Why do people criticise CPI as a measure of inflation?

Are you sick of reading inflation articles yet? Perhaps. I’m nearly bored of writing them, anyway.

The CPI, for the uninitiated, is the Consumer Price Index, which is the most commonly used metric in measuring inflation. The monthly release has been drawing a lot of eyeballs recently for obvious reasons, and is seen as the “official” inflation rate. It was 7.1% in the US for November, the most recent reading. Which is curious, you know, when heating your home hurts your wallet, a Magnum ice cream is $5 and a jar of real (100% nut) peanut butter is $10. That feels like a lot more than 7.1%, don’t you think?

Are you looking for fast-news, hot-tips and market analysis?

Sign-up for the Invezz newsletter, today.

Well, about that. There are reasons that the CPI may be wrong.

Fantasy Football and CPI

Last year in Fantasy Premier League – which is a virtual fantasy football game where you pick a team for each of the 38 weeks of the season – I finished 74K out of 9.1 million players.

Officially, that means I am in the top 0.8% of players worldwide. It sounds like an incredible achievement and that I have a truly rare gift. Unfortunately, it’s completely untrue. I get far too emotional about my players, with logic often going out the window in a bid to catch my more-gifted friends (although last year I actually won our league for the first time ever, possibly the greatest achievement of my life so far).  

In reality, that ranking is entirely unrepresentative – exactly like the CPI. With the football season 38 weeks long, the vast majority of players give up after a few weeks. Some even enter late – meaning they have less weeks to accumulate points. By virtue of simply playing for all 38 weeks, you are essentially guaranteed a place in the top 15%.

So with that extremely tangential point made (and the flex of my league title complete), what is it that makes the CPI number so unrepresentative?

CPI Shortcomings

The CPI is calculated by assessing the price of a basket of goods and services from one period to another. One big problem is that the method assesses the eroding purchasing power of one dollar, instead of one person. High-income individuals spend more, meaning they bear a greater weight on the index. In other words, each dollar casts a vote, instead of each person.

This is why the inflation rate is often way off the reality of what poorer, or even middle-of-the-road families, are experiencing. Whereas the average Joe and Jane may be spending 50% more on homebrand fusilli, a Wall Street finance-bro earning $300,000 per year has only seen a 2% increase on the price of their Patagonia vest. It’s a bit of a stretched example – but you see my point.

Lower-income groups spend a greater portion of their budget on food and energy. These rising more than other goods increases the hit to poorer individuals. Similarly, they have less options for budgeting. To stick with the fusilli example, a 2kg bag may be cheaper than a 250g gram on a per gram basis, but the lower income group may be forced to pay more per gram as a result of the inflation than the higher group – causing a knock-on effect on the CPI reporting.

There are other shortcomings, too. The CPI technically doesn’t measure the difference in prices, but rather it is a cost-of-living index. This means that if prices rise and consumers switch products as a result, the CPI may be biased and not report the rising prices. It is not limited to substitution either – changes in the quality of goods can also impact the accuracy of the CPI.

Why not just…fix it?

While the above shortcomings are relatively straightforward, we are wading into shakier ground by answering why exactly these errors persist. Certainly, in terms of the per-dollar measurement rather than the per-person measurement, this is an adjustment that could be examined.

The conspiracy theorists point towards the fact that governments want lower inflation numbers. High inflation is bad, to make an obvious point. And bad news makes people unhappy, and that doesn’t enamour politicians to voters. But high numbers also kick up inflation expectations, which makes the problem worse, because inflation expectations fuel further inflation.

Whether conscious or not, it is definitely true that elected officials have every incentive to produce as low an inflation statistic as possible. Inflation being lower will facilitate keeping interest rates low, which reduces the government cost of borrowing and allows a load of money to be spent without taxes being increased. Las time I check raising taxes was not a popular decision.

Low interest rates facilitates debt increasing, which will then allow money printer to go brrrrrr, and the next person can deal with the fallout (or repeat the same thing again).

Asking government to report inflation is like asking mafia to report on crime

Peter Schiff

The slightly less cynical answer is that it’s inertia, or an unwillingness to change the system. Although this point does weaken slightly when the changes over the years are considered – the measurement has been adjusted several times already.

Whether it’s cynical or not, the CPI is a plutocratic statistic – i.e. by/for the wealthy. In a cruel twist of irony, it thus matches inflation itself which hurts the poorer brackets significantly more than the wealthy.

Then again, no metric is perfect, and while some may say it understates the true inflation rate, it does give a reasonable picture as to what the situation is, even if by virtue of it being consistent for relative comparisons. But it is important to note that like any statistic, context is important and it is far from perfect.