The Implementation Phase of Burden Reduction

New OIRA report offer snapshot of Biden efforts to cut the time tax

As a scholar of administrative burdens, I’ve been tracking the efforts of the Biden administration in this area. They have been unprecedented in terms of drawing attention to the topic, creating a legal framework including an executive order, and institutionalizing a series of processes to push agencies to reduce burdens.

Ultimately, burden reduction will depend on how federal employees in agencies use their discretion to comply with the new legal framework. In other words, a governmentwide effort to improvement implementation (via burden reduction) is itself subject to the vagaries of the implementation process, depending on local factors like agency capacity, leadership interest, culture and history.

On July 10, the Office of Information and Regulatory Affairs, an obscure but very powerful part of the Office of Management and Budget, released a new report. Tackling the Time Tax¹ aims to track the progress the federal government has made in burden reduction.

OIRA is in charge of the Paperwork Reduction Act, which is a law that supposed to reduce the paperwork hassles the federal government imposes on the public. And it has used this implementation power to advance the burden reduction project in government, which OIRA Associate Administrator Sam Berger described as “a central priority for the Biden-Harris Administration.”

Here are my quick impressions of what the document tells us about progress in burden reduction efforts.

A change in culture and language

The most important aspect of the report is perhaps not any specific action it reports, but the change in government culture and language it represents. OIRA is the home of cost-benefit analysis in government, a tool accused (sometimes unfairly) of tilting government regulation in favor of businesses. Here, OIRA is very much directing attention not to powerful businesses, but individuals struggling with government processes.

The preface starts by describing an individual with disability talking about processes to renew their disability benefits:

The ordeal they described—confusing notices, complicated questions, and underlying it all, the deep anxiety of potentially losing life-saving assistance—is part of what are called “administrative burdens.” These include the time, money, and psychological costs involved in interacting with government. Forms and other information collections are an essential part of how Americans seek support from the government—but too often paperwork burdens like these end up serving as a tax on the time and well-being of individuals seeking assistance. And while reducing paperwork burdens matters for all Americans, we also know that those most in need of government assistance are often those least able to navigate complex forms or processes, so time taxes and burdens entrench social and economic disparities across underserved communities.

It is easy to dismiss this as simply a type of shallow politicking, one not dissimilar to the President telling the stories of individuals at the State of the Union. This misses, I think, the potential for such centering on user experiences as a means to transform how public organizations think about and structure their relationship with the public.

Some agencies have managed this. The Department of Veteran’s Affairs, for example, has institutionalized a set of customer experience values in its statement of organizational core values, which is codified into law.

(a) Ease. VA will make access to VA care, benefits, and memorial services smooth and easy. 

(b) Effectiveness. VA will deliver care, benefits, and memorial services to the customer's satisfaction. 

(c) Emotion. VA will deliver care, benefits, and memorial services in a manner that makes customers feel honored and valued in their interactions with VA. VA will use customer experience data and insights in strategy development and decision-making to ensure that the voice of veterans, service members, their families, caregivers, and survivors inform how VA delivers care, benefits, and memorial services.

Not coincidentally, VA is the poster child for successfully improving customer experience in the federal government, helping to boost user trust in VA services from about 55% to close to 80% since 2016.

It is amazing to have government institutions like OIRA take seriously the idea of psychological costs, or to have our second-largest federal employer discuss emotions as central to how they run their business. It reflects a broadening of how government thinks about citizen-state interactions, one that is necessary to putting the public at the center of how the government provides services.

Building momentum, sharing successes.

Governments are generally terrible at highlighting successes. This is partly a response to their environment, since positive news is overlooked in favor of negative stories of failure. (And don’t blame the media — that is on us for clicking on those stories of failure, reflecting a tendency toward negativity bias).

If you want good government, you need to draw attention to success, not just failure. The OIRA report does not offer a comprehensive account of where agencies are succeeding or failing with burden reduction efforts. Instead, it highlights success stories. But that is what is needed at this moment. Federal employees need to believe the momentum is real, that this reform is lasting, and they they can play a role.

I don’t think this is naive or pollyannaish. After all, the report includes plenty of evidence of where government is failing to reduce burdens. It includes an estimate of $140 billion of annual unclaimed benefits as a measure of this failure, as well as naming the welfare programs where people reporting the greatest hassles.

But then it pivots to a series of success stories. Such as this one about form simplification from the US Department of Agriculture.

Burden reduction as an ongoing journey, not a destination

The most interesting success story for me was around student loan forgiveness (pages 36-40), since it, intentionally or not, underlined how demanding burden reduction efforts can be.

I’m not talking about the $10K loan forgiveness cancelation that the Supreme Court nixed, but other loan forgiveness programs. Here, the Biden administration has forgiven $66 billion in loans for 2.2 million borrowers, which is pretty extraordinary, but has been done incrementally, and so has gotten less attention relative to the loan forgiveness policy that did not pan out. (Negativity bias again).

To get there, the Biden administration has made considerable progress in some areas that the Trump administration largely ignored or opposed, such making the Public Service Loan Forgiveness program, and support for students who were defrauded by for-profit universities, actually accessible.

It has also built on prior efforts. The OIRA report highlights loan forgiveness for students with total and permanent disabilities. Initially, individuals had to apply for this loan forgiveness, and many did not. After years of negotiation, the Department of Education worked with the Social Security Administration to use SSA data to identify those who were eligible. First, it contacted them, significantly boosting applications. Next, it moved to automatically discharge the debt. Finally, it eliminated income verification checks that were causing lots of eligible claimants to lose their benefits.

The case points to some lessons in burden reduction.

Third parties made the process more complicated: For a long time, those seeking disability-based loan forgiveness could not submit a single application to the government. If they had loans with multiple lenders, they had to work with each lender separately. About half of borrowers fell into this category, meaning they were reliant on on these third parties to complete the loan process.

Another third party that slowed loan forgiveness were physicians. Lots of people did not get the benefit because physicians did not provide the right information to verify their disability status. This did not make a lot of sense, because if you are totally and permanently disabled you have probably applied to the SSA for disability benefits. Which eventually led the government to decide, hey, why not use that data rather than asking people to undertake a separate medical determination?

Nudges helped, but auto-enrollment had a huge impact. Using SSA data to identify eligible individuals, the Department of Education contacted them in 2016, telling them to apply. Many did. But hundreds of thousands did not. Why? They may not have opened the mailer, or thought it was a scam, or did not understand what was offered. The informational nudge certainly helped, but it did not help everyone. Ultimately, using administrative data to automate the debt forgiveness process helped to extend the benefit to almost all who were eligible.

Its not just application processes; back end verification processes are a huge source of burdens. However, the benefit was tied to income, meaning that individuals had to document their income status every year to keep receiving the benefit. This became a huge, and mostly pointless, tripping point for borrowers. More than half of borrowers had their debt obligations reinstated for not completing income verification processes even though less than 1 in 10 had actually gone over the income limits. In other words, the documentation requirement generated the wrong inference by the government (that the person was not eligible) and the wrong outcome (debt being reinstated) the vast majority of the time.

There are two sub-points here. Sometimes burdens make government stupid, i.e. compels it to act on inaccurate inferences about the population they are serving. Second, the post-application process of verification is a really important venue where burdens can go unnoticed but matter a lot. (See also, tax audits of EITC recipients).

In this case, Biden’s Department of Education ultimately got rid of the income monitoring requirement. It accepted that some people might go over the income limit, but the vast majority would not. In other word, it engaged in a rough cost-benefit analysis of burdens, coming to the conclusion that income monitoring was vastly more likely to hurt low income borrowers than catch someone who was cheating the system (by working). More of this sort of sensible decisionmaking!

The process took a long time, and a lot of persistence. A federal rule had to be promulgated to allow SSA medical determinations to be acceptable. As with any federal rules, this was a slow process. It took years for the Department of Education and SSA to organize a data sharing agreement. Overall, it took more than a decade to go from a policy (people who have experienced permanent and total disability are not going to be in a position to repay student loans, so the government should forgive them) to the full realization of that policy, resulting in billions of dollars of forgiveness. This decade was spent engaged in incremental tinkering, successively identifying and reducing burdens. Other success stories, such as HUD simplifying application processes, also involve multi-year processes and rulemaking.

On the one hand, this might seem frustrating if you want progress quickly. On the other hand, it points to the fact that fundamental change is a long-haul process. It never ends. The wonks at the Department of Education now turn their attention to how to make other debt forgiveness programs, like income-driven repayment programs, more widely used.

This is what the bureaucratization of burden reduction looks like — and thats fine

A while back Rebecca Williams wrote an excellent essay subtitled “The Limits of Customer Service and Administrative Burden Frameworks.” She argued (or at least I interpreted her as saying) that the focus on customer service and administrative burden by civic technologists and government officials will drain attention away from the role of politics and power in creating those burdens. We become so focused on feasible solutions that we miss the big picture causes of problems.

Here, I think the Biden executive order deserves some credit. Unlike previous customer service efforts in the federal government, it directs attention to people who are eligible and not receiving government help, not just the people who make their way to government. Moreover, the customer experience EO explicitly calls for attention to “root causes” of burdens. From the start the Biden administration has tied reductions in administrative burdens to the broader goal of improving equity and racial justice efforts.

All that said, as government institutionalizes academic ideas like administrative burden, it is in its nature to bureaucratize them. By bureaucratize, I mean formalize, making legible using bureaucratic roles and processes. It also means focusing on what is technically and politically feasible via bureaucratic processes, which generally does not mean dealing with root cause structural sources of policy problems.

Thus, the source of OIRA’s power is that federal forms that collect data from the public have to be reviewed every three years. This is an incredibly nuanced vehicle by which an administration injects an administrative reform into government. There are others, including OMB Circular A-11, Section 280 (budget reporting guidance issued by OMB), or GPRA Modernization Act strategic planning and performance reporting requirements. These, not state of the union addresses or big pieces of legislation, have become the vehicles to mainstream burden reduction efforts.

This is what the bureaucratization of burden reduction looks like. And thats ok.

More precisely, this is the way that institutionalization works. The way in which administrative changes require “the slow boring of hard boards” that Max Weber described. A bureaucrat may know they cannot change eligibility requirements established in statute, but they can still look for solutions that mitigate the negative effects of those requirements within their domain of discretion. They are more likely to do so in the context of a formal requirement to identify burdens, more likely to do it well when that requirement is accompanied by well-reasoned justifications and guidance.

Just because all such work does not address structural factors does not mean it is trivial. The nature of bureaucratic organizations means that they will often miss the burdens they create. There might be no benefit, and even some budgetary reward, for imposing negative externalities on the public. Status quo bias favors not improving processes as new options become available, and negativity bias encourages adding more requirements to discourage fraud. Processes that correct for these tendencies by telling bureaucrats to prioritize burden reduction can therefore play a significant role.

The implications for researchers like myself is to continue to fight against administrative burdens in ways that are not possible within government. The implications for policymakers is that they and not the bureaucrats, are responsible for taking on the big structural questions.

Some notes for the future

Maybe I am biased here, because I think burden reduction is a good thing. Also, OIRA footnoted Can We Still Govern in their report, drawing attention to Jordan Kyle’s excellent guest blog about why progressives create burdens out of a desire to add nuance and inclusion.

So let me add a couple of critiques and suggestions.

First, in future years, there will be a need for more positive examples which were initiated under the Biden administration. Many of the strongest success stories cited have been in the pipeline for years, even if the Biden administration brought them to fruition. There is more to come. The State Department has been piloting online-only passport applications. The IRS is piloting direct electronic filing.

Second, we need progress on cross-agency measurement of burdens, ones which carries the same weight as cost-benefit analyses. For example, the OIRA report points to the Department of Homeland Security pushing to reduce the burdens it imposes on the public by about 10%, or 21 million hours. Without such specificity in metrics its going to be hard to convey to the public and policymakers the value and impact of burden reduction efforts. There is a risk of Campbell’s Law: once agencies start coming up with standard metrics of burden reduction, the metrics themselves will become corrupted. But the opposite risk seems more pressing: what doesn’t get measured, doesn’t get managed.

Third, if burden reduction is only an executive branch enthusiasm it will struggle to become embedded in agency culture. Ultimately, Congress needs to be more involved. Federal officials I’ve talked to wish that Congress would ask them more questions on burdens, as opposed to the scandal of the month. Until they do, the incentives for those officials is to focus on scandal prevention rather than long term strategies to fix problems.

1. Shout out to Annie Lowery who coined the term “time tax” and, to lots of scholars who have worked on the concept and empirical evidence around administrative burdens.