What SBA Default History Shows About a Real Downturn
No MCA-specific recession default rate exists in any source found for this piece, and none is invented here. The nearest disclosed benchmark comes from a different product entirely: per Crestmont Capital’s analysis of SBA program data, SBA 7(a) loans default at 2.5% to 6% across a normal economic cycle, but that figure exceeded 12% in some cohorts at the peak of the 2008 to 2010 financial crisis, and loans originated in 2007 and 2008 specifically showed 15% to 20% default rates by 2012 in some program segments. SBA 504 loans, which are backed by real assets rather than a general credit judgment, ran lower in ordinary conditions at 1.5% to 3.5%, but still climbed to 7% to 9% default during 2009 to 2011.
SBA 7(a) and 504 are traditional term loans, not merchant cash advances, and this figure should not be read as an MCA statistic. It is used here as the closest disclosed default-rate benchmark for what a lending portfolio does in a downturn, because nothing MCA-specific was found in this research pass, not because the two products behave identically.
The Lag Effect: Why Deferral Programs Delay the Reckoning
The same Crestmont Capital analysis of SBA data found something more useful than a single default number: a documented lag between a shock and its default impact. Post-COVID SBA default rates remained elevated at 4% to 8% for pandemic-era loans well into 2021 through 2023, not because the pandemic caused new stress that late, but because deferral programs run in 2020 and 2021 masked underlying stress that only surfaced once those programs ended.
An ISO’s own book can produce a similar lag. A merchant sliding toward trouble does not usually stop paying the day revenue drops. Daily or weekly holdback debits keep clearing for a while even as the merchant’s cushion erodes underneath, which means a portfolio can look stable for months into a downturn before the actual damage shows up in missed payments and default requests.
Why an ISO’s Book Concentrates Risk Differently Than a Bank’s
This is reasoning, not a separately cited statistic. A bank term loan is repaid on a fixed monthly schedule set independent of how the business happens to be performing that month. An MCA advance is repaid through a daily or weekly draw tied directly to the same revenue stream a downturn hits first, which is a structurally different risk shape than the SBA default data above describes for a traditional loan.
That difference cuts both ways. A merchant sliding toward trouble can keep clearing a shrinking daily draw for a while even as the cushion underneath erodes, the same kind of lag Crestmont Capital’s SBA analysis documents for a bank product, except the debit here is pulling directly from the account where the stress is showing up first, not a separate monthly bill a merchant can quietly fall behind on without anyone noticing right away.
What a Growth Run Hides
This is reasoning, not a separately cited statistic. During a growth run, strong new-deal flow does most of the work a portfolio review should be doing. When submissions are up and renewals are coming in, there is little pressure to look hard at how existing positions are performing, because the new numbers are covering for whatever is happening underneath.
That cover disappears the moment new-deal flow slows, which is usually the same moment a downturn starts. What is left at that point is whatever the back book looks like, not the growth story that had been obscuring it, and the SBA lag-effect evidence above suggests that reckoning can take a year or more to fully surface once it starts.
What This Argues for Tracking Before a Downturn, Not After
The practical argument follows directly from the lag-effect pattern in the Crestmont Capital SBA analysis cited above: the visible signal, a merchant missing a payment or requesting a workout, arrives well after the underlying stress starts. Waiting for that signal means finding out how exposed a book already is at the worst possible time to learn it.
Tracking early signs of strain in the existing book now, while a growth run is still underway, is the version of this an ISO can act on. That is not a claim that doing so prevents a downturn from hurting, it is a claim that it replaces a surprise with a picture the broker already had going in.
This is not only a lesson from 2008. Business bankruptcy filings are already climbing in the current cycle: per the Administrative Office of the U.S. Courts, business bankruptcy filings rose 16.9% year over year, to 26,941, in the twelve months ending June 30, 2026, up from 23,043 the year before, part of a broader climb that has continued every quarter since June 2022, though filings remain far lower than historical highs. That is a live signal an ISO can track today rather than a historical comparison to reason from after the fact.
Riding Out the Next Growth Run Differently
None of this is a reason to slow down submission volume during a strong stretch. It is a reason to be honest about what a growth run is measuring, which is new-deal flow, not portfolio health, and those are not the same thing until something forces them to be compared directly.
A portfolio built on submissions that were properly vetted going in, rather than pushed through to hit a monthly number, carries less of the repayment-risk exposure this piece describes before a downturn ever starts, since the quality of what came in the door was never a question that had to wait for a recession to get answered.
Sources
The external data in this article draws on the sources below. Figures described in the text as estimates or industry triangulations are directional and are not attributed to a single dataset.
- Crestmont Capital, SBA Loan Default Rate Statistics: What Happens When Businesses Can’t Repay
- Administrative Office of the U.S. Courts, Bankruptcies Rise 12.2 Percent