How important are petrol prices for inflation and inflation expectations

Many goods and services make up the consumer bundle that is used to measure the Consumer Price Index (CPI) and inflation. However, some of those goods are weighted more heavily in the bundle, because they make up a higher proportion of the average consumer's spending. One example is petrol. 'Private transport supplies and services', which includes (among other things) petrol, makes up 7.27 percent of the consumer bundle (and this Stats NZ page also suggests that petrol is 3.5 percent of the CPI).

Petrol is also one of the most visible prices. Many of us would see the petrol price every day, on the large fuel price boards outside service stations. So, from the perspective of a consumer, a change in petrol prices will likely be an important contributor to how inflation feels, both because petrol is something that consumers spend a lot on, and because the petrol price is something that is very visible.

How important are petrol prices to inflation, and to consumers' inflation expectations? That is the question that this 2024 article by Puneet Vatsa (Lincoln University) and Gabriel Pino (Universidad Diego Portales), published in the journal Energy Economics (open access), attempts to answer. Vatsa and Pino use data from 1995 to 2023 on petrol prices (from the Ministry of Business, Innovation and Employment), CPI inflation (from Stats NZ), and one-year (and five-year) inflation expectations from a quarterly household survey run by the Reserve Bank of New Zealand. They analyse the data using structural vector autoregression (SVAR) models, which are basically just fancy time series models that try to estimate how different variables move together over time, and figure out which changes are likely to cause which effects.

Using this method, Vatsa and Pino find that:

...petrol price shocks explained 32.61 % of the variation in the quarterly headline inflation rate from 1995 Q1 through 2023 Q3; they explained 34.05 % of the variation in one-year inflation expectations over the same period.

So, even though petrol makes up less than four percent of the CPI, changes (shocks) in the petrol price explained nearly one-third of the change in the CPI over that 28-year period. Part of that will be that petrol prices change frequently, so a lot of the fluctuations in the CPI will be due to fluctuations in petrol prices. Another part will be that petrol prices affect transport costs (or rather, diesel prices, which are highly correlated with petrol prices, affect transport costs), which in turn affect the prices of lots of other goods and services. This latter effect is demonstrated by Vatsa and Pino, who show that core inflation (which excludes petrol) is also affected by petrol price shocks, but to a much smaller degree, explaining 7.79 percent of the change in core inflation.

Consumer expectations about future inflation are also significantly affected by petrol price shocks. However, Vatsa and Pino find that this is true only of expectations about inflation one year into the future, but not for five-year inflation expectations. They don't offer a good explanation of why the effect of petrol prices on consumers' inflation expectations is only for short-term expectations. However, it is likely that immediate price changes affect consumers short-term beliefs about inflation, whereas their long-term beliefs about inflation are anchored by long-term policy, such as the Reserve Bank's goal of maintaining low and stable inflation. Maybe.

Overall, this paper shows the importance of petrol prices for consumer inflation in New Zealand. So, if you want to know what is happening to inflation, and to inflation expectations, take a closer look each time you fill up your car (it is likely better than relying on the media).