Interview: Will Jarvis, CEO at ValueBase
Interview: Will Jarvis, CEO at ValueBase
Or Henry George and a couple startup guys walk into a bar
The following is a conversation with Will Jarvis, founder and CEO @ ValueBase. ValueBase is on a mission to bring property assessment into the future – and enable policies like land value taxation (LVT) along the way.
For those unfamiliar with LVT or Georgism, it’s enough to say that taxing the value of land (as opposed to both land and the structures on top) is a pro-growth policy reform with a long history of advocacy behind it.
Separating out land value from structural value, though, is challenging and provides one of the more common criticisms of Georgist reform. We talk about all this and more below.
What Does ValueBase Do and Why Does Land Valuation Matter?
Jeff Fong: Will, thanks for taking the time today. Just to get things started, can you explain what ValueBase does? What problem do y’all solve?
Will Jarvis: Great question. The problem ValueBase solves – for a variety of different customers – is determining what a property would sell for if it were sold today (or in any given time period).
We determine the market value of a property, what it would sell for in a market that has a couple of different buyers. And then for us, what is the land value of the property? And what's the value of the building worth as well.
There's a third thing in there, which is the option value, but that also gets into the weeds. But that’s why we break it out as land value, the present value, and the total overall market value of any kind of given property.
JF: Why would we want to disaggregate those (land, structure, option) different values?
WJ: Well, there's a couple different reasons. If you're a Georgist interested in land value taxes or urban policy, understanding the locational value of the property is incredibly important. It’s also useful for insurers.
And in the context of cities or policy, understanding land value is incredibly important. You can understand the policy implications of different choices you might make.
So if you want to build a new mass transit system, you want to understand the uplifted [land] value from that mass transit system. And you can actually understand how you can improve the value of your city in different ways.
That's been the big barrier for Georgism in the past – actually getting good assessments. If you don't have good assessments, or you can't accurately do that at scale, you can't make it equitable, you can never actually implement the policy.
JF: You just said something interesting. You gave the example of building public infrastructure, I think you mentioned building a transit system, specifically.
In theory, if you can properly assess land values, you can use those to model out how proposed public infrastructure will increase land values and therefore tax revenues (under a regime that taxes land values)?
WJ: Yeah! How investments might possibly increase tax revenue, but other things as well. For example, we had HUD reach out to us recently, after the Maui fires. They wanted to be able to provide people an understanding of what the actual value of their land is worth now that the improvements are all gone.
JF: Super interesting. That perhaps raises the question…who exactly are your customers? I’m assuming folks in county assessors’ offices?
WJ: Absolutely. Our core customers and our core focus are the 15,000 valuation jurisdictions in the United States that set values for property taxes.
Currently, it's a very manual process to value properties. Most jurisdictions, about 85% - 90% of them, use a methodology called sales adjusted cost. That involves estimating the land value for every parcel, estimating the value of reconstructing the improvement, and then depreciating that based on the age. It's very manual and very time intensive. We use statistical modeling and valuation models to do that much more quickly.
It’s also an industry that has seen massive waves of retirement over the last five years. And so there's a problem getting enough talent to actually do this kind of work. We automate a lot of things – like data collection – to help people be more efficient and spend more time on higher leverage tasks.
Productizing Property Assessment
JF: In terms of how this gets done today, is there a typical proprietary software that everybody uses? Or does the degree of standardization just boil down to how everyone does the math?
WJ: The latter. There’s a traditional method of going about this that most jurisdictions use.
JF: It sounds like, from a product perspective, y’all are providing not only more accurate / consistent valuations, but also a faster / easier way to get them, yes?
WJ: That's correct.
JF: When you talk to prospective customers, is there some amount of education you have to do? Or is it clear that ValueBase is just an easier way for them to do their jobs?
WJ: Yeah, there's some amount of education required.
There's a general understanding. And there's a lot of work that has been done over the past 20 years in the assessment field itself, through organizations like IAAO (International Association of Assessing Officers).
They’ve been talking about how to use regression models to arrive at values and the benefits of that. The only problem is that you have to have a statistics background, you have to have someone in the office trained to do that, and these people are expensive.
So it's something where people have general knowledge. I would say most of the people we talk to are at least aware or have taken classes. Most people, though, have not deployed it because of the [technical] difficulty there.
JF: I realize I know very little about the profession of property assessment. Are there professional certifications required to do the job?
WJ: Absolutely. It varies by state. Most states have certification requirements, like in Texas, you need to be an RPA registered property appraiser. Some states offload that to the IAAO, they have courses and certifications as well. So there are certifications that are required in most states.
JF: Ah, so there’s an occupational licensing setup, similar to a CPA certification.
I’m curious, are the folks that end up using your product directly the same as the ones paying for it? Are your users the same as your buyers?
WJ: They’re generally the same person. They often have to get buy-in from the board or the county commissioner or something like that. But yes, generally, the person that's buying it is using it.
Why this, Why now?
JF: What prompted you to solve this problem at this time? What's the founding story behind ValueBase?
WJ: Yeah, it's a great question. So we were looking at the problem of slower economic growth in the developed world.
At least for me, personally, it was understanding that we have had slower, systemic economic growth in the developed world since about the 1970s and it’s not really clear why that happened. But it is clear that it is happening. And I think this [slower growth] leads to a lot of the weird craziness in our politics.
I don't know if you have had this feeling, but I definitely feel like we have much lower expectations than our parents did for how our lives will turn out – despite working in tech, one of the few areas where there still is economic growth. So there's this real sense that things have slowed down a lot.
It seems to me that, for example, cars are very similar to they were in the 1970s. Maybe they're a little faster, maybe they're a little safer, maybe a little more fuel efficient, but going from a car in the 1970's, say my dad's Mustang, to my present day Mazda…that’s a much smaller jump than from the horse to the car.
And you can repeat this over and over again, primarily the built environment, our urban cores have not changed very much. Your house looks very similar. And so I think for most of the economy, there's just been much slower growth. And the question is, what do we do about that?
Originally, I thought the answer was just found more startups. But then I realized, back in 2021, you have this massive influx of capital into the economy, it just drove valuations up, we didn't get more good startups.
So I started looking at other solutions and land policy was actually I think the biggest $20 bill on the sidewalk I found, particularly taxing land instead of income or sales or consumption. It could provide large benefits towards economic growth. If you shifted half of taxes off of income, and on to the land rent, the US would increase US GDP by about 5% a year.
Every marginal dollar that goes on land is making that better. And it's not like you have to win everything. If you win a little bit, you're gonna bump growth. This is why I work 80 hours a week at property valuation – it’s because I see this as the core way to kind of escape the great stagnation.
JF: So you came from this background of asking “Why does it feel like economic growth is slowing down?” And came to the conclusion that, in fact, “An LVT would fix that”?
WJ: Yeah, exactly. Urban policy, like Yimbyism, that's another great example. Essentially, the built environment and land policy is the doable lever. You can actually go and solve these things and, in a way, I think inventing new technology is much harder than straightforward policy shifts.
JF: As you were thinking through these things, where were you working? Did you spend any time in think tank land?
WJ: No, I was really convinced that startups were the right way. So I joined a startup as the first employee, number three out of college. We built recommendation engines, machine learning for big banks and financial institutions. That company got acquired. I then went to work for On Deck with Eric Kornberg for a while. So I've been in tech my whole life, essentially, and then won a big research grant to do mass appraisal research and get started on this path.
JF: And from that funded research, pivoted into a for-profit enterprise?
WJ: That's right.
Founding a Company, Funding a Mission, Selling a Product
JF: Once you made that pivot from doing a research project to building a company, how did you go about getting seed capital? Was this a difficult thing to pitch?
WJ: Well, it's never easy to raise capital. And, you know, I went to a state school, I'm from North Carolina, I've got a drawl. I'm not a Harvard educated yuppie who's like in the club, right? So it was more difficult. And we're all state school kids. And so that's…it's a whole thing.
But, you know, fundraising is mostly about running a good process. We worked really hard, we ran a really good process and we found people that were sympathetic to the mission. And that was how we closed and qualified a lot of investors early.
JF: So this is more of a matching problem…you got matched with the right people and they were already sold.
WJ: Yes, and generally, I think this is a good approach to fundraising. What people should spend more time on is “how do you qualify the investors?” you're talking to and find the right people. Because you can also do that in whatever kind of business you're building.
JF: That's an interesting insight. Going back to your customers, has it been easy to get traction with folks? You mentioned that assessment becoming more of a pain point as folks retire out of it as a profession. Are y'all just cold call county assessors?
WJ: Yeah, we call assessors all day long. That's our go-to market. And, ya know, I don't think it's ever easy to go to market. But it was straightforward. And I think it was actually easier than a lot of the businesses you've worked on1 where a product led growth, that seems like almost a harder problem than ours, which is almost old school.
We just try and get people on the phone, we try to talk to them, have conversations, understand their problems, and make sure we can meet their problems. So it’s a good feedback cycle, because if you talk to enough people, you start to see the patterns, and you can figure out how to address them.
JF: Has there been any amount of education that's needed to get folks on-boarded? Is it as simple to use as, say, clicking on an arrow in some Google Maps type UI and getting an output showing a parcel’s valuation?
WJ: There's a little bit of education primarily because property taxes themselves are a little complicated. Understanding all of the accuracy statistics, the compliance requirements, there's a lot of complexity.
Also actually deploying an implementation. And so that's primarily where a lot of the work comes in is making sure we do a good job on the implementation side, get it deployed correctly, and get people comfortable with it.
JF: Let me ask you about your onboarding and implementation process. When you take on a new customer, presumably you need to go about assembling the relevant data for that new area, correct? And then there's all these downstream transformations for that data to produce an output for the customer, is that right?
WJ: That's exactly right, yeah.
JF: Is that a pretty onerous process right now?
WJ: It has become a process. It took some time to get there. We get a new customer, we hop on a big call, do a huge data checklist. You see everything they have, if they don't have the supplement, you match it. You get the model built and they kind of deploy from there.
JF: One last question about your customers and the conversations you've had with them.
Have y’all talked to anyone yet that’s not only interested in ValueBase for valuation, but also has an eye to implementing policies like an LVT? Or is that a conversation that hasn't come up so much yet?
WJ: You know, sometimes…it's actually interesting. There’s a good group of assessors that are actually Georgist and very interested. For others that are less aware, it's one of those things that…you're on site with them, or you go out and have dinner and explain things, these people really get it and they really like the idea. But generally, people are not super aware of it.
Vision for the Future
JF: That is fascinating. Going back to ValueBase, what’s the bigger picture for the company? If you were to achieve everything you can imagine achieving through your startup, what does success look like?
WJ: Ideally, we're doing valuations for the majority of assessors in the United States, Western Europe, and Eastern Europe within five years – that's the plan.
JF: I imagine that you're already thinking through localization, both in terms of language but also different tax laws?
WJ: Absolutely. You know, we talked to several different countries, a lot of different finance
JF: I'm assuming that – between different countries, but probably also between different states – there are differing interpretations or methodologies?
WJ: It was actually interesting, I would say the methodology is surprisingly robust across the United States. And even in Western Europe and Eastern Europe, most people follow the same kinds of procedures to value now.
How implementation gets put out, what exemptions there are is always different, whether they take 100% of market value or like a third, those are always different. But generally, the process is pretty similar.
JF: So the differences are more downstream of the actual valuation of the property.
Well, this has been super educational. Thank you so much for walking us through what you’re up to. We’ll definitely keep abreast of all y’alls progress!
Closing Thoughts
Personally, I took a few things away from this conversation.
The first is a much better understanding of how property assessment works in the Western world. It’s something I’ve thought about from a policy perspective, but not something I knew the execution of in detail.
The second is how accurate land valuation data opens up the door for cities to think about public investment in much more rigorous ways. I can already imagine the financial model showing how much a new bus stop or public park could impact land values (and therefore tax receipts). If cities can assess the financial ROI on public investment2, they can make more sustainable long term decisions with regard to those investments.
The third takeaway is really an appreciation for what Will and the ValueBase team are trying to do. They’re solving a specific problem – valuations are labor intensive to do and it’s technically difficult to employ the latest methods. But they’re also removing an impediment to a set of policies that could make for a better world.
For folks interested in learning more about ValueBase, you can check out their website or follow them on Twitter. And for anyone interested in Georgism or land value taxation, check out the Progress & Poverty Substack to learn more.
Once upon a time I worked at Lyft. A little later I was at Postmates. Both were consumer focused businesses where growth was largely product led (as opposed to firms relying more directly on a sales team to acquire more customers).
That’s not to say every single piece of public investment needs to be ROI positive. We might want to build things for other reasons and have positive revenue streams cross subsidize the expense. But cities should know when that's the case – and accurate data would enable them to do that.