PRINCIPLES, NOT PROMISES

I was at Dorset Street last March. I saw what ICE is, in person, up close. 

So let's not pull any punches here. ICE is a modern day version of slave patrols. It's motivated by the same racial hatred that has marred our republic from the beginning, but it is supported by the power of the federal government and its impact and is amplified by 21st-century technology. And it incorporates some of the techniques used by authoritarian regimes around the world: community raids, denials of due process, and public incidents designed to instill fear in ordinary people.

Because of the erosion of our civil liberties, in which ICE plays a pivotal role, our democracy is currently rated by many political scientists as "backsliding". (Just as slave patrols and Jim Crow have led some political scientists to argue that the US wasn't actually a democracy until 1965.)

Resisting ICE will protect our freedom and the human rights of everyone in our community. 

Resisting ICE is also essential because Vermont's economic well-being depends on immigrant and migrant labor, particularly in our dairy and tourism sectors. Research shows that increased ICE deportations reduce employment  (see also this analysis) and lower hourly wages for US workers. They also decrease consumer spending at local retail stores.  If we want to keep our consumer dollars here in Vermont, we need to keep ICE away from our downtowns.

ICE deportations harm older adults and people with disabilities because deportations decrease the number of home-care workers. Deportations also slow the construction of urgently-needed new housing. 

Trump's Big Bad Bill allocated more than $170 billion to ICE over four years, making ICE's enforcement budget larger than the budgets of most of the world's militaries. 

ICE appropriations cost each US taxpayer an average of $2358. 

Mass deportations simultaneously take away from overall tax revenues by forcing out immigrants who pay income, sales, and Social Security taxes (and who don't qualify for most public benefits, making them net contributors to our national budget).

ICE enforcement has even hurt a beloved Vermont culinary tradition.

We can't stop ICE from operating in Vermont, but we can make it harder and more socially costly.

Vermont has already passed several important pieces of legislation:

• S.209/Act 150, which protects anyone traveling to, returning from, or present in sensitive community locations (libraries, hospitals, polling places, social services establishments, places of worship, day care centers, and all government-owned or -controlled facilities) from warrantless immigration arrests. The federal government previously provided such protections, but they were rescinded by ICE in 2025, making state-level protection necessary. 

• S.277/Act 125, which prohibits schools from sharing student information with immigration enforcement and designates superintendents and principals as the only person authorized to permit immigration entry into non-public areas of a school, and allowing such entry only with a warrant.

• H.849/Act 87, which allows Vermonters to sue federal law enforcement for unlawful searches, excessive use of force, and other constitutional violations.

We should also:

• Protect tenants from having their PII shared with ICE by their landlords. Vermont should immediately pass legislation similar to a recent Maine law that prohibits landlords from disclosing personal information about their tenants to immigration authorities. This would be similar to the information-sharing provisions of Act 125, and would have the added benefit of preventing landlords from intimidating vulnerable tenants to stop them from reporting health and safety code violations. The law should also include a provision requiring landlords to post "no entry without a judicial warrant" notices at the entrances to their buildings, similar to the state requirement to post notices asking tenants to report any deteriorated paint and the requirement that schools distribute an immigration resource guide to all students and families. While out door-knocking, I've already seen posters to this effect from the Champlain Housing Trust (bravo!).

• Prevent ICE from wearing masks. Last year, S.208 stopped short of doing this because of constitutional constraints, but we might be able to find a creative solution if we keep trying.

• Clarify and codify Vermont's Fair and Impartial Policy Policy to ensure Vermont law enforcement never collaborates with ICE.  Unfortunately, ICE is controlled at the federal level and we are not allowed to obstruct their work. But we don't have to cooperate (this is the so-called "anti-commandeering principle"). What happened on Dorset Street this past March was a clear departure from the minimum standards of Vermont's Fair and Impartial Policing Policy, which prohibits officers from facilitating immigration detentions. Many officers likely acted in good faith, following orders and believing that those orders fell under the policy’s public safety exception. However, this grey area leaves too much room for misinterpretation. The Legislature needs to clarify the policy and ensure that what happened on Dorset Street doesn't happen again, and it needs to codify it into law.

• Find ways to strongly disincentivize sheriff's offices from collaboration with ICE. We should look for any opportunity to do so, including financial penalties (since Vermont sheriffs' offices effectively operate as quasi-businesses, contracting law enforcement and security services for a profit, financial disincentives would likely have a powerful impact).

• Increase funding for Vermont's Immigrant Legal Defense Fund.

• Make it as inconvenient as possible for Homeland Security to maintain facilities in Vermont that support ICE. I'm particularly thinking of the Homeland Security Investigations Tip Line office in Williston which, like the Gestapo*, takes tips and denunciations from the public. For example, public utility rules may be one option: utilities can be shut off in Vermont to protect the "health or safety of the ratepayer or the general public". ICE has clearly proved itself to be a danger to the public. This would be a bold move, but one worth exploring.

What we should NOT, do, however, is refuse to house ICE detainees in Vermont. While this may sound like a good idea, on-the-ground legal activists say it's better to keep people here where advocates can help them than to send them to less "hospitable" states.  

Note: You can view information about the number of detainees in Vermont on the state's Immigration Detainee Dashboard. I applaud all information-sharing measures of this kind and believe they should be part of our state's anti-ICE strategy.

I hope very much to be assigned to the Government Operations and Military Affairs Committee. If I am, I'm going to do everything I can to make it as difficult as possible for ICE to operate in Vermont.

• Gellately, Robert. The Gestapo and German Society: Enforcing Racial Policy, 1933-1945 (Oxford, Clarendon Press: 1989).
• Gerhards, Sascha. “Tracing the Stasi in the Televised German Krimi: Tatort and Polizeiruf 110 Episodes as Precursors to the Post-Reunification Stasi Debate,” German Studies Review 40, no. 3 (October 2017).
• Grabowski, Jan and Zbigniew. “Germans in the Eyes of the Gestapo: The Ciechanów District, 1939-1945.” Contemporary European History 13, no. 1 (2004).
• Joshi, Vandana. Gender and Power in the Third Reich: Female Denouncers and the Gestapo (1933-1945), (New York, Palgrave Macmillan: 2006).

Political scientists identify two kinds of rights.

• "Negative rights" ensure your right to be left alone when you do certain things as long as you aren't endangering anyone else. In this country, you must be left alone to say what you want, worship how you want, hang out with the people you want, and live in your home without the government just walking in without a warrant.  (Note to trolls: your right to free speech is a negative right, meaning that all we as a society have to do is let you speak your piece. There's no concomitant obligation for anyone else to listen to you or give a crap about what you say.)

• "Positive" rights are rights to basic health, safety, and well-being. When people think of education, health care, or safe, affordable housing as rights, they're arguing in favor of positive rights. 

Negative rights make us free, and positive rights make us thrive as a community. I believe that everyone has both negative and positive rights - and that the role of government is to ensure both kinds of rights. 

Because who doesn't want to thrive? I sure do.

Contrary to popular rhetoric, a good social safety net is an investment in us all, not a handout. It gives us time and headspace to care for our families and communities. It lets people take essential and fulfilling but lower-paying jobs (like non-profit work, elder care, paraeducation, etc.). It lets aspiring entrepreneurs start a business and create jobs for Vermonters without worrying about health or child care while they get off the ground. 

How can we create a smart and sustainable social safety net? 

• By looking for places where one expenditure can have multiple or ripple effects. 

  • A positive example is childcare: we've already seen the positive impacts of Act 76, which lowered childcare costs for Vermont families - something parents described as a "game-changer".

  • A negative example is healthcare: it is the primary cost driver of our education budgets and our property taxes are not going to go down meaningfully until we make healthcare more affordable and accessible. 

• By not dipping into emergency money. Whatever we do has to be doable every year, not for just a single year.

Caveat: One exception I might make for this is that sometimes a little more money has to be spent up front to get a program off the ground to realize savings later. I'd take these scenarios on a case-by-case basis.

Here's the amazing part: providing a stronger social safety net might actually decrease our overall cost of living and increase entrepreneurship. Check out some data that makes this case.

Research supports the common-sense proposition that providing a basic social safety net increases the rate of business creation.* We already know that Vermonters are smart and hard-working. Imagine if we could unleash all that creative potential and let people create their dream businesses without worrying about health insurance or childcare for their families! 

There are a lot of programs that provide start-up and emergency funding for local businesses and entrepreneurs, including for people who traditionally have had less access to capital. I have not yet done a nerdy deep dive into these programs and assess their efficacy, but conceptually, they are the right thing to do. I'll add more about these programs here once I learn more.

One additional thing we should work on is reducing administrative overhead for small businesses by improving the state's IT systems, which is how most Vermont business owners now interact with state agencies. 

People have cars to fix, bread to bake, plants to grow, and customers to serve. They don't want the state to waste their time with antiquated systems.

I'm going to be blunt: I have a background in web design, plus extensive—and infuriating—experience working with state systems as a small business owner. The kindest words I can think of to describe the State of Vermont's IT systems are "well-intentioned" or "quaint". Less kind words include "outdated" and "wasteful" (the Vermont State Auditor agrees). Our spending on IT is incredibly inefficient and antiquated, clunky systems waste hours and hours of Vermonters' time. We can do better, and we should.

I should also note here that every time I gave up on a website and called a state office, the people I spoke with were helpful and kind. Our people are terrific! But we must have systems that better support these wonderful state employees—and all Vermonters.

*Here's a helpful summary of some of that research. 

Vermonters pay some of the highest health care premiums in the US and spend a far higher proportion of our income on health care than most of the rest of the country—and much more than the rest of the world.  While our actual clinical care is quite good, the average wait time for an appointment at UVM Medical is more than 100 days, so many people in our community suffer for a long time before seeing a doctor.  

Clearly, our current system of insurance financing is a resounding failure. It costs a fortune and lots of people still aren't getting the care they need! We could do a lot better by our community, and we should.

Furthermore, healthcare for school employees is also the primary driver of our ever-increasing educational budgets. We can consolidate our schools all we want, but we're not going to really solve our property tax problem until we change our health care system.

Data consistently show that universal health care programs deliver better and less expensive care than the US's incoherent, fragmented, and inefficient system.

I don't just know this because of policy papers. I've lived it.

In my twenties, I experienced two universal health care systems first-hand: the Austrian Krankenkasse and the British National Health Service. In both countries, my wait times were shorter, my access to care was far easier, and the administrative hassle was virtually non-existent. I always say that it was easier for me to understand the communications I got from the Austrian health care authorities, in German, than it is for me to figure out my bills from UVM Medical!

Ultimately, I believe we must move to a universal health care system if we want to solve our healthcare problems.

As an entrepreneur,  I've learned that sometimes you have to demonstrate "proof of concept" in order to get people on board with an idea. To demonstrate the efficacy of direct government funding for universal health care, I think we should expand the community nurse model that some small rural Vermont towns have pioneered.

The community nurse model is innovative and unique—a true example of our small state's demonstrated ability to lead the nation. The local, personal care it provides is both efficient and humane. Nurses who live in the same communities as their patients have a personal knowledge not only of their patients' medical histories but also the broader context of their lives. And the home visits provided by community nurses help elders to stay in their homes, which is likewise more efficient and humane than institutional care.

There have been some excellent local-care focused models elsewhere, such as Costa Rica's EBAIS system (Equipos Básicos de Atención Integral en Salud, or Comprehensive Basic Health Care Team). Here, a team of five people, including one doctor, serves a community by providing basic preventative care and home visits. They even ensure that family pets keep their vaccinations up to date! The system was implemented in 1995 in response to poor outcomes and low public satisfaction with the existing health care system (sound familiar?). Over the next twelve years, the number of people with access to primary health care increased from 25% to 93% and infant and adult mortality declined significantly. 

I therefore believe that community nursing is a program that the Legislature should support and extend to other parts of the state. Right now, these programs rely on grants, donations, and town support, all of which are uncertain. A Dartmouth policy analysis suggested that one possible model for providing sustainable funding would be to integrate community nursing programs into the work of the regional planning commissions. We should look at this, immediately and seriously.

The idea of having some local control over health care is also exciting. Health care today is so remote and bureaucratic and people rightly feel entirely at the mercy of a system they can't control. While we don't want health care quality to be determined entirely by zip code (though, let's be real, it already is), it would be empowering for citizens at town meetings to be able to vote on the way first-line health care is delivered in their community. 

Once people see that single-payer, voter-accountable health care delivers results, I think it will be much easier to persuade the Legislature to implement it at the state level.

I live in a large multigenerational home (my husband, me, his mother, and my father until he passed away). We pay $3000 a month in property taxes. We take in multiple roommates to cover it. I really get it.

Property taxes are already a terrible burden on many Vermonters, and the rapidly rising cost means that they're hurting more of us every year.

But education itself is not the problem.

A Hollywood screenwriter once famously advised young writers, "If Act V isn't working, rewrite Act III". In other words, fix the problem at its root, not where you see it. 

The same is true for property taxes. Education costs drive our property taxes—but healthcare is the single biggest driver of education costs. Healthcare is the root problem.

Therefore, district consolidation is not going to be enough. We won't get our property taxes under control until we fix healthcare.

Property taxes are a regressive and inequitable method of taxation that hurts lower-income Vermonters.

Although our current laws do provide some property tax relief for the lowest-income families, their tax bills are proportionately higher. Like many places in Vermont and nationwide, lower-value homes in Chittenden County are assessed at a higher percentage of their value than higher-value homes. The New York Times pulled no punches, saying that property taxes cheat lower-income families. 

We should get rid of property taxes and pay for education via income tax, plus a capital gains tax, a tax on second homes, and a substantial transfer tax on any property purchased by an institutional investor. This ensures that Vermonters spread the burden of paying for education more equitably and that those who benefit from Vermont's way of life but do not pay income tax in Vermont also contribute to Vermont.

This is a complicated problem, but when I put on my small businessperson's hat, I see it as fundamentally a math problem. 

Let's start with three statistics:

1. The primary driver of our increased housing need is not population growth, but shrinking household size.  About 70% of Vermont households are now only 1-2 people (a higher percentage than any other state except Maine).

2. Recent affordable housing developments have cost between $500,000 and $600,000 per unit. (And the process is slow - often more than two years.)

3. Most Vermonters can only afford to spend between $158,000 and $270,000 on a home. 

So, clearly, we need to prioritize creating homes for 1-2 people at a price point between $150K - $250K. 

So how can we cut the cost of a new affordable housing unit in half?

I think the answer lies in three more statistics: 

1. Regulation accounts for only about 10% of building costs. The primary costs are labor time, materials, and adding new infrastructure. Manufactured homes can cut down the cost of time and materials, but new infrastructure remains challenging - the opportunities for savings are few, and not every potential rural homesite is suitable for septic. 

2. Based on the census data, I calculated that there are at least 900,000 bedrooms in Vermont, more than enough for all 645,000 Vermonters. 

3. About three-quarters of Vermont households are “overhoused”, with more bedrooms than they need. (And the sad flip side of this is that many large families are overcrowded because smaller households are in too-big homes but have nowhere to downsize to.)

So we should figure out how to use those extra rooms, which are already served by existing infrastructure, to house more people.

Some people share their homes (my husband and I do). However, that doesn’t work for everyone - some people want more privacy, others need more financial contribution from a tenant than is allowed by HomeShare, and it's hard for parents with young families to also take on eldercare responsibilities.

But splitting up existing homes into separate, private, living units would be much more appealing to many more people - and economically smart.

Programs like these have been very successful around the country. 

One program, called “Apartments in Homes”, has already been tried in Brattleboro. The program has already created more than 50 apartments for between $5000 and $50000 each (that is just 1% to 10% the cost of building a new unit!) They're also quicker to build.

The "Apartments in Homes" program is run entirely by volunteers, but I think it should be fully funded and extended all over the state. 

We could even take a page out of the "802Homes" playbook and smooth zoning and permitting hassles by creating a set of pre-approved designs for additional home entrances. 

We should also create new insurance options for people who share or split their homes because the cost of insurance can be a huge barrier. 

(Case in point: we bought a large home to share with my father and my husband's mother. After my father died, we decided to do our part in relieving the housing crisis by renting out the rooms he used. We found that we fell into an insurance black hole because we have a large home that would normally be considered a "luxury property", but because we take in roommates, that makes us...something else for which literally no American insurance company had any options. We now pay $1500 per month to Lloyd's of London, which insures the otherwise uninsurable. They cover the space shuttle, the tankers in the Strait of Hormuz, and our house.) 

In this next session I’d like to see us put financing, zoning, and permitting, and insurance changes in place to expand "home-splitting" programs all over the state. 

This seems to me to be the only logical way to tackle the housing problem quickly and at scale.

According to the most recent census data, 32% of residents in the Burlington/South Burlington area are renters. Rents are higher than most renters can reasonably afford. This is because of Burlington’s historically tight housing market, which is now loosening somewhat, providing renters with (some) relief. Even so, I believe that rents should be stabilized and eviction should be regulated to prevent price gouging and housing discrimination.

Before I begin to outline my ideas, I want to note that while there are plenty of neglectful, discriminatory, and and even extortionary landlords, there are also good property owners in Vermont who provide decent housing for a reasonable price and without prejudice. It is imperative that our rental policies stop exploitative and discriminatory practices AND that they support good property owners. We need those good property owners! They provide essential rental housing to the community.

In addition, I would urge us all to bring our conceptions of rental property ownership into the 21st century. The word “landlord” is a holdover from the Middle Ages, where the landed aristocracy inherited (often poor-quality) rental housing that they leased to more or less captive “tenants” (another medieval word we still use). Some landlords do still inherit property, particularly landlords who rent out their parents’ old home because high interest rates and changing population patterns make it difficult to sell. But our policies should reflect the reality that today, the vast majority of rental property owners are businesspeople who purchase their own properties. 

To underline this point, from now on I will refer to rental property owners as “housing providers” rather than “landlords”, and to “tenants” as “housing consumers”. This brings our outdated feudal terminology into the 21st century and, more importantly, focuses our attention on what we as a community have the right to expect: that housing providers, like all businesspeople, will provide a good product to consumers at a fair price.

All this notwithstanding, some significant things are not that different from centuries past. Rental properties remain concentrated in the hands of a few. Recent data collected by the real estate industry suggests that about a third of rental property owners in the US have only one rental unit, and another third of rental property owners have only 2-4 units. Only 13% of property owners have more than 10 rental units—but these housing providers own a disproportionate share of all the total rental properties. 

In Burlington, just a few institutions and families control a significant share of all the properties in the city. Quick back-of-the-napkin math suggests that if the above real estate statistics hold true for the Burlington/South Burlington metro area, which (according to census data) has 31,596 renter households, then perhaps as many as 80% of those renter households live in housing controlled by large property management companies.

We also need to be aware that an increasing number of Vermont properties are being purchased by institutional investors. This is a growing problem that needs to be addressed through legislation and regulation, as has already been done in other states.

Sadly, housing discrimination also continues to be a pervasive practice. Black and Latina/o renters face higher eviction rates than white renters, and female renters of all ethnicities are evicted at a higher rate than their male counterparts. Housing discrimination must be addressed forcefully.

What makes a good vs a bad housing provider? 

First, a good housing provider does not discriminate on the basis of race, gender, sexuality, or any other protected characteristic.

Second, good housing providers charge fair rents. To maintain a sustainable business, housing providers must cover their fixed costs (mortgage, property taxes, insurance, and city fees); the equipment and supplies to maintain and upgrade their buildings as needed (furnaces, hot water heaters, tools, paint, etc.); and the hiring of staff and expert contractors. Plus, like any other businessperson, housing providers need to make enough to cover the work they themselves do for their business. 

The difference between good housing providers and bad ones is simply that good housing providers act without prejudice and charge an appropriate amount of money to enable them to maintain their buildings well and pay themselves fairly. Bad housing providers charge more than is needed and/or don’t take care of their buildings so that they can divert money into their own pockets. 

Some Attainable and Affordable Solutions to Improve Renters' Lives

Housing Consumers’ Union

I am a strong supporter of unions in general, and I likewise support the right of housing consumers to organize. Currently, housing consumer unions are not covered under the National Labor Relations Act. I therefore support a “Wagner Act for housing consumers”. Of course, because housing consumers would be bargaining with many different housing providers, any Vermont law would have to include a right to coordinated bargaining and/or for a “master contract” with an organization representing housing providers.

The reason I support a housing consumer union is that I believe it would be a positive thing for everyone. Both housing consumers and housing providers would likely benefit from good-faith negotiations in which each learns more about the realities the other faces.

In addition, a housing consumers’ union would offer an effective grievance procedure for housing consumers facing discrimination and harassment by an unethical housing provider.

“Housing Security Fund”

The high cost of security deposits and housing providers’ reticence to take a chance on housing consumers with imperfect financial, criminal, or rental histories are the biggest impediment to housing for many Vermonters. 

However, I believe that it is possible to remove these barriers in a win-win way.

I have developed an initial proposal for a “Housing Security Fund” that leverages Vermont’s nationally-recognized captive insurance industry. Currently, security deposits are held in individual small bank accounts all over the state, accruing virtually no interest and providing no benefit to the public at all. However, if all these security deposits, plus a modest fee imposed on housing providers, were pooled and invested using a captive insurance policy administered by the State or a non-profit entity, they could have a powerful positive impact. 

Housing consumers would pay their security deposit into the fund in manageable installments over the course of a year. At the end of a tenancy, security deposits would be fully refundable to the housing consumer if there is no willful damage or non-payment.

Payouts from the fund could be used to:

• Generate sustainable revenue for eviction avoidance programs for housing consumers facing financial hardship (such programs are currently funded by one-time appropriations).

• Insure housing consumers against accidental damage, which currently is deducted from security deposits; and

• Insure landlords against malicious damage not covered by typical insurance policies, and cover unpaid rent.

A Housing Security Fund would be a smart, self-sustaining way to remove two of the biggest barriers to safe housing.

You can read the full proposal here.

Housing Provider and Housing Consumer Ratings

Rental registries have been tried in a number of American cities (including Washington, DC and here in Burlington). 

Unfortunately, although attempts have been made to establish a statewide rental registry in Vermont (S.79 and S.210), they have not been successful. 

One of the most important benefits of rental registries like Burlington’s is that they are generally implemented together with mandatory property inspections, ensuring that rental units are held to a baseline standard. While S.210 did at least provide for rental unit inspections across the state, they are complaint-driven rather than proactive. 

If we were to try again to create a statewide rental registry, I would like to have us aim for something more comprehensive that includes information for both housing consumers and housing providers, including not only code compliance history but also rent and utility costs and even narrative descriptions of housing provider and housing consumer behavior. 

Accurate rent costs are not readily available to housing consumers other than through commercial services, and having such information could put housing consumers in a much stronger negotiating position (particularly if they had a union).

Similarly, positive narrative reviews could encourage a housing provider to take a chance on a housing consumer who due to hardship doesn’t currently pass income screenings, but has had good rental outcomes in the past. 

Narrative reviews would also help housing consumers avoid discriminatory or harassing housing providers (and would likely improve the behavior of those housing providers).

Of course, privacy protections would have to be rigorously maintained in any rating system, but there are models of reciprocal rating systems that work well on sharing economy platforms such as Airbnb. We could learn from and emulate those, with adjustments to reflect that, unlike travel accommodations, renting primary housing is non-optional.

While there has not been much academic work on rental registries, studies of other industries suggest that a more comprehensive rental housing registry might offer significant benefits. These include a study of online restaurant reviews in Rome that linked greater availability of restaurant reviews to an overall increase in restaurant quality; a study that showed that reciprocal rating systems tend to improve behavior (see also this study); and data that shows that in countries where salaries are published, the gender pay gap is generally lower. 

These studies of other industries are only suggestive, not conclusive, but they point to the intriguing possibility that a rental registry could improve rental unit quality, improve the behavior of both housing providers and housing consumers (including reducing discrimination), and make rents more equitable. Of course, “double blind” feedback system and moderation would be essential to prevent false, retaliatory, or discriminatory statements.

Just Cause Eviction Protection

“Just cause” eviction protections require housing providers to prove that an eviction is not arbitrary or punitive. Basically, it changes the burden of proof in eviction proceedings to require the property owner to prove that the eviction is fair rather than requiring the housing consumer to prove that the eviction is unfair. This is a sound idea, in my view, not least because housing providers are far more likely than housing consumers to have the assistance of a lawyer during eviction proceedings. 

A few years ago, I strongly opposed just cause evictions because of my experience as a small-time landlord with a frightening, harassing resident. At the time, all I could imagine was being stuck, in perpetuity, with a large, powerfully-built man who made me scared to enter his apartment to do necessary maintenance and who actively sought me out with spurious complaints to force me come to his apartment so that he could intimidate and belittle me. More than once, he made it clear that he was coming after me specifically because I am a woman. (He also would not stop smoking in his apartment, even though the smoke was drifting into the room of a newborn baby next door.) This is not an experience I would wish on anyone, and if we hope to have a more diverse set of housing providers in Burlington, we cannot allow them to be subjected to a hostile workplace. 

However, my own distress about being intimidated and discriminated against made me unable to see the benefits of just cause eviction for preventing intimidation and discrimination by housing providers against housing consumers. While my failure was the result of fear, I nevertheless regret very much that I was not more clear-sighted.

With a few years’ distance, my position on the proposed charter change has changed, though my conviction that harassment and discrimination must be prevented remains the same. Therefore, I would support a just cause ordinance if it ensured that housing consumers, housing providers, and neighbors are all strongly protected against discrimination and harassment. 

While the language of the currently proposed charter change allows the city to institute protections against discrimination and harassment, protections against discrimination and harassment are not currently required by the charter change. For me to be willing to cast a vote to ratify the charter change, these protections would have to be explicitly added. (Adding and adjusting proposed city and town charter changes before approving them is within the purview of the Legislature.) 

Furthermore, thinking back to the way in which my harasser harmed the newborn sleeping next door with his smoking, as well as about the greatly increased drug activity in Burlington in recent years, I would want any charter change to include explicit provisions to enable rapid eviction for violations of health and safety rules, destruction of property, and/or criminal activity. 

These important changes could be made with a single additional clause in Article 66:

(66)  To provide by ordinance protections for residential tenants, as defined in Chapter 137 of Title 9 of the Vermont Statutes Annotated, from eviction without ‘just cause,’ where just cause shall include, but is not limited to (a) a tenant’s material breach of a written rental agreement, (b) a tenant’s violation of state statutes regulating tenant obligations in residential rental agreements, (c) non-payment of rent, and (d) a tenant’s failure to accept written, reasonable, good faith renewal terms, and (e) discriminatory, harassing, destructive, dangerous, or criminal behavior by a tenant.

Of course, simply enabling evictions of residents who are destructive and dangerous is not, on its own, sufficient to protect others from those residents.

A Burlington-area law firm that often represents housing providers advises prospective clients to plan on having the eviction process last five to six months. A long, drawn-out eviction process may result in the departure of other housing consumers, especially those with small children, unwilling to put up with harassment, excessive noise, criminal activity, or health violations. (In other words, many good people might be effectively forced out of otherwise safe housing because of one bad person.) 

I don’t believe that anyone, especially children, should be subjected to dangerous, criminal, or harassing behavior for half a year or more. The sooner a bad person goes, the better everyone’s quality of life becomes.

Therefore, I would want to see a provision added to the charter change along these lines:

"The ordinance shall include robust mechanisms to promote the rapid removal of harassing, destructive, dangerous, or criminal tenants while maintaining housing for non-offending members of a household; and shall require the City Council to petition the Vermont General Assembly for statutory changes permitting accelerated judicial ejectment proceedings for criminal behavior and other severe lease violations." 

Here, I believe we can leverage existing procedures that protect survivors of domestic violence. Emergency ex parte possession writs are already permitted in Vermont to prevent domestic violence. These could be extended to include criminal, harassing, and destructive behavior. And, as with domestic violence, an emergency writ could be followed by something analogous to a permanent relief from abuse order, which offers more comprehensive, longer-lasting protection. One of the important benefits of these legal instruments is that they are person-specific, so that if only one person in a rental unit is a problem, the other residents can stay—a very positive change from current evictions, which can dispossess entire families, even if most family members (especially children!) are innocent of wrongdoing.

“Accelerated judicial proceedings” could also include 72-hour removal dockets for violent residents, similar to the special “Accountability Court” created by State’s Attorney Sarah George in collaboration with the state.

“Compliance incentives” for housing providers could include the possible forfeiture of housing grant and loan eligibility and rental registry certification; compliance incentives for housing consumers could include access to relocation assistance, back rent waivers, and record sealing. 

To maintain equity, any rapid-removal process should be conducted in accessible language and with extra support for housing consumers, who do not typically have access to lawyers.

Preventing Deterioration of Our Housing Stock Due to Just Cause

While just cause evictions do not seem to affect the overall creation of new housing, there have been a number of studies that show that just cause and rent control policies can have deleterious long-term effect on the quality of housing stock. While the data is not conclusive on this point, it makes sense: if they don’t need to attract new residents, housing providers have little incentive to make improvements; and if an eviction proceeding for non-payment of rent lasts a long time, housing providers have less income to make quality improvements.

I think everyone can agree that Burlington does not need worse housing than it already has and that it would be wise to ensure that well-intentioned rules do not have this unintended effect.

Therefore, I believe there should be modifications to the charter change provision that limits “unreasonable rent increases to prevent de facto evictions or non-renewals”. The definition of “unreasonable” is left up to those developing the final ordinance, but I think it would be wise to clearly define “unreasonable” in the charter change itself to ensure that allowable rent increases reflect the financial realities faced by housing providers and to encourage, rather than disincentivize, quality increases.

The Consumer Price Index is often used to judge the “reasonableness” of a rent increase in jurisdictions with rent regulations. However, Vermont’s property taxes are forecast to go up by 11.9% next year, the cost of property insurance has increased by 18% in New England in recent years, and the cost of building and repair materials has gone up about 10% since 2025. This is dramatically more than increases in the overall CPI. Any determination of “reasonable” rent increase must take these skyrocketing costs into account. 

Equally importantly, quality improvements must be accounted for in any definition of a “reasonable” rent increase. In jurisdictions where fair market price can be charged for upgrades, housing quality does not deteriorate. Burlington should be one of those jurisdictions.

Therefore, I would not require, but would strongly urge the addition of explicit language that requires the definition of a “reasonable” rent increase to reflect the real-world expenses of maintaining property to a high standard, as well as allowances for quality upgrades: 

“The ordinance shall define what is ‘reasonable’ and ‘adequate notice’ in defining just cause, the which definition shall take into account the owner’s real costs of providing housing and the cost of improvements in housing quality made by the owner during the term of the lease, and shall require that landlords provide notice of just cause and other legal requirements as part of the rental agreement.”

How to Get Housing Providers to Support Just Cause

Just cause was passed by Burlington voters and the Vermont Legislature, but was then vetoed by the Governor and failed to get enough votes to override the veto. 

The Governor’s primary objections to just cause were that it would create “perpetual tenancy” laws that violate property rights—and that limiting evictions would discourage housing providers from taking risks on Vermonters new to the area, just entering the rental market, or making a good-faith effort to recover from financial or personal setbacks. 

The Governor’s first objection has already been addressed by the Supreme Court, which upheld a lower court decision that eviction restrictions are not a violation of property rights.

The Governor’s second concern is, in my view, a more legitimate one, but it can be readily overcome with smart support by the State in the form of the above-described “Housing Security Fund”. 

THIS STATEMENT CAN BE DOWNLOADED IN PDF FORMAT HERE.

More than 80% of Americans use AI directly or indirectly nearly every day. Some use cases will not surprise you, but others may:

• Autocorrect on your phone

• Google searches and Gmail

• Streaming services such as Spotify, YouTube, Netflix, and more

• Social media

• Gaming

• Ring doorbells

• Weather forecasting apps or websites (in world-majority countries AI weather prediction may actually use less computing power than traditional models)

• Google Maps navigation

• Car “infotainment” systems, safety cameras, and parking assist systems

The State of Vermont also deploys or plans to deploy many AI-supported applications, including:

• Scanning state IT systems for cybersecurity vulnerabilities

• ChatVT: an internal, private, state-built AI chatbot for state employees (i.e. any time you ask a state employee for help, they may be using AI to assist them in carrying out your request)

• Inventorying and auditing all active AI services in state agencies

• The “GenTax” tax fraud system, tax parcel mapping, and property tax equity management

• Roadway data analysis to audit pavement conditions and manage safety features like traffic signs and rumble strips

• Monitoring land use changes

• Emergency and flood response systems

• Law enforcement use of force monitoring

• Public safety monitoring to reduce violence and drug use in our communities

If we want Vermont to boycott AI for environmental reasons, then our state government and every single person in Vermont would have to agree to stop using all these tools. I think this is probably unfeasible. These tools are simply too deeply embedded into our daily lives and many contribute to our safety and quality of life. 

Most importantly, refusing to allow the state to use AI would pose a massive cybersecurity threat to Vermont’s electrical grid and IT systems, including our tax and election systems. Unless we decide to go back to paper-only records (which are vulnerable to fire and water) and rely entirely on hyper-local power, the reality in 2026 is that our state needs AI to protect not only our personally identifiable information (PII), but also our democracy itself.

Realistically, the question we face is: how can we regulate AI and the generation of AI “compute” to maximize its benefit and minimize its harms?

One thing is absolutely clear: Vermont’s environment is our state’s greatest treasure and must be protected. I personally delight in our state’s natural beauty every day when I run, bike, and ski. Ecologists are deeply aware of the irreplaceable “ecosystem services”, including pure air and water, that our environment provides. But even Vermonters who are purely commercially-minded know that we must guard the resource that powers our tourism economy, especially our ski industry. 

Would I love to say, “Vermont’s environment takes precedence over all other considerations and there should never be an AI data center in Vermont?” Absolutely. 

However, I don’t think it’s morally acceptable for us to take a “NIMBY” approach and let some other place deal with the environmental consequences of our AI use. If Vermonters deserve clean air and water and affordable electricity (and we do), so do people in Texas, Indiana, Mexico, and China. If we’re going to use a technology (and it seems highly likely we will), we must take responsibility for it. Outsourcing our AI-related pollution is elitist, often racist (see also this article), and therefore wrong.

Therefore I believe that Vermont must position itself to be a leader in environmentally sustainable AI “compute” generation. This is yet another way in which our small state has the opportunity to lead the nation.

But this does NOT mean “Hey, let’s let Elon Musk build a data center on Camel’s Hump right now!”. No. Effing. Way. Vermont can and must do data centers differently.

What Has Been Tried So Far

The Democratic Party has already made a serious attempt to regulate AI data centers. H.727, “An act relating to sustainable data center deployment,” had overwhelming support during the last legislative session, but was vetoed by the Governor (his 62nd veto!). 

I am deeply concerned about that veto because it prevented us from implementing some vital protections, including a mandatory Act 250 review of any prospective data center of 20MW or larger, a requirement that any large-scale data center sign a 10-year contract with its utility and provide an annual payment to that utility to help Vermonters improve energy efficiency and cut their utility bills, and mandatory utilization of as much renewable energy as possible at the data center site.

What I’d Like to See

We should definitely resurrect this legislation - but I believe it should go even further.

First, the legislation should include at least a two-year moratorium on the construction of any AI data centers in Vermont. More than two is probably wiser.

Why a moratorium? Right now, economic projections about AI “compute” needs are highly speculative. AI token efficiency is likely to increase more rapidly than Moore’s law (which says that the cost of computing power is halved every two years). There are also token-reduction techniques like model pruning, quantization, and knowledge distillation, as well as on-device and cloud-edge AI models that can make AI more energy efficient (though only for certain tasks). On the other hand, overall AI usage will likely increase in line with Jevons’ Paradox, which says that the more efficient a resource becomes, the more the overall usage of that resource increases. In short, we simply don’t know how increased efficiency and usage will balance out.

As the moratorium winds down, we may find that existing data centers are sufficient for the US’s “compute” requirements and that no more need to be built. A more likely scenario, however, is that we’ll need more data centers - but that technology will have evolved that has the potential to make them far less damaging. A moratorium will give us time to assess whether Vermont needs to build infrastructure that contributes to the U.S.'s AI capacity and, if so, whether that infrastructure can be built in a way that aligns with our environmental values.

Second, we should ensure that corporate actors have zero role in any data center built in Vermont, ever. 

Corporations that construct data centers have a lousy track record of greenwashing, “localwashing” (pretending to support the community but really only doing a few symbolic acts), and requiring NDAs that prevent local officials from being honest with their constituents during the permitting phase. The benefits these corporations promise rarely materialize and the harms are far worse than they admit.

Because we can’t trust oligarchs to protect our well-being, we must require any AI data centers built in Vermont to be publicly-owned and completely publicly-controlled. There should never, ever, ever be a privately-owned data center in our brave little state! 

Therefore, any new legislation should create something akin to the existing Vermont Public Power Supply Authority that will own and manage all AI data centers built in our state.

If we build our own data centers, there are two major benefits:

1. We get to call the environmental shots - we get to decide when, where, and how we build. Right now, AI data centers are huge - but if we built them as public utilities, we’d get to control the size. And right now, AI data centers have massive impacts on the environment. They can use millions of gallons of water a day (including not only water for cooling but also water for power generation and chip manufacturing) and hundreds of terawatt hours of electricity a year. But there are AI data center technologies out there that seem promising: closed-loop cooling and direct-to-chip cooling instead of steam towers, and renewable power that greatly reduces demand on local grids (and that doesn’t indirectly consume even more water through technologies such as steam generation). The International Energy Agency forecasts that up to 85% of the additional demand created by AI centers in Europe will be met by renewables. If we build centers as public utilities, we can require that they be at least as efficient as that, if not net-zero on energy and water. If that’s not currently possible, we can say “nope, not yet”. We can mandate full server recycling, particularly for rare earth metals. We can mandate stringent noise limits. In short, we can make sure that our data centers don’t harm Vermont communities like data centers in other locations currently do.

2. We get the income and control the distribution of the benefits of the center. We can require that current Vermont residents are hired preferentially to staff the centers. We can ensure that utility bills don’t go up. And we can put the income generated by the center into a sovereign wealth fund that is used to reduce overall fossil fuel use and mitigate impacts on surrounding residents and municipalities. 

Third, no AI data center should be built without significant public input. 

As in H.727, any proposed data center should be subject to the most stringent level of Act 250 review, and then a referendum should be held in all towns served by the designated utility and that are part of the impacted watershed.

Conclusion

We are lucky that the horse is not yet out of the barn on data centers here in beautiful Vermont. That means that Vermonters are still able - and absolutely must - demand total state control of, and rigorous environmental and public impact standards for, any AI data centers we build.

Nearly everyone I've met in the last few weeks is deeply concerned about the state of our downtown. I wish I could offer concrete policy proposals for what the State should do to help right now! However, the reality is that this issue is so complicated that doing so would be premature and frankly irresponsible. I think it's important for legislators to take in many different perspectives and dive deeply into evidence and data instead of making purely ideological pronouncements, and I'm going to model that by listening first before I speak about this issue.

Here's what I'm thinking right now, but my views may evolve as I learn more. However, my commitment to making sure we all feel safe and that we treat everyone in our city with compassion will never change.

• Violent crime is actually down compared to ten years ago, but people feel less safe, in part because of some high-profile incidents of violence. Our downtown economy depends on increasing people's sense of safety and well-being, but we can't do it by simply pushing bad behaviors out of our downtown into surrounding neighborhoods. That isn't a sustainable solution. We must address the problems at the root.

• Solutions are being hindered by a political turf war. We have enough on our hands without that. We're not going to make real progress until we start working together more constructively. The Legislature should provide support to promote collaboration - what specific form that should take is something that must be thoughtfully worked out. (Some examples of what I mean by turf wars: the poorly-planned "defund the police" initiative a few years back, the lack of dialogue with key city stakeholders during the initial development of "Burlington Forward", and Governor Scott's deployment of state police officers to Burlington without even notifying Mayor Mulvaney-Stanak and only for a few weeks in mid-August, which just happened to be right before the primary. And a wonderful example of what we can accomplish when people collaborate: the highly effective "accountability court" for repeat offenders proposed by State's Attorney Sarah George and funded and supported by the State.)

• Evidence and data must drive solutions. Community members are clear about accountability. They don't want programs they consider to be unconditional handouts. However, I've heard service providers be equally clear that placing too many conditions on access to services can be a deterrent and can actually drive people away from help and back onto the streets. Our policies about providing services should be governed by a combination of compassion, evidence, and pure pragmatism rather than preconceived notions. We must focus on proven ways to get people off the street and help them get clean.

• The more we can restrict or deter guns downtown, the better. One possible way to do this is through insurance incentives for bar owners to ban guns rather than an outright ban in statute, which faces likely Second Amendment challenges.

• It is completely unreasonable for the State to kick unhoused people out of motels before shelters are ready and then wag the finger at Vermont cities and towns for the problems created when those people come there for help. First of all, I don't think anyone should ever be scolded for compassion. Second, evidence does not support the claim that places with higher levels of social services are not a magnet for the unhoused population (the so-called, and unsubstantiated, "honeypot effect"). The reality is that lack of housing and high rents drive homelessness, and Vermont has both.

• The State should play a role in the location of shelters and other services that everyone thinks should be in some other neighborhood. I'm also not persuaded that larger shelters are the best thing for our community. They bring advantages in terms of efficiency, but they create a real burden on the people and businesses right next door to them. As a result, these services often get concentrated in higher-poverty neighborhoods with less political clout, reinforcing negative spirals. Distributing smaller shelters and services is not only more equitable - I think it may also have the beneficial effect of promoting more human interaction and dialogue between the people receiving services and neighboring businesses and residents.  

• We need more Community Service Officers downtown. Evidence supports "hot spot policing", which is being implemented by Mayor Mulvaney-Stanak in Burlington. This approach, which identifies and focuses resources on specific known trouble spots, is based on routine activity theory (see also this summary), which posits that a crime requires a motivated offender, a suitable victim, and a viable location - generally a place where the potential offender feels comfortable, has a routine, and there are few "guardians". Therefore, the presence of "guardians" such as security guards, attendants, clerks, cleaners, other public employees, and public safety officers, can be a real deterrent to crime rather than simply displacing it into neighboring streets. This is because neighboring streets don't have the same number/type of potential victims and potential offenders don't know the routine of those streets. However, hot spot policing has been shown to have adverse effects on minority-majority neighborhoods, and we cannot allow this to happen in Burlington. To minimize racial profiling and other forms of abuse, downtown patrols should be regularly audited by the Police Commission and consist primarily of Community Service Officers and Community Support Liaisons, with backup provided by sworn officers only when essential for public safety.

• We should consider a "Night Mayor" for Burlington - a person who serves as an advocate and policy-maker for those who participate in and are affected by a city's nocturnal life. This role is now part of the city government in at least 40 places around the world. This podcast provides an accessible overview of the idea and how it's working out in practice.

Obviously, I would attend the NPAs each month (probably by video during the session, as I plan to stay in Montpelier during the week).  I also plan to hold virtual office hours, just as I did when I was teaching. I found that it encouraged more participation than in-person office hours, though if I hear that a lot of people want in-person meeting opportunites too, I'll make myself available that way as well. 

I would of course be available by email too. 

I'll share regular updates during the session on Front Porch Forum and a blog. 

However, I don't plan to use social media. It's a dangerous place for my under-18 constituents to be, and no Vermonter needs to wade through long scrolls of hateful nonsense posted by out-of-state trolls. We all have better things to do.