(Roughly) Daily

As the recent election reminds us, inflation is a central issue to millions. How we calculate inflation has always been a subject of debate. And, as Carola Conches Binder explains, small changes that might seem trivial can lead to enormous changes in how well-off we think we are…

Every month, the US Bureau of Labor Statistics releases its newest data on the consumer price index (CPI). The CPI report is eagerly awaited by economists and policy wonks and investors. It garners heavy news coverage as a key piece of information in macroeconomic policymaking and analysis. The CPI and related measures affect monetary and fiscal policymaking and are often used to adjust Social Security payments, income tax brackets, and wages for millions of workers. Because of these far-reaching impacts, even relatively small changes in the measurement of the CPI can have major implications for households, firms, and the government’s budget. Thus, the technocratic task of measuring the price level is often at the center of political controversies. The evolution of inflation measurement in the United States has reflected both technical progress and these political forces.

The government’s role in the collection and publication of price indexes has been politically controversial from its origins, which were surprisingly late. Wesley Clair Mitchell, the former president of the American Economic Association, in 1921 called it:

a curious fact that men did not attempt to measure changes in the level of prices until after they had learned to measure such subtle things as the weight of the atmosphere, the velocity of sound, fluctuations of temperature, and the precession of the equi­noxes . . . Perhaps disinclination on the part of ‘natural philosophers’ to soil their hands with such vulgar subjects as the prices of provisions was partly responsible for the delay…

[Binder recounts the history of price measurement, starting in Italy in the 18th century, explaining that economic and political pressures first resisted having indices at all, then struggled to shape them. She then compares the current approaches in use and unpacks the recent [and current] debate over whether we have inflation and if so, how much…]

… At the time of writing in 2024, inflation is falling by nearly any measure. But as Krugman’s super core episode [see here, here, and here] illustrates, the past few years have intensified public scrutiny of official price indexes and led to debates about their interpretations. In light of this scrutiny, it is important for national statistics agencies to maintain their credibility by adopting methodological improvements, learning from both the private sector and academic researchers, and communicating clearly with the public.

Just as the Bureau of Labor Statistics responded to the Stigler and Boskin Commissions by revising its methods, it has also responded to the Covid-19 pandemic and post-­pandemic inflation. For example, the pandemic demonstrated that biennial (every other year) updates to the CPI expenditure weights are too infrequent in times of rapid economic changes. The pandemic very quickly shifted the types of goods and services that people were buying, so expenditure weights based on survey data from 2018 became out of date. People were spending more on food and other items facing large price increases, and less in categories experiencing falling prices, like transportation, implying that the official CPI measure was underestimating inflation.

The Bureau of Labor Statistics could not move quickly enough to change its estimates of expenditure weights, but private researchers could. The economist Alberto Cavallo used data collected from credit and debit card transactions to build his own set of weights that he used to construct a new Covid CPI measure, which indeed rose more quickly than the official CPI in the first months of the pandemic…

Cavallo’s experience constructing alternatives to official inflation statistics began when his home country, Argentina, began doctoring its inflation statistics in 2007 to hide inflation that rose above 12 percent in 2006 and likely averaged above 20 percent from 2007 to 2011. Cavallo and a group called the Billion Prices Project at MIT used web-scraping techniques to collect the prices of goods sold online in Argentina and four other Latin American countries. For all but Argentina, the price indexes based on online prices closely tracked official price indexes, but for Argentina, Cavallo’s estimates of inflation were three times higher than official estimates, and Cavallo’s estimates soon became more trusted than the official statistics.

Cavallo and the other researchers behind the Billion Prices Project have since extended their methodology to other countries, including the United States. In 2011, they started a private company called PriceStats that produces daily-­frequency inflation measures for central banks and financial-­sector customers in 25 countries, including the United States, using data on millions of product prices from hundreds of retailers.

In the United States, private inflation estimates may supplement the official estimates, but are unlikely to replace them. In part, this reflects the statistical agencies’ willingness to refine their methods, learn from private researchers, and maintain methodological transparency. For example, having learned that biennial expenditure weight updates are too infrequent, the BLS will update its expenditure weights every year beginning in 2023. The BLS also recently sponsored a study, Modernizing the Consumer Price Index for the 21st Century, to investigate additional improvements to the CPI that could be adopted in years to come. The study’s panelists considered a variety of innovations by Cavallo and other researchers, and recommended that the BLS experiment with using a wider variety of data sources, including online transactional data, to improve the timeliness and accuracy of its estimates.

The development of price and inflation measures has often been driven by political controversies, especially during times of war or during labor disputes. The development of the consumer price index arose from a need to ensure that wages and benefits would keep up with the cost of living. The recommendations of several different commissions have led to changes in how the index is computed – changes that have major impacts on the federal budget and on the distribution of resources. Especially in recent years, alternative inflation measures have proliferated. Overall, the official price indexes represent a tremendous intellectual and public achievement, despite the debates that continue to surround their use and interpretation…

Measuring price changes: “Where inflation comes from,” by @cconces in @WorksInProgMag.

(Image above: source)

* H. Jackson Brown Jr.

###

As we muse on measurement, we might note that today marks the anniversary of another measurement regime that supplanted what had been a largely an informal (and often intuitive) understanding of a basic fact of life: on this date in 1883, precisely at noon, North American railroads switched to a new standard time system for rail operations, which they called Standard Railway Time (SRT). Almost immediately after being implemented, many American cities enacted ordinances adopting the standard, thus resulting in the creation of time “zones” in the U.S.– Eastern, Central, Mountain, and Pacific. Though tailored to the railroad companies’ train schedules, the new system was quickly adopted nationwide, forestalling federal intervention in civil time for more than thirty years, until 1918, when daylight saving time was introduced.