Gross Domestic Product (GDP) and Gross Domestic Income (GDI) are two measures of the same thing. But the difference is now over two percentage points, the third largest in history.
Discrepancy Notes
- When GDP is greater than GDI the numbers are positive. When GDI is greater than GDP the numbers are negative.
- Last quarter the discrepancy was 2.3 percentage points and this quarter 2.2 percentage points. Only twice in history has the discrepancy been higher.
- The average discrepancy is +0.5. This suggests a tendency to overstate GDP relative to GDI.
- There is no discernable pattern other than a tendency to revert to the mean, eventually, with long trends in one direction or the other.
Real GDP and GDI in Billions of Dollars
Chart Notes
- Gross Domestic Product (GDP) and Gross Domestic Income (GDI) are two measures of the same thing. Product produced should match sales and income. They do over time, but this is a large ongoing discrepancy.
- Real Final Sales is the bottom line estimate of GDP. The difference between GDP and Real Final Sales is inventory adjustment which nets to zero over time.
- Real means Inflation adjusted using the GDP deflator as calculated by the BEA as the adjustment.
In dollar terms, the discrepancy is the largest ever. However, percentage comparisons are a better measure which is why I created a new chart today to show the percentage point differences.
Reversion to the Mean
There is a strong tendency to revery to the mean. However, reversion to the mean does not imply GDP will fall. GDI could overshoot to catch up.
However, based on where the economy is right now, I would expect GDI to drop with GDP dropping more on a relative basis.
More Soft Economic Data, Q1 GDP Revised Lower, Q4 GDI Significantly Lower
Earlier today, I reported More Soft Economic Data, Q1 GDP Revised Lower, Q4 GDI Significantly Lower
Significant Negative Revisions
- 2024 Q1 GDP went from 1.6 percent to 1.3 percent.
- Based on updated data from the Bureau of Labor Statistics Quarterly Census of Employment and Wages program, Wages and salaries are now estimated to have increased $58.5 billion in the fourth quarter, a downward revision of $73.0 billion.
- Real gross domestic income is now estimated to have increased 3.6 percent in the fourth quarter, a downward revision of 1.2 percentage points from the previously published estimate of 4.8 percent.
Regarding the first bullet point, I calculated revised GDP at 1.2509 percent which the BEA rounded to 1.3.
Is GDP or GDI a Better Measure?
The Philadelphia Fed prefers GDI over GDP but it prefers a blend (not an average) which it calls GDPplus even more. Note that GDPE = GDP and GDPI = GDI in the discussion below.
The GDPplus Working Paper is mostly geekish math, but there are some readable snips.
Aggregate real output is surely the most fundamental and important concept in macroeconomic theory. Surprisingly, however, significant uncertainty still surrounds its measurement. In the U.S., in particular, two often-divergent GDP estimates exist, a widely-used
expenditure-side version, GDPE [GDP], and a much less widely-used income-side version, GDPI [GDI].
Nalewaik (2010) and Fixler and Nalewaik (2009) make clear that, at the very least, GDPI deserves serious attention and may even have properties in certain respects superior to those of GDPE. That is, if forced to choose between GDPE and GDPI , a surprisingly strong case exists for GDPI . But of course one is not forced to choose between GDPE and GDPI , and a GDP estimate based on both GDPE and GDPI may be superior to either one alone. In this paper we propose and implement a framework for obtaining such a blended estimate.
We will have an update to GDPplus later today and I will post some more charts.
Height vs Shoe Size Why the Huge Discrepancy and Which Is the Better Measure of the Humanity?
GDP uses a larger data set and is used is more timely and also is preferred by BEA. One the other hand, GPD-GDI grew for a while back in 2006 when the housing market started weakening which with the benefit of hindsight would have been a leading indicator for the financial crisis; so based on a statistical sample size of one there shows some advantage for using GDI.
They are designed by economist to track exactly for the United States economy. The GDI version includes an estimate of tax evasion which has been shown to be materially off in the past. The current net business interest income currently also does not seem to be right; some of it is falling into a sinkhole or business records are getting progressively worse (i.e. dishonest).
The question as to whether the difference of the (GDP – GDI) time series is mean reverting or not is contaminated by the efforts of economists and statisticians to make it so by “improviing” their data collection techniques.
That is a good point. The GDI methodology of today may not be the exactly the same as the GDI of 2006.
Sorry, but these are both just random numbers.
I appreciate that Mish probably spent a lot of time looking up the definitions of what the stats are supposed to be, and the differences, and the pros/cons in an ideal situation.
Lets be honest with ourselves. The problem is the criminal class in Washington (all the bureaucrats, not just the politicians) look at us as a bunch of deplorables. They admit as much when speaking about the geographic middle of the country, but public servants have no respect for the folks in NYC or San Fran or LA either.
The problem with GDP and GDI is the career bureaucrats that allegedly collect and process the data. They clearly are not doing that, and there is plenty of reason to think they are just making up the supporting data to get the final result that they (or the politicians) want.
Both numbers are garbage.
Garbage meaning LIES.
There really isn’t much point lying about attempted measures of something simultaneously entirely arbitrary and entirely irrelevant.
The only people concerned about this nonsense, are exactly the same ones who believe it doesn’t matter, at all, whether everyone in America is completely debt free; or everyone but Krugman is in debt to Krugman to the point of needing generations of indentured servitude to have any hope of paying it all back. Since, you know; It doesn’t matter, since “we” “owe it to ourselves.”
That sort of trivially obvious nonsense, is what’s being passed of as “Economics” these days. At this point: Whether the morons are lying on top of it all, is no more significant than whether a mosquito is lying or not about a problem in quantum physics.
I will dig it up but the people who do these jobs go out of their way to not respond to anyone or anything but the real data. John Mauldin had a letter a few years ago on this. Economic data is very noisy at a baseline.
Government employees contribute negatively to GNP.
1) The 30Y-3M dropped 4% from +2.51% in Mar 2021 to (-)1.64 in May 2023. It’s in negative territory. It might rise to close the gap at 1% and up to 2% ==> the Fed will cut rates and the yield curve will normalize.
2) Between Dec 2008 and Oct 2018 the 30Y ranged between 4.8% and 2%. Available
For Sale were sold at break even. The banks bought at higher rates accumulating them at higher rates, in order to sell them when rates drop, taking profit.
Casue they don’t measure the same thing.
Production vs income doesn’t strike me as the same.
As automation & technology replaces labor, there’s bound to be a lag
That said, having a celebration right now with family & friends, today marks the beginning of the end of wealthy privilege.
It’s safe to walk down fifth avenue once again.
.
On the grounds that the answer is usually ‘somewhere in between’, I averaged the three for the latest and also one year earlier, realiting in a rise of 2.5%. Allowing for a modest downward revision, I’d say we’re growing at a modest 2%. Our big competitors are doing better, granted we’ve got the job policing and running the world. Be great if we focused on ourselves.
The best measure of the economy is a combination of tax receipts, rail freight traffic, and energy consumption; oil , gas, electric. Everything else is an excuse to employ government liars.
The best measure is CFNAI-MA3.
https://www.chicagofed.org/research/data/cfnai/current-data
“ rail freight traffic”
I’d debate this, CRT’s vs flat screen, cast iron boilers vs condensing boilers the size of a suitcase, desk top’s vs laptops.
Everything’s getting smaller and lighter.
trump trial verdict in 30 minutes….get your popcorn ready!
Guilty on all counts.
GUILTY on 34 counts. ROFLMAO.
You’d’ve thought I was watching the Superbowl, I just lost it.
I thought I was watching the jury handing trump 34 spankings.
Kabuki theater is now complete.
Let the appeal process begin.
Trump lived most of his life in New York, he truly got judged by his peers. Appeal is irrelevant, he is now officially a convicted felon. Will cost him votes from sane and moral people.
I doubt being a felon will change a single persons vote (I know people from both parties and none of them is basing their vote on Trump’s trial).
98% of the country has already decided how they are going to vote. The last 2% is probably going to decide based on economy and other factors.
Plus he will obviously appeal. If you win the appeal you aren’t a convicted felon.
“I doubt being a felon will change a single persons vote.”
Wrong. It will change the votes that matter the most, independents. I’m sure we’ll get updated polls over the next few weeks.
And you are assuming Trump wins on appeal, what if he loses?
Absolutely correct, this just got real for millions who didn’t realize how serious the accusations were.
“I doubt being a felon will change a single persons vote”
It will, he’ll be serving a sentence, swing voters who were on the fence will have to rationalize a confined man trying to carry out the duties of office.
Debs would have told him so, too.
Errr … Appeal can overturn conviction. In this case with high probability. But you do you.
Not the NY court system, no Thomas or Alito to save the day in exchange for friendly gifts.
You can’t “appeal” facts. They can appeal procedure etc. but not the facts of the case.
As soon as the verdict was read, my phone lit up.
He’s definitely getting time, any judge bases sentencing on remorse, Trump went after his daughter, that’s as low as you can go.
I highly doubt moderates & independents will buy the idea that an imprisoned man can effectively carry out the duties of office.
.
I think we’ll look back at this moment as the point Trump ended Trump.
De Niro celebrates with his buddies P Diddler and Epstein!
Meanwhile, just a little drive around town shows the signs of rapidly advancing inflationary depression everywhere.