Essays on the Origins of Aggregate Fluctuations

Understanding the origins of business cycles has always been an important focus for the field of macroeconomics. Mechanisms that have been proposed in the previous literature include productivity shocks, financial frictions, demand shocks and informational frictions. The three essays in ray dissertation try to either empirically validate some of the mechanisms proposed by the previous literature, or theoretically propose new mechanisms to the current debate.;One mechanism that has attracted attention in the previous literature is the shock on uncertainty. If there are irreversibility or fixed costs associated with fixed investments, firms will reduce their investment in the fact of heightened uncertainty, which increases the option value of waiting. However, previous literature generally fails to come up with causal inferences on whether this mechanism is economically significant in the real world. Two chapters in my dissertation aim to answer this question by utilizing different natural experiments that are unique to the Chinese economy.;The first chapter utilizes the turnover of government officials as a natural experiment for shocks on policy uncertainty. We find that political turnover in China can be predicted by a set of variables that are exogenous to the local firms' investment decisions. We also show that political turnover is closely correlated with traditional measures of policy uncertainty. Based on these findings, we then document the causal relationship between policy uncertainty and firm-level investment rates. We find that an increase in policy uncertainty causes firm-level investment rates to fall, and makes firms less responsive to demand shocks. These findings are robust under a number of specifications and robustness checks. The third chapter utilizes the reform of exchange rate in China as a natural experiment for shocks on exchange rate uncertainty. Result shows that increased RMB volatility induced by China's policy shift had significantly reduced exporting firms' responsiveness of investment to demand. It is also found that exporters in the industries with higher degree of capital irreversibility were more sensitive to the exchange rate uncertainty shock, which is consistent with the capital irreversibility hypothesis in the literature. Moreover, investment response to exchange rate uncertainty is heterogeneous across firms. Exporters with non-state ownership, less exporting experience, and lower productivity were more affected by the increased RMB volatility. Another mechanism that has been proposed in the past is the amplification of idiosyncratic shocks on individual sectors. The previous literature has demonstrated that two otherwise identical economies can have different business cycle fluctuations simply because their input-output networks are organized differently. The economy with a few "star" sectors, which are suppliers to many other sectors, are more vulnerable to aggregate fluctuations than the economy without "star" sectors. However, previous literature focuses primarily on the case where each sector is perfectly competitive. Ignoring imperfect competition makes the economy immune to demand shocks. Chapter two in my dissertation introduces imperfect competition to the input-output literature. We provide a theoretical framework to show that the network structure of input-output linkages and the degree of competitiveness of individual industries jointly determine how sectoral shocks are transmitted. We then use an innovative data set from China and an exogenous exchange rate shock to empirically identify our results from the model. We find evidence on the amplification of demand shock within the input-output network, which is affected by the degree of competitiveness of the connecting industries. We propose that, when investigating how idiosyncratic shocks can turn into aggregate shock, it is necessary to analyze not only the network structure of the economy's input-output linkages, but also the degree of competitiveness of each industry.;To sum up, all the chapters of my dissertation are centered on one theme: to better understand the different mechanisms that can lead to aggregate fluctuations. All the papers in the dissertation contain clear hypotheses for empirical testing and an empirical identification that helps us to differentiate between our theory and alternative hypotheses. Our findings in these chapters contribute to the debate on the origins of business cycles.