Peer Group Distributions

Two patterns run through nearly every chart on this page: First, a size gradient: as credit unions get larger, capital cushions and net interest margins compress while funding costs, charge-offs, operating efficiency, and long-term-asset exposure all climb. The large-credit-union model trades thick capital and fat margins for scale, lower unit costs, and more leverage. Second, the distributions tighten as size rises — the smallest credit unions are wildly heterogeneous on every metric, while the two dozen institutions above $10 billion converge into a narrow band. The practical consequence: the typical (median) credit union and the asset-weighted system aggregate are different animals, and most institutions look nothing like the dollar-weighted averages the headlines cite.

Box plots show the interquartile range (P25–P75) with the median (P50) marked. Whiskers extend to P10 and P90; the ◆ marker is the dollar-weighted (system) value — where it sits relative to the median is the gap between the typical credit union and the asset-weighted aggregate. Each peer group is defined by total assets; data as of March 31, 2026 .

The seven asset-size tiers and how many credit unions sit in each. The skew is the point: the largest tier is a couple dozen institutions while the smallest tiers hold thousands — which is why the distributions tighten as size rises (fewer institutions cluster more naturally), and why the largest-tier boxes are a small sample.

No Results

Capital Adequacy

The gradient is starkest here. The median net worth ratio falls from roughly 17.6% at credit unions under $10 million to about 10.2% above $10 billion — and the spread collapses alongside it. The middle half of the smallest tier spans more than twelve points; for the largest tier it spans barely two. Small credit unions hold thick, idiosyncratic buffers; large ones run lean and remarkably uniform. The GAAP equity ratio sits below the regulatory net worth ratio at every size, the gap being unrealized securities losses still carried in AOCI.

Asset Quality

Credit quality splits the two ends of the system from the middle. Delinquency is U-shaped — elevated at the smallest tier (a noisy distribution where many report zero but a long tail runs high) and at the largest (a 0.9% median, the system's highest), with the mid-size tiers cleanest near 0.6%. Charge-offs, by contrast, rise monotonically with size: the median credit union under $10 million writes off essentially nothing, while the median institution above $10 billion charges off 0.74%. The largest tier is therefore both the most delinquent and the most loss-taking — the signature of its consumer- and auto-heavy book.

Earnings

Profitability improves with scale, and gets more reliable doing so. Median ROAA climbs from 0.41% at the smallest tier to 0.83% at the largest, and the distribution narrows from a wide, partly-negative range — many of the smallest credit unions lose money — to a tight, uniformly-positive one. The driver is cost structure, not pricing: cost of funds rises steeply with size (a 0.44% median at the smallest tier versus 1.98% at the largest, as big credit unions compete harder for deposits) and net interest margin falls in step, from about 4.0% to 3.2%. What more than offsets it is operating efficiency — the median efficiency ratio improves from roughly 76% to 44% across the size range, the clearest scale economy in the data.

Liquidity & Sensitivity

Balance-sheet structure tracks size too. The share of assets in long-term holdings — a proxy for interest-rate-risk exposure — rises from a median near 46% at the smallest tier to about 77% at the largest. Small credit unions sit short and liquid; large ones carry more duration and lean harder on wholesale borrowings. It is the same trade as everywhere else on the page: scale bought with more leverage and more rate risk.

Growth

Growth is the exception to the size gradient: median share and membership growth do not trend cleanly with size, clustering instead once you move past the smallest tier. What varies is dispersion. Growth at the smallest credit unions is enormously variable — a single new employer group or a lost sponsor can swing a tiny balance sheet — while larger institutions grow within a tighter band. Read the small-tier boxes as noise, not signal.

Data sourced from NCUA 5300 Call Report filings, latest reporting quarter. Box = interquartile range (P25–P75); whiskers = P10–P90; midpoint = median (P50). Distributions are of per-credit-union values within each asset-size peer group; income and growth ratios are annualized. The median credit union differs from the asset-weighted system aggregate — the larger the size skew in a metric, the wider that gap.