Many people have constructed alternative inflation measures to correct what they consider flawed and/or manipulated government statistics. Shadow Stats is a well-known example. Publisher John Williams believes the 1990s introduction of hedonic adjustments artificially depressed CPI. He also takes issue with other changes to how CPI is calculated.
The problem is that using the 1990 methodology shows inflation to be almost 4% higher than the CPI has it today, and using 1980s methodology shows it to be better than 7% higher.
I have trouble believing those levels are real. Long-term bond investors tend to be fairly sophisticated. Those I know wouldn’t buy 30-year bonds at 3% if inflation were really running 5–6% a year. The Federal Reserve can greatly influence shorter maturities, but the longer end of the yield curve is still market driven.
Another alternative is MIT’s Billion Price Project. They measure price inflation in 22 world economies by gathering prices online every day. This has the potential to be much more precise than CPI. Over time, however, it has actually tracked CPI closely.
See related podcast interview: MIT's Roberto Rigobon on the True Rate of Inflation
The Shadow Stats’ alternative CPI also highly correlates to official CPI, perhaps because it is simply the same number with an upward boost to remove the hedonic and other adjustments Williams believes are unnecessary or bogus.
I respect what Shadow Stats tries to do, but I disagree with it. Economic changes have been affecting statistics for many decades. The transition from an industrial economy to an information-based economy sped up the pace. The quality of electronic devices, among other technological marvels, improved far faster than the quality of eggs or lawnmowers.
Hedonic adjustment tries to account for this disparity. Is it perfect? No, but pretending the economy hasn’t changed is not a good option either.
Related: Interview with Diane Coyle on her book, GDP: A Brief but Affectionate History
We all need to let go of the fantasy that economic statistics are perfect. They’re not. Each monthly CPI number is a snapshot of a moving target. It’s blurry, and you don’t see the other activity that occurred outside the frame. The quality of the picture depends on your camera, lighting conditions, and many other factors.
None of this means the photographers are dishonest and showing us fakes. More likely, it means we expect too much from our stats. Despite all their limitations, those numbers can help us if we use them properly.
Truly sincere economists and statisticians debate the very arcane topics surrounding inflation. They really want to get it right. In reality, the numbers we get are the best they can do. The way we calculate inflation changes all the time.
Knowing that our best economists make it up as they go along is not comforting. We would like to believe that somewhere a little man is sitting behind a curtain, pulling levers to make the economy move. If that were true, we could understand the economy by understanding the little man.
Sadly, there is no one behind the curtain. There’s not even a curtain. There is a planet full of people with different resources and desires. The myriad ways they interact add up to a global economy.
To read more on this topic, see John Mauldin’s Thoughts from the Frontline letter, originally published Nov. 1, 2015. You can read the full issue here.
