ShadowStats data is an alternative to the official inflation data from the U.S. Bureau of Labor Statistics. In general, ShadowStats says that actual inflation is much higher than the government's official statistics. Although it doesn't match my personal experience, some people want an alternative to the U.S. government data. ShadowStats data is proprietary and copyrighted. Although ShadowStats has granted me permission to use some of its data in my Inflation Calculator, my program is not permitted to display the data. For information about ShadowStats, see www.shadowstats.com. (Note: ShadowStats appears to be inactive since mid-2023; the latest data is from 2022.)
What does the "Show Inflation Data" button do?
You can see inflation rate percentages by clicking the Show Inflation Data button. Note that this button is disabled (grayed out) until after you have performed an inflation calculation. If you click the button after it's enabled but the Inflation Data window still doesn't appear, your web browser is probably blocking pop-up windows to avoid unwanted advertisements. You can change your browser's settings to temporarily enable pop-ups.
When the Inflation Data window opens, it will display all the inflation percentages for your selected data set. If the inflation percentage is a negative number, it indicates deflation for that year. The window also displays the cumulative inflation rate for the range of years you selected. However, ShadowStats inflation data is proprietary and cannot be displayed by my program.
Back to frequently asked questions
What is the "cumulative inflation rate"?
The window that appears after clicking the Show Rates button also displays the cumulative inflation rate for the range of years selected. This is the total amount of inflation for that period. (Thanks to Cy Coleman for suggesting this feature.)
Note that cumulative inflation is derived iteratively, not merely by summing the inflation percentages for each year. For example, assume we have two successive years with 10% inflation per year. Something that costs $1.00 before inflation would cost $1.10 after the first year:
1.00 x 0.10 + 1.00 = 1.10
After the second year of 10% inflation, it would cost $1.21:
1.10 x 0.10 + 1.10 = 1.21
So the cumulative inflation rate for those two years would be 21%, not 20%, because 1.21 is 21% greater than 1.00.
Back to frequently asked questions
Why don't the calculator's inflation rates match the rates I've seen elsewhere?
There are many ways to measure inflation. When Tom's Inflation Calculator first runs, it defaults to the most common measure: Consumer Price Index (CPI), Annual Average. This is the Consumer Price Index for All Urban Consumers (CPI-U). It is the mean average of monthly inflation rates during a given year.
Another common measure is the December-to-December inflation rate. It compares prices from December 31 of one year to prices from December 31 of the previous year. Over many years, it varies little from the CPI-U Annual Average rate.
Some news stories quote the "core inflation rate," which is the Consumer Price Index (CPI-U) excluding food and energy inflation. Food and energy prices tend to be more volatile than other prices, so the core rate excludes those swings. But they are included in the CPI and PCEPI numbers that my programs use.
Remember, you needn't use the default CPI-U data; you can choose any of these six data sets.
Back to frequently asked questions
What is the calculator's source of data?
My program uses inflation data from several sources, including the Economic History Association (data no longer public), the Federal Reserve Bank of Minneapolis, the U.S. Bureau of Labor Statistics, the Congressional Budget Office, and the Social Security Administration. The Personal Consumption Expenditures Price Index (PCEPI) is compiled by the U.S. Bureau of Economic Analysis and is available from the Federal Reserve Bank of St. Louis. This data is the year-over-year percent change as measured each January 1 for the previous year, seasonally adjusted.
The most widely quoted inflation rate is the U.S. government's Consumer Price Index (CPI), 1913 to present, as compiled by the U.S. Bureau of Labor Statistics. By default, my Inflation Calculator uses the CPI-U (CPI All Urban Consumers) data, not seasonally adjusted (data series CUUR0000SA0). The CPI-U includes medical-care inflation since 1935 (data series CUUR0000SAM). The CPI-U December-to-December retail inflation is also from the U.S. Bureau of Labor Statistics and extends from 1913 to present (data series CUUR0000SA0).
Data before 1913 was reconstructed by historians and economists from old records and is less reliable. For inflation rates from 1801 to 1912, the Federal Reserve Bank of Minneapolis cites three sources: "Index of Prices Paid By Vermont Farmers for Family Living" (18001851), "Consumer Price Index by Ethel D. Hoover" (18511890), and "Cost of Living Index by Albert Rees" (18901912).
The ShadowStats retail-inflation data is from a private company, not the U.S. government.
Inflation rates for 2025 through 2034 are forecasts from the Congressional Budget Office, a nonpartisan source. Inflation rates for years beyond 2034 are based on the mean average inflation rate (3.26 percent per year) since the U.S. government began collecting these statistics in 1913.
Back to frequently asked questions
What if I don't trust the government's official inflation data?
Try using the ShadowStats alternative data set. This data is from a private source that believes the U.S. government consistently understates actual inflation.
Some people believe the U.S. government deliberately distorts its inflation data for various reasons. For instance, some Social Security retirees believe they are cheated out of larger cost-of-living adjustments because the government understates the Consumer Price Index (CPI), which is the basis for adjusting Social Security benefits. Other people assert that the government overstates the CPI. People who occasionally write to me on this subject never provide evidence to support their claims.
To address at least some of their concerns, Tom's Inflation Calculator now includes an alternative data set from ShadowStats, a private company. For more information about this data, see ShadowStats.com. My thanks to ShadowStats author John Williams for permitting me to use his data as an alternative in my Inflation Calculator. (Note: ShadowStats appears to be inactive since mid-2023; the latest data is from 2022.)
Back to frequently asked questions
Why can't the calculator adjust inflation for different months during the same year?
Tom's Inflation Calculator uses annual inflation data, not monthly inflation data. Adding monthly data would make the program much larger (taking longer to download) and would not be useful for calculations spanning several years. Also, I don't want to bother making updates every month!
If you want to make monthly calculations within a year, try using the U.S. Bureau of Labor Statistics inflation calculator. It uses the same CPI-U (annual average) data that my Inflation Calculator uses by default, but it includes the latest monthly inflation data.
For my calculator, I update the data only once a year (in January), when the data is more settled. (The government sometimes revises its monthly inflation data.) When converting between other years, rounding errors during calculations may cause insignificant differences between the answers from my calculator and the BLS calculator.
Note that the BLS calculator covers only the years 1913 to present. Unlike my calculator, it doesn't include historical data going back to 1665 or forecasts to 2100. Nor does it have alternative data sets.
Back to frequently asked questions
Why doesn't the calculator include the starting year when making calculations?
The target year's inflation rate is the inflation since the previous year. For instance, if you calculate inflation from 2016 to 2017, the program uses the 2017 inflation rate. If you calculate inflation from 1990 to 2017, the program uses the inflation rates for 1991 to 2017.
This program uses only annual inflation data, not monthly data, so it has a worst-case error of 12 months and an average error of 6 months. In the worst case, it will overstate the actual inflation by 12 months if your starting date is December 31 and your target date is January 1. To the program, a year has passed, but a year's worth of inflation doesn't happen overnight.
Back to frequently asked questions
Which inflation index does the Federal Reserve use to guide its monetary policy?
The Federal Reserve doesn't use the popular Consumer Price Index (CPI) to measure inflation against its ideal 2.0% target. Instead, the Fed uses the Personal Consumption Expenditures Price Index (PCEPI), a lesser-known measure. The PCEPI is compiled by the U.S. Bureau of Economic Analysis under the Department of Commerce. The CPI is compiled by the U.S. Bureau of Labor Statistics under the Department of Labor.
When you hear that the Fed's inflation-rate target is 2.0%, the PCEPI is the Fed's official yardstick not the CPI commonly quoted in news reports. Often, the PCEPI is lower than the CPI. That's important, because the Fed hesitates to raise its key interbank interest rate unless inflation exceeds the 2.0% target.
The PCEPI may trend lower than the CPI partly because it surveys the goods and services that people are actually buying instead of the constant "market basket" of goods and services that the CPI measures. Also, the PCEPI assigns a lower weight to housing costs. Historically, however, both the PCEPI and CPI have nearly the same average inflation rate: about 3.3%.
Because of its importance, I added the PCEPI data to my program in 2016 and revised it in 2024 to match the data set that the Federal Reserve cites. It's the annual year-over-year percent change, end of period, seasonally adjusted.
Back to frequently asked questions
Why is 2.0% the Federal Reserve's ideal inflation target? Why not zero percent?
The Federal Reserve's 2.0% target is arbitrary but not random. Because deflation is worse than inflation, the Fed aims a little high. Aiming for 2.0% allows some margin for missing on the low side. If the target were zero percent, missing on the low side could trigger a deflationary spiral.
Although falling prices may seem attractive, wages must fall, too, for businesses to stay profitable. But fixed-rate loan payments remain the same. As incomes decline, those payments soon become unaffordable. As more people and businesses are unable to repay their loans, the lenders force them into bankruptcies and foreclosures.
I have downloaded and analyzed the same inflation data that the Fed uses. The Fed hits its 2.0% target (plus or minus 1.0%) only about half of the time, and it hits the target (plus or minus 0.5%) only about one-third of the time. So missing the target is historically normal, and missing it by a small amount is no cause for alarm.
Back to frequently asked questions
Was U.S. inflation in 20212024 historically high?
Not even close. Even as recently as the 1970s and early 1980s, U.S. inflation was higher. Whereas recent inflation peaked at 8.0% in 2022, it reached 13.5% in 1980. In 1947, shortly after World War II, it was 14.4%. In 1918, the last year of World War I, inflation reached 18.0%. In 1864, during the Civil War, it hit 27.0%. And in 1778, during the Revolutionary War, it peaked at 29.7%.
If recent inflation seems unusually high, perhaps the Great Recession raised our expectations. After the Crash of 2008 (the low point of the Great Recession), the U.S. enjoyed 12 years of below-average inflation. Everybody grew accustomed to it, but nobody remembers it. Without that crash, if inflation had continued normally, today's prices would actually be higher. So we're still ahead of the game, even if we don't feel like it.
One problem is that today's workforce has less protection from inflation than in the past. In the 1970s, about 25% of American workers belonged to a labor union, and many of their union contracts included a cost-of-living adjustment (COLA) that automatically raised wages in step with inflation. Today, only about 10% of American workers are unionized, and their contracts are less likely to have a COLA. Nonunion workers are even less likely to receive raises that match inflation. This problem especially afflicts the private sector: only 6% of those workers are unionized, versus 33% in the public sector.
Ironically, the largest group of people who have COLAs today are retired folks despite the myth that they subsist on "fixed incomes." Each year, the federal government adjusts Social Security benefits for inflation according to a formula based on the Consumer Price Index (CPI). More than 72.5 million people who receive Social Security and/or Supplemental Security Income (SSI) benefits are guaranteed the annual COLA. Working people are more likely to be on fixed incomes!
Back to frequently asked questions
Is inflation always bad for everyone?
Nope! Individuals, businesses, and governments can actually profit from inflation, even if they don't realize it. Stock market investors don't complain about share price inflation, because they can sell their higher-priced shares for a profit. Likewise, inflated home prices discourage buyers but encourage owners, who can sell their property for a profit, too.
Another often-overlooked advantage of inflation is that borrowers can repay their loans with cheaper inflated dollars than they borrowed a huge advantage for long-term borrowers, such as homeowners with a 15- or 30-year mortgage. Businesses and governments can repay their debts with cheaper dollars, too.
Smart investors can buy inflation-indexed bonds that yield higher interest when inflation rises. Gold is a traditional hedge against inflation, although gold prices can vary unpredictably. Businesses can use inflation as an excuse to raise their prices above the real inflation rate (price gouging). Contrary to popular belief, profiting from inflation is actually quite common.
Back to frequently asked questions
What happened to the Java version of Tom's Inflation Calculator?
I discontinued the Java version after 2016. The newer JavaScript version is now the only one. Functionally, they were almost identical. Technically, the older version was written in the Java programming language and the newer one is written in the JavaScript language. (Despite their similar names, they are very different programming languages.)
To run the Java version, most web browsers required users to download and install Oracle's Java Runtime Environment (JRE) and customize their security settings. The JRE is free but is incompatible with some browsers, and some people prefer not to install it. Also, many people find it difficult to customize the security settings or worry about lowering their browser security. The JavaScript version rarely requires a plug-in because almost all popular browsers support it as a standard feature. It also runs faster.
Apart from appearances, the main difference between the two versions is their data sets. Both versions default to the Consumer Price Index for All Urban Consumers (CPI-U, Annual Average). Both versions also offer the following alternative data sets: U.S. Medical-Cost Inflation (a CPI component), CPI-U (December to December instead of Annual Average), and ShadowStats (a private-sector source). In addition, the Java version had the Consumer Price Index (CPI) for Urban Wage Earners and Clerical Workers (CPI-W), and the JavaScript version has the Social Security Wage Index and the Personal Consumption Expenditures Price Index (PCEPI). If popular demand warrants it, I will add the CPI-W data to the JavaScript version in the future.
Back to frequently asked questions