When Shared Exposure Makes Stress Correlated

The portfolio question in SR 26-4 is not simply how many borrowers share a description. It is whether several loans depend on the same employer, geographic market, industry, or workforce condition, and could therefore respond to the same outside change. This chapter separates those exposure dimensions from the possible shocks that affect them, then follows the path from a shared dependency to simultaneous repayment pressure. The result is a way to recognize correlation without inventing a concentration threshold or assuming that every borrower will experience the same outcome.

Concentration is a shared dependency

A portfolio becomes vulnerable to concentration when a significant amount of lending depends on a condition that is common to more than one loan. SR 26-4 names three dimensions of that dependence: geographic markets, employers, and industries. These dimensions describe where the loans are connected. They do not, by themselves, describe the event that will create stress.

A geographic concentration means that several borrowers' circumstances may be tied to the same local economy or labor market. An employer concentration means that multiple borrowers may depend on the same organization as a source of employment. An industry concentration means that borrowers may be exposed to conditions affecting the same type of work or business activity. The categories can overlap. A group of borrowers may work for one employer, operate in one industry, and live in the same market. In that case, the portfolio has more than one route by which a common change could reach several loans.

The useful unit is therefore not just the borrower and not just the loan. It is the connection between the loan and the condition that supports repayment. At the individual level, the lender asks whether a borrower's income is current, sustainable, and likely to continue. At the portfolio level, it asks how many repayment assumptions are connected to the same employer, place, or industry. The first question measures the condition of one obligation. The second reveals whether a change in that condition could affect other obligations at the same time.

This is why concentration is more than a large number of similar files. Similarity can be superficial. Two borrowers may have comparable loan products but different employers and different sources of income. Conversely, loans that look separate in individual files may share a dependency that is not obvious until the portfolio is viewed as a whole. SR 26-4 does not provide a required data structure or a universal way to map those connections. It does identify the dimensions that deserve attention.

The word “significant” in the letter also requires restraint. The source does not define significance as a percentage of capital, a borrower count, an employer share, or any other numerical threshold. It would therefore be inaccurate to attach a number to the concept and present it as a Federal Reserve limit. The safer interpretation is that the degree of exposure matters, but the source leaves the institution to assess materiality within its existing risk-management framework.