Look: I am eager to learn stuff I don't know--which requires actively courting and posting smart disagreement.

But as you will understand, I don't like to post things that mischaracterize and are aimed to mislead.

-- Brad Delong

Copyright Notice

Everything that appears on this blog is the copyrighted property of somebody. Often, but not always, that somebody is me. For things that are not mine, I either have obtained permission, or claim fair use. Feel free to quote me, but attribute, please. My photos and poetry are dear to my heart, and may not be used without permission. Ditto, my other intellectual property, such as charts and graphs. I'm probably willing to share. Let's talk. Violators will be damned for all eternity to the circle of hell populated by Rosanne Barr, Mrs Miller [look her up], and trombonists who are unable play in tune. You cannot possibly imagine the agony. If you have a question, email me: jazzbumpa@gmail.com. I'll answer when I feel like it. Cheers!

Wednesday, October 5, 2011

A Different Look at Real GDP

Over at Asymptosis, Steve presents a graph of the annual change in average GDP growth for 5-Yr rolling periods.  Have a look.

He points out that the differing inflation regimes had a profound influence on the shape of the curve - which they do.  But even with inflation out of the picture, there is still a regime change, and it still happens in the early 80's.  To take away the inflation effect, I've made what I hope is a similar graph.  The GDP change line is color coded by President's Party, Blue for Dems and Red for Rethugs.

I've also included a 13 year average line in yellow, to smooth out the long term trend.  This clearly shows the two regimes, and when the change occurred.  Here, in a nutshell, is the Great Stagnation, brought to you by Reagan, Bush and Shrub.  I'm not a big Bill Clinton fan, but that dismal competition makes him look awfully good.

I'm not sure I used the same methodology as Steve.  I went to FRED, downloaded series GDPC96: Real GDP, quarterly data.  I then 1) took a five year (20 Q) average of the data, and 2) computed a percentage change from 4 quarters earlier for that average, for each quarter, starting in Q1 1952.  This percentage change is plotted as the red/blue line. The yellow line is simply a 13 year (52 Q) average of that rate of change.  Note that with the exception of B. Hoover Obama, having a Dem Pres does pretty wonderful things for GDP growth.  In contrast, having a Rethug Pres means taking giant steps toward economic stagnation.  I hope you knew that.  

From 1970 through Q1 1981, the 13-Yr. average hovers between 4.8 and 4.9% 3.8 and 3.9%.  By the end of 1984, the average dropped to 3.04%.  It never goes above 3.15% again until 1999 - very late in the Clinton administration.  Remember - this is a 13 year average, and by the middle of '02, it had already passed its late peak value of 3.43%.  From the middle of '05 to about the middle of '09 it hovered around 3.05%.  The effect of the great recession has been a slow decline to  2.78% for the most recent data point in Q2 '11.

Note also that the two big drops are at the starts of the Reagan and Shrub administrations.  This is hardly a coincidence.


The Arthurian said...

Oh, I like that you put your methodology in the post.

A typo I think: "From 1970 through Q1 1981, the 13-Yr. average hovers between 4.8 and 4.9%." Sounds good, but maybe you meant 3.8 and 3.9%... in the neighborhood of 4%.

The "Percent Change from Year Ago" view of GDPC96 at FRED shows big drops at recessions in the 1950s and 1970s, which practically disappear from your 5-year view. Your big drop there in the '80s doesn't get lost -- presumably because that was the double-dip recession & had more effect on your numbers. I was surprised that the others disappeared.

I'm wondering about your 13-year trend: Each point on the curve shows the average of the previous 13 years? So maybe your "By the end of 1984, the average dropped" decline should be associated with a time some 6½ years earlier? Centering the decline on the period it shows, I mean.

Jazzbumpa said...

Typo corrected. Thnx.

For ref, here is % change from year ago for GDPC96. I think the reason the recessions aren't dramatic in the 5 Yr averages is that the recoveries were sharp, and the events were short, so the gyrations get washed out.

The regime change in the 80's comes partially from the depth and breadth of the double dip at the time and the weakness of the recovery between.

Sure, all 13 preceding years have equal weight. So the performance of Nixon-Ford - which is really quite poor - is also indicated in the numbers. That drop reflects the deep dive of 1969-70, and the '74 recession as well.

Note in contrast that the much maligned Carter administration is an island of almost-robust growth in a sea of anemia.

The other notable feature is that once the 13 yr average went down, it stayed down. To your point about lags, the Clinton prosperity doesn't move the yellow line above three until the beginning of 1998.

I like to point the finger at Reagan - and with ample good reason, I think. But things really did start going sour in the 70's.

I distinctly remember during the Nixon tern saying that nobody will ever look back at this time as "The good old days."

I think it's clear, though, the the Great Stagnation and the slump we' can't get out of now are clearly the result of 40 years of mostly Rethug rule. Which is why I say policy matters, and there's a lot more to the story thna just the Fed.


Steve Roth said...

So if I understand, the yellow line is a 13-year rolling average of the 5-year rolling averages? Why not just use a 13-year rolling average?

Jazzbumpa said...

Steve -

Yes, that is correct. It's a matter
of sequencing. The methodology is to
1) use a 5-yr avg to smooth GDP.
2) determine the annual RoC for the smoothed GDP measure
3) smooth the RoC with a 13 Tr Avg to further reduce the effects of transitory events.

Basically, I wanted to asssure myself the the two regimes is see in the red/blue RoC line is not a figment of my imagination. I thnk that does the trick.

Makes sense to me, anyway.


Jazzbumpa said...

Actually, now that I read your question again, it's not correct.

Hope my description above is helpful.


The Arthurian said...

"to further reduce the effects"

Averages of averages. Like re-fried beans (the only analogy that comes to mind. Sorry.). I thought Steve's question was a good one, and my initial reaction was that you should average the numbers, not the averages. But there's more than one way to skin a cat, no doubt.

I like the idea of discussing methodology. I know so little about it, I'm not even sure "methodology" is the right word. Before your post above, Jazz, the word was a complete mystery to me.

Early this morning I tried to duplicate your results and I got close, but I wouldn't call it a match. I *thought* I followed the procedure in your post, but... it WAS early in the morning.

Steve, Jazz, what do you think of setting up a "standard" where in our posts we try to remember to describe the methodology, and where we provide a link to a Google Docs spreadsheet showing the work ??

Jazzbumpa said...

Art -

Your suggestion is great. I've never used Google Docs, but I suppose I could.

I used a 5 Yr average of GDP to try to match what Steve did. The YoY change should be the inflation adjusted version of his original graph.

I then took a 13 Yr average of the RoC numbers. So it's not an average of averages. The yellow line is a rolling average of the data points in the red/blue line.