DICGC deposit insurance

Bring the hierarchy down to the instrument an ordinary depositor actually holds. A bank deposit's moneyness (its promise to convert to central-bank money at par, on demand) rests on two supports: the bank's own solvency and liquidity, and, standing behind that, deposit insurance up to a stated limit. India's DICGC (Deposit Insurance and Credit Guarantee Corporation, a wholly owned RBI subsidiary) insures deposits up to ₹5,00,000 per depositor, per bank, raised from ₹1,00,000 in the 2020 Union Budget, effective February 2020 (check the current limit against the DICGC Act, 1961, as amended, and DICGC's own notifications). The US's FDIC insures deposits up to $250,000 per depositor, per insured bank, per ownership category, made permanent by the Dodd-Frank Act (2010), after a temporary crisis-era increase from $100,000 (check the current limit against the FDIC's own rules). Below the limit, a deposit's moneyness is, for practical purposes, complete: the depositor is made whole regardless of what happens to the bank. Above the limit, the balance is an uninsured creditor claim on the bank, real, usually recoverable in an orderly resolution, but no longer carrying the state-backed par guarantee that makes the insured portion behave like money in the fullest sense.

Why does the deposit insurance limit matter?

This is not a small technical footnote for most retail balance sheets (the overwhelming majority of individual accounts sit comfortably under either limit) but it becomes the decisive fact for any institution whose deposit base skews toward large corporate, payroll, or institutional balances, exactly the concentration that turned a single mid-sized US bank's failure into a systemic event in March 2023. Worked Example 5 quantifies this precisely on two synthetic banks, one India-side and one US-side, and shows the same structural result on both: the overwhelming majority of accounts, by count, are fully insured, while a much smaller majority, or even a minority, of the deposit base, by value, actually is, because large balances concentrate in a small number of accounts. That gap between "percent of accounts covered" and "percent of money covered" is the retail-deposit version of exactly the same moneyness question this section opened with, and it is Guardrail #2 (Match) territory the instant you read any bank's "insured deposit ratio" without asking which of the two percentages it means.

Modern Data Analysis. Every T-account chain in this section, a loan creating a deposit, that deposit draining reserves as it is spent, a bank's daily reserve position against its requirement, is a small, deterministic state machine: given a starting balance sheet and a stream of transactions, apply each one's effect to the relevant accounts and check the identity holds. In pandas, this is a handful of lines: a small function that takes a bank's opening balance-sheet dict and a DataFrame of transactions (loan originations, settlements, interbank borrowings) and returns the running balance-sheet history, with an assertion after every row that assets equal liabilities plus equity. Scaled up, the identical logic reconciles a real, multi-bank system's published call-report or Pillar 3 data, computing each bank's actual capital-ratio headroom from its disclosed capital and RWA, the way Worked Examples 2 and 3 do by hand here, but across an entire banking sector's filings at once. The by-hand version in this module is not a lesser stand-in for that code; it is how you know, before the code runs, what the answer has to be, so a vectorized reconciliation that disagrees tells you immediately that the code, not your understanding: has the bug.