MIT vs. ShadowStats – Who’s Right on Inflation?

Few financial topics elicit as strong an emotional reaction as inflation. That's probably due to a number of factors including its "hidden tax" nature as well as the seemingly ambiguous process of calculating this illusive figure. If you don't properly factor inflation and inflation expectations into your investments and planning, you could miss the mark by a wide margin. But what inflation measurement should you use?

Do you go with the Consumer Price Index? Personal Consumption Expenditures? The figures reported by Shadowstats? The spread between similar maturity Treasuries and Treasury Inflation Protected Securities (TIPS)?

Each option has its drawbacks. The CPI and PCE include nebulous processes like substitution, weighting and hedonics. Shadowstats uses outdated methodologies, which, in my opinion, are rather poor as they don't reflect or capture changes in the evolution of our economy. And the spread between Treasuries and TIPS only provides us an idea of what the market expectations of inflation are, not what the real level of inflation is.

I'm frequently asked what I think the real level of inflation is. I always answer the question truthfully with, "I don't know" before explaining further. I don't believe in the figures Shadowstats puts out, because portions of the old methodologies are not appropriate and result in overstated inflation. So I tend to use this as an "upper bound." I also understand the distortions that may be at play in the CPI and PCE, so I tend to use these as the lower bound. Somewhere in that vast middle ground lies the true inflation level; at least that's my opinion.

There is another group that is taking a stab at measuring inflation using a modern, big-data approach. It's called the Billion Prices Project. I first heard about this from my buddy Cris Sheridan, Senior Editor at Financial Sense (financialsense.com). Cris had the opportunity to interview MIT's Roberto Rigobon, one of the two founders of the Billion Prices Project. The intriguing interview can be found on their website.

[Hear: MIT’s Roberto Rigobon on the True Rate of Inflation]

The Billion Prices Project was founded by MIT economists Alberto Cavallo and Roberto Rigobon. The goal is to use technology and real-time tracking methods to assess price level fluctuations, rather than rely on physical price-checking and a lot of complex statistics. Cavallo initially created the BPP to provide alternative inflation readings in Argentina, in an attempt to publicize the widespread manipulation of data. Since then, the BPP has expanded to track inflation and other economic statistics across the globe.

Rigobon, the other founder, is actually a member of the National Bureau of Economic Research (NBER) and the Census Bureau, allowing him unique insight into the limitations and drawbacks of the current, outdated approaches to measurement. The goal of the Billion Prices Project is to use real-time tracking methods and the plethora of price data available online from retailers around the globe, to conduct economic research. By scouring high-frequency price data to the tune of more than a billion prices per week, the BPP is able to provide another view of inflation.

So what exactly does it show?

You may be pleasantly surprised, or downright shocked, to hear that the BPP measures inflation to be relatively in-line with the Consumer Price Index. Conspiracy theorists out there may be quick to assume that the BPP is "in bed" with the other government agencies who report inflation, but Rigobon has been outspoken about the antiquated collection and measurement processes currently in use. Tracking changes across billions of products has not only allowed the BPP to assess inflation trends across the globe, it has allowed them to pinpoint which countries are providing accurate official readings and which are not. During his interview with Cris from Financial Sense, he said that the UK, US, and Netherlands show the most accurate official readings, while places like Russia, Argentina and Venezuela showed the least.

Here's a chart of annual inflation, comparing the BPP index to the CPI. As you can see there are minor discrepancies but overall the two series track each other quite well. The BPP is showing annual inflation tracking just above 2%, as of the end of August.

The relative consistency between the BPP and the CPI may provide another advantage. As we know, half the battle for profits simply entails being on the right side of the Fed. We also know what data the Fed looks at and how it will generally respond under various scenarios. For example, if inflation pressures remain subdued, the upward glide path of rates will likely be much more gradual than if inflation pressures are a concern.

This means that CPI and PCE data, when released, are critically important to assessing future actions of the Fed. Wouldn't it be nice to have an advanced reading of where the CPI and PCE are heading? It's possible that the BPP could provide some indication. The real-time nature of the data allows for a real-time assessment of inflation trends.

[See Also: A Long-Term Look at Inflation]

Below we can see an up-to-date assessment of inflation. In the chart below, "PriceStats" (the red line) refers to the same BPP index we've been discussing (PriceStats is the data collection company). You can see that the PriceStats data does a relatively good job at anticipating changes in the CPI. Most recently, we can see that prices may have rebounded a bit during the last month, indicating that we may not see another decline in the (monthly) CPI. The current US monthly inflation rate, as measured by the BPP and PriceStats, is 0.147%. This equates to an annual inflation rate of about 1.76%.

Inflation will never be an exact science but a variety of measurement approaches by different organizations are all pointing to similar levels of inflation. The CPI, PCE, spread between Treasuries and TIPS, and now the BPP, all point to roughly 2% or sub 2% inflation. Just remember that 2% compounded inflation cuts the dollar's value in half every 35 years.

There is a good chance that economic data tracking will continue to trend towards real-time information. Even policy makers understand how beneficial this could be. As an example, when Lehman Brothers collapsed in 2008, the BPP showed an almost immediate drop in aggregate prices. Businesses quickly responded to falling demand, but the CPI didn't pick this up until two months later. More timely information could give policymakers critical advance warning when storms are on the horizon.

The above content was an excerpt of Richard Russell's Dow Theory Letters. To receive their daily updates and research, click here to subscribe.

About the Author

Chief Investment Strategist
matt [at] modelinvesting [dot] com ()
randomness