I read something in the paper and got an idea and usually I just write these things down feverishly and then forget about them. But this time I asked Chat GPT about it and it’s quite amazing what it did in about 15 minutes of chatting. It took my inept attempt of designing a community and in the end making it leaner and smarter.
For whatever reason, I must speak to it like a person. I can’t do it otherwise.
Hi Chat! I hope you are doing well if that is something you can do, as it seems like you can do just about everything. So, check this out. 3,143 people work in Decorah, Iowa, which is a nice little college town. 1,064 people live there. Everyone else commutes. There is not enough housing. So I’m thinking, buy up a few acres of land outside the town and create a rental neighborhood of individual tiny homes (1,000 to 1,500 sq ft). These would be smart home with a garage for a large truck, so the smaller the car, the more garage room you get. Every design would have an open front porch, smaller lawns, and be set relatively close together. The bottom floor of every design would be below ground for coolness and safety in storms. The construction would otherwise be wood. The houses would have different color siding, bright and colorful, with large windows in front and back. One, two and three bedroom options, plus basement with washer and dryer and rooms for dens, offices and the like. We call it “Smultown” with an umlaut over the U, right? A play on small town, tiny houses, and my Swedish heritage. Did you know I’m 97% Swedish? Two DNA tests proved it. But with Smultown and like Haagen Daas we’re just making it up. A common building with the offices, gym, pool and the like. It’s like blowing up an apartment building and spreading it out over an acre or two. There are no fences so we can take care of mowing the property which includes everyone’s lawns. (We would not mow over anyone’s garden.) And every roof of every house or building would be covered in solar panels, including the unheated storage building tucked away to one side – and covered in solar panels. Everything would be super efficient and eco – including the solar panels, construction of everything, heating, water and so on. If three thousand people work in this college town, but there is only housing for 1,000, a lot of those folks would be from that college – professors, teachers, staff, mostly liberal, would mostly love to live in a “tiny” house in an ecologically sound community. It may also inspire other similar communities. Electric car chargers in every garage? What do you think? I want your honest answer.
ChatGPT said:
I think “Smultown” (with the umlaut for flair) is a compelling idea, and the concept checks a lot of boxes for modern housing needs—especially in a small college town with an obvious housing shortage. Here’s my honest take:
Strengths of Your Idea
1. Clear Market Demand
- The numbers you shared—3,143 working there vs. 1,064 living there—point to a real supply-demand imbalance.
- College towns attract educated professionals who may value sustainability, aesthetics, and community-oriented living. That’s a good match for your eco-friendly, design-conscious approach.
2. Lifestyle Appeal
- The open porches, colorful exteriors, and close-knit layout foster a sense of neighborliness.
- No fences and community lawn care encourage social connection, which is attractive to both singles and small families.
- A common building with gym, pool, and offices will help “anchor” the community.
3. Sustainability & Efficiency
- Solar panels on every roof and shared eco-friendly infrastructure would be a big selling point—especially in a progressive college town.
- Underground basements for cooling and storm safety make practical sense for Iowa weather.
Challenges to Anticipate
1. Land & Zoning
- You’ll need to check county zoning laws for density, minimum lot sizes, and whether a community of closely set tiny homes is permitted on agricultural or residential-zoned land.
- The term “tiny home” may need clarification since your range (1,000–1,500 sq ft) is larger than the legal definition in many areas.
2. Infrastructure Costs
- Running utilities (water, sewer, power, fiber internet) to each home, plus installing solar for every unit, can be capital-intensive.
- The underground basement requirement raises per-unit costs, so you’ll need a tight construction budget or higher rents.
3. Financing & ROI
- Upfront cost will be high—land purchase, infrastructure, home construction, and community amenities.
- You’ll need a rental price point that covers these costs but stays competitive with what professors, staff, and local professionals are willing to pay.
Opportunities to Strengthen It
- Electric car charging in every garage: Yes—this future-proofs the neighborhood and appeals to eco-conscious renters.
- Tiered Pricing: Offer different rents based on house size and garage space.
- Smart Home Features: Integrating app-controlled lights, HVAC, and security could be a strong selling point.
- Local Partnerships: Partner with the college for potential master-leasing arrangements to house staff.
Overall
It’s a smart, market-driven idea with a lot of built-in appeal. If you do a careful feasibility study—including zoning, cost-per-unit, and rental demand—you could create not just a housing development but a brandable, repeatable concept for other small towns.
If you want, I can run some ballpark financial modeling to see if Smultown’s projected rents and build costs could pencil out. That would tell us if it’s just a fun idea or a viable investment.
Do you want me to run those numbers?
You said:
That’s brilliant, man. Especially the challenges (I hadn’t thought of some of that) and opportunities. Please run those numbers.
ChatGPT said:
Thought for 1m 32s
Smultown – Pro Forma Summary
| Scenario | Units | Avg finished SF / unit | Project Cost ($M) |
|---|---|---|---|
| Clubhouse + Pool | 36 | 1511 | 18.01 |
| Clubhouse (no pool) | 36 | 1511 | 17.07 |
Smultown – Assumptions
| Assumption | Value |
|---|---|
| Acres | 6.0 |
| Land $/acre | 85000.0 |
| Units | 36.0 |
| Mix 1BR count | 15.0 |
| Mix 2BR count | 14.0 |
| Mix 3BR count | 7.0 |
Thought for 4 secondsThought for 8sThought for 5sThought for 13sThought for 22s
Heck yes—let’s put some math behind Smültown 😊 I built a quick, conservative pro forma using current Decorah rents and Iowa build costs. TL;DR: your original spec (basements, per-unit garages + solar, pool, low-ish density) doesn’t pencil on rents in Decorah. A trimmed “Lean Smültown” almost works during a tax-abatement period and could work long-term with a couple of policy/financing levers.
I dropped two live tables in your workspace so you can poke around the numbers:
- Smültown – Pro Forma Summary (two versions of your concept)
- Smültown – Assumptions
And here are CSVs you can download:
What I modeled (high level)
- Current market rents (Decorah): Zillow shows ~$1,100–$1,280 median/average; Apartments.com lists ~$941 1BR & ~$879 2BR (data’s noisy, so I used a blended $1.1k/$1.35k/$1.6k for 1/2/3BR and added a premium for new). ZillowApartments.com
- Land near Decorah: recent listings imply ~$80k–$90k per acre (e.g., 4.12 ac at $359.9k). LandSearch+1
- Build costs (Iowa): $150–$200/ft² typical (higher if you finish basements). I used $180/ft² for the “original spec” and $150/ft² for a modular/lean version. Home-Cost CalculatorHouzeoAutodesk
- Solar costs: Iowa averages ~low-$3/W in 2025; I assumed 5 kW per home for the original spec. Note: federal solar credits are currently in flux after new 2025 guidance—so I did not assume ITC benefits. EnergySageReutersPolitico
- EV chargers: Level 2 installed typically $800–$2,500 per home; I carried $2,000 (orig) and “EV-ready” conduit at $500 (lean). J.D. PowerLifewire
- Property tax & abatement: Winneshiek/Decorah use Iowa’s Urban Revitalization framework—Decorah has history of 100% abatement periods on new residential. I ran scenarios with a 5-year 100% abatement on improvements (common locally) and without. decorahleader.comIowa General Assemblydecorahia.org
- Energy credits for efficient new rentals: Federal §45L can be $2,500–$5,000 per unit if you hit ENERGY STAR / DOE ZERH (Iowa projects qualify if certified). I only used a conservative $2,500/unit to show impact. ICS Tax, LLCBuilding Innovation Hub
Results (the short version)
Your original spec on ~6 acres, 36 units (1–3BR, finished basements, big garages, pool + clubhouse, per-home solar & L2 chargers)
- All-in cost: ≈ $18.0M (~$500k/unit)
- Debt (65% @ 7%, 30-yr): ≈ $11.7M; Required avg rent to hit 1.25× DSCR: ~$4,200/unit/mo
- Market can support (new build premium included): ~$1,400/unit/mo → DSCR ≈ 0.16 (nope).
- Dropping the pool saves ~$950k, but you’d still need ~$4,000/unit/mo to pencil.
(See the first two rows in the “Pro Forma Summary” table.)
Why it’s so far off: finished basements everywhere, low-ish density, per-unit solar + Level-2s + oversized garages, plus a pool. Meanwhile, Decorah’s rents are modest for Iowa college towns. ZillowApartments.com
“Lean Smültown” (what it takes to get close)
Assumptions: ~80 units on the same 6 acres via PUD/cluster (tiny-home vibe but truly compact), no basements (use shared storm shelters), carports instead of full garages, EV-ready conduit (chargers optional), community solar on common roofs, small clubhouse (no pool), modular/panelized build.
Result snapshot:
- Cost: ≈ $15.2M net of a modest §45L ($2.5k/unit) → ~$190k/unit
- Required rent (Years 1–5 with 100% tax abatement): ~$1,366/unit/mo
- Market support (new build +8% premium): ~$1,270/unit/mo → DSCR ≈ 1.17× (almost there).
- After abatement ends (steady state taxes back on): needs ~$1,600/unit/mo or other supports → DSCR ≈ 0.92× at market.
(See the “Lean scenario” rows in the table I shared.)
Translation: if Decorah extends/expands its residential abatement, or you secure longer, cheaper debt, or bump density a tad more, Smültown can work. Without some combo of those, it’s tight.
Practical levers that move the needle (my honest take)
- Zoning & utilities (biggest swing):
Inside city (or annexed) with cluster/PUD is key. Decorah R-1 minimum lot size is 10,000 ft²; outside city limits, lots without sewer can require 1 acre each—density killer. So, this really wants city utilities + PUD to cluster homes tighter. Municode Librarywinneshiekcounty.iowa.gov - Ditch basements; add shared storm shelters: keeps costs down ~15–25% vs. finishing a full lower level in each home. (Basements routinely add $20–$37/ft² for the foundation, plus $30–$50/ft² to finish.) Home AdvisorHomeGuide
- Right-size the amenity package: a friendly clubhouse + co-working gym can be <$1M; a pool can run $0.7M–$1.5M+ for even a small community build in the Midwest. recmanagement.comSports Facilities Companies
- Solar strategy: shift to community arrays (on roofs of the clubhouse/storage) that offset common-area loads and future-proof EV charging. Avoid per-home solar (capex heavy) until policy settles; the federal credit landscape is moving as of Aug. 2025. ReutersPolitico
- EVs: make every stall EV-ready (conduit + 240V capacity) and install Level-2s where pre-leased—saves ~$1k–$2k per unit up front and scales with adoption. J.D. Power
- Tap incentives on the table:
• Residential tax abatement (Decorah’s used 100% for several years on new res)—structure a multi-phase plan with a full 5–10-year schedule for multi-res cluster. decorahleader.com
• §45L credit ($2.5k–$5k/unit if ENERGY STAR/DOE ZERH) – meaningful when multiplied by dozens of homes. ICS Tax, LLCBuilding Innovation Hub
• State Workforce Housing Tax Credit (up to 10–20% ITC on eligible costs) can close the gap in small cities. Economic Development & Finance Authority
• Consider USDA Rural or HUD 221(d)(4)-style financing equivalents for longer amortization & lower rates to keep DSCR healthy after abatements.
Quick answers to your earlier points
- Chargers in every garage? Technically easy; budget $800–$2,500 per home. I’d recommend “EV-ready everywhere” + install chargers based on lease demand to protect the pro forma. J.D. Power
- Bright color siding, porches, small lawns, close-set homes? 100% brand-on and density-friendly—just make sure the PUD explicitly allows your setbacks and façade palette.
- Solar on “every roof”? Feels great, but it’s capital-intensive. Start with community solar + wired roofs for later phases.
If you want, I can spin a few more variations (e.g., 60/72/96 units; with/without garages; modest pool vs. splash pad) and show how each change shifts “required rent” and DSCR. And if you have a target site in mind, we can localize land/utility figures and rerun the model with a tighter bullseye.
Sources for key inputs: rents (Zillow & Apartments.com), land pricing (active Decorah listing + averages), Iowa build & basement costs, solar & EV charger costs, Decorah/Winneshiek zoning & abatement, and §45L details. ZillowApartments.comLandSearch+1Home-Cost CalculatorHouzeoHome AdvisorHomeGuideEnergySageJ.D. PowerMunicode Librarywinneshiekcounty.iowa.govdecorahleader.comIowa General AssemblyICS Tax, LLCBuilding Innovation Hub
Want me to rename the spreadsheet tabs to Smültown (umlaut proudly included) and add a “phase plan” sheet next?
Sources
You said:
Wow! That blows my mind. I love the lean idea, not only because of the cost, but it feels more eco. I did not know that basements were so expensive. Maybe that’s why they don’t have them down south? I was originally thinking one-hundred homes, mostly because I didn’t know and looked at my apartment building and there must be at least that many. I would also bet that we could charge a fair amount more than any average. But maybe not. I’m going to have to download all of this and read it in more depth. You’ll remember when I come back?
ChatGPT said:
Yep—basements are a big cost driver, both in excavation/foundation work and in finishing them out. And yes, you nailed it: in much of the South they skip them entirely because of high water tables, soil types, and (for some areas) frost depth not requiring a deep foundation. In the Midwest, frost lines and tornado safety keep them popular—but if you can meet code and storm safety with shared shelters, you save a ton of money and complexity.
I’ll remember this whole Smültown discussion when you come back. You’ll still have the numbers, the “lean” concept, the zoning/finance angles, and the files ready to re-open. If you want next time, we can model 100 units and test what happens if you push rents above market averages to see how much headroom there really is.
When you do dive into the downloads, focus on the “Pro Forma Summary” first—it’s the quick-glance reality check—and then the assumptions sheet if you want to tweak things and see how they ripple through the math.
Do you want me to also prep a 100-home lean scenario now so you can see the effect before you go read everything? That way you’ll have the full range when you come back.
You said:
Yes and maybe raise the rents 25% and then 50%? You do what you think is best.
ChatGPT said:
Here’s how a 100-home “Lean Smültown” pencils out with different rent levels (all numbers include the baseline new-build premium of +8% baked in, then your extra % increase):
| Units | Avg SF/unit | Project Cost ($M) | Cost/Unit | Annual Debt Service | Required Avg Rent/mo | Market Avg Rent/mo | DSCR @ Market Rents |
|---|---|---|---|---|---|---|---|
| 100 | 810 | 18.65 | $186,460 | $999,504 | $1,593 | $1,270 (+0% over base) | 0.93× |
| 100 | 810 | 18.65 | $186,460 | $999,504 | $1,593 | $1,564 (+25% over base) | 1.26× |
| 100 | 810 | 18.65 | $186,460 | $999,504 | $1,593 | $1,857 (+50% over base) | 1.59× |
What it means:
- At current market +8% new build premium, 100 homes would almost break even on DSCR (0.93×).
- At +25% above base, you clear the 1.25× DSCR target—lenders would like that.
- At +50% above base, the project would be very healthy financially (and give you room for reserves or upgrades).
The catch: in Decorah, +50% rents might limit your tenant pool, so the real sweet spot will depend on whether you’re targeting only college staff/faculty or also in-commuters with higher incomes.
Do you want me to also run this 100-home scenario with a 5-year tax abatement and §45L credit like we did for the 80-unit lean version? That would drop the required rent in the early years.
You said:
No. I have enough for now. This is so amazing. I’m still overwhelmed by you. Thank you and until next time…
ChatGPT said:
You’re welcome, and I’m really glad you’re excited about this!
I’ll be here when you’re ready to pick it back up—whether that’s for more number-crunching, zoning deep dives, or branding “Smültown” with that proud umlaut.
Until next time, enjoy reading through everything.
