Hostname: page-component-cd9895bd7-jkksz Total loading time: 0 Render date: 2024-12-26T03:40:21.383Z Has data issue: false hasContentIssue false
  • English
  • Français

The Anglo-German Trade Rivalry, 1887-1913: A Counterfactual Outcome and its Implications

Published online by Cambridge University Press:  04 January 2016

Hugh Neuburger  and
Houston H. Stokes
Hugh Neuburger
Affiliation:
Department of History, Columbia University
Houston H. Stokes
Affiliation:
Department of Economics, University of Illinois at Chicago Circle
Get access Rights & Permissions [Opens in a new window]

Extract

We are probably fools not to find a reason for declaring war on Germany before she builds too many ships and takes away our trade.

Arthur Balfour 1907.

Projecting economic variables is a difficult art but one which statesmen practice with relish. In parliamentary democracies it is a rare candidate who does not forecast that under his tenure the pots of the electors will be graced with chickens. A more sober, careful kind of forecasting is required in conducing foreign affairs. Sound judgments about tariffs, subsidies, military and naval force levels, and even alliances depend upon prior analysis of the effects of each such policy over some specified length of run. Although one suspects that some kind of forecasting was always a part of statecraft, industrialization in Western Europe necessitated a change in the time frame of the predictions of statesmen. As the pace of economic change quickened, the rate at which the economic and military potential of states might change also accelerated. For many Englishmen observing the growth of the German economy during the 1890s, learning this lesson produced more than a little anxiety. So great was the attention which the British gave to German economic growth that predictions of Germany’s overtaking of Britain came to be a staple item of British politics in the two decades before World War One. Should these predictions have been taken seriously by contemporaries? The best answer is that given by the outcome of the pre-1914 Anglo-German rivalry: Germany did not overtake Britain in that era. Because war intervened, we cannot know the outcome of continued peaceful economic rivalry, but by using a relatively new statistical method we can chart the course this rivalry may well have followed had the peace been kept in 1914. Our projections suggest a sequence of events that never occurred but may nonetheless be used as a guide in evaluating the views and expectations of contemporaries.

Type
Research Notes
Information
Social Science History , Volume 3 , Issue 2 , Winter 1979 , pp. 187 - 201
Copyright
Copyright © Social Science History Association 1979 

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

Notes

1 See: Kindleberger, Charles P., “Germany’s Overtaking of England, 1806-1914,” Weltwirtschaftliches Archiv (1975)Google Scholar.

2 Hardach, K. W., “Anglomanie und Anglophie wahrend der industriellen Revolution in Deutschland,” Schmollers Jahrbuch für Wirtschafts und Sozialwissenschaften, 91 (1971)Google Scholar and Tilly, R. H., “Los von England: Probleme des Nationalismus in der deutschen Wirtschaftsgeschichte,” Zeitschrift für die gesamte Staatswissenschaft, 124 (1968)Google Scholar.

3 The rise of powerful United States producers was, of course, also a source of competitive pressure. For a good discussion of the emergence of foreign competition to British industry see: Aldcroft, Derek H., ed. The Development of British Industry and Foreign Competition 1875-1914 (Toronto, 1968)CrossRefGoogle Scholar.

4 Hoffman, Ross J. S., Great Britain and the German Trade Rivalry 1875-1914 (New York, 1964)Google Scholar.

5 Banze, Angelike, Die deutsche-englische Wirtschaftsrivalität (Berlin, 1935), 96Google Scholar.

6 Hoffman, Great Britain and the German Trade Rivalry 1875-1914, 303.

7 Hofrichter, Anton, Krieg undHandelsrivalitat (Berlin, 1917)Google Scholar; Seligman, Edwin R. A., An Economic Interpretation of the War (New York, 1915)Google Scholar.

8 Our counterfactual result is a modest one in as much as it is not the product of an elaborate econometric model. It is not clear whether theory would enable such a model to be constructed; in any case, the required data are not available even to contemplate such an effort.

9 In the model employed in this paper, Markov analysis is limited to seven countries (shares). Extension of this technique beyond that number would be very difficult and could be expected to yield little relative to the effort required.

10 We discuss the theory of the Markov model using imports only because all the mathematics remains the same in analyzing exports; only the estimated matrix would be different.

11 Sources for import and export data are: Austria-Hungary - Österreichisches Statistisches Handbuch; France - Annuaire Statistique (1966); Germany, - Hoffman, W. G., Das Wachstum det deutschen Wirtschaft seit derMitte des 19 Jahrhunderts (Berlin, 1965)Google Scholar; Italy, - Annali di Statistica, serie VIII, vol. 9Google Scholar; Russia, - Khromov, P. A., Economic Development of Russia in the 19th and 20th Centuries, 1800-1917 (Moscow)Google Scholar; United Kingdom - Mitchell, B. R. and Deane, P., Abstract oj British Historical Statistics (Cambridge, 1962)Google Scholar; United States - Historical Statistics of the Unitea States, Colonial Times to 1970, Part 2 (Washington, D.C., 1975)Google Scholar. All European series have beer converted into United States dollars using gold standard par value exchange rates given in Bloomfield, A. I., Short-Term Capital Movements Under the Pre-1914 Gold Standard (Princeton, 1963), Appendix I, 95Google Scholar.

12 The compound rate of growth is calculated assuming annual compounding. For example, in the case of Germany

865.13 × (1.065935)16 2403.1

13 For data on the relative percentages of exports and imports of each country, see Tables 2 and 3.

14 More detailed discussion of the methods whereby transition matrix P can be estimated are contained in Estimating the Parameters of the Markov Probability Model from Aggregate Time Series Data by Lee, T., Judge, G., and Zellner, A. (North Holland, 1970Google Scholar). Statistical Decomposition Analysis by Theil, H. (North Holland, 1972)Google Scholar indicates that by a spectral decomposition it is possible to measure the convergence rate of the estimated transition matrix. This analysis has been performed and the convergence rate of the export and import matrices reported in Tables 2 and 3 are .9941 and .7879, respectively. Additional requirements that all transition matrices must meet are

and

and the requirement that the largest eigenvalue must be equal to one. All transition matrices reported have been tested as noted above and meet these requirements. The accuracy of the forecasts of a transition matrix can be tested by inspection of the mean squared errors for within sample predictions. Mean squared errors accompany the forecasts reported below.

The use of the reported Markov transition matrix can be seen if we show the predicted proportionate shares relate to the estimated Markov matrix and the shares for the past year. For example in Table 3 we indicate that the predicted share of imports for France in 1898 is .1330. This share can be obtained by multiplication of the share vector for 1897 by the second column of the estimated Markov matrix. In this case we note that .1330169981 = (.054412) (.0532) + (.582591) (.1325) + (.1933) (0) + (.0373) (0) + (.473963) (.05) + (.016343) (.3809)+ (.1527) (.150635).

15 For example, 0.2286 x 0.5042734E = 04 1152.7689 or the level of exports of the United States in 1897.

16 Colonel C. E. H. Vincent was one of the founders of the United Empire Trade League and a member of Parliament for Sheffield.