A developer calls us after a pre-application meeting with the city. They went in with a 60-unit market-rate project. They came out with the city suggesting they could build 81 units if they set aside a handful as affordable. The architect is already redesigning the floor plans. The lender is excited about the increased unit count. Nobody has called the civil engineer yet.

That call — the one that happens six weeks after the pre-app, when the site plan already shows 81 units and a parking structure that doesn't quite fit — is the one we get most often. This post is for the people who want to make that call before the site plan is locked.

What California's Density Bonus Law Actually Says

Government Code Section 65915 — California's Density Bonus Law — gives residential developers a by-right density increase in exchange for including a percentage of affordable units. It's not discretionary. If you meet the affordability threshold, the city has to grant the bonus. They don't get to vote no on the extra units because the neighbors don't like it.

The bonus percentages work on a sliding scale based on how many units you make affordable and at what income level. A project with 10% of units at low income (80% AMI) gets a 20% density bonus. Push that to 15% at low income and you're at 33%. Offer 24% at low income and you hit the maximum: a 50% density bonus under current law (AB 2345 raised the cap from 35% to 50% starting January 2021).

On top of the density bonus, Section 65915 entitles you to up to three "concessions or incentives" — project modifications that the city must grant unless they can make specific findings that the modification would cause a specific adverse impact. These concessions can include:

In theory, this is a clean trade: you take on some affordability obligation, the city gives you more density and some flexibility on development standards. In practice, each concession has a cascade of civil engineering consequences that nobody tells you about at the pre-app meeting.

The Civil Engineering Cascade

Here's the chain reaction that happens when you add 21 units to a 60-unit project:

More units means more bedrooms, which under most Bay Area city parking standards means more required parking stalls — even with a density bonus parking reduction applied. Our 60-unit baseline might have required 90 stalls. At 81 units, you're looking at 120 stalls, even after reduction credits.

More parking means more impervious surface. Thirty additional stalls at standard 9' x 18' dimensions, with drive aisles, add roughly 14,000 to 18,000 square feet of new pavement. On a tight urban infill site, that's significant.

More impervious surface means larger stormwater treatment requirements. Under the MRP C.3 provisions, your bioretention sizing is a direct function of tributary impervious area. Add 15,000 square feet of pavement, and your bioretention cells need to grow accordingly — which takes space you no longer have because you used it for parking.

Less buildable area means tighter grading constraints. You're fitting more building footprint and more parking onto the same land area, which forces steeper cut slopes, more retaining walls, or tiered parking structures. Each of those has its own cost and engineering complexity.

This cascade is predictable. It's not a surprise to a civil engineer who's looked at a density bonus project before., it's frequently a surprise to the rest of the project team, because nobody mapped it out when the extra units looked like free money.

Parking Reductions: Density Bonus vs. AB 2097

Section 65915 allows reduced parking minimums as a concession. The specific reduction amounts depend on unit count and proximity to transit, a typical density bonus project can reduce minimums to 0.5 stalls per bedroom for affordable units, and in some cases lower. That's a meaningful reduction from standard city minimums of 1.0–1.5 per unit.

AB 2097, which took effect January 1, 2023, goes further: it eliminates parking minimums entirely for projects within a half mile of a major transit stop. If your site qualifies under AB 2097, you may not need any minimum parking at all — which completely changes the civil engineering equation. Less parking means less impervious area, which means smaller stormwater BMPs, which means more site area for building and landscaping.

The catch is that "major transit stop" has a specific legal definition (Gov Code 21064.3 — a rail station, a ferry terminal, or an intersection of two or more high-frequency bus routes). Not every site within walking distance of a bus stop qualifies. We check this early in site planning because it's one of the most consequential decisions you'll make on a dense urban project.

What a 35% Density Bonus Does to Your Grading Plan

Let's get specific. Take our 60-unit project, a 1.8-acre infill site in Oakland with about 6 feet of topographic relief from the street to the rear property line. The original grading plan had one level of at-grade parking, a single building pad, one retaining wall along the rear, and two bioretention cells totaling about 2,400 square feet.

Add 21 units via a 35% density bonus. Here's what changed on the civil drawings:

None of these changes were insurmountable., all of them added cost, time, and coordination that didn't exist in the original project scope. The retaining walls alone added roughly $180,000 in construction cost. The bioretention redesign required two rounds of review with the city's Public Works department.

Stormwater: C.3 Bioretention Grows With You

Under the Bay Area MRP C.3 requirements, bioretention sizing is calculated using the volumetric method: you treat the volume of runoff from a 1-inch storm event over the tributary impervious area, using a curve number approach or the directly-connected impervious area method, depending on your jurisdiction's preferred protocol.

When you add 15,000–18,000 square feet of impervious parking surface, the bioretention treatment volume increases in proportion. On a site where you've already used every square foot of grade-level open space to meet the original BMP requirement, that growth has to go somewhere. The options, in rough order of cost:

  1. Expand existing bioretention cells — requires available adjacent space, which is usually the landscaping you were planning to use for project amenities
  2. Add a second bioretention location — sometimes the parking layout creates new drainage tributary areas that can support their own BMP
  3. Linear biofiltration strips along drive aisles — works well with tiered parking layouts where you have retaining wall toes that can accommodate planting
  4. Underground proprietary treatment systems (StormFilter, Modular Wetlands, etc.) — more expensive, they work where you literally have no surface area. We use these as a last resort because the O&M cost over the life of the project is meaningful

The right answer depends on the site., you can't make that call without a civil engineer who knows the C.3 math in the room when the extra units are being discussed. We do this work during early site planning — it's a few hours of analysis that can save weeks of redesign later.

Fire Access: The Conflict Nobody Plans For

Denser sites and tighter layouts create a recurring fire access problem. When you're squeezing in extra units by requesting a setback reduction as a density bonus concession, the space you recovered from the setback often becomes the space your fire apparatus needs for access or turnaround.

California Fire Code Section 503 requires a 20-foot unobstructed fire apparatus access road. On sites with buildings over 30 feet in height, the apparatus needs to be able to position within 15 feet of the building face. On a tiered parking layout where the building steps up from the parking podium, that positioning requirement can conflict directly with the parking layout.

We've resolved this through aerial ladder positioning studies (working with the fire marshal to confirm that ladder truck reach covers the building without a grade-level setback), and through designing hammerhead turnarounds into the parking tier transitions. Neither is elegant. Both require early coordination with the fire department — ideally before the entitlement submittal, not after the grading plan is 60% complete.

Utility Demand: The Numbers Scale Up

Twenty-one additional units isn't just a civil grading problem. Utilities scale up too.

Water demand increases roughly 70–90 gallons per unit per day for multifamily residential, per AWWA residential end-use data. For 21 units, that's about 1,500–1,900 gallons per day added to your peak demand calculation. Depending on your water purveyor and the capacity of the main serving the site, you may need a larger service connection. EBMUD, for example, requires a water capacity analysis for projects over 50 units — trigger that threshold because of your density bonus, and you're in a process that adds time and a connection fee calculation you hadn't budgeted.

Sewer is similar. Most Bay Area agencies size sewer laterals based on fixture unit counts, and 21 more units pushes the fixture count up meaningfully. If you're on the border of triggering a upsized lateral requirement, the bonus units can tip you over.

Electrical is often the longest lead-time item. PG&E transformer sizing and service entrance capacity are based on the project's total connected load. More units means a larger service, which may require a different pad configuration, a larger vault, or a new transformer location entirely. We've seen density bonus additions delay a project's PG&E application by six weeks because the original load calculations were submitted before the unit count was final.

When to Push Back vs. When to Make It Work

The density bonus is usually worth pursuing. The financial math — additional units at market rate with a small affordable set-aside — almost always pencils., the civil engineering consequences have a cost, and that cost needs to be in the pro forma before the entitlement strategy is locked.

We push back on architects when the bonus unit count requires site improvements that outweigh the revenue gain. We've seen projects where adding 15 units triggered a full parking structure that added $2.2 million in construction cost. The 15 units didn't come close to covering that. The right call was to cap at the base density and use the cost savings elsewhere.

We make it work when the civil impacts are manageable — when tiered parking fits the topography, when the BMP expansion has somewhere to go, and when the utility upsizing is straightforward. On those projects, the extra units are genuinely additive.

The only way to know which situation you're in is to get the civil engineer involved early. Not after the site plan is drawn. Not after the entitlement application is filed. Before the architect starts laying out units, so we can tell you what the site can actually absorb.

If you've got a density bonus project in the Bay Area and you want to know what the civil drawings are going to look like before the pre-app meeting, give us a call. That's exactly what our site planning work is designed for. We can also walk through stormwater requirements and help you understand what construction administration looks like on a project that changed scope mid-entitlement.