Unemployment and inflation range in opposite directions. Higher unemployment is the 'cost' of lower inflation.

-> while directly creating jobs may be difficult for a government, the central bank can engage in expansionary monetary policy, increasing the rate of inflation and, via the Philips relationship, achieve a lower rate of unemployment.

• Adaptive expectations: people use data and policy actions from the past. When a new policy is introduced, beliefs do not change quickly, but rather slowly adapt as new data becomes available. One problem is that it provides considerable scope for people to be systematically fooled by policy makers.
  
• Rational expectations: people are feared looking and know how the world works. It is hard to have much faith in the predications of a theory that is based on people being systematically tricked, so rational expectations are widely assumed in theoretical work. However, models with adaptive expectations perform well empirically.

• Hitting the target at all times is impossible in practice, an inflation target of 0 per cent increases the risk that the economy will experience periods of deflation.
  
• Some evidence suggests that the conventional measures of inflation tend to overstate the true rate of inflation by about one per cent.
• Some economists believe that a small amount of inflation is necessary to 'grease' our economic engine.