Robert P. Henri

The 15% cap is not a proposal anymore. It was issued, it was challenged, and it lost in court. What has not been fixed is the process that made it possible. The way OU’s funding almost disappeared, with no hearing and no vote, is a common pattern across our government, and changing it is what my campaign is about. On this question specifically, I would support making the protection for negotiated research rates permanent law, and I would support the rules changes below that would stop it from being undone quietly.

The question refers to an announcement made in February 2025 by the NIH that it would pay every university a flat 15% for indirect costs instead of the rate each university had negotiated. It did this with no warning and no public comment, and universities sued. In January 2026, a federal appeals court blocked the cap permanently, and the Justice Department did not appeal. OU’s rate is safe for now.

But here is why it can happen again. NIH did not write a rule, which would have required public notice and a comment period. It issued “guidance,” which requires neither. The court said that broke NIH’s own regulations, but nothing stops an agency from trying the same shortcut next time.

The second problem is that OU’s and other institutions’ protections hinge on a single sentence in a larger bill that needs passing each year. Every year since 2018, Congress has included a line in the health spending bill, known as Section 224, that requires NIH to honor negotiated rates. That line has never been voted on by itself. Instead, it rides inside a bill thousands of pages long that usually passes under a shutdown deadline. The President’s proposed budget for FY2027 asks Congress to write the 15% cap into law and to leave Section 224 out of the bill. If that happens, there is no debate and no repeal. The sentence is simply gone, and so is the protection.

Neither of those problems are about research. They are about how Washington makes decisions.

Here is what I would support with a federal vote:

1. Make the protection permanent. Section 224 should move out of the yearly spending bill and into standing law, so taking it away requires a bill, a hearing, and a recorded vote. No bill does this today. I would support one, and if nobody writes it, I would.
2. One subject per bill. The One Subject at a Time Act, H.R. 4324 in the House and S. 59 in the Senate, would bar spending bills from carrying unrelated policy. I would cosponsor it.
3. Real notice before a vote. The House already has a 72-hour posting rule, and it gets waived whenever leadership wants. I would support a version that covers both chambers with no waivers.
4. Show the math. Any agency setting a rate that decides whether labs stay open should publish how it got there and take public comment, whether it calls the change a rule or guidance.

Some universities’ rates may be too high. If so, an audit can solve that, not an arbitrary figure with no explanation.

Keywords: OU’s research funding was nearly cut with no hearing and no vote. That is the pattern I am running to change.

My position is that tribal nations in Oklahoma should have the same authority to protect their own land and water that tribes in every other state have, and that the federal money for that work should not depend on a number that can change every year without a vote. The way that authority was taken, and the way the money stays thin, follow the same pattern I described in Question 1. Decisions that should have been made in the open were made where nobody was looking.

The first problem is how that authority was taken away. In 1987, Congress amended the Clean Water Act so that a tribe could apply to EPA to be “treated as a state,” meaning it could set and enforce its own water quality standards on its own land. That change was made in a water bill by a Congress debating water policy. The Pawnee Nation applied in 1998, and after six years of review, EPA approved them in 2004. The state sued, and lost. So the fight moved to Washington. In 2005, less than a year after the Pawnee approval, one of Oklahoma’s own senators added a section to that year’s federal highway bill that applied to Oklahoma alone. Section 10211 requires EPA to let the state run its environmental programs inside tribal land whenever the state asks. It also says an Oklahoma tribe cannot be treated as a state under any EPA law unless it first reaches an agreement with the state, and even then the tribe can only share the program, never run it. None of that was about highways, and none of it was debated as water policy. The Pawnee kept what they had earned, but no other Oklahoma tribe has been able to follow them since. And because it is written into law rather than agency policy, no President and no court can undo it. Only Congress can.

The second problem is funding. EPA’s General Assistance Program is the main federal funding for tribal environmental work, and it is a single line in the Interior spending bill each year. EPA’s baseline is $110,000 per tribe per year, a figure set in 1998 and never adjusted since. That is roughly one scientist and a truck. And because it is an annual line item, cutting it does not require repealing anything. It requires one number to change in a bill written under deadline.

The third problem is who is actually failing. In 2017, GAO added federal management of programs serving tribes to its High Risk List. The findings were about federal agencies, not tribes. GAO found that the Bureau of Indian Affairs, a federal office inside the Interior Department that must approve any energy project on tribal land before it can start, had no process and no data for tracking how long its own reviews took. One tribe reported that a review dragged on for eight years and cost it more than $95 million in lost revenue. GAO also found unsafe conditions at schools run by the federal Bureau of Indian Education and weak oversight of hospitals run by the federal Indian Health Service. It listed nearly 40 recommendations. Nine years later, the designation is still there.

Here is what I would support with a federal vote:

1. Put Section 10211 to its own vote. Congress made the 1987 decision in the open. If it wants to reverse that for one state, it should do so in the open. No bill repeals Section 10211 today. I would support one, and if nobody writes it, I would.
2. Write a floor under the General Assistance Program. The Interior spending bill should say “not less than” a set amount for tribal environmental programs, adjusted for inflation, so a tribe’s water monitoring is not one deadline away from disappearing and does not sit at a 1998 number for another 28 years. I would support that language in the bill.
3. Put a clock on the High Risk List. A 2019 law already requires agencies to report on open GAO recommendations. It has not worked. Every open recommendation should carry a deadline and a name, and the agency should answer for it every year until it closes. I would support that change, and if nobody writes it, I would.
4. One subject per bill. The One Subject at a Time Act, H.R. 4324, would have kept Section 10211 out of a highway bill in the first place. I would cosponsor it.

The state may have a real case for running some of these programs. If so, a hearing with the tribes at the table can settle that, not a short section buried in a highway bill.

Keywords: Congress gave Oklahoma’s tribes authority over their own land and water in a water bill, then took it back in a highway bill. It should give it back in the open.

My position is that Congress should know what its own water programs add up to before it votes on them, and that it should get current data on the Ogallala Aquifer every year. The state’s new reporting and enforcement laws do what a state can do. The limit on how much they can protect the aquifer is that the biggest levers, the money and the data, are federal. I am not going to grade a state law I would have no vote on. I have never administered a water permit, and I will not claim to have a settled opinion about one. What I can tell you is what Congress actually controls, because that is the job I am asking for.

Four things, all federal:

• The Farm Bill funds EQIP and the Ogallala Aquifer Initiative. That is where the money for efficient irrigation in western Oklahoma comes from.
• The Interior spending bill funds the USGS monitoring that tells us how fast the aquifer is dropping.
• USDA Rural Development makes the loans that fix leaking water lines in small towns.
• The water compacts Oklahoma shares with Texas, Kansas, and Arkansas cannot change without the consent of Congress.

Here is the part that does not come up. The Farm Bill pays producers to use less water through conservation programs. The same bill pays producers more to grow water-intensive crops through commodity supports. Both programs are real, both are funded, and Congress does not require anyone to put the two numbers on the same page. It votes without ever seeing the net figure. It is voting now. The House passed its Farm Bill in April, the Senate released its version in June, and the current law expires September 30. This is the same pattern I described in Questions 1 and 2: a decision with real consequences for Oklahoma, made without the information that would let anyone judge it.

The second problem is that Congress stopped asking for the data. In 1986, Congress directed USGS to report Ogallala water levels every year. In 1995, it cut that to every two years in a bill about reducing paperwork. Oklahoma’s water was not the subject of that bill. The most recent report, published in 2024, covers water levels through 2019. Oklahoma is making decisions about a shrinking aquifer with numbers that are seven years old.

Here is what I would support with a federal vote:

1. Require the net number. Not a mandate on a single farmer. A published figure showing what federal programs, taken together, do to water use in the aquifer states, on the table before the vote, not two years after. No bill requires this today. I would support one, and if nobody writes it, I would.
2. Restore the annual aquifer report. Congress asked for it every year in 1986. It should ask for it every year again, with a deadline USGS has to meet. I would support that language in the Interior spending bill.
3. Get the rural water money to the towns that need it. The House Farm Bill includes low-interest loans and loan forgiveness for small-town water system repairs. That money only helps if it reaches towns that cannot staff a competitive grant application. I would support that provision, and I would support directing USDA to reach those towns first.

On data centers and water, the federal piece is that water rules and power rules get reviewed separately, and I cover that in Question 6.

I am not putting agriculture against industry here. I am saying the federal government does not currently know whether its own water programs cancel each other out. It should, and Oklahoma should be the state that makes it find out.

Keywords: The federal government does not know whether its own water programs cancel each other out. Oklahoma should be the state that makes it find out.

17,000 wells did not get orphaned because nobody studied the problem. They got orphaned because walking away was cheap. My position is that the federal government should make walking away expensive, keep the plugging money flowing, and pass the well-repurposing bill the House already passed.

Before a company drills on federal land, it posts a bond, which is money set aside so that if the company quits, the bond pays to plug the well. The minimums were set in 1960 at $10,000 for a single lease and $25,000 for all of a company’s wells in a state, and they stayed there for sixty-four years. By 2019, the average bond BLM held was $2,122 per well. Average cleanup ran about $71,000, according to GAO and Interior’s Inspector General. In 2024, Interior finally raised the minimums and tied them to inflation. In June 2026, BLM proposed putting them back where they were.

That is the first problem. A bond minimum is a number in a regulation, not in law. An agency can leave it alone for six decades, fix it, and unfix it two years later, and Congress never votes on any of it. A fixed minimum is a countdown.

The second problem is that the money that plugs wells is temporary. The 2021 infrastructure law put $4.7 billion into orphan wells nationwide. In one year, federal dollars plugged 938 Oklahoma wells. State dollars plugged 27. When that money runs out, the plugging rate goes back to what the state can afford.

Here is what I would support with a federal vote:

1. Bond to the real cost, in law. Small operators cannot carry a half-million-dollar bond, and Oklahoma runs on small operators. That objection is real. The answer is a bond scaled to what each well will actually cost to plug, indexed to inflation, and written into the Mineral Leasing Act so no agency can quietly reset it. A bill to do this was introduced in the last Congress and has not been reintroduced. I would support one, and if nobody writes it, I would.
2. Guard the plugging money. Keep the federal orphan well program funded past the infrastructure law, and write the rules tight enough that plugging does not become its own cash grab. I would support that language in the Interior spending bill.
3. Pass the bill that already passed. In April 2024, the House voted 333 to 75 to have the Department of Energy find abandoned wells, improve plugging, and study which ones can be converted to geothermal. Rep. Bice co-led it, and she was right to. The Senate let it die in committee. Senator Mullin is a cosponsor of the Senate version, S. 818. I would cosponsor the House companion.
4. Bond every repurposed well. A well that becomes a geothermal well or a storage well is still a hole in the ground with somebody’s name on it. Oklahoma’s HB 3173 says the Corporation Commission “may” require financial assurance on a repurposed well. May. Weak financial assurance is how we got 17,000 wells. Convert every suitable well, and attach a bonding requirement to any federal dollar that helps convert one. I would support that condition in the DOE program above.

One more thing, and it is not my vote. All three Corporation Commissioners took hundreds of thousands of dollars from the industry they oversee. None of it is illegal, and that is why the state’s bonding rule stayed weak. That fight belongs to Oklahoma’s legislature, and I would be prepared to introduce that language for any legislator who wants to take it up.

Keywords: Federal well bonds were set in 1960 and left there for sixty-four years. A fixed number nobody has to vote on is a countdown.

My position is that the money for severe weather research should be written into law with a floor under it, so that it cannot be cut without a vote. That is not how it works today, and the way NSSL lost ten percent of its funding last year shows why it should.

Start with what the investment has already bought. The research done in Norman produced:

• The first radar track of a tornado’s full life, at Union City in 1973.
• Dual-polarization radar, which shows debris the moment a tornado touches down and now runs on every Weather Service radar in the country.
• Warn-on-Forecast, a system aimed at closing the gap between “conditions are favorable” and “it is on the ground.”
• The phased array radar that is meant to replace the aging national network.
• Hail research, in a country that loses about $10 billion a year to hail damage.

When the Enid tornado hit in April, NSSL was the lab that surveyed it.

That work is cheap for what it prevents. All of NOAA’s research labs and institutes combined are less than 5 percent of NOAA’s budget. Hurricane research alone, which comes out of the same research arm as NSSL, is credited with saving about $5 billion in property damage for every major storm that makes landfall. This is a property argument, not a climate argument. You can want government smaller and still want this.

The threat to that investment is not a vote against it. It is a process that cuts it without a vote. In 2025, the administration proposed eliminating NOAA’s research arm, NSSL included. Congress refused. The House and Senate committees each set research funding between $710 and $730 million. But the final bill, passed in January, came in at $630 million. No committee voted for that number. It appeared in the conference agreement, the final version negotiated behind closed doors and passed without amendments. The instructions to move programs out of research and into operations were written in report language, the explanatory text that travels with a spending bill but is not law. Nobody had to vote against NSSL. The lab survived, and ten percent of its funding did not.

The same proposal is back for 2027, and the starting point has moved. Last year, the members defending research asked for $656 million. This year, they are asking for $605 million. The new floor is the old cut.

Here is what I would support with a federal vote:

1. Write the floor into the bill. The spending bill should say “not less than” a set amount for NSSL and for weather research, in law, not in a report. A transfer out of research would then have to be written down and voted on. That includes the radar lab, which should not lose funding while Congress is still deciding which radar to buy. I would support that language in the Commerce spending bill.
2. Fence any money that moves. Research money sent to operations stays research money, with quarterly reporting on what it bought. I would support that requirement in the same bill.
3. Two-year money. A field campaign should not die because Congress passed a temporary spending extension instead of a budget. I would support two-year funding for weather research.
4. Reauthorize the Weather Act with dollar floors. The Weather Act Reauthorization Act of 2025, H.R. 5089, is led by Rep. Frank Lucas of Oklahoma and passed the House Science Committee last September. It reauthorizes NSSL’s programs through 2030. It should also set the dollar amounts, so next year’s fight starts from the authorized level, not last year’s cut. I would cosponsor it and support adding the floors.

A freshman independent will not sit on the Appropriations Committee. What I would have are floor amendments, letters with the right number on them, and cosponsorships. That is enough to make somebody vote, and a vote is the whole point. Some of that research may belong in operations. If so, move it with a line in the bill that every member votes on, not with a paragraph in a report that nobody does.

Keywords: NSSL lost 10% of its funding, and no committee ever voted for it. If Congress wants to move that money, it should have to vote on it.

The question says it plainly: data center development is continuing. One project is proposed in Luther, in this district, and the neighbors are already at the zoning board. Some people would rather stop it all until we know more, and I understand why. But a moratorium is a bet that the building stops, and if the bet loses, Oklahoma has spent its time arguing instead of writing the rules. My position is that the rules should be written now, so that households are not the ones paying when a project’s promises fall short. The company that announces the load should pay for the upgrade, households should never cover the difference, and both rules should be in federal law rather than left to a regulator that can change its mind.

The question is not whether to want data centers. It is who pays for the grid they need, and what happens when the forecast is wrong.

Your electric bill is set by the state, and Oklahoma has already acted. HB 2992 requires large loads to pay their own costs, be billed separately, and give 60 days’ notice before buying land. It passed the Senate unanimously.

The transmission system is federal. The Federal Energy Regulatory Commission, or FERC, decides who pays for a new high-voltage line or substation. That is the half that gets skipped, and it is the half I would have a vote on.

The risk is one FERC has named itself. A company announces a 1,000-megawatt data center. The utility builds the lines for 1,000 megawatts. Only 200 shows up. The lines are in the ground, and everybody else’s bill pays for them. The risk is not that data centers use too much power. It is that they promise to.

The rules that decide who pays are being rewritten right now. Last October, the Energy Department directed FERC to write a national rule for connecting large loads, and its proposal would have assigned 100 percent of the upgrade cost to the company that asked for it. In June, FERC chose not to write that rule. Instead, it told each regional grid operator, including the Southwest Power Pool that runs Oklahoma’s grid, to justify its own rules or fix them within 60 days. So the rule that protects Oklahoma households now lives in a grid operator’s tariff, which can change without anyone in Congress voting on it.

Here is what I would support with a federal vote:

1. The company that announces it pays for it. The upgrade gets built to the company’s forecast, with the company’s money, before it plugs in. If a company promises 1,000 megawatts and brings 200, that is its miscalculation and its bill. The Energy Department already proposed this. It should be in the Federal Power Act, not in a proposal. I would support that bill, and if nobody writes it, I would.
2. The public keeps the asset. There is no refund for capacity a company did not use. I would support that as part of the same bill.
3. Households do not pay for a data center’s load, and nobody subsidizes it. If a data center raises residential rates, the data center covers the difference. If that gets expensive, the company can cool its servers a better way. A company spending billions on one campus can pay its own bill, and any federal help, whether loans, land, or fast-tracked permits, comes with that condition. No protection for households, no federal help. I would support attaching that condition to every federal program that touches these projects.
4. Households come first when the grid is tight. SPP already does this. Its fast-track service for large loads cuts the data center off before any house. I would support making that the national standard.

Water is mostly a state question, and SB-259 already handled it: no open-air groundwater cooling, and low-consumptive systems required. The federal point is that water rules and power rules are reviewed separately. Closed-loop cooling saves water and burns more electricity, and the two never get read side by side.

If the building stops, these rules cost nothing. If it does not, they are the only thing standing between a data center’s forecast and your electric bill.

Keywords: The risk is not that data centers use too much power. It is that they promise to. Whoever makes the promise should pay for it.

My position is that rural Oklahomans should be able to plan on care being there when they need it, and right now they cannot, because every protection rural hospitals have is either temporary, delayed, or unmeasured. Fixing those three things is what I would support.

This lands harder in Oklahoma than in most states. About one in three Oklahomans, 1.3 million people, live in rural areas, compared with about one in five nationally. Oklahoma ranks second in the country in the share of rural hospitals that have stopped offering general surgery since 2014. That matters more than it sounds, because a hospital cannot safely deliver babies without surgical capability, so when surgery goes, obstetrics usually follows. More than half of Oklahoma’s counties now have no maternity care at all.

The first problem is that the replacement money is smaller, slower, and shorter than the cut. Last year’s reconciliation law reduces federal Medicaid spending by about $911 billion over ten years, and about $137 billion of that comes out of rural areas. The same law created a $50 billion Rural Health Transformation Program over five years to soften the impact, and Oklahoma received $223.5 million in the first year, one of the largest awards in the country. That money is real, but it pays by reimbursement, so a hospital spends first and is paid back later, and the hospitals closest to closing are the ones least able to front the cash. And the fund runs five years while the cuts run ten. Nobody has said what happens in year six.

The second problem is that nobody is required to measure the new work requirement. In 2020, Oklahoma voters wrote Medicaid expansion into the state constitution so the legislature could not roll it back. It held. Then the rollback came from Washington instead. Starting in 2027, federal law requires working-age adults on expanded Medicaid to document 80 hours a month of work, school, or volunteering. Some will lose coverage because they did not meet the requirement, which is the policy working as written. Others will lose it because they missed a form while working full time, which is not. Right now, nobody has to tell the difference.

The third problem is that the rules rural hospitals are paid under keep expiring. The Medicare payment adjustments that keep small rural hospitals solvent, and the telehealth coverage rural patients rely on, are renewed in short extensions attached to must-pass spending bills. Last fall, the telehealth coverage lapsed for six weeks during the shutdown. The rural hospital payment programs expire again on December 31, 2026, unless Congress acts. A hospital cannot plan around that, and neither can a patient.

Here is what I would support with a federal vote:

1. Advance payment for hospitals in distress. The Rural Health Transformation Program should be able to pay facilities in financial trouble up front instead of by reimbursement. I would support that change.
2. No quiet expiration. Before the fund runs out, CMS should be required to report to Congress what it built and what happens when the money ends. I would support that requirement in the next health spending bill.
3. Make the rural payment rules permanent. The Rural Hospital Support Act, S. 335, and the Save America’s Rural Hospitals Act, H.R. 3684, would make the rural Medicare payment adjustments permanent instead of renewing them every year or two. I would cosponsor the House bill, and I would support making the telehealth coverage permanent the same way.
4. Measure the work requirement. CMS should be required to publish enrollment losses by cause, state by state: how many people lost coverage because they did not work, and how many lost it because of paperwork. If the requirement is doing what its supporters say, that data proves it. If most of the losses are paperwork, that is a fixable problem. I am not for or against the requirement here. I am for finding out. I would support that reporting requirement.

Rural hospitals were in trouble before this law, and the state’s share of the expansion cost is a real and growing line in Oklahoma’s budget. Both of those things are true. Neither one means the answer is fewer people covered and less information about why.

Keywords: Oklahoma voters locked Medicaid expansion into the constitution so it could not be rolled back. The rollback came from Washington instead. Before Congress decides whether that worked, it should have to find out.