The first point is that it's early days. We actually wouldn't expect to see much broadening this early into the oil price shock. We have measures of core inflation that take out volatile components. The two we've been focusing on are the median and the trim mean. When there's a big change in oil prices or gasoline prices, they get pulled out. If you saw core inflation start to go up, even when it's pulling out the direct effects of higher oil prices, that would be a sign of spreading. That would be one thing we'd look at.
Another thing we can look is at the distribution of price changes. In the monetary policy report, there's a chart of the percentage of components of the CPI that are rising faster than 3%. It has shown recently that it has actually been coming down. It's a little bit higher than historically, largely because food price inflation is high and there are a lot of food components in the CPI, but it has been coming down.
If you started to see that going up, that would suggest that not only are your oil prices going up but a lot of other prices are going up, too. That would be another kind of thing we would look at.
Another thing we look at closely is expectations about future inflation. Right now, people have filled up their gas tanks and they've gone to the grocery store, so they know that inflation has gone up recently. Their perception of inflation and their short-term inflation expectations have gone up, but they remain confident that inflation will come back down. If we started to see some erosion of that confidence and if those longer-term expectations were going up, that would be a real sign of worry that inflation was getting entrenched and that people were starting to believe inflation was going to stay.
There's no single thing that we focus on. We look at a range of indicators, but if we start to see it spreading or if we start to see the persistence going up, we will be starting to think about the need to raise interest rates.