Assessments and Property Taxes in Albany

Property taxes are Albany's primary source of revenue. Without them our city couldn't function. They're also a major expense for city residents, whether we're paying those taxes directly or indirectly through rent. In this series of articles we'll try to gain a better understanding of Albany's property tax system, how it works, and what problems it has in practice.

The mechanics of property taxes are simple. They get determined by taking the value of every property in the city (its assessment) and then charging each owner a percentage (the tax rate) of that value every year. There are some other complications (e.g. there are different rates for residential, commercial, and non-profit properties) but for now let's ignore all that.

At first blush, the system sounds pretty fair and straightforward: properties that are larger and more expensive pay more to support the city. Let's take a closer look at a single block and see how well that holds up.

The Block

We'll be looking at the block bounded by Broadway, N. Pearl, Wilson, and Livingston. It's an interesting mix of vacant lots and active uses, and it's located in a historically important area that is once again up and coming. This block wasn't cherry-picked, it's the first we looked at. We expect that similar results would be found for almost any block in the city. Below are a satellite image of the block and the city's tax map for it.

Tax maps allow you to see each lot that the city assesses and how those lots relate to each other. The maps also include useful information like the tax id of each lot, various dimensions, etc. To make it easier to see what's what, here's the tax map over the satellite map:

Assessments

This block has 33 separate lots. The block's total assessed value was $3,366,260 in 2015. The least expensive property was the vacant land at 84 Livingston, assessed at $5,000. The most expensive was the commercial building at 744 Broad, assessed at $361,900. Here's the tax map again, with properties shaded by their assessed value (darker is more expensive):

Let's see what changed in 2016. The overall assessment for the block increased by $230,000 to $3,597,700. The least expensive property is 84 Livingston again, but that assessment was halved to $2,500. The most expensive property is now the parking lot at 776 Broadway, with an increased assessment of $572,200. Here's the updated map:

Comparing the two maps emphasizes changes that occurred with this block's reassessment. The most visible is at 744 Broadway (corner with Wilson). In 2015 that building had the highest assessment on the block, over $360,000. In 2016 the assessment is only $40,000, a fraction of what the other improved lots are valued at. It's even lower than the assessment of the similarly sized parking lot at 191 N Pearl.

In 2015 the buildings at 221-229 N Pearl had vastly different assessments, from $40,000 at the low end to almost $170,000 at the top, giving them a zebra-like appearance on the map. In 2016 they range from $115,000 to $178,000 and the increase in parity is visually obvious - they're much more uniform in color.

Another obvious change happens at 76 and 78 Livingston. In 2015 those empty lots were valued at $25,000 and $75,000, far above their neighbors. This year they've been updated to $4,500 and $4,200 respectively, which is more in line with the neighboring lots.

All of these changes highlight a big problem with Albany's tax system: assessments are very difficult to reason about and to get right. Subjective changes like these need to happen every year. It's impossible to estimate the value of every property correctly: how much more is a building worth if it has 2½ baths instead of 2? What about 3 bedrooms instead of 4? The process is also opaque to the public: where do we go to find out why 744 Broadway lost 89% of it's value last year? Even the most well-intentioned assessor has a nearly insurmountable task to deal with.

Incentives

The colorization above is not necessarily the most useful for analysis. It makes it obvious that most of the block's value is concentrated at 776 Broad, but that's to be expected. We could have just as easily identified it as the most valuable lot by its size alone.

Let's try to adjust for the intrinsic value of each lot so we can see how the assessments vary relative to that baseline. There are a number of factors that affect the value of a lot (e.g. which street it faces, its size, its orientation, its shape, the state of neighboring lots) but for this analysis we'll use the length of each lot that is touching a street, or its frontage, as a proxy for value. This is simplistic, but the data was readily available and it does a decent job. Most lots in this block are similar in depth and shape so frontage mostly mirrors land area. Choosing frontage as our metric does overvalue corner lots somewhat (although they are more valuable than area alone would suggest) and does a poor job of valuing strangely shaped lots (luckily for us the only real outlier here is 243 Pearl).

The following map is for 2016 again, but this time the assessments have been normalized by dividing them by the frontage of each lot. Darker colors mean higher assessments relative to the land value.

The results here are largely as expected. Vacant / parking lots have low assessments compared to their land value and buildings in active use have high ones. One glaring exception is again 744 Broadway. The new $40,000 assessment for that property seems out of line with other improved properties on the block.

While not surprising, the results do a great job of highlighting a second major problem with Albany's tax system. When land is put to good use it's highly taxed and when it's left empty the tax bill is low. The incentives created by property taxes are out of line with the city's goals. We generally desire new development, but we punish valuable new properties with higher taxes. We like it when our neighbors maintain their buildings, but we give tax breaks to those who let their properties go to seed. We want vibrancy and activity, but we reward vacant buildings and empty lots.

Speculation

Why does this block (and Albany in general) have so many vacant lots and abandoned buildings? The incentives problem mentioned above hints at one cause, but another huge problem is real estate speculation. If the owner of 76 Livingston were to put that vacant land to productive use, say by adding a mixed-use building with retail and apartments, then the properties at 74 and 78 Livingston would dramatically increase in value. It's obviously much more valuable to have land next to a vibrant building than land next to a vacant lot. What's more, the owners of numbers 74 and 78 wouldn't need to do any work at all to have their land increase in value - the work of their neighbor makes them a profit.

Real estate investing in blighted areas like this often ends up resembling a high stakes game of chicken - nobody wants to make the first move. Speculators buy up land for cheap and sit on it, hoping that through the work of others the block will "turn around". When it does, they realize a gain. This behavior is obviously detrimental to the city. When everybody is sitting and waiting nothing will ever improve.

The problem of speculation, however, is implicitly encouraged by the city's tax structure. Since we assess vacant buildings and unused land at such low levels, speculators can afford to hold those properties for long periods of time without paying much of anything in taxes. What little they pay is a worthwhile investment in the rare case that somebody else's work adds value to their land.

Next time

The three problems we've highlighted above might not be news to some people. In fact, they're well known enough that the government already has mechanisms in place for mitigating each of those problems. In the next article we look at those mechanisms and examine their effectiveness.

Comments

Thoughtful comments on this article are greatly appreciated! Email your comments or questions to albanyargus@gmail.com and we'll post them here. We won't post your email address.

Appendix

The following table was manually transcribed from the 2015 assessment roll, 2016 tentative assessment roll, and the tax map. Land value figures are from the assessment rolls. As it turns out, they're largely useless - it seems that the land value figures are dramatically higher for active uses than vacant land, even for seemingly identical lots of land. I'm not sure how the assessor's office computes those figures, but whatever they're doing makes no sense. C'est la vie. Full value figures are also from the assessment rolls. These figures are exclusive of exemptions (e.g. star, veteran's exemptions). Frontage was computed by summing values from the tax map.

If you notice any transcription errors in the table below please let us know! It would be great if this data were made available in a machine-friendly format so this analysis could be done on a larger scale.

The table is also available as a CSV, here.

Address Tax ID ’15 Land Value ’15 Full Value ’16 Land Value ’16 Full Value Frontage
68 L 10 12700 63400 13400 67000 25.01
70 L 9 14300 55000 11600 58000 24.96
72 L 8 8000 8000 9100 9100 24.96
74 L 7 7500 7500 4500 4500 24.96
76 L 6 25000 25000 4500 4500 26
78 L 5 75000 75000 4200 4200 28.12
80 L 4 8000 8000 4200 4200 28.13
82 L 3 6000 6000 3000 3000 20.08
84 L 2 5000 5000 2500 2500 20.02
191 P 30 61800 61800 61800 61800 198.16
199 P 34 10200 10200 3200 3200 20.38
201 P 35 1500 155600 45800 229000 16.55
203 P 36 31100 155600 36000 180000 16.52
205 P 37 28200 140900 29400 147000 15.46
207 P 38 21600 108100 22400 112000 15.46
207.5 P 39 32200 152079 38200 191000 15
221 P 46 22900 114500 27800 139000 22
223 P 47 8200 40900 32800 160000 20
225 P 48 35300 125000 26600 133000 20.39
227 P 49 17900 89400 23000 115000 20.35
229 P 50 33700 168300 35600 178000 20.67
231 P 51 28700 143300 31800 159000 36
235 P 52 12000 12000 13400 13400 34
235.5 P 53 13800 13043 4300 4300 14.83
237 P 54 28800 124600 31800 159000 19.38
239 P 55 28500 142600 31200 156000 19.74
241 P 56 9200 87700 2900 85000 20
243 P 57 27100 135500 31800 159000 159.59
5 W 27 56700 212665 45800 229000 20.29
744 B 26 60300 361900 8000 40000 172.93
750 B 23 42500 236673 34700 180100 22.07
752 B 22 20000 21000 34700 34700 22.07
776 B 15 300000 300000 572200 572200 569.14