Wages and prices of consumer goods in Germany, 1850 to 1889.
- Pfister, Ulrich (Universität Münster, Historisches Seminar)
- Historical Social Research
AbstractThe study constructs new series of nominal wages in industry and crafts as well as a new consumer goods price index for the period 1850-1889; the coefficient of the two series gives the real wage. While such information was collected and published by government agencies from the late 1880s onwards, the decades before are part of the pre-statistical age. After all, information is available from municipal authorities, from branches of territorial state authorities and from individual companies. For the construction of a new nominal wage series, the current study refer to Kuczynski´s material (1961/62), supplements it with information from individual studies of the past 50 years, and constructs wage indices for the heavy ironware, machine construction, mining, printing, and municipal construction industries on this basis by means of unbalanced panel regression with fixed effects. Of the 38 individual wage series on which these sector indices are based, 27 come from Kuczynski, the remainder from more recent studies. Wages in the textile sector are represented by those in the cotton industry. The study uses the wage series published by Kirchhain (1977). Weighted according to employment figures, all these sector-specific series (excluding miners´ wages) are aggregated into a Fisher index of nominal wages in industry and crafts. Both this index and the indices at sector level are linked in 1888/89 with the series by Hoffmann (1965); the resulting values denote annual earnings in Marks. The sector indices differ little from those of Kuczynski and Hoffmann despite the expansion of the database and the different methodology of index construction, but the aggregated index shows a stronger growth rate than that of Kuczynski; the latter index is obviously erroneous (Pfister 2018, 576). The consumer goods price index is based on five sub-indices for (1) food, (2) beverages and luxury foods, (3) rent, (4) furniture, household goods and heating, and (5) clothing. The sub-indices for food and rent are new, the other three are from Hoffmann (1965). Weights are determined for 1848/49 and 1889 on the basis of research literature, values in between are interpolated linearly. Both the sub-index of food prices and the overall index are constructed as Fisher indices. Both the rental index and the food prices rise more strongly in the long term than the two corresponding Hoffmann indices (Pfister 2018, 578 and 582). Hoffmann constructs the rental price index only indirectly by multiplying the estimated building capital by an assumed interest rate. The rent index of the current study is based on data from three major cities. Only if it is assumed that large cities are completely unrepresentative for the entire real estate market should Hoffmann´s series still be considered. In the case of food prices, the comparatively stronger long-term increase - compared to previous research - results from the higher weight of prices from the southern parts of the country far from the sea in the new sub-index. Here, the price dampening effect of growing imports of American grain had a weaker effect than in the coastal regions in the north. Thus, one of the main findings of the study is that the assessment of the development of the living standards of urban workers from the 1850s to 1880s strongly depends on how one determines the effect of the first wave of modern globalization on the German price structure. The greater consideration given in this study to food prices in areas distant from the sea results in a more pessimistic view of the development of real wages during this period than has been the case with some previous research. To the data: 1. individual wage series (table set A.01) This set of tables contains wage series from six branches at the level of regions, cities, individual enterprises and in one case (cotton industry) an entire branch. Only series containing data for at least 15 years were taken into account. In detail, the series are the following: Heavy Ironware Bochum 1869-1889: Average annual income of the workers of the Bochumer Verein (steelworks) in Mark; Däbritz (1934, Annex Table 4). Essen 1848-1889: Average annual income of the workers of the Krupp works in Mark; Kuczynksi (1961-62, vol. I, 377, vol. II, 227, vol. III, 426). Ruhr 1855-1889: Average annual income of the workers at the blast furnaces in the Ruhr district in Mark; banks (2000, Table A59). Saar 1869-1889: Day wage of workers at the blast furnaces of the Burbach Ironworks in Mark; Kuczynksi (1961-62, vol. III, 426). Silesia 1869-1889: Average annual income of workers at the blast furnaces in Silesia in Mark; banks (2000, Table A59). Machine construction Augsburg 1851-1889: Average annual income of the workers of the Machine Factory Augsburg in Mark; Vol. II, 227; Kuczynski (1961-62, Vol. III, 426). Chemnitz 1860-1887: Weekly wage of machinists in Mark; Kuczynski (1961-62, vol. II, 227; vol. III, 426). Esslingen 1848-1889: Average annual income of workers at the Esslingen machine factory in Mark; Schomerus (1977, 296 f.). Hanover 1875-1889: Weekly wage of Hannoversche Maschinenbau AG in Mark; Kuczynksi (1961-62, vol. II, 227; vol. III, 426). Munich 1867-1889: Average annual income of the workers of the locomotive factory Munich in Mark; Kuczynski (1961-62, vol. II, 227; vol. III, 426). Nuremberg 1851-1889: Average annual income of the workers of the iron foundry and machine factory Klett in Mark (later MAN); Gömmel (1978, 204 f.). Szczecin 1870-1887: Day wage of the workers of the Vulkanwerk (shipyard, mechanical engineering). The figures for the annual income from 1882 to 1887 were divided by 300 to determine the daily wage; Kuczynski (1961-62, vol. II, 227; vol. III, 426). Mining Aachen 1818-1884 (hard coal): Wages of miners per shift in the Aachen mining district in Mark. The value of 1889 is an outlier and was not included in the analysis; Kuczynski (1961-62, vol. I, 373-4; vol. II, 223; vol. III, 419). Dortmund 1818-1889 (hard coal): Average annual net income of miners in the Dortmund mining area in Mark; until 1833 Kuczynski (1961-62, vol. I, 373-4), from 1850 Holtfrerich (1973, 54-56). Mansfeld 1818-1884 (copper): Miners´ wages per shift in the vicinity of Mansfeld (Harz) in Mark. Values from 1874-1884 refer to the average wages in the upper and lower districts; Kuczynski (1961-62, vol. I, 373-4; vol. II, 223; vol. III, 419). Lower Rhine (left of the Rhine; iron) 1818-1884: Miner´s wages per shift in left Rhine iron mining in Mark; Kuczynski (1961-62, vol. I, 373-4; vol. II, 223; vol. III, 419). Saar 1824-1889 (hard coal): Annual wage of underground workers in Prussian state mines in Mark to Müller (1904); taken over from banks (2000, vol. I, 139; vol. II, 103-5, appendix Table A15). Saxony 1869-1885 (hard coal): Average annual earnings of workers working in mining and coal extraction in the state mines of Zauckerode in Mark; Böhmert (1885, 202). printing industry All series are from Kuczynski (1961-62, vol. I, 376; vol. II, 225; vol. III, 42). Until 1872 they refer to Marks per 1000 letters set, from 1872 to weekly wages in Marks. Data are available for the following cities and periods: Berlin 1830-1889, Halle 1830-1872, Hamburg 1872-1886, Hanover 1872-1889, Kassel 1848-1872, Leipzig 1830-1889, Munich 1848-1889 and Stuttgart 1830-1889. Textile industry (cotton industry) Annual earnings in the cotton industry in Mark; Kirchhain (1977, 158-160). Data for 1815-59 refer to the earnings of textile spinners in central workshops and factories. The values for 1860-89 are the weighted average of the average annual earnings of spinners and weavers in central workshops and factories. The figures for the second subperiod are weighted by the relative employment shares of spinners and weavers; the employment shares between census years are determined by linear interpolation. The resulting series is calibrated with wages in the whole textile sector 1888/89 (Hoffmann 1965, Table II/108, 468-471; cf. GESIS ZA8257, Table 02). Municipal building trade As far as the source is not expressly noted, the information comes from Kuczynski (1961-62, vol. I, 375; vol. II, 224; vol. III, 420-422). Berlin 1868-1889: 1868-9: daily wage of carpenters in marks times 6; 1870-89: wage of masons in marks per week. Bochum 1871-1885: Wages of carpenters in Marks per week. Braunschweig 1840-1877: Wages of skilled construction workers in grams of silver per week; Schildt (1986, 383). Chemnitz 1841-1885: Wages of masons in marks per week. The values in the years 1841-1850 refer to the daily wages of the journeymen and are multiplied by 6; until 1850 Strauß (1962-64, part 1962(4), 159-161), 1860-1885 Kuczynski (1961-62, vol. II, 224; vol. III, 420-422). Gelsenkirchen 1871-1885: Wages of masons in Marks per week. Glauchau 1865-1885: Wages of the masons in Marks per week. Hamburg 1840-1889: Day wage of the carpenters in Mark. Leipzig 1840-1860: Daily wage of carpenters during the summer in grams of silver. The linear relationship between the wages of the carpenters and the wages of the masons from 1795-1860 serves to extrapolate the values in the years 1853 and 1859; Pfister (2017, online appendices S1 and S4). Nuremberg 1840-1889: Annual earnings of construction workers in Mark; Gömmel (1978: 204 f.). Rhineland 1856-1870: Wages of masons on the Lower Rhine in dollars [sic] per two weeks. Rostock 1840-1885: Wages of carpenters and masons in Marks per week (values for carpenters and masons are identical). 2. nominal wage indices for individual sectors or in industry and crafts (table set A.03) On the basis of the wage series described, nominal wage indices were constructed for the respective sectors. Since the individual series extend over different periods and show missing values, an unbalanced panel regression with fixed effects for years and places is chosen as the method (Pfister 2018, 572 f.). The fixed effects for locations also take into account the fact that wages even within the same sector sometimes refer to different currency units (mark, gram silver) and periods (daily wage, weekly wage, annual wage). With the exception of the regression for mechanical engineering, the error variants of the OLS estimate suffer from heteroskedasticity. Feasible GLS with the grouping of the error variants alternatively by location and year served to clarify the sensitivity of the results obtained with OLS with regard to heteroskeleton elasticity of the error variants (Pfister 2018, 573-575). The differences between the wage indices estimated with feasible GLS and OLS at industry level are small and are mainly limited to the 1870s. Heteroscede elasticity is mainly due to the fact that the crisis of the 1870s affected wages to varying degrees in the individual regions. The study is also based on the OLS estimates, but Table A.03.01 also contains the series estimated with feasible GLS, in which the error variants are grouped by city (excluding mechanical engineering). All sector indices are linked in 1888/89 with the corresponding series in Hoffmann (1965, Table II/104, 461; Table II/108, 468-471; cf. GESIS ZA8257, Tables 01 and 02), so that the values are given in Marks per year. In detail, the following new sector indices were constructed: - Heavy ironware/iron-creating industry 1848-1889. Because the data before 1868 are essentially limited to the wages of the Krupp factories in Essen, the GLS estimate only refers to the period 1868-1889. For earlier years the values estimated on the basis of the OLS 1868 are linked with this series. - Mechanical engineering 1848-1889. - Mining 1818-1889. - Printing 1830-1889. As mentioned, wages until 1872 refer to Marks per 1000 letters set, from 1872 to Marks per week. Therefore, separate indices were estimated for the two subperiods and the two series were chained together in 1872. - Tree craft 1840-1889. For the period 1848-1889, the sector indices for heavy hardware, machinery and equipment, printing, construction and textiles were combined into a nominal wage index for industry and crafts (Table A.03.02). It is constructed as a linked Fisher index where weights are adjusted each year. The weights are derived from the number of persons employed in the sectors covered (Hoffmann 1965, Table II/15, 196-199). The values frequently missing before 1875 are replaced by linear interpolation. The resulting series is linked in 1888/89 with Hoffmann´s nominal wage in industry and crafts (1965, Table II/108, 468-471), so that the values are available in Marks per year. 3. consumer goods prices (table set A.02) The study develops new indices for rental expenditure and food prices and constructs an index of consumer goods prices for the period 1850-1889. Index of rental expenditure (1871=100) The data are based on the following cities´ reference years: - Berlin 1846, 1855, 1864, 1871, 1881, 1887-1889; Kuczynski (1961-62, vol. I, 381; vol. II, 231; vol. III, 437). - Hamburg 1840, 1850, 1860, 1870, 1875; Kuczynski (1961-62, vol. II, 232). - Nuremberg: 1849, 1857, 1874, 1883, 1889; Gömmel (1974, 216-220). For Nuremberg, Gömmel has already developed an annual series; for Berlin and Hamburg, the values between cut-off years are interpolated with the exponential trend. An overall index of rental expenditure is formed by geometric means of the three series (Pfister 2018, 577 f.). Index of food prices (1871=100) The index is based on three sub-indices for the Kingdom of Prussia and the two cities of Munich and Nuremberg (1848-1889; Table A.02.03). - Prussia: Prices for six basic foodstuffs on 81 (until 1866) and 101 markets respectively; data from magazine (1907, 84-86), obtained from GESIS ZA8302. - Munich and Nuremberg: data from the municipal authorities on the prices of eleven basic foodstuffs; data from communications (1894, 380); Gömmel (1978, 216-220). The determination of the weights of the prices of the individual products is based on assumed expenditure structures in the years 1849 and 1889 (Pfister 2018, 580). On the basis of these key figures, the portions of expenditure for the intervening years are interpolated on a straight-line basis. On this basis, the three indices for Prussia, Munich and Nuremberg are constructed as Fisher indices with annually adjusted weights (Pfister 2018, 582). The national food price index is the weighted geometric average of the three sub-indices, with the Prussian index weighted 0.6 and the two cities weighted 0.2. These weights are based on Prussia´s share of the German population (after 1866). Index of consumer goods prices (1871=100) The consumer goods price index is based on five sub-indices for (1) food, (2) beverages and tobacco, (3) rent, (4) furniture, household goods and heating and (5) clothing. Those for food and rent are the new series described above, the remaining three come from Hoffmann (1965, Table II/148, 598-601; data taken from: GESIS histat data file ZA8254, Table 11). The weights are initially based on the proportions of expenditure reported for Nuremberg households in the years 1849, 1857, 1883 and 1899 (Gömmel 1978, 211). However, they are changed according to the expenditure structure assumed by Saalfeld (1984, 238) for about 1847 and by Fischer (2011, 328 f.) for the period around 1900 on the basis of budget studies for the workforce (Pfister 2018, 579 f.). Values between the reference years are obtained by linear interpolation; the result can be found in Table A.02.01. On the basis of these weights and the sub-indices in Table A.02.02, a Fisher index of consumer prices is formed (Table A.03.02). The following data series are included in this study: A.01 Individual wage series A.01.01: Individual wage series (in marks) - Heavy ironware, 1848-1889 A.01.02: Individual wage series (in marks) - machine construction, 1848-1889 A.01.03: Individual wage series (in marks) - Mining, 1818-1889 A.01.04a: Individual wage series (in marks) - Printing: per 1000 letters, 1830-1872 A.01.04b: Individual wage series (in marks) - Printing: weekly wage, 1872-1889 A.01.05: Individual wage series (in marks) - Textiles (cotton processing), 1815-1889 A.01.06: Individual wage series - urban building, 1840-1889 A.02 Price indices A.02.01: Price indices - weights, 1848-1889 A.02.02: Price indices - consumer goods categories (1871=100), 1848-1889 A.02.03: Price indices - Food price index (1871=100), 1848-1889 A.03 Aggregated indices, real wage A.03.01 Aggregated indices: Nominal wages by sector (OLS estimate, estimate on ´feasible GLS´ basis), 1818-1889 A.03.02 Aggregated indices: Nominal and real wages in industry and crafts, consumer goods price index, 1848-1889 Notice: Studies closely related to this study are: ZA8636: Pfister, U., nominal wages and consumer goods prices in 18 German cities, 1500-1850. ZA8709: Pfister, U., wage inequality in Germany from the late 15th century to 1889.
1850 / 1889
German Reich (1871-1945) (DXDE)German Reich
Number of Variables:
- ZA8710 (Type: ZA-No.)
Ulrich Pfister (2018), Real Wages in Germany during the First Phase of Industrialization, 1850-1889. In: Jahrbuch für Wirtschaftsgeschichte 2018; 59(2): 567-596. https://doi.org/10.1515/jbwg-2018-0019
Update Metadata: 2021-04-07 | Issue Number: 14 | Registration Date: 2019-08-06