Commentary
The latest inflation numbers offer grim news. The Bureau of Labor Statistics reports a 4.2 percent year-over-year rate of increase in the Consumer Price Index (CPI), mostly driven by energy prices. What’s more alarming is that the news might even be worse than this posted number. The BLS has inherited such a confused thicket of mathematical messes that we struggle even to know the real rate.
The enormous bout of inflation we’ve experienced over six years—a result of $7.4 trillion in money printing—has usually been accompanied by strange public messaging. When it began, it was said to be merely “transitory,” a peculiar word that sounds like temporary but it really meant something else. We were transitioning from one thing to another.
As it turns out, the transition was to a dollar with 50 percent less purchasing power. This means that the cumulative inflation rate over these years has been 100 percent.
If you look for that number at the Bureau of Labor Statistics, you will not find it. The agency calculates its CPI in a way that disguises the effects of policy on the prices that people actually pay. This has been going on since 1986 with eight significant changes in the way prices are measured and rendered.
During low inflation times, these changes did not matter that much. They were annoying but did not fundamentally distort reporting. But after 2021, all the sectors that had been subject to technical changes in reporting began to soar. This included cars, medical services, housing, and groceries that are famously subject to substitutions (instead of beef, people buy chicken).
As a result, the numbers we have been presented for six years have wildly underreported the actual inflation rate. This is why with each monthly print for all these years, the headlines have usually been about inflation having been cooled, tamed, reduced, and otherwise improving. Meanwhile, the dollar was buying less and less.
It felt like gaslighting. It is technically true that no one was actually lying. The employees at the BLS were merely doing what they were told. The problem was the models themselves, which had rejiggered in ways that made their jobs impossible.
And what is their job? To build an index: prices averaged in one period are compared with the same prices average in another period, calculating the change. Doesn’t sound that hard. But when you get the theorists involved, matters get complicated. That’s precisely what happened across a huge range of sectors. Each excuse for manipulating outcomes had a seemingly rational basis. Once you add it all up, you end up manufacturing fiction.
We just did not know how far from reality this fiction had become until the real test arrived. That test was a near-hyperinflation born of the pandemic crisis. All these years later, we can easily see the problem. The BLS data is no longer useful doing what it is supposed to do.
This really matters. It’s not just a matter of curiosity. The CPI impacts tax rates, benefits adjustments, salary increases, business cycle dating, and valuations on financials. When it is wrong, everything else goes wrong along with it.
There have been many attempts through the years to construct alternative indexes. The trouble is that doing so is a herculean task. Even if you do it for one period, calculating inflation rates requires doing it for many periods, month by month, dating back years and decades. That has been beyond the ability of a single individual or even an entire team. The computing power just has not been there.
The advent of Artificial Intelligence has changed that. One American entrepreneur living in Spain, Tom Elliott, finally saw an opportunity. As a weekend project, he figured out how to eliminate all the noise in the system and just look at actual prices. He uses pre-adjusted numbers from the BLS: actual home prices, actual car prices, stable items in a grocery basket, actual insurance rates, and so on.
The result was realityindex.co which is the site he made to explain his entire methodology and the results. He finally posted it complete with charts dating back to 1980. His headline result is that prices have gone up 5.4 times over 45 years.
To be sure, there will be criticisms. One that has been floated is that he uses real college tuition prices without accounting for grants and loans. That’s a strength, not a weakness. To say they should be adjusted is no different from claiming that grocery prices should be adjusted because some people use food stamps. It doesn’t matter. We are seeking real and not subsidized prices.
Looking at his eye-popping numbers, I decided to take this two steps further. I used AI to telescope in on the most recent inflation period from 2020 until 2025. Here is where Elliott’s charts show the highest divergence between official numbers and his own Reality Index. The results initially stunned me and I wondered if I had done something wrong. No matter how I rewrote my prompts, the result was the same: a 50 percent loss in purchasing power and a doubling of prices.
This is the point at which I had to do a mental reset. How does this fit with my intuition? It fits perfectly with my experience. Just to be sure, I called one of the most credible economists to ask him, explaining the method and approach and giving my conclusion. He thought for a minute and said that this actually sounds correct. That gave me confidence.
Even so, it has taken me a full week to intellectually process the implications. It means that if you made $100,000 a year in 2019, and have not received a raise, you really make $50,000. In fact, very few people’s wages or incomes have kept up with this level of inflation. Nearly everyone is poorer now than they were back then.
That means that household income needs to be adjusted. Indeed, with these new numbers a range of other data points can be modified: inventory valuations, income, wages, salaries, retail sales, and so much more.
Above all else, we can use these new numbers to adjust output as recorded by the Gross Domestic Product. To find real GDP, the Commerce Department uses a GDP deflator. What happens if we use the new Reality Index and adjust the GDP? Prepare for another shock: we have lost 12 percent in GDP over six years.
Once you mentally process that figure, you can compare it against your perceptions of the world around you. It does make sense. And this is why we are seeing the lowest rates of consumer confidence ever recorded, and also why previous metals prices have soared over two years. These markets never believed official numbers. The gold and silver price tends toward truth telling in the long run.
I’m very sorry to be the bearer of this bad news but there is a silver lining. It is better to know than not know. Being aware of a 50 percent purchasing power loss over six years has a major impact on household budgeting. It’s obviously time to cut back, to focus on saving money by cutting expenditures, and otherwise refashioning one’s lifestyle to fit with the underlying economic reality.
Who is to blame? It is the usual suspects. It’s Congress spending more than it has at a rate that breaks all records, and the Federal Reserve that steps in to buy the debt to keep it off the markets and creating new money to do so. It’s a financial shenanigan as old as the hills.
Since the inflation of the 1970s was tamed, we were fortunate to live through four decades of relatively stable prices or, at least, a manageable rate of inflation. Our good fortunes ended in 2021 and now we must live with the results.
Views expressed in this article are the opinions of the author and do not necessarily reflect the views of The Epoch Times.




















