Return on equity and return on assets: measuring bank profitability

Shareholders' equity is assets minus liabilities: the owners' stake. This is the number investors care about most, because it tells them the return on their capital.

A worked example

Let's take a simple, illustrative bank. Call it Meridian Bank (fictional, round numbers for teaching).

• Total assets: $100 billion
• Shareholders' equity: $10 billion
• Net income: $1.2 billion

ROA = 1.2 / 100 = 1.2%

ROE = 1.2 / 10 = 12%

Notice equity ($10bn) is only 10% of assets ($100bn). That 10-to-1 ratio is the engine that turns a modest 1.2% ROA into a punchy 12% ROE. That engine is leverage.

The DuPont link: how leverage bridges ROA and ROE

The DuPont framework (named after the company that popularized it in the 1920s) breaks ROE into components. The banking version is elegantly simple:

ROE = ROA x Leverage

Where Leverage = Total assets / Shareholders' equity (also called the equity multiplier).

Plugging in Meridian:

• ROA = 1.2%
• Leverage = 100 / 10 = 10x
• ROE = 1.2% x 10 = 12% ✓

This equation is the single most important relationship in bank profitability. It says a bank can boost ROE two ways:

1. Earn more per asset (higher ROA): better margins, lower costs, fewer loan losses.

2. Use more leverage (higher multiplier): fund the same assets with less equity.

Here is the catch. Route 2 is dangerous. More leverage means thinner equity cushions to absorb losses. In 2008, banks with 30x or 40x leverage discovered that a small drop in asset values wiped out their entire equity. That is precisely why regulators now cap leverage.

The regulatory ceiling on leverage

After the 2008 crisis, the Basel III framework (global standards set by the Basel Committee on Banking Supervision) introduced a leverage ratio: a minimum floor of equity-like capital against total exposure, regardless of how "safe" the assets look on paper.

The Basel III minimum leverage ratio is 3%, with a surcharge for the largest global banks. In the US, the biggest banks face an "enhanced supplementary leverage ratio" that is higher. In practice this caps the equity multiplier: a 3% capital floor implies leverage cannot much exceed roughly 33x on that measure, and real-world banks run well below that.

Read the source directly: the Basel III framework overview from the Bank for International Settlements is free and authoritative.

The takeaway: post-2008, banks can no longer juice ROE with unlimited leverage. So ROE improvement now has to come mostly from ROA, meaning real operating performance.

Why 10% ROE is the magic number

Investors do not fund a bank for free. They demand a return for the risk they take. That required return is the cost of equity (part of the broader cost of capital).

For a typical large US bank, the cost of equity is commonly estimated at around 10% (an estimate; it moves with interest rates and market risk appetite). This gives us the hurdle:

• ROE above cost of equity → the bank creates value. Investors reward it with a share price above book value.
• ROE below cost of equity → the bank destroys value. Shares trade below book value.

So "10% ROE" became shorthand for "clearing the hurdle." A bank earning 12% (like Meridian) is comfortably above it and creates value. A bank earning 7% is destroying it.

The US vs Europe gap

Here is where the opening scene resolves. For years, large US banks have consistently earned ROEs above their cost of equity, while many large European banks have struggled to reachreachThe number of unique people exposed to your message in a given period. Unlike impressions, reach counts each person once, no matter how often they see it.Voir la définition complète → it.

Rough, widely cited estimates (directional, not precise, and they shift year to year):

• Large US banks: ROE frequently in the 11% to 15% range in recent strong years.
• Large European banks: ROE often clustered around 7% to 11%, with many banks historically below the 10% hurdle for much of the 2010s. Higher interest rates from 2022 onward lifted European ROEs meaningfully, so the gap narrowed but did not close.

Why the persistent gap? Several structural reasons, all inside finance:

1. Net interest margins (NIM). NIM is interest income minus interest expense, divided by earning assets. It tells you the spread the bank earns on lending. US banks generally enjoy wider NIMs than European banks, partly because of a deeper capital-markets economy and, for much of the 2010s, negative policy rates in the euro area that crushed European margins.

2. Fee and capital-markets income. US banks (think JPMorgan Chase, Bank of America, Goldman Sachs) have huge investment-banking and trading arms that lift ROA. Europe's market is more fragmented across many national champions (BNP Paribas, Deutsche Bank, Santander, and others), so fewer achieve the scale economics of the US giants.

3. Cost efficiency. The cost-to-income ratio (operating costs divided by operating income; lower is better) is often higher at European banks, dragging down ROA. Many European banks carry legacy branch networks and fragmented IT.

4. Fragmentation. The US has one large integrated banking market. Europe still lacks a completed banking union, so banks cannot fully consolidate across borders. Less scale, lower ROA, lower ROE.

Notice how the DuPont logic ties it together. Since Basel III limits the leverage lever equally on both sides of the Atlantic, the US advantage shows up almost entirely in higher ROA: wider margins, more fee income, lower cost ratios.

Reading the ratios like an analyst

When you see a bank's ROE, always decompose it before judging:

• High ROE from high ROA = healthy, sustainable, well-run bank.
• High ROE from high leverage = fragile, flattered by financing structure. Check the capital ratios.
• Low ROE despite decent ROA = the bank is over-capitalized (lots of equity). Safer, but investors may push for buybacks.

Quick sanity check on Meridian: 12% ROE on a 10x multiplier and 1.2% ROA is the profile of a solid, conventionally financed US bank. Nothing exotic.

One more benchmark to keep in your head: a bank trading at a price-to-book ratio above 1.0 is one the market believes earns above its cost of equity. Below 1.0 signals the opposite. This ratio and ROE move together for exactly the reasons above.

Key Takeaways

• ROA = Net income / Assets; ROE = Net income / Equity. For US banks, ROA around 1% or higher and ROE around 10% or higher are the healthy benchmarks (estimates, as of the mid-2020s).
• DuPont link: ROE = ROA x Leverage. Leverage (assets / equity) is the bridge. A 1.2% ROA at 10x leverage produces a 12% ROE.
• Basel III caps leverage (a 3% minimum leverage ratio, higher for the biggest banks), so post-2008 ROE gains must come mainly from ROA, not from piling on debt.
• 10% ROE is the rule-of-thumb hurdle because it approximates the cost of equity. Above it, banks create value and trade above book; below it, they destroy value.
• The US-Europe gap is an ROA gap, driven by wider net interest margins, richer fee income, better cost efficiency, and greater scale, not by different leverage rules.