ATL With No A/C

It’s summertime in Atlanta. The temperatures haven’t gotten unbearable quite yet, but there’s no reason to expect anything but another re-run of our recent years of record-breaking highs. This heat is, of course, distributed unevenly across the city, being felt more intensely in places without tree canopy cover. And while most of us are able to retreat to the comfort of our air conditioned homes, not all Atlantans have the same luxury. The problem is that there has not previously been a reliable, granular dataset that allows researchers and policymakers to understand who might not have access to air conditioning and where they live.¹

The US Census Bureau’s new experimental Local Air Conditioning Estimates (or LACE) data seeks to fill this gap by modeling the number households without air conditioning of any kind, using a combination of variables drawn from the American Housing Survey and American Community Survey. According to the LACE dataset, just 3,657 households (or 0.8% of the countywide total) in Fulton County live without A/C, which seems like a reasonably good mark for a place where access to air conditioning could be the difference between life and death during a heat wave.²

But the LACE data is still limited in a few key ways. First, it’s only available down to the scale of the census tract, which prevents some of the more fine-grained analysis we like to do. Second, it doesn’t represent actual observed phenomena, but rather a modeled estimate, which is itself based on a fairly small sample of the population. Third and finally, it’s only capturing the general presence of air conditioning, without differentiating it by type or efficacy. And as anyone who’s lived in a home or apartment without central air conditioning can tell you, an ad-hoc window unit or a mini-split can’t always get the job done in the middle of an Atlanta summer.

Using a somewhat hidden piece of data in the Fulton County Tax Assessor’s CAMA (or Computer Assisted Mass Appraisal) supplementary file, we were able to approximate a slightly different version of the Census’ modeled data, identifying all of the residential parcels in the county that actually lack central A/C based on the value in the data’s “Heat” column. Essentially, any property that had a value for this field and wasn’t listed as having central heat and air conditioning was assumed to be a property without central air, the distribution of which can be seen below.

Using this building-level data, we can say that of the 320,457 residential parcels in Fulton County, 24,764 (or 7.7%) do not have central A/C installed. That number is nearly 7x greater than the LACE estimate of those without air conditioning at all, meaning that there’s a pretty big difference in how one might assess the severity of the problem here in Atlanta based on which data you’re using.³ Even more troubling is the fact that, as the map above and table below show, homes without central air are clearly clustered in historically Black neighborhoods on Atlanta’s west and south sides, largely mirroring the ever-familiar pattern of Atlanta’s ur-choropleth.⁴ And even further, these neighborhoods are already some of the most vulnerable to extreme heat in other ways, exacerbating the already serious concerns for future climate adaptation planning in the city.

Neighborhood	Properties Without Central A/C	Heat Vulnerability Score5
#1. Grove Park	726	9
#2. Oakland City	613	8
#3. Collier Heights	608	8
#4. Sylvan Hills***	533	10
#5. Westview	519	8
#6. Venetian Hills	496	6
#7. Cascade Avenue/Road***	438	10
#8. Center Hill***	423	10
#9. West End***	363	9
#10. Carroll Heights	354	5

Even though practically everything in American society runs downhill from racism, the causal mechanism here might not be so simple. We know, for instance, that air conditioning only became widespread in newly built homes in the late 1960s, a development that was itself responsible for the boom in sunbelt urban population during this time.⁶ Given that starting point, we can also double check this pattern against the distribution of homes built prior to 1965, which is also available (though not always entirely accurate) from the Fulton County Tax Assessor’s data.

Indeed, the geographic distribution of the 79,514 parcels built prior to 1965 largely mirrors the distribution of homes without central air, showing a clear connection, albeit with some caveats. As you might have noticed, that number dwarfs the number of homes without central air mentioned previously by a factor of more than three. This is because, 57,216 (or about 72%) of these pre-1965 properties actually do have central air installed. The remaining 22,298 homes built before 1965 make up the vast majority (about 90%) of the properties without central air.⁷

So while building age certainly plays an important role in shaping thermal inequity, clearly so too do the complex racial and class dynamics of housing and neighborhood change. Older homes can either be considered run-down, worthless and in need of wholesale replacement, or as diamonds in the rough waiting for investment so that they can be restored to their former glory. Which direction any given property or neighborhood goes, however, is predominantly shaped by the existing racial and class makeup, leading to gentrification and displacement for some neighborhoods and persistent disinvestment for others.

1. This, of course, does not account for our homeless neighbors who have no reliable shelter from the elements, whether heat, cold, precipitation or anything else.[↩]
2. Heat exposure is the #1 cause of weather-related fatalities in the United States.[↩]
3. Which unfortunately just confirms that we still don’t have the kind of singular reliable dataset needed to create consistent measures of thermal inequity across space and time.[↩]
4. The fact the tax assessor-based measure generally mirrors the distribution seen at the tract level in the LACE data provides some additional confidence in the validity of this measurement.[↩]
5. This heat vulnerability score comes from UrbanAdapt LLC’s 2023 Atlanta Heat and Flood Risk Assessment. It is worth noting, however, that the heat vulnerability score already takes into account the same data on properties without central air. So while it’s a useful indicator of the broader vulnerability to extreme heat across neighborhoods, comparing it to our tax assessor-based measure raises some multicollinearity issues (for all of our nerds out there).[↩]
6. Which in turn helped to produce the realignment of 20^th century American politics towards the south, not to mention further contributing to greenhouse gas emissions and anthropogenic climate change, which then requires us to use the A/C even more.[↩]
7. Though this of course leaves an anomalous 2,466 properties across the county without central air that were built after 1965, with dozens as recently as the last few years.[↩]