Lesson preview · Credibility And Policy
Credibility, Forward Guidance & the Lucas Critique
~12 min · Free to read
A new central banker takes office and promises to keep inflation low — say 2% a year. Workers and firms believe her, so they sign wage contracts assuming prices will rise only 2%. Now she faces a tempting choice. If she lets inflation run a bit higher than promised, real wages dip, hiring gets cheaper, and unemployment falls for a while. The promise was sincere yesterday. Today, breaking it looks like free money.
That temptation is the whole story of this lesson. A promise that was the best plan yesterday stops being the best plan once everyone has acted on it. We will see why this drags inflation up for no lasting gain in jobs, how central banks fight back by making themselves believable, and why a famous economist warned that policy changes change behaviour.
First, a quick recall from the last two lessons. In the short run a central bank can trade a little more inflation for a little less unemployment — the short-run Phillips trade-off. But that trade-off vanishes in the long run. Unemployment returns to its natural rate (the rate the economy settles at once wages and prices have fully adjusted, here 5%) no matter what inflation is. The long-run Phillips curve is a vertical line at that natural rate. Inflation can be high or low; unemployment ends up at 5% either way.
Key Term
Time inconsistency
Time inconsistency is when a plan that is best today stops being best once people have acted on it.
The banker’s 2% promise is the textbook case. Before wages are set, promising 2% is the right plan — it anchors expectations cheaply. But after contracts lock in at 2%, she is tempted to surprise everyone with higher inflation to shave unemployment below 5% for a while. The plan reverses on itself. What was optimal beforehand is no longer optimal once people have committed.
The catch: this only looks like free money if the public is fooled. They will not be fooled twice.
Here is why the public refuses to be fooled. They have watched this game before. They know the banker is tempted to break her word, so they stop believing the 2% promise. Instead they expect the inflation she is actually tempted to deliver — say 5% — and they bake that 5% straight into their wage demands.
Now the banker is boxed in. Expected inflation is already 5%. If she delivers only 2%, real wages turn out higher than firms planned, hiring slows, and unemployment rises above 5% — a recession she will not accept. So she validates the 5% everyone expected. Inflation lands at 5%, unemployment sits right back at its natural rate of 5%, and she got nothing for the extra inflation.
Key Term
Inflation bias
Inflation bias is the extra inflation a non-credible central bank ends up with — over and above what it actually wanted — for no lasting fall in unemployment.
The sequence: the bank is tempted to surprise-inflate, the public foresees the temptation, the public pre-sets high expected inflation, and the bank then has to deliver that high inflation just to avoid a recession. Both the credible and the discretionary outcomes sit at the natural rate of unemployment. The only difference is the price tag in inflation.
This is the Kydland–Prescott result (1977), later sharpened by Barro and Gordon (1983). It won Kydland and Prescott the Nobel Prize. The headline lesson: the ability to make a binding promise is itself valuable, because not being able to promise costs you permanent excess inflation.
Tip
So what
Without credibility you pay an inflation bias — here 3 extra percentage points of inflation — and get exactly nothing for it, because unemployment stays glued to its natural rate of 5%.
Worked Example
A central bank’s true goal is 2% inflation, and the natural rate of unemployment is 5%. Compare two regimes. Regime A: the bank can credibly commit, and the public believes it. Regime B: the bank uses discretion (it is free to break its promise), and the public knows it. (a) What inflation and unemployment result under each regime? (b) What is the inflation bias, and what did the economy buy with it?
So how does a bank escape the trap? By tying its own hands. If the bank cannot easily break its promise, the public believes the promise, and the inflation bias disappears. Countries do this with an independent central bank (politicians cannot order it to inflate before an election), a published inflation target it is judged against, and a reputation it has spent years building. Economists sum this up as rules over discretion: a bank bound by a rule it cannot bend is more believable, and therefore better off, than a bank free to do whatever looks best in the moment.
Key Term
Forward guidance
Forward guidance is when the central bank steers expectations by telling people, in advance, what it plans to do.
Expectations are doing the heavy lifting in this whole lesson, so the bank tries to shape them directly. It might announce: “We will keep rates near zero until unemployment falls below 4%,” or “We will not raise rates before 2026.” If people believe the announcement, they spend and invest today on the strength of cheap borrowing they expect tomorrow.
Forward guidance only works if the bank is credible. A promise from a bank with a reputation for keeping its word moves expectations; the same words from a bank that flip-flops move nothing. Guidance is credibility put to work.
Key Term
The Lucas critique
The Lucas critique says you cannot predict the effect of a new policy from old data, because people change their behaviour the moment the policy changes.
Robert Lucas pointed out a deep flaw in naive forecasting. Suppose past data shows that whenever the central bank raised inflation, unemployment fell. A planner concludes: “So I can buy lower unemployment by inflating.” But the old relationship only held because people were surprised by the inflation. Announce a permanent inflation policy and people will see it coming, build it into their wage demands, and the old link breaks. The relationship the planner was relying on dissolves precisely because they tried to exploit it.
Concrete example: in the 1960s, policymakers read a stable inflation-unemployment trade-off off the data and tried to ride it for lower unemployment. By the 1970s the public had wised up, expected the inflation, and the economy got high inflation and high unemployment at once. The data-driven trade-off had been an illusion. That is exactly the inflation bias and time-inconsistency story playing out in history.
Check yourself · no marks
A finance minister says: “Our records prove that surprise inflation reliably cuts unemployment. So let’s commit to permanently higher inflation — it will keep unemployment low forever.” Using the Lucas critique, explain why this plan backfires.
Commit to one first. Guessing and being wrong beats reading the answer cold.
Trap · Common Misconception
"A central bank should keep its options open — discretion beats tying its hands"
It sounds obviously right: more freedom must be better, so a bank that can do whatever looks best in the moment should outperform one bound by a rule. The time-inconsistency result flips this on its head.
A bank free to break its promise (discretion) is worse off, not better. The public foresees that it will be tempted to inflate, so they pre-set high expected inflation. The bank then has to validate that high inflation to avoid a recession. It ends up with the inflation bias — permanently higher inflation and the same unemployment as a bank that tied its hands. The bank that cannot renege is believed, gets low inflation, and reaches a better outcome. Here, binding yourself is a strength: it is the only way to be credible, and credibility is what keeps inflation low at no cost in jobs.
Practice · 1 / 4
What does it mean that a central bank’s low-inflation promise is “time inconsistent”?
Next
Like what you read?
Sign up free to save your progress, get spaced reviews tuned to you, and unlock all 50 microeconomics knowledge points with interactive graphs and adaptive practice.