Do The Math: South City Residential

Tweet about this on TwitterShare on Facebook0Share on Reddit0Share on LinkedIn0Print this pageEmail this to someone

If this kind of analysis excites you, it’s similar to what a firm called Urban3 out of Asheville, NC led by Joe Minicozzi does. We’re trying to get them to do an analysis of St. Louis. That takes considerable funding. If that’s something you’re interested in helping with please email me at richard at

I’ve mainly been examining commercial corridors around St. Louis so far. While looking at a heat map of assessed value per acre, I noticed a cool spot in south St. Louis city, in the Southampton neighborhood, that warranted a closer look. There seems to be a negative correlation between frontage and value per acre. Let’s look at the area bounded by Chippewa, Sulphur, Landsdowne, and Brannon. Let’s do the math!

Leaving out the four-families, commercial properties, and the school, the study area is 66.44 acres with total assessed value of $11.2M or $169k per acre. The average assessed value per parcel is $21k. The area was 90% built-out by 1950.

Let’s separate the parcels into two groups. The “svelte” group are parcels with less than 43.5 feet of frontage and the “fat” group with more frontage. Per parcel the fatties average $21,864, and the sveltes are at an average of $20,446.

{Examples of the “sveltes” – Google Streetview}

What is striking is the contrast between the two groups in assessed value per acre. The sveltes come in at $196,894 per acre while the fatties come in at $123,322 per acre. That’s a difference of $73,572. Had the entire study area been of the svelte variety the total assessed value would be $13,082,560, $1,878,330 or 16.8% more. That’s $142,471 in potential total property taxes each year. The city made a poor development decision when these were platted many decades ago, thus have come at a high opportunity cost. Worse many of the entities that receive property taxes had no say in the development decision (Some didn’t exist like the ZMD).

{Examples of the “fatties” not pulling their weight – Google Streetview}

This amounts to a hidden subsidy for the fatties. With more frontage they have more feet of street, more feet of water and sewer pipes, more feet of power lines, more streetlights, service providers have to travel further, etc. One saving grace is that a couple of the blocks don’t have alleys to maintain, though this ruins the aesthetic of the street in my opinion.

{Recent sales indicate that the market doesn’t heavily favor the fatties.}

What is done is done in the study area, but in future development decisions, the city, planners, neighborhood stakeholders, and developers should consider the impact of large frontage parcels. It is crucial for the city. Skinny houses on wide lots are too common in recent developmetns. Decision makers need to think of the buildings and land as a portfolio, consider which ones are higher returning and take a long view.

{Skinny houses on wide lots on 4200 N Market. Those side lots come at a high cost the city. – Google Streetview}

A developer will move on, but the city is a corporation that will endure. Even if the market doesn’t value the land highly, the city (and any city) must treat it so, especially if it can’t add land area. Without higher productivity land uses we’ll never be able to afford the level of infrastructure and quality of services we desire.

Tweet about this on TwitterShare on Facebook0Share on Reddit0Share on LinkedIn0Print this pageEmail this to someone
  • gmichaud

    How about a do the math for how many steps people have to get to transit, or to a commercial enterprise or other public place like a park? The distances traveled would be included in the lot size discussion above. And if financial value was the goal then the impact of walkability on prices could be included. In fact the most important point mentioned above isn’t value but that the policies of lot widths should be under discussion. I would argue that the future, maybe sooner than we think, will include more row houses like in Soulard for the energy efficiency.
    In recent years communities like U City and Webster have changed lot size minimum requirements from 30 to 50 feet, invalidating the use of many vacant lots until they are combined with others. This is a new standard, why is it better than an old standard?
    I do like the point of the fatties not pulling their weight in regard to utilities. Taking that one step further the far suburbs like Chesterfield are subsidized by those living in the city and the inner ring of suburbs.

    • rgbose

      Good ideas on the distances to amenities. Someone has already done it for transit. Just drag the marker to your starting point. I suppose the next step would be to take into account how many people live in an area and multiply that by the area they can access in a certain amount of time.

      Depends on what the city’s goals are. I’ll speculate and guess U City and Webster want to maintain a housing price floor and figure a larger lot will command a bigger, fancier, and therefore more valuable home be built. A small lot might make for a small cheaper house and thus a lower-income person might afford it. Webster also forbids renting apt-above th garage apts to non-related people, blocking a path to cheaper housing, impeding the market force of higher housing prices pushing towards density.

      Not sure if/how Ameren separates areas. I’ve wondered the same thing. Charter charges the same whether you’re on a small or big lot. Signal strength drops over distance so I’d think fewer homes could be served per foot of cable and per location that houses their equipment.

  • Aristippus

    Looking at the scatterplot at the top, I think there’s an issue with the density of points in a few ranges that skews the analysis. The points from ~33′ to ~37′ skew highly above the trend line while ~42′ to ~47′ are skew way below the line. After 50′, it looks like the (few) big fatties actually do rather well by outperforming the trend line significantly. I don’t know these neighborhoods, so I can’t say for certain, but I think something else could be in play as causing this variation. The theory behind this model makes sense and the R^2 value looks solid, but I think RMSD as opposed to R^2 would show that there’s a weaker fit between the regression model and the data.

    • rgbose

      Yeah, I was thinking perhaps averaging the values at each amount of feet and then fitting might be more illustrative. It does seem to flatten out once you get to the larger lots. Maybe a linear fit isn’t the right function type. Trouble is a polynomial function with more terms will always fit better (R^2 goes up), but those additional terms might not be statistically significant. There is a way to test for that. I’ll have to dig into my stats and econometrics notes from long ago! Also your eye is a poor regression device. That was the first thing we learned in stats class!

  • D. Boone

    My professional opinion is that yall need to generally chill out on these analyses of assessed values but in this one you got it right. I wouldn’t use this to come to any conclusions about the market value of ranch houses versus bungalows or whatever, but I think most would agree that people who consume more services should pay for them. Unfortunately there are about a billion things about the city’s property tax assessment model I would fix before this.

    • Dude

      Agreed. I also find it interesting that the author chose not to mention the big differences in building materials and quality used. A full masonry home with thicker walls will be vauled higher than any vinyl sided or brick veneer home regardless of lot size. Also aesthetics will come into play with the neighborhood comps. The majority of buyers in the city value full masonry over vinyl therefore they have a higher selling price which leads to a higher assessment. I’m sure many buyers would prefer a larger lot but pay more for the “sveltes” for their full masonry. A lot of these “Do the Math” articles fail to see the bigger picture.

      • rgbose

        Not sure I see your point. In this area the masonry sveltes are both assessed and sell about the same as the siding-clad with driveway fatties. My point is from the city’s perspective the fatties aren’t puling their weight and this should be considered in development decisions.

        The highest performing house in the area on a per acre evaluation is a home built in 2013 on a 50′ lot. If any other wide lots come up for redevelopment, the city should be interested in encouraging a more valuable home be built than what existed there before or that lots be replatted into thinner ones if practical. The ramifications for the city last for many decades as we see here for the past 6+ decades.

    • matimal

      You get paid for opinions like this?

  • JZ71

    If 40′ is “better” than 50′ or 60′ or 75′, would 30′ or 25′ of frontage (for a single family home) be even “better”? How about 20′? I agree, “skinny houses on wide lots” both look silly and are inefficient, but the other half of the equation is what sells – if you look at the construction dates, the svelte ones date to the early 20th century, while fatties date to the middle. People’s priorities changed, and developers responded with attached garages and no alleys.

    In a (your?) perfect, rational, world, we’d all, apparently, live either in shotgun houses or in multi-story, multi-family or mixed-use structures. Unfortunately, multi-story, mixed-use buildings cost more per square foot to build (structure, elevators, fire protection) than typical tract housing, and most people want the biggest single-family home they can afford, even if they have to sacrifice both quality and density (based on actual sales in the real world)! Shotgun houses would the right answer given your criteria, but I doubt that there would be many buyers for them – if you answer a question no one is asking, how successful will you be?

    Another criteria worth investigating/comparing would be front setbacks. We have side setbacks to minimize fire spread. We have front setbacks purely for aesthetics. Front yards are purely decorative, and minimizing them would increase density with few negative (perceived or real) impacts, unlike reducing the size of rear or side yards. (In this example, the fatties also have larger front yards than the svelte homes, further reducing their density.)

    Finally, where do duplexes and zero lot line homes fit into the discussion? Urban side yards have more negatives than positives – they provide light, ventilation and fire separation, but, depending on window placement, can create both privacy and security challenges. Assuming equal face widths, say 24′ per structure, you can do things at least three different ways in the same 120′ distance:

    – four single units with “standard” 3′ side yards, each, on both sides

    – four zero lot line single units with 6′ side yards, each, on one side

    – two duplex units with 6′ side yards, each (on one side, obviously)

    The only real advantage of the “standard” answer is cross ventilation, since you can do windows on both sides. But given the ubiquity of central air conditioning, especially in St. Louis, is this really still a priority? Would more buyers choose a narrow-but-usable side yard over an unusable, half-as-wide gangway?

    Bottom line, density takes many forms, and it’s more than just lot width, it’s how intensively each lot is used, in total. To have “more productive” land uses, the city needs to provide a range of attractive solutions, not just say everything must be narrower. And our biggest challenge isn’t 36′ vs. 60′, it’s the fact that we’ve lost thousands and thousands of structures, on lots of all widths, and vacant land generates far less tax revenue than occupied land!

    Land values drive density – expensive land naturally leads to more-dense development. Conversely, large amounts of unwanted, cheap, vacant, land leads to less intensive uses and less-dense development. The same negatives that apply to “fatties” – “they have more feet of street, more feet of water and sewer pipes, more feet of power lines, more streetlights, service providers have to travel further, etc.” – applies tenfold to vacant lots. Just filling in all our vacant lots, with whatever fits, would go a long way toward “fixing” our desire to “be able to afford the level of infrastructure and quality of services we desire”!

    • rgbose

      Remember though that what sells is a reflection not only of the market, but also the policies and subsidies that perturb the market. If frontage were a significant portion of property taxes don’t you think there would have been fewer fatties in this area?

      • Alex Ihnen

        Last year I sold my 25ft frontage home in The Grove at a price I was happy with. And I way overpaid for it in 2006. Most homes that sell in the city are narrow – perhaps the market in speaking.

        • JZ71

          I don’t disagree that there is a market for city houses in good condition in good neighborhoods – EVERY real estate transaction involves two parties agreeing on an accetable price. My point was there is no one, “right”, answer, just many shades of grey, especially when it comes to new construction. And the reason why many city homes are narrow is a reflection of when their subdivisions were platted and how people relied on public transit AT THAT TIME – in the first half of the 20th century, people relied on public transit far more than they did in the second half (and our built environment reflects that). Until more of the St. Louis region embraces public transit (both funding and actually using it), and quits looking at it as something just for losers and poor people, the SOV will continue to drive our development patterns. And, for better or for worse, even though there is a market for narrow lots and narrow homes in the city, there continues to be a much larger market for wider homes in the surrounding counties.

          • matimal

            There IS an most economically beneficial answer for St. Louis. That is the point of this article.

          • JZ71

            Huh? You seem to be making the one-size-fits-all argument. The “most economically beneficial answer for St. Louis” is to provide a range of attractive options, not just one, theoretically “right” one, IF that means more residents, in total. A vacant, unsold, low-taxed, potentially-LRA-owned, lot, of any width, is a far greater drain on our resources than an occupied parcel. The aging sewers, the cast iron water mains and the crumbling streets are there, and they need to be maintained and, eventually, replaced. With low property taxes, no water bill and no earnings taxes, vacant land does very little to pay for maintaining our basic infrastructure.

            The issue in St. Louis is not an academic analysis like this one, the issue is simple supply (high) and demand (low), in most residential areas in the city. If you want more density, one needs to look no futher than Kirkwood, or one can look at Chicago, Houston, Denver, DC or NYC to see where high demands = high land values = more-dense redevelopment. Small single family on large lots are being replaced by larger, many times multi-family, structures on smaller lots.

          • matimal

            The most economical benefit for St. Louis is that which increases property values and tax income most. If that involves a “range” of choices, then so be it. St. Louis may NOT need to maintain all of it’s infrastructure. It may make more sense to put all our eggs in one basket to show how it can work and to create a model for other neighbors. Think of it as a ‘local stimulus’ plan.
            Cincinnati has focused public money on it’s downtown much more than St. Louis and it’s been a huge success. So successful that suburbanites outside of Cincinnati are becoming resentful.

      • JZ71

        Sure, frontage could be an explicit factor in property taxes (and it already is, implicitly, in percieved values). But I think you’re missing the real problem – people choose to drive (and choose not to walk or use public transit) for two big reasons, comfort and convenience, and economics plays only a tiny role in individual decisions. As long as parking remains free (most places), monthly parking costs less (downtown) than a monthly transit pass, and it takes 3+ times as long to get from Point A to Point B, creating walkable communities won’t encourage people to drive much less (and we can do nothing about our weather).

        When it comes to new development, the cost of distributing utilities is included, directly, in the cost of each lot. It’s the replacement costs, for century-old distribution systems, that the city is struggling with. (And as was noted in the original post, the cost of maintaining alleys also needs to be factored into any analysis, something the “fatties” avoid.) The real issue isn’t that it’s going to cost X dollars to maintain the system, it’s that no one wants to pay for it, especially since costs have been avoided for decades (not budgeted for, not billed, not collected and not escrowed), and now current users are left stuck paying for the “free ride” that our parents and grandparents received! The incremental cost of 20′ or 30′ of wire or pipe, the final distribution, spread over decades, is miniscule. The cost of treatment or production is much higher, so the only way to limit that would be to limit population, not lot size.

  • matimal

    If the secret is density, shouldn’t St. Louis be focusing it’s resources on the most dense areas to make them the most efficient areas? Wouldn’t this strategy be the quickest way to build the value of property and therefore the tax base of St. Louis? Isn’t this what seems to be happening of it’s own accord in the CWE now? How can the city accelerate this even more without being accused of ‘abandoning’ other areas, mostly north city?

    • Bernard Finucane

      Maybe if density is all that matters maybe the city should focus on the least dense areas suitable for dense development.

      • matimal

        but wouldn’t that approach take much longer to produce increases in property value and property tax income and decreases in the per household cost of providing services?

        • Bernard Finucane

          I don’t follow.

          • matimal

            It’s comparative advantage. If new public investment in Old North St. Louis and CWE increases property values in both places by 5%, but the total value of property in the CWE is worth 20 times the total value of property in ONSL, then it would be of a much greater benefit to St. Louis City. 5% of the value of CWE is worth vastly more than 5% of the value of ONLS. If St. Louis city invested EVERYTHING in CWE and nothing in ONLS, it would receive an even greater net benefit to its tax income at no additional expenditure.

  • Rob M

    The math behind why it is a bad idea to build suburban houses in an urban environment.

    • Bradlykf

      Those “fatties” houses also don’t have basements and are generally less desirable in the area.

      I think the issue is really that in these neighborhoods (I live in Southampton) the “svelte” 30’s construction brick homes are what people want.

      Look at St Louis Hills Estates. Those ranch style with attached garage homes on wide lots would fit your designation as a fatty but those houses sell for well above everything else in the area.

      Correlation is not causation

      • rgbose

        Yes it’s not just about frontage, a very valuable home can take up a bit more land because it’ll pay more. That’s why a $/acre or $/ft of frontage is a better metric than $ per parcel in my opinion.

        • Andy Struckhoff

          Would love to see this analysis applied to St. Louis Hills, and the Hill. Also, how about doing a real tax $$ per parcel analysis and a real tax $$ per front-foot analysis?

          • rgbose

            In the city multiply the assessed value by 0.075874 plus any special district taxes. In the county total property tax levies vary a lot. Here’s the voluminous rate book. That’s why I’ve been using assessed value instead of actual taxes. If I can get the Urban3 study funded we’ll definitely want to overlay the tax book.

            Here’s Eureka’s breakdown.

          • Eddie in NorCal

            see reason #7 in this article about the demise of the starter home


          • JZ71


          • rgbose

            Here’s The Hill’s single families. There’s probably a few commercial spots included unintentionally.
            Count: 1008
            Minimum: $20,761
            Maximum: $750,547
            Sum: $20,329,550
            Per Acre of entire 112.64 acres: $180,482
            162 parcels have more than 43 ft of frontage. There are a few listed as 0 so there’s a handful more.

      • matimal

        Correlation means that there MIGHT be a causative relationship.