State Space Models

All state space models are written and estimated in the R programming language. The models are available here with instructions and R procedures for manipulating the models here here.

Monday, August 18, 2014

War and 'Normalcy': 1914-29


Peter Fearon has written a great book on the US Great Depression titled War, Prosperity & Depression: The U.S. Economy 1917-45. It's an excellent summary of the literature up to 1987, when the book was written. Prof. Fearon is a British historian. He writes from a very broad perspective and is not wedded to any one pet theory of the US Great Depression, as are many US economists. If we're going to understand the Inter-War years and the Economic Bubble Machine, Fearon's book is a great place to start. I'm going to go through every chapter in the book in upcoming posts and compare results from my own State Space models of the period. The intent is to see whether the theory of Complex Dissipative Systems will provide any insight into the causes of the US Great Depression.

Fearon starts out Chapter 1 by describing growth from 1870 as having been strong enough to make the US the "world's leading industrial power" in 1914 on the eve of WWI. After WWII, the US would clearly be the dominant world power. To understand this period of hegemonic succession (Great Britain was the dominant power in the 19th Century being challenged in Europe by Germany), we will have to, at some point, look back at the 19th Century. For now, it is enough to say that the InterWar period was chaotic and that the future dominance of the US was not preordained. The Great Depression could well have crippled the US. Two Wars had to be fought to decide that Germany would not succeed Great Britain as the dominant power in Europe. The Wars did not have to turn out the way they did. There will be no simple, monocausal explanation for the period; randomness and uncertainty will always remain dominant factors.

On the eve of WWI, Fearon lists a number of attributes of the US Economy: growing monopoly power, spread of mass production methods, vertical integration of firms, the ascendancy of professional managers, the use of scientific management (Taylorism), the growth of new technologies, economic specialization (division of labor), standardization and use of the continuous assembly line (Fordism), urban expansion, growth of retailing, and a large agricultural sector producing surplus output. In summary "...the US was a vigorous, high wage economy, capable of either taking European inventions, such as the automobile, and adapting them to the American market or exploiting to the full the ideas of her own people" (p. 6).

The outbreak of WWI, however, was a shock to the US economy. European markets were closed. War financing was needed. Workers had to be shifted from agriculture to industry and labor shortages developed. The War created inflationary pressures and prices began to rise. In the graph at the top of this page, the US Wholesale Price Index (P.WPI.) is plotted from 1913-1923 (Fearon presents price data for six industrial commodities on page 10). The dashed red line in the figure is the attractor path for P.WPI. From the start of WWI well into the 1920, the price bubble is quite evident. However, in 1923 wholesale prices had returned to the dynamic attractor path.


The attractor path for P.WPI. is being driven by the overall price level (P.GDP.) which is plotted above along with it's attractor path. The attractor path for overall inflation was stable and was driven by Gross Domestic Product. The inflationary bubble popped in 1920 but inflation was still not on the attractor path by 1923. Fearon (page 11) seems to attribute the continuing inflationary pressures to the US Federal Reserve, which had been founded in 1913 and was still pushing expansionary monetary policy in 1919. What is somewhat interesting about the period is that many regulatory agencies were founded to deal with War mobilization and the War Industries Board (WIB) did have "wide-ranging powers to set priorities and even fix prices," but seemed to have little actual impact on controlling inflation.


Fearon devotes a section of Chapter 1 (starting on page 15) to the Post-War Boom and Bust which can be seen clearly from the time series plot above. The attractor path (dashed red line) is being driven by the state of the US economy generated by the US20E model (a simple Kaya Identity model, all in real variables). The red and the green dashed lines are the upper- and lower-98% bootstrap prediction intervals for GNP. What is interesting about the attractor path is that the US economy under-performed during the years before WWI and really did not start to take off until 1915 (World War I lasted from 1914-1918) a year after the war had started. After 1917, the US economy moved into highly improbable territory and was not on the attractor path even after the Post-War Bust.

The Post-War bust is conventionally described as a recession made worse by inept monetary policy:

There can be little doubt that the actions of the Federal Reserve made the ensuing recession a good deal worse. The restriction in credit hit business generally but especially the construction industry and also automobile sales, which depended upon the ability of consumers to borrow cheaply. In addition, commodity dealers and speculators panicked when faced with the rising cost of borrowing and quickly unloaded their stocks, causing massive price falls. (Fearon, 1987: 17)

The argument is based simple straight-line extrapolation from historical data points. In the graphic above, a straight-line extrapolation from some historical data point defines the "slump" afterwards. Lines A and C make the slump look particularly "severe" while lines B and D a little less so. Each extrapolation line assumes that some historical trend (however short) can be maintained into the future. Taking a longer perspective and deriving a model-based attractor path gives a different picture, a return to normalcy after WWI (in Warren G. Harding's words) that was never really completed (if "normalcy" means the dynamic attractor path which was not Harding's definition of "normalcy").



Over the entire Roaring Twenties, as seen from the GNP attractor plot from 1919-1930 above, the US Economy never really returned to the attractor path until the start of the Great Depression. And, rather than a soft landing, the economy crashed. Partly, these are definitional issues. A recession should be defined as a drop below the lower 98% confidence interval not just any drop in production from some level. Partly, these are modeling issues. Any historical GNP point reflects the sum of prior "errors" in the model. GNP(t3) = f[(GNP(t2) + E(t2)+GNP(t1) + E(t1)) + E(t3) while the attractor path starts at (t1) and moves forward without including modeling errors. If the errors are predominantly positive, as they were from 1914 to 1919, we have an Economic Bubble (the Roaring Twenties). The Age of Prosperity, as the 20's was also called, really was a bubble and was not prosperous for farmers, working-class laborers, African Americans and other minorities.



What is also unusual after WWI is that there was no unemployment crisis even though the economy was in "recession". The time series plot for the US Labor Force (L.US.E. above) shows that employment was below the attractor path until 1923. After 1924 until the start of the Great Depression, employment was actually above the dynamic attractor path. Fearon presents data (page 24) to show that the jobs were created mostly in the Wholesale and Retail sectors in response to a strong period of consumption expenditure. What is important to point out here is that Economic Bubbles create employment bubbles. The "bubble" jobs are lost when the bubble pops and will not return unless the bubble is reflated.

In Chapter 1, Fearon brings up many issues in passing that, hopefully will be dealt with in future chapters: (1) the impact of the New Deal vs. WWII, (2) the growth of Inequality, (3) the role of economic growth and competition in the 19th Century, (4) Weakness in Agriculture, (5) the role of Urbanization in expansion of the internal market, (6) the impacts of Federal Reserve "stop-and-go" monetary policy, (7) Price inflation and deflation, (8) Bank Failures and financial markets, (9) Stock Market speculation, (10) the end of European Immigration during the Inter-War years and (11) Crisis in the Housing Market. All of these issues and the analysis above should ring bells for those of us who have lived through the current Great Recession and Financial Crisis of 2007-2008. The difficult task will be to disentangle what is common and what is different about the two periods.