capacity of Labor Force In the context of economic downturns such as the recent severe recession, the question is sometimes raised as to whether there is excess capacity in the labor force to the extent that consumption impacts generated by new households will be in part, absorbed by existing jobs and workers, thus resulting in fewer net new jobs. In response, an impact analysis of this nature is a one-time impact requirement to address impacts generated over the life of the project. Recessions are temporary conditions; a healthy economy will return and the impacts will be experienced. The economic cycle also self-adjusts. Development of new residential units is not likely to occur until conditions improve or there is confidence that improved conditions are imminent. When this occurs, the improved economic condition of the households in the local area will absorb the current underutilized capacity of existing workers, employed and unemployed. By the time new units become occupied, economic conditions will have likely improved. The Burden of Paying for Affordable Housing Honolulu's proposed affordable housing requirements do not place all burden for the creation of affordable housing on new residential construction. The burden of affordable housing is also borne by many sectors of the economy and society. A most important source in recent years of funding for affordable housing development comes from the federal government in the form of tax credits (which result in reduced income tax payment by tax credit investors in exchange for equity funding). Additionally there are other federal grant and loan programs administered by the Department of Housing and Urban Development and other federal agencies. Local governments play a large role in affordable housing. In addition, private sector lenders play an important role, some voluntarily and others less so with the requirements of the Community Reinvestment Act. Then there is the non-profit sector, both sponsors and developers that build much of the affordable housing. In summary, all levels of government and many private parties, for profit and non-profit contribute to supplying affordable housing. Residential developers are not being asked to bear the burden alone any more than they are assumed to be the only source of demand or cause for needing affordable housing in our communities. Disclaimers This report has been prepared using the best and most recent data available at the time of the analysis. Local data and sources were used wherever possible. Major sources include the U.S. Census Bureau: 2011-2013 American Community Survey, U.S. Bureau of Labor Statistics, and the IMPLAN model. While we believe all sources utilized are sufficiently sound and accurate for the purposes of this analysis, we cannot guarantee their accuracy. Keyser Marston Associates, Inc. assumes no liability for information from these and other sources. Keyser Marston Associates, Inc. Page 59 \\Sf-fs2\wp\14\14100\002\001-002.d ocx APPENDIX 1: MARKET SURVEY Keyser Marston Associates, Inc. Page 60 \\Sf-fs2\wp\14\14100\002\001-002.docx MARKET RATE PROTOTYPES & PRICING Since the purpose of the Residential Nexus Analysis is to quantify the impacts from new market rate residential development, it is necessary to identify the types of new market rate housing that is being built in Honolulu today. Identification of the market rate housing prototypes and sale prices and rents associated with those prototypes, forms the initial input into the Nexus Analysis. This section starts with a general overview of the Honolulu housing market as context for the types of housing being built in Honolulu, where it is being built, and how different geographic areas of the island differ from each other. It then describes the market rate prototypes selected for the Nexus Analysis and then concludes with islandwide average sale price and rent conclusions for each of the prototypes. I. Overview of Honolulu Housing Market Most residential development in Honolulu is concentrated in the southern and western areas of the island. KMA has assessed the extent to which residential development characteristics and pricing vary from part of the island to another. This analysis assessed market data in relation to the City's eight Development Plan Areas. In terms of residential sales activity, the largest number of 2014 sales was in the "Primary Urban Center" (PUC) plan area followed by the "Ewa" plan area. Lower shares of residential sales were located in the remaining plan areas. uarffiT1ttt t ��1 II -11111/1 North Shore io\ Central Oahu II waianae i l Ewa t G ;r d s l'' 40 "II Primary Urban Center x East Honolulu EiEjserow wry Urban Center MINI Urban-Fringe lllllllll Rural Keyser Marston Associates, Inc. Page 61 \\Sf-fs2\wp\14\14100\002\001-002.docx In general terms, the housing stock is older in the PUC and leeward areas of the island. On average, homes are newer in the Ewa and Central areas where growth has occurred more recently. As would be expected, housing units tend to be larger in the more suburban areas and smaller in the urban areas. Home values are highest in the East Honolulu, PUC, and Koolaupoku plan areas but this is partly a function of the larger size of homes in these areas. These three areas have the highest home values even after adjusting for unit size but the differences compared to the other plan areas are less significant. Figure 1.Median Single Family Home Price (units built since 1990) Data quick $1,800,000 - $1,600,000 $1,400,000 $1,200,000 $1,800,000 $800,000 -- $600,000 --- — $400,000 -•__.W. $200,000 $0 a0 • Ja0\ • �J Q JG °\J\) New residential development going forward is planned to be concentrated in the PUC, Ewa, and Central plan areas, with the largest numbers planned for Ewa. As shown in the following chart, comparatively fewer housing units are planned for the other plan areas. Figure 2. Growth in Honolulu Housing Units(cumulative) City and County of Honolulu 30,000 25,000 20,000 ■Ewa 15,000 ■Primary Urban Center a Central Oahu 10,000 Other Plan Areas 5,000 - 0 2010-15 2010-20 2010-25 2010-30 2010-35 Keyser Marston Associates, Inc. Page 62 \\Sf-fs2\wp\14\14100\002\001-002.docx The City is also targeting residential growth in areas along the new elevated rail system in compact, mixed use development. II. Residential Prototypes The City and County of Honolulu, like many large municipalities with both urban and suburban areas, has a diverse and wide range of housing types. On one end of the spectrum, Honolulu has relatively large single family detached homes at low densities. On the other end of the spectrum are high-rise residential towers. It is recognized that all prototypes are not built in all areas of the island. For example, high-rises are more highly concentrated in the more urbanized areas of the PUC plan area. With the assistance of City staff, KMA has identified five residential prototypes for the Nexus Analysis: Figure 3. Market Rate Prototypes Average Unit Size Single Family Detached 1,700 square feet Town homes 1,200 square feet Mid-Rise Condos 1,000 square feet High-Rise Condos (PUC area) 1,000 square feet Rental Apartments 900 square feet The intent of the prototypes is to be generally representative of the types of housing being built. It is not necessary to include every variation of unit size or density. III. Market Survey In order to estimate the current pricing of new construction projects, KMA undertook a survey of current home prices and rents throughout the island. The survey included a review of asking prices of new construction projects currently on the market, re-sale of units, and asking rental rates. a. For-Sale Prototypes For for-sale homes and condominiums, KMA surveyed both new construction projects and re- sales. The following chart plots 2014 sales of homes in new construction projects currently in the market. The new construction projects were identified by Hanley Wood, a third party market data provider. The survey identified projects in various parts of the island but some areas are better represented than others. Most of the projects are located in the Ewa, Central, and Waianae plan areas with a few projects in the Koolaupoku plan area. Keyser Marston Associates, Inc. Page 63 \\Sf-fs2\wp\14\14100\002\001-002.docx Figure 4. Honolulu 2014 New Construction Home Sales Hanley Wood, Dataquick $1,200,000 $1,100,000 • ♦ $1,000,000 ♦.__.._ $900,000 • • $800,000 $700,000 atiVAL $600,000 ------` .F $500,000 ■A UN it ... $400,000 $300,000 ♦ $200,000 ♦• ♦ $100,000 $0 , 0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 Unit Square Feet ♦Hoakalei (Ewa Beach) x Mahinui-Kaneohe ■Gentry Homes(Ewa Beach) xAwakea at Mehana(Kapolei) •Kahiwelo at Makakilo(Kapolei) +Makaha Oceanview(Waianae) Sea Country-Makalae(Waianae) •Plantation Town Apts(Waipahu) •Ka Malanai Kailua Keyser Marston Associates, Inc. Page 64 \\Sf-fs2\wp\14\14100\002\001-002.docx Y Z� New For-Sale Residential Projects Honoluluotaf M fir ()sea GaurtrY Makaiae ,' �"'.3� Q.Ka.MaYtii _>d R3 KaNwalo et Mak*kilo Awakxa at Mehaaa G .-- $a GennYHomaa Pa hahia M..•._. , iQ Sonny wawa sandalwood .. _ .. O Noakalai .._,. . 55 In an effort to have better geographic representation, as well as to increase the number of home sales for statistical validity, KMA also pulled data on re-sales of existing homes in the market. Re-sale data was focused on "newer" homes in the market (homes built since 1990) because newer homes are more likely to share price similarities to new construction projects than older homes. By including re-sales, there was a much larger number of sales to assess the market (from January to October 2014, there were over 2,000 sales of"newer" units). All geographic areas of the island were represented by that data set. The following charts separate the sales of single family homes from condominiums. To better graphically present the data, sales of very large homes were not included in the chart. Foreclosure sales were also not included. Keyser Marston Associates, Inc. Page 65 \\Sf-fs2\wp\14\14100\002\001-002.docx Figure 5. Honolulu 2014 Single Family Home Prices(built since 1990) Dataquick $3,000,000 $2,750,000 • • $2,500,000 ♦ ♦ ♦ $2,250,000 — ---- • $2,000,000 ♦♦ • • $1,750,000 ♦ ♦ • $1,500,000 ♦ • • • ♦ • • $1,250,000 ♦••♦ ♦ ♦~ ♦!I . ♦ • • • $1,000,000 $750,000 • e $500,000 - ._ ♦• ♦ ♦ $250,000 $0 I I • • , I T , I I , 0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500 5,000 Unit Square Feet Figure 6. Honolulu 2014 Condominium Prices(built since 1990) Dataquick $2,000,000 ♦ ♦ S $1,750,000 _.._...............__._..._ ♦ . _.__._.._...... ..... • • • $1,500,000 - ♦ - S:- . ♦ ♦♦ -_.♦♦ • $1,250,000 • ♦- —.t-♦.._._ ..._ _ • ♦ • $1,000,000 ♦ *--__ -_- --_.__...♦.-.. ...."10 ♦ • $750,000 *UV • t.�.. �♦ 1 ,� - r< • $250,000 s 1. *...._..._. ♦" • $0 1 1 F I 0 500 1,000 1,500 2,000 2,500 Unit Square Feet b. Rental Prototypes Unlike many real estate markets on the mainland, in recent years there has been little new development of market rate rental projects in Honolulu. City staff has identified two proposed apartment projects that are currently in the development pipeline— 7000 Hawaii Kai and Kapolei Lofts. Both of these projects require a rezoning of the land in order to allow residential development and, consequently, both will have market rate and affordable units. Keyser Marston Associates, Inc. Page 66 \\Sf-fs2\wp\14\14100\002\001-002.docx • 7000 Hawaii Kai. This 269-unit project is proposed to consist of two ten-story buildings with units ranging from 700 to 1,100 square feet. According to a Pacific Business News article, the market rate rents are expected to range from $2,200 to $3,700/month. • Kapolei Lofts. This project is proposed to consist of 499 rental units including units at 80% and 140% of AMI. According to an article in the Pacific Business News, this project will be Honolulu's first Class A new construction apartment project. In addition, KMA surveyed existing apartment developments for asking rents. Even though some of these properties have been recently renovated, the rents for new construction projects would likely be somewhat higher than the rents from older properties on the market given the premium that many renters are willing to pay for modern amenities and unit layouts. Figure 7.Honolulu Apartment Rents Project websites, other internet research $4,000 $3,500 $3,000 )I KK s $2,500 ______ + ++ .I++ X $1,500 j XA .-._ X $1,000 $500 $0 0 250 500 750 1,000 1,250 1,500 1,750 Unit Square Feet 7000 Hawaii Kai XWaena Apartments AWaikiki Walina Apartments Palms of Kilani Kalaeloa Rental Homes -Waterfront at Pu'uloa Apartments ♦Moanalua Hillside Apartments xWaimanalo Keyser Marston Associates, Inc. Page 67 \\Sf-fs2\wp\14\14100\002\001-002.docx Apartment Developments Honolulu ' "'- j F1a.aiu.x,nade Aparmenl. ' i.e* cnbeba RanulHana Yaaterfreme at -- Emoetawaa carQ ' 0 Wt{k441'1!a?Yia Apnt A+rentt IV. Prototype Pricing The sale prices and rents for the five market rate prototypes have been estimated by KMA based on the market survey. The rents and prices for the apartments and the lower density for- sale prototypes have been estimated based on market averages. Since the City's inclusionary program will be an islandwide requirement, the use of averages is appropriate for purposes of the Nexus Analysis. The pricing for the High-Rise Condos is the one prototype not based on an islandwide average. All the comparable sale price data used to estimate the high-rise prototype was from developments in the PUC plan area, where most high-rise projects are currently being built. As shown in the following table, the for-sale prototypes are estimated to have average sale prices ranging from $525,000 for the Mid-Rise Condo Prototype to $700,000 for both the Single Detached Home and High-Rise Condo Prototypes. Keyser Marston Associates, Inc. Page 68 \\Sf-fs2\wp\14\14100\002\001-002.docx Fi•ure 7 Surma of Market Rate Rest •n .,.�. _ _ For-Sale Prototypes Rental Prototype 1 Prototype 2 Prototype 3 Prototype 4 Prototype 5 Single Family Low Rise Mid-Rise High-Rise Rental Detached Townhomes Condos Condos(PUC) Apartments Homes Average Unit Size 1,700 sf 1,200 sf 1,000 sf 1,000 sf 900 sf Market Rate Price/Rent $700,000 $575,000 $525,000 $700,000 $2,500 /mo. $/SF $412 /sf $479 /sf $525 /sf $700 /sf $2.78 /sf Source:Estimated by KMA based on market survey Keyser Marston Associates, Inc. Page 69 \\Sf-fs2\wp\14\14100\002\001-002 docx APPENDIX 2: SUPPORTING TECHNICAL TABLES Keyser Marston Associates, Inc. Page 70 \\Sf-fs2\wp\14\14100\002\001-002.docx APPENDIX 2, TABLE 1 WORKER OCCUPATION DISTRIBUTION, 2013 SERVICES TO HOUSEHOLDS EARNING $75,000-$100,000, RESIDENT SERVICES RESIDENTIAL NEXUS ANALYSIS HONOLULU, HI Worker Occupation Distribution' Services to Households Earning Major Occupations (2% or more) $75,000-$100,000 Management Occupations 4.2% Business and Financial Operations Occupations 4.5% Education, Training, and Library Occupations 2.8% Healthcare Practitioners and Technical Occupations 7.7% Healthcare Support Occupations 4.1% Food Preparation and Serving Related Occupations 12.8% Building and Grounds Cleaning and Maintenance Occupations 5.7% Personal Care and Service Occupations 5.1% Sales and Related Occupations 13.7% Office and Administrative Support Occupations 16.3% Installation, Maintenance, and Repair Occupations 3.7% Transportation and Material Moving Occupations 5.4% All Other Worker Occupations-Services to Households Earning 14.1% $75,000-$1 00,000 INDUSTRY TOTAL 100.0% Distribution of employment by industry is per the IMPLAN model and the distribution of occupational employment within those industries is based on the Bureau of Labor Statistics Occupational Employment Survey. Source: Bureau of Labor Statistics, IMPLAN Keyser Marston Associates, Inc. \\Sf-fs2\wp\14\14100\002\75-100k Honolulu 1-16-15; 9/8/2015; dd Page 71 APPENDIX 2,TABLE 2 AVERAGE ANNUAL WORKER COMPENSATION,2013 SERVICES TO HOUSEHOLDS EARNING$75,000-$100,000 RESIDENTIAL NEXUS ANALYSIS HONOLULU,HI %of Total Households Earning %of Total $75,000-$100,000, 2013 Avg. Occupation Resident Services Occupation' Compensation' Group 2 Workers Page 1 of 4 Management Occupations Chief Executives $150,500 3.5% 0.2% General and Operations Managers $100,200 32.7% 1.4% Sales Managers $81,100 5.4% 0.2% Administrative Services Managers $68,500 4.1% 0.2% Financial Managers $95,600 8.2% 0.3% Food Service Managers $48,500 4.6% 0.2% Medical and Health Services Managers $109,600 6.7% 0.3% Property,Real Estate,and Community Association Managers $52,400 9.6% 0.4% Managers,All Other $91,800 4.4% 0.2% All other Management Occupations(Avg.All Categories) $90,600 20.9% 0.9% Weighted Mean Annual Wage $90,600 100.0% 4.2% Business and Financial Operations Occupations Claims Adjusters,Examiners,and Investigators $62,100 6.1% 0.3% Compliance Officers $65,500 3.0% 0.1% Human Resources Specialists $58,200 5.5% 0.2% Labor Relations Specialists $53,900 3.2% 0.1% Management Analysts $77,100 6.2% 0.3% Training and Development Specialists $65,600 3.3% 0.1% Market Research Analysts and Marketing Specialists $57,700 6.0% 0.3% Business Operations Specialists,All Other $67,200 13.2% 0.6% Accountants and Auditors $62,200 17.1% 0.8% Financial Analysts $72,800 4.9% 0.2% Personal Financial Advisors $98,700 5.5% 0.2% Loan Officers $62,000 4.5% 0.2% All Other Business and Financial Operations Occupations(Avg.All Categories) $66,700 21.6% 1.0% Weighted Mean Annual Wage $66,700 100.0% 4.5% Education, Training, and Library Occupations Vocational Education Teachers,Postsecondary $53,200 3.9% 0.1% Preschool Teachers,Except Special Education $32,400 15.8% 0.4% Elementary School Teachers,Except Special Education $54,800 7.6% 0.2% Middle School Teachers,Except Special and Career/Technical Education $54,500 3.4% 0.1% Secondary School Teachers,Except Special and Career/Technical Education $56,500 5.2% 0.1% Self-Enrichment Education Teachers $46,900 9.4% 0.3% Substitute Teachers $44,700 4.2% 0.1% Teachers and Instructors,All Other,Except Substitute Teachers $50,000 6.8% 0.2% Teacher Assistants $27,800 15.6% 0.4% All Other Education,Training,and Library Occupations(Avg.All Categories) $42,000 28.3% 0.8% Weighted Mean Annual Wage $42,000 100.0% 2.8% • Sources:U.S.Bureau of Labor Statistics,IMPLAN Keyser Marston Associates,Inc. Page 72 \\Sf-fs2\wp\14\14100\002\75-100k Honolulu 1-16-15;9/8/2015;dd APPENDIX 2,TABLE 2 AVERAGE ANNUAL WORKER COMPENSATION,2013 SERVICES TO HOUSEHOLDS EARNING$75,000-$100,000 RESIDENTIAL NEXUS ANALYSIS HONOLULU,HI %of Total Households Earning %of Total $75,000-$100,000, 2013 Avg. Occupation Resident Services Occupation' Compensation' Group 2 Workers Page 2 of 4 Healthcare Practitioners and Technical Occupations Pharmacists $114,900 4.2% 0.3% Physicians and Surgeons,All Other $206,800 4.5% 0.3% Registered Nurses $88,500 31.0% 2.4% Dental Hygienists $70,700 3.6% 0.3% Pharmacy Technicians $36,700 5.5% 0.4% Licensed Practical and Licensed Vocational Nurses $46,700 8.6% 0.7% All Other Healthcare Practitioners and Technical Occupations(Avg.All Categorie: $87,300 42.6% 3.3% Weighted Mean Annual Wage $87,300 100.0% 7.7% Healthcare Support Occupations Home Health Aides $22,900 20.2% 0.8% Nursing Assistants $29,100 30.5% 1.3% Dental Assistants $33,300 10.8% 0.4% Medical Assistants $33,700 18.4% 0.8% Phlebotomists $34,200 3.1% 0.1% All Other Healthcare Support Occupations(Avg.All Categories) $29,300 17.0% 0.7% Weighted Mean Annual Wage $29,300 100.0% 4.1% Food Preparation and Serving Related Occupations First-Line Supervisors of Food Preparation and Serving Workers $34,700 7.0% 0.9% Cooks,Fast Food $19,800 4.6% 0.6% Cooks,Restaurant $26,000 8.9% 1.1% Food Preparation Workers $23,300 6.7% 0.9% Bartenders $29,800 5.1% 0.7% Combined Food Preparation and Serving Workers,Including Fast Food $19,700 25.9% 3.3% Counter Attendants,Cafeteria,Food Concession,and Coffee Shop $21,500 3.8% 0.5% Waiters and Waitresses $26,900 20.7% 2.6% Dining Room and Cafeteria Attendants and Bartender Helpers $24,600 3.2% 0.4% Dishwashers $22,900 4.1% 0.5% Hosts and Hostesses,Restaurant,Lounge,and Coffee Shop $23,500 3.1% 0.4% All Other Food Preparation and Serving Related Occupations(Avg.All Categories $24,400 7.0% 0.9% Weighted Mean Annual Wage $24,400 100.0% 12.8% Building and Grounds Cleaning and Maintenance Occupations First-Line Supervisors of Housekeeping and Janitorial Workers $40,200 3.4% 0.2% Janitors and Cleaners,Except Maids and Housekeeping Cleaners $25,300 50.8% 2.9% Maids and Housekeeping Cleaners $32,400 11.6% 0.7% Landscaping and Groundskeeping Workers $28,800 25.4% 1.4% All Other Building and Grounds Cleaning and Maintenance Occupations(Avg.All $27,700 8.8% 0.5% Weighted Mean Annual Wage $27,700 100.0% 5.7% Sources:U.S.Bureau of Labor Statistics,IMPLAN Keyser Marston Associates, Inc. Page 73 \\Sf-fs2\wp\14\14100\002\75-100k Honolulu 1-16-15;9/8/2015;dd APPENDIX 2,TABLE 2 AVERAGE ANNUAL WORKER COMPENSATION,2013 SERVICES TO HOUSEHOLDS EARNING$75,000-$100,000 RESIDENTIAL NEXUS ANALYSIS HONOLULU,HI %of Total Households Earning %of Total $75,000-$100,000, 2013 Avg. Occupation Resident Services Occupation 3 Compensation 1 Group 2 Workers Page 3 of 4 Personal Care and Service Occupations First-Line Supervisors of Personal Service Workers $42,000 3.7% 0.2% Nonfarm Animal Caretakers $25,100 4.5% 0.2% Amusement and Recreation Attendants $20,800 6.4% 0.3% Hairdressers,Hairstylists,and Cosmetologists $36,400 14.4% 0.7% Manicurists and Pedicurists $20,000 3.1% 0.2% Childcare Workers $19,100 13.6% 0.7% Personal Care Aides $25,500 26.0% 1.3% Fitness Trainers and Aerobics Instructors $32,000 5.3% 0.3% Recreation Workers $34,400 4.9% 0.2% All Other Personal Care and Service Occupations(Avg.All Categories) $27,400 18.3% 0.9% Weighted Mean Annual Wage $27,400 100.0% 5.1% Sales and Related Occupations First-Line Supervisors of Retail Sales Workers $48,500 9.2% 1.3% Cashiers $22,500 24.6% 3.4% Counter and Rental Clerks $29,100 4.8% 0.7% Retail Salespersons $25,500 34.9% 4.8% Securities,Commodities,and Financial Services Sales Agents $77,000 3.2% 0.4% Sales Representatives,Services,All Other $48,400 3.8% 0.5% Sales Representatives,Wholesale and Manufacturing,Except Technical and Scie $44,600 5.8% 0.8% All Other Sales and Related Occupations(Avg.All Categories) $31,500 13.7% 1.9% Weighted Mean Annual Wage $31,500 100.0% 13.7% Office and Administrative Support Occupations First-Line Supervisors of Office and Administrative Support Workers $51,400 6.7% 1.1% Bookkeeping,Accounting,and Auditing Clerks $37,300 7.3% 1.2% Customer Service Representatives $34,500 10.9% 1.8% Receptionists and Information Clerks $29,000 6.8% 1.1% Stock Clerks and Order Fillers $26,500 10.0% 1.6% Executive Secretaries and Executive Administrative Assistants $50,700 3.2% 0.5% Medical Secretaries $37,200 3.9% 0.6% Secretaries and Administrative Assistants,Except Legal,Medical,and Executive $38,300 10.0% 1.6% Office Clerks,General $31,300 13.2% 2.1% All Other Office and Administrative Support Occupations(Avg.All Categories) $35,500 28.0% 4.6% Weighted Mean Annual Wage $35,500 100.0% 16.3% Installation, Maintenance, and Repair Occupations First-Line Supervisors of Mechanics,Installers,and Repairers $71,900 7.9% 0.3% Telecommunications Equipment Installers and Repairers,Except Line Installers $60,300 3.6% 0.1% Automotive Body and Related Repairers $47,900 3.9% 0.1% Automotive Service Technicians and Mechanics $47,200 16.0% 0.6% Bus and Truck Mechanics and Diesel Engine Specialists $63,200 3.6% 0.1% Maintenance and Repair Workers,General $42,400 33.5% 1.2% All Other Installation,Maintenance,and Repair Occupations(Avg.All Categories) $49,300 31.5% 1.2% Weighted Mean Annual Wage $49,300 100.0% 3.7% Sources:U.S.Bureau of Labor Statistics,IMPLAN Keyser Marston Associates,Inc. Page 74 \\Sf-fs2\wp\14\14100\002\75-100k Honolulu 1-16-15;9/8/2015;dd APPENDIX 2,TABLE 2 AVERAGE ANNUAL WORKER COMPENSATION,2013 SERVICES TO HOUSEHOLDS EARNING$75,000-$100,000 RESIDENTIAL NEXUS ANALYSIS HONOLULU,HI %of Total Households Earning %of Total $75,000-$100,000, 2013 Avg. Occupation Resident Services Occupation 3 Compensation' Group 2 Workers Page 4 of 4 Transportation and Material Moving Occupations Bus Drivers,School or Special Client $34,500 7.1% 0.4% Driver/Sales Workers $29,400 7.7% 0.4% Heavy and Tractor-Trailer Truck Drivers $44,500 11.3% 0.6% Light Truck or Delivery Services Drivers $30,600 8.7% 0.5% Taxi Drivers and Chauffeurs $26,800 4.1% 0.2% Parking Lot Attendants $21,700 4.2% 0.2% Industrial Truck and Tractor Operators $40,100 3.3% 0.2% Cleaners of Vehicles and Equipment $25,000 5.5% 0.3% Laborers and Freight,Stock,and Material Movers,Hand $30,900 23.1% 1.2% Packers and Packagers,Hand $21,700 7.7% 0.4% All Other Transportation and Material Moving Occupations(Avg.All Categories) $31,300 17.3% 0.9% Weighted Mean Annual Wage $31,300 100.0% 5.4% 85.9% The methodology utilized by the Bureau of Labor Statistics(BLS)generally assumes hourly employees are employed full-time.Annual compensation is calculated by BLS by multiplying hourly wages by 2,080 hours per year(40 hours per week and 52 weeks). 2 Occupation percentages are based on the 2013 National Industry-Specific Occupational Employment survey compiled by the Bureau of Labor Statistics. Wages are based on the 2013 Occupational Employment Survey data specific to Honolulu from the Bureau of Labor statistics. 3 Including occupations representing 3%or more of the major occupation group. Sources:U.S.Bureau of Labor Statistics,IMPLAN Keyser Marston Associates,Inc. Page 75 \\Sf-fs2\wp\14\14100\002\75-100k Honolulu 1-16-15;9/8/2015;dd APPENDIX 2,TABLE 3 WORKER OCCUPATION DISTRIBUTION, 2013 SERVICES TO HOUSEHOLDS EARNING $100,000-150,000, RESIDENT SERVICES RESIDENTIAL NEXUS ANALYSIS HONOLULU, HI Worker Occupation Distribution' Services to Households Earning Major Occupations (2% or more) $100,000-$150,000 Management Occupations 4.2% Business and Financial Operations Occupations 4.5% Education, Training, and Library Occupations 3.3% Healthcare Practitioners and Technical Occupations 7.4% Healthcare Support Occupations 3.9% Food Preparation and Serving Related Occupations 12.7% Building and Grounds Cleaning and Maintenance Occupations 5.6% Personal Care and Service Occupations 5.2% Sales and Related Occupations 14.1% Office and Administrative Support Occupations 16.2% Installation, Maintenance, and Repair Occupations 3.5% Transportation and Material Moving Occupations 5.4% All Other Worker Occupations-Services to Households Earning 14.0% $100,000-$150,000 INDUSTRY TOTAL 100.0% 1 Distribution of employment by industry is per the IMPLAN model and the distribution of occupational employment within those industries is based on the Bureau of Labor Statistics Occupational Employment Survey. Source: Bureau of Labor Statistics, IMPLAN Keyser Marston Associates, Inc. \\Sf-fs2\wp\14\14100\002\100-150k Honolulu 1-16-15; 9/8/2015; dd Page 76 APPENDIX 2,TABLE 4 AVERAGE ANNUAL WORKER COMPENSATION,2013 SERVICES TO HOUSEHOLDS EARNING$100,000-$150,000 RESIDENTIAL NEXUS ANALYSIS HONOLULU,HI %of Total Households Earning %of Total $100,000-150,000, 2013 Avg. Occupation Resident Services Occupation 3 Compensation 1 Group 2 Workers Page 1 of 4 Management Occupations Chief Executives $150,500 3.6% 0.1% General and Operations Managers $100,200 33.1% 1.4% Sales Managers $81,100 5.3% 0.2% Administrative Services Managers $68,500 4.1% 0.2% Financial Managers $95,600 8.4% 0.3% Food Service Managers $48,500 4.6% 0.2% Medical and Health Services Managers $109,600 6.4% 0.3% Property,Real Estate,and Community Association Managers $52,400 7.8% 0.3% Social and Community Service Managers $58,000 3.2% 0.1% Managers,All Other $91,800 4.5% 0.2% All other Management Occupations(Avg.All Categories) $90,200 18.9% 0.8% Weighted Mean Annual Wage $90,200 100.0% 4.2% Business and Financial Operations Occupations Claims Adjusters,Examiners,and Investigators $62,100 6.1% 0.3% Compliance Officers $65,500 3.0% 0.1% Human Resources Specialists $58,200 5.4% 0.2% Labor Relations Specialists $53,900 3.3% 0.1% Management Analysts $77,100 6.2% 0.3% Training and Development Specialists $65,600 3.4% 0.2% Market Research Analysts and Marketing Specialists $57,700 5.8% 0.3% Business Operations Specialists,All Other $67,200 13.3% 0.6% Accountants and Auditors $62,200 16.7% 0.7% Financial Analysts $72,800 5.0% 0.2% Personal Financial Advisors $98,700 5.8% 0.3% Loan Officers $62,000 4.7% 0.2% All Other Business and Financial Operations Occupations(Avg.All Categories) $66,900 21.2% 1.0% Weighted Mean Annual Wage $66,900 100.0% 4.5% Education, Training,and Library Occupations Vocational Education Teachers,Postsecondary $53,200 3.9% 0.1% Preschool Teachers,Except Special Education $32,400 16.1% 0.5% Elementary School Teachers,Except Special Education $54,800 7.4% 0.2% Middle School Teachers,Except Special and Career/Technical Education $54,500 3.3% 0.1% Secondary School Teachers,Except Special and Career/Technical Education $56,500 5.1% 0.2% Self-Enrichment Education Teachers $46,900 9.1% 0.3% Substitute Teachers $44,700 4.0% 0.1% Teachers and Instructors,All Other,Except Substitute Teachers $50,000 6.8% 0.2% Teacher Assistants $27,800 15.5% 0.5% All Other Education,Training,and Library Occupations(Avg.All Categories) $41,800 28.8% 0.9% Weighted Mean Annual Wage $41,800 100.0% 3.3% Sources:U.S.Bureau of Labor Statistics,IMPLAN Keyser Marston Associates,Inc. Page 77 \\Sf-fs2\wp\14\14100\002\100-150k Honolulu 1-16-15;9/8/2015;dd APPENDIX 2,TABLE 4 AVERAGE ANNUAL WORKER COMPENSATION,2013 SERVICES TO HOUSEHOLDS EARNING$100,000-$150,000 RESIDENTIAL NEXUS ANALYSIS HONOLULU,HI % of Total Households Earning %of Total $100,000150,000, 2013 Avg. Occupation Resident Services Occupation 3 Compensation I Group 2 Workers Page 2 of 4 Healthcare Practitioners and Technical Occupations Pharmacists $114,900 4.5% 0.3% Physicians and Surgeons,All Other $206,800 4.4% 0.3% Registered Nurses $88,500 30.7% 2.3% Dental Hygienists $70,700 3.6% 0.3% Pharmacy Technicians $36,700 6.0% 0.4% Licensed Practical and Licensed Vocational Nurses $46,700 8.5% 0.6% All Other Healthcare Practitioners and Technical Occupations(Avg.All Categorie ) $87,000 42.2% 3.1% Weighted Mean Annual Wag $87,000 100.0% 7.4% Healthcare Support Occupations Home Health Aides $22,900 20.7% 0.8% Nursing Assistants $29,100 30.3% 1.2% Dental Assistants $33,300 10.7% 0.4% Medical Assistants $33,700 18.1% 0.7% Phlebotomists $34,200 3.0% 0.1% All Other Healthcare Support Occupations(Avg.All Categories) $29,300 17.2% 0.7% Weighted Mean Annual Wag $29,300 100.0% 3.9% Food Preparation and Serving Related Occupations First-Line Supervisors of Food Preparation and Serving Workers $34,700 7.0% 0.9% Cooks,Fast Food $19,800 4.6% 0.6% Cooks,Restaurant $26,000 8.9% 1.1% Food Preparation Workers $23,300 6.8% 0.9% Bartenders $29,800 5.1% 0.7% Combined Food Preparation and Serving Workers,Including Fast Food $19,700 25.9% 3.3% Counter Attendants,Cafeteria,Food Concession,and Coffee Shop $21,500 3.8% 0.5% Waiters and Waitresses $26,900 20.6% 2.6% Dining Room and Cafeteria Attendants and Bartender Helpers $24,600 3.2% 0.4% Dishwashers $22,900 4.1% 0.5% Hosts and Hostesses,Restaurant,Lounge,and Coffee Shop $23,500 3.1% 0.4% All Other Food Preparation and Serving Related Occupations(Avg.All Categorie) $24,400 7.0% 0.9% Weighted Mean Annual Wage $24,400 100.0% 12.7% Building and Grounds Cleaning and Maintenance Occupations First-Line Supervisors of Housekeeping and Janitorial Workers $40,200 3.4% 0.2% First-Line Supervisors of Landscaping,Lawn Service,and Groundskeeping Wor - $52,200 3.0% 0.2% Janitors and Cleaners,Except Maids and Housekeeping Cleaners $25,300 51.0% 2.9% Maids and Housekeeping Cleaners $32,400 11.3% 0.6% Landscaping and Groundskeeping Workers $28,800 25.5% 1.4% All Other Building and Grounds Cleaning and Maintenance Occupations(Avg.All C $28,500 5.9% 0.3% Weighted Mean Annual Wag- $28,500 100.0% 5.6% Sources:U.S.Bureau of Labor Statistics, IMPLAN Keyser Marston Associates,Inc. Page 78 \\Sf-fs2\wp\14\14100\002\100-150k Honolulu 1-16-15;9/8/2015;dd APPENDIX 2,TABLE 4 AVERAGE ANNUAL WORKER COMPENSATION,2013 SERVICES TO HOUSEHOLDS EARNING$100,000-$150,000 RESIDENTIAL NEXUS ANALYSIS HONOLULU,HI of Total Households Earning of Total $100,000-150,000, 2013 Avg. Occupation Resident Services Occupation 3 Compensation I Group z Workers Page 3 of 4 Personal Care and Service Occupations First-Line Supervisors of Personal Service Workers $42,000 3.7% 0.2% Nonfarm Animal Caretakers $25,100 4.6% 0.2% Amusement and Recreation Attendants $20,800 6.3% 0.3% Hairdressers,Hairstylists,and Cosmetologists $36,400 13.5% 0.7% Childcare Workers $19,100 15.4% 0.8% Personal Care Aides $25,500 25.7% 1.3% Fitness Trainers and Aerobics Instructors $32,000 5.3% 0.3% Recreation Workers $34,400 4.8% 0.3% All Other Personal Care and Service Occupations(Avg.All Categories) $27,400 20.8% 1.1% Weighted Mean Annual Wage $27,400 100.0% 5.2% Sales and Related Occupations First-Line Supervisors of Retail Sales Workers $48,500 9.6% 1.4% Cashiers $22,500 25.6% 3.6% Counter and Rental Clerks $29,100 4.2% 0.6% Retail Salespersons $25,500 36.7% 5.2% Securities,Commodities,and Financial Services Sales Agents $77,000 3.3% 0.5% Sales Representatives,Services,All Other $48,400 3.6% 0.5% Sales Representatives,Wholesale and Manufacturing,Except Technical and Scier $44,600 4.5% 0.6% All Other Sales and Related Occupations(Avg.All Categories) $31,200 12.4% 1.7% Weighted Mean Annual Wage $31,200 100.0% 14.1% Office and Administrative Support Occupations First-Line Supervisors of Office and Administrative Support Workers $51,400 6.7% 1.1% Bookkeeping,Accounting,and Auditing Clerks $37,300 7.1% 1.2% Tellers $26,000 3.1% 0.5% Customer Service Representatives $34,500 10.9% 1.8% Receptionists and Information Clerks $29,000 6.6% 1.1% Stock Clerks and Order Fillers $26,500 10.6% 1.7% Executive Secretaries and Executive Administrative Assistants $50,700 3.2% 0.5% Medical Secretaries $37,200 3.7% 0.6% Secretaries and Administrative Assistants,Except Legal,Medical,and Executive $38,300 9.8% 1.6% Office Clerks,General $31,300 13.0% 2.1% All Other Office and Administrative Support Occupations(Avg.All Categories) $35,100 25.2% 4.1% Weighted Mean Annual Wage $35,100 100.0% 16.2% Installation,Maintenance,and Repair Occupations First-Line Supervisors of Mechanics,Installers,and Repairers $71,900 8.0% 0.3% Telecommunications Equipment Installers and Repairers,Except Line Installers $60,300 3.4% 0.1% Automotive Body and Related Repairers $47,900 4.1°A) 0.1% Automotive Service Technicians and Mechanics $47,200 17.3% 0.6% Bus and Truck Mechanics and Diesel Engine Specialists $63,200 3.6% 0.1% Maintenance and Repair Workers,General $42,400 30.9% 1.1% Installation,Maintenance,and Repair Workers,All Other $54,500 3.1% 0.1% All Other Installation,Maintenance,and Repair Occupations(Avg.All Categories) $49,700 29.5% 1.0% Weighted Mean Annual Wage $49,700 100.0% 3.5% Sources:U.S. Bureau of Labor Statistics,IMPLAN Keyser Marston Associates,Inc. Page 79 \\Sf-fs2\wp\14\14100\002\100-150k Honolulu 1-16-15;9/8/2015;dd APPENDIX 2,TABLE 4 AVERAGE ANNUAL WORKER COMPENSATION,2013 SERVICES TO HOUSEHOLDS EARNING$100,000-$150,000 RESIDENTIAL NEXUS ANALYSIS HONOLULU,HI %of Total Households Earning %of Total $100,000-150,000, 2013 Avg. Occupation Resident Services Occupation' Compensation' Group 2 Workers Page 4 of 4 Transportation and Material Moving Occupations Bus Drivers,School or Special Client $34,500 7.9% 0.4% Driver/Sales Workers $29,400 7.4% 0.4% Heavy and Tractor-Trailer Truck Drivers $44,500 10.9% 0.6% Light Truck or Delivery Services Drivers $30,600 8.6% 0.5% Taxi Drivers and Chauffeurs $26,800 4.4% 0.2% Parking Lot Attendants $21,700 4.3% 0.2% Industrial Truck and Tractor Operators $40,100 3.1% 0.2% Cleaners of Vehicles and Equipment $25,000 5.5% 0.3% Laborers and Freight,Stock,and Material Movers,Hand $30,900 22.4% 1.2% Packers and Packagers, Hand $21,700 7.8% 0.4% All Other Transportation and Material Moving Occupations(Avg.All Categories) $31,300 17.6% 0.9% Weighted Mean Annual Wage $31,300 100.0% 5.4% 86.0% 1 The methodology utilized by the Bureau of Labor Statistics(BLS)generally assumes hourly employees are employed full-time.Annual compensation is calculated by BLS by multiplying hourly wages by 2,080 hours per year(40 hours per week and 52 weeks). 2 Occupation percentages are based on the 2013 National Industry-Specific Occupational Employment survey compiled by the Bureau of Labor Statistics. Wages are based on the 2013 Occupational Employment Survey data specific to Honolulu from the Bureau of Labor statistics. Including occupations representing 3%or more of the major occupation group. Sources:U.S. Bureau of Labor Statistics, IMPLAN Keyser Marston Associates, Inc. Page 80 \\Sf-fs2\wp\14\14100\002\100-150k Honolulu 1-16-15;9/8/2015;dd