skip to content

Keynes Fund

 

Summary of Project Plan


The recent financial crisis and the events surrounding Brexit negotiations have brought the impact of economic uncertainty to the forefront of macroeconomic research. Uncertainty of future economic policy and conditions can distort and hinder investment, delay firm hiring decisions and adversely impact consumption (Bloom, 2009; Christiano et al., 2014; Jurado et al., 2015). Consequently, variation in economic uncertainty has been found to be an important driver of economic conditions in recent years - both during recessions and economic recoveries (Caldara et al., 2016).

This project aims to apply Natural Language Processing (NLP) methods to over 150 years of data from four major newspapers to better understand economic uncertainty. This research requires funding both to acquire the digitized data from the newspapers and to get access to remote high-powered computing to analyse over 100 million words and sentences using machine learning methods. This can help us distinguish between shocks to economic uncertainty, to business sentiment and to financial conditions, which in turn is crucial for our understanding of the drivers of business cycles. For example, based on existing data, Jurado et al. (2015) and Caldara et al. (2016) note that financial shocks and uncertainty shocks are often closely correlated, making it difficult to disentangle the impact and the relative importance of the two (Baker et al., 2016). An additional obstacle to identificiation is highlighted in a recent contribution by Nimark, 2014, showing that modest but unusual changes to macroeconomic aggregates themselves can drive fluctuation in uncertainty. Our use of natural language processing methods coupled with a time-varying model and a penalty function based identification strategy aims to overcome these issues, while at the same time creating a new long-run index of economic uncertainty. We then propose to proceed to ascertain how changes in economic uncertainty have influenced economic conditions over the long run.

Our project is directly in line with the objectives of the Keynes Fund, as it aims to draw on history to understand an important source of market frictions and imperfections. Indeed, Keynes, 1937 himself highlighted economic uncertainty as an economically important source of market ineffi . More broadly, our project directly contributes to a prolific line of macroeconomic research on the drivers of the behaviour of market participants.

 

 

Walter Jansson and Alain Naef

 

Walter Jansson is an Analyst at the Bank of England, London. He obtained his PhD from the University of Cambridge. His thesis studied linkages between the financial system and the economy in Britain from 1850 to 1913, with a particular focus on the role of various credit shocks in driving macroeconomic conditions.

 

Alain Naef is a postdoctoral fellow at the Department of Economics of the University of California, Berkeley. He obtained his PhD from the University of Cambridge. His research expertise is in Economic History, International Economics, Monetary Economics and Environmental Economics.

 

Share this Project