![Sample our Humanities journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5939/TOC-Subject-Sample-2021-Humanities.jpg"})
ABSTRACT
There are a number of reasons why investor portfolio characteristics are of interest. First, there is limited evidence of what individual investors actually held in their portfolios in the past, including, for example, whether there were significant differences between male and female investors. Second, investors’ portfolio holdings are relevant to the debate on the ‘democratisation’ of investment and, third, they inform the debate on whether investors in the past made efforts to reduce portfolio risk through diversification, before the full ‘scientific’ approach of the early-twentieth century and the Markowitz optimisation approach of the mid-twentieth century. This research explores the portfolio choices made by a sample of 508 investors – 263 men and 245 women – between 1870 and 1902. Evidence of diversification exists, with the average holding of the sample being 4.6 securities. There is also evidence of increasing levels of diversification over time, of international diversification, and greater diversification by wealthy men and women. Investors in the past clearly made efforts to reduce portfolio risk before Markowitz optimisation.
Acknowledgements
We gratefully acknowledge the support of the collaborators on this project − David Green, Josephine Maltby and Alastair Owens − as well as research assistants Stephen Ainscough, Carien van Mourik and Claire Swan.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1. Most of this work has been on the UK rather than the USA where lack of access to shareholder records hinders research in this area. Similar research is also progressing on other countries. For work on Swedish, Canadian, Australian, Canadian and Italian shareholders, see Laurence, Maltby, and Rutterford (Citation2009) and Green et al. (Citation2011).
2. See, for example, Goetzmann and Ukhov (Citation2006), Chabot and Kurz (Citation2010), Edlinger, Merli, and Parent (Citation2013) and Scott (Citation2002).
3. A trust is an arrangement whereby a person (trustee) holds property as its nominal owner for the good of one or more beneficiaries. In this instance, trusts were often set up for widows and children, on the death of the husband. Prior to 1893, trustees who were restricted to investing only in so-called ‘trustee investments’ could only purchase Consols. The Trustee Act of 1893 allowed trustees to purchase safe British and colonial government stocks, in particular those of India, UK and Indian Railway debentures and some ‘safe’ railway preference shares, as well as Bank of England and Bank of Ireland stock.
4. This categorisation is not explicit in the texts of the period, but it is implied by them. For further discussion, see Rutterford (Citation2004) on how yields were used as a valuation tool to take account of risk.
5. Another way for investors to improve information flow was to live close to the company’s headquarters, area of operations and/or location of annual general meetings. For more discussion on local investment bias at the time, see Rutterford, Sotiropoulos, and van Lieshout (Citation2015).
6. IR19 Board of Stamps: Legacy Duty Office and successors: Specimens of Death Duty Account, 1796–1903, The National Archives, Kew, UK. The data were collected as part of ESRC research project RES-000-23-1435. For further discussion and analysis of this data sample, see Green et al. (Citation2011). See also Green et al. (Citation2007).
7. For the sample of 1276 of the 1446 Residuary Accounts for which it was possible to establish age at death, the average age for males was 60.3 and for females 64.4. See Green et al. (Citation2011).
8. For further discussion of this point, see Green et al. (Citation2011).
9. This figure reflects the total number of records preserved in the IR19 series which ends in 1902. This was a small proportion of the total actually submitted, the remainder having been destroyed. However, the sample itself appears to be representative of the broader population from which it was drawn.
10. The MWPA of 1870 did not allow married women to own real estate in their own name. Married women also required their husbands’ permission to own assets separately. It was not until the MWPA of 1882 that married women acquired the same rights as single women and men with respect to owning financial assets (Rutterford and Maltby Citation2006, 115–116).
11. Average life expectancy at birth for both men and women born in England and Wales was still less than 50 by 1900. However, these figures reflect high death rates in childhood. See Longevity Science Advisory Panel (Citation2012), Paper 2, page 8, Figure 1.
12. During this period, individuals had to save for retirement, especially as ill health might prevent working until death. The aim was to live off income and not sell off assets or securities if at all possible. See Morris (Citation2005).
13. Note that the probate definitions say ‘shares’ for some categories. They include, however, all types of corporate security, such as debentures, loan stock, preference shares and ordinary shares.
14. Although canal and shipping shares were not important sectors by the end of the nineteenth century, shows women with higher average shareholdings in these sectors in the 1870s. For further discussion of women’s investments in ships and canals, see, respectively, Doe (Citation2009) and Hudson (Citation2001).
15. Indeed, Lowenfeld went further and recommended equal amount in uncorrelated securities, so that losses in one security would be balanced by gains in another (Lowenfeld Citation1911, 79–87).
16. These differences highlight the variation in optimal portfolio weights which small changes in data inputs in a Markowitz optimisation can imply. The figures quoted do not allow short selling and also require no more than seven asset classes in any optimal portfolio. The variation in results is, in part, due to the stellar performance of world railway debt securities during the period, leading to an optimal percentage of around 50% in such securities using actual returns in the optimisation process. See Goetzmann and Ukhov (Citation2006, 290–296).