The Housing Production Funnel

What happens when a city legalizes fourplexes? Opponents warn it will destroy neighborhood character. Proponents celebrate the thousands of new homes they assume will follow. Neither is true.

"It's not a house, it's a oneplex."
Illustration by Alfred Twu; used with permission.

Zoning reform opens the possibility of more homes, but the number that actually get built is determined by a funnel of feasibility, site turnover, and construction constraints. Walk through that funnel, Grand Rapids’ likely outcome becomes clear: 43 to 500 additional homes per year, depending on how aggressively the city fixes the rest of its zoning ordinance. The difference between those numbers and what we choose to do about them matters far more than the word “fourplex” ever will.

Grand Rapids approved its new 20-year Community Master Plan at the end of 2024, setting the stage for citywide fourplexes once our consultant completes the zoning rewrite, expected by early 2028. The natural question is: how many incremental units of housing can we expect?

Why Zoning Reform Frequently Disappoints

Headlines love to cite the theoretical denominator (“Units could quadruple!”), but residents live in the numerator: what actually gets built.

A 2020 UCLA study shows a large gap between permission and production. Most lots never change hands. Most redevelopments don’t pencil. Many approved projects never break ground. It’s like job hunting or dating: 100 leads → 10 matches → 3 conversations → 1 viable outcome. Zoning is only the first gate.

The Four Housing Production Funnel Gates

Gate 1: Zoning — What’s Allowed

Right now, Grand Rapids allows duplexes by right only on corner lots—a reform that produced few units because most of those lots never sell or already have newer homes that aren’t worth demolishing. Selective upzoning like this creates artificial scarcity: only a slice of the city receives new development rights.

Citywide fourplex zoning removes this structural bottleneck. It ensures that when a parcel does become feasible and does come up for sale, zoning won’t be the reason it stays frozen as a single-family home.

But legal permission is only the first filter. The next gates narrow things dramatically.

Gate 2: Economic Feasibility — What Pencils

Of the city’s 57,093 parcels, only a fraction make financial sense to redevelop. A project must clear several economic hurdles at once:

A redevelopment only pencils if the expected revenue outweighs construction costs, land prices, financing, and required returns:

• Construction costs (materials and labor fluctuate constantly)
• Financing conditions (a project that works at 4% interest may die at 7%)
• Market rents or sale prices (which vary by block)
• Required returns (institutional investors want 15–20%; locals may accept less)

The UCLA study found that fourplex zoning produces 23 to 275 additional units per 1,000 parcels, depending on land values and rents-a more-than 10x difference within the same state.

Scaled to Grand Rapids, that implies 1,311 to 15,675 net new market-feasible units under current market conditions.

But these aren’t all fourplexes. Using the unit mix reported in the study’s Appendices, the average upzoned redevelopment adds about 3.33 units per parcel (a mix of duplexes, triplexes, fourplexes, and rare 5-unit outcomes). Dividing gives an estimated 400–4,700 economically feasible parcels in Grand Rapids (≈0.7%–8.2% of all parcels).

Gate 3: Site Transactions — What Gets Sold

Even feasible sites don’t redevelop unless their owners decide to sell.

According the Grand Rapids Assessor and the Grand Rapids Association of Realtors, Grand Rapids sees only 3.6%¹ turnover per year, or ≈ 2,000 residential sales. That turnover caps the number of annual redevelopment opportunities.

Figure: Residential sales in Grand Rapids have fallen by more than half since 2018, from ≈ 4,300 annual to ≈ 2,000. This directly constrains redevelopment opportunities, regardless of zoning. Lower turnover means the funnel works more slowly, so zoning reforms must be MORE aggressive to compensate.

Gate 4: Housing Production — What Gets Built

Once a feasible parcel sells, construction still isn’t guaranteed. A project can stumble because of:

• financing lapses
• labor shortages
• permitting delays
• inspection bottlenecks
• contractor availability
• developer capacity

Though this completion rate varies significantly by project size and developer experience, I estimate a 90% completion rate for permitted projects that contain 4 or fewer units. The Development Center likely tracks this ratio, but does not publish it.

The Math: From 57,093 Parcels to 43 - 500 units

Table: Annual housing production under fourplex zoning, low and high scenarios

For context, the city currently permits 750–800 units/year. Fourplex zoning adding 43-500 incremental units represents a 6–67% increase. Where Grand Rapids lands in that range depends on two factors:

1. local market conditions: interest rates, construction costs, achievable rents, etc.
2. how aggressively the city reforms complementary regulations during the zoning ordinance rewrite.

Why Complementary Reforms Matter

The UCLA analysis assumes fourplexes must follow existing single-family rules for:

• greenspace
• setbacks
• height limits
• lot area per unit

Grand Rapids’ current zoning ordinance heavily restricts all four of these. Eliminating or reducing them would shift many more parcels into the “feasible” category. I’ve written about these previously if you want more details.

Zoning sets the size of the top of the funnel. These reforms widen the middle.

Hyperlocal Feasibility: Why Outcomes Will Vary by Neighborhood

Feasibility isn’t uniform: areas with better amenities, transit access, parks, or job proximity will hit feasibility thresholds sooner. Lower-amenity neighborhoods may need complementary reforms (e.g., reduced parking) for redevelopment to pencil.

Citywide zoning allows the market to find the economically efficient locations rather than having city hall guess.

This volatility is why zoning reforms must be bold; we should allow more density than we think we need, because market conditions will prevent most of it from being built anyway.

What Grand Rapids Should Expect

For policymakers:

Fewer than 8% of parcels are likely feasible for fourplexes in normal conditions. Only 3.6% of those sell annually. Only 90% of those become housing. Even at the high end, 500 units per year is less than 1% of the city’s housing stock.

This is incremental, not transformative. Your policy choices determine whether we end up closer to 43 units or 500.

What you can influence is how wide the feasibility funnel opens. Policies like eliminating parking requirements, reducing combined side setbacks from 14 to 10 feet, and removing the 2,000 sq ft per unit minimum make more projects financially viable, which translates directly into more homes built.

For housing advocates:

The gradual pace of change is both the good news and the challenge. When skeptics worry about neighborhood transformation, you can point to the math: the funnel filters so aggressively that even at the high end, we’re talking about less than 1% annual change.

But that same filtering means your work isn’t done when fourplexes become legal. The difference between 43 units per year and 500 units per year depends on the complementary reforms happening during the ordinance rewrite right now. This is the moment to push for reducing setbacks, eliminating parking minimums, removing per-unit size requirements, and streamlining permits.

Fourplex legalization opens the top of the funnel. Those follow-up reforms widen the middle. Without both, we get the symbolism of reform without the housing.

For residents:

Real estate development is slow, capital-intensive, and risk-averse. Market conditions cause variation year to year, but the pace will be gradual. Over 10-20 years, a few hundred extra homes annually add up to thousands more neighbors: including your kids when they’re ready to move out, your coworkers, the teachers at your children’s school, the nurses at your hospital, and your aging neighbors who want to downsize without leaving the community.

The numbers show fourplex zoning won’t remake your block overnight. But it will create space for the next generation. Our Community Master Plan says it plainly: “all neighborhoods must allow for some change.” Grand Rapids cannot stay affordable or keep its talent if every neighborhood insists on staying exactly the same.

The choice isn’t between change and no change; it’s between a little bit of gradual change spread across the city, or watching your own children, neighbors, and essential workers get priced out entirely.

The Bottom Line

Citywide fourplex zoning will increase housing production, but only if we let it work.

The difference between 43 units and 500 units per year isn’t luck—it’s policy.

Fourplex legalization is the beginning of reform, not the end. Feasibility depends on setbacks, greenspace rules, height limits, parking minimums, and permitting timelines. Fix those, and the funnel widens. Ignore them, and it narrows.

Grand Rapids has a rare opportunity: to design a zoning ordinance that actually produces the housing its residents need. Let’s not squander it.

Methodology Notes

This analysis uses Grand Rapids assessor data (57,093 residential parcels), GRAR market data (3.6% annual turnover), and applies feasibility insights from Paavo Monkkonen, Ian Carlton, and Kate Macfarlane’s UCLA study “One to Four: The Market Potential of Fourplexes in California’s Single-Family Neighborhoods” (2020). Grand Rapids’ lower land values suggest feasibility closer to lower-cost California markets, though strong local demand and job accessibility may push results higher.

To ensure the 57,093 residential parcels were the correct number to use as a starting point, I compared it to the Census’s ACS B25024 table, which breaks down housing units by type. Just adding 1-4 unit properties gave me an estimated 58,147 parcels. Because the ACS counts units (not parcels) and because small multi-unit buildings appear as multiple units but a duplex usually sits on a single parcel, the ACS-derived parcel count should be slightly higher than the assessor’s. A 1.8% difference is exactly what we’d expect.

Source: U.S. Census Bureau (2023). Units in Structure American Community Survey 1-year estimates. Units/Parcel and Est. Parcels calculated by author.