Has equity always earned a premium? Evidence from nineteenth-century Britain

Financial economists have become increasingly interested in the historical returns of financial assets.1 This interest largely stems from a desire to calculate the expected equity risk premium. In particular, academics and practitioners are interested in discovering whether or not the high returns on stock markets over the past half-century are an aberration or are somehow intrinsic to equity as an asset. Recently, William Goetzmann and colleagues have estimated the performance of the US market in the nineteenth century, finding that a sizable premium existed on equity in that period.2

Although previous studies have analysed the performance of the British equity market in the twentieth and late nineteenth centuries3, no studies examine the performance of this market in its formative period.4 In a recent conference paper, we estimate the returns earned by investors in the nineteenth century by assembling a monthly dataset of stock prices for the London market from a stockbroker list for the years 1825-70.5

This period is particularly important because it contained three major transformations in wider society that had substantial consequences for the equity market. First, the growing and increasingly prosperous middle class increased their demand for alternative investments to low-yielding government debt. Second, Parliament increasingly bypassed the conservative common law and legislated to ease the creation of corporations. Third, the development of the steam locomotive resulted in capital-intensive industries requiring substantial injections of equity capital.

Stock market returns in the nineteenth century

In our indices, there are 1,015 common equity securities for 681 companies, with banks, insurance companies, canals, mines and railways constituting approximately 80% of our indices in terms of market capitalisation. The main findings are as follows:

• The annual average total return weighted by market capitalisation was 12.1%. Omitting the railways, the dominant sector at the time, has little effect on this figure.
•  The unweighted annual average total return was 16.9%, which was higher than the weighted return, suggesting that small firms earned a premium.
• If a person invested £100 in the equivalent of an index tracker fund in 1825, it would have been worth £16,684 by 1870.
• Unlike modern investment performance, most of the return for nineteenth-century investors came from dividends rather than capital appreciation. Depending on the weighting methodology used, dividends constituted between 63% and 70% of total annual returns. Whereas in the second half of the twentieth century, dividends only constitute 31% of total returns.
• The stock market experienced negative total returns in only four years between 1825 and 1870. Three of these years (1825, 1826 and 1847) had financial crashes after a period of promotional mania on the stock market. The negative returns in 1853 can be attributed to concerns over the impending Crimean war.
• Despite the serious financial crisis of 1866 following the collapse of several banks, the market produced positive total returns in that year.

The above estimates of stock returns overestimate the total earned by investors, as they do not take account of companies that entered bankruptcy, where shareholders lost some or all of their capital. We therefore adjusted our stock returns for survivorship bias by making the stringent assumption that all other delisting companies that had existed for longer than three years delisted because of bankruptcy and shareholders of those firms did not have anything returned to them.6

Assuming that investors endure a negative 100% return on the month of delisting reduces our estimate of total returns weighted by market capitalisation from 12.1% to 10.0%. In addition, taking account of survivorship bias increases the share of the dividend component in total returns.

High returns

When the real returns for the 1825-70 period are compared with the United States in the same period and with later epochs in the British market, it appears that returns in this formative period of the British market are relatively high. Although differences in construction, constituents, and survivorship bias may explain some of this difference, it is unlikely to explain it all. Furthermore, using crude risk measures, it appears that the British market was less risky in this period than in subsequent periods.

Another possibility as to why total real returns are lower in the latter periods is that the stock market simply reflected Britain’s industrial hegemony and its subsequent decline. As is well known, 1825-70 was the period when Britain was at the peak of its industrial power, enjoying above market rates of return due to it being the first industrial nation. Also, as a result of imperial expansion, companies enjoyed high profit streams as demonstrated by high dividend payments.

In contrast, the period after 1870 was one where Britain increasingly faced competition from imperial rivals, and this could explain the relative performance of the equity market in this period. However, the high rates of return for investors in the period 1825-70 translate into high costs of capital for companies, and in an efficient capital market, such high returns should be competed away.

This implies that there may have been inefficiencies or barriers to entry in the capital market at this time. The list of possible market imperfections includes:

• Factors that made portfolio diversification costly, such as high share denominations, unlimited liability and uncalled capital.
• Almost non-existent investor protection laws and poor corporate governance.
• Highly illiquid financial markets.
• In the period 1825-70, large sums of capital were tied up in land. But the large movement of funds out of land and into equity after the 1880s may have resulted in the substantial fall in the cost of capital.

Lessons for the present

In the light of historical experience, it would appear that high returns are an intrinsic part of equity. But will current woes in the financial system result in a fall in the equity premium?

The historical experience provides some insight into this. Despite the financial crisis of 1825-26, the dramatic collapse of the railway mania in the late 1840s and the severe financial crisis of 1866, returns in the 1825-70 period were high. The stock market proved robust to these temporary setbacks, which in many senses were more severe than what the financial system is experiencing today.

The historical experience also sheds light on the impact on equity returns of a wholesale shift of funds into the stock market. The movement of funds out of land and into equity in the late nineteenth century appears to have contributed to reduced returns for investors. Will the current fall in the property market result in lower stock returns due to substantial funds shifting into the stock market?

Finally, unlike modern-day investors, nineteenth-century investors earned most of their return from dividends rather than capital appreciation. This raises the question as to why dividends have disappeared in the last three decades.7 Possibilities include the changing characteristics of firms, increased use of stock repurchases, improvements in investor protection law, and better alignment of managerial interests with those of shareholders.

Footnotes:

1 See, for example, Dimson, E., Marsh, P. R., and Staunton M. (2002). Triumph of the Optimists: 101 Years of Global Investment Returns, Princeton, NJ: Princeton University Press; Goetzmann, W. N and Ibbotson, R. G. (2006). The Equity Risk Premium: Essays and Explorations, New York: Oxford University Press

2 Goetzmann, W. N, Ibbotson, R. G. and Peng, L. (2000). ‘A New Historical Database for the NYSE 1815 to 1925: Performance and Predictability’, Journal of Financial Markets, vol. 4, pp.1-32

3 Dimson, E. and Marsh, P. (2001). ‘U.K. Financial Market Returns, 1955-2000’, Journal of Business, vol. 74, pp.1-31; Grossman, R. S. (2002). ‘New Indices of British Equity Prices, 1870-1913’, Journal of Economic History, vol. 62, pp.121-146.

4 Price indices do exist for this period, but they ignore dividends, are effectively unweighted, and are based on small samples of stocks. See Gayer, A. D., Rostow, W. W. and Jacobson-Schwartz, A. (1953). The Growth and Fluctuation of the British Economy, Oxford: Clarendon Press.

5 Acheson, G. G., Hickson, C. R., Turner, J. D. and Ye, Q. (2008). ‘Rule Britannia!: British Stock Market Returns, 1825-1870’, Paper presented at Economic History Society Conference, Nottingham. Available at http://www.ehs.org.uk/ehs/conference2008/Assets/AchesonFullPaper.pdf This research was supported financially by the ESRC (RES-000-22-1391).

6 Although our stockbroker lists don’t inform us as to why firms delisted, we were able to use other sources to help us identify the railways which delisted because of merger or firms which moved their listing from the London exchange to a regional exchange.

7 On this, see Fama, E. F. and French, K. R. (2001). ‘Disappearing Dividends: Changing Firm Characteristics or Lower Propensity to Pay’, Journal of Financial Economics, vol. 60, pp.3-43