Kristina Bluwstein and Julieta Yung
Financial markets provide insightful information about the level of risk in the economy. However, sometimes market participants might be driven more by their perception rather than any fundamental changes in risk. In a recent Staff Working Paper we study the effect of changes in risk perceptions that can lead to a mispricing of risk. We find that when agents over-price risk, banks adjust their bank lending policies, which can lead to depressed investment and output. On the other hand, when agents under-price risk, excessive lending creates a ‘bad’ credit boom that can lead to a severe recession once sentiment is reversed.
Risk perceptions and the term premium
Risk perceptions are often highly subjective and notoriously hard to quantify, yet can have sizable effects on the economy. To measure movements in bond markets that are not driven by changes in fundamentals, we turn to something called the term premium. The term premium is the compensation that investors require in order to hold a bond that exposes them to duration risk instead of holding a consecutive series of short-term assets for the same amount of time. As such, movements in the price of a long-term bond can reflect (i) changes in the ‘expectations’ component, as we learn new information about the state of the economy, or (ii) changes in the compensation for duration risk, the term premium component.
Figure 1 shows the term premium corresponding to a 10-year U.S. Treasury yield, as estimated by the New York Fed. The figure indicates that the term premium varies over time, and has generally been positive over the past six decades, with an average of around 1.6%. However, the term premium has been trending downward since the mid-1980s, with a value close to zero in the most recent years prompting questions on whether duration risk is being mispriced by the markets.
Importantly, the term premium is highly correlated with other measures related to financial market risk, interest rate volatility, economic policy uncertainty and inflation uncertainty and can therefore serve as a useful indicator of financial market participants’ risk perception.
Figure 1: The U.S. Term Premium
What do the data say?
To understand whether risk perception shocks can affect not only bond markets but also economic activity, we collect monthly data on bank lending (loans to the private and public sector) and the economy (output, inflation and interest rates), and using a vector autoregressive model we explore the role of term premium shocks. We find that a 90 basis-point increase in the term premium is associated with a 2% decline in short-term commercial loans granted by banks (see Figure 2). This shock is even more persistent than the effect on output, with the median response only returning to the steady state after roughly five years. Interestingly, banks increase their long-term lending to the public sector by more than 4% in response to the same shock, suggesting a portfolio rebalancing strategy following heightened risk perceptions.
Figure 2: Effect of a Term Premium Shock on Growth (based on empirical model)
(a) Short-Term Loans to the Commercial Sector
(b) Long-Term Loans to the Public Sector
(c) Annual Output Growth
Although we establish a relationship between the term premium and the economy, a caveat of this analysis is that it remains agnostic on the mechanisms leading to the effects we observe in the data. In order to investigate the channels through which risk perceptions impact bank lending and the economy, we need a structural model to help quantify the different mechanisms at play.
What does the model say?
There are currently only a few structural models that allow us to combine banking, asset pricing, and expectations. We therefore develop a new dynamic stochastic general equilibrium (DSGE) model to study the role of risk perceptions in shaping banks’ decisions to tighten bank lending, while maintaining econometric tractability. In this framework, households, entrepreneurs, banks, the central bank and the government make choices to achieve their objectives (maximise utility, stabilise the economy, increase profits). Figure 3 illustrates how all sectors in the economy interact with one another, providing us with a simplistic but intuitive representation of the channels through which financial markets influence the economy and vice versa.
Figure 3: Bank Lending in the DSGE Model
Our model has three important features. First, we introduce a more realistic financial sector, in which short-term rates are set by the central bank according to inflation and economic growth conditions, while long-term bonds issued by the government are priced by investors following general asset pricing rules. Banks take this into account when choosing how much and at what cost to lend to the private or public sector, influencing the availability of credit in the economy.
Second, we introduce a feedback loop between the financial sector and the economy via the term premium. That is, when agents suddenly perceive more/less risk in the economy, they misprice their compensation for risk exposure which leads to a change in the term premium. That change in the term premium influences the pricing of long-term bonds, thus affecting financial markets. We remain agnostic about what drives the unexpected shock in risk perception that allows the term premium to change. However, pessimistic expectations about the future or lower risk tolerance could lead to an over-pricing of risk in the form of a higher term premium. The opposite holds if agents under-price the actual underlying risk due to eg exuberant expectations, which would translate into a lower term premium.
Third, our model can explain both salient features of the economy and the financial sector, a notoriously difficult task given that the economy is slow moving and financial markets are more volatile. Our model therefore captures important dynamics that describe the way in which the economy operates at an aggregate level.
We find that when financial market participants ‘panic’, ie their perception of risk increases without any change in macroeconomic fundamentals, banks restrict their lending to the private sector and increase their holdings of long-term government bonds, as previously identified in the data (see Figure 4). However, we are now able to identify the mechanisms that link risk perceptions to bank lending, and ultimately the economy.
Figure 4: Effect of a Risk Perception Shock (based on theoretical model)
(a) Short-Term Loans to the Commercial Sector
(b) Long-Term Loans to the Public Sector
(c) Annual Output Growth
In this model, when perceived risk increases, investors seek insurance against bad outcomes and over-price risk. Therefore, their compensation for risk (the term premium) increases, pushing long-term interest rates higher. Elevated perceived risk increases the price at which banks are willing to provide loans to the commercial sector, as they themselves are indifferent between lending short-term to the private sector or long-term to the government. Unlike the government, the private sector needs to provide collateral, when they borrow money from the bank. Therefore, with higher costs of borrowing, fewer people can afford to take on a new loan and the total quantity of short-term loans to the private sector drops. As a consequence, investment and hence real economic activity drops (Figure 4.c).
Policy implications
Understanding bank lending and the factors that influence the availability of credit is of particular interest for financial stability and monetary policy transmission throughout the business cycle. Using our model, we can simulate different types of credit booms and find that a ‘bad’ credit boom, ie a boom driven by agents under-pricing risk in the economy, is less supportive of economic growth. A ‘good’ credit boom, ie one that is driven by strong economic fundamentals, however, allows for higher investment and consumption, and once over, remains supportive of economic growth.
Kristina Bluwstein works in the Bank’s Macroprudential Strategy and Support Division and Julieta Yung works at Bates College.
If you want to get in touch, please email us at bankunderground@bankofengland.co.uk or leave a comment below.
Comments will only appear once approved by a moderator, and are only published where a full name is supplied.Bank Underground is a blog for Bank of England staff to share views that challenge – or support – prevailing policy orthodoxies. The views expressed here are those of the authors, and are not necessarily those of the Bank of England, or its policy committees.
