An Accurate Understanding Of U.S. Inflation: A Deep Dive with John Williams Of Shadow Stats
27:02 Share

Forget frequently asked questions. Common sense, common knowledge, or Google. How about advice from a real genius? 95% of people in any profession are good enough to be qualified and licensed. 5% go above and beyond. They become very good at what they do, but only 0.1%.

are real geniuses. Richard Jacobs has made it his life's mission to find them for you. He hunts down and interviews geniuses in every field. Sleep science, cancer, stem cells, ketogenic diets, and more. Here come the geniuses. This is the Finding Genius Podcast with Richard Jacobs.

Hello, this is Richard Jacobs with the Finding Genius podcast. I have a very interesting guest today, John Williams. He runs a website called ShadowStats. ShadowStats.com. Just a quick note for listeners, ShadowStats.com is undergoing some renovation. There's some tech issues. So John has an email list where he can update you on current stats as he sees them.

So, again, first of all, welcome, John. And then how can people get on your email list to start getting stats from you directly? Well, thank you, Richard, for having me on your program. All they need to do is send me an email, johnwilliams, all one word, at shadowstats.com. I do receive the emails there, shadowstats.com. I've expressed an interest in us and that material explaining what's involved.

That's great. So regardless, your site's very popular, at least with all the financial YouTubers. I always hear the reference to every single one. So maybe some background, like who are you? How did you first conceive of creating these stats and researching them and the website and everything? What's your background? Well, I've been an active economist, a

pretty much since I got out of Tuck Business School back in 1973, I think it was. So, 72. And I'd always had an interest in economics and got into econometric modeling, developing models on a variety of areas, including the chainsaw industry, which was family business. And as I got into looking at all the government programs

numbers that went into the econometric models. With an econometric model, you're trying to forecast something based on a history of other factors so that if you have all the other factors and

you have enough lead time going into the future, you can plug those numbers into your equation and it'll tell you where things are going. I found that a number of the government statistics had a number of problems and a number of them had been changed over time and were inconsistent and

And so I had to go back and look at what had been changed and started modeling the numbers for what they should be or looking at alternate data that was still published. I'll give you a very specific example. Being a very important number, the Consumer Price Index. Back in the early 80s, the government had a circumstance where you had a large jump in the

and the, and the CPI coming into, um, social security payment time. And, uh,

And Newt Gingrich was heading the economic committee in Congress. And this meant that they'd have to make a very big adjustment to the Social Security payments going forward, which was eating into the budget that they were looking at using otherwise. And he talked to Catherine Abram, who was the head of the Census Bureau, the Bureau of Labor Statistics.

And she noted in her memoirs, they said, hey, gee, if you could just find a way to restate the inflation numbers so it would be lower, we might be able to find more money for the Bureau of Labor. They redefined the CPI. Yeah, I heard about the CPI.

The CPL, it's very good. It knocked one and a half percentage point. The changes they made at that time knocked one and a half percentage point off the annual headline consumer price index, which was used, again, for adjusting Social Security payment.

Well, I took offense at that. I wasn't on Social Security at the time, although I am now. Well, when people agreed to go on to the Social Security system, it was your understanding that the payments would be ejected for inflation. It wasn't that, well, it'll be ingested for inflation unless the inflation is too high, in which case we're going to lower the inflation rate. And the changes...

over the decades. And right now, the aggregate difference between what the Consumer Price Index was, say, back in 1982 and where it is today, based solely on the definition of the CPI components, is a little over 9% at points. And all those redefinitions were aimed specifically at

reducing the headline inflation. So that if you're looking at 2.5%, 3% inflation, you're looking at something that's in terms of the headline CPI, you're looking at something that really should be up closer to 10%, 11% point. And that's a significant amount that's built up over time in that differential. What's the, like recently over the past year or so, what's been the

stated inflation rate and what's the real inflation rate? Well, the differential has stayed fairly constant for several years. But let's say you're looking at headline inflation at 3%. It should be up at 11% or 12% year over year. That's what it could be. That's what the cost of living adjustment should be.

So that was the first time I got into what the government was doing with its numbers and published what I've referred to as the alternate CPI. It's the way the Consumer Price Index used to be looked at, the way that the Fed chairman of the time always in place and fighters looked at it at the time. It was truly a political gimmick.

put forth by the Congress to cut the Social Security inflation adjustments. But that wasn't the only area that changed. Of course, once you change the inflation, that affects 12 because...

They had like core inflation and this inflation and that inflation. Now, they did multiple methods. Those are different measures. The core inflation is net of food and energy, where food and energy are big, volatile, and excuse me, I think it's the other way around. I think, yeah, the food and energy. So...

eliminate the other areas, but that's not used for the Social Security adjustments. Social Security adjustments, just the pure GPI. And the pure CPI right now is what's been redefined over time very specifically to cut the Social Security adjustments. And again, running in that 8% to 9% range understatement.

So do you think that continued through the past few years of the whole pandemic and everything, or has it changed the ratio? Well, actually, that's minimal going into the pandemic. And again, they publish these numbers over time. Now, in the pandemic, they did a couple of funny things where you had unusual inflation patterns with the pandemic, and they made some changes in terms of using some monthly averages that were smooth for a period of time. All

All I do is I take the government numbers as they report them, and when they make a statistical change, when they make a methodological statistical change, they publish a paper on it and says, here's what we've done, and then they estimate the effect of the change on the headline CPI. And going back to 1982, there had been a...

order magnitude of 15-20 changes, again, the aggregate of 8-9 percentage points. And that's just from what they have published. Now, what happened with the pandemic, they've not published any numbers on yet, and they may well have reversed it, so there may not be any lingering impact from that. But I haven't added any of those numbers into my estimate yet, because I use the government's estimates on what they've done. And

just make the adjustment to the headline numbers as to what they should have been otherwise if they were not being offered politically. So you're using just an older model that you feel is more accurate than the current model? Well, it's not the model so much. It's the actual number. In other words, if the CPI was 5% in a year, but the CPI

They've reduced the headline reporting by three percentage points aggregate. And since they started making the changes, as opposed to just giving a consistent reporting over time, consistent methodology. Well, if they use a different model, I could see they'd have an excuse.

But if they just change it without any rationale, like what is their reasoning? Or do they just hide? They have each, again, going back to the early 80s. And we're looking at a period here approaching 40 years where they've been doing it. Although, again, we don't have anything in the last five years or so at the moment because they haven't reported it all. And again, a lot of it was tied to the COVID crisis that may never be

It may have been reversed out. But going back before that, you've had a large number of changes methodologically where they've redefined an element

in the way they measure it. Every time it is reduced, the headline inflation from what it would have been otherwise, and the Bureau of Labor Statistics published a paper on it explaining what they have done and their estimated impact on the headline CPI. And every single change that has been made, redefinition that has been made, has lowered the headline CPI.

And we go back to the point where they started the redefinition. And keep in mind that the CPI used to be Sankratank. It was though you didn't go back and redefine things, you just, what was reported was reported. And the definitions usually carried forward. That's no longer the case. What gets reported is still reported, but the definitions get changed.

So that when they get a reporting with a new definition, you're not seeing a particular component. And they don't go back. Like when people pull data and they show trendline charts, they're probably not adjusting for this. And I don't know if they've gone back and changed the old data. How do you draw a trendline properly if there's been so many adjustments to it? Well, it depends.

that then. Let me put it this way. If you can put all I do, the government publishes when they made a change. When they make the change, prior to when they make the change,

then there's nothing to adjust for. But after they've made the change and they're, say, reducing it year after year to 1% below where it would have been using the old methodology, then you have a series that is not consistent over time. You're not seeing consistent inflation numbers today calculated the same way they were 10 years ago. So that, again, if you take the changes that they have made

And in each case here, the changes have reduced the new headline inflation against what it would have been under the old definition. Those changes go forward and if you aggregate them, you'll find that over enough time as we've had here, you're getting the CPI underreported in that 8 to 9 percentage point range. It's significant.

Before we continue, I've been personally funding the Finding Genius Podcast for four and a half years now, which has led to 2,700 plus interviews of clinicians, researchers, scientists, CEOs, and other amazing people who are working to advance science and improve our lives and our world. Even though this podcast gets 100,000 plus downloads a month, we need your help to reach hundreds of thousands more worldwide. Please visit findinggeniuspodcast.com and click on support us. We have three levels of membership.

from $10 to $49 a month, including perks such as the ability to see ahead in our interview calendar and ask questions of upcoming guests, transcripts of podcasts you're interested in, the ability to request specific topics or guests, and more. Visit FindingGeniusPodcast.com and click Support Us Today. Now, back to the show. Again, from what people are actually experiencing or used to experiencing.

what was used initially in estimating cost of living adjustments for Social Security. If I went and went to government websites and tried to reconstruct a chart myself from 1980 till now, would I just be getting garbage data? You have to go through all what you know in order to construct a proper chart. Is that what you're saying? Well, what I've done, and again, the website at this moment is inactive.

And it's been active and active for two years while we're rebuilding it. But all my subscribers are receiving email updates on a regular basis that show exactly the same thing that was shown on the site. Show the headline reporting, but also what...

The reporting would be if you use the original CPI, if you're not using gimmick CPI that had been developed to deliberately interstate cost of living adjustments for Social Security. That's why the changes were made. And Congress is very open about that. Again, I just reversed the...

understatement and from what the government has said the understatement is and add it back in and again you're looking up in that 8-9% range instead of down where the headline numbers are today at 3% or 2.5%

half percent, whatever they're running. So what has been the actual inflation rate over the past, like, you know, five years, let's say? Well, let me put it this way. The over the last five years ago, the last five years, we're getting an average of about 3%. It would be up in the, you know, the 5% to 8% range, depending on the year and the circumstance. You've been through a pandemic and all sorts of things. Well, that sounds pretty low. I thought it'd be much higher than that.

Again, the differential, instead of seeing 3%, you're up around 11% or so with the 8% differential, 8 percentage point differential. Well, that's built up over time. When was peak inflation according to the data that you have and how much was it? You're saying the latest numbers?

Just over the past few years, when was the highest increase in PPI? Over the last several years, you've seen headline inflation 3% plus or minus. It was down lower this year because of the pandemic. There's been some variability in that. But let's say you've averaged around 3% the last several years.

that'd be up around 11% or the 8% difference. For a lot of items, food, et cetera, they seem to be up 30 to 100%. I don't know if that's baked into the feed. There are things that go up 100% or such, and they build...

And they normally get balanced out. I mean, there's a good reason for balancing something out. If it's unusual in the year, you get seasonal adjustments and things like that. And you get different cost increases in a wide variety of things, ranging from food to clothing to automobiles.

The CPI brings them all together and weights them. And if it's weighted consistently, what I'm saying is you'd see inflation about 8 percentage points higher than what they're reporting. The reason you don't see it higher is because they've changed...

the weightings over time to reduce the headline inflation. And all the changes have been to the downside and the very specific purpose, as it was expressed in the earlier days of Congress when this was set up, was to reduce

to cut the social security cost of living adjustments because that was becoming too big an expense in the federal budget. That's crazy. So when the economy was quote-unquote healthy or things seemed to be going well, what was the inflation rate, the CPI back in those days versus now, let's say?

I don't have all my numbers in front of me. You've seen... Let me put it this way. Back in the 80s, it was around 3%. You got up to maybe 5%, 6% different points in time. But after they started lowering the numbers, that also...

Even when the numbers were higher, they weren't as high as it would have been without the gimmicks. Right now, we're at about, again, 3%, let's say, the average in this last year. And you add the 8 percentage points that has been redefined out of the aggregate series, you're up around 11% as to where it would be had they not made the changes. And with the redefinition,

And the redefinitions were made deliberately for purposes of lowering the cost of living adjustments for social security. That's why it was done. Right, I make them. Okay. And Catherine Abrams at least explained it that way. And it was requested by the Congress and they accommodated. Okay. So what other stats or BS, you know, what about employment?

You know, they've monkeyed with everything, it sounds like. Well, I look at the numbers the way I view it. However, the numbers were set up and allowing for, you know, getting an initial stable reporting and such going forward in time. That gives you something of a consistent base. And if you continue to report the numbers on a basis, then you're getting a fair picture of what's happening with unemployment or the GDP or the Consumer Price Index.

There have been some changes that have been made to the unemployment over time. Very simply, in order to be counted as unemployed, you have to be actively looking for work. If you're not actively looking for work, you won't be counted in the headline unemployment number. But there are broader unemployment numbers. Well, let me back up. In the headline unemployment number, they will count people who are given up looking for work because there are no jobs to be had.

at least that's their opinion, but they're not counted after six months. What you'll find is that there are a lot of people out there in the broader unemployment measures that, you know, after six months, if you run the numbers where they're not no longer counted in even the government's broader numbers, you get an unemployment rate that goes well into the double digits. And that's it. These are people who've given up looking for work because there are no jobs to be had.

And after six months or a year that the government just stops counting them in their unemployment numbers. So what do you think the differential is in employment then? Well, if you look at it, as I'm estimating it based on the total number of people who are unemployed and say there's

there's no work available, you're up over 20%. Holy cow. I mean, that's, so they're no longer in the, they're no longer in the employment universe of the government. And they have the GDP. GDP is generally a happy number. You have it, you have year to year growth, I think right now that it's, it's still,

still holding up around 3%. If you look at it, you had a lot of volatility there with the pandemic. I'll give you the actual numbers. If you look at the December 2024 consumer price and

the latest published year over year inflation is 2.89%, 2.9% I'll take. It's close to the three that's been sort of the average over time. It's been coming down recently. It hit a peak, hit a peak of eight and a half percent at the, in 2022. Again, this is pandemic related and it,

There are a lot of distortions tied to the pandemic, but right now at 2.9%, we're sort of back to where it was pre-pandemic, and that would be within the bounds of what the Bureau of Labor would be reporting. My estimate based on that, given the differential of all these factors over time, would be at 10.7%. So you've got a differential there of magnitude 8 percentage points.

Okay, and then for GDP, we talked about GDP, we talked about employment. I'll give you the GDP very precisely. Here's yours. The unemployment, let me give you that one. Yeah, unemployment. So what was the actual then according to your calculation? And then GDP. I can give you the actual numbers here. You said it's about 8%, so 12%, I guess.

Actually, these are numbers that I posted. Okay, right up through December, the headline TPIU, that's the one that's most broadly followed. It's very close to the one used for the Social Security payment. That was year over year in December 2024 was 2.9% based on my estimate and the changes that they've made, again, the aggregate changes.

They've made over time defining the CPI lower, which I add back in to where it would, this is what it would be showing. Had they not made all those changes, it's 10.7%. Changes in one month or yearly? This is year over year. Okay, got it. The way inflation usually look at, what's the CPI versus last year? How much has inflation increased in the last year? So it's 2.9%.

The 2.9 headline, if you go back to November, it was at 2.8. It was 2.6 in October, 2.4 in September, 2.5 in August, and then it goes back up to that. This is the headline numbers. The numbers that go back in time were higher. It was 3.5% back in March of 2024. That's the headline CPI. My number rides high.

Order of magnitude about 8 percentage points higher than that. So that you're right now, the December's at 0.7%. Yeah, what about employment and then GDP? What do you think the real numbers are? I'll give you the moment, Sean. Headline GDP, this is the annualized quarter-to-quarter inflation adjusted.

For the third quarter of 2024, we're coming up on the... This is the final estimate for the third quarter. We're coming up on the fourth quarter estimate at the end of this month. It could be the next couple of days. It'll be out shortly. But for the third quarter, this is annual year-to-year change, not the quarter-to-quarter where it gets... It's raised to the fourth power. This is just how it was against...

The official is at 2.7%. That's for the third quarter of 2024. Now it's down from 3.1% in the second quarter versus 2.9% in the fourth quarter. Excuse me, first quarter of 2024. It's been varying between sort of flat and 3-4% in the last couple of years. You get back to the pandemic, it actually hit a peak of 4%.

12% in the second quarter of 2021, but that was coming after the collapse of the pandemic. Is that an employment or what is it? This is the gross domestic product. Right now, the latest effects of the pandemic effectively are behind us.

So right now, if you're looking at, and this is a year-over-year change, this is the third quarter of 2024. It's the most recent quarter they've posted. The year-over-year change in real GDP, inflation-adjusted GDP, 2.7%. That was lowered for 3.1%. The government's got it at 2.7%. I'm down, my numbers show down six-tenths of a percent. And...

There's a lot of gimmicking in the GDP. Let me put it this way. The GDP is booming. If you look on a year-over-year basis, the government right now is showing it up 3.5%. Yet, if you look at all the monthly numbers,

Using the headline numbers on the real retail sales, industrial production, housing start, the cash trade index, they're all flat to minus year over year. How's the GDP up 3.5% when the monthly numbers are flat to minus?

We just got an election-boosted GDP number, which didn't work too well for the administration. Okay. Well, very good. Thank you so much for coming on the podcast. I really appreciate your time. If you like this podcast, please click the link in the description to subscribe and review us on iTunes. You've been listening to the Finding Genius Podcast with Richard Jacobs.

Thank you.