What is CPI and how is it Calculated?
Common sense tells us the Consumer Price Index is not an adequate measure of inflation. For the second year in a row the Consumer Price Index for All Urban Consumers (CPI-U) remained under 2 percent. On average, consumer prices increased 1.5 percent, according to the government. However, the government has incentives to keep this statistic as low as possible. In fact, the CPI doesn’t even measure inflation, rather a range of consumer spending behaviors. The CPI is perhaps one of the most important government statistics because it affects a number of public programs and is used as a benchmark to set public policy. But it’s accuracy is questionable, especially when compared with other agency’s inflation measures.
Why does the government want low inflation numbers?
The CPI is tied to the incomes of about 80 million Americans, specifically: Social Security beneficiaries, food stamp recipients, military and federal Civil Service retirees and survivors, and children on school lunch programs. The higher the CPI, the more money the government needs to spend on these income payments to keep pace with the cost of living. However, this same government is about $17 trillion in debt. If the CPI is low, the less money the government needs to spend on cost of living adjustments, something seniors are astutely aware of.
The government has a few resources at its disposal to manipulate the CPI. First, the Bureau of Labor Statistics operates under a veil of secrecy. The raw data used to calculate the CPI is not available to the public. When I asked why, I was told “so companies can’t compare prices.” This makes very little sense because companies can easily compare prices with data openly available on the internet. It also makes it impossible to audit their findings. Additionally, over the past 30 years, the government has changed the way it calculates inflation more than 20 times. These ‘methodological improvements’ to the CPI are said to give a more accurate measure of consumer prices. However, these changes could also be a convenient way to include or exclude certain products that give favorably low results, but there’s no way to know, given the lack of transparency.
So how is the CPI calculated?
Although this is not a transparent process, the BLS gives some insight on their website as to how it calculates the CPI. The economic assistants track about 80,000 consumer products each month known as the Market Basket of Goods. However,
“If the selected item is no longer available, or if there have been changes in the quality or quantity (for example, eggs sold in packages of ten when they previously were sold by the dozen) of the good or service since the last time prices were collected, the economic assistant selects a new item or records the quality change in the current item.”
This data is then plugged into a formula along with other factors including census information and consumer spending patterns. In other words, the CPI doesn’t measure changes in consumer prices, rather it measures the cost-of-living. Further, the government makes the assumption that consumer spending habits change as economic conditions change, including rising prices. So if prices rise and consumers substitute products, the CPI formula could hold a bias that doesn’t report rising prices. Not a very accurate way to measure inflation.
The CPI doesn’t even meet the government’s definition of inflation
The Bureau of Labor Statistics defines inflation “as a process of continuously rising prices or equivalently, of a continuously falling value of money.”
As I outlined above, the CPI is not a measurement of rising prices, rather it tracks consumer spending patterns that change as prices change. The CPI doesn’t even touch the falling value of money. If it did the CPI would look much different.
The CPI doesn’t meet other government agency’s inflation measurements either
The most obvious is the Federal Reserve’s measure of monetary inflation. M2 measures the supply of US dollars, which includes cash, checking deposits, saving deposits, and money market mutual funds. The more money that’s created and put into circulation, the less valuable it becomes. And the Fed has created a lot of money recently. The Fed’s unprecedented bond buying program, Quantitative Easing, created $116 million an hour for the entire year last year. It doesn’t make sense that the BLS’s measurement of inflation was only 1.5% last year, while at the same time, monetary inflation grew 4.9%.*
Another example where the BLS doesn’t meet other agency’s inflation measurements is the U.S. Department of Agriculture. According to the BLS the average price of beef and veal increased 20 percent over the past five years. However, according to the USDA, beef prices have increased 26 percent over the past five years. I asked a statistician at the BLS about this discrepancy and he said “I would expect those numbers to be a little closer together.” When even the federal government gets different numbers on the same products, how could this possibly be an accurate measurement of inflation?
It’s important to have an accurate measure of inflation because consumers, especially those on fixed incomes, are negatively impacted by rising prices. Also, the federal government and the Federal Reserve use CPI trends to help set tax, monetary and fiscal policies, which affects economic growth. Given that the CPI is calculated secretly, is no where near comparable to monetary inflation, and doesn’t even meet its own definition of inflation, we should use our common sense to count the value of our cents, not the CPI.