I have historical data where the CPI and GDP are correlated. Does this make sense? If not, how do I test for it?

  • 1
    $\begingroup$ They could be correlated, but they don't have to be. What's the problem you're trying to solve here? $\endgroup$ – 410 gone Nov 16 '16 at 11:04
  • $\begingroup$ You should definitely plot your data because the correlation may be driven by large outliers or non-stationarities. It is not clear why absolute real GDP and CPI should be highly correlated unless you are working on a economy that is highly dependent on commodities. $\endgroup$ – Andrew M Nov 16 '16 at 11:47
  • 2
    $\begingroup$ @Dqingqong please could you edit in that extra information, into your question? $\endgroup$ – 410 gone Nov 16 '16 at 16:30

There was a degree of economic stability between the UK's exit from the Exchange Rate Mechanism in 1992 and the World Financial Crisis in 2008: positive real GDP growth typically between 2% and 4% and low steady inflation typically between 1% and 3%.

The Governor of the Bank of England called this "The Great Moderation".

The effect is what you have observed: the level of real GDP and the level of the CPI both grew, apparently highly correlated with each other, as shown in the red line below based on quarterly data.

But this correlation was an artefact of each indicator growing itself over time and of auto-correlation, as well as not having a recession in that period or sudden spikes in prices; looking from 1991-2009 would have produced a less smooth chart.

If instead you look at the percentage changes over the previous 12 months, the apparent linearity collapses, as shown in the blue line below.

enter image description here

  • $\begingroup$ This seems to be nominal, not real GDP. $\endgroup$ – Giskard Nov 16 '16 at 15:00
  • 1
    $\begingroup$ @denesp - I do not know what makes you think that. It is the ONS time series ABMI while nominal GDP (YBHA) typically grew by 3%-7% a year over the same period $\endgroup$ – Henry Nov 16 '16 at 15:06
  • $\begingroup$ My mistake. The growth seems to large and sustained to be real. Having looked it up, I now see this was indeed the case. Thank you for providing the source though. $\endgroup$ – Giskard Nov 16 '16 at 15:38

The CPI encompasses the price of consumed goods (some of them are domestically produced, other are imported). The GDP deflator takes into account only the goods produced domestically (some of them are consumed here, other are exported). Thus there is some correlation between the GDP deflator and the CPI as you can see below

Inflation (my own plot, the series are from FRED database (GDPDEF and DPCERD3Q086SBEA), seasonnaly adjusted, growth rate between the same quarter from the previous year)

Having this in mind, recall that the real GDP is calculated as the nominal GDP divided by the GDP deflator. Thus the price of the domestic goods procuded and consumed in the country are in both CPI and real GDP. Here comes your correlation.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.