Output, Stock Volatility, and Political Uncertainty in a Natural Experiment- Article Review

Academic Research Insight:

George Bittlingmayer, “Output, Stock Volatility, and Political Uncertainty in a Natural Experiment: Germany, 1880–1940” (Published in The Journal of Finance, Vol. 53, No. 6 (Dec., 1998), pp.2243-2257)
Published by:   Blackwell Publishing for the American Finance Association).

- Article Review by Dipendra Karki

 

I.                  Research Background:

a. The theory of investment under uncertainty implies that political uncertainty may simultaneously increase volatility and reduce output.

b.       Schwert (1989b) points out  in  a  more  general context, we  can  easily construct links  between volatility and  the  business cycle,  but  causation is hard to pin  down.

c.        Political uncertainty may represent an exogenous factor.

d.       Until the start of World War I in 1914, Imperial Germany had a stable, limited government.

e.       Weimar Germany after the  war was  marked by  revolution;

                                                               i.      Unstable  republic

                                                              ii.      Insurrection

                                                             iii.      The   rise of anti-capitalist parties on  the  left and  right

                                                            iv.      Hyperinflation

                                                              v.      Protracted struggle over  reparations

f.         Economic historian Gerald Feldman (1993) calls  the  period from  1914  to 1924  “ The  Great Disorder.”

II.                Purpose:  The purpose of this study is to examine at stock volatility and output in a case marked by a clear exogenous political shock.

III.              How this paper differs from its predecessors;

- This paper adopts a different strategy and examines a dramatic political shift. In contrast to the United States, where the sources of volatility are subtle or controversial.

IV.             Importance of the Study: The study is quite important in the given research context because it aims to bridge the gap between empirical research works and real research evidences of stock volatility and output in a case marked by a clear exogenous political shock.

V.                   Literature Review:

a.       The Authors have made number of related literature review.

b.       Huang and  Kracaw (1984), Schwert (1989a), Romer(1990), Pindyck  (1991a): the  causal link  between volatility and  business slumps is unclear. Slumps may cause volatility, volatility may cause slumps, or both  may  be  the  consequence of some  other, more  clearly exogenous factor.

c.        Pindyck’s  (1991b) review of the  literature on  uncertainty and   irreversible investment also  notes that  political uncertainty may  depress investment.

d.       Fama (1990) and Schwert (1990) show that real and financial factors explain about half  of the  variance in annual returns.

VI.                 Data Collection:

a.       Output: For the period from 1880 through1913, the study uses the production index from Hoffmann (1965). This  series is based on a variety of industries covering metals, metal working, chemicals, textiles, food, utilities, and  construction.

- For the period from 1914 to1924,  The study uses the esti- mates of national income from  Witt (1974). Witt’s  estimates are  based on tax  information.

- For   1925   to 1940,   it uses   the   industrial production series in Petzina, Abelshauser, and Faust (1978).

b.       Stock Index: The  stock  price  index stems from  Gielen (1994),  who  brings together information from  various sources to construct a real, dividend-adjusted series for the  period 1870 to 1993. 

-For the  years up to 1913,  the  series is composed of a  simple average of 20 large stocks until 1889.  Starting in  1890,  the series becomes a  capitalization-weighted average of 48 stocks. The series was  further expanded to 71 stocks in  1905. 

VII.               Methodology:

a.       The transition from  Imperial to Weimar Germany offers  a natural experiment.

b.       Author makes three contributions to the  study of stock  volatility:

                                             i.      The  switch in  regimes throws light on  the  sources of volatility: Stock  volatility was  low before  1914,  it increased markedly after the  war.

                                            ii.      The author use the  theory of investment under uncertainty to  build  a  natural  connection between political uncertainty,  stock volatility, and   output.

                                           iii.      Empirical results for Germany confirm  earlier work  for the  United States; volatility increases before  and  during business slumps. 

Figure 1. German political events, log  of  stock prices, and log  volatility, monthly intervals, 1890–1940. The  figure shows  selected political events affecting Germany, the  natural log  of real, dividend-adjusted German stock  prices (DD) in  the left scae, and monthly returns (D D D) are shown on the right scae. Both  series are  shown at monthly intervals for January 1890 through December 1940.  Data are from Gielen (1994) [Source: article by author]

c.        Use of graph is well predicted with Summary Statistics for  German Stock Returns and Changes in Output, finding mean, geometric mean and standard deviation.

d.       Regression analysis is done taking lagged changes in log volatility and introducing dummy variables of deflation and world war I.

e.       See Table I: Summary Statistics for  German Stock Returns and Changes in Output, 1880–1940.

f.         See Table II: Regression of changes in industrial production on current and  lagged changes in  volatility and changes in stock prices:

ln (IP_t ) - ln(IP_t-1) = a + b₁ [ln(SD_t) - ln(SD_t-1)] + b₂ [ln(SD_t-1) – ln(SD_t-2)] + b₃ [ln(S_t) – ln(S_t-1)] + e_t 

g.       See Table III: Regression of Annual Percentage Changes in Production on  Changes in Current and Lagged Log  Volatility, the Rate of  Deflation, a World War I Dummy, and Stock Returns, 1880–1940

ln (IP_t ) - ln(IP_t-1) = a + b₁ [ln(SD_t) - ln(SD_t-1)] + b₂ [ln(SD_t-1) – ln(SD_t-2)] +b3DEFLAT_t +b4WWI_t

          + b₅ [ln(S_t) – ln(S_t-1)] + e_t 

VIII.              Major Findings: See fig. 1, 2 & 3  and  table I, II & III for detail.

a.       German stock  prices experienced a one-time down- ward shift in  the  1914  to  1920  period, coincidental with the  war  and   its consequences.

b.       Stock   returns  became more   volatile temporarily  at the  beginning of the  war,  then especially at the  end  of the  war  and  in  the early and  mid-1920s, and  again in  the  early 1930s.

c.        The  s.d of monthly returns was  2.46 percent for 1880 to 1913,  17.26  percent for 1914 to 1923  (World War I through the  hyperinflation of 1923),  and  7.04  percent for 1924  to 1940.

d.       The  geometric mean monthly real return  was  0.45 percent for the  period 1880  to 1913.  It declined to 21.18 percent for 1914  to 1923,  and  increased to 0.71 percent for 1924  to 1940. 

e.       Stock  volatility increased dramatically after 1913.

IX.                 Limitations of the study:

a.       Making regression analysis, no test is mentioned regarding the model specification error, Normality test for residuals, Heteroscedasticity, Multicollinearity and Autocorrealation problem, if any in the model.

X.                   Further scope of the study:

a.       Future work  on  Germany would  benefit from  daily   stock   prices and   at least quarterly if not  monthly output series.

b.       Future work  on the  sources of stock  volatility in general might benefit from  the  study of other “natural experiments”—other instances in which  the  political system and  the  stock  market experienced large, clearly identifiable   exogenous shocks.

XI.                 Conclusion/Implication:

a.       The study explains the sources of stock volatility: Why was  volatility high  during the  Great Depression? Even for the Great Depression—the increase in German volatility in  the  late teens and  early and  mid-twenties seems closely linked to the  shift  from  ascendant empire to beleaguered republic.

b.       The  start of World  War I, the  1918 Armistice, and  the  political turbulence of 1920  each pushed the  German stock  market down  and  volatility up.  Fluctuating hopes for a  solution to  the  reparations  problem, and  the  occupation of the  Ruhr and  the  hyperinflation of 1923  were  also  associated with large stock  market movements. Finally, the  political stabilization  of Germany in  the  mid  and late 1920s  was  accompanied by steadily declining volatility.

_______------------****------------________