EVE vs NII: The Two Lenses of Interest Rate Risk

Every question about interest rate risk in the banking book (IRRBB) can be asked in two ways, and the two answers routinely disagree. Economic value of equity (EVE) asks what the balance sheet is worth if rates move; net interest income (NII) asks what happens to the margin. Understanding why they conflict, and why every modern regulatory framework requires banks to measure both, is the single most important piece of intuition in this field. It is also, as Silicon Valley Bank demonstrated in March 2023, a matter of survival. 

The economic value lens: EVE

The economic value of equity is the present value of all the banking book’s asset cash flows minus the present value of all its liability cash flows: the liquidation value of the book, in effect, computed on a run-off basis, meaning positions amortize and are not replaced. The headline metric, delta EVE, measures how that value changes when the yield curve moves.

EVE captures the full lifetime effect of a rate change on everything currently on the balance sheet. A 20-year fixed-rate mortgage contributes twenty years of repriced discounting, not just next year’s coupon. This makes EVE the long-term, “total economic truth” view of the balance sheet.

It also has two structural blind spots. EVE says nothing about when losses arrive: a present-value hit that materializes over two decades looks identical to one that lands next quarter. And because it runs the book off, it ignores future business entirely.

One technical choice matters here: EVE is computed by discounting cash flows on a risk-free (or near-risk-free) curve per currency, and banks may include or exclude commercial margins from the cash flows, a choice the EU’s supervisory outlier test prescribes to keep results comparable (the internal mechanism behind that choice is funds transfer pricing; see [link: D16]).

The earnings lens: NII

Net interest income analysis asks a nearer-term question: how does the interest margin evolve over the next one to three years under a given rate scenario? It is usually computed on a constant or dynamic balance sheet, in which maturing positions are replaced, so it reflects the bank as a going concern rather than a book in run-off.

Delta NII captures what will actually hit the P&L, and therefore capital via retained earnings, in the near term. Unlike EVE, it works with full contractual rates: the commercial margin is precisely the income being measured.

Its blind spot is the mirror image of EVE’s: NII sees nothing beyond its horizon. A bank can show stable, even improving, earnings for the next three years while the economic value of its book quietly erodes. 

EVE vs NII at a glance  

Why the two lenses conflict

Consider a bank funding long fixed-rate assets with short-term deposits: the classic profile produced by maturity transformation, the core banking business of turning short-term, often instantly withdrawable funding into long-term assets.

If rates rise, NII often improves: the existing assets are locked, but new lending and reinvestment reprice upward faster than sticky deposit rates rise. EVE, meanwhile, falls, because the locked-in fixed-rate assets are now discounted at higher rates. If rates fall, the reverse happens.

The two lenses are not giving contradictory information; they are answering different questions about the same position. The earnings lens sees the near-term cash effect of repricing. The value lens sees the full lifetime cost of having locked in yesterday’s yields. A bank optimizing only NII will happily build exactly the position that destroyed Silicon Valley Bank.

Why every framework requires both 

Because either lens alone can be gamed, or simply misread, every modern regulatory framework requires banks to measure both. Basel Principle 4 demands it globally. The EBA Guidelines embed it in EU law, and the EU now runs a supervisory outlier test on each measure: delta EVE against 15% of Tier 1 capital, and delta NII against 5% of Tier 1. In the US, the 2010 Interagency Advisory (SR 10-1) carries the same requirement, with supervisory practice historically weighted towards the earnings lens. 

The measurements themselves run off a common set of prescribed rate scenarios, six instantaneous shocks in the Basel canon, recently recalibrated.

Duration of equity: the summary statistic

A useful way to compress the economic value lens into one number is the duration of equity. If the weighted duration of assets exceeds the duration of liabilities, equity value falls when rates rise, and the bank is “positioned long”: it benefits from falling rates in value terms. Most traditional lenders naturally sit there, because maturity transformation makes them so.

The SVB lesson

For decades, IRRBB was regarded as a second-tier risk: real, but rarely fatal. March 2023 ended that view. Silicon Valley Bank, then the 16th largest US bank, failed in substantial part because of textbook IRRBB, unmanaged.

During the 2020-21 deposit boom, SVB invested heavily in long-duration, fixed-rate securities, much of the portfolio classified as held-to-maturity, funded by large, uninsured, highly concentrated corporate demand deposits. When the Federal Reserve raised rates by over 400 basis points in 2022, the economic value of those securities collapsed: unrealized losses grew to a magnitude comparable to the bank’s entire equity base, kept out of sight by amortized-cost accounting and the regulatory capital rules. 

The two lenses gave opposite signals, and the bank chose to look only at the comfortable one. Internally, SVB had focused its risk management on short-term earnings rather than economic value, removed hedges that protected against rising rates, and adjusted the behavioral assumptions in its economic-value model when the results breached limits. When depositors ran, withdrawing $42 billion in a single day, the bank had to sell securities and crystallize the losses. It failed within 48 hours.

The lesson is not that EVE is the “right” lens and NII the “wrong” one. It is that each lens is blind to what the other sees, and a bank that manages only one of them is managing half its risk. 

Key takeaways 

• EVE is the present value of banking book cash flows on a run-off basis: full lifetime effect, but blind to timing and future business.

• NII is the margin over a 1-3 year horizon on a constant or dynamic balance sheet: the going-concern view, but blind beyond its horizon.

• The classic long-fixed-assets, short-deposits profile means rates up often lifts NII while cutting EVE; the two lenses conflict by construction, not by error.

• Basel Principle 4, the EBA Guidelines and the US 2010 Advisory all require banks to measure both; the EU runs a supervisory outlier test on each (EVE 15%, NII 5% of Tier 1).

• Duration of equity compresses the value lens into one number; most traditional lenders are positioned long. 

• SVB is the cleanest real-world proof: optimizing only short-term earnings builds exactly the position that kills you when rates move.

Mirai ALM & Liquidity runs EVE and NII stress testing side by side, across the full Basel scenario set, with the deposit behavior models that make or break both measures.

For the full picture, including the Basel BCBS 368 framework, the deposit-modeling debates behind the SVB failure, and how the EU, UK, and US regimes each address it differently, read the complete guide "IRRBB: A Complete Reference, From Zero to Expert."

To see how Mirai Regulatory Reporting automates IRRBB measurement, EVE and NII stress testing, and regulatory reporting across these frameworks, learn more here.