What Credit Tightening Really Costs
The subject of this series has been the full cost of a lender-driven credit tightening: not only the credit that disappears today, but the productive capacity that may fail to be built tomorrow. The final question is therefore broader than whether firms cut output after a financial shock. It is whether the financial system changes the path by which firms finance, protect, and renew their ability to produce. The evidence supports a clear but conditional answer: credit tightening can scar productivity when current liquidity pressure forces firms to sacrifice investments with delayed returns, even when prices, inventory sales, or replacement lenders cushion the immediate decline.
That answer rests on a chain rather than a single statistic. Bank balance sheets and funding conditions create different exposures for firms connected to different lenders. Credit then becomes less available, more expensive, or both. Firms allocate scarce cash among production, debt service, employment, inventory, and capability-building investment. Some of the first responses are visible in prices and current sales; the loss from reduced R&D, training, technology adoption, or organizational improvement arrives later. What looks like a temporary financing disturbance can thus alter the firm’s future production function.
The cost arrives in stages
The strongest firm-level evidence comes from the New York Fed study of Belgian manufacturing firms around the 2010 European sovereign debt shock. Firms whose pre-shock lenders held more distressed sovereign securities experienced a larger credit contraction. The design supports a lender-side interpretation through pre-trend checks, bank controls, within-firm comparisons across lenders, and detailed fixed effects. It does not make every observed fall in borrowing a supply shock, but it makes the specific exposure comparison credible enough to follow into the firm’s decisions.
The first stage is financial, not technological. A one-standard-deviation difference in exposure to the sovereign shock reduces total bank credit by about 18 percent cumulatively over the following three years and eventually raises the study’s financing-cost proxy by about 3.5 percent. The quantity and cost of finance move in the directions expected from a supply tightening. That pressure increases the value of current liquidity. The firm must decide whether to find a new lender, sell inventory, reduce ordinary investment or employment, cut innovation, or combine several responses.
The second stage is nominal and operational. Exposed firms lower output prices by about 2 percent in the immediate aftermath of the shock. Firms with more inventory and greater financial pressure cut prices more aggressively, consistent with using discounted sales to convert existing stock into cash. This response can protect current consumption and operations. It also changes what a revenue-based productivity measure sees: the firm can look less productive because it is charging less even when its physical efficiency has not yet changed.
The third stage is capability-related. Exposed firms reduce R&D and training spending, and the effect on physical productivity becomes significant only around three years after the shock. By the end of the 2009–16 sample, the one-standard-deviation exposure difference corresponds to a 6.4 percent contraction in cumulative physical-productivity growth. The timing is the central clue. A financial shock can be absorbed first by margins, inventories, and budgets, then appear later as a lower stock of knowledge, weaker adoption capacity, and slower technical improvement.
The fourth stage is cost pass-through. Firms more exposed to the shock eventually raise prices relative to less-exposed firms, by up to about 4 percent in the years following the shock before the difference moves back toward zero. Higher financing costs and lower physical productivity raise the cost of operating. The initial price cut and later price increase are therefore not contradictory. They belong to different phases of the same liquidity and capability adjustment.
The Italian evidence broadens this chain. Manaresi and Pierri use a much larger panel of Italian corporations from 1997–2013 and identify lender-driven variation through bank-firm relationships and the 2007–08 interbank-market freeze.