Migration and Labour Market integration, British Engineers

Transcription

1 Migration and Labour Market integration, British Engineers Kentaro SAITO (Kyoto Sangyo University) I. Introduction Markets are thought to become better integrated regionally and then nationally as industrialization proceeds. Britain in the nineteenth century witnessed this process in the labour market integration. It has been discussed in connection with migration, especially on skilled workers because workers with highly skills are expected to have had more frequent traveling, or tramping, for jobs and better working conditions. The more mobile commodity or labour force is, the more integrated their markets is. That s why the labour market of skilled workers is thought more integrated than those of the unskilled. Regarding researches on migration and mobility of skilled workers in the historical perspectives, E. Hobsbawm and H.R. Southwall understand that frequent migration of skilled workers took place in the 19 th century. 1 Hobsbawm claims that migration was caused not mainly by market forces, and wage differentials did not narrowed down. On the other hands, Southall argues that migration took place mainly by economic reasons and was accelerated during recessions. Recently, G. Boyer and T. Hatton, using statistically more elaborate techniques, have extended the area and methods of research on labour migration. 2 Regarding discussions on market integration, although Hobsbawm claimed in the 1960s that levelling [of wages] must have been a slow, gradual, long-term and not startling affair, E.H. Hunt showed more optimistic views of integration. 3 On the other hand, studies in market integration developed with new time series analysis of statistics, being based on an idea of Law of One Price. They regard the decreases in wage-differential as a measure of integration. In this context, again, Hatton and Boyers researched labour market integration of carpenters as an example of skilled workers. Boyers and Hatton claims that various patterns of integrations were proceeding, in urban and rural sections of workers. 4 Against these backgrounds, this paper discusses labour markets of skilled workers in Britain between 1865 and 1914, focusing on skilled engineers, to examine it. Data for engineers wages are combined with data for their migration to develop an argument about the drivers of market integration. Although some traditional sections of workers, e.g. carpenters and agricultural labourers, have been well-researched, this paper focuses on 1 E. Hobsbawm, The Tramping Artisan, in Labouring Men (London, 1964); R.S. Southall, The tramping artisan revisits: labour mobility and economic distress in early Victorian England, Economic History Review, 44 (1991). 2 G.R. Boyers and T.J. Hatton, Migration and labour market integration in late nineteenth century England and Wales, Economic History Review, L,4(1997), pp E.H. Hunt, Industrialization and Regional Inequality: Wages in Britain, , Journal of Economic History, Vol. 56, No.4 (1986); E.H. Hunt, Regional Wage Variations in Britain (Oxford, 1973). 4 G.R. Boyers and T.J. Hatton, Regional Labour Market Integration in England and Wales, , in G. Grantham and M. MacKinnon (eds.), Labour Market Evolution: The Economic History of Market Integration, Wage Flexibility and the Employment Relations (London, 1994). 1

2 modern section of skilled workers, engineers, one of typical skilled group in modern Britain. The engineering industry began to develop from the late eighteenth century, and its importance and them the employment increased a lot throughout the nineteenth century to the twentieth century as Table 1 shows its increase in the employment of between the 1851 and Table 1. Employment in engineering, Year/Branches metal manufacture mechanical engineering electrical engineering Total Source, C. H. Lee, British regional employment statistics, (Cambridge, 1979) While migration and labour market integration have been discussed as seen, the reasons why this paper deals with them are the followings. Firstly, although Hobsbawm s discussion on tramping artisans by Hobsbawm is almost classical work it is not very comprehensive and quantitative. Secondly, Southwall only discussed a part of modern skilled section, steam engine workers, mainly in the first half of the nineteenth century. Thirdly, in the discussion of Hatton and Boyers, wages series of carpenters has been used as urban and skilled workers wages. However, carpenters are traditional section and may not be the best example for testing labour market integration during modernization in industries. In fact, although it is not easy to compare, the labour market of carpenters in Japan was disintegrated in the process of its modernization. 5 Finally, as, the case of engineers is so important in the context of the British industrial relations and labour history, especially on its institutional dimensions, needless to say on the establishments of their trade unions and employers organizations of the nineteenth century to the twentieth century. In the followings, in the first section, we will examine the regional labour market integration of engineers. In the second section, their migration and its feature will be seen. In the third section, the relationship of the both above and some other factors influencing on the integration will be explored, before summarizing them in the concluding section. II. Labour Market Integration In this section we will examine the inter-regional labour market integration of the engineers. The data for this research are the records of the Engineering Employers Federation (thereafter EEF). EEF was established in 1899 and played a major role in industrial relations in the engineering industry. We have used its wage records for 5 Kentaro Saito, Regiocal Labour Market Integration and Standard of Livings in Japan: Four Section of Workers, the 1890s to 1930s,Working Paper for The 14th International Economic History Congress, Helsinki, Finland, 21st-25th August,

3 examining the market integration. Certainly there are governmental records on wages and Labour Gazette recorded them from 1893 by Labour Department of the Board of Trade. 6 However, governments researched them only on irregular basis. On the other hand, EEF recorded the movements in wage rates of skilled engineers, i.e., fitters and turners, in more than 70 British towns continuously over the period and the country from roughly the 1850s to the 1960s. Although the EEF was established in the end of the nineteenth century, it made Statistics Department in the 1910s and recorded wages of employees in their federated firms, tracing back to the 19 th century. 7 Fig. 1 Nominal wages, engineers, , 8 regions Note: As for the 8 regions, see Appendix. Source: MRC, MSS. 237/13/1-2, EEF Wage Book The EEF wages records cover many cities and towns, and therefore, we have broken them down geographically into 8 regions by using the EEF s classification; Scotland, North West, North East, Yorkshire, West Midlands, East Midlands, South East and West Country. Fig. 1 shows average weekly wage rates of 8 regions at the nominal level. Average means average wages of wages of branches in each region. From Fig.1, firstly we can see that from the middle to the end of the 19 th century, wage rates were quite diverse and fluctuated freely, and after the early 20 th century they became quite flat. Secondly, wage rates in the South-East are always high and it is, needless to say, partly because it contained London. Lowest region is Scotland and others show some complex movements. And then, we have converted the wage rates into real wage series from nominal one to compare wage rates between the area and over years. Cost of livings has to be considered adequately because it is quite large depending on the areas, especially in urban areas, although some pointed out that regional variations in prices were small in the first half of the nineteenth century. For some limit of cost of living index, we adjusted nominal data in two ways. Firstly they are put chronologically by using Phelps-Brown s consumable index. 8 6 Official wage statistics over this period can be seen in the followings; Labour Gazette(Board of Trade, ); Board of Trade Gazette(Board of Trade, ). Unpublished Report by Board of Trade in 1908, Rates of Wages and Hours of Labour in Various Industries in the United Kingdom (1908) and its supplements are available. 7 MSS.237/7/11, Modern Records Centre of Warwick University (Thereafter, MRC) 8 Henry Phelps Brown and Sheila V. Hopkins, A perspective of wages and prices (London, 1981) 3

4 Secondly, they are applied by regional cost of living data by Board of Trade in 1907, which shows the indices in Fig.2 Real wages, engineers, , 6 cities Note: 6 cities; London, Manchester, Newcastle, Leeds, Bristle, Birmingham Source: MRC, MSS. 237/13/1-2, EEF Wage Book. Table 2 Regression of CVs of the wages Constant CV on Time/100 R 2 no. of sample Nominal wages (t-value) (-35.1) (-18.1) Real wages (t-value) (-33.1) (-13.9) Source: MRC, MSS. 237/13/1-2, EEF Wage Book Fig. 2 shows the real wages. In this Figure, 6 series of real wages are shown. Because the limits of data of cost of living index, we can only show the cases of London, Manchester, Leeds, Newcastle, Birmingham and Bristle. 10 One of interesting things in this Figure is that London did not always enjoy highest real wages. And another point is that the differentials in the real wages decreased over the period. From here on, we see quickly if the labour market integration was proceeding over the period. Table 2 shows Regression of coefficients of variation of wage rates and real wages on a time trend. In the first place, the convergence of the wage differentials, which was softly rejected by Hobsbawm, can be examined. As index of wage differentials, coefficient of variations (CV, hereafter) is used. CV is value given as standard deviations divided by averages, and shows dispersion of data. Fig. 2 plots CVs of each region over the period to show the trend. Table 2 shows a result of regression of the CVs on the time-trend, in order to evaluate changes. If there is a tendency of decrease in CVs, parameters have to be minus and certainly not zero. The result seems to satisfy it. It has been shown that in the process of modernization and industrialization, local markets were integrated into the national market. 9 Board of trade, Enquiry into Working Class Rents, Housing and Retail Prices in London=100, Newcastle=94, Cardiff=91, Leeds=86, Manchester=86, Birmingham=85. There are no continuous wage records in Cardiff and we calculated, instead, for real wages of Bristle. 4

5 III. Migration of Engineers Although the migration of skilled engineers between towns, or tramping, has been relatively well-documented by previous researches, its quantitatively comprehensive picture has not been researched yet. This paper, therefore, provides data constructed from the travel records of the Amalgamated Society of Engineers (ASE), which from 1865 recorded monthly travels of its members in England, Scotland and Ireland, illustrating characteristics of engineers inter- and intra-regional migration, and the age of workers. We will see their migration between the beginning of their tramping system and As a total number recorded, there are 254, 351 migrations between branches, or house of call, of the ASE over the period. Fig. 3 shows the total migration over the period. Among them, we have chosen 152,940 cases to meet the places of the branches in the EEF and have broken them down into the 8 regions as in the case of the EEF. In this section, we briefly examine the characteristics of the migration among the engineers. Firstly we overlook the comprehensive picture of the migration. The number tended to decrease and fluctuated with some peaks; in 1868, 1879, 1886, 1994, 1904 and In 1879, more than 18,000 engineers travelled, but, in 1914 only 388 migrations were recorded. The largest peak was the one in 1879 and the peaks in the 20 th century were small. This fluctuation seems to have been influenced largely by the trade cycle and matched with the waves of unemployment. Southall showed the periods of the unemployment in the engineering industry and the fluctuation almost exactly followed it. 11 Fig.3 Migration of engineers, Source: MRC, MSS. 259/4/1-52, ASE Monthly Reports and Journals, Secondly, we will see ages and generations of the engineers who migrated and had tramping. The ASE Monthly Journals recorded also their ages and we can reconstruct the age structure among the tramping engineers. Fig.4 shows the proportion of the age group among tramping engineers. Fig. 4 shows that most of them were on their 20s to 60s and the largest group was engineers on their 20s. However, the proportion of the 20s tended to decline and, on the other hands, the group of the 40s to 60s increasingly rose over the period. After 1906, the elder group overtook the younger although the number of the migrants themselves decreased a lot. Because there still repeated severe industrial contraction also in 11 H. Southall, Regional unemployment patterns among skilled engineers, , Journal of Historical Geography, 12,3 (1986), pp

6 the 20 th century, these might be able to be explained by social and even cultural reasons as L.A. Leeson explained. 12 Fig.4 Proportion of age-group, engineers, by 20s, 30s, 40s to 60s Source: MRC, MSS. 259/4/1-52, ASE Monthly Reports and Journals, Thirdly we can see some features of the migrations themselves. Although the destinations to travel and the pattern of travel were, of course, various, the major destination throughout the period was the North West regions, followed then by Yorkshire and South East. While there were many cases in which an engineers travelled over many regions, eg., starting from a town in Wales to travel North East and then Scotland before coming back home, major migrations tended to take place between neighbour regions. Here, we have to divide migrations into two types; inter-regional migration and intra- or internal- regional migration. Inter-regional migration is movement between regions and Internal- or Intra- regional migration is the one within regions. And, from this classification, two migration rates can be defined; Inter-regional migration rate and Intra-regional migration rate. Migration rate should be defined as the number of migrants divided by the number of people in the region. In the case of migration among workers in a trade union, the number of people should be its membership. However, in this paper, we use the ratio of inter- migration to the total migration as inter-migration rate. Table 4 shows inter- regional migration rates and intra- regional migration rates of the engineers. It shows that inter- regional migration rate among the engineers was quite in the middle of the nineteenth century, continuously around 50 per cent and more. However, the rates tended to decrease over the period, in the late nineteenth century; it fell to around 40 per cent although some exceptional yeas can be seen in the twentieth century. VI. Drivers of integrations In this section, we explore the labour market integration and its drivers. Firstly, we try to link migration of engineers between regions with changes in labour market integration, ie., the decrease in wage differentials. Secondly, we will examine the influence of changes in institutions, such as EEF, on the integration. 12 L.A. Leeson, Travelling Brothers(London, 1979) 6

7 Table 4 Inter-regional migration rate and Intra-regional migration rate, Engineers, Scotland North East Yorkshire North West East Midlands West Midlands South East West Total Inter- Intra- Inter- Intra- Inter- Intra- Inter- Intra- Inter- Intra- Inter- Intra- Inter- Intra- Inter- Intra- Inter- Intra Sources: MSS. 259/4/1-52, ASE Monthly Reports and Journals. Firstly we will see the relations between the wage-differentials and labour market integration. The framework of this analysis is the followings. When workers are out of work, they are expected to travel for jobs. In addition, it can be thought that they, including the case of unemployment, seek for better wages than they earn in their present situations although we have to consider their costs to move around. In the case of engineers, it is quite famous that their trade union, ASE, gave them information of their trade over the country. 7

8 Therefore, it can be thought that the larger wage differentials, the more simulative to migration. Therefore, we have tested the relationship between inter-regional migration rates and CVs of wages at the national level. Table 5 shows the results of regression. The result shows that the nominal differentials do not support significantly to induce the migration, and shows that the differentials of real wages were more influential on the migration. Fig.6 shows relation between inter-migration rates and CVs. Fig. 6 Inter-regional migration rate and CV of real wages, Source: MSS. 259/4/1-52, ASE Monthly Reports and Journals, ; MRC, MSS. 237/13/1-2, EEF Wage Book Table 5 Inter-regional migration rates on CVs, Constant Migration rate on CV R2 no. of sample(=years) CV of real wages (t-value) (-12.5) (-4.25) CV of nominal wages (t-value) (-7.44) (-0.35) Notes: The number of sample is 46 because there are no migration records in 1895, 1896, 1897 and Sources: MSS. 259/4/1-52, ASE Monthly Reports and Journals, ; MRC, MSS. 237/13/1-2, EEF Wage Book As next analysis, we will try to see the labour market integration from a different angle. If two or several markets have a stable relationship during a certain period, it can be thought that there is an integration relationship. This stable relation means stability or equilibrium in which the relation comes back soon after a shock takes place and once disturb the relation. This type of relation is called co-integration statistically, and we tried if co-integration relationship exists between two sample markets in our data. In this case, as shocks, we will examine two; the Great Depression in the late nineteenth century and the establishment of the national collective bargaining. The former took place roughly in the late 1870s and influenced on the economy and industries all over the Western countries. In the engineering, 1878 and 1879 were especially severe years, and therefore we 8

9 examine 15 years before and after the slump. 13 We will also see the influence of some institutional changes in the industry. In the British industry, system of collective bargaining was gradually developed throughout the 19 th century. In the engineering, large size of industrial dispute took place in London in 1897 and just after that EEF was established. 14 Therefore, we will see its influence taking 17 years before and after 1898 to see how market stability was changed. Table 6 to 7 shows the results. In these Tables, name of places, regions, and town are written vertically and horizontally, and each square shows its combination. As combination only is needed, only upper half is used in the table. If there is a stable relationship, figure is filled in. These figures are called Trace Value, which is used co-integration test. Table 6-1 shows changes before and after the slump at nominal level. Before the slump, there was no integration but, after it, 2 integration can be seen; Northern-Yorkshire and Northern-Lancashire. Table 6-2 shows the changes at real wage level. Before the slump there was one integration; between London-Bristle. After the slump, the number increased; 6 integrations; Newcastle-London, Newcastle-Bristle, Manchester-Birmingham, Manchester-London, Manchester-Bristle, London-Bristle. Table 7-1 shows changes before and after the establishment of the EEF at nominal level. Before the EEF, there were two integration; Northern-Yorkshire and Yorkshire and Lancaster. After the EEF, the number of integrated market became 3; Northern-Lancashire, Lancashire-West and West-Midlands. Table 7-2 shows its change at real wage level. Before the establishment of the EEF, there were 3 integrations; Newcastle-Leeds, -London, -Bristle. After the establishment, again, the number increased; 7 integrations. V. Conclusions Labour market of skilled workers is expected to be integrated as the industrialization proceeds. Certainly, on the case of engineers, their labour markets were being integrated steadily from the middle of the nineteenth century to the middle of the 20 th century. In the long run, at the national level, it can be thought that the labour market of engineers was integrated by migrations of engineers between regions. The scale of the migration among the engineers was very large in the middle of the nineteenth century but then decreased and faded away before the First World War. In accordance with it, the characteristic of the migration itself changed. We need more researches to know the reasons. On the other hand, in the shorter period, the labour market integration was boosted by the industrial contraction and changes in institutional factors such as the establishment of employers organizations. As engine of the integration, it might be thought in the former case that external shock such as national wide depression accelerated the integration, and in the latter that institutional change such as the establishment of national bargaining influenced positively on the integration. In this paper we have shown only some the results of our on-going research briefly, and obviously we need further examinations on these important topics in more details. 13 As for the influence of the slump to the members of the ASE, see J.B. Jeffery, The Story of the Engineers (London, 1970), Chapter V. 14 A.I. Marsh, Industrial Relations in Engineering (Oxford, 1965) 9

10 Table 6-1 Before and after the Slump; in nominal wages Source: ; MRC, MSS. 237/13/1-2, EEF Wage Book Table 6-2 Before and after the Slump; in real wages Source: ; MRC, MSS. 237/13/1-2, EEF Wage Book Table 7-1 Before and after the establishment of the EEF, in nominal wages Source: ; MRC, MSS. 237/13/1-2, EEF Wage Book 10

11 Table 7-2 Before and after the establishment of the EEF, in real wages Real wage Newcaslte Leeds Manchester Bermingham London Bristol Newcaslte * 16.5* 17.9* Leeds Manchester Bermingham London Bristol * 5% significant, figures in parenthesis show Trace value Real wage Newcaslte Leeds Manchester Bermingham London Bristol Newcaslte -- Leeds * 19.1* 26.9* Manchester * 18.1* 17.0* Bermingham London * Bristol * 5% significant, figures in parenthesis show Trace value Source: ; MRC, MSS. 237/13/1-2, EEF Wage Book Primary Sources Modern Records Centre at University of Warwick MSS. 259/4/1-52, ASE Monthly Reports and Journals, MSS. 237/13/1-2, EEF Wage Book References Boyers, G.R., and Hatton T.J., Regional Labour Market Integration in England and Wales, , in G. Grantham and M. MacKinnon (eds.), Labour Market Evolution: The Economic History of Market Integration, Wage Flexibility and the Employment Relations (London, 1994). Boyers, G.R., and Hatton T.J., Migration and labour market integration in late nineteenth century England and Wales, Economic History Review, L.,4(1997)pp Ejinaes, M., and Persson, K. G., Market Integration and Transport Costs in France : The threshold error correction approach to the Law of One Price, Exploration in Economic History 37, (2000), pp Hunt, E.H., Industrialization and Regional Inequality: Wages in Britain, , Journal of Economic History, Vol. 56, No.4 (1986). Hunt, E.H., Regional Wage Variations in Britain (Oxford, 1973). Jacks, D., Intra and international commodity market integration in the Atlantic economy, , Exploration in Economic History 42(2005), pp Jacks, D., What drove the 19 th century commodity market integration?, Exploration in Economic History 43(2006), pp Leeson, L.A., Traveling Brothers(London,1979) Marsh, A., Industrial Relations in Engineering (London, 1965) Persson, K.G., Grain Market in Europe, , Integration and deregulation (Cambridge, 1999). Rothenberg, W.B., The Emergence of Farm Labor Markets and the Transformation of the Rural Economy: Massachusetts, , Journal of Economic History, Vol. 58, No.3, (1988). Southall, H.R., The tramping artisan revisits: labour mobility and economic distress in early Victorian England, HER, Vol.XLV, No.2, 1991 Southall, H.R., Mobility, artisan community and popular politics in early nineteenth century England in Urbanizing Britain, Essays on class and community in the nineteenth century, Kearn G. and Whiters W.J.(eds.), (Cambridge, 1991). 11