A few weeks ago I wrote about the latest high-profile crash involving Tesla’s level 2 automation system, in which a car drove into a two-story brick house on a quiet residential street in Katy, Texas, killing a woman who was standing in her living room. Some developments since then merit an update.
Both the NTSB and the NHTSA have opened investigations into the incident. The NTSB has preliminarily confirmed Tesla’s initial reaction to the event: the driver had “overridden” the automation system by pressing the accelerator, in this case to 100%, causing the car to accelerate to a speed of 73 miles per hour before it hit the house.
The above facts will likely cause many to conclude that this crash was not Tesla’s fault. A bit more context complicates the picture. First, it is worth noting the similarities to Benavides, last summer’s blockbuster verdict against Tesla. In that case, the driver also “overrode” Autopilot by putting his foot on the accelerator. As the evidence showed at trial, accelerating with Autopilot engaged does not completely disable the system; rather, it overrides “traffic aware cruise control” (the system that adjusts the car’s speed in response to traffic in the lane ahead) but not “autosteer,” automatic emergency braking, or forward collision warning. I am not certain how Tesla’s latest system, which it calls “Full Self Driving (Supervised)” responds in these situations, but to say that the automation system has been “overridden” by a foot on the accelerator is probably at least somewhat misleading.
My sense is that the involvement of criminal law is a welcome development here. Crashes like these implicate complex questions of responsibility. To what extent do we expect drivers to understand the nuances of the automation systems they use and take responsibility for maintaining vigilance on the roads? To what extent do we expect manufacturers to design their systems to mitigate the effects of known human imperfections, which products liability law would call “foreseeable misuse”? The driver in Benavides, George McGee, who killed a young woman and caused permanent injuries to her boyfriend, got little more than a slap on the wrist (he was fined $1,000 and settled a tort suit for an undisclosed sum).
It will be interesting to watch how criminal and tort law develop in this area.
Wisconsin candidates are required to submit their next campaign finance reports by Wednesday, July 15th. These will record all their fundraising and spending through June 30th. The gubernatorial candidates fundraising hauls will draw the most attention, but I’m also looking forward to seeing the numbers for Wisconsin’s state legislative candidates.
In November, Wisconsin will elect all 99 state assembly and 17/33 state senate seats. The 2026 elections will be the best chance for Democrats to win legislative majorities since they lost them in 2010. Currently, Republicans hold a 3-seat majority in the upper chamber and a 9-seat majority in the lower.
The simplest path to a majority for Democrats is to hold the seats they’ve already got while winning the handful of Republican-held seats Kamala Harris won in 2024, despite narrowly losing statewide. This would be enough for a two-seat majority in the senate and a 1-seat majority in the assembly.
These are not the only possible battlegrounds. Democrats could cast a wider net and attempt to flip some seats in the assembly that Trump won only narrowly. Republicans could take a swing at the two Democratic-held seats which Trump actually carried in 2024. The July 15th fundraising figures will give us a sense of where the two parties’ strategic priorities lie.
This post lays out some background on how much these campaigns usually raise and spend, both cumulatively and through July 1st. I also calculate a simple measure of money’s influence in these races.
Campaign Spending
In assembly races, the median Democratic campaign spending fluctuates between about $20,000 and $45,000. Median Republican spending is a bit higher–hovering around $50,000 in recent cycles. These statistics are only for elections where both parties fielded a candidate.
State senate races are much more expensive. In 2010, the median Democratic campaign spent $83,000 and the median Republican spent $124,000. Those levels of spending stayed about the same in 2012 and 2014, but in 2016 campaign spending went through the roof. The median Republican campaign spent $334,000 to $433,000 for the median Democrat.
In most years, Republican spending has exceeded Democratic spending for the median candidate. But, since 2020, Democrats have consistently outspent Republicans in the biggest battleground races.
In 2024, the most expensive Assembly race for both parties was District 21, in the southern Milwaukee suburbs. Republican incumbent Jessie Rodriguez spent $1,439,000 en route to defeating the Democratic challenger David Marstellar whose campaign spent $2,526,000.
The Democratic advantage was even larger in the most expensive senate race that year. In the 8th district, Democratic winner Jodi Habush Sinykin outspent Republican Duey Strobel by more than $3 million.
Most campaign spending happens late in the campaign. In Wisconsin, after all, the partisan primary isn’t held until August; although, in many races the ultimate nominee is known long before then for each party.
Here are the historical July 1st fundraising statistics for those campaigns which ultimately did become their party’s nominee to the state legislature. In 2024, the median assembly campaign had still only raised about $10,000 dollars by July 1st and the median senate campaign about $60,000. These are the numbers against which to benchmark the fundraising data released tomorrow.
Fundraising through July 1st by Major Party Nominees
In Races Contested by a Candidate from Both Major Parties
Assembly
Senate
campaigns1
mean
median
max
campaigns
mean
median
max
2010
118
$12,735
$10,120
$59,273
22
$30,545
$31,374
$85,204
2012
136
$11,605
$9,926
$62,114
20
$21,567
$11,632
$79,135
2014
80
$16,531
$13,173
$98,868
28
$34,640
$21,746
$166,214
2016
86
$16,107
$10,873
$70,175
14
$55,624
$53,220
$105,909
2018
118
$20,280
$10,989
$424,077
26
$69,591
$49,667
$352,529
2020
152
$15,604
$9,932
$132,183
20
$53,263
$44,487
$187,413
2022
140
$15,481
$6,807
$547,974
24
$36,442
$23,095
$134,094
2024
154
$25,345
$9,902
$638,893
22
$89,199
$59,980
$270,159
1 Campaign finance records could not be located for a small number of campaigns, which are ommitted from this analysis
Most Expensive Races
The table below shows the 10 most expensive assembly races since 2010. The top 9 were in 2024 and the 10th was in 2022. Democrats outspent Republicans in each race, by amounts ranging from $138,000 to $2 million. Despite this spending, the Democratic candidate trailed the top-of-the-ticket (either Harris in 2024 or Evers in 2022) in all but 3 districts. Ultimately, Republicans won six of these races, despite being outspent by $9.5 million cumulatively.
Top 10 Most Expensive Wisconsin State Assembly Races from 2010 – 2024
district
Election Results
Campaign Spending
Legis. Dem
Gov. or Pres Dem
Over/Under performance
total
Dem candidate
Rep candidate
Dem minus Rep
2024
21
48.6%
52.1%
−3.5
$3.97M
$2.53M
$1.44M
+$1.09M
2024
61
48.4%
51.1%
−2.7
$3.76M
$2.45M
$1.31M
+$1.14M
2024
53
49.4%
52.2%
−2.8
$2.84M
$1.77M
$1.07M
+$693.82K
2024
85
46.8%
49.7%
−2.9
$2.81M
$1.77M
$1.04M
+$723.94K
2024
30
45.7%
47.7%
−2.0
$2.67M
$1.40M
$1.27M
+$137.94K
2024
88
49.7%
50.2%
−0.5
$2.48M
$1.66M
$822.58K
+$833.93K
2024
94
50.3%
48.9%
+1.4
$2.47M
$1.60M
$863.14K
+$739.80K
2024
26
51.5%
51.4%
+0.1
$2.28M
$1.90M
$383.60K
+$1.51M
2024
89
51.4%
51.0%
+0.4
$2.15M
$2.08M
$67.21K
+$2.01M
2022
94
51.3%
51.9%
−0.6
$1.72M
$1.16M
$562.72K
+$597.30K
Here are the 10 most expensive state senate races since 2010. Democrats had more success in these races than in the most expensive assembly costs, winning 6 of them.
Top 10 Most Expensive Wisconsin State Senate Races from 2010 – 2024
district
Election Results
Campaign Spending
Legis. Dem
Gov. or Pres Dem
Over/Under performance
total
Dem candidate
Rep candidate
Dem minus Rep
2024
8
50.7%
51.2%
−0.5
$7.83M
$5.47M
$2.37M
+$3.10M
2024
14
51.1%
50.8%
+0.3
$4.62M
$3.00M
$1.63M
+$1.37M
2024
30
52.5%
51.8%
+0.7
$4.18M
$3.21M
$971.77K
+$2.23M
2020
30
45.3%
45.9%
−0.6
$2.24M
$1.45M
$791.34K
+$656.22K
2020
8
45.7%
49.9%
−4.2
$2.24M
$1.27M
$962.65K
+$310.87K
2020
32
50.3%
53.1%
−2.8
$2.16M
$1.56M
$601.70K
+$956.01K
2020
24
43.5%
43.8%
−0.2
$1.87M
$1.15M
$723.62K
+$426.90K
2024
18
53.3%
54.4%
−1.1
$1.72M
$1.00M
$716.25K
+$287.74K
2018
17
45.9%
50.6%
−4.7
$1.46M
$573.79K
$887.45K
−$313.66K
2022
31
50.5%
50.9%
−0.4
$1.40M
$930.33K
$470.54K
+$459.79K
How Much Does Campaign Spending Help?
It’s clear that money doesn’t “buy elections” in any crude way. Candidates routinely win after being outspent by their opponent–a fact not limited to the Wisconsin legislature. Donald Trump, for instance, was outspent by $336 million in 2016 and $542 million in 2024, according to calculations by OpenSecrets.
But, measured across hundreds of elections, the money does matter. In a simple measure of this, I regressed the share of total money spent by each campaign against the election’s difference from the top-of-the-ticket. The hypothesis is that in districts where the Democrat accounts for more of the spending, that Democrat will do better relative to the performance of the Democratic candidate for governor or president in that district.
This is, indeed, what we see, but the effect of money is small. In 2024, a 1% increase in the Democratic share of a campaign’s total spending was associated with a 0.04-point improvement in vote share, relative to Kamala Harris.
This small effect size has stayed about the same in recent elections, but it used to be higher. In 2010, a 1% increase in a party’s share of the spending was associated with a three-times larger increase in the vote. Money just doesn’t buy what it used to.
With few exceptions, the development and administration of water law (and environmental law more generally) have been the responsibility of either the federal government or the states. In matters related to the control of water pollution, for example, a “cooperative federalism” approach predominates. A federal statute, the Clean Water Act, provides the substance. Yet it is largely administered and enforced by the states, with the federal Environmental Protection Agency (EPA) acting mostly in a standard-setting role. By contrast, matters of water quantity and allocation are largely governed by state law. The balance between federal and state power has not been entirely stable over the years.
Increasingly, both the federal and state governments have been unable to effectively and conclusively address some pressing water-related matters. They may be underfunded, understaffed, or politically paralyzed, among other obstacles, or they may simply choose not to regulate. This has created an opportunity for local governments to exert greater control over decisions about water resources. The remainder of this post examines the role of Wisconsin municipalities related to three water-related issues: data centers, the privatization of critical infrastructure, and the rights of nature movement.
Data centers. Data centers are large warehouses containing computer servers, data storage equipment, and related digital infrastructure supporting technologies upon which our society increasingly relies, such as artificial intelligence, cloud computing, and video streaming. They can be significant economic development engines but have become highly controversial because of their heavy use of water and energy. Recent results from the Marquette Law School poll show that 71 percent of respondents nationally and 69 percent within Wisconsin believe that the costs of data centers outweigh the benefits.
That left Wisconsin municipalities to fend for themselves when considering how to deal with data center developers. Municipalities must decide, for example, whether to allow a proposed data center to hook up to a publicly owned water system (and how to control the amount of water the data center uses); how much information to release about a data center’s proposed water use, given that many developers have requested non-disclosure agreements; whether to grant tax incentives to developers in the hope of economic development; and whether to enact moratoria on data center developments. A proposal pending before the Milwaukee City Council would effectively ban data centers of more than 60,000 square feet, restrict water use at smaller data centers, and mandate certain transparency measures. Other Wisconsin municipalities, including Madison, have enacted moratoria on new data center projects.
Ownership and management of critical water infrastructure. Many municipalities own or manage water infrastructure through their operation of water and wastewater treatment systems. Increasing financial pressures have caused some local governments to contract out the management of those systems, or even to sell the systems outright to private operators. The Milwaukee Metropolitan Sewerage District has recently been subjected to criticism over the actions of its contractor, Veolia, in operating its wastewater infrastructure. Interestingly, on the energy front, Milwaukee officials have debated the possibility of attempting to take over energy infrastructure currently owned by We Energies within the city limits. The company holds a monopoly on electricity generation in the region.
Rights of Nature movement. Finally, though it’s certainly not front-page news in the same sense as the preceding two issues, in late 2023 Milwaukee County became the first county in Wisconsin, and one of the first nationwide, to pass a “Rights of Nature” (RON) resolution to ensure that “human activities do not interfere with nature and its ability to be healthy, robust, and resilient.” The resolution emphasized the County’s interest in protecting its “waterways and bodies of water.” This represents a relatively mild formation of the RON movement; in other places, it has been crafted to allow natural entities to assert legal rights to counter alleged wrongs committed by governments and private parties. Scholars have expressed doubt about the approach and its viability.
Nevertheless, in late 2025 Wisconsin Republicans were concerned enough to introduce a bill banning “rights of nature” ordinances in Wisconsin, warning that such measures could threaten rights in private property. The bill passed the legislature, but in March 2026, Democratic Governor Tony Evers vetoed the law. His veto message was most interesting. He wrote: “I trust our local governments and the Tribal Nations of Wisconsin to know best how to address environmental concerns within their communities and how to protect the natural resources that are vital to local health, economies and quality of life.” That sounds like a recipe for even more local involvement in water law and environmental law.