How do I compare profitability across banks of different sizes?

Focus on ratio-based metrics that normalize for size, especially ROAA (return on average assets). Dollar-based measures like net income are meaningless across size categories because a $50 billion bank will always earn more dollars than a $500 million bank. ROAA, combined with peer group analysis, gives you the clearest picture of which bank is actually more profitable relative to its resources.

A $500 million community bank and a $50 billion regional bank operate in fundamentally different ways. They earn revenue from different mixes of activities, run at different cost structures, hold different levels of capital relative to assets, and face different competitive pressures. Comparing their profitability with a single number, or with metrics designed for same-size comparisons, will produce misleading conclusions almost every time.

The core problem is that raw dollar figures tell you nothing useful. A large regional bank earning $400 million in net income looks far more profitable than a community bank earning $8 million, but that comparison is meaningless without knowing how many assets each bank deployed to generate those earnings. Ratio-based metrics solve this by expressing profitability relative to a common base.

ROAA as the Primary Comparison Tool

ROAA (return on average assets) is the single best metric for comparing profitability across different-sized banks. It measures net income as a percentage of total assets, which removes two major distortions at once: the raw scale difference and the capital structure difference.

Consider a community bank with 11% equity-to-assets and a regional bank at 8% equity-to-assets. If you compare them on ROE (return on equity), the regional bank gets an artificial boost from its thinner capital cushion. ROAA sidesteps this entirely because both banks are measured against their full asset base, not their equity alone.

A practical example: if the community bank posts ROAA of 1.25% and the regional bank posts 1.10%, you know the community bank is generating more profit per dollar of assets regardless of its higher capitalization. That 15 basis point difference is a genuine performance gap, not a leverage artifact.

Why ROE Misleads Across Size Categories

ROE comparisons between banks of different sizes should be interpreted with real caution. Larger banks tend to carry less equity relative to assets for a few reasons:

Geographic and business line diversification reduces their risk profile, which regulators and investors accept as justification for lower capital ratios

More sophisticated treasury and capital management functions allow them to operate closer to regulatory minimums

Market expectations push them toward capital returns (buybacks and dividends) that keep equity levels leaner

This structural gap in capital ratios means a large bank can post a competitive ROE even when its ROAA is mediocre. Two banks might both report 12% ROE, but if one achieves it with 1.20% ROAA and 10x leverage while the other needs just 0.85% ROAA with 14x leverage, the first bank is clearly the stronger operator. The DuPont decomposition (ROE = ROAA x equity multiplier) is the standard way to untangle this.

NIM Needs Size Context

Net interest margin (NIM) varies structurally by bank size, and ignoring that structure leads to bad comparisons. Community banks typically run NIMs in the 3.50% to 4.50% range, while large banks often operate at 2.00% to 3.00%. These differences are baked into how each type of bank does business.

Community banks make smaller, relationship-based loans that command wider spreads. They fund themselves heavily with sticky core deposits from local customers, which keeps their cost of funds low. Large banks originate a higher proportion of lower-spread assets (large corporate loans, traded securities) and rely more on wholesale funding, which is rate-sensitive and more expensive.

A community bank with 3.60% NIM is not outperforming a money center bank with 2.50% NIM. Both may be executing well within their structural realities. Peer NIM comparisons are only meaningful within the same size category.

Efficiency Ratio and Scale Advantages

The efficiency ratio (non-interest expense divided by revenue) is another metric where size creates structural differences. Larger banks benefit from economies of scale that smaller institutions cannot replicate: centralized operations, technology platforms spread across a larger asset base, and the ability to negotiate better vendor pricing.

A 55% efficiency ratio at a $100 billion bank is a very different achievement than 55% at a $500 million bank. The smaller bank hitting that number is demonstrating exceptional cost discipline given its inherent scale disadvantage. Typical efficiency ratios for well-run community banks fall in the 58% to 65% range, while large banks often target 55% to 60%.

When comparing across sizes, pay attention to the gap between a bank's efficiency ratio and its size-appropriate benchmark rather than the absolute number.

Revenue Mix Complicates the Picture

One factor that often gets overlooked in cross-size comparisons is revenue mix. Community banks generate the vast majority of their revenue (often 80% or more) from net interest income. Larger banks typically have more diversified revenue streams: wealth management fees, capital markets activity, mortgage banking, service charges, and trading income can represent 30% to 50% of total revenue.

This matters for profitability comparison in two ways. First, a large bank with significant fee income can maintain solid profitability even with a lower NIM, because non-interest income fills the gap. Comparing only NIM between a fee-heavy large bank and an interest-income-dependent community bank misses half the picture. Second, fee income tends to be less capital-intensive than lending, so a dollar of fee revenue often generates higher returns on assets than a dollar of interest income.

The non-interest income to revenue ratio is worth checking alongside the core profitability metrics. A bank with 35% of revenue from fees operates in a fundamentally different mode than one at 15%, even if they are similar in asset size.

Building Effective Peer Groups

The most reliable approach to cross-size profitability analysis is constructing a proper peer group. Rather than comparing a community bank directly against a money center bank (which creates all the structural distortions described above), compare each bank against others with similar characteristics.

A good peer group typically includes 8 to 15 banks matched on:

Asset size range (for example, $1 billion to $3 billion)

Geographic footprint or market type (urban, suburban, rural)

Business model similarity (commercial-focused, consumer-heavy, or balanced)

Once you have a well-constructed peer group, rank each bank on ROAA, efficiency ratio, NIM, and ROE. A bank that consistently places in the top quartile across multiple metrics is genuinely outperforming, not just benefiting from structural advantages. One that ranks well on ROAA but poorly on efficiency ratio might have strong revenue generation but a cost problem.

For comparing across size categories (rather than within them), ROAA is still the best single number. But even ROAA comparisons are more informative when you understand the structural context behind each bank's results.

Common Mistakes in Cross-Size Comparison

A few patterns regularly trip up investors comparing banks of different sizes:

Comparing absolute dollar earnings or revenue growth rates without adjusting for asset base. A $50 billion bank growing earnings 10% is adding far more dollars than a $1 billion bank growing at 15%, but the smaller bank may be the better investment.

Using ROE as the primary comparison metric without checking whether differences reflect operating performance or just leverage. The DuPont decomposition takes about 30 seconds and eliminates this blind spot.

Treating NIM or efficiency ratio differences as performance gaps when they are actually structural features of different business models. Always compare against size-appropriate benchmarks.

Ignoring revenue mix entirely. Two banks with identical ROAA may achieve it very differently: one through a wide NIM and minimal fee income, the other through a narrow NIM supplemented by diversified fees. The second model may be more stable through interest rate cycles.

Related Metrics

Related Valuation Methods

Peer Comparison Analysis