Natalie Burr
The challenge of gauging financial conditions
Imagine that you are tasked with thinking about how financial conditions may change during a policy tightening cycle. Different economists may come to very different conclusions, and none of them will necessarily be wrong. Why? Because measuring financial conditions is challenging—for a variety of reasons. The Financial Conditions Index (FCI) is a popular solution, and its advantage lies in the disadvantage of the alternative: it is simpler than making a judgment across a range of individual variables. In this post, I am proposing one way to set up an FCI in the UK. I find that financial conditions have tightened significantly over the last couple of years, coming off a period of accommodative conditions in the wake of Covid.
What are the financial conditions?
Financial conditions are generally defined as the ease with which households and businesses can obtain financing. But FCIs can also be used as an (imperfect) measure of the effectiveness of the first stage of the MTM. It is imperfect because factors other than monetary policy can also influence financial conditions, but it is nonetheless useful for assessing the effectiveness of changes in policy rates in feeding financial markets, such as money market interest rates, credit spreads, asset prices, risk premiums and the exchange rate.
How do you measure financial conditions?
When it comes to choosing methodology, there are generally two “segments” of FCI fields. Some indicators have an economic explanation, as the weights of the individual components depend on their historical alignment with macroeconomic variables such as gross domestic product, or inflation. Other indicators use a data-driven approach, deriving weights statistically. My new FCI falls into the latter group. The advantage over the first set of indicators is that they do not impose a fixed relationship between financial conditions and GDP. I would argue that while making this connection is useful, these pointers run the risk of trying to achieve too much in just one pointer. The assumption that the relationship between fiscal and credit variables and GDP is constant over time is strong. In fact, these relationships can be variable over time and depend on the state.
methodology
Introducing a UK FCI generated using principal components analysis (PCA), inspired by Angelopoulou et al (2013). Why PCA? The definition of financial conditions can include a wide range of variables. PCA is a dimension reduction technique that decomposes the covariance structure of a given string into factors common to all, and special noise. PCA is simple and intuitive, allowing the incorporation of a set of closely related explanatory variables, improving the information contained in the input variables. The choice of methodology was motivated by insights from Arrigoni et al (2020), who found that simpler measures of financial conditions perform better than complex statistical methods.
The index uses monthly data on a range of nominal short- and long-term government bond yields, spreads (the difference between some short- and long-term yields), credit spreads (mortgage and unsecured lending), exchange rate and risky asset prices. Data selection is motivated by asset prices and credit variables that are likely to be important to economic activity. Risky asset prices are important because of the wealth effects that influence spending decisions, and they represent the cost of market-based financing for firms. A wide range of interest rates (including mortgage rates) are important because they represent the cost faced by businesses and households who borrow. In the same way, the exchange rate is important because it represents the cost in pounds sterling to other currencies. The question is, what information is most relevant to overall financial conditions? This is what PCA can help with.
First, I prepare the data for analysis. The indicator contributes to the literature on existing FCIs by addressing the issue of instability. Input constancy is particularly important for the FCI to obtain a reasonable interpretation over a longer period of time. I remove these trends by subtracting an estimate of the long-run equilibrium real interest rate (R*) from interest rates, and applying logarithmic differences to stock prices. In the longer term, structural factors not directly related to financial conditions have driven these trends, which would otherwise mean continued easing of financial conditions over time. By backtracking, I ensure a reasonable interpretation of the index, which I will discuss in the next few paragraphs.
Second, the variables that enter PCA have very different sizes and units. Therefore, all variables are standardized by subtracting the mean and dividing by its standard deviation. Finally, the variables are also normalized to account for the way the series affects financial conditions. In order for an increase in FCI to indicate a tightening in financial conditions, variables in which the increase reflects a decrease in financial conditions are entered into the model with an inverted sign.
Finally, the index is compiled using the weights implied by the first three principal components (which explain approximately 80% of the total variance in the data) on the normative variables. Graph 1 shows the index, and decomposition of the contribution from different groups of variables. Financial conditions tightened sharply during the last tightening cycle, but movements have eased somewhat since the sharp rally in September 2022, due to lower yields and a stronger pound.
Chart 1: A new indicator of financial conditions in the UK
Sources: Bloomberg Finance LP, Moneyfacts, Refinitiv Eikon from LSEG, Tradeweb and bank accounts. Last note: January 2023.
By measuring the index to be an average of zero, the index should be interpreted as a relative, rather than an absolute measure of financial conditions. Although an increase (decrease) in the index indicates a tightening (easing) in financial conditions, it is not clear to what extent they are “tight” or “loose” in absolute terms. In other words, financial conditions above zero are aggravating conditions only relative to the historical average.
Similar to current competitive financial indicators, it is not a perfect or aggregate measure of financial conditions. This indicator is a largely reductive model measure and tells us nothing about “why”, for example, interest rates rise and fall. Individual variables are not omitted from their responses to macroeconomic variables (because of the difficulty in doing so accurately), and thus the index does not capture purely external shifts in financial conditions. Movements may be internal to changes in the macroeconomic environment or monetary policy, and movements in asset prices that are attributable solely to macroeconomic shocks are not necessarily meaningful changes in financial conditions.
Thought experiment
This indicator is built on the basis that financial conditions are difficult to measure. Maintaining the belief that a single index cannot deliver the final word, I am testing three alternative index specifications.
First, I don’t allow the weights to be variable over time, so they are generally not robust to changes in the sample. So the index was only re-estimated during the post-financial crisis (GFC) sample period, as shown in Chart 2. Conditions appear to be tighter in the aftermath of the global financial crisis, with spreads and a steeper yield curve contributing the most. But since the 2016 Brexit referendum, when the Bank cut interest rates and conducted quantitative easing, the index has softened relative to historical experience.
Chart 2: Assessment of risk assessment in the UK during the post-financial crisis sample period
Notes: The solid line represents the post-GFC, while the dotted line shows the release of Graph 1.
Sources: Bloomberg Finance LP, Moneyfacts, Refinitiv Eikon from LSEG, Tradeweb and bank accounts. Last note: January 2023.
Second, I’m exploring a “real” version of the indicator, which uses variables of real interest rates, the exchange rate and stock prices. Chart 3 shows that in real terms, financial conditions have contracted to a lesser extent, reflecting the fact that real interest rates in recent cycles of tightening remain largely in negative territory.
Chart 3: The UK’s “real” financial conditions index
Sources: Bloomberg Finance LP, Moneyfacts, Refinitiv Eikon from LSEG, Tradeweb and bank accounts. Last note: December 2022.
Finally, I am exploring a version of the index that removes the effects of the international ramifications on interest rates and stock prices. I use model-based estimates that are determined by the inelasticity of asset prices, which quantify the geographic origin of the underlying shocks. Comparison of post-GFC FCI (dotted) and FCI excluding fallout (solid line) In Chart 4, the international fallout (primarily in the US and EA) slowed the relative tightening of UK financial conditions through 2021, but moderated the uptick in financial conditions around the micro-budget disruption in September 2022.
Chart 4: UK Financial Conditions Index, removing the international spillover
Notes: The solid line represents the FCI excluding the indirect effects, while the dotted line shows the post-GFC release.
Sources: Bloomberg Finance LP, Moneyfacts, Refinitiv Eikon from LSEG, Tradeweb and bank accounts. Last note: January 2023.
After seeing a variety of specifications, Chart 5 brings the correlation back into policy making, by depicting the relationship between financial conditions and bank rate, with a focus on tightening cycles in the UK since the bank’s operational independence in 1997. For each episode, I use a linear trend to illustrate the relationship. There are reasonable arguments for why this relationship is not linear, not least because of the difference in frequency, speed, and size of the spikes. But in general, bank exchange rate tightening has been associated with tightening financial conditions (except for the early period of operational independence). And in the most recent tightening cycle, each unit of bank rate increase resulted in a significant amount of tightening, even more than in previous cycles.
Figure 5: Scatter plot of the UK Financial Conditions Index vs. Bank Rate over previous tightening cycles
Observations: For the hiking cycle 2021–2222, two observations (indicated by gray diamonds) were excluded from the linear trend estimation. These are observations for September and October 2022, which will likely be affected by the micro-budget disruption in September 2022. The flexible indicator used for this graph is the one shown in Chart 1.
Sources: Bloomberg Finance LP, Moneyfacts, Refinitiv Eikon from LSEG, Tradeweb and bank accounts. Last note: January 2023.
In short, financial conditions are a difficult concept to capture in just one indicator. I have argued that indicators of variable rates are nonetheless useful, for assessing how changes in policy rates are transmitted to aggregate financial conditions. I find that financial conditions in the UK have tightened significantly during the most recent tightening cycle, but the degree of tightening is subject to a lot of uncertainty. The hardness tests done by looking at the different shapes of the FCI are indicative of this. Therefore, it is important to focus on a variety of indicators to make a robust and comprehensive assessment of financial conditions.
Natalie Burr Works in the bank’s external MPC unit.
If you would like to get in touch, please email us at bankunderground@bankofengland.co.uk or leave a comment below.
Comments will appear once approved by the moderator, and will only be posted when full name is submitted. Bank Underground is a blog for Bank of England staff to share opinions that challenge – or support – mainstream, conventional politics. The opinions expressed herein are those of the authors, and are not necessarily those of the Bank of England or its policy committees.
Share the post “The Challenge of Measuring Financial Conditions”
