![Sample our Economics, Finance,Business & Industry journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5931/TOC-Subject-Sample-2021-Economics-Finance-Business-Industry.jpg"})
Abstract
In this article, we propose an original analysis of advice given by financial analysts prior to WW1. Our article focuses on the writings of A. Neymarck, one of the most popular French analysts in the early 20th Century. The creation of portfolios from a new database composed of the monthly returns of all the security types listed on the official Paris Stock Exchange from 1903 to 1912 has provided results demonstrating that Neymarck correctly identified the risk in a number of sectors. The performances of these portfolios, which were built according to Neymarck’s guidelines, confirm Neymarck’s ranking in terms of both risk and return: the richer the investor, the riskier and the more profitable his portfolio was seen to be. Finally, the Modern Portfolio Theory enables us to pinpoint the few imperfections in Neymarck’s advice, which globally appears to be driven by reliable financial analysis.
Acknowledgements
We wish to thank Jan Annaert, Hugh Rockoff and Patrick Roger, who all provided helpful comments on a former version of this paper. We also thank the anonymous referees and the editors of this journal.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes on contributors
Maxime Merli is Professor of Finance at the University of Strasbourg and an affiliated Professor at EM Strasbourg Business School. His research interests are in behavioral finance. Maxime Merli has published numerous articles on the behavior of individual and professional investors. His interdisciplinary approach has allowed him to publish articles on the rationality of investors behaviors in a historical perspective.
Antoine Parent is Affiliate Researcher at OFCE (Sciences Po), Professor of Economics at Sciences Po Lyon and a member of the LAET CNRS research laboratory. He is the founding Director of the Team CAC ‘Cliometrics and Complexity’ that he launched in September 2014 (hosted by the Complex Systems Institute (IXXI) of the Ecole Normale Supérieure de Lyon. With the creation of the Team CAC, Antoine Parent aims at fostering interdisciplinary collaboration in the modeling of long historical macroeconomic and financial data and at stimulating new approaches to economic history by drawing inspiration from other disciplines, most notably complex systems modeling in Physics, Econophysics, Mathematics, Signal Processing and Data Analytics.
Cecile Edlinger has done doctorate in finance is now working as a Senior Risk Management Analyst at FundRock Group, Luxembourg.
Notes
1 Bignon and Miscio, (Citation2010, p. 397) recall that Neymarck advised investors to check the identity and interests of the management and journalists of a journal before trusting it.
2 For an analysis of investor financial portfolio strategies in England during the second half of the nineteenth century, see Sotiropoulos and Rutterford (Citation2018) and Rutterford, Sotiropoulos, & van Lieshout (Citation2017).
3 According to Neymarck, (Citation1913, p. 348), ‘For risk division to be effective there must be several placements of different types: national or local government loans, railway stock, various industrial companies and so on […] but this is not enough […] the diversification and decrease of risk should be carried out by distributing investments across stock from various countries’.
4 For instance, he describes the ‘stock for additional income’ as ‘exposed to more vagaries than […] sound investments’, and considers that they could boost the average income of a portfolio because they are ‘likely to see capital gains’ (1906, pp. 75, 81–82).
5 A more detailed description of this advice can be found in the second part of the paper.
6 Arbulu, Citation1998; Parent & Rault, Citation2004; Gallais-Hamono, Citation2007; Vaslin Citation2007; Rezaee, Citation2010; Lebris & Hautcoeur, Citation2010.
7 Paul Leroy-Beaulieu, who became a Professor at the Collège de France, published many books dealing with economics that revealed the concerns of French society prior to World War One. He also founded his own journal, L’Économiste Français, in 1873. He managed it until his death, supplying the readers of the journal with articles and information about economics, business and finance. Paul Leroy-Beaulieu presented and developed his advice in a successful handbook entitled L’Art de Placer et Gérer sa Fortune (1906).
8 We decided against using the Cours Authentique, published every day by the Compagnie des Agents de Change, because we assumed that this daily publication was designed for professionals operating on the Paris Stock Exchange rather than for ordinary investors. In addition, not all securities were listed on a daily basis. For example, a third of the available securities were not exchanged on July 1st 1912, and were therefore not quoted in the edition of Cours Authentique published that day.
9 Note that although Neymarck suggests investing 10% of funds in industrial French and foreign bonds, he does not add foreign industrial bonds to the portfolio of investors with large fortune, despite it is the largest portfolio. Edlinger and Parent (Citation2014) therefore consider that he advises well-off investors to invest 10% in industrial French bonds alone.
10 «At the outset, can we ever be certain that we will choose the best assets? It is difficult to label assets as goods and rank them with numbers so that people can say, «buy N° 1 assets rather than N° 2 assets. A given asset may be the best one at one moment in time, but not at another. For example, a state fund may be the best asset to acquire when an event makes the price fall to a moderate level, yet it had not been the best value asset on the previous day. We can therefore say that an asset is good, very good, or excellent, but we cannot affirm that it is absolutely the best » (Neymarck, Citation1913, p. 368).
11 As a major part of our analysis is run using indices, this choice has no impact on our main conclusions.
12 See Table 4.
13 Note that asset category 13 is the only index made by just one security. This asset category weights 2% of one portfolio advised by Neymarck and is never included in the optimal portfolio. The results are not modified by the presence or absence of this category.
14 The bank shares index (18) is particularly interesting for the portfolio, since it shows very little additional risk for much higher return.
15 Note that the out-of-sample performance of this strategy is clearly debated. DeMiguel, Garlappi, and Uppal (2009) conclude that the out-of-sample performance of the sample-based mean-variance model (and its extensions designed to reduce estimation errors) is not consistently better than the 1/N rule ratio (called naïve diversification) in terms of Sharpe ratio. For a comparison between buy-and-hold and naïve strategies before WW1, see also Sotiropoulos and Rutterford (Citation2018). This is tested in the next section.
16 ‘Toutes les valeurs, sans aucune exception, présentent des risques ; l’expression valeurs de tout repos, valeurs de toute sécurité n’est qu’une expression; en réalité, les placements mobiliers de même que les placements immobiliers, présentent des aléas…’.
17 Over the period considered (1903–1912), we identified four securities that were less risky than the 3% French Rentes: one Paris city bond and three bonds issued by the Crédit Foncier.
18 ‘Ce livre ne s’adresse pas aux professionnels de la Bourse’ (Neymarck, Citation1913, p. V).
19 Our main conclusions are not modified by this choice.
20 This choice of distribution has very little effect on our results.
21 Note that by extending our data from January 1913 to July 1914, the results dealing with the ex-post performance of Neymarck’s advice against the performance of the mean variance optimization over two periods (one year and eighteen months) are mixed. The results show that the advice of Neymarck outperform the optimization for two of the four investor categories (small investors and middle-class investors) and that the ex-post returns are very close for a third category (well-off investors). For instance, over a horizon of 18 months, the Sharpe ratio associated to the portfolio advised by Neymarck for the 4 categories of investors was 0.89, 1,08, 1,58 and 1.22, respectively. For the mean-variance portfolio, these values were 0.75, 1.02, 1,61 and 1.92, respectively.