The State answered the fear and rates fell. The entire thesis, proven by decree.

The debtor did not change, the world did not improve, and the bank did not become virtuous; what changed was who answers for the loss — that is what structure does: it does not lecture against risk, it decides where it lives.

The government created Pronampe to keep small businesses alive during the pandemic. A public guarantee covers a decisive share of the risk, the rate is capped, and the tenor is extended. The result is revealing: it took only the State answering the fear for rates to fall. The entire thesis, proven by decree.

Before the guarantee, banks were looking at companies with interrupted revenue, stale data, and a future impossible to model. The rational price of credit rose, but in many cases the problem was not price; it was absence of supply. No rate compensates a creditor for risk he can neither distinguish nor bear. The public guarantee changes the expected loss and releases capital. The debtor remains the same. The structure changes around him.

This exposes the falseness of debates that treat the rate as a moral attribute of the bank. Banks charge margin and use market power, but they also respond to loss. When the State assumes a share of the worst case, the contract can become cheaper. Not because the banker became charitable, but because the fear found another balance sheet. The taxpayer started selling protection.

A guarantee is not free money. It transfers risk, creates incentives, and needs a price, a limit, and governance. If it covers everything without discipline, the bank can originate volume and deliver the losses to the guarantee fund. If it covers too little, it does not change the decision. If it demands excessive bureaucracy, it arrives after the bankruptcy. The right design must divide responsibility: enough to unlock supply, not enough to eliminate care.

Pronampe shows that small companies do not pay dearly merely because they are worse. They pay dearly because they are opaque, heterogeneous, and expensive to analyze. A portfolio of thousands of businesses demands data, standardization, and distributed collection. The bank compensates with generic models and personal guarantees. In an unprecedented crisis, those models lose their reference and the owner's guarantee loses value along with the company. The public guarantee temporarily replaces the information that does not exist.

The long-term question is why that information does not exist. The government knows invoices, payroll, taxes, and declared revenue. Banks know account movement and payment behavior. Acquirers know sales. Platforms know orders. The company, however, must re-present fragments to each institution, and the creditor cannot form a verifiable real-time view. Risk is expensive also because the data is locked in silos that charge rent on ignorance.

A modern infrastructure could reduce dependence on public guarantees in future crises. Consented data, receivables registries, payment histories, and operational information would allow distinguishing the company that lost revenue temporarily from the company that was already unviable. The State could guarantee the systemic part of the shock, while the market priced the idiosyncratic part. Today, both arrive mixed and the program must treat different cases with similar rules.

The entrepreneur will see the rate and conclude he finally received fair credit. Perhaps. But he must ask whether the tenor matches the recovery period, how the amortization begins, and what obligations he assumes. Cheap credit that starts maturing before revenue returns merely displaces the crisis. The contract must finance the whole bridge, not half of it. A bridge interrupted in the middle of the river is more dangerous than none.

The program also creates a natural experiment. Similar portfolios, with and without the guarantee, may reveal how much of the rate stems from loss, how much from cost, and how much was margin. If the data is published, we will learn. If it is hidden inside political aggregates, only propaganda will remain. Credit policy should produce public knowledge, not just disbursement.

My position would be favorable to emergency use, but skeptical of indiscriminate permanence. In a pandemic, the shock was not chosen by the entrepreneur and private liquidity retreated collectively. The State is the only balance sheet able to absorb the systemic component. After the emergency, the guarantee should be recalibrated to prevent businesses from depending on it and banks from abandoning their own analytical capacity.

As an investor in banks, I would watch not only the volume originated, but the quality of the risk sharing. Guaranteed programs can generate revenue with reduced capital consumption, yet they expose institutions to operational, reputational, and documentation risk. A poorly formalized guarantee may not be honored. In credit, the administrative detail ignored at origination tends to reappear as a legal loss.

In fintechs, I would look for companies able to use the experience to build better data and products, not merely distribute the public line. The crisis offers subsidized client acquisition. The durable value lies in tracking flow, offering cash management, collecting, adjusting limits, and financing specific assets after the program ends. Whoever treats Pronampe as a sales campaign will get growth; whoever treats it as a laboratory may get an advantage.

Moral hazard also exists on the debtor's side. A low rate and a public guarantee can encourage unviable companies to postpone closure. Preserving jobs during a shock is a legitimate goal, but credit does not replace capital when revenue disappears indefinitely. In some cases, direct subsidy would be more honest. Lending to those who cannot pay turns social policy into future default and blames the entrepreneur for accepting the only help offered.

It is necessary to distinguish liquidity from solvency. Liquidity is a temporary lack of cash to cross a flow. Solvency is the economic capacity to generate enough value to pay obligations. A public guarantee solves the creditor's willingness, not the business's economics. If the company does not recover demand, the guarantee fund merely chooses who will book the loss. Structure distributes the damage; it does not eliminate it.

Even so, the program teaches something that should outlive the pandemic: guarantees can be decomposed and shared. The bank does not need to demand property, personal guarantee, and receivables for every operation. Partial coverage, a granular portfolio, a subordinated reserve, or a buyer anchor can reduce risk more efficiently. The Brazilian problem is using collateral as an accumulation of power, not as a design variable.

The small entrepreneur usually offers personal wealth because he has no organized business assets. That mixes family and operation, amplifies the trauma of bankruptcy, and does not necessarily improve recovery. A well-built portfolio guarantee can protect the creditor without destroying the borrower. Pronampe socializes part of the risk out of necessity; the market should learn to mutualize it out of technique.

The value of granularity will also become evident. A single small company is hard to analyze and can fail entirely. A portfolio of thousands, spread across sectors and regions, makes losses statistically tractable, provided origination is consistent. The FGO works as collective protection; private funds can build similar layers. The secret is not predicting each debtor perfectly, but limiting concentration and aligning whoever originates.

The state guarantee answers fear because it has the power to tax and a longer horizon. That does not mean it is free. Future losses will be debt, taxes, or less spending. The program should be judged by the jobs and capacity preserved relative to the fiscal cost, not by the number of contracts announced. Disbursement is the input. The result is how many viable companies crossed and repaid.

For private credit, 2020 will be a dividing line. Investors will see that risk is not merely an attribute of the company; it is a product of the regime. Excellent businesses can lose revenue by sanitary decision. Models based on history fail when the world changes state. Good structure must include pause, extension, and sharing mechanisms that keep an exogenous shock from becoming a destructive liquidation.

It took only the State answering the fear for rates to fall because the rate was compensation for concentrated fear. When the fear is shared, the price changes. The challenge is not concluding that every high rate demands a public guarantee. Some demand better information, others private collateral, competition, collection, or simply refusal. The State proved the design thesis. Now the market must learn to apply it without summoning the taxpayer for every contract.

Pronampe will be celebrated as cheap credit. I would read it as a demonstration: the debtor did not change, the world did not improve, and the bank did not become virtuous. What changed was who answers for the loss. That is what structure does. It does not lecture against risk. It decides where it lives.

Leo Bentier