This title sounds like an oxymoron. Almost by definition, bailouts are hastily arranged affairs, negotiated behind the scenes by lobbyists and legislators, with everyone involved seeking their own advantage. Much in the CARES Act fits that description, particular the hidden provision that allowed each restaurant in a restaurant chain and each hotel in a hotel chain to separately qualify as a small business and obtain loans from the CARES Act's centerpiece, the Payroll Protection Program (or PPP). We have now been through two rounds of the PPP, and it is time for some stock-taking. How well did it work? How could it be redesigned to be more efficient or fairer—or both?

Admittedly, we are also at a moment when Congress would not throw a life preserver to a drowning man—without first ascertaining his political party (and then would deadlock with one House supporting and the other opposing the toss of the life preserver). Holding that small political difficulty aside, how should a bailout that intends to save the jobs of employees at small businesses be structured? This analysis will avoid political name-calling and assumes that at some point Congress may be willing to reconsider the CARES Act, which has received almost universal criticism.

The initial key decision made by the CARES Act was to use the Small Business Administration (SBA) and have it guarantee loans made by banks (and other lenders) to qualifying small businesses. Ordinarily, this is a time-proven technique which allows the government to maximize the impact of its subsidy. Take, for example, the paradigmatic case of the Federal Housing Administration (or FHA). When it guarantees home mortgages, this allows banks to make loans to less creditworthy borrowers and the cost to it is only the cost of those loans that default. If the default rate is, say, 4%, this may effectively enable banks to make 100 loans at the cost to the FHA of the four that default—in short, this technique enables the agency to leverage the subsidy and create a greater bang for the buck.

But does this strategy have any real applicability to PPP loans? The short answer is no, because PPP loans were intended to be forgiven (and not repaid), so long as the loans are used for the intended purposes. Why does this matter? Here, we need to identify the relevant criteria for evaluating a bailout program. Logically, four criteria stand out: (1) relative costs; (2) relative speed; (3) potential for bias (will some eligible applicants be systematically disfavored?); and (4) overbreadth (are the eligibility rules so loose that a limited subsidy is wasted on borrowers who should have been ineligible?).

How does the PPP experience measure up against these criteria? Let's take them one by one:

(1) Relative Costs. Congress understood that banks were not likely to make loans (particularly at the 1% interest rate that the PPP legislation mandated) unless there was something in it for them. Thus (and realistically), Congress provided in §1102(P) of the CARES Act for processing fees to be paid by the SBA to the lending banks according to a sliding scale:

• 5% for loans up to $350,000;
• 3 % for loans between $350,000 and $2 million;
• 1% for loans not less than $2 million.

The understandable intent here was to incentivize banks to make loans to smaller businesses by paying them more. But did this work? Not at all, as we will see in a moment.

(2) Relative Speed. The whole point of the PPP effort was to get money into the hands of poor employees fast. But the PPP used two distinct intermediaries: the lending banks and the SBA. This implied that the truly small business entity (for example, a sole proprietorship with 15 employees) had to find a bank to sponsor its application, and then they had together to approach the SBA. Some small proprietors may not have had an established banking relationship, or they may have split their business (using, for example, another bank's credit card services). Next, the SBA had to guarantee the loan, which required documents to be processed and possibly a loan closing (unnecessarily as this was not truly a loan, but more a welfare transfer payment). Because the SBA had never handled anything remotely similar in terms of the number of transaction, it encountered problems, and its computer system regularly crashed. Who could have done better? Both the Social Security Administration and the Internal Revenue Service have had far more experience with massive distributions (and the IRS has just handled the distribution of stimulus check with relative ease).

(3) Bias. Logically and predictably, banks favor their established clients. For these preferred clients, larger banks ran a virtual "concierge service," preparing the application and interfacing with the SBA. Small businesses with lesser connections to a bank waited at the end of the line. This was neither illegal nor even shady, but, to this extent, CARES was poorly designed. The goal of the PPP was to reach the smaller business, and for this purpose banks were precisely the wrong intermediary to use.

(4) Overbreadth. Subsidies should have fairly strict eligibility conditions. The CARES Act required principally that a business have less than 500 employees to qualify as a small business (with some much-lobbied exceptions for hotel and restaurant chains), and it had few exclusions. Thus, a profitable business could apply for loans at 1% interest to cover employee salaries and rent. Even if this loan was not later forgiven under Section 1006 of the CARES Act, a 1% loan is close to free money. Similarly, only the first $100,000 of an executive's salary is eligible for a PPP loan under the CARES Act, but this does not explain why an executive salary over $1 million should be eligible for even a $100,000 loan from the SBA at 1%. Given that funds are limited, that amount should have better gone to lower level employees. Nonetheless, according to a Wall Street Journal survey, "more than 235 public companies received more than $880 million in paycheck loans." See Yuka Hansen, "Delays, Glitches Mar Loan Relaunch," The Wall Street Journal, April 28, 2020 at p. 1, 2.

If this statute were rewritten, the first change should be a requirement that an eligible borrower be losing money and on the brink of insolvency if it did not receive the loan. Larger, solvent firms should stand at the rear of the line (even if they might layoff some employees without a PPP loan). After all, larger borrowers can qualify for the Federal Reserve's independent loan program for mid-sized companies.

Where did the PPP money actually go? According to the SBA, loans for more than $1 million accounted for only 4% of applications, but nearly 45% of the approved dollars under the PPP program. See Ruth Simon and Peter Rudegeair, "Small Businesses Bank Ties Help Determine Federal Aid," The Wall Street Journal, April 21, 2020 at p. 1, 4. That implies that larger entities received the lion's share of PPP funding. This is even clearer if we look at a more granular level at the behavior of participating banks. The SBA has reported that the largest lender under its PPP Program (apparently, JP Morgan Chase) distributed some $14 billion in loans and its average PPP loan was for $515,304. See Emily Flitter, "Businesses Suing Banks For Prioritizing Bigger Borrowers," The New York Times, April 21, 2020 at p. B-6. A truly small business could not possibly borrow that amount to cover its payroll.

In short, although participating banks may have received a larger fee (5%) for making small loans, that incentive appears not to have adequately motivated them, and they remained loyal to their existing (and larger) clients.

Nothing in all this was unlawful, shady or even surprising. Predictably, banks will favor their established customers. Banks have also emphasized that "know your customer" rules make it difficult for them to assist smaller borrowers with whom they have had little prior financial relationship. All that said, the clear policy implication here is that the CARES Act is not working to the benefit of the truly small business. Indeed, small businesses have been more than vocal in asserting this theme. See Pete Vegas, "PPP Loan Terms Amount To Legalized Fraud," Op-Ed, The Wall Street Journal, April 21, 2020 at p. A-19.

What then should a revised Act (let's call it "TRULY CARES") provide? First, the use of banks as the initial intermediary seems unnecessary and counterproductive. Banks might be needed if they were truly making a judgment about creditworthiness, but they are not when these loans are intended to be forgiven. Funds could be directly disbursed by the Treasury Department, with either the IRS or the Social Security Administration using its manpower to review the application. The real need here is to audit for fraud (which neither the bank nor the SBA is likely to do effectively).

Second, the "first-come, first-served" approach of the first two rounds of PPP funding should be ended in favor of a two-tier review: loans for under $350,000 (or some similar cutoff) should be reviewed and funded first. The "first-come, first-served" system has favored the larger borrower, assisted by special "concierge service" from the largest banks, and the fund is exhausted before the smaller borrower at the end of the line is reached.

Third, given limited funding, it should be a pre-condition to show that the loan applicant is in dire financial need. Currently, showing that you have not laid off employees is the principal condition for loan forgiveness. Instead, showing that you will have to lay off employees (unless the loan is made) should be a condition of loan eligibility.

Fourth, the SBA has not shown that it can play a useful role in this process. Although it may by now have learned much, it does not have the scale or manpower to audit what borrowers have done with their loans. That is the next and predictable scandal on the horizon. In contrast, the IRS has the manpower and the prosecutorial attitude.

John C. Coffee Jr. is the Adolf A. Berle Professor of Law at Columbia University Law School and Director of its Center on Corporate Governance.