The Chain from Bank Balance Sheets to Productive Capacity

At this point, the subject is the complete chain linking a lender-side credit shock to a firm’s later productive capacity. The useful judgment is not a new estimate; it is knowing which links are directly measured, which are supported by mechanism tests, and which depend on a calibrated model. Put differently, a credible credit-supply design tells us where the shock begins, but we still need to understand how scarce liquidity moves through a firm before deciding what the productivity result means.

One shock, several observable paths

Consider two firms operating in the same market when a financial shock hits their lenders. One firm borrows mainly from banks whose balance sheets are weakened by the event. The other borrows from less-exposed banks. If the firms had similar pre-shock conditions, their different lender exposure provides a way to compare the credit pressure they face. The exposed firm may receive less credit or pay more for it. That is the first observable change, not yet a productivity loss.

The firm then has several uses for a limited pool of cash. It must keep production running, pay workers and suppliers, service existing debt, and decide whether to buy equipment or continue spending on research and training. If it has several lenders, it may replace part of the lost funding. If it has inventory, it may sell existing goods at a lower price to generate cash. If those buffers are inadequate, it may reduce investment in capabilities whose returns arrive later. These options can coexist; they are not a fixed sequence in which every firm passes through every state.

The distinction between the immediate and delayed paths is the key connection among the evidence. In Belgian manufacturing, a one-standard-deviation difference in exposure to banks holding distressed sovereign securities reduces total bank credit by about 18 percent cumulatively over three years and eventually raises the financing-cost proxy by roughly 3.5 percent. In the short run, exposed firms lower output prices by about 2 percent while the quantity-based physical-productivity estimate shows no significant effect. Around three years after the shock, physical productivity begins to fall and reaches a cumulative decline of 6.4 percent by the end of the 2009–16 sample.

Those numbers are not three independent stories. They are different positions along one chain. Credit quantity and cost describe the financing shock. Prices describe an immediate liquidity response and, later, the pass-through of higher costs. TFPQ describes the delayed change in the physical production relationship. The Italian evidence adds breadth to the same logic: in a large panel without firm-level output prices, a 1 percent increase in the credit-supply shifter raises value-added productivity growth by about 0.1 percent and revenue productivity growth by about 0.02–0.03 percent. The measures differ, but both settings indicate that lender-side finance can reach beyond the amount of input purchased.