PDA

View Full Version : The Great Depression tight parallels... busted (v 2.0)


bart
06-28-09, 06:00 PM
<center>The Great Depression tight parallels... busted</center>

<center>by bart</center> <center>June 20, 2009 (r)
</center>


The following monthly charts compare a few important data series during the periods starting September 1929 and October 2007, and simply measure the changes in growth or contraction. The actual facts in the monetary and fiscal areas are much different than most think or believe.




<center>http://www.nowandfutures.com/images/depression_comp_stack1.png</center>

There are truly huge differences between then and now in the following statistics:

M1 is up about 25% more now than during an equivalent period during the Great Depression, and shows positive growth.
M2 is up about 22% more now than during an equivalent period during the Great Depression, and shows positive growth.
M3 is up about 23% more now than during an equivalent period during the Great Depression, and shows positive growth.
Bank credit is up about 22% more now than during an equivalent period during the Great Depression, and shows positive growth.
Current GDP is up about 5% more now than during an equivalent period during the Great Depression, and still shows positive growth.

As rough definitions, M1 is composed of cash and checking accounts, M2 adds in savings deposits and CDs under $100,000 and money markets funds (excluding IRA or Keogh money market fund balances), and M3 adds in CDs over $100,000 and institutional money market mutual funds and a few other items. Bank credit is "Bank Credit of All Commercial Banks" (TOTBKCR).





<center>http://www.nowandfutures.com/images/depression_comp_stack3.png</center>



The Industrial Production index is about 14% higher now than then, but has fallen about 14% since October 2007. It's not as bad as then. Additionally, industrial production in the U.S. fell over 58% in the recession in the mid 1970s from peak to trough.
Federal government debt is up about 24% more now than during an equivalent period during the Great Depression, and shows significant positive growth.
Velocity (the speed with which money moves through an economy) is down about 39% now when compared to the Great Depression, and in our opinion is the primary item holding back much higher inflation. Inflationary expectations are being heavily managed by the Fed and government, etc. Also in our opinion, while debt deflation is quite real it's also a roughly equivalent situation to what was occurring in the 1930s.
The Dow is about 24% higher now than then, but volatility makes the comparisons a bit dicey.







<center>http://www.nowandfutures.com/images/depression_comp_stack2.png</center>

These are simply various rate comparisons at the beginning of the most recent month, with a comparison to the equivalent month in the 1930s. They are not differences since the beginning of either the Great Depression or the stock market peak in October 2007.


CPI is about 7.5% higher now than during an equivalent period during the Great Depression. If John Williams of shadowstats.com (http://www.shadowstats.com/) CPI corrections are used, then CPI is about 13% higher now, and also shows positive growth.
The discount rate (the rate at which banks borrow from each other) is 2.5% lower now than during an equivalent period during the Great Depression.
The 10 year Treasury Note is about the same when compared to an equivalent period during the Great Depression. It is a true parallel, if one does not adjust the rates for inflation differences, in which case it's about 7% lower now than during the Great Depression.
Unemployment is about 7% lower now than during an equivalent period during the Great Depression, when using the "official" U3 unemployment rate. However, in our opinion the U6 unemployment rate from the Bureau of Labor Statistics is a much better rate to use for comparisons since its definition is much closer to the U3 rate definition from the 1930s. It is a true parallel when U3 is compared to U6, and the unemployment rates are virtually identical.










<center>http://www.nowandfutures.com/images/depression_comp_stack4.png</center>




The trade balance is in much better shape currently, being only about 2/3 down when compared to the start of both periods. The trends are also diverging between the two periods - now being much better than then... assuming protectionism does not get worse and that trade deficit financing does not turn into a major issue, and both are far from assured.
Debt to GDP comparisons are much worse, and it is well above 75% currently. It peaked in mid 1933 at 40%, dropped and then peaked again higher at 43% in 1938.
Federal tax receipts are also worse now than then, being down about 34% since the 2007 peak. Do note that the 1930s data is sparse, and has been interpolated from yearly data.
Housing prices have dropped about 2.5 times more now than then, but do note that the 1930s data is quite sparse, and has been interpolated from yearly data.







It's not possible to show a truly full picture and comparison between now and the 1930s, and there certainly are items which are tracking the Great Depression quite well (like bank failures when total actual bank locations are counted), but its important to note the many items which are not tracking the Great Depression so that a fuller and more accurate perspective may be achieved.


Data sources: Federal Reserve, Federal Reserve archives, Bureau of Labor Statistics
Changes between then and now were calculated by taking the percentage growth from September 1929 mostly to April 1931 figures, which is the same amount of months between October 2007 and May 2009. Adjustments were made when data was not available for May 2009.

The charts above are maintained on our Great Depression page.
Now and The Future (http://www.nowandfutures.com/)

<hr><hr>

<center>An alternate view on charts from Barry Eichengreen and Kevin O’Rourke (http://www.voxeu.org/index.php?q=node/3421)

</center> In their article A Tale of Two Depressions (http://www.voxeu.org/index.php?q=node/3421) from April 2009 and updated on June 4th, some charts were included which could be quite misleading, and have been for many in our experience. As Barry Eichengreen and Kevin O’Rourke note, "World industrial production, trade, and stock markets are diving faster now than during 1929-30."... "The update shows that trade and stock markets have shown some improvement without reversing the overall conclusion -- today's crisis is at least as bad as the Great Depression."

We note that industrial production fell about 48% from peak to trough in the recession in the mid 1970s. It fell about 53% during the U.S. Great Depression. In other words, one could also say that the mid 1970s recession was very close to as bad as the Great Depression when judging by industrial production. The S&P 500 index fell about 48% during the mid 1970s recession, just for comparison.

And we have seen many sites link to the article without noting the starting date or current date issues, as well as not noting that there are many other statistics which show very different pictures and perspective as noted above.

Our basic concern is that they used a starting date for most of the charts of June 1929, which is the world industrial production peak, while the Great Depression itself didn't start for many months after the production peak. The charts also start of the current period at April 2008, which is 5 months after the U.S. official recession started in December 2007 and seven months after the stock market peak in October 2007.

Here are the Great Depression starting dates by quarter, sorted earliest to latest. As can be seen, the average starting date was significantly later than June 1929. The current date of April 2008 is also significantly later (a total gap of at least 8 months when compared to using actual Great Depression and current U.S. stock market peak dates), which quite significantly affects the picture shown in the various charts.

<table> <tbody><tr><td>Country</td><td>Year, quarter</td></tr> <tr><td>Germany </td><td>1928,1</td> </tr><tr> <td>Brazil </td><td>1928,3</td> </tr><tr> <td>Poland </td><td>1929,1</td> </tr><tr> <td>Canada </td><td>1929,2</td> </tr><tr> <td>Argentina </td><td>1929,2</td> </tr><tr> <td>United States </td><td>1929,3</td> </tr><tr> <td>Italy </td><td>1929,3</td> </tr><tr> <td>Belgium </td><td>1929,3</td> </tr><tr> <td>Switzerland </td><td>1929,4</td> </tr><tr> <td>Czechoslovakia </td><td>1929,4</td> </tr><tr> <td>Netherlands </td><td>1929,4</td> </tr><tr> <td>India </td><td>1929,4</td> </tr><tr> <td>Great Britain </td><td>1930,1</td> </tr><tr> <td>Japan </td><td>1930,1</td> </tr><tr> <td>South Africa </td><td>1930,1</td> </tr><tr> <td>France </td><td>1930,2</td> </tr><tr> <td>Sweden </td><td>1930,2</td> </tr><tr> <td>Denmark </td><td>1930,4</td> </tr> </tbody></table> Source (http://www.nowandfutures.com/d2/great_depression_romer.pdf)



<center>World stock markets, using U.S. Great Depression and current day stock market peak start dates

</center>
<center>http://www.nowandfutures.com/images/bear_markets_world1929_today.png</center>

<center>Barry Eichengreen and Kevin O’Rourke chart</center>
<center>http://www.nowandfutures.com/d2/worldstockmarkets1920-2009.png</center>

<hr width="60%">

<center>World money supply, using U.S. Great Depression and current day stock market peak start dates

</center>
A similar situation with start and end dates is also apparent when looking at world money supply, then and now. Our chart below also shows detail from 2009, which again significantly affects the picture and perspective.


<center>http://www.nowandfutures.com/images/money_supply_world1929_today.png</center>

<center>Barry Eichengreen and Kevin O’Rourke chart "Money Supplies, 19 Countries, Now vs Then"</center>
<center>http://www.nowandfutures.com/d2/worldmoneysupply1925-1938.png</center>

ggirod
06-28-09, 06:09 PM
Thanks, Bart, Don't give up publishing this stuff. I can see it will take a while to wrap my head around this, but that is what you do best. Thanks.

cjppjc
06-28-09, 06:20 PM
Bart:
Thank you for all your labors. Very informative.:)

jk
06-28-09, 06:39 PM
illuminating, bart. thanks. i find the velocity comparison especially interesting, in light of all the efforts made lately to keep the economy from crashing.

kartius919
06-28-09, 08:16 PM
I wonder what fraction of industrial production consists of defense then and now. What would the industrial production decline be excluding defense expenditures and defense related productivity?

Also, could you offer any insight into what you think would be the catalyst(s) for a turnaround in velocity? Foreign or domestic?

TheServant
06-28-09, 08:34 PM
This is a very interesting take on the comparison of the two, and really hits home how much a relatively small shift in time can change the complexion of the statistics a great deal.

I don't know that I agree with using the stock market peaks as the starting point in both, but it is difficult to say what would be the best approach. Anyway, your main point, I believe, was to show how misleading stats can be if taken verbatim without exercising some scrutiny.

I am actually a bit more inclined to follow the idea that Mar 2000 is akin to Sept 29, and Oct 07 akin to Mar 37, shifted +/- a few months, depending on the data set you find most relevant to both of those economic cycles, which complicates any attempts are comparison even more. There are just so many ways to look at the two events in relation to one another.

bart
06-28-09, 10:42 PM
I wonder what fraction of industrial production consists of defense then and now. What would the industrial production decline be excluding defense expenditures and defense related productivity?

Excellent point, and at least some of the data is available at the Fred & BEA to answer it, but its not something that I track. Perhaps someone else has input or charts?

Also, could you offer any insight into what you think would be the catalyst(s) for a turnaround in velocity? Foreign or domestic?

I'm pretty much aligned with iTulip here. The chances of some kind of sudden shock like 9/11, or a "black swan" like what happened in Argentina in 2001, or even something like the assassination of Rathenbauer in mid 1922 in Weimar Germany (which in my opinion caused velocity to move into a whole different range due to confidence loss) are very definitely non zero probabilities. The possibilities are almost endless with so many and varied hot spots like North Korea, Pakistan & even California... or in tinfoil hat mode, "accidents" at Goldman Sachs or similar. I think that China isn't very likely to be a catalyst though - they have too much to lose, and they're not stupid.

It's almost trivial to put together comparison charts like the ones above when compared to trying to predict politics and that weird thing called "humans". ;)

I wish I had something better and less dark to offer other than "problems are everywhere", although I do have a newer long term US velocity chart that may help some. Velocity is currently around the levels of the early 1970s (the red line being the average of my three prefferred velocity measures), and that's a far cry from where they were in the 1930s. They also show significant signs of stabilizing lately after the unprecedented recent drop.



http://www.nowandfutures.com/images/velocity_preferred.png






http://www.nowandfutures.com/images/velocity_preferred_short.png




And for comparison's sake, here's my best shot so far on velocity in Argentina.

http://www.nowandfutures.com/d2/argentina_velocity1995-2005.png

bart
06-28-09, 10:55 PM
I don't know that I agree with using the stock market peaks as the starting point in both, but it is difficult to say what would be the best approach. Anyway, your main point, I believe, was to show how misleading stats can be if taken verbatim without exercising some scrutiny.

Yes, you got it and I'm also trying to make a statement about if one is talking about and comparing against the Great Depression, then the starting date used should be as close as possible to it's actual date.

Since the US depression started in 3Q 1929, it was a little bit arbitrary to use Sept 1929 but the charts change little if using July or August.
I also could have chosen December 2007 (the official recession starting date) instead of October 2007 for the current starting date but decided that since Sept 1929 was the stock market peak, I also used October 2007 for some consistency... and who knows if history will look back and change the recession starting date to October.


I am actually a bit more inclined to follow the idea that Mar 2000 is akin to Sept 29, and Oct 07 akin to Mar 37, shifted +/- a few months, depending on the data set you find most relevant to both of those economic cycles, which complicates any attempts are comparison even more. There are just so many ways to look at the two events in relation to one another.

You might want to check the other thread ( The Great Depression parallels... busted (http://www.itulip.com/forums/showthread.php?t=10477) ) since I did produce one off charts using alternate start dates - one set of which started in January 2000... and doesn't look hugely different from the ones starting 10/2007 - at least to me.


I think the main point I'm trying to make though is that now is not a clean parallel to the Great Depression especially in the monetary and fiscal areas, whether just in the US or world wide.

D-Mack
06-29-09, 09:53 AM
I don't know much about it, but I think you are also using charts with alternative numbers for GDP.

Just looking at the non-subscriber data at shadowstats, I would guess that GDP from 2007-2009 would be something like -4%

Wouldn't it be better to use the same methods they used to calculate GDP decades ago?

bart
06-29-09, 11:06 AM
I don't know much about it, but I think you are also using charts with alternative numbers for GDP.

Just looking at the non-subscriber data at shadowstats, I would guess that GDP from 2007-2009 would be something like -4%

Wouldn't it be better to use the same methods they used to calculate GDP decades ago?

Don't you know its dangerous for you to pay that much attention to facts & details? ;)

You're basically correct on GDP, the numbers being used in the charts are current dollar GDP. That's partially to deal with "normal" economists etc. who don't give much credence to John Williams work and partially to avoid the "tinfoil hat" accusations or attacks.

I did add in a note about CPI and shadowstats.com, but decided that using corrected GDP would hurt the general comparison with "normal" economists or investors, etc... while also making many go WTF and perhaps question current day GDP reporting accuracy. It's one of those damned-if-I-do and damned-if-I-don't areas... but I just did add a short note about the GDP data in the article original on my site. Thanks for the nudge.




One additional point, I just added a "total credit" for the current day period to the first chart in order to help show a fuller picture and avoid accusations of cherry picking just Comm'l bank credit... and as can be seen, total credit is still up about 4% since 10/2007. The actual current day facts on credit growth almost always produce surprise when I mention them to those who don't track the area closely and assume that total unadjusted credit creation has crashed.

grapejelly
06-29-09, 11:07 AM
huge difference between the Great D and today's D is that the dollar was convertible, and hence banks were allowed to go out of business, shrinking the money supply by defaulting on their obligations to depositors.

Today, banks are not allowed to default, money is not convertible into anything, so the comparisons between now and then in some ways are hard to make.

bart
06-29-09, 11:50 AM
huge difference between the Great D and today's D is that the dollar was convertible, and hence banks were allowed to go out of business, shrinking the money supply by defaulting on their obligations to depositors.

Today, banks are not allowed to default, money is not convertible into anything, so the comparisons between now and then in some ways are hard to make.

Indeed, and there's also this:

"Essentially, all models are wrong, but some models are useful."
-- George E.P. Box (Professor Emeritus of Statistics at the University of Wisconsin)
;)



And here's something else in the dollar value area that will likely be a surprise to many:

http://www.nowandfutures.com/d2/dollar_1900-1951%28armstrong%29.png

TheServant
06-29-09, 04:17 PM
I'll check that thread out. I believe you have done a good job in making your point, or perhaps better said, attempting to quantify it. It is certainly an interesting subject, and a difficult one to decipher precisely.

Btw, have you done a comparison with starts of fall 2008 -vs- fall 1937? It seems the amount/speed of monetary stimulus in response to the 1937-38 "crash" was perhaps more similar to the current response to the '08-'09 "crash". At the same time, this comparison may not be as valid since '37-'38 seemed to result more from the removal of stimulus -vs- the initial "sudden stop" of '08, to use the iTulip vernacular. Just curious.

bart
06-29-09, 05:21 PM
Nope, haven't done a chart set with those parameters. I think it would be of limited usefulness since data gets sparse (or gov't controlled) prior to and during WWII.

Here's something about 2000 and on vs. 1929 and on that might amuse you and others.

http://www.nowandfutures.com/images/dow1929_nasdaq2000.png

Chomsky
06-29-09, 06:15 PM
Nope, haven't done a chart set with those parameters. I think it would be of limited usefulness since data gets sparse (or gov't controlled) prior to and during WWII.

Here's something about 2000 and on vs. 1929 and on that might amuse you and others.

http://www.nowandfutures.com/images/dow1929_nasdaq2000.png



Why not stick the Nikkei c. 1990 in there as well. :)

Finster
07-02-09, 06:05 PM
I am actually a bit more inclined to follow the idea that Mar 2000 is akin to Sept 29, and Oct 07 akin to Mar 37, shifted +/- a few months, depending on the data set you find most relevant to both of those economic cycles, which complicates any attempts are comparison even more. There are just so many ways to look at the two events in relation to one another.

Me too. The 2000 and 1929 tops are analogous, as are the 2007-2008 and 1937-1938 declines. Of course that does not mean we can't find any differences. In the earlier epoch, the first stock market downdraft was the more severe, while this time around the second is more severe. That difference, in turn, can be traced at least in part to the differing monetary response. Then, the Fed was slow in responding to the first decline with stimulus; this time it was swift. So this time, part of the deflation was put off, only to return with greater vengeance when the second wave hit.

Verrocchio
07-05-09, 11:48 AM
Bart, you are a bloodhound! Thank you for the informative graphs and following your nose on the data trail of some of the most vital economic questions that we face today. :D

bart
07-05-09, 12:03 PM
Me too. The 2000 and 1929 tops are analogous, as are the 2007-2008 and 1937-1938 declines. Of course that does not mean we can't find any differences. In the earlier epoch, the first stock market downdraft was the more severe, while this time around the second is more severe. That difference, in turn, can be traced at least in part to the differing monetary response. Then, the Fed was slow in responding to the first decline with stimulus; this time it was swift. So this time, part of the deflation was put off, only to return with greater vengeance when the second wave hit.


Looks like I'll be adding them to the article along with

"This is not just a deep recession or depression"




http://www.nowandfutures.com/images/depression_comp_stack1_2000.png



http://www.nowandfutures.com/images/depression_comp_stack3_2000.png




http://www.nowandfutures.com/images/depression_comp_stack2_2000.png




http://www.nowandfutures.com/images/depression_comp_stack4_2000.png