A new working paper out of Princeton and the University of Chicago’s Booth School puts real numbers behind something loan officers have always sensed. Build the same single-family home in Indianapolis and it’s move-in ready in six months. Build it in San Francisco and the clock runs closer to four years. That difference changes how long a construction loan stays open, and how much can go wrong before it closes.
Same house, eight times longer to build
Economists at Princeton and Chicago Booth pulled roughly 1.1 million permit records covering 2.7 million housing units across 60 U.S. cities, going back to 2000. To compare cities fairly, they built a standardized project for each one, the same size single-family home and the same size 20-unit apartment building, and measured how long that identical project would take from permit application to completion, as ResiClub covered this month.
The fastest cities in the study, Indianapolis, El Paso, and Colorado Springs, get a standardized single-family home from application to move-in ready in around six months. San Francisco takes nearly four years for the same project, with Miami and Philadelphia close behind at roughly 3.4 years each. Permitting alone, not construction, just the wait for approval, explains 30% to 43% of why the slowest cities take so much longer than the fastest ones. In Miami that’s about 1.4 years spent waiting on approval before a crew ever breaks ground. In Indianapolis or Albuquerque, it’s a matter of weeks.
Multifamily follows the same pattern, stretched further. A standardized 20-unit building takes about two years in Raleigh, Grand Rapids, or Reno, and 4.6 to 4.8 years in San Francisco, Miami, and New York. Nationally, multifamily permitting time roughly doubled between 2000 and 2023, and single-family permitting nearly tripled over the same period.
Twenty draws instead of five
None of this changes what a construction loan does. It changes how long that loan stays open while it does it.
A loan in a fast-permit market might run four to six draws over six months. The same size loan in a slow-permit market can run twenty or more draws stretched across three or four years. Take a retainage figure that’s off by a decimal on draw three. In a six-month loan, the next draw catches it within weeks. In a four-year loan, that same error can ride along for a year or more before a reconciliation cycle turns it up, compounding with every disbursement that follows it.
A build that runs three or four times longer runs through more inspections and more lien waivers along the way, and it gives a line item that much more time to drift from budget before anyone catches it. A draw process built around the assumption that a construction loan is a six-to-nine-month product starts breaking down the moment a loan runs three times that long.
One draw policy, two risk profiles
Most banks and credit unions aren’t lending in just one city. A portfolio that spans a fast-permit Sunbelt market and a slow-permit coastal or historic-district market is running two different risk profiles under one draw policy, whether that’s the plan or not.
It’s tempting to treat this as a staffing problem, add headcount to keep closer tabs on the longer-running loans. That doesn’t change how long those loans stay open, or how many draws run through them before a project closes. It calls for a draw process built to handle a loan that closes in six months and one that’s still open four years later, with the same discipline applied at draw one and draw thirty. CoFi Blueprint pairs purpose-built technology with hands-on construction expertise to run the program with you, backed by 100+ years of combined construction lending experience across the team reviewing every draw, for as long as a project stays on the books.
A city isn’t always a market
A real limitation in the data is that permit records are city-level, not metro-level, and in a lot of the country’s fastest-growing homebuilding markets, particularly Sunbelt metros, a large share of single-family construction happens on unincorporated county land or in suburban jurisdictions outside the central city’s own limits. A metro’s overall building climate can look very different from what its downtown permitting office reports on its own.
That unpredictability shows up inside the dataset too. Charleston, South Carolina posts one of the longest permitting stretches in the entire study, despite being a smaller city that doesn’t fit the coastal-versus-Sunbelt pattern the rest of the data follows. A portfolio built on the assumption that permitting speed tracks neatly with a city’s size or region would miss a market like Charleston completely.
The trend line offers a little more nuance too. The multifamily doubling and single-family tripling in permitting time both come from a much smaller starting point than the headline numbers suggest, and the authors note tentative signs of improvement in both figures over just the past couple of years. Tentative is the right word for it. Eighteen straight quarters of tightening AD&C credit haven’t exactly been offset by a couple of quarters of modest permitting relief.
Slower-permit markets aren’t the risk
At CoFi Blueprint, the position is simple—you shouldn’t have to choose growth over risk. A market with a four-year permit timeline can be worth lending in just as much as one with a six-month timeline, provided the draw process behind it is built to run that long.
Talk to the Blueprint team about what a fully managed draw process catches on both a six-month build and a four-year one.
