Why a Bank Loan Can Close the Next Door
To understand how FinTech lending can complement bank credit, this chapter focuses on the contracts behind the two loans: who gets paid first, which assets are tied up, and what remains available for the next lender. The reader’s task is to separate seniority from security and to see why a collateral-free loan may improve future borrowing capacity without becoming a cheap loan in the present.
Two claims on the same borrower
A small firm does not borrow into an empty space. It already has assets, cash flows, existing lenders, and obligations that shape what a new lender can safely do. The crucial distinction is not simply between a bank and a FinTech platform. It is between the claims those lenders hold on the firm and on the firm’s assets.
Seniority describes the repayment order. A senior creditor has priority over a junior creditor if the borrower fails. Security describes whether a claim is attached to specified collateral. A loan can be unsecured but still senior; it can also be junior and secured in some other arrangement. The terms overlap in ordinary conversation, but they answer different questions. Who is paid first? Which asset backs the claim? What can a later creditor reach?
Traditional bank credit is commonly senior and often secured, particularly for young or small firms. Suppose a bank lends against a machine. The machine is not merely an object on the balance sheet; it is part of the bank’s protection. If the firm later seeks another loan, it cannot normally offer that same machine as though no claim existed. The first loan has converted part of the firm’s productive capacity into an encumbered asset. That can be sensible from the bank’s perspective: collateral helps protect the bank from losses and from dilution by a later creditor. But it can reduce the firm’s room to finance its next opportunity.
This is the first half of the collateral problem. A firm may have assets in an accounting sense while lacking assets that are freely pledgeable. The difference matters when a bank decides how much more to lend. A business that owns a machine already pledged to one senior lender may look stronger than an asset-poor business, yet have less collateral available for a new bank loan. A contract can therefore protect a lender today while narrowing the borrower’s choices tomorrow.
Collateral is not valuable only because it can be sold after default. It also changes the bargaining position before default. A bank that can point to a specified asset has more protection against a borrower taking on another claim that would compete with it. That protection makes the original loan safer from the bank’s perspective, but it can make the firm’s future financing more rigid. The business may still own the machine, use it to produce revenue, and report it on its balance sheet; what it may not have is the freedom to promise the machine to another lender. Pledgeability is therefore a form of financial capacity that can be consumed by a contract.
The reverse possibility is equally important. If a firm acquires an asset with a junior unsecured loan, the first lender has accepted weaker protection so that the asset remains outside a specified security claim. The firm does not become unlevered, and the asset is not risk-free. But a later bank can potentially take a senior claim over it. In that sequence, the first loan changes the set of contracts the firm can write next. The key mechanism is not that FinTech credit creates an asset from nothing; it is that the priority of the first claim can determine whether a new asset remains usable as collateral.
A junior unsecured loan changes the sequence. The lender does not take a specified asset as collateral, and the junior claim stands behind senior creditors in the repayment hierarchy. In the French setting studied by Paul Beaumont, Huan Tang, and Eric Vansteenberghe, FinTech platforms supplied exactly this combination to small and medium-sized firms. A firm could use the money to acquire a new asset without giving the FinTech lender a senior secured claim over that asset. If a bank later accepted the asset as collateral, the asset could support a second claim.
The contrast is easiest to see as a pair of financing paths. In the first, a bank finances a machine and attaches a senior claim to it. The firm gets money, but the machine’s future pledgeability is constrained. In the second, a FinTech platform provides junior unsecured credit, the firm buys the machine, and a bank later lends against it. The firm has not escaped debt. It has changed which lender receives the first claim on the new asset.
The diagram does not say that every FinTech loan produces a bank loan. It shows the contractual possibility that the study calls the collateral channel. The asset is the scarce resource shared by the two financing relationships. A loan that leaves the asset free can create an option for a later bank creditor; a loan that encumbers it may remove that option. The value of the first loan depends partly on what it leaves possible for the second.
This also explains why the phrase “unsecured” should not be treated as a synonym for “unconstrained.” A FinTech lender may not take a particular machine, but the firm still owes principal, interest, and fees. A bank loan that is described as unsecured may still be senior, may rely on personal guarantees, or may be protected by the firm’s unpledged assets. The relevant question is not what label appears in an advertisement. It is whether the firm can still offer a usable asset to a later creditor and whether that creditor has a claim strong enough to lend.
Why the French product was possible—and not cheap
The study’s French setting made this product contrast unusually visible. For decades, a banking-monopoly regime prevented nonbanks from carrying out lending activities. A 2014 legal change allowed crowdfunding intermediaries—what the study calls FinTech platforms—to enter the corporate-loan market. These platforms were not banks, so they did not operate under the same capital and liquidity requirements. At the same time, rules limited the size of corporate loans and the amount an individual investor could place in one loan, which kept the platforms focused on the small-business market.
The regulatory difference helps explain product supply, but it does not by itself explain borrower welfare. Banks have reasons to prefer secured lending. Under the capital framework described by the authors, unsecured loans to SMEs are more costly for banks to support with regulatory capital than secured loans. A non-deposit-taking platform outside the bank capital regime can offer the unsecured product more readily. That is a supply-side explanation: a product appears where the institutional constraints make it feasible. It is not a claim that a platform has eliminated the underlying credit risk.
This distinction keeps regulation in its proper place. A rule can make one type of lender more willing or able to offer a product, but it does not make the borrower’s project more profitable, the asset more valuable, or the repayment schedule easier. It changes who can intermediate the risk and how the claim is structured. In the French case, the combination of bank constraints and platform entry created a product that banks typically did not supply in the same form. The resulting choice was not between a risky loan and a riskless loan. It was between different allocations of risk, time, and control over the firm’s assets.
For a borrower, that allocation can be useful only if the timing lines up. A bank’s lower rate may be valuable but arrive after the opportunity has passed. A platform’s rapid decision may preserve the opportunity but create payments that arrive before the new asset produces enough cash. The first lender’s claim also affects the second lender’s incentives: a bank may be more comfortable lending against a clean, pledgeable asset than against an asset already tied to another creditor. The product comparison therefore has at least three clocks—the decision clock, the asset’s cash-flow clock, and the repayment clock. A sensible interpretation has to keep them together.
The product also used a different operating model. Applications were made online. Platforms collected accounting and credit-history information and typically made decisions within 48 hours; the paper describes approval as taking less than a week and sometimes less than a day. Banks often took more than a month to process a comparable request. For a firm facing a narrow window to buy equipment, pay a supplier, or preserve operations, the value of time can be real even if it is difficult to measure separately from the price of credit.
That speed came with a price structure more complicated than a single interest-rate number. The French platforms could charge a fixed application fee, an origination fee proportional to the loan—typically 3% to 5% across platforms—and interest paid to investors. Late or early repayment could bring additional charges. In the study’s sample, the average FinTech interest rate including fees was 7.8%, with substantial variation and a maximum of 16.8%. The average loan was about €150,000, the median was €50,000, and average maturity was 38 months.
The comparison with bank credit makes the premium clearer. After accounting for loan size, maturity, and observable borrower characteristics, FinTech loans carried an average rate 5.5 percentage points above comparable fixed-term bank loans. The baseline bank rate in that comparison was 1.8%, and both rates included fees. These are historical French observations, not a current universal price quote. They establish that the collateral-free, junior product was materially more expensive in the studied market.
Why would a borrower accept that premium? One answer is risk allocation. A junior unsecured lender is less protected if the firm fails, so investors may require a higher return. Another is convenience: a faster decision can have value when waiting would mean losing an opportunity. Neither explanation should be overstated. The study observes the product’s prices and discusses speed as a possible reason for the premium; it does not isolate a precise “value of speed” from the risk premium or other differences.
The more important point is that the first loan’s price and its future effect are separate dimensions. Consider a firm that borrows €50,000 at a high all-in cost to buy a productive asset. If the asset generates cash flow and can later support a bank loan at a lower rate, the firm may rationally accept an expensive first step. But the logic fails if the money covers recurring expenses with no asset left to pledge, if the bank will not recognize the asset, or if the repayment schedule consumes the cash flow the asset was supposed to create. The product is not made attractive by being expensive. It can be attractive despite being expensive when it changes the firm’s financing possibilities in a valuable way.
This is why the study’s question is more demanding than “Do FinTech borrowers pay more?” They do, in the French comparison. The question is whether the extra cost buys a different position in the financing sequence. A senior secured bank loan may be cheaper because the bank is protected, but that protection can make the asset less available for later borrowing. A junior unsecured FinTech loan may be dearer because the lender is exposed, yet leave the asset available for a later bank claim. The two prices cannot be interpreted without considering the two claims.
There are also clear stopping points. The collateral channel requires a new asset or another form of pledgeable capacity. Paying wages, utilities, or an old bill may be necessary for survival but does not automatically create collateral a bank can use. The later bank must be willing to lend, and the legal priority must allow the asset to support that loan. Finally, the firm must remain solvent long enough for the second step to occur. The absence of collateral can provide flexibility, but it can also leave the first lender with a weak recovery position and the borrower with a demanding payment obligation.
Seen this way, the French product was neither a technological shortcut around all lending constraints nor a simple replacement for bank credit. It was a new layer in the stack, made feasible by a different regulatory position and valuable under a particular asset-and-credit sequence. The next question is empirical: when firms actually received these loans, did their bank credit rise because the FinTech loan changed their borrowing capacity, or because they already wanted to borrow more for reasons the data cannot see?