Disaggregated financial flows and economic development: Evidence from pre-1913 Germany

In this paper we analyse income formation patterns throughout the German industrialisation process (1860-1913) through the analysis of different financial flows. Similar to Neuburger & Stokes (1974), we make use of flow statistics originally estimated by Eistert (1970) with regard to four different types of financing, i.e. bills of exchange credit, lombard credit, current account credit, and securities credit which together comprise the total flow of credit provided by the banking system. We also enlarge the data set of Eistert (1970) and Eistert & Ringel (1971) by making use of different sources in order to allow for a representative statistical analysis. To our knowledge, we have compiled the first dataset on German financial flows spanning from 1860 to 1913. Our goal is to provoke a fundamental discussion about the suitability of stock vs. flow statistics - a question which has been disregarded for too many years. Moreover, we would like to shed more light on the question whether a qualitative differentiation of the different types of financing is needed in order to make more precise estimations of the influence of finance on real economic activity in general and non-agricultural income formation in particular. This paradigm derives from the theory of disaggregated credit formalised by Werner (1997) which is amongst others advocated by Eistert & Ringel (1971) as well. Statistical analysis will be conducted by utilising the General-to-Specific (Gets) approach presented in Sucarrat & Escribano (2012). Contrary to Neuburger & Stokes (1974), we have found a significantly positive relation between current account credit flows and non-agricultural output among other findings. Besides, the results might lend further support to the theory of disaggregated credit and might have implications for renowned models of income formation especially the IMF Polak (1957) model (for developing economies).