Creating Expectations

Prof. Dr. Alexander Kriwoluzky, Berlin (Principal Investigator)

Prof. Dr. Moritz Schularick, Bonn (Principal Investigator)

Stephanie Ettmeier M.Sc., Berlin


If the present behavior of forward-looking economic agents is driven by their expectations about the future, can governments influence the economy by creating expectations? More precisely, can governments, by skillfully managing and guiding the public’s perception of present and future policies, boost economic activity and employment in the present without actually spending much? Do words speak louder than deeds after all?

We study this question in the context of one of the most consequential macroeconomic success stories of the 20th century that has, to date, defied the explanations advanced within the literature: the recovery of the German economy from the Great Depression under the Nazis from 1933 onwards. Revisiting the causes of the German recovery in the 1930s, our project asks if the role of fiscal policy as a driver of the recovery has to be fundamentally reassessed once the effects of expectation creation are considered.

We start from the idea that fiscal policy begins to be effective when it is expected and not only when it is implemented. Thus, our central hypothesis is that Nazi communication and propaganda efforts boosted the effectiveness of the fiscal stimulus much beyond its actual size. Fiscal propaganda likely influenced perceptions through three channels. Firstly, and most importantly, people came to expect a long and sustained public investment effort as the heavy use of propaganda about highway construction and public work programs credibly tied the legitimacy of the new regime to the success of these programs. This is the forward guidance channel. In addition, in a world of imperfect information, propaganda can also increase the effects of public spending through two additional channels: by making them appear bigger than they are, and by spreading the knowledge of the policy across larger parts of the population through the use of mass media.

Our project will be divided into three parts. The objective of the first part is to compile a new historical dataset on fiscal propaganda and fiscal spending. To obtain a quantitative measure for fiscal propaganda we will first digitize and preprocess a range of German national and regional newspapers as well as radio broadcasts. We will then apply quantitative text analysis techniques in order to extract empirical evidence on the intensity of Nazi fiscal propaganda from these sources. The second project will use the new data on propaganda to study how it affects the transmission of fiscal policy in a Bayesian VAR model. In the third project we take a regional perspective and compute the effect of regional fiscal spending. Using military spending as an instrument for fiscal spending, we estimate the effects of fiscal policy on employment in regions with low (or high) propaganda intensity. This allows us to explore how strong fiscal propaganda affected the multiplier and thus its role for the German recovery from the Great Depression.