Chapter 1 · Part 1 — The KFI Table and How to Read It
The KFI Table and How to Read It
What Nepal Rastra Bank publishes every quarter, the three disclaimers printed on it, and the five families of columns.
Every quarter, Nepal Rastra Bank's Bank Supervision Department publishes a single sheet of paper that tells you more about a Nepali bank than its entire annual report will. It is called Key Financial Indicators. Almost nobody outside the industry reads it.
This week you will learn to read it line by line.
What the document actually is
Two separate releases, published on the same date, in the same format:
┌────────────────────────────────────────────────────────────────────┐ │ KEY FINANCIAL INDICATORS — Chaitra end 2082 (mid-April 2026) │ ├────────────────────────────────────────────────────────────────────┤ │ │ │ Release A Commercial Banks 20 institutions │ │ Class "A" Rs 7,136,173 million │ │ of deposits │ │ │ │ Release B Finance Companies 15 institutions │ │ Class "C" Rs 139,692 million │ │ of deposits │ │ + 2 listed separately as "Problematic" │ │ │ └────────────────────────────────────────────────────────────────────┘
The commercial-bank sheet is about fifty-one times larger by deposits than the finance-company sheet. That size difference is the single most useful thing to hold in your head all week: the two sheets look alike, but they describe institutions operating under different pressures, and a ratio that is ordinary on one sheet can be alarming on the other.
Three warnings printed on the document itself
NRB puts three disclaimers on the sheet. Investors ignore all three.
"Provisional." These are unaudited numbers submitted by the institutions themselves. The audited annual figures will differ, sometimes materially, and the direction of the difference is not random — provisioning and impairment are exactly the lines an auditor adjusts.
"Based on regulatory requirement format, and so, the NFRS based statistic may differ." This is the sentence that trips up analysts who come from reading published accounts. The bank's NFRS financial statements and this table are two different measurement systems over the same underlying business. A bank can show one impairment number to its shareholders and a different provision number to its regulator, and both can be correct.
"Monthly Average" marked with an asterisk on CD Ratio, SLR, Base Rate and Spread. Those four are not point-in-time snapshots. They are averages across the month, which is precisely how NRB stops an institution from window-dressing on the reporting date.
POINT-IN-TIME MONTHLY AVERAGE
(measured on the last day) (measured every day, averaged)
│ │
│ ● ← what you │ ●●●●●●●●●●●●●●●●●●●●
│ ╱ report │ ← what you report
│ ●●●●●●●●╱ │
│ ●╱ │
└────────────────── └──────────────────
you can arrange you cannot arrange
one good day thirty good daysThe column map
Fourteen numeric columns on the bank sheet. They fall into five families, and the rest of this week is one day per family.
┌──────────────┬──────────────────────────────────────┬──────────┐ │ FAMILY │ COLUMNS │ DAY │ ├──────────────┼──────────────────────────────────────┼──────────┤ │ SOLVENCY │ Core Capital │ Day 2 │ │ can it │ Total Capital Fund │ Day 2 │ │ absorb │ CCAR % │ Day 3 │ │ losses? │ CAR % │ Day 3 │ ├──────────────┼──────────────────────────────────────┼──────────┤ │ SCALE │ Total Deposit / LCY Deposit │ Day 1 │ │ how big? │ Total Loan / LCY Loan │ Day 1 │ ├──────────────┼──────────────────────────────────────┼──────────┤ │ LIQUIDITY │ CD Ratio % │ Day 4 │ │ can it pay │ Net Liquidity % │ Day 4 │ │ tomorrow? │ SLR % │ Day 4 │ ├──────────────┼──────────────────────────────────────┼──────────┤ │ ASSET │ NPL % │ Day 5 │ │ QUALITY │ Net NPL % │ Day 5 │ │ will it be │ │ │ │ repaid? │ │ │ ├──────────────┼──────────────────────────────────────┼──────────┤ │ PRICING │ Base Rate % │ Day 6 │ │ what does │ Spread % │ Day 6 │ │ it earn? │ │ │ ├──────────────┼──────────────────────────────────────┼──────────┤ │ MANDATE │ Prescribed sector columns (6) │ Day 7 │ │ who must │ Total Prescribed Sector Loan % │ Day 7 │ │ it lend to? │ Deprived / Specified (Class C) │ Day 7 │ └──────────────┴──────────────────────────────────────┴──────────┘
The order is not accidental. It is the order a supervisor asks the questions in. Solvency first, because an institution with no capital is a resolution problem rather than an investment. Liquidity second, because banks die of illiquidity long before they die of insolvency. Asset quality third, because it is what turns into a capital problem twelve months later. Pricing fourth, because it determines whether the institution can earn its way out. Mandate last, because it constrains all four.
LCY, and the column most people skip
Two columns look redundant: Total Deposit next to LCY Deposit, Total Loan next to LCY Loan. LCY is Local Currency — Nepali rupees. The difference between the two is the institution's foreign-currency book.
For most Nepali banks the gap is small. For a few it is the whole story.
TOTAL vs LCY DEPOSIT — the FCY gap, Chaitra end 2082 (Rs million; gap = Total − LCY = foreign-currency deposits) Standard Chartered 147,395 total 119,645 LCY gap 27,750 18.8% Nabil 588,835 total 566,477 LCY gap 22,358 3.8% Himalayan 344,139 total 319,297 LCY gap 24,842 7.2% NIC Asia 336,710 total 336,444 LCY gap 266 0.1% Rastriya Banijya 601,181 total 592,307 LCY gap 8,874 1.5% System 7,136,173 6,896,215 gap 239,958 3.4%
Standard Chartered funds nearly a fifth of its book in foreign currency; NIC Asia essentially none. That is not a quality judgement — it reflects who the customers are. But it tells you which bank has currency and cross-border exposure worth asking about, and which does not. A remittance shock or an exchange-rate move does not touch those two banks equally.
Reading a single row
Take row 18 of the bank sheet, and read it the way a supervisor would.
┌───────────────────────────────────────────────────────────────────────┐ │ NMB BANK LIMITED — Chaitra end 2082 (provisional) │ ├───────────────────────────────────────────────────────────────────────┤ │ │ │ SOLVENCY Core capital Rs 31,548 m │ │ Total capital fund Rs 41,297 m │ │ CCAR 9.76 % │ │ CAR 12.77 % │ │ │ │ SCALE Deposits Rs 300,468 m │ │ Loans Rs 261,585 m │ │ │ │ LIQUIDITY CD ratio 84.68 % <- highest of the 20 │ │ Net liquidity 26.32 % │ │ SLR 20.89 % <- lowest of the 20 │ │ │ │ QUALITY NPL 4.95 % │ │ Net NPL 1.77 % │ │ │ │ PRICING Base rate 5.23 % │ │ Spread 3.50 % │ └───────────────────────────────────────────────────────────────────────┘
One row, and a picture already forms: this is the most lent-out bank on the sheet. Highest CD ratio, lowest SLR, liquidity near the bottom of the range. It is working its balance sheet harder than any of its nineteen peers.
That is not automatically bad — a bank that lends more earns more. But it is a bank with less room to absorb a surprise, and the row tells you to go looking for what happens if deposits leave.
Compare it with row 6.
┌───────────────────────────────────────────────────────────────────────┐ │ STANDARD CHARTERED BANK NEPAL — Chaitra end 2082 (provisional) │ ├───────────────────────────────────────────────────────────────────────┤ │ CCAR 16.14 % CAR 17.82 % <- highest capital of the 20 │ │ CD ratio 53.84 % <- lowest lending intensity │ │ Net liquidity 53.84 % <- highest liquidity │ │ SLR 48.48 % <- four times the minimum │ │ NPL 1.81 % Net NPL 0.37 % <- near-cleanest book │ │ Base rate 4.37 % Spread 3.67 % │ └───────────────────────────────────────────────────────────────────────┘
The mirror image. Enormous capital, enormous liquidity, very little lending, very clean assets. A fortress — and a bank leaving a great deal of earning power on the table.
Neither row is "the right answer". The point of the KFI table is that it makes the trade-off visible and comparable, in the same format, for every institution in the class.
Where the numbers come from
The institutions submit them. NRB does not compute them independently for this release; it compiles returns filed under the NRB reporting forms. The finance company sheet says so explicitly: "Based on Unaudited Reports submitted by Financial Institutions."
This matters for how much weight you put on any single cell:
- Structural figures (deposits, loans, paid-up capital) are hard to misstate and reconcile to the published accounts.
- Averaged figures (CD ratio, SLR, base rate, spread) are hard to manipulate because they are monthly averages.
- Judgemental figures (NPL, and therefore Net NPL and the capital that depends on provisioning) rest on the institution's own loan classification. This is the softest number on the sheet, and it is the one that moves most between the provisional release and the audited accounts.
The two sheets are not the same sheet
The finance-company release carries columns the bank release does not, and drops some it has:
BANK SHEET has FINANCE SHEET has
───────────────── ──────────────────
LCY Deposit / LCY Loan Paid up Capital
Net NPL % Total RWE (the denominator, in rupees)
6 prescribed-sector columns Deprived Sector %
Total Prescribed Sector Loan % Specified Sector %
Individual Deposit
+ a separate "Problematic" blockTwo differences are worth flagging on day one.
The finance sheet prints Total RWE in rupees. The bank sheet does not. That single column lets you verify CCAR and CAR arithmetic yourself, which you cannot do from the bank sheet. We will use it on Day 3.
The finance sheet has a "Problematic Finance Companies" block below the main table — institutions NRB has separated out. Two are listed at Chaitra end 2082. Their numbers are extraordinary and we will look at them properly on Day 5.
What this table cannot tell you
Be clear about the limits before you rely on it.
It contains no profit figure. No net interest income, no operating profit, no EPS, no ROE. You cannot value a bank from this sheet.
It contains no trend. One date, one column each. Every judgement you make from a single release is a level, not a direction — and for most of these ratios the direction matters more. Keep the previous releases.
It contains no group view. These are the licensed institution's regulatory returns, not consolidated group accounts including subsidiaries.
It says nothing about governance, management or strategy, which is what actually determines whether next quarter's NPL goes up or down.
What you should be able to do now
- Say what the KFI release is, who publishes it, and how often.
- Name the three disclaimers printed on it and why each one matters.
- Group the fourteen columns into the five supervisory families.
- Explain why four of the ratios are monthly averages rather than snapshots.
- Read one institution's row and describe its balance-sheet posture in a sentence.
- State three things the table cannot tell you.
Tomorrow: the capital columns — what Core Capital and Total Capital Fund actually contain, and why the gap between them is the most informative number in the solvency family.
Saved in this browser only — there is no account to create. Clearing your browser data clears your progress.