The bi-directional causality between financial variables and business cycle

Our paper aims to document how macroeconomic conditions and financial variables can influence and affect each other. In the first part of our paper, we study how the business cycle affects various financial variables; in particular how firms’ debt and equity issuance changes over the business cycle....

Full description

Saved in:
Bibliographic Details
Main Authors: Lim, Guan Wei, Pua, Robin Suan Jin, Tee, Guang Ying
Other Authors: Wu Guiying Laura
Format: Final Year Project
Language:English
Published: 2017
Subjects:
Online Access:http://hdl.handle.net/10356/69738
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Our paper aims to document how macroeconomic conditions and financial variables can influence and affect each other. In the first part of our paper, we study how the business cycle affects various financial variables; in particular how firms’ debt and equity issuance changes over the business cycle. We find that both debt and equity issuance are pro-cyclical for small firms. For large firms, debt issuance is pro-cyclical whereas equity issuance of large firms may be weakly pro-cyclical or counter-cyclical depending on which definition of net equity issuance is used. This suggests that large firms may be able to substitute between different financing methods during different phases of the business cycle. In the second part of our paper, we reversed our direction to study how various financial variables affect the business cycle; in particular the feasibility of predicting business cycles using financial variables. We find that in both Probit and Logit model, change in capital expenditure in property, plant and equipment has statistical significant effect on the probability of recession occurrence. Although the impact is about 0.2% increase in probability of recession for every 1% increase in change in capital expenditure in property, plant and equipment, our model is able to classify 75.16% of the data correctly.