How Cheap Credit Becomes a Boom
The subject here is the domestic mechanism linking credit-supply liberalization to a lending boom. The useful question is not simply whether private credit rises, but how to tell whether it rose because banks could supply more loans, where that credit went, and when the expansion began to change the system’s exposure to loss. Those signals turn an abstract policy result into a sequence that can be observed in lending, sectors, asset prices, and bank funding.
Quantity first, risk later
A direct credit-supply reform changes the conditions under which banks originate loans. If the reform removes a binding limit, banks can expand lending even without a sudden increase in borrowers’ appetite for debt. That does not mean demand is irrelevant. It means the first movement in the market may come from the amount of credit made available rather than from borrowers bidding up the price of a fixed loan supply.
The Federal Reserve study finds a pattern consistent with that interpretation. After supply-side deregulation, real private credit increases, while real short- and long-term interest rates do not rise. They remain broadly flat for several years and then decline, with the short-term rate falling more noticeably. If a surge in borrowing demand were meeting an unchanged supply of loans, one might expect stronger upward pressure on borrowing costs. Instead, the main adjustment appears to occur through quantity.
The credit-spread response adds a time dimension. Spreads—the difference between the yields on lower-grade debt and a safer benchmark—remain largely unchanged in the first few years. Later, they rise above their pre-reform level. This is consistent with a system in two phases. At first, more credit is available while stronger activity and improving borrower balance sheets keep expected losses contained. Over time, the accumulated leverage makes the system riskier, and lenders or investors begin to demand compensation for that risk.
“Consistent with” is important here. Interest rates and spreads do not provide a direct view of every bank’s counterfactual loan decision. They are diagnostic evidence, not a perfect experiment. But they line up with the policy classification and with the credit response: the reforms designated as directly related to supply produce a much stronger lending expansion than the other banking reforms. Taken together, the patterns make it harder to explain the result as a simple demand surge.
A simple comparison helps clarify the logic. Suppose borrowers suddenly become more optimistic while banks’ lending capacity stays fixed. More borrowers compete for the same pool of funds, which would ordinarily put upward pressure on rates, spreads, or both. Now suppose a binding lending limit is removed while borrower demand changes little. Banks can make more loans, so the market can adjust mainly through quantity; rates need not rise and may fall as competition increases. Neither scenario is pure in the real world. Monetary policy, inflation, exchange rates, and expectations also move prices. The point is not that a flat rate proves a supply shift. It is that flat or falling real rates, stable early spreads, a strong quantity response, and a distinctive sectoral allocation form a more coherent supply-side pattern when they occur together.
The study’s use of real private credit is part of that discipline. A nominal increase could reflect nothing more than a higher general price level. A real measure is intended to track a larger volume of purchasing-power-adjusted claims on the private sector. Even that measure does not tell us whether each claim finances productive capacity, a property purchase, or another asset. It answers a narrower question: did the stock of private credit expand in real terms after the policy change?
The same care applies to the timing of spreads. A spread is a relative price: it compares the yield on riskier borrowing with a safer benchmark. If the benchmark falls because of broader monetary conditions, the spread can remain stable even while the absolute cost of borrowing changes. If the spread widens only after several years, it suggests that the market’s assessment of aggregate credit risk changed later, but it does not identify every reason for that repricing. Reading rates, spreads, quantities, and sectoral allocation together is therefore more informative than asking any one series to carry the whole causal argument.
Where the new credit goes matters as much as how much appears. The study finds that supply-side deregulation sends credit disproportionately toward non-tradable sectors, especially construction and real estate. Real-estate credit rises by more than 10 percent in the five years after liberalization. Mortgage and household credit also increase, though the reported responses are smaller than for total private credit and real-estate credit.
Non-tradable firms often have characteristics that make the allocation consequential. They tend to be smaller, more likely to face financing constraints, and more dependent on loans secured by real estate. A manufacturer that sells into foreign markets may have revenue that provides a wider economic cushion and may be evaluated against international demand. A local builder or property firm is more directly exposed to domestic land values, rents, and credit conditions. The contrast is not a moral judgment about construction or property. It is a statement about sensitivity: when collateral values fall, credit-dependent non-tradable activity can contract sharply.
The sectoral pattern is therefore another piece of supply evidence. A general improvement in economic prospects could raise borrowing across many sectors. A credit expansion concentrated in collateral-sensitive, non-tradable activity is more characteristic of a shift in the financing constraint facing those borrowers. It also creates a channel through which the initial boom can feed on itself.
The loop from credit to collateral is the critical transition. A borrower’s ability to obtain a loan depends partly on the value of the asset offered as security. When more lending raises demand for property, property prices can rise. The higher valuation then makes existing owners appear better collateralized and can support additional borrowing. That second round of credit can push prices higher again. In a rising market, the feedback can look like prudent balance-sheet improvement: loan-to-value ratios may appear manageable because the denominator is increasing.
The feedback reverses when the asset price stops rising. A fall in property values reduces the collateral available to support new loans and weakens the balance sheets of borrowers that borrowed against the earlier valuation. The same leverage that amplified the expansion now magnifies the contraction. Defaults can reduce banks’ willingness to lend, which tightens credit precisely when borrowers are trying to repair their finances. The resulting decline in demand can deepen the asset-price and credit reversal.
The study reports a hump-shaped response in house prices after supply-side deregulation. That shape is not proof that every reform creates a property bubble. It does show why housing is a useful place to look for the interaction between credit supply and financial fragility. A house is both an asset whose price can rise with easier finance and collateral against which more finance can be obtained. The two roles link the asset market to the banking system.
How funding turns expansion into fragility
The asset side is only half of the mechanism. A bank can expand loans by attracting more stable deposits, by using other liabilities, or by relying on funding that may be more sensitive to market conditions. The Federal Reserve study finds that banks’ loans-to-deposits ratios and non-core funding ratios rise after banking deregulation. A rising loans-to-deposits ratio indicates that lending is growing relative to the deposit base; a rising non-core ratio signals greater reliance on liabilities outside the bank’s most traditional funding source. These are not crisis diagnoses on their own, but they help reveal how the expansion is being financed.
Funding structure matters because a loan is usually longer-lived and less liquid than the liability used to finance it. A bank that funds a five-year loan with a liability that can leave quickly must repeatedly renew or replace that funding. In calm markets, this can look efficient. It allows credit to grow faster than ordinary deposits and connects banks to a wider pool of investors. In stressed markets, the same arrangement can turn a concern about asset quality into a liquidity problem. If funding cannot be rolled over, the bank may sell assets, stop making new loans, or do both at once.
Competition can intensify this choice. Liberalization may bring new entrants or reduce protected margins for existing banks. Lower franchise value can weaken the cushion that makes a bank willing to absorb losses, while pressure to win market share can encourage looser lending terms. The bank does not need to believe that every borrower is safe. It may believe that competitors are expanding, that property prices will continue to rise, or that a profitable opportunity will disappear if it waits. Those incentives help explain why a credit boom can persist even when individual decisions appear defensible at the time.
Early conditions can make the risk difficult to see. When credit is newly available, borrowers who were previously constrained may use it for productive investment, and stronger economic activity can improve repayment. Low defaults then validate lenders’ expectations. Rising house prices improve collateral. Spreads remain calm because the market is pricing current performance more readily than a distant, system-wide reversal. The absence of immediate stress is real, but it is not the same as evidence that the balance-sheet expansion is harmless.
Eventually, risk can be repriced in aggregate. The study’s later spread widening, higher loans-to-deposits and non-core funding ratios, and hump-shaped house-price response fit a transition in which the quantity of credit has outpaced the system’s ability to absorb a shock. Once expectations turn, borrowers may default, banks may pull back, and high debt may constrain consumption and investment. A negative shock is the trigger; the prior credit expansion is the vulnerability that gives the trigger larger consequences.
That distinction helps avoid two mistakes. The first is to call the initial lending increase irrational simply because it later becomes dangerous. Some of the new credit can finance opportunities that were blocked by the old rules. The study’s early GDP response is consistent with real economic gains, and aggregate evidence does not establish that every loan to a non-tradable firm is unproductive. The second mistake is to treat a successful early repayment record as proof that the expansion has no systemic cost. A system can look healthy while its borrowers and banks are becoming more exposed to a common collateral or funding shock.
A practical reading of a proposed credit liberalization should therefore combine several observations. First ask whether lending quantities rise after the rule changes and whether prices move in a way consistent with increased supply. Then examine the destination of the credit: is it reaching a broad set of productive uses, or is it concentrating in collateral-sensitive sectors? Next inspect the funding side. Are loans growing faster than deposits? Is non-core funding becoming more important? Finally, follow asset prices and spreads over a longer horizon. A quiet first phase and a later repricing are not contradictory signals; together they may be the expected shape of a delayed risk cycle.
No single indicator can answer the question. A falling interest rate may reflect monetary policy as well as credit supply. A rising house price may reflect population growth or construction constraints as well as lending. A higher loans-to-deposits ratio may be sustainable in one banking system and dangerous in another. The strength of the account comes from the alignment of several measures with the timing and type of reform, not from any one ratio treated as a warning light.
The domestic mechanism can now be stated without reducing it to a slogan. A direct liberalization makes more lending feasible; the additional credit is allocated unevenly; collateral and optimism can support another round of borrowing; banks may finance the expansion with more fragile liabilities; and the resulting exposure is repriced only after the boom has had time to mature. The initial benefit and the later danger are not separate stories. They are different phases of the same credit-supply process.
That conclusion also marks the boundary of the domestic view. It tells us how a bank’s lending and funding decisions can create a boom, but not yet how the result changes when the lender, borrower, and currency are spread across countries. Once credit can be booked offshore, supplied by a foreign affiliate, or passed through several institutions, the same sequence may become harder to locate on any single balance sheet. The first-half model is therefore complete enough to test, but not complete enough to close the story.