How Administrative Burdens Make Government Less Effective and What to Do About It

by Pamela Herd et Donald Moynihan

 

 

Bureaucracy, confusing paperwork and complex regulations – or what Pamela Herd and Donald Moynihan call administrative burdens – often introduce delay and frustration into our experiences with government agencies. Administrative burdens diminish the effectiveness of public programmes and can even block individuals from fundamental rights like voting. Because burdens affect people’s perceptions of government and often perpetuate long-standing inequalities, understanding why administrative burdens exist and how they can be reduced is essential for maintaining a healthy public sector.

In this article for Government in Action: Research and Practice, Pamela Herd and Donald Moynihan summarise, for the benefit of practitioners, the issues they explore in their recently published book Administrative Burden: Policymaking by Other Means.  

 

 

The frictions that people encounter in their interactions with government – searching for information about a program, the filling out of forms, and the accompanying frustrations – are easy to dismiss as a nuisance necessary to the functioning of the administrative state. But such administrative burdens play a more fundamental role in the quality of governance than traditionally acknowledged. They are more than a mere irritant: they affect whether people will be able to exercise fundamental rights of citizenship, like voting, or can access public services that can improve their quality of life. Burdens alter the effectiveness of public programs. Government programs aimed at reducing inequality often reach only a fraction of their target population, automatically weakening their effectiveness by shutting out those who fail to negotiate the required procedure. Burdens may sometimes be unavoidable – costs can serve important political values or reflect administrative realities – but in many instances the value of burdens seems, at best, debatable, and at worse, entirely unjustified by the costs they impose.

Administrative burdens therefore are central to big questions about the administrative state: Are people able to enjoy in practice the rights and benefits provided to them in theory? Are public policies working? How does the state facilitate or minimize inequality? To answer these questions we need to be able to ask yet another set of queries: How do we recognize burdens? How do we determine when they are unjustified? How do we design and manage public programs in ways that shift administrative burdens away from citizens? Whose role is it to shine a spotlight on these burdens, and how do they make determinations? To frame such practical questions, however, first requires the type of conceptual language we present in this paper. In doing so, we offer a toolbox to make government simple, accessible, and respectful to the citizens it serves.

What are administrative burdens? They are the experience of policy implementation as onerous (Burden et al. 2012). We focus on the costs that people encounter when they search for information about public services (learning costs), comply with rules and requirements (compliance costs), and experience the stresses, loss of autonomy, frustrations or stigma that come from such encounters (psychological costs) (Herd and Moynihan 2018). Table 1 provides more detail.

 

Table 1: The Components of Administrative Burden
Learning costs	Time and effort expended to learn about the program or service, ascertaining eligibility status, the nature of benefits, conditions that must be satisfied, and how to gain access.
Compliance costs	Provision of information and documentation to demonstrate standing; financial costs to access services (such as fees, legal representation, travel costs); avoiding or responding to discretionary demands made by administrators.
Psychological costs	Stigma arising from applying for and participating in an unpopular program; loss of autonomy that comes from intrusive administrative supervision; frustration at dealing with learning and compliance costs, unjust or unnecessary procedures; stresses that arise from uncertainty about whether citizen can negotiate processes and compliance costs.

Adapted from Herd and Moynihan 2018

 

We limit our focus to the experience of burdens by individuals. An existing literature already addresses regulatory burdens on businesses. International organizations such as the World Bank and OECD encourage governments to adopt a less burdensome regulatory regime. The burdens on individual citizens are, by comparison, an after-thought. Such an imbalance reflects the success of private organizations in making their case for lower regulation. By contrast, there is no well-organized opposition to administrative burdens. Nonetheless, complaints about the hassles of dealing with government are constant in almost every public setting, with much of the dissatisfaction that people have with government coming from negative experiences with complex and frustrating processes.

 

Why Care about Administrative Burdens?

Governments should pay attention to burdens because they have real effects on the lives of the public. Administrative burdens can impose significant costs on citizens, especially if they result in programs failing to reach their intended recipients.

 

Burdens are Consequential

One reason we don’t fully understand the impact of administrative burdens is there is a paucity of studies modelling their effects. One important exception is a study based in South Africa. Here, Carolyn Heinrich (2016) examined an ambitious cash-transfer program, the Child Support Grant, which aimed to improve economic security for 11 million children. By giving these families access to cash, the program was intended to increase home stability, and educational outcomes. However, participants faced significant compliance costs such as extensive documentation requirements, delay at welfare offices, and learning costs that were exacerbated by changes in the policy rules. As a result, most beneficiaries experienced disruptions in cash transfers, with four out of five of those disruptions made in error. The lost resources had a negative effect on adolescent outcomes resulting in greater rates of sexual activity, alcohol consumption, and criminal behavior. In other words, a program designed with the best of intentions was undercut by administrative burdens.

Burdens also matter to whether people practice basic democratic rights, such as voting. When we think of voting, we think of the citizen casting their ballot at a polling place. But voting is just the final stage in the election process. Registration is the more consequential part of the election process when it comes to burdens. Countries with a requirement that voters register separate from the voting process add a layer of learning and compliance costs onto political participation, which reduces turnout. Studies from the US find that states that allowed people to register on election day had turnout that was three to seven percentage points higher, controlling for other factors (Burden et al 2014).

A field experiment from France illustrated how an intervention that reduced administrative burdens in registration could increase turnout (Braconnier, Dormagen, and Pons 2017). Conducted as part of the 2012 French Presidential and parliamentary elections, the researchers conducted a randomized control trial with 20,500 homes. Compared to a control group, citizens who had campaigners come to their homes, offer information, and help them register were more likely to register and vote. Visits increased registration by 29 percent, and 93 percent of those who registered because of the visit voted in at least one election since 2012. Citizens who received such help also increased their civic knowledge, becoming more interested and knowledgeable about the election according to post-election surveys. The study offers strong causal evidence that both learning and compliance costs are barriers that lower registration, and that reducing those costs helps foster more engaged and active citizens.

 

Burdens Matter to Inequality

Studies of political participation also point to a second reason why governments should care about administrative burdens: the effects are distributive. Burdens hurt some groups more than others, and therefore can worsen existing inequality in society. In the case of voting, higher income groups participate more because they are more likely to be registered: low income groups vote at similar rates to higher income groups if they are registered (Herd and Moynihan 2018). Therefore, policies that make registration easier facilitate the participation of low-income, younger and minority groups. The greatest beneficiaries of the French experiment to help people to register from their home tended to be immigrants, those who did not speak French at home, and younger and less educated citizens (Braconnier, Dormagen, and Pons 2017). Similarly, a study of changes to US election policies between 1978 and 2008 found that easing the registration process increased the share of low-income voters (Rigby and Springer 2011). In other words, reducing burdens makes the electorate more representative of the citizenry.

Burdens can make it more difficult for vulnerable groups to claim standing and fully participate in society. In Pakistan, Nisar (2018) documents the burdens that gender non-conforming individuals face in accessing identity documents, which have become increasingly necessary to get a job, claim property rights, access welfare services, or negotiate encounters at security checkpoints. More vulnerable groups who are classified as less deserving are subject both to being targeted by higher burdens, as well as informal discrimination by bureaucrats.

Policies targeted towards the poor are more likely to be burdensome relative to universal policies that all people use. In the United States, compared to the near 100 percent take-up for more universal programs like Social Security and Medicare, take-up rates by eligible beneficiaries of means-tested programs typically aimed at poor people in the United States are much lower (Herd and Moynihan 2018): 40 to 60 percent for Supplemental Social Insurance, two-thirds for the Supplemental Nutrition Assistance Program; 30 to 60 percent of unemployment insurance benefits; about 50 to 70 percent for Medicaid. For the Earned Income Tax Credit, a reimbursable tax credit tied to work for low-income earners, the take-up rate is about 80 percent. Aid to Families with Dependent Children had an estimated take-up rate of between 77 to 86 percent, participation rates declined dramatically after 1990s welfare reform. Its successor, Temporary Assistance for Needy Families (TANF), has a much lower take-up rate of between 42 to 52 percent. A central difference between these programs and universal programs such as Social Security is that means-tested programs must do more to distinguish between the eligible and ineligible, and in creating administrative processes to do so, they add more burdens.

Human capital – the stock of innate abilities and characteristics that people possess, and knowledge and skills they acquire over time – can further exacerbate the ways in which burdens contribute to inequality. Those with more human capital will be generally better able to overcome administrative burdens. For example, the voter in better physical health is more likely to walk to the local polling station (Burden et al. 2017). People with more social connections or education have a greater ability to learn about a program and understand the requirements they need to satisfy in order to participate. Those with higher technological literacy or greater perseverance will be better able to negotiate online administrative processes. The result is that people who might need services the most – such as the sick, those in poverty, or older adults facing cognitive decline – also struggle the most to overcome burdens to access those services. For example, a wide variety of studies point out that people put in a position of resource scarcity have lower cognitive performance (Mullainathan and Shafir 2013). They can be expected to struggle with administrative processes that reward diligence and perseverance.

An example of how human capital matters to the effects of burdens comes from the US state of Arkansas, which in 2018 adopted work requirements for beneficiaries of Medicaid, public insurance for the poor. Unsurprisingly, thousands of Medicaid beneficiaries soon lost coverage because they could not manage the compliance costs involved. Coverage of low-income adults declined from 70.5 percent to 63.7 percent, even though 95 percent of those who lost benefits were actually completing the required work (Sommers et al. 2019). The problem was that they could not overcome the onerous reporting processes. Learning costs played a significant role. One third of those affected had not heard about the new requirements, and 44 percent were unsure if they applied to them. Compliance costs also played a role: the state required that reporting processes be online only, but one-third of adult Medicaid recipients in Arkansas lacked access to the internet. In short, a policy aimed at encouraging work failed in that goal, but instead caused large numbers of low-income beneficiaries to lose health insurance because of administrative burdens.

Reforms that promise to simplify administration may actually make things worse if they do not think through how they impose burdens on citizens. In the United Kingdom, reforms to the disability payment system, specifically the switch from the disability living allowance to the personal independence payment substantially increased compliance costs. The goal of the reforms was to reduce spending while making the system “fairer, more straightforward to administer, and for it to be easier and clearer to understand” (Office for Budgetary Responsibility 2019, p. 62). The promise of lower burdens gave way to a reality of participants undergoing more frequent reassessments for eligibility and benefit levels. Neither spending on the program nor the number of beneficiaries has been reduced by the more stringent evaluations, but it has led to increases in delays in receiving benefits and substantial compliance costs for beneficiaries (Office for Budgetary Responsibility 2019). For example, an independent review of the program in 2017 found that 65 percent of benefit denials that were appealed were overturned. This high error rate is compounded by delays, with the average appeal process taking 31 weeks (BBC 2017). Moreover, complaints regarding the assessments have risen rapidly, from142 in 2015-16 to 9,320 between 2018-19 (Bulman 2019). A representative of the Multiple Sclerosis Society, noted: “This surge in complaints is not at all surprising, given the assessment process is so completely inadequate for disabled people. We frequently hear about how the process causes great anxiety, stress and harm to people’s health. Even those who’ve had increased support say the process takes a toll on their emotional and physical wellbeing.” (Bulman 2019).

 

How to Reduce Burdens

Governments are fond of reforms to reduce red tape on businesses, but are less attentive to reducing administrative burdens on individuals. In this section, we briefly consider what a reform agenda might look like.

 

Provide Information, Simplify Processes and Offer Help

For governments willing to reduce burdens, a variety of options are at hand. Some of these center on reducing learning costs and some on compliance costs (there has been little research on efforts to directly minimize psychological costs).

Reminders and outreach can reduce learning costs. One field experiment found that reminders that told potential beneficiaries in simple language and with specific dollar numbers what they might gain from participating increased participation in a US program that provided a tax reward for the working poor (Bhargava and Manoli 2015).

However, for some administrative burdens, telling people they are eligible, or reminding them of deadlines is not enough. For example, in the case of the French voter registration experiment, those who were given help registering to vote, thereby minimizing compliance costs, were more likely to vote than those who just received information. The key point is that for demanding administrative processes, people need more than to be pointed in the right direction.

One strategy to reduce compliance costs is to simplify processes. For example, reducing the frequency of renewal processes, allowing people to do interviews on the phone rather than in person, or having a single form for multiple overlapping programs are all proven techniques to increase participation in social programs (Herd and Moynihan 2018).

If processes can’t be simplified, another strategy is for government or other actors to directly help people to deal with compliance costs. An example of the value of help comes from a cash transfer program designed to help poor women in India. The program struggled with low participation, with only one-third of eligible participants being enrolled. A field experiment with over 1,200 people examined the effects of different interventions designed to increase enrollment. Providing information increased participation, but only for those literate enough to read the information. Providing help in filling out forms, or working directly with bureaucracies to claim benefits, had large positive effects, increasing applications by 41 percent and 70 percent respectively. The findings also reinforce the distributive effects of burdens. Directly helping claimants had the greatest benefit on the most vulnerable groups even among a relatively poor eligible population, most benefiting those who were illiterate, lacked political connections, or had limited autonomy to partake in activities outside the household without permission (Gupta 2017).

Two field experiments in US higher education further illustrate the potential of offering help. One found that telling families that they were eligible did little to increase participation in a program that provided financial aid for higher education, but that simply helping families complete the complex forms led to dramatic increases in applications and a 29 percent increase in actual college enrollment (Bettinger et al. 2012). Another study examined a program that was designed to make it easier and cheaper to repay student debt. Simply sending people a electronic pre-populated form that they only had to sign and return increased participation from 24 percent to 60 percent (Mueller and Yannelis 2019). Providing such help is made vastly easier with information technology that can pre-populate forms, or provide calculators for people to estimate benefits.

 

Shift Burdens onto the State Using Information Technology and Administrative Data

Information technology can be used to reduce administrative burden. For example, simply having an online option can reduce the time it takes to travel to a physical location to collect and submit a form. By investing in IT systems that can integrate data across programs, the state can reduce the need for applicants to provide the same information multiple times, while improving accuracy. For example, information based on tax returns is more likely to be accurate than self-reported income data in verifying eligibility.

The goal of reducing compliance costs is at the heart of Estonia’s once-only principle: the idea that citizens should be asked to share information with the state just once, rather than again and again (Heller 2017). This reduces the compliance costs of citizen-state interactions as governments move more and more of those interactions – such as paying taxes, enrolling in school, or renewing prescriptions – online. Service providers, such as doctors, can pull up client’s records (but not other aspects of their digital profile) rather than have them fill out forms. The technology depends on a mandatory chip ID card, pin numbers and blockchain, but it is the relentless commitment toward a vision of a digital state that sets Estonia apart.

India’s national identification system, Aadhaar, is a huge bet on the potential for information technology to reduce burdens on an extraordinary scale. It has enrolled over a billion people, tying 12-digit number to photos, iris scans and fingerprints. The stated purpose of the program was to replace the compliance costs and corruption that came from providing food entitlements via paper cards. Aadhaar been controversial, raising privacy concerns. Even as the Indian Supreme Court allowed the program to continue for government services, it prevented private companies from demanding the new ID, and urged the Indian government to pass a data protection law to restore faith in the program. On the other hand, a forerunner to the Aadhaar system using biometric cards in the Indian state of Andhra Pradesh demonstrated that such technology reduced compliance costs by saving participants time engaging with the state, and minimized “leakage” of payments between the government and beneficiaries. Nine out of ten users preferred the smartcard approach to the old system (Muralidharan, Niehaus, and Sukhtankar 2016).

Another option is to use administrative data to automatically register individuals deemed eligible for programs or rights (Herd and Moynihan 2018). For example, seventeen US states have adopted auto-enrollment for voter registration. A technique like auto-enrollment dramatically reduces the psychological costs of opting-in, as well as learning and compliance costs. Some of these costs might arise if the individual chooses to stay in the program, but the barriers to entering the program are significantly reduced.

Technology offers both great promise but also great risk; it can be used to reduce burdens but can also enhance them. Automated systems might be used to intrusively target people with hassles and rob administrators of flexibility to ameliorate burdens when the circumstances require it. More generally, without proper data protection and data privacy rights, it could leave people more vulnerable to intrusive or autocratic policies. Technology, or any other innovation, will only reduce burdens if that’s the goal of those controlling it.

 

Recognize the Role of Third Parties

While we have discussed citizen-state interactions, many public services are provided by third-parties, who have the opportunity to buffer or amplify burdens just as public employees can, but with greater flexibility. A good example comes from a field experiment in Belgium. Jilke, Van Dooren and Rys (2018) use an ingenious design to demonstrate how people encounter learning costs differently in public-private settings. The researchers sent emails from fictional retirees seeking information about how to apply to Belgian nursing-homes. Half of the emails came from a typically Flemish name, and the other half came from an Arab-sounding name, a standard research technique to detect discrimination among recipients. Private providers were about 20 percentage points less likely to provide information to the Arab applicants. The findings confirm that the selective provision of information to minimize learning costs is one tool that organizations can use to exclude participants, and in this setting at least, private actors were more willing to use it to limit access to public services. The influence of third party providers on administrative burdens can be particularly pernicious when they stand to profit from such burdens. Private providers, contracted by the state of the Florida to provide services and supports for welfare beneficiaries, generated unnecessary burdens for recipients, which reduced their use of services and increased profits for the provider (Soss et al. 2011).

The role of third parties in creating burdens raises a regulatory challenge for governments. Governments regulate private actors that provide public services, monitoring adherence to standards, quality of services provided, and financial performance. Governments could also regulate for administrative burden, ensuring that providers do not profit by making public services more onerous than they need to be. Such regulation could include examining how transparent and responsive private service providers are, how onerous are their application processes, and surveying users to assess how well they are treated.

 

Invest in Administrative Capacity

While we offer suggestions to shift administrative burden onto the state, such a shift is not a free lunch. These techniques demand smart and capable government, with investments in information technology and administrative data. Estonia’s model had been successful because it was supported by public opinion in favor of digital democracy, as well as a massive investment in secure digital portals that can share information when a citizen tries to pay taxes, open a business, or go to a doctor.

Financial resources, administrative expertise, and organizational capacity all influence the degree to which the state can reduce administrative burdens on citizens. Financial resources affect the ability of the state to minimize or shift burdens away from citizens. This trade-off is easy to miss if we only examine government costs in the provision of services, and neglect citizen costs. The use of biometric cards in Andhra Pradesh cost $4.1 million for the government. However, it saved an estimated $4.3 million of citizen time, and increased the flow of benefits by $32.8 million. (Muralidharan, Niehaus, and Sukhtankar 2016). Capacity also includes administrative expertise. Ultimately, administrative actors design the systems that individuals interact with to obtain benefits. Those with more expertise can use those skills to design more easily accessible systems, or to put more barriers in place.

Heinrich’s (2016) study of South African cash transfer programs underlines the importance of a functioning administrative state. Quotes from program participants illustrated that the problems they encountered were magnified by a slow and unresponsive bureaucracy:

You can lose your ID and go to Home Affairs to get another one, and you find that you do not get it for a long time and so you cannot register. When you get there they tell you to go and get an affidavit, and when you come back they tell you it is wrong.

Heinrich argues that the “bite” of administrative burden may be larger in poorer countries, given that the state has less ability to identify and resolve burdens, and individuals with lower human capital, such as literacy and familiarity in dealing with bureaucratic procedures, are less able to overcome burdens.

 

Conclusion: Embedding Attention to Administrative Burdens

Policymakers and public managers should regularly evaluate the benefits of burdens with a bias toward reduction. The concepts outlined in this paper provide a framework to recognize and evaluate burdens, complemented by some simple principles.

Consideration of administrative burdens should be evidence-based. A rational approach to policymaking would consider both the costs and benefits of burdens. Policymakers and public managers need to weigh the relative benefit, for example, of fraud reduction in welfare programs, with the relative costs of limiting access to benefits for which individuals are eligible. Such considerations must be informed by empirical evidence to the greatest degree possible.

Public organizations should cultivate a professional norm of assessing burdens. In aggregate, however, the language or concepts of administrative burdens are not impressed upon public managers or frontline employees in the way, for example, that the language and concepts of strategic planning or performance measurement have become ubiquitous. Public managers do not face administrative burden equivalents to routine strategic planning, performance reporting, or audit requirements. And while standard cost models and cost-benefit analyses direct attention to administrative burdens, typically these tools focus only on how businesses experience state-created burdens, not on the experience of individuals (although Germany provides a partial exception (Nationaler Normenkontrollat, 2013). Training for the public service could easily incorporate the type of concepts discussed here.

There is no single best way to apply the insights of the administrative burden framework to practice, but there are more and more tools that can help. A human-centered design perspective is inherently attuned to mapping and avoiding the burdens that people encounter in public services (Milkowska 2018). Sunstein (2019) urges governments to apply the tools of cost-benefit analysis to what he labels as “sludge” in public services, via mandates that costs of paperwork burdens should not exceed the benefits. Further, he argues that governments could prioritize sludge-reduction by identifying vulnerable groups they want to help, or policy areas where burdens are having an outsize impact on key political goals. For managers of public organizations, Herd and Moynihan (2018) offer simple diagnostic tools (e.g. Table 2) that public organizations can use to help them identify where learning, compliance and psychological costs arise, and setting the stage for discussions of solutions. As such tools become more available and better understood, they promise to help the public sector to improve the experience of public services.

 

Table 2: Diagnostic questions about administrative burdens

Take-up: What is take-up rate for eligible beneficiaries? Inequality : Does take-up rate vary across populations? Learning costs Is it easy for potential participants to: find out about the program? establish if they are eligible? understand what benefits are provided? learn about application processes? Compliance Costs How many questions and forms are there to complete? How much documentation is needed? Does the participant have to input the same information multiple times? Is the information sought already captured via administrative data? Is it possible to serve the person in a less intrusive way, e.g. phone rather than in-person interviews? Do applicants have easily accessible help? How frequent is re-enrollment? How much time must people commit to the process? What are the bottlenecks? How much financial costs must people commit? Psychological Costs Are interactions stressful? Do people receive respectful treatment? Do people enjoy some autonomy in the interaction?

Source : Herd et Moynihan (2018)