1. What a central bank actually does: mandate, policy rate and why it does not control long rates

Macro 2: Central banks

The previous level closed on an uncomfortable idea: the price of almost everything is a future stream of cash divided by a number that macro moves. It is time to look squarely at the institution that moves part of that number, and it is worth starting by pinning down the limits of its power, because almost everyone exaggerates it in one direction and underrates it in the other.

The mandate is not a legal footnote: it drives the decisions

A central bank does what its founding legislation obliges it to do, and those laws are not the same everywhere. That asymmetry explains why two institutions can look at identical data and act differently.

• The American Federal Reserve operates under a dual mandate: maximum employment and price stability. When unemployment rises and inflation falls, both halves point the same way. The conflict appears when they pull apart: inflation climbing while employment weakens forces it to choose which of the two objectives it betrays.
• The European Central Bank has a hierarchical mandate: its primary objective is price stability, and only where that is not compromised may it support the general economic policies of the Union. Employment is subordinate, not an equal partner.

This is not institutional folklore, it predicts behavior. Faced with the same shock, the Fed has legal room to hesitate and the ECB has less. When you see two different reactions to the same data point, look at the mandate before you look at the governor's personality.

Add one more layer: almost all of them now pursue an inflation target of around 2%. That is not a law of physics but a convention, high enough above zero to leave a buffer before deflation, low enough that nobody gives it a thought when signing a contract.

The policy rate is not decreed: it is manufactured in the market for reserves

Here sits the most widespread misunderstanding. When it is announced that rates are going up to 4%, no decree obliges anyone to lend at 4%. What is being set is the price in one small, very specific market: the market for reserves, the deposits commercial banks hold at the central bank to settle payments among themselves.

The central bank controls that price because it is the only institution that can create the asset. Under the current system it does so by setting a floor and a ceiling:

• A floor: it pays interest on the reserves banks park with it. No bank will lend to another below what the central bank pays it at no risk whatsoever.
• A ceiling: it lends against collateral at a slightly higher rate. No bank will pay another more than it costs to go to that window.

Between those two rails sits the effective overnight rate: a rate that almost no economic agent pays directly and which, even so, is the anchor from which the entire ladder of maturities is built.

The transmission chain, and why it breaks halfway along

Several links separate the overnight rate from your mortgage, and each one adds delay and loses signal:

• From the overnight rate to short market maturities. This is the most reliable link and it is close to automatic.
• From short maturities to bank credit. It depends on whether the banking system is healthy, whether it has capital and whether it wants to lend. A damaged banking system transmits badly.
• From credit to the real economy. The lag is measured in quarters, not days: the bulk of the effect of a rate rise takes four to six quarters to show up in inflation.
• From all of the above to asset prices. This link is instantaneous, because markets do not wait for the real effect: they price the expectation.

That lag explains a scene that keeps repeating itself: the central bank carries on raising rates while the economy is already cooling. It is not blindness, it is driving while looking at a rear-view mirror running a year and a half behind.

Where its power ends: the long end is not its own

Level 1 established that the market sets the long end. Now the why can be spelled out: the ten-year rate is the average of the overnight rates the market expects over those ten years, plus the term premium. The central bank controls the starting point of that average and influences the expected path with its words. What it does not control is the credibility it is granted or the price of risk.

From this comes a conclusion that helps you read any meeting: the central bank does not set the price of money at long maturities, it sets the anchor and competes for the narrative. Its most powerful tool is not the 0.25% it announces, it is what it manages to make the market believe about the next three years.

Real case: July 2012, when a sentence cost less than a program

In July 2012 the euro area was coming apart. The Spanish ten-year bond went above 7% and its spread over the German bond widened past 600 basis points; the Italian was hovering around 6.5%. This was not an inflation problem: the market was putting a price on the possibility that Spain or Italy would leave the euro and pay in a different currency.

On July 26, Mario Draghi declared in London that the ECB would do whatever it takes to preserve the euro, and added that it would be enough. In September it was turned into a program of conditional sovereign bond purchases, with one decisive detail: it was never activated and never bought a single bond. Even so, spreads collapsed and the Spanish bond was back around 4% in 2013. More telling still: the ECB's balance sheet actually shrank during 2013, while the crisis receded.

The lesson is precisely the one in this module. The tool that worked was not a purchase, it was a credible expectation about what the central bank was prepared to do. The next module examines the other lever, the one that does involve buying for real.