The cost of capital is an entitlement risk

Twelve extra months of Loudoun County's special-exception review, on a representative $40 million construction draw, costs somewhere between $2.7 million and $3.5 million in interest alone. Nobody sends that number to committee. It should be the first number on the page.

Here is where it comes from, and why it did not matter a decade ago and matters enormously now.

A timeline used to be free. It is not anymore.

Entitlement has always carried timing risk. Loudoun's own post-March 2025 case file puts it plainly: Virginia special-exception review means a Planning Commission hearing, a Board of Supervisors vote, public comment, and written findings, and "6-18 months is a reasonable planning range, not a guarantee." That range has existed in some form in every jurisdiction with discretionary review for as long as discretionary review has existed. What changed is what a month of waiting costs.

Between 2009 and 2021, the effective federal funds rate spent most of its time near zero. A construction loan floating a few hundred basis points over an overnight index that is itself near zero is, for underwriting purposes, close to free money. Extra entitlement months were a calendar problem: they pushed a delivery date, annoyed a leasing team, occasionally blew a rent-comp assumption. They rarely showed up as a line item, because the line item rounded to nothing.

That stopped being true. SOFR sits at roughly 3.66 percent as this is written. A construction loan for a strong sponsor on a straightforward project prices at something like SOFR plus 250 to 450 basis points; a mid-market deal or a private-credit take, filling in where banks have pulled back, prices at SOFR plus 400 to 600. Both of those are real spreads quoted in the market right now, not historical footnotes. Multiply either one out and a month of delay has a dollar sign on it again, for the first time in most underwriters' careers.

Run the math once, on a real range

Take a $40 million construction draw, fully out, at two representative all-in floating rates: 6.66 percent (SOFR plus 300, a clean bank deal) and 8.66 percent (SOFR plus 500, closer to where a private-credit lender prices a transitional or entitlement-exposed asset today). Interest-only carry on that balance runs about $222,000 a month at the low end and $289,000 a month at the high end. Loudoun's own planning range spans twelve months, from a fast six-month track to a slow eighteen. Carry the difference across that full twelve months and you get $2.66 million on the cheap end of financing and $3.46 million on the expensive end, for the identical project, sitting in the identical hearing queue, with the only variable being how long the Board of Supervisors takes to vote.

That figure assumes a full draw from day one, which overstates a typical ramping construction schedule. Haircut it however you like for a realistic draw curve and the number still lands solidly in seven figures on a single mid-size deal. Scale it across a multi-site pipeline and the entitlement calendar stops being a scheduling exhibit in the appendix. It becomes a cost line that competes with land basis and hard costs for committee attention, and on a deal running the private-credit spread, it can exceed either one.

Most screening models do not carry that line at all. They carry a probability of approval, sourced from comparable outcomes, which is the right instinct: RealClear's own read on entitlement risk starts there, because a project that gets denied has no carry cost worth computing, it has a dead deal. But conditional on approval, the time axis is its own risk, priced in real dollars at whatever the lender is actually quoting that quarter, and treating it as free is a modeling choice nobody would defend out loud if you asked them directly.

The carry math compounds with how the loan gets sized

There is a second-order effect that makes the twelve months worse than the interest bill alone. Construction loans today are increasingly sized to a debt-yield test, stabilized net operating income divided by loan amount, rather than to loan-to-cost, because rates moved further and faster than cap rates did. Most banks want 8 to 10 percent debt yield before they will size proceeds at all. That test does not care how long entitlement took. It cares what the asset earns once it opens. A project that spends eighteen months instead of six in Loudoun's queue pays twelve extra months of interest on the same loan amount, and it is frequently carrying that interest on a loan sized under an older, tighter debt-yield assumption than the one the market will demand by the time it finally closes, because underwriting standards for construction and land-development loans have kept drifting even while banks eased elsewhere. A slow entitlement clock costs more in carry and can shrink the loan that carry is attached to, both at once. Nobody budgets for the second half of that sentence.

Time and money are not the same risk, and jurisdictions trade between them

Loudoun's range is a time cost with no cash outlay attached. Compare it against Saline Township, Michigan, which we have written about before for a different reason: a 575-acre rezoning denied 4 to 1 on September 10, 2025, settled by consent judgment on October 15, thirty-five days later. Fast. But the settlement carried a $4 million farmland trust, a $2 million community fund, $7 million for the local fire department, and $500,000 each for two neighboring departments, fourteen million dollars in cash commitments in exchange for compressing the clock from an open-ended fight down to five weeks.

Fourteen million in cash, paid once, against three and a half million in carry, paid slowly. Different jurisdictions are quietly offering the same trade in opposite directions: pay cash up front and buy speed, or hold cash and pay the lender instead. Almost no pro forma frames it that way. It should, because once you can price both sides in dollars, "which jurisdiction is cheaper" becomes an answerable question instead of a vibe.

The scale of the exposure is not a one-off either. The Mortgage Bankers Association counts $875 billion of commercial and multifamily mortgage balances maturing in 2026, 17 percent of the $5.0 trillion outstanding, with hotel loans at 30 percent maturing and industrial loans at 23 percent. A meaningful share of that stack sits behind assets that are still moving through some jurisdiction's entitlement or re-entitlement process on the way to stabilization. Every month those specific loans wait on a hearing date is a month compounding against a refinancing rate nobody has locked yet, on top of a maturity date the borrower does not get to move.

The number that should be on the first page of the memo

I think within two underwriting cycles, standard site memos on any project with a discretionary approval step will show a carry-cost sensitivity table the same way they already show a rent-comp sensitivity table: fast case, base case, slow case, each with a dollar figure attached, run at the actual quoted spread on the actual deal rather than a stale rate card. Today almost none of them do. That gap is not because the arithmetic is hard. It is four numbers and a calculator. It is because entitlement time has been modeled as a scheduling risk for so long that nobody rebuilt the template when the discount rate stopped being free.

My bet, stated so it can be graded: by 2028, a materially expensive entitlement timeline, not a materially uncertain one, is what kills marginal deals in committee, because the market will have already priced the uncertain ones out. If that is wrong, if committees in 2028 are still asking "will it get approved" and nobody at the table is asking "what does the wait cost," then the industry never actually learned this lesson, it just survived one expensive cycle and forgot why.

Pull up your own pipeline this week. Take the slowest jurisdiction on it, apply the spread your lender actually quoted last month, and multiply by the months your own file admits it might sit. If that number surprises the room, the room was underwriting the wrong risk.

This analysis is a source-cited research summary drawn from public records, not legal advice. It can contain errors and should be verified independently before any investment decision.