SAN FRANCISCO in 2040

Summary of Building Plan

286,150 new housing units across 30 designated growth zones, adding approximately 515,000 residents to San Francisco by 2040. Funded by $22 billion in infrastructure: $5.5B GO bonds, $5.9B developer agreements, $10.6B Mello-Roos districts. Per-unit development cost cut 49% through entitlement reform. Developers build and pay for everything on their own sites.

Document Outline

1. Executive Summary

• 286,150 new units, 515,000 new residents, 30 growth zones, 2027-2040
• Per-unit cost cut 49% ($989K → $502K) through radical entitlement reform
• $22B infrastructure via hybrid financing: $5.5B GO bonds + $5.9B developer agreements + $10.6B Mello-Roos

2. A Picture of the Future

• SF current density vs. global cities: 6,800/km² today (London, Amsterdam) → 10,000/km² (NYC, Lyon, Montréal)
• What 2040 looks like: 22,600 new businesses, 181,000 jobs, 162 schools, $2,500 studio rents
• Unchanged neighborhoods: Sunset, Richmond, Noe Valley, Pacific Heights, the Haight

3. Where We Build: 30 Growth Zones

• Underbuilt transit corridors only: Mission, Geary, 19th Ave, India Basin, East SoMa, and 7 more
• Zone-by-zone unit targets, max heights, and delivery notes
• No changes to any low-density residential neighborhood

4. How We Build: Phased Delivery

• Four milestones: 17K (2030), 72K (2033), 166K (2037), 213K (2040)
• Fast-start zones first (India Basin, East SoMa); politically complex zones last (Ocean Beach, Wharf)
• Bond pause triggered if delivery falls >15% behind milestone

5. Entitlement Reform

• Impact fees: $101K → $15K. Inclusionary suspended — developers compensate nearby residents directly
• Prevailing wage and local hire waived for private construction; retained for public infrastructure
• Parking minimums eliminated. Flat $2K permit fee. 200 sq ft minimum unit size
• All exemptions expire at 213K units or Dec 31, 2042

6. Infrastructure & Financing

• $22B across 11 categories: schools ($7.9B), stormwater ($3.1B), water/sewer ($2.9B), power ($1.8B), and 7 more
• Two approaches: full $22B GO bond (7.1yr payback) vs. hybrid $5.5B GO bond (1.8yr payback)
• Infrastructure detail for all 11 line items
• Why direct payments are affordable housing: older buildings are naturally affordable; new BMR construction wastes hundreds of thousands per unit through layers of middlemen

7. How It Pays for Itself

• $3.1B/year recurring at build-out: property tax, sales tax, payroll tax, GRT, utilities
• $22.7B cumulative through 2040 vs. $22B infrastructure cost
• Every assumption sourced: SF Assessor, BTRC, BLS, DBI schedules, SFPUC rates

8. The Referendum

• Voter initiative locks in framework for 13 years — Board cannot repeal
• Discretionary review → ministerial. Project EIRs → programmatic EIR. Fragmented zoning → codified height/density
• Cannot fix: CEQA litigation, Coastal Commission (Ocean Beach), federal cleanup (Hunters Point)
• Full list of provisions, key terms, and what the measure does and does not do

9. Next Steps: Referendum Calendar

• Draft & file ballot language Mar-May 2026. Signature gathering Jun-Aug. Vote Nov 2026
• First bond tranche and Mello-Roos formation Q1 2027. Construction starts Jan 2028
• First housing delivered Q4 2029 in fast-track zones

10. Addendum: Full Referendum Text

• 9 sections of proposed ballot language: findings, financing, infrastructure, growth areas, permitting, entitlement reform, accountability, fiscal, effective date
• Ballot question names every entitlement suspension and the direct resident compensation requirement
• Subject to City Attorney review

11. Addressing the Opposition

• "Too expensive": $22.7B revenue vs $22B cost; hybrid needs only $5.5B voter approval
• "What about existing residents": developers pay neighbors directly — no fund, no nonprofits, no middlemen
• "Protecting a broken system": $1.1B/year in housing nonprofits with documented waste; this plan sends money to people, not institutions

1. Executive Summary

Over the last two decades, San Francisco has created extraordinary abundance. It has become the most important city in the world for the world's most important industry. The companies, products, and ideas that define modern life were built here. No other city on earth can claim that.

And yet San Francisco has lacked the ambition to grow alongside this creation. The city that invented the future has refused to build for it. Housing costs have spiraled. Teachers, nurses, firefighters, and the workers who keep the city running have been priced out. Families leave. The state has mandated 82,000 new homes by 2031, and the city is not on track.

San Francisco should inspire the world not only with its entrepreneurship, but with its quality of life. It should demonstrate that the city where the future is invented is also the city where the future is lived: a place that grows, that welcomes, that builds the schools and parks and transit that a great city requires, and that maintains its leadership as the global technology hub by being a place people actually want to live.

This is what that looks like.

The Plan

This proposal is a 13-year plan to build approximately 286,000 new homes across 30 designated growth zones, adding roughly 515,000 new residents to San Francisco by 2040. Infrastructure is funded by $22 billion through GO bonds ($5.5B), developer agreements ($5.9B), and Mello-Roos districts ($10.6B). Developers build and pay for everything on their own sites.

Radical entitlement reform within the growth zones cuts per-unit development cost from $989,000 to $502,000 — a 49% reduction. Impact fees drop from $101,069 to $15,000. Inclusionary mandates are suspended and replaced by a simple requirement: developers compensate existing residents near construction sites directly. No fund, no nonprofit intermediaries, no city administration. Prevailing wage, local hire, and parking minimums are suspended for private construction. All exemptions expire at 213,000 units or 2042. Public infrastructure retains all existing labor standards.

The math works. At full build-out, the new residents and businesses generate an estimated $3.1 billion per year in recurring tax revenue. Add one-time construction-phase revenues and cumulative revenue through 2040 reaches $22.7 billion against a $22 billion infrastructure cost. Under the hybrid financing approach, the GO bond pays back in 1.8 years.

Where, Not Everywhere

This is not a plan to turn quiet residential streets into construction zones. It is a plan to build on the underbuilt core arteries of the city: corridors that already have transit, commercial activity, and wide streets but are lined with single-story retail, surface parking lots, and underused industrial land.

The Mission has two BART stations and miles of corridor zoned for three stories. Geary Boulevard is the longest commercial street in San Francisco, served by the most-ridden bus line, and most of it is one story tall. 19th Avenue is a major arterial lined with gas stations and strip malls, slated for BRT. India Basin is largely vacant.

The Sunset, the Richmond, Noe Valley, Pacific Heights, the Haight, and every other low-density residential neighborhood are not part of this plan. Nothing in this proposal changes the character of those areas.

What 2040 Feels Like

A city of 1.34 million is a fundamentally different place. Density comparable to New York City, Lyon, and Montréal — still well below Barcelona or Paris. Every growth zone corridor transforms from single-story vacancy into walkable, active streetscape. Families stay. Commutes shorten. Housing costs fall because supply meets demand.

2. A Picture of the Future

SF Current Density

Density is often misunderstood. When people hear "denser city," they picture Hong Kong or Manhattan. But San Francisco is already a moderately dense city. The question is whether it should remain at a density that does not support enough housing for the people who want to live here.

Cities Like SF Today (~5,500-8,500 per sq km)

Cities Like SF Post-2040 (~8,000-12,000 per sq km)

At 10,012 per square kilometer, SF 2040 sits right alongside New York City, Lyon, and Montréal. Not Manhattan. NYC averaged across all five boroughs.

3. Where We Build: 30 Growth Zones

The plan identifies 30 growth zones. These are underbuilt commercial corridors, industrial areas, and transit-adjacent sites.

Table 1: Complete Build Plan. Population at 1.8 persons per unit.

Zone Profiles

4. How We Build: Phased Delivery 2027-2040

Construction is phased over 13 years with four milestones. Shovel-ready zones start first; zones requiring remediation or new transit ramp up later.

5. Entitlement Reform: Cutting the Cost Per Unit in Half

Removing regulatory barriers without addressing cost barriers produces expensive housing faster. The average all-in development cost per unit in San Francisco today exceeds $900,000. Roughly half of that premium comes not from land or materials but from city-imposed requirements: impact fees, inclusionary mandates, prevailing wage rules, parking minimums, and design review processes that add months of carrying costs. These requirements apply uniformly across the city regardless of context, scale, or public benefit.

The referendum suspends all of them within the 12 growth zones for the duration of the build-out. This is not permanent deregulation — it is a project-specific exemption scoped to 213,000 units over 13 years, after which standard city requirements resume. The goal is to cut the per-unit cost low enough that middle-income households can afford the housing without subsidies.

Before and After: Per-Unit Development Cost

What Each Reform Does

Impact Fee Reduction: $101,069 → $15,000

San Francisco's impact fees are the highest in the country. They fund schools, parks, transit, childcare, and affordable housing. For this project, the infrastructure bond directly funds schools, parks, transit, and fire/police — making most of the fees redundant. The $15,000 retained fee covers sewer connection and a minimal school assessment. Affordable housing funding is replaced by a direct program (see below).

Inclusionary Housing: 20% BMR → Direct Developer Payments

The current inclusionary ordinance requires 15-20.5% of units to be rented or sold below market rate. This is effectively a cross-subsidy — every market-rate buyer pays an additional ~$140,000 to cover the cost of the BMR units. The referendum suspends this requirement and replaces it with a single obligation: developers must compensate existing residents near their construction sites directly. No city fund, no nonprofit intermediaries, no administrative apparatus.

Prevailing Wage: Required → Exempt

Prevailing wage adds approximately 12-15% to construction labor costs in San Francisco, where market wages are already among the highest in the country. The exemption applies only to private residential construction in the 12 growth zones — not to the public infrastructure funded by the bond, which retains prevailing wage. This is the most politically difficult reform in the package. The building trades will oppose it. The tradeoff is ~$84,000 per unit in savings and tens of thousands of additional construction jobs that would not exist at the higher cost structure.

Parking Minimums: 0.5-1 space → 0

A structured parking space costs $60,000-$120,000 to build. Requiring parking in transit-served growth zones is a direct subsidy from non-car-owners to car-owners. Zero parking minimums do not mean zero parking — developers will build parking where the market demands it. They just won't be forced to build it where it isn't needed, particularly in zones with BART, Muni Metro, and BRT access.

Local Hire: 30% → Waived

The local hire mandate requires 30% of construction labor hours to be performed by San Francisco residents. At the scale of this project — 213,000 units over 13 years — there are not enough SF-resident construction workers to meet 30% of demand. The mandate creates compliance costs, limits the labor pool, and slows production. Waiving it for the growth zones allows the project to draw on the full Bay Area labor market.

6. Infrastructure & Financing: $22 Billion

The entitlement reform extends to public works. Streamlining public works permitting within the growth zones and competitive bidding without restrictive labor carve-outs reduces the infrastructure budget by 22% — from $28.3 billion to $22 billion. The same infrastructure gets built. It costs less.

Two Approaches to Financing

The $22 billion can be funded entirely through general obligation bonds, or through a hybrid structure that shifts most of the cost off the city's balance sheet. Both approaches deliver the same infrastructure on the same timeline. The difference is political: how large a bond measure voters need to approve.

Approach A: Full GO Bond — $22 Billion

The city issues $22 billion in general obligation bonds over 13 years, secured by property tax revenue. Simplest structure — one ballot measure, one funding source, one administrative framework. Requires a 2/3 supermajority vote, and $22 billion would be the largest municipal bond in American history. At $3.1 billion/year in recurring revenue, the bonds pay back in 7.1 years with a cumulative surplus of $0.7 billion through 2040.

Approach B: Hybrid Financing — $5.5 Billion GO Bond

Developer Agreements are binding contracts where the developer builds zone-specific infrastructure — trunk sewer lines, power connections, road improvements — as a condition of approval. San Francisco already uses these for large projects: Parkmerced committed ~$1 billion. Scaling across 12 zones shifts $5.9 billion to the developers who benefit from it.

Mello-Roos / Community Facilities Districts are special tax districts where new residents pay an additional property tax assessment funding infrastructure bonds. This is how most large California developments have been financed for 40 years — Irvine, Sacramento, San Jose. San Francisco has barely used this tool. Mello-Roos districts in each growth zone fund schools, parks, fire stations, and police at $10.6 billion with zero city cash.

The hybrid approach reduces the voter-approved bond from $22 billion to $5.5 billion. At $3.1 billion/year in recurring city revenue, the $5.5 billion bond pays back in 1.8 years. Cumulative city revenue surplus through 2040 is $17.2 billion. The tradeoff: Mello-Roos assessments add approximately $200-$400/month to each new household's costs, partially offsetting the savings from entitlement reform.

Infrastructure Detail

K-12 Schools — $7.9 billion

Schools are the single largest line item because they are the single most important determinant of whether families stay in San Francisco. At $49 million average construction cost per school (prevailing wage exempt), the budget builds approximately 162 new schools across the 12 growth zones. SFUSD enrollment has been declining for over a decade. This reverses it.

Stormwater & Sea Level Rise — $3.1 billion

San Francisco's combined sewer system — which mixes stormwater and wastewater in a single pipe — cannot handle the additional volume from 213,000 new units without separation in the growth zones. This budget also covers waterfront barriers and pump stations for zones exposed to sea level rise, particularly India Basin, Mission Bay, and the Southern Waterfront.

Water & Sewer Trunk Lines — $2.9 billion

New trunk mains connecting growth zones to the city's water and sewer backbone. The existing system was sized for a city of 800,000. Expanding to 1.2 million requires new mains under the major corridors.

Power Grid Substations — $1.8 billion

PG&E's distribution grid does not have capacity for 213,000 new electrified units. New substations, feeder lines, and transformer capacity in each growth zone.

Wastewater Treatment — $1.7 billion

Upgrades to the Southeast Treatment Plant and Oceanside Water Pollution Control Plant to handle increased flow from 383,000 additional residents.

Parks & Open Space — $1.4 billion

San Francisco currently has approximately 3.4 acres of parkland per 1,000 residents. Maintaining that ratio for 383,000 new residents requires approximately 1,300 acres of new park space across the growth zones.

Road Resurfacing & Traffic Signals — $1.2 billion

No new highways. No new freeways. No road widening. This budget covers resurfacing of existing streets in and around the growth zones, new traffic signals, pedestrian improvements, and intersection redesigns for increased foot traffic.

Fire Stations — $840 million (22 new)

SFFD's response time standard is 4 minutes or less. Meeting that in 12 new density zones requires 22 new stations at approximately $27 million each, plus apparatus.

Bus Route Expansion — $560 million

New Muni routes serving India Basin, the Southern Waterfront, and the Outer Mission. Increased frequency on the 38-Geary, 19th Avenue BRT, and Caltrain corridor connections.

Police Facilities — $450 million

District stations and equipment to serve 12 new density zones. Staffing is operational budget, not capital.

911 / Emergency Dispatch — $170 million

Call center expansion and system upgrades to handle increased call volume from 383,000 additional residents.

Why Direct Payments Are Affordable Housing

Building new affordable housing makes no sense. New construction is the most expensive housing in any market — $600,000 to $1,000,000 per unit in San Francisco. Taking that unit and renting it at $1,200/month to one lottery winner requires a subsidy of $400,000-$700,000, routed through nonprofit developers, consultants, attorneys, and city agencies. The current system delivers a small number of expensive new units to a small number of people while consuming enormous public resources.

Older buildings are naturally affordable. A 40-year-old apartment with no construction debt, a locked Prop 13 tax base, and stabilized operating costs charges lower rent because the capital costs are already paid. This is how affordable housing works in every city in the world: people live in older buildings at lower rents. The filtering process — yesterday's luxury housing becomes today's middle-income housing and tomorrow's affordable housing — is the only mechanism that has ever produced affordable housing at scale.

Direct developer payments to nearby residents make existing older buildings more affordable while encouraging construction of new units. A renter in a 1970s building near the Geary Corridor paying $2,400/month who receives $300/month from the developer next door has an effective rent of $2,100 — affordable to a household earning $84,000. This reaches tens of thousands of households instead of the ~1,500 lottery winners per year the inclusionary system produces. And it eliminates every middleman, every nonprofit fee, every consultant, every compliance bureaucracy. Developer pays resident. That's it.

San Francisco spends over $1.1 billion per year on homelessness and housing nonprofits. The affordable housing pipeline produces roughly 1,000-1,500 units per year at average costs exceeding $700,000 per unit. Multiple Controller audits have documented waste, cost overruns, and lack of measurable outcomes. This plan eliminates every organization that profits from the housing shortage and replaces them with nothing. The developer writes checks to the neighbors. The money goes to people, not institutions.

7. How It Pays for Itself

All rates, fees, and formulas sourced from current SF tax code, the 2026 Impact Fee Register, and state law.

Property Tax Assumptions

Transfer Tax Tiers (SF Business & Tax Code Art 12C)

Development Impact Fees (per unit, reformed)

Business & Consumer Tax Assumptions

Construction-Phase Assumptions (reformed)

Revenue Projections by Milestone

Data Sources

• SF Citywide Development Impact Fee Register (eff Jan 1, 2026) - sfplanning.org
• SF Business & Tax Regulations Code Art 12C - Transfer Tax tiers
• Prop M (Nov 2024) - GRT restructuring, 7 categories, $5M threshold
• SF Controller FY25-26/26-27 Revenue Letter
• SFPUC Rate Schedules (eff 7/1/2025)
• CA Education Code 17620
• DBI Development Impact Fee Collection Unit

8. The Referendum

Why a Referendum

San Francisco's housing crisis is fundamentally a regulatory crisis. The city has the land, the demand, and the capital. What it lacks is a system that allows housing to be built at the speed and scale the market would otherwise support. A voter-approved initiative is the strongest possible vehicle for fixing this because it creates law that the Board of Supervisors cannot undo without another public vote. Ordinances passed by the Board can be repealed by the Board. A referendum cannot.

This matters because the political dynamics that created the housing shortage are still in place. The Board has repeatedly downzoned land, added discretionary review layers, and expanded opportunities for project opponents to delay or kill housing. A referendum locks in the regulatory framework for 13 years and removes the ability of future Boards to backslide.

What the Referendum Fixes

Discretionary Review → Ministerial Approval

The single biggest obstacle to housing in San Francisco is discretionary review. Every project, no matter how code-compliant, goes through a multi-year approval process where neighbors, the Planning Commission, and the Board of Supervisors can block, shrink, or delay it. A single appeal can add 18 months. This referendum mandates ministerial (by-right) approval for any project in the 12 growth zones that meets the building code and the height/density limits established by the plan. If it's code-compliant, it gets a permit. No hearings, no appeals, no commission votes. SB 35, signed into state law in 2017, already established the legal precedent that ministerial approval for housing is constitutional in California.

Project-by-Project EIR → Programmatic EIR

Under current practice, every significant housing project requires its own Environmental Impact Report under CEQA — a process that takes 18-36 months and costs $1-5 million per project. This referendum funds and mandates a single Programmatic EIR covering all 12 growth zones, completed within 18 months of the effective date. Individual projects that fall within the scope of the programmatic EIR do not need separate environmental review. Multiple California cities have used programmatic EIRs successfully, including the Downtown Plan EIR that San Francisco itself adopted in 2023. This is proven, legally tested, and the single most effective way to collapse the permitting timeline from years to months.

Infrastructure Chicken-and-Egg → Pre-Funded Backbone

Developers will not build where there is no sewer capacity, and the city will not build sewer capacity until developers commit. This deadlock has stalled growth in every potential expansion area for decades. The $22 billion infrastructure financing — GO bonds for citywide systems, developer agreements for zone-specific connections, Mello-Roos for schools and parks — breaks the deadlock by pre-funding all backbone infrastructure so that developers can build with confidence that the public systems will be in place when residents arrive.

Fragmented Zoning → Codified Height and Density

San Francisco's current zoning is a patchwork of overlapping districts, special use districts, and design guidelines that make it nearly impossible for a developer to know what they can build on a given parcel without months of pre-application review. The referendum establishes clear, codified height and density limits for each of the 12 zones. A developer can look at a parcel, check the zone map, and know exactly what is allowed. This certainty is itself a form of infrastructure — it reduces the cost and risk of development, which translates directly into lower housing costs.

What the Referendum Cannot Fix

CEQA Litigation

A programmatic EIR reduces exposure but does not eliminate it. Project opponents can still file CEQA lawsuits on project-specific grounds — inadequate traffic analysis, shadow impacts, historic resources, or claims that a project falls outside the scope of the programmatic EIR. Each lawsuit adds 1-2 years even if it ultimately fails. At this scale, expect dozens of lawsuits across the 12 zones. The plan accounts for this by building legal defense costs into the infrastructure budget and by setting delivery milestones that assume some litigation delay. But no local ballot measure can reform CEQA itself — that requires state legislation.

Coastal Commission Jurisdiction

The Ocean Beach / Great Highway zone falls under California Coastal Commission jurisdiction. A local ballot measure cannot override state coastal authority. The 10,000 units proposed for that zone require Coastal Commission approval through a Local Coastal Program amendment, which is a separate multi-year process with its own environmental review and public hearings. This zone is the most legally fragile in the plan and the most likely to underdeliver or experience significant delays.

Federal Jurisdiction at Hunters Point

Portions of the India Basin / Bayview zone include the former Hunters Point Naval Shipyard, a federal Superfund site. Remediation timelines are controlled by the U.S. Navy and the EPA, not the city. A local ballot measure cannot accelerate federal environmental cleanup. The delivery schedule for that zone assumes remediation proceeds on the current federal timeline, but delays are possible and outside the city's control.

What It Does

• Authorizes $22B in infrastructure: $5.5B GO bonds, $5.9B developer agreements, $10.6B Mello-Roos districts
• Mandates ministerial (by-right) approval for code-compliant projects in all 12 growth zones
• Funds a single Programmatic EIR covering all 12 zones, eliminating project-by-project environmental review
• Codifies height and density limits that cannot be reduced by future Boards without a public vote
• Suspends inclusionary housing mandates; developers compensate nearby residents directly — no fund, no nonprofits, no city administration
• Reduces impact fees from $101,069 to $15,000 per unit (sewer + school only); other infrastructure funded by bond
• Exempts private residential construction in growth zones from prevailing wage and local hire requirements
• Eliminates parking minimums in growth zones
• Flat $2,000 permit fee replaces percentage-based fee schedule
• All exemptions are project-specific and expire at completion of 213,000 units or December 31, 2042
• Public infrastructure retains prevailing wage and local hire
• Controller audits bond spending annually. Requires 2/3 voter approval (66.67%)

Key Provisions

Financing: $22B total. GO bonds ($5.5B) for citywide systems. Developer agreements ($5.9B) for zone-specific trunk lines and power. Mello-Roos ($10.6B) for schools, parks, fire, police.

Infrastructure: $7.9B schools, $3.1B stormwater, $2.9B water/sewer, $1.8B power, $1.7B wastewater, $1.4B parks, $1.2B roads, $0.8B fire, $0.6B bus, $0.5B police, $0.2B 911.

Zones: 12 zones with codified height/density limits. Board can reallocate up to 10% between zones by supermajority.

Milestones: 2030: 17K. 2033: 72K. 2037: 166K. 2040: 213K. Miss by >15% triggers bond pause.

Permitting: Ministerial approval for code-compliant projects. Programmatic EIR covers all 12 zones. Flat $2,000 permit fee.

Entitlement Reform: Impact fees reduced to $15K/unit. Inclusionary suspended — developers compensate nearby residents directly. Prevailing wage and local hire waived for private construction. Parking minimums eliminated. All exemptions expire at 213K units or Dec 31, 2042.

Revenue: $3.1B/yr recurring at build-out. $6B one-time through 2040. $22.7B cumulative.

Affordability: Inclusionary mandates replaced by direct developer payments to existing residents near construction. No city fund, no nonprofits, no intermediaries. Older buildings are naturally affordable — direct payments make them more so.

Labor: Private residential construction: market-rate wages. Public infrastructure: prevailing wage and 30% local hire retained.

Legal: GO bonds = 2/3 vote. Prop 13/218 compliant. Voter initiative — cannot be repealed by Board.

9. Next Steps: Referendum Calendar

Phase 1: Draft & File (March-May 2026)

Phase 2: Signature Gathering (June-August 2026)

Phase 3: Campaign & Vote (Sep-Nov 2026)

Phase 4: Build (Q1 2027 onward)

Immediate Next Steps

• Controller models debt service + property tax rate impact
• City Attorney reviews draft for legal sufficiency
• Decide: full GO bond ($22B, 2/3 vote) vs. hybrid ($5.5B GO + developer agreements + Mello-Roos)
• Build coalition: SPUR, YIMBY Action, Housing Action Coalition, building trades, SFUSD, SFPUC
• Draft full referendum text (see addendum)
• Finalize developer fee approach

10. Addendum: Full Text of Proposed Referendum

Full text of proposed ballot measure. Subject to City Attorney review and certification.

11. Addressing the Opposition

"It's Too Expensive"

The infrastructure costs $22 billion. Cumulative tax revenue through 2040 totals $22.7 billion. Under the hybrid financing approach, only $5.5 billion requires voter-approved GO bonds — the rest is funded by developers and new residents. Property tax alone from 213,000 new units generates over $2.4 billion per year, every year, permanently. The cost of doing nothing is losing $3.1 billion per year in revenue the city will never collect.

"What About Existing Residents?"

The old approach required 15-20% of new units to be rented below market rate. This added ~$140,000 to every market-rate unit, produced expensive new apartments for a small number of lottery winners, and funneled hundreds of millions through nonprofit developers, consultants, and city agencies. The fundamental premise — that the way to provide affordable housing is to build brand-new units and sell them at a loss — is wrong.

New construction is the most expensive housing in any market. An older building with its mortgage paid off, its Prop 13 base locked in, and its operating costs stabilized is naturally affordable. That is how affordability works in every city in the world: people live in older buildings at lower rents. The inclusionary system fights this reality by spending $600,000-$1,000,000 to build a new unit and then renting it for $1,200/month to one household. The subsidy per household is staggering. The number of people served is tiny. And the money passes through layers of organizations that take their cut before it reaches anyone.

This plan does the opposite. Developers compensate existing residents near their construction sites directly. A renter paying $2,400/month who receives $300/month from the developer next door has an effective rent of $2,100 — affordable to a household earning $84,000. No fund, no nonprofit, no consultant, no compliance report. Developer pays resident. Every dollar goes to the person.

Why This Is Better

• Tens of thousands of households benefit instead of ~1,500 lottery winners per year under the old system.
• Older buildings become the affordable housing. Direct payments reduce effective rents in existing stock — which is where affordable housing actually exists, not in expensive new construction.
• Zero middlemen. No fund, no nonprofits, no consultants, no city administration. Developer to resident.
• It changes the politics. A new building going up next door means a check in the mail. Existing residents become allies of construction instead of opponents.
• Lower market rents for everyone. Removing the $140K/unit cross-subsidy means new units rent at $2,500 instead of $4,500. More supply at lower prices benefits the entire market.
• It eliminates the corruption. San Francisco spends $1.1 billion/year on homelessness and housing nonprofits. Multiple Controller audits have found waste, cost overruns, and lack of outcomes. This plan routes money directly to people and eliminates every organization that profits from the housing shortage.

The Real Opposition: Protecting a Broken System

The loudest opposition comes from organizations that have built careers and budgets around keeping the city small. The housing shortage is not an accident. It is a business model.

San Francisco spends over $1.1 billion per year on homelessness and housing-related nonprofits. The problem has not improved. The incentive structure rewards managing the crisis, not ending it.

Multiple audits by the Controller and Budget & Legislative Analyst have found widespread waste, lack of accountability, and in some cases fraud in nonprofit contracting. Organizations receive tens of millions in annual city contracts with minimal performance requirements. Their political influence blocks housing development, which would reduce the crisis that justifies their funding.

Paying for It by Cutting Waste

• $1.1B/year on homelessness nonprofits. A 10% efficiency gain frees $110M/year.
• Controller has identified hundreds of millions in contracts with inadequate performance measurement.
• Multiple nonprofits charge 30-40% administrative overhead.
• Direct resident benefit at $1K/household/year for 95K households costs $95M/year — easily covered.

About

I'm Brad. I work in tech and I've lived in San Francisco for 14 years. I want the city to grow, but in a way that benefits new residents and old.

SF2040 started with the idea that the only way to fix San Francisco real estate is to fix everything at once in one giant referendum. Without this, we'll be stuck fighting over every building and street in the city, one by one, to no one's benefit.

The bill must please all major coalitions in the city, while also angering all of them a little. Everyone wins and everyone compromises a little. That's the only path forward.

If this excites you, please sign up here.