IRS data doesn’t tell us much about who is being helped by stimulus payments. Here’s what the data we really need would look like.

When Congress passed the CARES Act in March, Americans readied to receive $2 trillion in stimulus money. But policy doesn’t happen without thinking through delivery. And on the delivery side of things, getting the money to needy Americans wasn’t quite as easy as passing the bill itself. The IRS has, to their credit, managed to move much faster than many advocates (including ourselves) expected.

 But there are still likely millions of payments to go, perhaps to the most vulnerable among us — so it is critical to understand who has not been paid, and why. The IRS has shown some transparency in this process, releasing statistics on a roughly biweekly basis of how many payments have been paid, disaggregated by state. But spend a little longer with these numbers and you’ll realize they raise more questions than they answer.

If the IRS makes a payment to an incorrect bank account or mails a check to an incorrect household, is that payment included in their count? If they make a second, correct, payment, is the payment included again? How many non-filers have used the non-filer tool, and are thus now in line to eventually get paid? How many households have used the tool for filers, and are getting — or have gotten — direct deposit instead of a check? 

What should we infer from these state-by-state numbers, anyway? 

Surely, if there are geographic trends — communities that have had an especially difficult time accessing the payments, for example — statewide numbers are far too coarse a measure to accurately identify them. What is the larger context to these numbers? How many payments does the IRS expect to pay in total, and what percent of the way there are we in each state?

 Add this all together and these IRS reports do not really tell us much at all. They provide only the vaguest notion of where we currently stand. And, they tell us nothing at all about where we’re going. Trying to use this data to figure out who still needs to be paid and what must be done to make that happen is like trying to find your way out of the woods with a malfunctioning compass and no map. It is simply not enough.

So what would useful data reporting from the IRS look like?

It would break out the payments by actionable categories that begin to explain why a family has not been paid and what might be done to fix it. For example, it would be useful to know the number of people who don’t have direct deposit information on file with the IRS who are waiting for checks.

Useful data would show a geographic payment breakdown, which in turn would highlight cities or counties where payment progress has been unusually slow. And finally, if the IRS told us how many payments they expect to make across all categories and payments, we would know how much farther we have to go.

What would happen if the IRS provided these numbers, when what matters is getting payments to actual people?

The federal government relies on external community organizations to implement programs like stimulus payments, with local leaders ensuring news reaches marginalized populations. Often these community organizations help people complete the steps they need to access benefits. But without a map and up-to-date data, these local leaders are flying blind. Perhaps the vast majority of people in a given community don’t reach the income threshold for filing taxes and have not registered with the IRS for payments, as they are required to do to receive a check. In this case, community leaders might undertake a broad campaign to help educate non-filers and assist them in completing the necessary forms. Maybe there are paperwork problems with payments to SSI beneficiaries, and community leaders need to focus on contacting payees to update their bank information, or target interim local assistance to these beneficiaries while waiting for the federal relief to come through. We can’t know. So instead these advocates are frantically spinning their wheels, desperate to help but not knowing where help is needed.

It is possible that the IRS has just such a map for internal use, outlining total goals by population, and progress toward them. But, given the lack of public transparency, it is just as possible they do not. And, if not, how is the agency confident it is prioritizing changes that affect the most people, who need the funds the most? How are they keeping track of edge cases, each one perhaps trivial in its own right, but together summing to surely millions of households? 

Historically, the IRS doesn’t have a great track record when it comes to figuring out who isn’t getting the money that they need to survive. Consider the Earned Income Tax Credit. The IRS helpfully publishes estimates each year, created in collaboration with the Census Bureau, of the portion of eligible families each year who claim the credit. But which families are the ones left out? Where are they geographically? Which of them file taxes without claiming the credit, versus those who do not file taxes at all? Details like these would prove invaluable to outside partners working to fill the gaps.

In our work with a state family leave program earlier this year, the PIT team found insufficient data hampering the team’s efforts in two significant ways. Expanded outreach around parental leave was a significant administration priority. But the agency was not tracking what portion of new parents actually used the program, much less which types of parents were left out. 

This led to all manner of contradictions. The state’s outreach efforts targeted entirely new beneficiaries. But in practice, over half of the targeted mothers had been accessing part — simply not all — of the state’s programs. And while the agency was aware that slow processing times could be a major impediment for vulnerable families, they were not tracking how long applications took to process, or which types of applications were seeing the longest delays, making it impossible to prioritize solving for the most problematic issues.

Especially in a crisis, with critical decisions being made daily, and huge amounts of work to do, there is a strong and reasonable urge not to waste time. And if there is a clear enough sense of what to do first, there is no need to spend time crafting a dashboard. There is no need for a GPS system and a map of the west when you are first setting out from New York to drive to a friend’s house in California. 

But, as government programs move forward, the lack of a clear destination and measure of progress becomes an enormous impediment, as we see now with stimulus payments. Lots of people like to talk about programs being “data-driven”, but that is not the same as publishing giant sets of numbers that don’t tell the story behind them. Imagine what the stimulus payments could look like with careful, rigorous, and actionable accounting. There would be a map, and it would make all the difference.