The Challenges of Building Rental Housing

SPENCER GARDNER is A STRONG TOWNS MEMBER WHO LIVES IN MADISON, WI.  Today he details his experience attempting to build an accessory dwelling unit on his property and why he ultimately decided against it.

Image from  YouTube

Image from YouTube

Much of the groundwork for this post was laid during the recent Strong Towns week dedicated to federal housing policy. If you want to learn more about why rental markets around the country are borderline dysfunctional, you’ll find some good answers there.

Today, I’d like to share my story about considering building a rental unit next to my house and why I ultimately have decided against it. While my case is anecdotal, it is clear that whatever the reason, the fact that not enough people are building rental units here in Madison, WI has created a significant housing crunch. My case is less about federal policy; it’s about how zoning is often explicitly biased against renters. And while there’s a lot of shared history between federal and local land use policy, the dynamic for reform is very different.

Madison prides itself (rightly or wrongly) on its forward-thinking governance and highly educated populace. Several years ago, this outlook resulted in a complete rewrite of the city’s zoning code. It was a huge undertaking and I can attest that the result is a far more sensible land use regime than any other place I’ve lived in or researched across the country.

I’ll get to why zoning is an obstacle shortly, but I’d first like to point out that the city got a lot of things right: no parking minimums in certain zones, legalizing what’s already there, and most importantly for my situation, official acceptance of accessory dwelling units.

Accessory dwelling units (I’ll call them granny flats from here on out because it sounds less jargony) are additional housing units that reside on the property of another house. In most cases, you can imagine taking a single family home with too much acreage and setting a smaller, more modest home in the back yard. This is classic missing middle housing. In a city where vacancy rates are effectively the time it takes for one tenant to move out and the next to move in, granny flats offer a relatively unintrusive way to boost the number of housing units in the city without adding new roads, sewers, and other attendant long-term liabilities.

Granny flats offer another significant benefit: as a homeowner, a granny flat on my land can turn otherwise unproductive space into a significant income generator. This was my primary concern when I started to consider the possibility sometime last year. Although we are able to make ends meet, the prospect of capturing some additional income by renting out an underutilized portion of our property is very attractive. Plus, as their name implies, granny units provide the possibility for housing aging family members who wish for some autonomy, but need some care. Although grandparents on both sides of our family are relatively young and healthy currently, it would be comforting to know that we could accommodate them if needed.

Given the enormous benefits and relatively minimal downside, what is keeping a person like me from building a granny flat? For all the things Madison got right in its zoning code, there’s one significant provision that completely spoils granny flats for me: on any lot that receives approval for a granny flat, the primary house must remain owner-occupied. In other words, if I build a granny flat, I am forced to either stay here in my house permanently, or sell upon moving elsewhere.

If I go in for the significant investment of building a granny flat, I will have severely restricted my options down the road.

This is a big deal to me because it violates one of Saint Nassim’s fundamental rules of antifragility: optionality. If I go in for the significant investment of building a granny flat, I will have severely restricted my options down the road. This is not an insignificant calculation; one of my best friends just moved across the country after an unexpected job change. Rather than quickly sell his house at a loss, he decided to rent it out and his mortgage is easily covered by the rental income. If I build the granny flat and find myself in a similar situation, I will likely not recover the costs of my significant investment, particularly given that many buyers in a neighborhood like mine are likely to view the granny flat as a liability, not an asset. If I stay put long enough, the financial math pencils out, but it’s not a risk that I’m willing to take.

I think the most vexing part of this is the fact that a zoning code needs to distinguish between renters and owners at all. In fact, it almost seems like a violation of basic human rights. We don’t condone discrimination on the basis of skin color, creed, or any other protected class, but for some reason it’s OK for a municipal zoning code to explicitly draw a distinction between people who own property and people who don’t? I fail to see how there’s a compelling government interest in ensuring that if I build a granny flat there are not renters occupying my house.

So in the end, we’re not building a granny flat. We’re getting chickens instead (which were legalized under the zoning rewrite). We’ll miss that extra income, but at least we’ll have an abundance of fresh eggs.

(Top photo from Jay Austin)

Related stories

About the Author

Spencer Gardner is a transportation planner based in Madison, WI. He spends his spare time chasing his children, riding bikes, doing hobbyist computer programming, and very occasionally writing about urban issues. You can read his thoughts about transportation at