The Lending Squeeze

Every quarter since 1996, the Federal Reserve Board has conducted the Senior Loan Officer Opinion Survey on Bank Lending Practices — known colloquially as SLOOS. About 80 domestic banks and 24 U.S. branches of foreign banks participate. The survey asks whether, over the past three months, banks have tightened or eased their lending standards across various loan categories. For credit cards, the key question boils down to: did you make it harder or easier for someone to get approved?

The DRTSCLCC series reports the net percentage — the share of banks tightening minus the share easing. A reading of +30% means 30 percentage points more banks tightened than eased. A reading of −20% means 20 points more banks eased than tightened. Zero means the banking system is holding steady.

What makes SLOOS so valuable isn’t what it measures today but what it predicts tomorrow. Lending standards are a leading indicator. When banks start tightening, credit growth slows 6–12 months later. When they ease, borrowing accelerates. The data is like watching a dam operator open or close the floodgates — you can see the water level change coming before it arrives.

The survey also tracks subprime credit card lending separately (SUBLPDCLCTENQ), asking banks whether they’ve tightened policies specifically for customers who don’t meet standard credit scoring thresholds. This subprime measure tends to move in the same direction as the overall reading but with wider swings — banks are quicker to cut off subprime borrowers when trouble appears and quicker to court them when times are good.

The DRTSCLCC series begins in 1996 with a reading of 25.0% — one in four net banks tightening — reflecting the aftermath of the mid-1990s credit card delinquency wave. Banks had been burned by aggressive expansion into subprime credit card lending during the early 1990s, and the hangover was still visible. Tightening peaked at an extraordinary 48.9% in the second half of 1996, the highest reading until the financial crisis over a decade later.

From that defensive posture, banks gradually eased. By 1999, the net tightening reading had fallen to single digits, and by mid-2000 it briefly turned negative (−2.6%) for the first time — more banks were making it easier to get a credit card than harder. The 2001 recession caused a brief reversal (back to 20%), but it was nothing compared to what was coming.

The mid-2000s represented the loosest credit card lending standards in the survey’s history. From 2004 through early 2007, the reading hovered between −11.1% and +3.2%, with banks competing aggressively for cardholders. This was the era of zero-percent balance transfer offers, pre-approved mailers flooding mailboxes, and credit limits being raised without request. Capital One, which had pioneered data-driven credit card marketing in the 1990s, was running 60,000 different experiments simultaneously to find the optimal combination of rate, limit, and fee structure for each consumer segment.

The financial crisis produced the first great squeeze. In Q1 2007, the net tightening reading was 0.0% — perfectly neutral. By Q2 2007, it had flipped to −11.1%, meaning banks were still easing even as the mortgage market was beginning to crack. The first signs of trouble didn’t appear until Q4 2007, when the reading turned positive at 3.2%. Then came 2008.

The tightening escalated quarter by quarter like an air raid siren: 9.8% in Q1 2008, 32.4% in Q2, 66.7% in Q3, 58.8% in Q4. In just twelve months, the banking system went from neutral to the most restrictive credit card lending environment since the survey began. Two-thirds of banks were actively making it harder to get a credit card. Lines were being cut, accounts closed, applications denied. As we traced in Episode 7, bank-held credit card balances would go on to decline 22% over the following two years.

The second great squeeze came with COVID, and it was even more dramatic. From a reading of 13.6% in Q1 2020 (already elevated by pre-pandemic concerns), tightening exploded to 71.7% in Q3 2020 — the highest reading in the survey’s entire 30-year history. Nearly three-quarters of banks simultaneously tightened credit card lending standards. Synchrony Financial, one of the largest store credit card issuers, cut credit limits for millions of dormant accounts. JPMorgan restricted new credit card applications from existing customers with limited credit history.

But the COVID squeeze differed from the GFC squeeze in one crucial way: it reversed almost instantly. The GFC tightening stayed above 30% for four consecutive quarters (Q2 2008 through Q1 2009) and didn’t turn negative until Q3 2010 — over two years of sustained restriction. The COVID tightening peaked in Q3 2020 and turned negative by Q1 2021 — just six months later. By Q3 2021, the reading hit −37.3%, the most aggressive easing in the survey’s history. Banks had gone from maximum tightening to maximum easing in twelve months flat.

The subprime measure (SUBLPDCLCTENQ) — tracking bank policies specifically for borrowers who don’t meet credit score thresholds — amplifies every move in the overall reading. When the overall DRTSCLCC hit 66.7% during the GFC, the subprime measure reached 57.6%. When COVID pushed the overall to 71.7%, the subprime reading peaked at 48.9%. And when banks eased after COVID, the subprime measure turned negative even faster — hitting −11.1% by Q2 2022.

The subprime data reveals something important about the current cycle. After the post-pandemic easing, banks began tightening again in Q4 2022, and the subprime measure climbed to 25.6% by Q3 2023 before gradually moderating. As of Q1 2026, the subprime reading stands at 4.5% — barely positive, suggesting banks are neither actively courting nor actively avoiding weaker borrowers. The overall DRTSCLCC has dropped all the way to 0.0% — perfect neutral.

This neutrality is notable because it comes during a period of rising delinquency. As detailed in Episode 5, credit card delinquency reached 2.94% in Q4 2025, above pre-pandemic levels. And charge-offs hit 4.11%, as we covered in Episode 6. In past cycles, delinquency rates this high would have prompted significant tightening. The fact that banks are holding steady suggests either that the current losses are within acceptable bounds or that competitive pressure is preventing any individual bank from pulling back.

The relationship between lending standards and actual credit growth follows a reliable pattern with an approximate two-quarter lag. When net tightening crossed above 30% in Q2 2008, revolving credit peaked in May 2008 and began contracting from Q3 2008 onward. When tightening finally turned negative in Q3 2010, credit card balances stabilized by early 2011 and resumed growth by 2014 (the long gap reflecting the depth of the crisis).

The COVID sequence was tighter: 71.7% tightening in Q3 2020, revolving credit trough in Q1 2021, easing at −37.3% by Q3 2021, and the great credit card rebound of 2022 (17.1% growth, as noted in Episode 7). The post-2022 tightening cycle, which peaked at 36.4% in Q3 2023, appears to have already moderated the growth rate — card balance growth decelerated from 17% in 2022 to 9% in 2023 to 5% in 2024.

With the reading now at zero, the outlook is genuinely uncertain. A neutral reading means no new headwinds from the banking system — credit will grow at whatever rate consumer demand and existing underwriting standards allow. But it also means there’s no easing tailwind to accelerate growth. The trillion-dollar plateau in bank credit card balances, discussed in Episode 7, reflects this equilibrium: banks are willing to lend, but consumers aren’t rushing to borrow.

The wild card is what happens if the economy weakens. The survey data shows that banks can go from neutral to extreme tightening in just two quarters — the move from 0% to 66.7% in 2008 took only nine months. If delinquencies continue to climb from their current 2.94%, the next SLOOS could be the canary in the coal mine for a broader credit contraction.