Firing Noem Wasn’t Enough. The $220M Is Still on the Ledger.

The ad campaign polls at net −65. The firing polls at +47. And 73% want independent investigations into federal enforcement killings. Voters are running three accountability ledgers at once.

Wednesday, March 11, 2026 | 2 min read

The conventional read on the Noem story was that the firing would satisfy accountability demands and close the loop. The data from this week says otherwise — and it's more complicated than one unresolved ledger.

Net −65 on the $220M DHS ad campaign is the most negative result in this week's data on spending questions — more lopsided than the Iran numbers, more lopsided than the DOJ law firm controversy. Sixty-four percent oppose the spending. Seventeen percent support it.

In the same survey, the same week: 63% support Noem's firing. Sixteen percent oppose it. Net +47. The combined swing between the two findings is 112 points.

That gap alone would be the story. But there's a third ledger — and it dwarfs both.

After seeing bodycam footage contradicting ICE's account of the Ruben Ray Martinez shooting, 73% support independent investigations into citizen deaths involving federal agents vs. just 11% who oppose. That's a 63-point margin — the widest consensus reading in the entire survey, bigger than the war powers authorization gap, bigger than the Epstein subpoena question. This isn't a partisan finding. It's a consensus.

Then there's Mullin. The nomination of Senator Markwayne Mullin to lead DHS draws 38% opposition vs. 27% support — but the standout number is the 36% who say they're not sure, the highest level of uncertainty on any question in the survey. Voters know something went wrong at DHS. They're not yet convinced the fix is right.

Taken together, these aren't four separate data points — they're one pattern. Voters are running three simultaneous accountability ledgers on DHS: fiscal (the $220M), personnel (Noem out, Mullin in), and enforcement (the 73%). The administration addressed ledger two. Ledgers one and three remain wide open.

The Kash Patel data from earlier this year showed the same dynamic: 52% of respondents cited taxpayer waste as their primary concern — even among voters who didn't call Patel's behavior inappropriate. Conduct questions and fiscal questions operate independently. So do enforcement accountability questions. Resolving one doesn't resolve the others.

What This Means in 30 Seconds

For campaigns: The Noem firing is settled. The $220M and the Martinez shooting are not. Fiscal accountability and enforcement accountability are separate lanes — and both are live vulnerabilities for the administration.

For lawmakers: The 73% independent investigation number is the most bipartisan finding of the week. That's not an attack line — it's a mandate. Oversight legislation framed around transparency into federal enforcement deaths has a 63-point tailwind.

For advocates: Three independent arguments, three independent audiences. The spending question reaches fiscal conservatives. The enforcement accountability question reaches civil liberties voters. The Mullin uncertainty (36% unsure) means that story is still forming — there's time to shape it.

Methodology: Tavern Research Public Pulse survey, online panel, n=526, fielded March 2, 2026. MOE ±4.3%. March 8 survey: n=552, fielded March 8, 2026, MOE ±8.0%.

Want the full memos behind these numbers? Email data@tavernresearch.com →

Next
Next

THE LAB: The Attention Problem