Who Can Develop When Nobody Can Develop

I spent the first ten years or so of my career on the development side of the business. Just when I was starting to get my feet under me — the requisite experience, the contacts, the confidence of the craft — the ship sailed. We are not in the golden age of development. I don't expect to return for another ten years or so.

But the truth is that new development isn't dead. It just got extremely selective.

There is a narrow lane where it still pays and where it should still be pursued. This is an attempt to draw the edges of that lane precisely, using our own small-bay development model, and to say plainly what has to be true before you step into it.

Why it doesn't pencil

Two things are elevated at the same time: the cost of capital and replacement cost. Neither is unwinding quickly. That combination is what pushed most of the pipeline to the sidelines. The math just doesn’t work most of the time.

It’s a factor of costs to develop, achievable rents, and unpredictability around exit cap rates 3-5 years out. You are underwriting a sale price you cannot see, in a rate environment nobody has called correctly in four years.

A healthy development project carries a 150–200 bps development spread — the gap between the yield on total project cost and the exit cap rate. That spread is the entire compensation for taking construction risk, lease-up risk, and time risk.

Now watch how fast it disappears:

• • 50–100 bps to cap rate expansion between underwriting and exit
• • 50–100 bps to an extended carry (and project cost inflation which usually comes with it) — permitting slips, the GC's schedule slips, lease-up takes two more quarters than the model says

You don't need both to happen. Either one alone takes a 175 bps spread down to where you’re basically working for free. Both, and you have built an asset that is worth less than it cost you to build.

The deal that fails

Here is a generic small bay development deal, underwritten with no advantages of any kind. 50,000 NRSF on 3 acres. Land at market. Construction debt at market. A normal timeline. Doesn’t take much analysis to realize this doesn’t work.

Uses

Sources

Stabilized operations — months 45 to 56

Cost against value

Nothing here is a mistake. There is no sloppy line item, no aggressive contractor, no bad site. It is a competently underwritten deal that does not clear, because a competently underwritten deal is no longer good enough.

This environment is where edges start to matter

When the market is easy, everyone can develop. Rent growth outruns mistakes. The marginal developer with no structural advantage gets the deal done anyway, because the market bails out the basis. That is what the last cycle was.

When the market is hard, that cover is gone, and everything runs through whatever structural advantage you actually have. Tough environments don't destroy competitive advantages. They reveal them, and they make them far more pronounced.

There are six that I think make the difference, more so when they compound on each other. Five of them are about the project. The sixth is about who’s building.

• • Land advantage
• • Construction advantage
• • Financing advantage
• • Lease up and rent advantage
• • Timeline advantage
• • Tax advantage

And nobody else is delivering

Here is the part that gets lost in all the hand-wringing about cost of capital. The same headwinds that make the math hard for you make it impossible for almost everyone else. The merchant developers are sidelined. The land buyers who need a construction loan at market are sidelined. The institutional programs have pulled back to their core markets and their core sizes, and small bay was never on that list to begin with.

Which means if you have an advantage and you step into the void, there is a very real chance you are the only new delivery in your trade area for the next two to three years.

That is not a small thing. Lease-up risk is the single largest risk in a development pro forma, and the void is precisely what removes it. You are not delivering into a wave of competing product. You are not matching someone else's free rent or TI package because they broke ground six months before you and needs to fill first. You set the ask. In the right submarket, you are the only ask.

And in the right submarket, rents are not merely holding — they are strong and growing. Small-bay vacancy in the good pockets is tight, the tenant pool is deep and local, and there is a replacement cost floor under the whole thing that keeps getting higher every year nobody builds. A pipeline that goes quiet for three years while demand keeps compounding is exactly the setup that produces a rent spike on the other side.

Here is what each lever is worth.

Lever #1 — Land Advantage

The best opportunities are owners who hold unimproved land they have owned for many years, at a basis well below today's market value.

The owner with cheap carry can afford to make it through the planning board hearings. He can also take the time to go through ZBA to obtain material variances without breaking the bank. He can walk away from a bad condition. He can wait out an abutter. The leveraged land buyer will wind up with a worse site at a higher basis (land + carry), because holding non income producing land is expensive when you buy at market land prices.

The icing on the cake is when the excess land is adjacent to, or part of, an existing asset that is already cash flowing. Now the carry is covered by the operating asset — the land is effectively free to hold. You may also inherit the curb cut, the detention, the utility runs, and a tenant base that already knows the address.

Lever 1 — where the improvement comes from

Lever 1 — the full lever against the base case

Lever #2 — Construction Cost Advantage

Market developers that just hire a GC have no construction cost advantage. By definition they are paying market and will build on a market timeline.

Owners who have the requisite skill set, background, infrastructure, etc to self-perform, or have the competency in-house to price and manage the work themselves have a huge edge. Hard cost is a line item that most players have no influence over, and it’s also the largest line in the budget by a wide margin — $150 a foot against $60 of land. Those who can genuinely generate savings here have a structural advantage.

And ten dollars a foot buys more than ten dollars a foot. Architecture and engineering are struck as a percentage of hard cost. Contingency is a percentage of hard plus soft, so it follows both down. Interest and loan fees follow the smaller loan. Take $10 off the hard number and $11.84 comes out of total cost.

Lever 2 — the full lever against the base case

Lever #3 — Financing Advantage

Higher rates and longer hold periods will crush the development model. But the coupon is the smallest part of the story. Even bank financing at an attractive headline rate has its pitfalls:

• • Full recourse, burning down at CO
• • Completion guarantee
• • Net worth and liquidity covenants that follow you across every other asset
• • A capitalized interest reserve sized into the loan proceeds — you are borrowing to pay your own interest, and paying interest on that
• • 30–45 day draw lags, which the GC prices into the bid whether or not you see the line

Filling the gap with pref equity or mezz at 12–15%, weighs on the capital structure further. Every one of these is a real cost that never shows up as a rate.

Those with established portfolios — who did not overleverage in the last cycle, who have optionality with a credit facility, or who could fund construction out of refinance proceeds on another asset — have a structural competitive advantage. They skip the reserve, skip the draw lag, skip the pref, and price the job as a cash buyer.

The cost, of course, is contamination risk. Using debt across the portfolio, or on one asset, to pay for the development of another ties their fortunes together. A stabilized building that was carrying itself comfortably is now underwriting a project with construction risk attached. That is a real risk and it deserves to be priced, not waved through — but it is a risk a strong balance sheet can hold and a merchant developer cannot.

Lever 3 — rate against structure

Lever 3 — the full lever against the base case

Lever #4 — Time

Shaving six months of downtime off the construction timeline, or off the lease-up period, is material.

It is the only lever that is earned rather than inherited. Land basis and balance sheet are positions you already hold or you don't. Time is execution.

What actually buys it:

• • Product designed to the demand curve throughout the process, not just at the start
• • An ability to pre-lease, or at minimum to build interest and demand so it coincides with TCO
• • Willingness to lease in process and deliver in phases
• • Simplified leases and an expedited negotiation

Small bay is unusually well suited to all four. You are not waiting on one 50,000 SF requirement to appear — you are filling twelve to fifteen units out of a deep, local, non-institutional tenant pool. Phased delivery is natural rather than a concession. And the lease itself should be a short form document. Nobody should spend ninety days and two rounds of outside counsel negotiating a 4,000 SF deal; the legal spend and the delay both come straight out of the return.

Lever 4 — where the improvement comes from

Lever 4 — the full lever against the base case

Lever #5 — Rents

This one is obvious and if realistic and achievable moves the numbers considerably. Achieving higher rents means a high valuation on the refi or exit.

It comes from two places. The first is the asset itself — the specific site and the strength of the submarket around it. Location within the trade area, access and truck movement, ceiling height and dock configuration, the depth of the local tenant pool.

The second is narrower and more often left on the table: the marginal dollar that comes from having the right broker on the building, one who knows the tenant base, holds the ask, and pushes for top dollar instead of filling space quickly to get paid. On twelve suites that difference compounds, because the first two leases set the comparable for the next ten.

A dollar of rent is not a dollar of value. Capitalized at 6.50% it is worth about fifteen dollars of value, and against the yield this deal actually earns it moves the development spread about seventeen times as far as a dollar taken off cost. That’s why the income side is the most impactful variable.

Here is what one additional dollar per foot across the whole rent roll does.

Lever 5 — the full lever against the base case

Lever #6 — The Depreciation Schedule

The first five levers change the project. This one doesn't touch the project at all. It changes who the project is worth more to.

A merchant developer who builds and sells on completion never sees this lever. A private investor with an income problem — real income, taxed at the top of the schedule, with nothing left to shelter it — is buying the same building on materially different terms, because a large share of what he spends comes back as tax he doesn't pay in the year he spends it.

The number to run this on is not total project cost. Land is not depreciable and nor is the carry on it, and the financing and leasing costs sit outside the building basis too — loan fees amortize over the loan, commissions over the lease term, and funded reserves are not an asset at all. Take those out of the $269.19 a foot in the model above and $199.36 is left: hard cost, soft cost, entitlement, contingency, and the construction interest that capitalizes into basis. Call it $199 a foot. That is what a study actually works on.

What is actually depreciable

Left alone, that $199 sits on a 39-year straight-line schedule and returns about $2.34 of deduction in year one. A cost segregation study breaks the building into its actual components and reassigns the ones that aren't really building. Small bay is unusually good at this: the paving, truck court, curbing, site utilities, yard lighting, and fencing are a large share of the job, and every one of them is a land improvement on a 15-year life rather than a shell on a 39-year life.

That reassignment matters now in a way it didn't for most of the last three years. The One Big Beautiful Bill Act, signed July 4, 2025, restored 100% bonus depreciation permanently for qualified property acquired after January 19, 2025, reversing the phase-down that had it scheduled to reach zero in 2027. Anything with a recovery period of twenty years or less is deductible in full the year it goes into service — which covers both the 5-year personal property and the 15-year land improvements.

Here is the whole lever on one page, at the federal level, on the $199 a foot that is actually depreciable:

So the year-one number is about $23 per square foot of federal tax not paid — a little under twelve percent of the depreciable basis, and under nine percent of total project cost, back in the first return.

This is a deferral, not a discount, and the last column is where you can see it. The $199 gets deducted either way. On a 39-year schedule it comes back as $81.34 of federal tax over four decades; with a study and full bonus, the same $81.34 comes back — just front-loaded. The full-life advantage is zero. Cost segregation does not change what the building costs after tax. It changes when you get the money.

Which means the honest way to size it is present value. Discounted at 8%, the federal shield on a straight 39-year schedule is worth $23.78 per foot. With the study and full bonus, it is worth $38.05. The lever is the difference: $14.27 per square foot, or seven percent of the depreciable basis.

That is a real number and it is worth pursuing.

Three conditions gate the whole thing:

• • The loss has to be usable. Rental real estate losses are passive. Without real estate professional status or other passive income to absorb them, the deduction suspends and waits — which eliminates the timing benefit that is the entire point of the lever
• • The excess business loss limitation under §461(l) still caps non-passive business losses, and OBBBA made it permanent
• • Recapture comes due on a sale. Short-life property recaptures at ordinary rates under §1245; the building recaptures at 25% under §1250. A deferral becomes something closer to permanent only if the asset rolls into a 1031 and the schedule restarts somewhere else — which is why a cost segregation strategy and an exchange ladder go hand in hand.

Put the gates together and the profile this lever is built for is narrow and specific: a private owner, with passive income to shelter, holding long term, exchanging rather than selling. Which is more or less a description of every good small-bay owner I know — and none of the merchant developers who are currently sidelined.

None of this is tax advice. Allocations and usability are facts-and-circumstances questions and belong with your CPA before they belong in a pro forma.

Stacking the levers

The important question isn't just what each lever is worth in isolation. It's how many of them one can stack to compound an advantage.

What each lever adds in stack position

Net basis after the depreciation shield

The theme

A theme I constantly see amongst our private clients: the people positioned to develop right now are usually not developers by trade.

They are operators and owner-occupiers. Second and third generation families who own the gravel yard next to the building. People with a bank relationship going back twenty-five years and a land basis that is a number from a different decade. They did not assemble these advantages in order to develop. They assembled them by staying in one place for a long time and not overreaching in the good years.

Meanwhile the professional merchant developers — the ones with the actual craft — are largely sidelined, because their model requires land at market and a construction loan at market, and that math does not work right now.

The edge in this market isn't craft. It's position.