A $42M Multifamily Value-Add We Passed On

1. Deal at a glance

The offering was straightforward on the surface. A 220-unit garden-style community built in the late 1990s, located in a Sun Belt secondary MSA with employment growth above the national median. The acquisition price implied a $191,000 per-unit basis. Going-in cap rate of 5.25 percent, exit cap rate assumed at 4.75 percent, 70 percent loan-to-value bridge financing for the first 24 months, with a planned refinance into agency debt at stabilization.

2. Sponsor profile

The sponsor presented an 11-deal track record across roughly nine years, with three full-cycle exits and eight deals still in their hold period. The team has a real operations function: in-house construction management, regional property management partnerships, and a CFO with prior REIT experience. None of that was the problem.

The problem was in the realized exits. Of the three full-cycle deals, two underperformed their projected IRR by more than 400 basis points. One delivered 9.8 percent against an 18 percent projection, the other delivered 11.2 percent against 16 percent. The third exit beat projection by 200 basis points. The sponsor's marketing materials reported the simple average across all three, blending the outlier upward.

3. The acquisition thesis as presented

The pitch had three legs.

1. Rent comps support a 22 percent rent bump over a 36-month value-add execution. Sponsor cited a comp set of three nearby properties with average rents $310 per month above the subject's in-place rents.
2. Value-add capex of $8,000 per unit covering kitchen and bath refresh, flooring, lighting, and unit-turn cosmetics. Total capex budget of $1.76M plus $400k of common-area work.
3. Exit at a 4.75 percent cap on stabilized year-5 NOI of $2.39M, implying a $50.3M sale price net of selling costs.

Each leg has a published comparable or assumption attached. The model is internally consistent. The question is whether the inputs are correct.

4. The numbers reality check, leg one: rents

The sponsor's three-property comp set used asking rents from listings, not signed-lease rents. We pulled CoStar and the local MLS for the trailing 12 months of actual signed leases at the named comps and found a different picture.

Subject in-place 2BR rents averaged $1,545. Against the actual signed comp average of $1,670, the achievable post-renovation rent is approximately 8 percent higher than current rents, not 22 percent. That single correction takes the year-5 NOI from $2.39M to roughly $2.04M, a 14.6 percent reduction before any other changes.

5. Reality check, leg two: capex per unit

An $8,000 per-unit interior capex budget on a late-1990s Class B asset is on the low end of what we see executed. Late-1990s vintage means original kitchens, original baths, original HVAC nearing end of life, original water heaters, original electrical panels in some unit configurations, and aging windows.

Across the last 36 months of value-add executions we have tracked at this vintage, interior capex has run $11,000 to $13,000 per unit when the scope includes kitchen cabinets, countertops, appliances, bath vanities, flooring, paint, and lighting. The sponsor's $8,000 budget covers about 65 percent of that scope. The remaining 35 percent either gets value-engineered out, which limits the rent premium achievable, or it gets spent anyway and shows up as a budget overrun.

6. The capital stack red flag

This was the section where the deal moved from optimistic to structurally fragile.

Senior debt was a 75 percent loan-to-cost bridge from a non-bank lender at SOFR plus 450 basis points, with a 12-month interest-only period. The sponsor's plan was to stabilize the asset over 18 to 24 months and refinance into a 10-year agency loan at a projected 5.25 percent fixed rate.

Three things have to happen for that refinance to clear. NOI has to hit the projected stabilized number. Cap rates have to hold. And the agency rate has to be roughly where the sponsor projected it. If any one of those misses, the deal is in trouble. If the rate environment moves against the sponsor, the property cannot generate enough debt service coverage to refinance at 70 percent LTV, and the bridge starts maturing without a takeout.

The 1.20x DSCR test at the first extension is the single most important number in the cap stack. If NOI does not stabilize on the sponsor's schedule, the extension is denied, and the deal is in workout. The model does not stress-test for this.

7. Exit cap rate gymnastics

The sponsor underwrote a 4.75 percent exit cap on a 5.25 percent going-in cap. That is 50 basis points of cap rate compression over a five-year hold. We pulled trades in the same submarket and product type for the last 18 months. Cap rates have decompressed approximately 75 basis points from the 2021 trough. The sponsor's exit cap assumes a full reversal of that decompression plus another 50 basis points of compression on top.

For context, that exit cap implies a 21x multiple of stabilized NOI. The going-in implies 19x. To justify 21x in year five requires a 10-year Treasury at or below 3.5 percent and a healthy spread on private real estate over Treasuries. Neither is the consensus view of the curve five years out.

If the exit cap is held flat at 5.25 percent instead of compressed to 4.75 percent, the sale price drops from $50.3M to roughly $45.6M, equity multiple drops from 2.1x to approximately 1.55x, and IRR drops from 18 percent to approximately 9 percent.

8. The waterfall

The distribution structure is where the deal economics quietly tilted toward the GP.

• 8 percent preferred return to LPs, accruing
• Return of capital to LPs
• 70 percent to LPs and 30 percent to GP on all distributions thereafter
• No second IRR hurdle
• No catch-up cap

In an upside scenario where the deal hits an 18 percent IRR, the GP captures 30 percent of every dollar above the 8 percent pref. There is no second tier where the GP share steps down or where additional outperformance gets shared more favorably with the LPs. Market institutional standard is a two-tier waterfall: an 8 percent pref, then 80/20 to a second hurdle (typically 14 to 16 percent IRR), then 70/30 above that. This deal collapsed the structure to a single split that favors the GP at exactly the IRR range the sponsor is marketing.

9. What we asked the sponsor

We sent five questions before the second meeting.

10. Verdict and decision rationale

We passed. The deal is not a bad deal. It is a deal that requires three independent variables to break in the sponsor's favor, where each variable is plausible but not probable, and where the compounded probability of all three landing within underwriting tolerance is materially below 50 percent.

More importantly, the structure does not protect LPs in the scenario where the variables miss. The bridge financing converts modest underperformance into capital risk. The waterfall captures GP economics at the marketed IRR rather than rewarding outperformance. The capex contingency does not survive even minor scope creep.

11. What would have changed our mind

• A revised rent assumption based on signed-lease comps (8 to 10 percent post-renovation premium, not 22 percent), with a corresponding NOI projection
• An interior capex budget of $11,000 per unit with a 10 percent contingency, supported by a fixed-fee construction contract
• A senior debt structure with a five-year fixed-rate component or, alternatively, an interest-rate cap purchased at close
• A two-tier waterfall: 8 percent pref, 80/20 to 14 percent IRR, 70/30 above
• Sponsor co-invest of at least 5 percent of common equity, alongside LPs, on the same terms

With those five changes, the deal returns roughly 12 to 14 percent IRR in a base case, 8 to 10 percent in a downside, and 18 to 20 percent only in a clear upside. That is the right shape of return distribution for the risk being underwritten.

How to apply this analysis to your own deal review

The framework that produced this teardown is portable across deals.

1. Verify rent comps with signed-lease data, not asking rents. CoStar, RealPage, and the local MLS will all give you actual transaction rents. Asking rents are aspirational.
2. Pressure-test capex against vintage benchmarks. If the budget is materially below market for the vintage and scope, either the scope is being trimmed or the budget will overrun.
3. Stress the capital stack at the maturity test, not at the exit. The most common failure mode for value-add multifamily over the last cycle is the bridge-to-agency takeout failing, not the exit underperforming.
4. Hold the exit cap rate flat to going-in as your base case. Compression should be an upside, not an assumption.
5. Read the waterfall in dollars, not percentages. Compute GP take at the marketed return and at a market-standard alternative structure. The delta is the alignment cost.
6. Ask the sponsor for a sensitivity to 1.0x equity multiple. If they have not run it, they have not thought about the downside hard enough.

The answer to "should I invest in this deal" is rarely yes or no. It is "yes, if these three things change," or "no, because the structure does not protect me when these three things go wrong." That second sentence is what produced this pass.