Not Every Rule Creates the Same Escape Route

The subject is still a banking group deciding where to place corporate credit, but the relevant comparison now shifts from one entity to the rule itself. This chapter asks whether all forms of “tighter regulation” leave the same room for an affiliated NBFI to step in, and whether every nonbank has the business model to use that room. The answer helps readers predict when a lending shift is structurally plausible rather than treating the IMF’s result as a universal law.

The rule decides where the pressure lands

Regulation does not act on a financial group as an undifferentiated force. It acts through a particular instrument, at a particular level, on a particular balance sheet or lending activity. That level determines whether the parent can reduce pressure on one subsidiary by moving a loan to another. A useful first distinction is between measures aimed at a lending margin and measures aimed at the group’s capital position.

Loan-supply-oriented measures are designed to influence the flow of credit. They can limit credit growth, restrict specified forms of lending, alter loan-loss provisioning or otherwise constrain a bank’s capacity to originate. A reserve requirement can affect the resources associated with bank intermediation. Stress-testing, profit-distribution and structural measures can also alter the prudential conditions under which a bank operates. In the IMF study’s baseline, these different measures are combined into a macroprudential shock, but the mechanism is clearest when asking where each measure binds.

If a rule acts mainly on a bank subsidiary’s local lending activity or on a particular product, the parent may face an uneven constraint. The bank has less room in the affected channel, while an affiliated NBFI may remain able to participate in the corporate-loan market. That does not mean the NBFI is free of all requirements. It means the direct effect of the measure is not identical across the two entities. The group can then compare reducing the loan with using the less directly constrained affiliate.

Capital-based measures pose a different problem for substitution. Capital requirements, countercyclical and conservation buffers, and leverage limits are more likely to be calculated against the parent’s consolidated accounts. When the assets of several subsidiaries count toward the group’s relevant capital position, moving a loan from a bank subsidiary to an NBFI may not remove the underlying group-level pressure. The parent can change the legal originator, but not necessarily the capital exposure that the rule is meant to constrain.

The IMF’s results follow this distinction. For its baseline combination of measures, affiliated NBFIs expand lending relative to bank subsidiaries after tightening. When the analysis isolates capital-based measures, NBFI subsidiaries instead reduce lending by 1.4 percent relative to bank subsidiaries. The high-frequency capital-based exercise points in the same direction. The contrast does not prove that capital rules eliminate every form of regulatory leakage, but it is consistent with a consolidated constraint limiting the parent’s ability or incentive to shift credit across entities.