Eh.net Book Review: The Great Crash of 1929: A Reconciliation of Theory and Evidence.

Published by EH.Net (August 2015)

Ali Kabiri, The Great Crash of 1929: A Reconciliation of Theory and Evidence. Basingstoke, UK: Palgrave Macmillan, 2014. xv + 236 pp. $115 (hardcover), ISBN: 978-1-137-37288-8.

Reviewed for EH.Net by Raphaël Hekimian, West Paris University and the Paris School of Economics, and David Le Bris, KEDGE Business School.

The New York Stock Exchange (NYSE) crash in 1929, which featured a 45 percent decline in stock prices over the last weeks of October, is one of the most studied topics in financial history, and academic researchers still fiercely debate many of its aspects. Among those, we can cite the crucial question of whether or not the stock market boom of the 1920s was justified by fundamental values. The Great Crash of 1929: A Reconciliation of Theory and Evidence aims at resolving this issue: could the rise in stock prices before the crash be seen as an asset “bubble,” and if so, could it have been anticipated?

Ali Kabiri relies on both contemporaries (Smith 1924) and more recent academic research (Shiller 1981, 2000; Goetzmann and Ibbotson 2006) to provide extensive and detailed empirical analysis. After checking some of their results with new hand-collected data, the book details a range of econometric tests and robustness checks in order to rigorously analyze ex ante and ex post stock prices movements between 1921 and 1932.

The book is organized into six chapters. After an introduction, chapter two reviews the modern literature on the financial history of the 1920s, the different types of tests used in this literature, the theory of asset “bubbles,” and the Efficiency Market Hypothesis (EMH).

Chapter 3 focuses on the historical background of the U.S. economy, providing a detailed look at debt levels in various economic sectors, in particular the housing debt held within the banking sector. In addition, Kabiri considers the Gold Standard system and the newly formed Federal Reserve and its interest rate policy, and the effect of credit expansion on U.S. corporate earnings and stock prices. The chapter ends with a look at productivity growth and expected inflation as drivers of the valuation of assets. The main results are that the first part of the boom (1921-1927) can be attributed to a credit/debt expansion coming from World War I monetary base expansion, along with an expectation of higher returns or lower risk premia on U.S. common stocks.

The fourth chapter looks at the dynamics of U.S. common stock prices from 1870 up to 2010. The author tests market efficiency with a long-term asset prices perspective, using historical data to observe both ex ante expectations and ex post realizations of stocks returns. The objective here is to test for a potential deviation from rational valuation during the second part of the boom (1927-1929) in three ways. First, at the aggregate level: the author estimates the scale of the overvaluation of stocks in September 1929. In order to replicate the 1920’s investors’ expectations, he applies a method of that time (Smith, 1924) to a set of common stocks of large firms between 1900 and 1929 to calculate the dividend growth rate; accepting the hypothesis that expectations of that time were formed according to financial theories of that time. The historical Equity Risk Premium (ERP) before 1927, taken from both recent (Goetzmann and Ibbotson 2006) and contemporary (Smith 1924) research, is used as a discount rate to solve a Dividend Discount Model. This market valuation is then compared to the actual level reached in 1929, so as to estimate the scale of the overvaluation, following the basic method of Schiller (1981). According to this method, the estimated overvaluation of U.S. common stocks is found to revolve around 50 percent. Secondly, an ex post analysis of the very long-run realized returns is run. The aim is to test if investors could have anticipated an upcoming growth in dividends in 1927, before the second phase of the boom. To do so, the author constructs a total return index (cautiously avoiding the survivorship bias) using data from 1925 up to 2010. This return is then compared with real returns of government bonds to deduct a realized equity premium, which is found to be similar to the historical ERP calculated in Smith (1924). According to the author, this indicates that high returns were not forthcoming on stocks when compared to historical figures, so perfect foresight should have prevented rational investors from expecting a higher dividend growth rate in the late 1920’s. Finally, the aviation industry, a technological sector potentially prone to overvaluation in the 1920s, is tested using a valuation model available in the Moody’s Manual of Investments (1930). The model is calibrated with historical data from 1904 to 1929 of the automobile industry’s growth path. This test implies that aviation stocks were overvalued in 1929 by around 300 percent, relative to the history of the automobile industry. It means that a rational investor in 1929 would not have held those stocks in his portfolio.

Chapter 5 investigates the role of the money market in the boom and bust of the NYSE. It is known that the Federal Reserve took measures to slow down speculation by restraining credit to banks and funds under regulation. The author argues that a sort of ‘shadow banking system” had been developing to lend to traders, as a regulatory arbitrage. Funds were coming mostly from U.S. unrestricted corporations, but also investment funds and foreign banks. In order to assess whether the crash was due to an exogenous shock stemming from a credit retraction of those unregulated sources, tests of the ratio of stock prices to credit are run. Data show that the crash was not induced by a credit contraction even if regulation arbitrage generated instability. The results are more in favor of a bubble reversal in stock prices.

Finally, chapter 6 studies stock prices movements with regards to their fundamental values but during the 1929-1932 period. According to the data, it seems as if an undervaluation occurred during the Great Contraction based on ex post analysis. The conclusion is that the low level of the market in 1932 cannot be fully explained by rational forecasts, meaning that irrational pessimism probably took place.

Ali Kabiri’s book provides a synthesis of the debates on the 1929 crash but also a new set of tests built on both existing and newly collected data to understand which forces drove the stock market to levels reached in the 1920s, based on both ex-ante and ex-post analysis. In financial history, the book provides new insights on how investors could have valued stocks with respect to available information and, this is an important hypothesis, methods at the time.  He also finds evidence of the deviation in prices based on ex post fundamental values. In addition, the book contributes to behavioral economics, estimating the rationality of the rise and fall in stock prices during the boom and bust as well as to history of economic thought, detailing financial theories and methods of the 1920s.

Raphaël Hekimian is a Ph.D. student at the West Paris University and the Paris School of Economics. He is also research assistant for DFIH, a database project collecting and keying financial historical data on the Paris Stock Exchange. David Le Bris is an assistant professor at KEDGE Business School. He has written several articles about the history of the French financial market. Using daily data, they work together examining the absence of any contagion of the 1929 U.S. crash to the French stock market.

Copyright (c) 2015 by EH.Net. All rights reserved. This work may be copied for non-profit educational uses if proper credit is given to the author and the list. For other permission, please contact the EH.Net Administrator (administrator@eh.net). Published by EH.Net (August 2015). All EH.Net reviews are archived at http://eh.net/book-reviews/