A central bank announces a rate and the world's borrowing costs adjust. The transmission from announcement to actual market rates involves specific machinery that has changed considerably in the past fifteen years.
The old system
Historically, central banks kept reserves in the banking system scarce and adjusted the quantity to move the rate at which banks lent reserves to each other overnight.
Small open market operations — buying or selling short-term securities — moved the quantity of reserves and therefore the price.
Which required continuous fine-tuning and worked because reserves were genuinely scarce.
What changed
Quantitative easing flooded the system with reserves. Banks held vastly more than they needed.
Which broke the old mechanism entirely, because adding or removing small quantities from an enormous surplus does nothing to the price.
The response was to shift to setting rates administratively.
The floor system
Central banks now pay interest on reserves held with them.
Which sets a floor, because no bank would lend to another below the rate it can earn risk-free at the central bank.
A lending facility sets a ceiling, because no bank would borrow above the rate at which it can borrow from the central bank against collateral.
Market rates trade in the corridor between, generally near the floor when reserves are abundant.
The policy rate is then set by moving the administered rates rather than by managing quantities.
The leakage problem
Not everyone can hold reserves at the central bank.
Money market funds, government agencies and some other institutions can lend in the same markets but cannot access the deposit facility.
Which means they will lend below the floor, since their alternative is worse.
Additional facilities open to a broader set of counterparties were introduced to address this, and their usage has become a closely watched indicator of how much surplus cash is in the system.
Transmission to actual borrowing costs
The policy rate affects overnight lending between banks. It does not directly set a mortgage rate.
Transmission runs through several channels — bank funding costs, expectations of future policy embedded in longer rates, asset prices, exchange rates and credit availability.
Each operates with a different lag and different reliability.
Which is why the effect of a rate change on the real economy is generally estimated in quarters rather than weeks, and why central banks describe policy as acting with long and variable lags.
Forward guidance
Because expectations of future policy affect long rates today, communicating the expected path is itself a policy tool.
Which is why statements, projections and speeches move markets as much as the rate decision, and frequently more.
The difficulty is that guidance which becomes a commitment constrains future flexibility, and guidance which is easily abandoned is not credible.
Central banks have moved toward conditional formulations that tie policy to economic outcomes rather than to dates, which is more robust and less clear.
Balance sheet policy
A separate instrument operating alongside the rate.
Buying long-dated assets pushes their prices up and yields down, affecting rates the policy rate reaches only indirectly.
Unwinding those holdings — reducing the balance sheet — has been done both by allowing maturing holdings to run off and by selling.
The effects of unwinding are less well understood than the effects of buying, since there is far less historical experience of it.
Independence and its limits
Most major central banks operate with statutory independence over instruments while pursuing objectives set politically.
Which is a deliberate design, based on evidence that politically controlled monetary policy produces higher inflation.
The independence is not absolute — mandates can be changed by legislatures, and central banks are accountable to them — and the tension between independence and accountability is permanent rather than solvable.
Why follow this
Because rate expectations drive the pricing of essentially every financial asset, and understanding the mechanism makes the commentary legible rather than mystifying.
Ample against abundant reserves
A current design question with practical consequences.
As balance sheets shrink, central banks must decide how far to go — back to scarcity, or stopping at a level where reserves remain plentiful enough that rate control stays administrative.
Most have signalled a preference for the second, which keeps the floor system and avoids the operational complexity of daily fine-tuning.
The difficulty is knowing where that level is, since demand for reserves is not directly observable and only becomes apparent when it is undershot.
One major system discovered this the hard way when overnight funding rates spiked after reserves fell below what the market required, which prompted a rapid reversal.
The exchange rate channel
For smaller open economies, this transmission route dominates the others.
A rate change affects the currency, which affects import prices directly and rapidly, and passes into consumer prices faster than any domestic channel operates.
Which means central banks in such economies pay closer attention to the currency than large-economy central banks do, without formally targeting it.