The coastal town of Ipswich, Massachusetts, 30 miles north of Boston, has about 6,000 homes built over the course of five centuries. There are the typical cul-de-sac Colonials, the new townhouses, and both modest and massive waterfront properties. But Ipswich is also awash in historic homes—including roughly five dozen “First Period” houses built before 1725, more than any other community in the United States. Lately, the town’s antique houses have been popular with homebuyers, fetching the kinds of multimillion-dollar sales prices usually associated with new construction.

Ipswich Chief Assessor Mary-Louise Ireland isn’t sure whether it’s a temporary blip or the start of a trend. But she does know one thing: it’s making her team’s task of assigning fair and accurate property tax values to every home in town a bit more challenging.

After all, one of the biggest difficulties for a local tax assessor isn’t just making accurate property valuations—it’s doing so consistently, across all price points, home styles, and neighborhoods. If a $1 million Colonial is assessed at $950,000, for example—or 95 percent of its market value—then a $100,000 condo in the same district should be assessed at $95,000. When that ratio is consistent across a community’s price tiers, the valuations have what’s called vertical equity.

That’s tricky enough to achieve in a homogenous postwar suburb. But when 300-year-old saltboxes share the streets with new luxury townhomes, and storied houses get converted to character-rich condos, making equitable assessments across such a sundry assortment of housing styles gets even more challenging. “We’re three people,” says Ireland, “and we do all of the field work on our own.”

Now, Ireland’s small department is using an innovative—and free—new online tool from the Lincoln Institute of Land Policy to evaluate and interpret the vertical equity of their assessments. “We don’t have a lot of money for extra tools,” she says. “So having this has been fabulous.”  

Evaluating the Valuations

Getting assessments right across the board is crucial to a fair and equitable property tax. But accurately assessing very low- and high-priced properties is notoriously difficult, partly because there are fewer market sales in those brackets. And in recent years, researchers analyzing national data sets have found headline-worthy evidence that lower-priced homes are being over-assessed—and therefore overtaxed—relative to higher-priced properties nearby.

“If assessments are equitable, then low-, medium-, and high-priced properties are all assessed at the same level relative to the market,” says Lincoln Institute of Land Policy fellow Ron Rakow. “But even though it’s a fairly simple concept, vertical equity is really tricky to measure.” 

The International Association of Assessing Officers (IAAO) has two vertical equity standards in place to guide assessors, says Rakow—former commissioner of the City of Boston Assessing Department—but even those measures are imperfect. The price related differential is a simple ratio most assessors use, but Rakow says it can be imprecise; the coefficient of price related bias is a little more robust, but also more complex—it requires a type of analysis that many small departments don’t have the resources or expertise to conduct. 

“Because of the difficulty of measuring vertical equity, there’s no single best, definitive measure,” Rakow says. “So rather than just looking at one indicator, it’s better to look at several indicators to paint a more complete picture.”

Needless to say, that’s no simple undertaking. So the Lincoln Institute partnered with the nonprofit Center for Appraisal Research and Technology (CART) to develop a new online tool to help assessors measure and understand the vertical equity in their own valuations.

The browser-based vertical equity app, which is free to use, instantly analyzes property data that any local assessor already has on hand, evaluating it against six different measures of vertical equity and providing a detailed report. “We wanted to give assessors a tool where they can not only get these measures calculated out, but also get some assistance in interpreting them,” Rakow says.

The new tool, launched in September, simply requires users to upload a data set of assessment records, which are anonymized to protect the privacy of property owners. The tool then runs a calculation based on two main ingredients: time-adjusted sale prices and assessed values.

From there, assessors can see different illustrated measurements of vertical equity in their data set, with customized graphs and explanations, and can download a full PDF of the results.

“If you can upload an attachment to an email, you can now do these complex statistical quality control studies—you don’t have to have a PhD, you don’t even have to have programming experience,” says CART founder and research scientist Paul Bidanset. “There are a lot of different ways to do it that would have been more complicated—but we thought if we could meet people exactly where they were, we would be helping the most people.” (Read our profile of Bidanset, a former C. Lowell Harriss fellow at the Lincoln Institute.)

Ireland says she’s thrilled to have access to such a powerful tool. “It was super simple—I have everything in Excel spreadsheets anyway, and you only needed two columns,” she says. “I can use this really beautiful report to go before the Select Board and say, ‘OK, here’s the data to support what we’ve done.’”

The professional look of the report was impressive, Ireland says—and not something her department of three could have put together on their own with their limited budget. And the illustrated graphs aren’t just useful for communicating vertical equity data to non-assessors. Paired with contextual explanations of what each measurement means and how it’s calculated, they helped Ireland wrap her head around some of the more complex and novel metrics. “I’ve taken all the classes, and we’ve talked about [these measurements], but for some reason it really hit home for me seeing it all put together this way,” she says. 

Six Sides to Every Story

The tool provides results based on six approaches. The first looks at the commonly used assessment-to-sale ratio, which simply divides assessed values by their sale prices; the tool then sorts and charts those results into price deciles.

“We basically split all the sales into 10 bins—lowest-priced properties in the first bin and highest-priced properties in the tenth bin—and then we compare that ratio and see if it changes,” Rakow explains. “If we have proportional assessments, the ratio should be the same in each of those bins. But what we commonly see is that the assessment ratios tend to be a little bit higher for the low-priced properties than they are for high-priced properties.”

The coefficient of dispersion analysis plots out how far each property’s ratio is from the median. While that’s more commonly used as a measure of horizontal equity, Rakow says, it still reflects the overall quality of the assessments. “Generally speaking, if you have problems with vertical equity, you’re also probably going to have a pretty high coefficient of dispersion,” he says. 

The tool also calculates the price related differential, one of two standards the IAAO uses to measure vertical equity (a PRD between 0.98 and 1.03 indicates vertical equity, according to IAAO guidance); the coefficient of price related bias, which can help users understand patterns in assessment-to-sales ratios at higher price points; and Spearman’s rank-order correlation, which compares rankings of assessments and sales from lowest to highest.

Finally, the tool includes Gini coefficients, which have long been used to measure inequality in economics. It’s only fairly recently that the assessment profession has begun to apply the Gini ranking technique to analyze vertical equity. “We’re really excited about these,” Rakow says. The Gini ranking not only offers an overall indicator of equity in the assessments, “but it also can point to where in the price distribution you’re actually having problems,” Rakow says. “It’s great to know whether or not the assessment distribution is equitable or not, but it’s even more important, if it isn’t, to know where to start looking and where you may have some issues.”

While any one of these six measurements in isolation might provide an imperfect analysis of vertical equity, Rakow says, they offer a more complete picture when taken altogether. And the app can also help an assessor look more closely at specific data. “If you suspect that the issue may be in certain neighborhoods, or within certain housing styles, you could basically cull your sales file and just feed those types of properties into the app and see whether or not that is in fact the case, and how severe the problem is,” Rakow explains.

Ultimately, the developers of the tool hope that it will make it easier for assessors not just to understand vertical inequity, but to take steps to address it. In future iterations, Rakow would like to add diagnostic elements. One feature currently in development is a geographically weighted tool to highlight areas with the most significant divergences between market values and assessments. “So then you can zoom in and see what’s going on there,” he says. “Maybe there’s a certain style of house in that neighborhood that you’re not capturing right in the model, or maybe it’s very large homes that tend to be in that particular location versus the rest of the community.” 

This kind of data could also help assessors make the case for their municipalities to consider targeted tax relief policies, such as a homestead exemption, that can help make assessments more equitable.

Like any good technology, the tool will never truly be finished, Bidanset says: “It’ll always be changing and evolving as the industry evolves, and as we get more feedback, and as the industry comes up with new metrics and better statistics.”

Jon Gorey is a staff writer at the Lincoln Institute of Land Policy.

Lead image: Houses in Ipswich, Massachusetts. Credit: Leigh Mantoni-Stewart.

The Lincoln Institute provides a variety of early- and mid-career fellowship opportunities for researchers. In this series, we follow up with our fellows to learn more about their work.

Determining the value of property is a complex and often controversial job, but new tools are making it easier for appraisers to ensure the fairness of their work. Those tools include an approach developed by Paul Bidanset, a doctoral candidate at Ulster University in the United Kingdom and former C. Lowell Harriss Dissertation Fellow. The fellowship, named for a longtime Lincoln Institute of Land Policy board member and Columbia University economics professor, assists PhD students whose research complements the Lincoln Institute’s interests in land and tax policy. As founder and research scientist at the nonprofit Center for Appraisal Research and Technology, Bidanset has now advised officials from the United Kingdom to Moldova. He described his efforts to help democratize and modernize the appraisal field in this interview, which has been edited and condensed for clarity.

JON GOREY: What is the focus of your work, and how did your fellowship help advance that research?

PAUL BIDANSET: I came from a data science background, where I was forecasting anything people wanted—forecasting revenues based on advertising expenditures, forecasting pass-fail rates based on number of hours studied—anything where you could put in some inputs and try to forecast an output. That led into predictive algorithms for appraising property, specifically for property taxes—looking at recent sales and creating models that would estimate how much certain property characteristics determine what a property would sell for, then using those to appraise all the properties within a jurisdiction, so the government can tax them based on their market value.

There’s a quality control that we do in this industry that tests how accurate those models are, and not only if they’re accurate, but if we’re being consistently accurate across all properties. Are we being consistent? Are we being fair? Are we being equitable? A lot of research I do goes into making these predictive models more accurate and more consistent for taxpayers.

In this dissertation, I took an algorithm that was already being used in the industry that brought in a lot of really granular location data, so it’s much more sensitive to local fluctuations across neighborhoods and even within neighborhoods, and I modified it to not only be more accurate with regard to location, but also to the current time of the market. So making sure that old sales, for example, if they happened before COVID, weren’t counted the same way as recent sales.

The research is all done, and all the algorithms were actually improved as far as government standards and property tax standards and governing documents are concerned. I don’t like to brag, but the valuation oversight authority in the UK actually took this algorithm and used it to revalue properties in Wales. So it was cool to see this research taken out and actually used.

JG: What are you working on now, and what are you interested in working on next?

PB: I founded a think tank, it’s a 501(c)(3) called the Center for Appraisal Research and Technology. I’ve been working in Moldova, and in Romania currently; I’ve done some work in Estonia and Ukraine, and I’m starting to work in Asia as well with the Asian Development Bank and the World Bank. A lot of the stuff that I’m teaching or working with them on is more basic modeling and technology, so it’s not directly tied to my thesis or my dissertation, but I think it is a result of my experience in the doctoral program.

And recently our nonprofit partnered with the Lincoln Institute to create this vertical equity app dashboard that governments can use. So when they’re done with their valuations, they can upload their spreadsheets . . . to test to make sure that taxes are fair across those price points. You upload it, you click a couple buttons, and you get this nice generated report that breaks things down for you very simply. We’re looking to get that type of help in the hands of governments all around the world.

JG: What’s the most surprising thing you’ve learned in your research?

PB: I think the most interesting thing to me is it doesn’t matter where you are, the issues and questions are the same. I started in Norfolk, Virginia, working in a government office, that’s where I cut my teeth in this industry. But [there is] continuity from Norfolk, Virginia, to Chişinău, Moldova, to post-Soviet countries, to developing countries in Asia—it’s amazing how similar it all is, when you’re talking about relationships between the government and taxpayers, limited budgets, outdated software, staff being spread too thin. Even the questions that the taxpayers have when they come in, their questions, their protests—I mean, it’s copy and paste. It’s fascinating.

JG: What do you wish more people knew about the appraisal industry?

PB: I wish people knew how much people in local government—at least the governments that we work with—care, and how much they actually do. Because I don’t think people realize that. I used to work for a different nonprofit and when I tell people that we would host conferences where government practitioners would come to learn how to get better at valuing properties and do things more equitably, they’re like, ‘Governments [care] about that? I just thought they threw a dart at the highest number they could get away with.’ I think if people just knew how much your average government assessor cares, how much work goes into this, how much due diligence and continuing education and hard work . . . the majority of them are really trying hard to get better at this and do a good job for the community.

JG: When it comes to your work, what keeps you up at night? And what gives you hope?

PB: Something that keeps me up would be just how much people ignore good statistics and research. It’s very convenient and easy for people to just dismiss something because it doesn’t jibe with their preconceived notions.

Something that gives me hope? I would say the open source ethos. We don’t want to foster a consultancy dependence, we want to empower these countries with limited resources. So in Moldova, for example, we were teaching them how to use free open source software that they don’t have to pay for, and really put the power in their hands, which is going to help them hopefully develop faster and comprehensively across the entire country.

JG: What’s the best book you’ve read lately, or best show you’ve streamed?

PB: As far as shows go, Silicon Valley. I’m a huge Mike Judge fan. The book that I’m reading now is by Nassim Nicholas Taleb, it’s called Fooled By Randomness. He talks a lot about financial markets, but it’s really just a very pragmatic way to look at statistics and make sure we’re not drawing the wrong conclusions or putting false hope in certain things, which I think is massive when it comes to vertical equity and ratio studies. We’ve got to make sure that we’re not drawing false conclusions and thinking we’re good when we’re not, or vice versa. Because it’s a tough job as it is—we don’t need any more confusion.

Related Articles

Fellows in Focus: Rethinking Stormwater Management in the West

Fellows in Focus: Building Affordable Homeownership Opportunities in New Orleans

Fellows in Focus: Mapping Our Most Resilient Landscapes

Jon Gorey is a staff writer at the Lincoln Institute of Land Policy.

Lead image: Paul Bidanset in Beirut, Lebanon. Bidanset traveled to the city for a project with the Lincoln Institute and Beirut Urban Lab. Credit: Courtesy photo.