The $2 Trillion Experiment: What the CARES Act Taught Us About Stimulus

Thirteen Days

On March 27 2020 President Trump signed the CARES Act into law. Two point two trillion dollars the largest rescue package in American history was approved by the Senate by a vote of 96 to 0. No one voted against it. Thirteen days before that signing the economy was operating normally offices full restaurants open nothing unusual on the calendar. That speed is the first thing I try to get people to understand about this law because almost everything goodand the bad thing about it goes back to the choice to act quickly instead of carefully. Tax stimulusThe economy wasn't slowing it was grinding to a halt and twenty-two million jobs disappeared in a single month when the lockdowns hit

What Was Actually in It

The law was based on four pillars and it's helpful to keep them in mind before judging any of them. First direct payments: $1,200 per adult and $500 per child deposited in almost every household with an income limit. It was the fastest broad cash transfer the government has ever attempted. Second it supercharged unemployment insurance a $600-a-week federal supplement added to what the state normally paid workers extended for the first time.and the self-employed people who had never qualified for unemployment in their lives. Third the Paycheck Protection Program $349 billion at launch and completed in a matter of weeks once it ran out gave small businesses loans that turned into grants if they kept their staff on payroll. Fourth $454 billion allows the Treasury to support the Federal Reserve's lending programs money that the Fed could leverage several times over to shore up corporate credit municipal borrowing and midsize businesses with portions dedicated to airlines andto the most affected industries

What Worked

Compared to what it was actually trying to do stop a depression the CARES Act worked better than almost anyone had predicted in real time. Household incomes rose during the sharpest economic contraction in modern history. Read that again. It's the only recession on record in which that happened. The personal savings rate hit an all-time high of over 30 percent in April 2020 because people were receiving cash and had nowhere to spend it. Bankruptcies stayed low.They suspended evictions. When the economy reopened demand returned quickly rather than dragging because balance sheets had been protected rather than ruined. Economists choose 2008 as a comparison point almost reflexively. Smaller slower support then produced a recovery that lasted the better part of a decade. This one took months

A Worked Example: What One Family Actually Got

Figures like $1,200 per adult are easy to read so let's run the actual program in a home. Imagine a family of four two adults and two children under 17 with income below the maximum limit. Direct payments only: $1,200 times two adults is $2,400 plus $500 times two children is $1,000 for a total of $3,400.deposited in a few weeks. Now let's say one of the two adults is laid off and spends eight weeks on unemployment before finding a new job. State benefits vary widely so we isolate just the federal part: $600 a week for eight weeks is $4,800 on top of what the state itself paid. Add that to the $3,400 in direct payments and this illustrative family received about $8,200 in CARES Act money in about two weeks.months before counting a dollar of state unemployment benefits which in most states would have added several thousand more. That's how much money the law pumped into a single ordinary household. Multiply it by tens of millions of households and the rise in the savings rate no longer seems mysterious

What It Cost Beyond the Price Tag

Speed sends three bills and they all arrived later. The first was targeting. Giving money to almost everyone meant giving a lot to homes and businesses that didn't need it and the $600 supplement paid many workers more than their old paycheck a real point of political tension once the initial panic subsided. The second was fraud. Later government estimates put tens of billions of dollars of PPP and unemployment money in the hands of ineligible or direct criminal claims the costpredictable way of moving money at maximum speed with minimal verification. The third is inflation and it has its own section below instead of a paragraph here because it is the part of this story that is still genuinely discussed

Case Study: Shake Shack and the Targeting Problem

The clearest small-scale example of the targeting problem is Shake Shack and I keep coming back to it because it's small compared to the trillions elsewhere in this law and telling nonetheless. In April 2020 Shake Shack received a Paycheck Protection Program loan supposedly around $10 million the kind of loan that existed to keep cash-strapped small businesses afloat. Except Shake Shack was a chainof publicly traded restaurants with access to the stock markets and credit that a neighborhood restaurant could only dream of and which had already raised tens of millions of dollars from investors days before. Once this became public the reaction was immediate. Within days CEO Danny Meyer announced that the company would repay the loan in full saying that Shake Shack had access to capital that the program's intended targets simply did not have. No one broke a rule. A publicly traded company with enough technically qualified employees. That's itexactly the problem: a first-come first-served design built for speed couldn't distinguish between a restaurant with no other options and one that could raise money on the stock market with a phone call

The Inflation Debate: Demand or Supply?

This is where I think the honest answer is that no one has completely figured this out and I'm skeptical of anyone who tells you otherwise with complete confidence. In 2022 the United States was going through the worst inflation in about forty years and the CARES Act was just the first chapter in a much longer pandemic support story about $5 trillion in total once you count the March 2021 round of checks and everything that came after. The real question is how much of that inflation actually caused thestimulus

The case for the demand side is this. Households ended 2021 with savings that were still high relative to before the pandemic money accumulated while spending options were limited and then topped off with another round of direct payments in March 2021. This is a lot of purchasing power hitting an economy at once and when demand increases faster than the economy's ability to produce goods and services prices are the safety valve. Economists likeLarry Summers argued exactly this loud and early before it was fashionable to say it and once the Federal Reserve itself removed the word transitional from its own vocabulary at the end of 2021 that argument seemed much stronger in retrospect than it did a year earlier

The supply-side case is equally real. Global shipping remained entangled for the better part of two years. Semiconductor shortages left auto lots half-empty which alone drove up new and used vehicle prices sharply and vehicles are a significant portion of the inflation basket. China's factories closed under the zero-COVID policy at unpredictable intervals.an event that has nothing to do with any stimulus check ever sent out. None of that is demand. It's about the economy's ability to produce and ship things collapsing at the same time demand was healthy and prices were rising because supply couldn't keep up and not because someone had too much cash

My honest read is that both stories are true at the same time and the real argument is about proportions not which side is right. A supply shock alone probably produces a bad run of inflation that fades faster than it actually occurred. Demand so strong hitting an economy with normal supply chains probably produces a milder version of what really happened. Having both at the same time is what produced something so severe and persistent. If you ever hear someone assign a precise percentage to the inflation ratio of2021 to 2022 that was stimulus vs supply chains ask how they isolated both. In such a tangled economy I don't think anyone can do that cleanly and I would be skeptical of a specific number expressed with confidence either way

The Precedent

Perhaps the most important legacy is the one that never appears on a graph. The CARES Act demonstrated that the government can when it chooses deposit cash into most American bank accounts within weeks of deciding to do so. That ability once demonstrated does not go away. It becomes a permanent line in the policy toolkit and a permanent public expectation. Since then every crisis begins with some version of the question of when the checks arrive and every discussion about moral hazard deficits and inflation now occurs morebelow that expectation rather than beyond it. Economists will spend decades arguing whether the package should have been smaller better targeted or canceled sooner than it was. Hardly anyone disputes that this should not have happened at all

How I Actually Think About Stimulus

When I read about a new stimulus proposal now pandemic relief disaster response or just a package to combat the recession in an ordinary crisis the CARES Act is the mental model I turn to first. The way I would actually use this story is as a checklist not a verdict

First I ask what the trade-off between speed and accuracy looks like in this specific case. The CARES Act went for speed and got a really fast recovery combined with real waste. That was probably the right decision for such a sudden shutdown. It's not automatically the right decision for every crisis and I try to ask myself if the emergency really requires days instead of weeks before assuming that more speed is always better

Second I look at who the money actually goes to before trusting the framework offered by the program's proponent. Shake Shack's obtaining a PPP loan is a small story compared to the billions appearing elsewhere in this law but it's exactly the kind of detail that tells you whether a program's design matches its stated purpose or simply its stated size

Third I now treat any major stimulus figure as a question of future inflation and a question of today's relief at the same time without assuming that it was the single cause of the last episode. I was wrong the first time I read about the CARES Act when it was passed. I remember thinking of it simply as emergency aid and not relating it at all to what would appear in the price of a used car eighteen months later. That connection is obvious now. It wasn't obvious to me then and closing that gap goes a long way.what this series is trying to do

None of this is a call to buy or sell anything and I'm not in the business of predicting the next response to the crisis. It's more like a series of questions I now ask out of habit whenever a headline says the government is about to send money to almost everyone at once

The Bottom Line

The CARES Act was a $2.2 trillion bet that overreacting quickly is better than underreacting carefully. On its own terms the bet paid off. The depression never came and the recovery was the fastest on record. The cost of that speed was waste fraud and a supporting role along with supply shocks that no one fully controlled in the worst inflation in forty years. For anyone who studies policy this is the clearest lesson of the decade about the true compromise behindthe answer to the crisis: speed precision and cost. You can only choose two