Impact analysis methodology

Metric methods

General calculation methods and model inputs used across development impact analyses. Project-specific values and sensitivity ranges are reported on each development page.

Metric	Calculation	Inputs and assumptions
New households Estimates how many of the proposed homes are likely to be occupied and contributing to local activity.	$H = U \times o$ Proposed housing units adjusted for expected occupancy	Proposed housing units Occupancy rate
New residents Translates occupied homes into the number of people expected to live at the development.	$R = H \times h ˉ$ Occupied households multiplied by persons per multifamily unit (bedroom-mix demographic multipliers)	Occupancy rate Multifamily household size
Aggregate household income Approximates the total purchasing power of the development's new households.	$I = H \times M_{tract} \times p_{new}$ Households use the mean income of the site's census tract (aggregate income / households), adjusted for new construction	Tract mean household income New-construction income premium Occupancy rate
New annual spending by category Shows how much new household spending may be available for groceries, dining, retail, services, and entertainment each year.	$S_{c} = I \times s_{c}^{CES} \times k_{CES}$ Consumer Expenditure Survey category spending per household, income-scaled with per-category Engel elasticities (necessities grow slower than income)	CES category spending share Category income elasticity CES scaling factor New-construction income premium
Annual capture at a destination Estimates how much of the new spending individual businesses or commercial areas are likely to receive.	$P_{j, c}^{(m)} C_{j} = \frac{A _{j, c}^{α} ω _{c, m} e ^{- β_{m} t_{j, m}}}{\sum _{k} \sum _{r} A _{k, c}^{α} ω _{c, r} e ^{- β_{r} t_{k, r}}} = c \sum S_{c} m \sum P_{j, c}^{(m)}$ Joint walk/drive Huff destination choice using CES category spending; named clusters are reporting rollups and on-site retail competes as a destination	Destination attractiveness Walk impedance decay Mode share by spending category CES scaling factor
In-city capture share Shows what portion of new spending is expected to remain inside the city instead of flowing elsewhere.	$P_{j, c}^{(m)} C_{j} q_{city} = \frac{A _{j, c}^{α} ω _{c, m} e ^{- β_{m} t_{j, m}}}{\sum _{k} \sum _{r} A _{k, c}^{α} ω _{c, r} e ^{- β_{r} t_{k, r}}} = c \sum S_{c} m \sum P_{j, c}^{(m)} = \frac{\sum _{j \in city} C _{j}}{\sum _{j} C _{j}}$ Joint walk/drive Huff destination choice using CES category spending; share of capture assigned to destinations inside the city boundary	City boundary Destination attractiveness Walk impedance decay Mode share by spending category
On-site jobs removed (existing space) Estimates the jobs associated with existing commercial space that would be displaced by redevelopment.	$J_{removed} = \frac{F _{existing}}{f _{office/job}}$ Existing commercial floor area converted to an estimated job count	Existing commercial floor area Square feet per office job
On-site retail jobs added Estimates the permanent retail jobs supported by the project's proposed commercial space.	$J_{added} = \frac{F _{retail}}{f _{retail/job}}$ Proposed retail floor area converted to an estimated job count	Proposed retail floor area Square feet per retail job
On-site office jobs added Estimates the permanent office jobs supported by the project's proposed non-retail commercial space.	$J_{office} = \frac{F _{office}}{f _{office/job}}$ Proposed office floor area converted to an estimated job count	Proposed office floor area Square feet per office job
Net on-site job change Shows whether the completed project is expected to support more or fewer on-site jobs than the current use.	$Δ J = J_{added} + J_{office} - J_{removed}$ New on-site retail and office jobs less jobs associated with removed commercial space	Square feet per office job Square feet per retail job
New annual spending arriving on foot Estimates the spending that nearby businesses may receive from residents who reach them by walking.	$P_{j, c}^{(walk)} W = \frac{A _{j, c}^{α} ω _{c, walk} e ^{- β_{walk} t_{j, walk}}}{\sum _{k} \sum _{r} A _{k, c}^{α} ω _{c, r} e ^{- β_{r} t_{k, r}}} = j \sum c \sum S_{c} P_{j, c}^{(walk)}$ Walk-mode share of the joint destination and mode Huff choice, summed across destinations	Mode share by spending category CES scaling factor Walk impedance decay
Spending arriving on foot at project retail Shows how much resident spending may reach the development's own shops and restaurants on foot.	$P_{j, c}^{(walk)} W_{own} = \frac{A _{j, c}^{α} ω _{c, walk} e ^{- β_{walk} t_{j, walk}}}{\sum _{k} \sum _{r} A _{k, c}^{α} ω _{c, r} e ^{- β_{r} t_{k, r}}} = c \sum S_{c} P_{own, c}^{(walk)}$ Walk-mode share of the joint Huff choice assigned to the on-site retail destination	On-site retail attractiveness Walk impedance decay Neighborhood walk share
Implied spending per resident walk trip Provides a reasonableness check by comparing walk-based spending with the total number of expected walking trips.	$w ˉ_{trip} = \frac{W}{365 T _{walk/day}}$ Walk-arriving spending spread across all estimated resident walking trips	Walk trips per resident per day Mode share by spending category CES scaling factor
New resident walk trips per day Estimates the additional daily foot traffic generated by people living at the development.	$T_{walk/day} = R \times r_{walk}$ New residents multiplied by daily walking trips per resident	Walk trips per resident per day Multifamily household size Occupancy rate
Foot-traffic index change, nearest commercial street segments Highlights the nearby commercial streets most likely to experience a noticeable increase in pedestrian activity.	$w_{ij}^{walk} Δ_{e} Δ_{10} = e^{- β_{walk} t_{ij}} = 100 \frac{F _{e}^{new}}{F _{e}^{baseline}} = \frac{1}{10} e \in N_{10} \sum Δ_{e}$ Exact marginal trip flows from the site using POI-weighted destinations and shortest paths, compared with seeded sampled baseline betweenness for the current population	Occupancy rate Multifamily household size Walk trips per resident per day Walk impedance decay
Current real estate tax (site) Establishes the site's existing annual property-tax contribution as the baseline for comparison.	$T_{RE, 0} = V_{0} \times \frac{r _{RE}}{100}$ Current assessed value multiplied by the real estate tax rate	Current assessed value Real estate tax rate
Current value per acre Shows how intensively the site currently contributes taxable value relative to its land area.	$V_{0, acre} = \frac{V _{0}}{A}$ Current assessed value divided by site area	Current assessed value Site area
Projected assessed value Estimates the taxable real estate value of the completed development, across housing, ground-floor retail, and office space. The value per dwelling unit is drawn from assessment comparables of the same product class — for-sale condominiums are valued against condominium comparables, not apartment buildings.	$V_{1} = U v_{unit} + F_{retail} v_{sf} + F_{office} v_{office}$ Housing units multiplied by assessed value per unit, plus retail floor area multiplied by retail value per square foot, plus office floor area multiplied by office value per square foot	Proposed housing units Assessed value per dwelling unit Proposed retail floor area Retail value per sq ft Proposed office floor area Office value per sq ft
Projected real estate tax Estimates the annual property-tax revenue the completed development could generate.	$T_{RE, 1} = V_{1} \times \frac{r _{RE}}{100}$ Projected assessed value multiplied by the real estate tax rate	Projected assessed value Real estate tax rate
Real estate tax increase Shows the expected gain in annual property-tax revenue compared with the site today.	$Δ T_{RE} = T_{RE, 1} - T_{RE, 0}$ Projected real estate tax less current real estate tax	Projected real estate tax Current real estate tax
Projected value per acre Measures how much taxable value the redevelopment could produce from each acre of land.	$V_{1, acre} = \frac{V _{1}}{A}$ Projected assessed value divided by site area	Projected assessed value Site area
Personal property tax on resident vehicles (rough estimate) Rough estimate of annual vehicle (personal property) tax from new households. The city's rate is real; the vehicle counts and values are assumptions, so treat this as order-of-magnitude.	$T_{pp} \approx H \times ν_{hh} \times v ˉ_{veh} \times \frac{r _{pp}}{100}$ New households multiplied by assumed vehicles per household and assumed value per vehicle, taxed at the city's personal property rate	Vehicles per household Average vehicle assessed value Personal property tax rate Occupancy rate
BPOL business license tax on project retail (rough estimate) Rough estimate of annual business-license (BPOL) tax from the project's own shops and restaurants, counting only the share of sales that is new to the city. Sales captured from existing city businesses stop being taxed at their old location, so they move the tax base rather than adding to it. The city's rate schedule is real; tenant sales are assumed from typical sales per square foot.	$B_{onsite} T_{bpol} = F_{retail} g_{sf} - C_{own} \approx B_{onsite} \times \frac{r _{bpol}}{100} \times η$ Proposed retail floor area multiplied by assumed gross sales per square foot, less resident spending already counted elsewhere, taxed at the city's BPOL retail rate and scaled by the net-new share	Proposed retail floor area Retail sales per square foot BPOL retail rate Share of on-site sales that is new to the city
BPOL business license tax on project office (rough estimate) Rough estimate of annual business-license tax from the project's office tenants, at the city's financial and professional services rate. Not reduced for displacement: professional and medical practices largely serve regional demand rather than recirculating local spending.	$T_{bpol, off} \approx F_{office} \times g_{office} \times \frac{r _{bpol, off}}{100}$ Proposed office floor area multiplied by assumed gross receipts per square foot, taxed at the city's professional-services BPOL rate	Proposed office floor area Office gross receipts per sq ft BPOL professional rate
Business tangible property tax (rough estimate) Rough estimate of annual tax on business furniture, fixtures, and equipment in the project's commercial space. Not reduced for displacement, since the equipment is physically new to the city — though a relocating tenant brings existing equipment with it.	$T_{bpp} \approx (F_{retail} + F_{office}) \times v_{bpp} \times \frac{r _{bpp}}{100}$ Proposed commercial floor area multiplied by assumed equipment value per square foot, taxed at the business tangible property rate	Proposed commercial floor area Business equipment value per sq ft Business tangible property rate
Meals tax on the project's own restaurants (net-new, rough estimate) Rough estimate of meals-tax revenue from restaurants in the development, counting only the share of sales that is new to the city rather than captured from existing city restaurants. Applicant analyses typically count these sales in full, which overstates the city's gain.	$B_{onsite} T_{meals, own} = F_{retail} g_{sf} - C_{own} \approx B_{onsite} ϕ_{rest} r_{meals} η$ Ground-floor sales less resident spending already counted elsewhere, multiplied by the assumed restaurant share, the meals tax rate, and the net-new share	Proposed retail floor area Retail sales per square foot Restaurant share of ground-floor space Meals tax rate Share of on-site sales that is new to the city
Local sales tax on the project's own retail (net-new, rough estimate) Rough estimate of the city's share of sales tax on the development's own retail sales, counting only the share that is new to the city.	$B_{onsite} T_{sales, own} = F_{retail} g_{sf} - C_{own} \approx B_{onsite} r_{local} η$ Ground-floor sales less resident spending already counted elsewhere, multiplied by the local sales tax share and the net-new share	Proposed retail floor area Retail sales per square foot Local sales tax share Share of on-site sales that is new to the city
Meals tax on captured in-city dining Estimates city meals-tax revenue from new resident spending at restaurants inside the city.	$T_{meals} = S_{restaurant} q_{city, food} r_{meals}$ Restaurant spending captured inside the city multiplied by the meals tax rate	Restaurant spending In-city dining capture share Meals tax rate
Local sales tax share on captured in-city retail Estimates the city's share of sales-tax revenue from new retail spending that remains in the city.	$T_{sales} = S_{retail} q_{city, retail} r_{local}$ Retail spending captured inside the city multiplied by the local sales tax share	Retail spending In-city retail capture share Local sales tax share
Estimated K-12 students Estimates how many public-school students the development is likely to add — the driver of the school-cost component.	$S_{K12} = U \times s_{unit}$ Proposed housing units multiplied by a student generation rate for high-rise multifamily	Proposed housing units Students per unit
Annual service cost - naive per-capita method Conservative estimate that assigns each new resident the city's average non-school service cost, plus per-pupil tuition for the students the project actually generates.	$C_{naive} = R \times c_{non - school} + S_{K12} \times c_{pupil}$ New residents multiplied by non-school general-fund cost per resident, plus estimated students multiplied by the net local cost per pupil (tuition contract minus state education aid)	New residents Non-school expenditure per resident Students per unit Net local cost per pupil
Annual service cost - marginal framing Estimates the added public-service costs most likely to grow because of the development. School costs follow the student estimate in both framings.	$C_{marginal} = R \times c_{non - school} \times k_{marginal} + S_{K12} \times c_{pupil}$ Non-school per-capita cost scaled by the marginal factor, plus estimated students multiplied by the net local cost per pupil (tuition contract minus state education aid)	Non-school per-capita service cost Marginal cost factor Students per unit Net local cost per pupil
Net annual fiscal impact - naive per-capita method Tests whether new recurring revenue covers costs under the more conservative per-resident approach. Revenue includes the rough-estimate personal property and BPOL lines.	$N_{naive} = r \sum T_{r} - C_{naive}$ New recurring revenue (including rough-estimate personal property and BPOL lines) less the per-capita service-cost estimate	New recurring revenue Naive per-capita service cost
Net annual fiscal impact - marginal framing Tests whether new recurring revenue covers the public costs most likely to increase in practice. Revenue includes the rough-estimate personal property and BPOL lines.	$N_{marginal} = r \sum T_{r} - C_{marginal}$ New recurring revenue (including rough-estimate personal property and BPOL lines) less the marginal service-cost estimate	New recurring revenue Marginal service cost
Net annual fiscal impact range Presents a practical range of possible yearly gains or losses rather than relying on one cost estimate. Revenue includes the rough-estimate personal property and BPOL lines.	$N \in [r \sum T_{r} - C_{naive}, r \sum T_{r} - C_{marginal}]$ Range bounded by the naive per-capita and marginal service-cost estimates	New recurring revenue Naive per-capita service cost Marginal service cost
External estimate — net annual fiscal impact Reports the fiscal estimate the applicant or city staff published for this project, so it can be compared against this analysis. These figures are reproduced as published and are never averaged into, or substituted for, the estimates above. Where the ranges disagree, the report says so and names the methodological differences.	Reported by the applicant's fiscal impact analysis or the city staff report; not computed by this pipeline
New annual spending arriving by bike Estimates the spending nearby businesses may receive from residents who reach them by bicycle (developments only; bike competes with walking and driving as a third mode).	$S_{bike} = c \sum j \sum S_{c} P_{bike} (j ∣ c)$ Bike-mode share of the joint destination and mode Huff choice, summed across destinations; the bike preference is carved from the drive remainder so walk figures keep their meaning	Bike share by spending category Bike impedance decay CES scaling factor
Bike catchment population (decay-weighted) Estimates how many residents can practically reach the corridor by bike, weighting nearby homes more than distant ones.	$P_{eff} = b \sum pop_{b} e^{- β_{bike} t_{b}}$ Census-block population weighted by exponential decay in bike travel time to the corridor over the OpenStreetMap bike network	Bike impedance decay
Induced bike visits per day (corridor) Estimates daily bike visits to corridor businesses that the new facility itself brings about.	$V = P_{eff} \times τ_{bike} \times σ_{induced}$ Decay-weighted catchment population multiplied by latent bike trips per resident and the induced corridor visit share (bounds calibrated to corridor before/after studies)	Bike trips per resident per day Induced corridor visit share Bike impedance decay
New annual spending at corridor businesses Estimates the yearly spending corridor businesses may gain from new bike visits after the facility is built.	$S_{corridor} = 365 g \sum V m_{g} s ˉ_{g}$ Induced daily bike visits split across the corridor's business mix and priced with surveyed cyclist spending per trip (Portland intercept surveys)	Bike trips per resident per day Induced corridor visit share Cyclist spending per trip
Implied corridor sales uplift (diagnostic) Reasonableness check comparing the model's new corridor spending against rough existing sales, flagged when it exceeds what corridor studies have ever observed.	$u = \frac{S _{corridor}}{n _{biz} \times F ˉ \times g _{sf}}$ New corridor spending divided by a rough baseline (business count at typical suite size and sales per square foot); compared with the 0-50% envelope from corridor natural experiments	Cyclist spending per trip Induced corridor visit share
Trail catchment population (decay-weighted) Estimates how many residents live within practical reach of the trail's access points.	$P_{eff} = b \sum pop_{b} e^{- β_{access} d_{b}}$ Census-block population weighted by exponential distance decay to the nearest trail access point over the walking network (best-powered curve in the trail-access literature)	Trail-access distance decay
Annual trail user-days Estimates how many visits the trail is likely to generate in a year from its catchment.	$D = P_{eff} \times u_{capita}$ Decay-weighted catchment population multiplied by annual user-days per catchment resident (North Carolina four-trail count program)	Trail-access distance decay Trail user-days per resident
Annual trail-user spending at nearby businesses Estimates the yearly spending trail users may leave at businesses near the trail.	$S_{trail} = D \times s ˉ_{day}$ Annual user-days multiplied by surveyed direct spending per user-day (ordinary urban greenways; destination-trail tourism excluded), allocated to businesses by walking distance from the access points	Trail user-days per resident Trail spending per user-day Walk impedance decay
Assessed value within the trail premium band Totals the existing taxable property value close enough to the trail to plausibly capitalize a premium.	$V_{band} = p \in band \sum AV_{p}$ Sum of assessed values for parcels within roughly a third of a mile of the trail line	Premium band width
Trail property value uplift (capitalization) Estimates the one-time increase in nearby home values that a well-used trail can capitalize, with a zero floor because low-profile trails often show none.	$Δ V = V_{band} \times π_{trail}$ Assessed value in the premium band multiplied by the trail property premium (hedonic literature range)	Trail property premium
Annual real estate tax increment (trail premium) Estimates the recurring property-tax revenue if the trail premium capitalizes into assessments.	$Δ T_{RE} = Δ V \times \frac{r _{RE}}{100}$ Property value uplift multiplied by the city's real estate tax rate	Trail property premium Real estate tax rate