Is GDP (via BEA) Measuring Growth or Inflation?
We noted earlier today that the Bureau of Economic Analysis reported revised Q2 GDP data at a better-than-expected annualized 3.3%.
As discussed in the comments, the measure of Inflation is crucial to getting an accurate read on GDP (or Durable Goods). Say you live in a country that produced $100X worth of widgets in Year 1. In Year 2, it produced $110X worth of widgets. What was your GDP gains? 10% ? 0% ? Or something in between?
If your inflation data is ~2%, then you can conclude that the bulk if those widget sales was growth.
Back to the US 2008 Q2 data: Here lies the gravamen of the issue. Part of the reason the GDP number looked so good was because the GDP price index for the second quarter was marked at just 1.2. In other words, BEA subtracted from nominal GDP 1.2% in order to produce their version of "real" (inflation-adjusted) GDP.
Mike Panzner sends along the chart below, along with these comments:
"Call me a skeptic, but based on the accompanying graph of the GDP inflation figure and headline CPI (which most people already believe is lower than reality), there seems to be something of a disconnect between the two (which would imply, of course, that U.S. economic growth is a lot lower than reported)."
That is precisely the issue at hand. The GDP Price index is even lower than the already laughable CPI inflation index.
>
Click for ginormous chart
Chart courtesy of Michael Panzner
>
Take growth. remove inflation ex inflation. What's left? I cannot even make up a term for the Frankesteinian mess that results.
But it sure as hell ain't Real GDP . . .
>
Previously:
Goldman Sachs’ Jan Hatzius: Don’t Be Fooled by Inflation (August 2008)
http://bigpicture.typepad.com/comments/2008/08/durable-goods-d.html
Q2 GDP = 3.3% (kinda) (August 2008)
http://bigpicture.typepad.com/comments/2008/08/gdp-33.html
Thursday, August 28, 2008 | 12:00 PM | Permalink
| Comments (45)
| TrackBack (0)
add to de.li.cious | 
digg this! | 
add to technorati | 
email this post
TrackBack
TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c52a953ef00e554915e6c8833
Listed below are links to weblogs that reference Is GDP (via BEA) Measuring Growth or Inflation?:
Comments
It looks like a similar divergence before three prior recessions, too.
Posted by: wally | Aug 28, 2008 12:07:26 PM
That's the way this central planning economy works.
If Bernanke and Paulson wish for zero inflation they'll be reported zero inflation.
Posted by: Aurora Borealis | Aug 28, 2008 12:09:54 PM
The graph shows similar divergence between the two series before the 78, 88, and 98 recession bars? Is this a natural pattern, as opposed to doctored data? If yes, it still does not change the dire predictions.
Posted by: tyaresun | Aug 28, 2008 12:16:20 PM
The logical explanation for the divergence could be that the CPI is increasing but the GDP index is decreasing as producers reduce prices in light or recession related reduced demand. These price reductions show up in the CPI with a lag as shown by your graph.
Posted by: tyaresun | Aug 28, 2008 12:21:15 PM
br -- i can make out qualitatively similar disparities in 1979, 1987, 1990, 1996, and 1999.
overlay those years with a chart of YoY WTIC. i think the answer is there.
the GDP price inflator doesn't account for price hikes in oil imports.
are we experiencing inflation? yes. but it is a result of import price increases. domestically, thanks to crashing credit availability, i think we're actually experiencing something closer to a deflationary environment.
i think we'll get the real GDP results we're "feeling" if the intensifying global slowdown results in a rising dolllar, lower oil and slowing overseas demand for american exports.
Posted by: gaius marius | Aug 28, 2008 12:21:24 PM
Sure is quiet over here....
I'm hoping we can start the global depression MEME...
not like we take requests..
Posted by: Eric Davis | Aug 28, 2008 12:25:16 PM
gaius marius has it right. The oil price is in CPI. It is actually TAKEN OUT OF GDP. When they add exports minus imports, the tick up in oil price actually depresses GDP deflator.
Voila! Big tick up in oil price = higher real GDP growth.
Beautiful isn't it?
Posted by: ReturnFreeRisk | Aug 28, 2008 12:37:19 PM
Maybe the lower demand drops input costs after a spike up of initial higher demand from a lower valued dollar. Just a thought I had.
Posted by: GB | Aug 28, 2008 12:39:04 PM
Skeptical minds think alike. Posted the historical difference between the GDP Deflator and CPI this morning:
http://econompicdata.blogspot.com/2008/08/gap-between-gdp-deflator-and-cpi-widest.html
As Gaius correctly points out, it seems to match the spike in oil extremely closely.
Posted by: Jake | Aug 28, 2008 12:39:20 PM
This is comical how low the defllator is right now.
Is this current divergence between the inflation rate and GDP inflation 2nd highest on record? Looking back at the chart, it looks like the only time it was this far apart was right before the giant recessions of the early eighties.
I fully agree with you Barry. We need to get a handle on this because without good information it is much harder to make good decisions.
Note that if you use a reasonable rate of inflation for the last few decades, we haven't had ANY economic growth. I don't think this is about right, as my life is probably about 15% better than my dads was, but this is mostly due to technological progress. In terms of house, car, time, food, I am not that much better. Certainly not 100% better as real GDP/capita figures would have us believe.
Posted by: mickslam | Aug 28, 2008 12:45:50 PM
I've been working the "real" GDP data a little this morning, I think the GDP number released this morning is WRONG USING BEA'S OWN APPROACH TO THAT CALCULATION.
One way it reports the GDP data is through two tables that show (a) the change in GDP on a current and real basis (which is where the 3.3% number shows up for Q2) and (b) the actual estimated annualized GDP for the quarter on both a current and chained 2000 dollar ("real") basis(put at $11.7T for Q2).
I have compared these two sets of data and the results are amazing. I compared the STATED change in real SAAR GDP from the first table with the CALCULATED real AR GDP from the second for the 34 quarters from 1Q2000-2Q2008.
For all quarters EXCEPT the last one (2Q2008), the difference between them is minuscule, averaging less than 5/1000 of a percentage point with no quarter showing a discrepancy of more than 5/100s percentage points. In short, they are virtually identical.
Now, the 2Q2008 comparison shows an absurd difference of 1.41 percentage points. That's the difference from the stated 3.3% growth in real GDP compared with the calculated change of 1.89% USING THE SAME OFFICIAL DOC/BEA DATA SET.
I have no idea how the two numbers can be virtually identical for 33 quarters (and probably much, much longer) and suddenly explode to a difference of 1.41% in the most recent quarter.
It's either horrendous homework or fraud. Go figure!
Posted by: Terry | Aug 28, 2008 12:46:24 PM
Same arguments apply to the retail numbers. Retail sales looked fine for a bit because everyone was buying expensive gas for their cars.
Now that inflationary component is being taken out, retail sales figures are going to be exposed.
Posted by: leftback | Aug 28, 2008 12:58:22 PM
BR: Does this mean you lost your bet regarding recession?
Posted by: johnnyvee | Aug 28, 2008 12:58:27 PM
what if the treasury held an auction for $22 billion, 5 year notes and nobody came? keep an eye on rates today...
Posted by: karen | Aug 28, 2008 1:02:29 PM
Terry:(11740.3/11646)^4-1 = 3.3%
Not sure if you were looking at the old #'s...
Posted by: Jake | Aug 28, 2008 1:06:19 PM
World markets are getting too big. With all the different news flows, it makes it too easy to put lip-stick on the financial pig.
Posted by: Concerned Citizen | Aug 28, 2008 1:19:15 PM
BR: It might help to understand how Real GDP is determined - Jim Hamilton anticipated your question back on 5 August 2008!
Posted by: Ironman | Aug 28, 2008 1:21:14 PM
@Karen:
If nobody wants the 5-year notes I will happily buy them all... although Bill Gross will beat me to it.
People who are bearish on Treasuries should consider the fact that when we finally see the deflation beast is at the door and the bottom falls out of this thing, the only place people will want to be is out of stocks, junk and MBS and into - Treasuries.
A look at JGB rates shows you where the path of deflation leads.
What happens when we get there? That is I agree a question for debate.
Posted by: leftback | Aug 28, 2008 1:33:37 PM
fudged ___? (fill in you favorite econ. stat acronym here: e.g.: CPI, GDP, NFP, etc.). Hey, it's an election year...
Good article at Bloomberg re:GDP vs. GDI:
http://www.bloomberg.com/apps/news?pid=20601087&sid=aIWiIsk9fzl8&refer=home
"...the income [GDI] and growth [GDP] figures should theoretically match..."
Posted by: fudged_numbers | Aug 28, 2008 1:36:17 PM
"Voila! Big tick up in oil price = higher real GDP growth.
Beautiful isn't it?"
I'll bet you my universe we could find quite a few economists who would tell us with a straight face that GDP deflator and CPI ,as defined works well.
Then, when asked how pertinent (you know, this thing about being practical and all that jazz) the resulting measure/calculations are, they'd look at us with a blank stare and an air of total incomprehension. I can already hear their reply:
"But...pertinence has nothing to do with it"
Posted by: Francois | Aug 28, 2008 1:36:53 PM
Ironman-
My response to Jim's post: I think James' example is simple and his logic is easy to understand, but it isn't necessarily applicable to the real world (or specifically to the U.S.). James' story involves an Island which grows one good (a coconut, which it doesn't "produce") and imports oil.
The U.S. does not have many "natural goods" like coconuts. Thus, we use that super-expensive imported oil as an input for our final goods. Thus, inflation has crept up as this expensive source of energy has flowed through the supply chain (YoY: energy up 70%+, PPI up 10%, CPI up 5%, Core CPI up 2.5-3%).
A better example would have been coconuts and a fertilizer that is essential to grow coconuts. In this case had fertilizer doubled in price, I have feeling that $515.10 worth of coconuts would not have gotten you the 510 initially assumed by the Island's BEA in the first round of its Real GDP calculation, but rather only 470 as the higher price of fertilizer increased the price of the final good, hurting demand. Thus, rather than an increase in GDP from 500 to 510 coconuts, there was a recession to 470 (same nominal GDP in both cases).
At the end of the day, we need to look at the numbers and ask ourselves if they make sense and 1.1% does not. Should the GDP Deflator be as high as that implied by the CPI? Probably not, but somewhere in between would cause another revision to GDP, which was my point all along...
Posted by: Jake | Aug 28, 2008 1:45:06 PM
dear leftback:
buy all the certificates of confiscation that you want as the dollar becomes worth less and less every year. and don't worry about bill gross beating you to it, cuz he's losing big on his mbs and distressed debt purchases. btw, i loved your ship analogy in comments the other day. the way money is printed on this planet, i'm not worried about universal deflation, just figuring out where the next bubble is gonna be. oh, and as an aside, you probably noted that the $22 billion auction was not gobbled up...
Posted by: karen | Aug 28, 2008 1:56:09 PM
@Karen
I think that they are in fact nibbling on that 5-year supply right now (5-year at 3.04), the voracious gobbling will come later.
Yes I do worry about reflationary nonsense, that's what gold is for. But I believe that we will see all of this stuff play out sequentially, and right now all of the danger is in equities not bonds. I'm not sure if this is 1929, but it feels like it.
Q: Did you really like my ocean liner analogy - or are you just taking the mickey? I did think it was a bit better than the bloke jumping out of a plane with no parachute...
Posted by: leftback | Aug 28, 2008 2:20:20 PM
Remember that in addition to excluding import costs, the GDP deflator includes a lot of housing related stuff that they've deliberately excluded from the CPI.
So, they've excluded inflationary things, and included deflationary things. Viola! A low deflator.
So, as previously mentioned, this might be calculated correctly, but it's not actually measuring anything that anyone intelligent should find useful.
Posted by: Jim D | Aug 28, 2008 2:22:31 PM
How much oil is produced _domestically_ in the USA? You can't exclude the oil price change entirely by pointing at imports.
Posted by: Aurora Borealis | Aug 28, 2008 2:32:28 PM
