Why $1.4 Billion Sent to Dead People is a Policy Success
No policy is perfect, so we should design them to fail correctly
On March 27 the CARES Act was passed, which included $1,200 direct payments to every adult as means of financial stimulus. On June 25, the Washington Post reported that an estimated $1.4 billion worth payments went out to people who were actually dead. The article described these erroneous checks as a major failure, saying that:
The news that so much money has gone to the dead could add to reluctance from some Republicans to agree to more direct relief payments.
I’m going to offer an alternative perspective: these errors are totally fine! While I am no fan of Secretary Mnuchin or President Trump, if these are the only issues with the stimulus money, then the program was a huge success! In fact, I want to shift the way you think about governing and policy!
False Positives, False Negatives, and Speed
Let’s consider a thought experiment: a country is choosing what method to use for Coronavirus testing. There are two methods — one has a 10% chance of giving a COVID-positive result to someone without the virus (false positive), and one has a 10% chance of giving a COVID-negative result to someone with the virus (false negative).
Which test is best? The “false negative” test seems risky, since 10% of those with COVID would go on spreading the disease. The “false positive” method has some downsides — a lot of people without COVID would be forced into isolation and otherwise inconvenienced. Still, if your goal is eliminating the disease, it is preferable to over-diagnose rather than under-diagnose. All things considered, I would argue that the method more likely to give “false positives” is better.
Some level of error is inevitable in any type of testing, so what kind of mistake are you more willing to make? All “tests” in society force policy-makers to decide how tolerant they are of false positives and false negatives. For crime, are you more comfortable with innocent people being locked up, or guilty offenders going free? For new airplanes, would you rather slower innovation or more plane crashes? There are countless examples.
There will be some number of false positives and false negatives in every real-world example, but institutions need to find a balance given what they are evaluating. Airplane crashes are horrible, so there are very strict regulation around aerospace technology. On the other hand, criminal justice uses the standard of “innocent until proven guilty,” which is meant to reduce the chance that an innocent person is convicted (well, at least in theory).
What if you want to make certain there are no mistakes? This requires more time-consuming tests. I recall a meme from my time in university that asked the reader to “pick two” of the following: “good grades”, “social life”, or “sleep”.
There is clearly some truth to this — there are only so many hours in a day, and you need to decide how you are going to spend them. In the case of developing a “policy test” you can imagine a different pick two: “low false positives”, “low false negatives”, and “time efficiency”.
This discussion about policy may sound over-simplified, but these trade-offs are quite real. It is extremely hard to develop policy that are both perfectly accurate and rapid. Well-intentioned legislators and technocrats would all prefer systems that are easy to use and error-free, but that simply isn’t possible in the real world.
The Right Kind of Mistake
Let’s run another thought experiment: you are designing the CARES Act. You are told there are two plans that can be implemented right now. The first has a chance of giving money to some number of dead individuals. The second plan has a chance that some eligible folks will not get their payment. Alternatively, you could take more time and try your luck developing a better program.
The CARES Act was passed at the end of March, which was a scary time in the economy. Employment numbers were in free-fall, and the stock market looked like it could crash. Delaying the program to develop a better approach would be very risky (and politically unpopular).
False negatives — meaning some individuals not receiving money — would be more undesirable than false positives. Millions of people had lost their jobs and were desperate to get cash to meet basic needs until unemployment benefits kicked in. Alternatively, sending some of the money to dead people is not particularly risky. At this point we don’t know how much of that money was actually claimed — some checks were likely sent back to the government, or are still sitting in their unopened envelopes. In these cases, the mistake of sending out the money cost roughly the postage expenses.
In some number of cases, family members pocketed their dead relative’s money. That’s unfortunate, but is it a disaster? It is unfair that some folks got $2,400 while the rest got $1,200, but it is hardly a crisis. Besides, the purpose of the stimulus money was to cover living expenses, and help families as they waited for unemployment benefits. The fact that some families got a bit of extra assistance does not undermine the goals of the program.
Inefficiencies versus Administrative Burden
The purpose of this article is not to heap praise on the administrative nuances of the CARES Act. It has other issues, including some people who have not yet received their $1,200 checks. Rather, the purpose is to adjust expectations of government programs. Not all “scandals” are created equal. Over the course of Scott Pruitt’s tenure in Trump’s cabinet, government funds were used to cover the cost of his security and travel expenses for his family. The overall cost of Pruitt’s indulgences is far less than the extra $1.4 billion sent out by the CARES Act. The fact that Pruitt’s situation looks like deliberate corruption makes his actions much more troubling.
The CARES Act included $269 billion for direct payments. The $1.4 billion that went to deceased individuals represents roughly 0.5% of the total. An error rate of 0.5% is exceptionally low, and while we would prefer it to be 0%, that is simply not realistic.
Stories about inefficiencies in government programs are often weaponized by politicians. Most famously, the “Welfare Queen” became an archetype of government waste in the welfare system. Such images were used as a cudgel against the welfare program as a whole, when the vast majority of the beneficiaries were not abusing the system. Did some people con the program? Yes, but that doesn’t mean the whole system is a failure. By becoming obsessed with stopping over-use of welfare (false positive) the consequence is making the program too difficult to access for others (false negative).
An unfortunate consequence of defining social programs by their inefficiencies is the creation of administrative burdens. We all have the experience of dealing with government bureaucracy and the associated frustration, but regulation is not a “natural” phenomenon. Red tape is often deliberately wrapped around programs to make them more difficult to access. On the surface, requests for extra documentation may seem logical, but these can be used to block deserving people from accessing government programs.
These barriers are also not equal for all people. Imagine if the CARES Act instituted the following restriction to collecting the $1,200: you must log on to a government website and fill out a simple form that requires basic information like date of birth, social insurance number, and current address. The vast majority of people would fill out this form, but who would have difficulty completing this task? Those living in poverty may not have access to the internet, or a fixed address. Individuals who are illiterate, non-English speaking, elderly, or disabled may struggle to complete the form. These are also the people who need the assistance most. It should also be remembered that these barriers are amplified in the time of COVID, since it was impossible to get help in a government office or community center.
This online form is the mildest administrative burden possible — imagine if applicants were asked to demonstrate proof of residency or photocopies of government IDs. Once again, these disadvantaged communities would feel the weight of these added burdens far more than individuals from privileged groups.
Obviously, some level of regulation is important to ensure government programs are not subject to flagrant abuse, but effort should be made to keep the administration as simple as possible. In the case of the CARES Act, more red tape probably would have lowered the payments to deceased individuals, but that would have cost precious time, and led to increased burdens on deserving recipients.
Understanding the trade-offs in policy making is important. All government programs will experience some inefficiency, and it is worthwhile to develop programs that minimize that waste, but we should not cut off our nose to spite our face. Administrative burdens should only be increased when necessary. In the case of the CARES Act, the $1.4 billion in extra payments are a very reasonable price if it means more people from disadvantaged groups were able to receive their money.