Financial Shocks or Productivity Slowdown: Contrasting the Great Recession and Recovery in the United States and United Kingdom

Abstract

This article contrasts the experiences of the United States and United Kingdom during and after the Great Recession to understand the role of financial shocks in the magnitude of the crises and length of the recoveries. It starts from the common consensus that the Great Recession first and foremost was a financial crisis. It shows that relative to the United States, the Great Recession in the United Kingdom was more closely associated with a decline in productivity. Motivated by the similar behavior of financial variables at a business cycle frequency, it contrasts the behavior of the U.S. and U.K. economies through the lens of a simple real business cycle model augmented with financial shocks. A credit channel that operates on firm hiring decisions captures the magnitude of the output decline in both the United States and United Kingdom but exaggerates the response of the hours margin for the United Kindgom. The conclusion is that the financial channel supported in the U.S. data seems less appropriate for understanding the U.K. experience.

Introduction

The Great Recession renewed academic interest in the role of financial factors in the determination of real macroeconomic quantities. The magnitude of the decline in output, fall in house prices, and failure in financial markets led economists to place greater emphasis on financial institutions, debt, and financial variables within macroeconomic thinking (Reinhart and Rogoff, 2009; Brunnermeier, 2009; and Guerrieri and Uhlig, 2016). The synchronized nature of the decline across developed economies has reasonably been interpreted as suggesting a common cause, principally a large negative financial shock. 

This article studies the behavior of the United States and United Kingdom economies during the Great Recession and subsequent slow recovery to study whether 

(i) differences and similarities in the responses of macroeconomic aggregates can reveal the underlying drivers of the decline and recovery, 

(ii) financial shocks were the most important factor in the U.S. and U.K. experiences during and after the Great Recession, and 

(iii) financial markets transmit shocks to the real economy in the same way in both countries.

The main finding is that the financial channel has greater explanatory power for the U.S. than the U.K. economy in precipitating the Great Recession. This is perhaps surprising given the prevailing consensus of a financial market origin of the crisis and the strong similarities in the unconditional moments of aggregates at a business cycle frequency in the two countries. Despite having private business sectors that exhibit similar financing dynamics and U.K. households being more indebted than their U.S. counterparts, through the lens of a simple model, financial shocks seem to have been more important for understanding the U.S. economy than the U.K. economy during the Great Recession.

First, in the United Kingdom, changes in productivity account for the majority of macroeconomic movements during the Great Recession. Second, while financial shocks help explain the precipitous decline in output in the United Kingdom in 2008, they also generate counter-­­factually high volatility in the U.K. labor market. Third, financial conditions took longer to return to normality in the United Kingdom than in the United States. During this period, financial factors appear more relevant in the U.K. setting. 

In this analysis, we first apply the classic business cycle accounting (BCA) framework of Chari, Kehoe, and McGrattan (2007) to the U.K. data. The idea of this approach is to understand the key drivers of business cycles in terms of "wedges" in the equilibrium conditions. This is useful, as it can aid model selection. Our analysis shows that a productivity decline was the primary determinant of the fall in output in the United Kingdom during the Great Recession. This contrasts with the finding for the United States, where the data suggest a sizable role for labor market frictions (Brinca et al., 2016). In the United Kingdom, the labor "wedge" mainly plays a roles in very short-run fluctuations in labor. Given this divergence, we assess the ability of a simple real business cycle (RBC) dynamic stochastic general equilibrium model augmented with financial shocks proposed by Jermann and Quadrini (2012) to explain the behavior of macroeconomic and financial variables in the U.S. and U.K. economies. In this model, financial shocks chiefly operate on the labor wedge, by constraining firms in their labor choice. Financial conditions are substantially more volatile in the United Kingdom than in the United States, though this is primarily due to a stronger feedback mechanism from productivity to financial conditions in the U.K. data. In both cases, financial shocks amplify the decline in gross domestic product (GDP) during the Great Recession, but elucidating evidence as to the relevance of this channel is revealed by the labor market. In the U.S. case, financial shocks also improve the model's predictions for hours worked, mirroring the importance of the labor wedge for understanding the Great Recession. In the U.K. case, financial shocks cause too large a response of hours worked, calling into question the validity of the mechanism and financial shocks for understanding the episode. 

From 2010 onward following the recession, financial variables returned to normal more quickly in the United States than in the United Kingdom. During this period the downward pressure generated by financial frictions helps explain the prolonged negative response of hours in the United Kingdom. By contrast, in the United States, hours in the data are below where they are predicted to be given the financial shocks observed. More recently, the United Kingdom has seen an hours boom, but through the lens of the model this would imply a loosening of financial constraints not supported by the data. 

This article is related to a broad literature that has sought to incorporate financial variables or a financial sector into macroeconomic models. The seminal contribution to this field was the Bernanke, Gertler, and Gilchrist (1999) financial accelerator model. Other notable contributions include Gertler, Kiyotaki, and Queralto (2012) and Gertler and Kiyotaki (2015). The key difference between the financial accelerator models and the Jermann and Quadrini (2012) set up is that the latter incorporates financial shocks as an important source of disturbances in themselves. The specification employed here owes a significant inspiration to the endogenous credit constraints literature, for example, Kiyotaki and Moore (1997). Christiano, Motto, and Rostagno (2010) have analyzed financial frictions in large-scale New Keynesian dynamic stochastic general equilibrium models, finding financial frictions were important for the empirical properties of such models. Recent work has stressed the importance of firm heterogeneity in the transmission of financial shocks. This heterogeneity can affect the representation of a credit crunch in terms of aggregates wedges (Buera and Moll, 2015) or generate a new precautionary channel, which may have been important in the U.K. experience (Melcangi, 2016).